Hospital Medicine

Digitalization of patient-specific information over the past 2 decades has dramatically altered health care delivery. Nonetheless, this technology has yet to live up to its promise of improving patient outcomes, in part due to data storage challenges as well as the emphasis on data entry to support administrative and financial goals of the institution.^1-4 Substantially less emphasis has been placed on the retrieval of information required for accurate diagnosis.

A new search engine, Voogle, is now available through Microsoft Internet Explorer (Redmond, WA) to all providers in the US Department of Veterans Affairs (VA) on any intranet-enabled computer behind the VA firewall. Voogle facilitates rapid query-based search and retrieval of patient-specific data in the VA Corporate Data Warehouse (CDW).

Case Example

A veteran presented requesting consideration for implantation of a new device for obstructive sleep apnea. Guidelines for implantation of the new device specify a narrow therapeutic window, so determination of his apnea-hypopnea index (AHI) was critical. The patient had received care at more than 20 VA facilities and knew the approximate year the test had been performed at a non-VA facility.

A health care provider (HCP) using Voogle from his VA computer indexed all Veterans Information Systems and Technology Architecture (VistA) notes for the desired date range. The indexing of > 200 notes was completed in seconds. The HCP opened the indexed records with Voogle and entered a query for “sleep apnea,” which displayed multiple instances of the term within the patient record notes. A VA HCP had previously entered the data from the outside sleep study into a note shortly after the study.

This information was found immediately by sorting the indexed notes by date. The total time required by Voogle to find and display the critical information from the sleep study entered at a different VA more than a dozen years earlier was about 1 minute. These data provided the information needed for decision making at the time of the current patient encounter, without which repeat (and unnecessary) testing would have been required.

Information Overload

Electronic health records (EHRs) such as VistA, upload, store, collate, and present data in near real-time across multiple locations. Although the availability of these data can potentially reduce the risk of error due to missing critical information, its sheer volume limits its utility for point-of-care decision making. Much patient-specific text data found in clinical notes are recorded for administrative, financial, and business purposes rather than to support patient care decision making.^1-3 The majority of data documents processes of care rather than HCP observations, assessment of current status, or plans for care. Much of this text is inserted into templates, consists of imported structured data elements, and may contain repeated copy-and-paste free text.

Data uploaded to the CDW are aggregated from multiple hospitals, each with its own “instance” of VistA. Often the CDW contains thousands of text notes for a single patient. This volume of text may conceal critical historical information needed for patient care mixed with a plethora of duplicated or extraneous text entered to satisfy administrative requirements. The effects of information overload and poor system usability have been studied extensively in other disciplines, but this science has largely not been incorporated into EHR design.^1,3,4

A position paper published recently by the American College of Physicians notes that physician cognitive work is adversely impacted by the incorporation of nonclinical information into the EHR for use by other administrative and financial functions.²

Information Chaos

Beasley and colleagues noted that information in an EHR needed for optimal care may be unavailable, inadequate, scattered, conflicting, lost, or inaccurate, a condition they term information chaos.⁵ Smith and colleagues reported that decision making in 1 of 7 primary care visits was impaired by missing critical information. Surveyed HCPs estimated that 44% of patients with missing information may receive compromised care as a result, including delayed or erroneous diagnosis and increased costs due to duplication of diagnostic testing.⁶

Even when technically available, the usability of patient-specific data needed for accurate diagnosis is compromised if the HCP cannot find the information. In most systems data storage paradigms mirror database design rather than provider cognitive models. Ultimately, the design of current EHR interaction paradigms squanders precious cognitive resources and time, particularly during patient encounters, leaving little available for the cognitive tasks necessary for accurate diagnosis and treatment decisions.^1,3,4,7

VA Corporate Data Warehouse

VistA was implemented as a decentralized system with 130 instances, each of which is a freestanding EHR. However, as all systems share common data structures, the data can be combined from multiple instances when needed. The VA established a CDW more than 15 years ago in order to collate information from multiple sites to support operations as well as to seek new insights. The CDW currently updates nightly from all 130 EHR instances and is the only location in which patient information from all treating sites is combined. Voogle can access the CDW through the Veterans Informatics and Computing Infrastructure (VINCI), which is a mirror of the CDW databases and was established as a secure research environment.

The CDW contains information on 25 million veterans, with about 15 terabytes of text data. Approximately 4 billion data points, including 1 million text notes, are accrued nightly. The Integrated Control Number (ICN), a unique patient identifier, is assigned to each CDW record and is cross-indexed in the master patient index. All CDW data are tied to the ICN, facilitating access to and attribution of all patient data from all VA sites. Voogle relies on this identifier to build indexed files, or domains (which are document collections), of requested specific patient information to support its search algorithm.

Structured Data

Most of the data accrued in an EHR are structured data (such as laboratory test results and vital signs) and stored in a defined database framework. Voogle uses iFind (Intersystems Inc, Cambridge, MA) to index, count, and then search for requested information within structured data fields.

Unstructured Text

In contrast to structured data, text notes are stored as documents that are retrievable by patient, author, date, clinic, as well as numerous other fields. Unstructured (free) text notes are more information rich than either structured data or templated notes since their narrative format more closely parallels providers’ cognitive processes.^1,7 The value of the narrative becomes even more critical in understanding complex clinical scenarios with multiple interacting disease processes. Narratives emphasize important details, reducing cognitive overload by reducing the salience of detail the author deems to be less critical. Narrative notes simultaneously assure availability through the use of unstandardized language, often including specialty and disease-specific abbreviations.¹ Information needed for decision making in the illustrative case in this report was present only in HCP-entered free-text notes, as the structured data from which the free text was derived were not available.

Search

The introduction of search engines can be considered one of the major technologic disruptors of the 21st century.⁸ However, this advance has not yet made significant inroads into health care, despite advances in other domains. As of 2019, EHR users are still required to be familiar with the system’s data and menu structure in order to find needed information (or enter orders, code visits, or any of a number of tasks). Anecdotally, one of the authors (David Eibling) observed that the most common question from his trainees is “How do you . . .?” referring not to the care of the patient but rather to interaction with the EHR.

What is needed is a simple query-based application that finds the data on request. In addition to Voogle, other advances are being made in this arena such as the EMERSE, medical record search engine (project-emerse.org). Voogle was released to VA providers in 2017 and is available through the Internet Explorer browser on VA computers with VA intranet access. The goal of Voogle is to reduce HCP cognitive load by reducing the time and effort needed to seek relevant information for the care of a specific patient.

Natural Language Processing

Linguistic analysis of text seeking to understand its meaning constitutes a rapidly expanding field, with current heavy emphasis on the role of artificial intelligence and machine learning.¹ Advances in processing both structured data and free-text notes in the health care domain is in its infancy, despite the investment of considerable resources. Undoubtedly, advances in this arena will dramatically change provider cognitive work in the next decades.

VistA is coded in MUMPS (Massachusetts General Hospital Utility Multi-Programming System, also known as M), which has been in use for more than 50 years. Voogle employs iKnow, a novel natural language processing (NLP) application that resides in Caché (Intersystems, Boston, MA), the vendor-supported MUMPS infrastructure VistA uses to perform text analysis. iKnow does not attempt to interpret the meaning of text as do other common NLP applications, but instead relies on the expert user to interpret the meaning of the analyzed text. iKnow initially divides sentences into relations (usually verbs) and concepts, and then generates an index of these entities. The efficiency of iKnow results in very rapid indexing—often several thousand notes (not an uncommon number) can be indexed in 20 to 30 seconds. iKnow responds to a user query by searching for specific terms or similar terms within the indexed text, and then displays these terms within the original source documents, similar to well-known commercial search engines. Structured data are indexed by the iFind program simultaneously with free-text indexing (Figure 1).

Security

Maintaining high levels of security of Health Insurance Portability and Accountability (HIPAA)-compliant information in an online application such as Voogle is critical to ensure trust of veterans and HCPs. All patient data accessed by Voogle reside within the secure firewall-protected VINCI environment. All moving information is protected with high-level encryption protocols (transport layer security [TLS]), and data at rest are also encrypted. As the application is online, no data are stored on the accessing device. Voogle uses a secure Microsoft Windows logon using VA Active Directory coupled with VistA authorization to regulate who can see the data and use the application. All access is audited, not only for “sensitive patients,” but also for specific data types. Users are reminded of this Voogle attribute on the home screen.

Accessing Voogle

Voogle is available on the VA intranet to all authorized users at https://voogle.vha.med.va.gov/voogle. To assure high-level security the application can only be accessed with the Internet Explorer browser using established user identification protocols to avoid unauthorized access or duplicative log-in tasks.

Indexing

Indexing is user-driven and is required prior to patient selection and term query. The user is prompted for a patient identifier and a date range. The CDW unique patient identifier is used for all internal processing. However, a social security number look-up table is incorporated to facilitate patient selection. The date field defaults to 3 years but can be extended to approximately the year 2000.

Queries

Entering the patient name in Lastname, Firstname (no space) format will yield a list of indexed patients. All access is audited in order to deter unauthorized queries. Data from a demonstration patient are displayed in Figures 2, 3, 4, 5,
and 6.

Structured Data Searches

Structured data categories that contain the queried term, as well as a term count, are displayed after the “Structured Data” toggle is selected (Figure 2). After the desired category (Figure 2: “Outpatient Rx”) is selected, Voogle accesses the data file and displays it as a grid (medication list, Figure 3). Filter and sort functions enable display of specific medications, drug classes, or date ranges (Figure 4).

Display of Terms Within Text Notes

Selecting a term from the drop-down list (Figure 5) opens a grid with the term highlighted in a snippet of text (Figure 6). Opening the document displays the context of the term, along with negation terms (ie, not, denies, no, etc) in red font if present. Voogle, unlike other NLP tools that attempt to interpret medical notes, relies on interpretation by the HCP user. Duplicate note fragments will be displayed in multiple notes, often across multiple screens, vividly demonstrating the pervasive use of the copy-and-paste text-entry strategy. Voogle satisfies 2 of the 4 recommendations of the recent report on copy-and-paste by Tsou and colleagues.⁹ The Voogle text display grid identifies copy-and-pasted text as well as establishes the provenance of the text (by sorting on the date column). Text can be copied from Voogle into an active Computerized Patient Record System (CPRS) note if needed for active patient care. Reindexing the following day and then repeating the search will demonstrate the newly copied-and-pasted text appended to the sequence.

Limitations

Voogle is unable to access all VA patient data currently. There are a dozen or so clinical domains that are indexed by Voogle that include prescriptions, problem lists, health factors, and others. More domains can be added with minimal effort and would then be available for fast search. The most critical deficiency is its inability to access, index, or query text reports stored as images within VistA Imaging. This includes nearly all reports from outside HCPs, emergency department visits or discharge summaries from unlinked hospitals, anesthesia reports, intensive care unit flow sheets, electrocardiograms, as well as numerous other text reports such as pulmonary function reports or sleep studies. Information that is transcribed by the provider into VistA as text (as in the case presented) is available within the CDW and can be found and displayed by Voogle search.

Voogle requires that the user initiates the indexing process prior to initiating the search process. Although Voogle defaults to 3 years prior to the current date, the user can specify a start date extending to close to the year 2000. The volume of data flowing into the CDW precludes automatic indexing of all patient data, as well as automatic updating of previously indexed data. We have explored the feasibility of queueing scheduled appointments for the following day, and although the strategy shows some promise, avoiding conflict with user-requested on-demand indexing remains challenging.

The current VA network architecture updates the CDW every night, resulting in up to a 24-hour delay in data availability. However, this delay should be reduced to several minutes after implementation of real-time data feeds accompanying the coming transition to a new EHR platform.

Conclusions

The recent introduction of the Joint Legacy Viewer (JLV) to the VA EHR desktop has enhanced the breadth of patient-specific information available to any VHA clinician, with recent enhancements providing access to some community care notes from outside HCPs. Voogle builds on this capability by enabling rapid search of text notes and structured data from multiple VA sites, over an extended time frame, and perhaps entered by hundreds of authors, as demonstrated in the case example. Formal usability and workload studies have not been performed; however, anecdotal reports indicate the application dramatically reduces the time required to search for critical information needed for care of complex patients who have been treated in multiple different VA hospitals and clinics.

The Voogle paradigm of leveraging patient information stored within a large enterprise-wide data warehouse through NLP techniques may be applicable to other systems as well, and warrants exploration. We believe that replacing traditional data search paradigms that require knowledge of data structure with a true query-based paradigm is a potential game changer for health information systems. Ultimately this strategy may help provide an antidote for the information chaos impacting HCP cognition. Moreover, reducing HCP cognitive load and time on task may lessen overall health care costs, reduce provider burn-out, and improve the quality of care received by patients.

Near real-time data feeds and adding additional clinical domains will potentially provide other benefits to patient care. For example, the authors plan to investigate whether sampling incoming data may assist with behind-the-scenes continuous monitoring of indicators of patient status to facilitate early warning of impending physiologic collapse.¹⁰ Other possible applications could include real-time scans for biosurveillance or other population screening requirements.

Acknowledgments
The authors express their sincere appreciation to Leslie DeYoung for documentation and Justin Wilson who constructed much of the graphical user interface for the Voogle application and design. Without their expertise, passion, and commitment the application would not be available as it is now.

Digitalization of patient-specific information over the past 2 decades has dramatically altered health care delivery. Nonetheless, this technology has yet to live up to its promise of improving patient outcomes, in part due to data storage challenges as well as the emphasis on data entry to support administrative and financial goals of the institution.^1-4 Substantially less emphasis has been placed on the retrieval of information required for accurate diagnosis.

A new search engine, Voogle, is now available through Microsoft Internet Explorer (Redmond, WA) to all providers in the US Department of Veterans Affairs (VA) on any intranet-enabled computer behind the VA firewall. Voogle facilitates rapid query-based search and retrieval of patient-specific data in the VA Corporate Data Warehouse (CDW).

Case Example

A veteran presented requesting consideration for implantation of a new device for obstructive sleep apnea. Guidelines for implantation of the new device specify a narrow therapeutic window, so determination of his apnea-hypopnea index (AHI) was critical. The patient had received care at more than 20 VA facilities and knew the approximate year the test had been performed at a non-VA facility.

A health care provider (HCP) using Voogle from his VA computer indexed all Veterans Information Systems and Technology Architecture (VistA) notes for the desired date range. The indexing of > 200 notes was completed in seconds. The HCP opened the indexed records with Voogle and entered a query for “sleep apnea,” which displayed multiple instances of the term within the patient record notes. A VA HCP had previously entered the data from the outside sleep study into a note shortly after the study.

This information was found immediately by sorting the indexed notes by date. The total time required by Voogle to find and display the critical information from the sleep study entered at a different VA more than a dozen years earlier was about 1 minute. These data provided the information needed for decision making at the time of the current patient encounter, without which repeat (and unnecessary) testing would have been required.

Information Overload

Electronic health records (EHRs) such as VistA, upload, store, collate, and present data in near real-time across multiple locations. Although the availability of these data can potentially reduce the risk of error due to missing critical information, its sheer volume limits its utility for point-of-care decision making. Much patient-specific text data found in clinical notes are recorded for administrative, financial, and business purposes rather than to support patient care decision making.^1-3 The majority of data documents processes of care rather than HCP observations, assessment of current status, or plans for care. Much of this text is inserted into templates, consists of imported structured data elements, and may contain repeated copy-and-paste free text.

Data uploaded to the CDW are aggregated from multiple hospitals, each with its own “instance” of VistA. Often the CDW contains thousands of text notes for a single patient. This volume of text may conceal critical historical information needed for patient care mixed with a plethora of duplicated or extraneous text entered to satisfy administrative requirements. The effects of information overload and poor system usability have been studied extensively in other disciplines, but this science has largely not been incorporated into EHR design.^1,3,4

A position paper published recently by the American College of Physicians notes that physician cognitive work is adversely impacted by the incorporation of nonclinical information into the EHR for use by other administrative and financial functions.²

Information Chaos

Beasley and colleagues noted that information in an EHR needed for optimal care may be unavailable, inadequate, scattered, conflicting, lost, or inaccurate, a condition they term information chaos.⁵ Smith and colleagues reported that decision making in 1 of 7 primary care visits was impaired by missing critical information. Surveyed HCPs estimated that 44% of patients with missing information may receive compromised care as a result, including delayed or erroneous diagnosis and increased costs due to duplication of diagnostic testing.⁶

Even when technically available, the usability of patient-specific data needed for accurate diagnosis is compromised if the HCP cannot find the information. In most systems data storage paradigms mirror database design rather than provider cognitive models. Ultimately, the design of current EHR interaction paradigms squanders precious cognitive resources and time, particularly during patient encounters, leaving little available for the cognitive tasks necessary for accurate diagnosis and treatment decisions.^1,3,4,7

VA Corporate Data Warehouse

VistA was implemented as a decentralized system with 130 instances, each of which is a freestanding EHR. However, as all systems share common data structures, the data can be combined from multiple instances when needed. The VA established a CDW more than 15 years ago in order to collate information from multiple sites to support operations as well as to seek new insights. The CDW currently updates nightly from all 130 EHR instances and is the only location in which patient information from all treating sites is combined. Voogle can access the CDW through the Veterans Informatics and Computing Infrastructure (VINCI), which is a mirror of the CDW databases and was established as a secure research environment.

The CDW contains information on 25 million veterans, with about 15 terabytes of text data. Approximately 4 billion data points, including 1 million text notes, are accrued nightly. The Integrated Control Number (ICN), a unique patient identifier, is assigned to each CDW record and is cross-indexed in the master patient index. All CDW data are tied to the ICN, facilitating access to and attribution of all patient data from all VA sites. Voogle relies on this identifier to build indexed files, or domains (which are document collections), of requested specific patient information to support its search algorithm.

Structured Data

Most of the data accrued in an EHR are structured data (such as laboratory test results and vital signs) and stored in a defined database framework. Voogle uses iFind (Intersystems Inc, Cambridge, MA) to index, count, and then search for requested information within structured data fields.

Unstructured Text

In contrast to structured data, text notes are stored as documents that are retrievable by patient, author, date, clinic, as well as numerous other fields. Unstructured (free) text notes are more information rich than either structured data or templated notes since their narrative format more closely parallels providers’ cognitive processes.^1,7 The value of the narrative becomes even more critical in understanding complex clinical scenarios with multiple interacting disease processes. Narratives emphasize important details, reducing cognitive overload by reducing the salience of detail the author deems to be less critical. Narrative notes simultaneously assure availability through the use of unstandardized language, often including specialty and disease-specific abbreviations.¹ Information needed for decision making in the illustrative case in this report was present only in HCP-entered free-text notes, as the structured data from which the free text was derived were not available.

Search

The introduction of search engines can be considered one of the major technologic disruptors of the 21st century.⁸ However, this advance has not yet made significant inroads into health care, despite advances in other domains. As of 2019, EHR users are still required to be familiar with the system’s data and menu structure in order to find needed information (or enter orders, code visits, or any of a number of tasks). Anecdotally, one of the authors (David Eibling) observed that the most common question from his trainees is “How do you . . .?” referring not to the care of the patient but rather to interaction with the EHR.

What is needed is a simple query-based application that finds the data on request. In addition to Voogle, other advances are being made in this arena such as the EMERSE, medical record search engine (project-emerse.org). Voogle was released to VA providers in 2017 and is available through the Internet Explorer browser on VA computers with VA intranet access. The goal of Voogle is to reduce HCP cognitive load by reducing the time and effort needed to seek relevant information for the care of a specific patient.

Natural Language Processing

Linguistic analysis of text seeking to understand its meaning constitutes a rapidly expanding field, with current heavy emphasis on the role of artificial intelligence and machine learning.¹ Advances in processing both structured data and free-text notes in the health care domain is in its infancy, despite the investment of considerable resources. Undoubtedly, advances in this arena will dramatically change provider cognitive work in the next decades.

VistA is coded in MUMPS (Massachusetts General Hospital Utility Multi-Programming System, also known as M), which has been in use for more than 50 years. Voogle employs iKnow, a novel natural language processing (NLP) application that resides in Caché (Intersystems, Boston, MA), the vendor-supported MUMPS infrastructure VistA uses to perform text analysis. iKnow does not attempt to interpret the meaning of text as do other common NLP applications, but instead relies on the expert user to interpret the meaning of the analyzed text. iKnow initially divides sentences into relations (usually verbs) and concepts, and then generates an index of these entities. The efficiency of iKnow results in very rapid indexing—often several thousand notes (not an uncommon number) can be indexed in 20 to 30 seconds. iKnow responds to a user query by searching for specific terms or similar terms within the indexed text, and then displays these terms within the original source documents, similar to well-known commercial search engines. Structured data are indexed by the iFind program simultaneously with free-text indexing (Figure 1).

Security

Maintaining high levels of security of Health Insurance Portability and Accountability (HIPAA)-compliant information in an online application such as Voogle is critical to ensure trust of veterans and HCPs. All patient data accessed by Voogle reside within the secure firewall-protected VINCI environment. All moving information is protected with high-level encryption protocols (transport layer security [TLS]), and data at rest are also encrypted. As the application is online, no data are stored on the accessing device. Voogle uses a secure Microsoft Windows logon using VA Active Directory coupled with VistA authorization to regulate who can see the data and use the application. All access is audited, not only for “sensitive patients,” but also for specific data types. Users are reminded of this Voogle attribute on the home screen.

Accessing Voogle

Voogle is available on the VA intranet to all authorized users at https://voogle.vha.med.va.gov/voogle. To assure high-level security the application can only be accessed with the Internet Explorer browser using established user identification protocols to avoid unauthorized access or duplicative log-in tasks.

Indexing

Indexing is user-driven and is required prior to patient selection and term query. The user is prompted for a patient identifier and a date range. The CDW unique patient identifier is used for all internal processing. However, a social security number look-up table is incorporated to facilitate patient selection. The date field defaults to 3 years but can be extended to approximately the year 2000.

Queries

Entering the patient name in Lastname, Firstname (no space) format will yield a list of indexed patients. All access is audited in order to deter unauthorized queries. Data from a demonstration patient are displayed in Figures 2, 3, 4, 5,
and 6.

Structured Data Searches

Structured data categories that contain the queried term, as well as a term count, are displayed after the “Structured Data” toggle is selected (Figure 2). After the desired category (Figure 2: “Outpatient Rx”) is selected, Voogle accesses the data file and displays it as a grid (medication list, Figure 3). Filter and sort functions enable display of specific medications, drug classes, or date ranges (Figure 4).

Display of Terms Within Text Notes

Selecting a term from the drop-down list (Figure 5) opens a grid with the term highlighted in a snippet of text (Figure 6). Opening the document displays the context of the term, along with negation terms (ie, not, denies, no, etc) in red font if present. Voogle, unlike other NLP tools that attempt to interpret medical notes, relies on interpretation by the HCP user. Duplicate note fragments will be displayed in multiple notes, often across multiple screens, vividly demonstrating the pervasive use of the copy-and-paste text-entry strategy. Voogle satisfies 2 of the 4 recommendations of the recent report on copy-and-paste by Tsou and colleagues.⁹ The Voogle text display grid identifies copy-and-pasted text as well as establishes the provenance of the text (by sorting on the date column). Text can be copied from Voogle into an active Computerized Patient Record System (CPRS) note if needed for active patient care. Reindexing the following day and then repeating the search will demonstrate the newly copied-and-pasted text appended to the sequence.

Limitations

Voogle is unable to access all VA patient data currently. There are a dozen or so clinical domains that are indexed by Voogle that include prescriptions, problem lists, health factors, and others. More domains can be added with minimal effort and would then be available for fast search. The most critical deficiency is its inability to access, index, or query text reports stored as images within VistA Imaging. This includes nearly all reports from outside HCPs, emergency department visits or discharge summaries from unlinked hospitals, anesthesia reports, intensive care unit flow sheets, electrocardiograms, as well as numerous other text reports such as pulmonary function reports or sleep studies. Information that is transcribed by the provider into VistA as text (as in the case presented) is available within the CDW and can be found and displayed by Voogle search.

Voogle requires that the user initiates the indexing process prior to initiating the search process. Although Voogle defaults to 3 years prior to the current date, the user can specify a start date extending to close to the year 2000. The volume of data flowing into the CDW precludes automatic indexing of all patient data, as well as automatic updating of previously indexed data. We have explored the feasibility of queueing scheduled appointments for the following day, and although the strategy shows some promise, avoiding conflict with user-requested on-demand indexing remains challenging.

The current VA network architecture updates the CDW every night, resulting in up to a 24-hour delay in data availability. However, this delay should be reduced to several minutes after implementation of real-time data feeds accompanying the coming transition to a new EHR platform.

Conclusions

The recent introduction of the Joint Legacy Viewer (JLV) to the VA EHR desktop has enhanced the breadth of patient-specific information available to any VHA clinician, with recent enhancements providing access to some community care notes from outside HCPs. Voogle builds on this capability by enabling rapid search of text notes and structured data from multiple VA sites, over an extended time frame, and perhaps entered by hundreds of authors, as demonstrated in the case example. Formal usability and workload studies have not been performed; however, anecdotal reports indicate the application dramatically reduces the time required to search for critical information needed for care of complex patients who have been treated in multiple different VA hospitals and clinics.

The Voogle paradigm of leveraging patient information stored within a large enterprise-wide data warehouse through NLP techniques may be applicable to other systems as well, and warrants exploration. We believe that replacing traditional data search paradigms that require knowledge of data structure with a true query-based paradigm is a potential game changer for health information systems. Ultimately this strategy may help provide an antidote for the information chaos impacting HCP cognition. Moreover, reducing HCP cognitive load and time on task may lessen overall health care costs, reduce provider burn-out, and improve the quality of care received by patients.

Near real-time data feeds and adding additional clinical domains will potentially provide other benefits to patient care. For example, the authors plan to investigate whether sampling incoming data may assist with behind-the-scenes continuous monitoring of indicators of patient status to facilitate early warning of impending physiologic collapse.¹⁰ Other possible applications could include real-time scans for biosurveillance or other population screening requirements.

Acknowledgments
The authors express their sincere appreciation to Leslie DeYoung for documentation and Justin Wilson who constructed much of the graphical user interface for the Voogle application and design. Without their expertise, passion, and commitment the application would not be available as it is now.

Digitalization of patient-specific information over the past 2 decades has dramatically altered health care delivery. Nonetheless, this technology has yet to live up to its promise of improving patient outcomes, in part due to data storage challenges as well as the emphasis on data entry to support administrative and financial goals of the institution.^1-4 Substantially less emphasis has been placed on the retrieval of information required for accurate diagnosis.

A new search engine, Voogle, is now available through Microsoft Internet Explorer (Redmond, WA) to all providers in the US Department of Veterans Affairs (VA) on any intranet-enabled computer behind the VA firewall. Voogle facilitates rapid query-based search and retrieval of patient-specific data in the VA Corporate Data Warehouse (CDW).

Case Example

A veteran presented requesting consideration for implantation of a new device for obstructive sleep apnea. Guidelines for implantation of the new device specify a narrow therapeutic window, so determination of his apnea-hypopnea index (AHI) was critical. The patient had received care at more than 20 VA facilities and knew the approximate year the test had been performed at a non-VA facility.

A health care provider (HCP) using Voogle from his VA computer indexed all Veterans Information Systems and Technology Architecture (VistA) notes for the desired date range. The indexing of > 200 notes was completed in seconds. The HCP opened the indexed records with Voogle and entered a query for “sleep apnea,” which displayed multiple instances of the term within the patient record notes. A VA HCP had previously entered the data from the outside sleep study into a note shortly after the study.

This information was found immediately by sorting the indexed notes by date. The total time required by Voogle to find and display the critical information from the sleep study entered at a different VA more than a dozen years earlier was about 1 minute. These data provided the information needed for decision making at the time of the current patient encounter, without which repeat (and unnecessary) testing would have been required.

Information Overload

Electronic health records (EHRs) such as VistA, upload, store, collate, and present data in near real-time across multiple locations. Although the availability of these data can potentially reduce the risk of error due to missing critical information, its sheer volume limits its utility for point-of-care decision making. Much patient-specific text data found in clinical notes are recorded for administrative, financial, and business purposes rather than to support patient care decision making.^1-3 The majority of data documents processes of care rather than HCP observations, assessment of current status, or plans for care. Much of this text is inserted into templates, consists of imported structured data elements, and may contain repeated copy-and-paste free text.

Data uploaded to the CDW are aggregated from multiple hospitals, each with its own “instance” of VistA. Often the CDW contains thousands of text notes for a single patient. This volume of text may conceal critical historical information needed for patient care mixed with a plethora of duplicated or extraneous text entered to satisfy administrative requirements. The effects of information overload and poor system usability have been studied extensively in other disciplines, but this science has largely not been incorporated into EHR design.^1,3,4

A position paper published recently by the American College of Physicians notes that physician cognitive work is adversely impacted by the incorporation of nonclinical information into the EHR for use by other administrative and financial functions.²

Information Chaos

Beasley and colleagues noted that information in an EHR needed for optimal care may be unavailable, inadequate, scattered, conflicting, lost, or inaccurate, a condition they term information chaos.⁵ Smith and colleagues reported that decision making in 1 of 7 primary care visits was impaired by missing critical information. Surveyed HCPs estimated that 44% of patients with missing information may receive compromised care as a result, including delayed or erroneous diagnosis and increased costs due to duplication of diagnostic testing.⁶

Even when technically available, the usability of patient-specific data needed for accurate diagnosis is compromised if the HCP cannot find the information. In most systems data storage paradigms mirror database design rather than provider cognitive models. Ultimately, the design of current EHR interaction paradigms squanders precious cognitive resources and time, particularly during patient encounters, leaving little available for the cognitive tasks necessary for accurate diagnosis and treatment decisions.^1,3,4,7

VA Corporate Data Warehouse

VistA was implemented as a decentralized system with 130 instances, each of which is a freestanding EHR. However, as all systems share common data structures, the data can be combined from multiple instances when needed. The VA established a CDW more than 15 years ago in order to collate information from multiple sites to support operations as well as to seek new insights. The CDW currently updates nightly from all 130 EHR instances and is the only location in which patient information from all treating sites is combined. Voogle can access the CDW through the Veterans Informatics and Computing Infrastructure (VINCI), which is a mirror of the CDW databases and was established as a secure research environment.

The CDW contains information on 25 million veterans, with about 15 terabytes of text data. Approximately 4 billion data points, including 1 million text notes, are accrued nightly. The Integrated Control Number (ICN), a unique patient identifier, is assigned to each CDW record and is cross-indexed in the master patient index. All CDW data are tied to the ICN, facilitating access to and attribution of all patient data from all VA sites. Voogle relies on this identifier to build indexed files, or domains (which are document collections), of requested specific patient information to support its search algorithm.

Structured Data

Most of the data accrued in an EHR are structured data (such as laboratory test results and vital signs) and stored in a defined database framework. Voogle uses iFind (Intersystems Inc, Cambridge, MA) to index, count, and then search for requested information within structured data fields.

Unstructured Text

In contrast to structured data, text notes are stored as documents that are retrievable by patient, author, date, clinic, as well as numerous other fields. Unstructured (free) text notes are more information rich than either structured data or templated notes since their narrative format more closely parallels providers’ cognitive processes.^1,7 The value of the narrative becomes even more critical in understanding complex clinical scenarios with multiple interacting disease processes. Narratives emphasize important details, reducing cognitive overload by reducing the salience of detail the author deems to be less critical. Narrative notes simultaneously assure availability through the use of unstandardized language, often including specialty and disease-specific abbreviations.¹ Information needed for decision making in the illustrative case in this report was present only in HCP-entered free-text notes, as the structured data from which the free text was derived were not available.

Search

The introduction of search engines can be considered one of the major technologic disruptors of the 21st century.⁸ However, this advance has not yet made significant inroads into health care, despite advances in other domains. As of 2019, EHR users are still required to be familiar with the system’s data and menu structure in order to find needed information (or enter orders, code visits, or any of a number of tasks). Anecdotally, one of the authors (David Eibling) observed that the most common question from his trainees is “How do you . . .?” referring not to the care of the patient but rather to interaction with the EHR.

What is needed is a simple query-based application that finds the data on request. In addition to Voogle, other advances are being made in this arena such as the EMERSE, medical record search engine (project-emerse.org). Voogle was released to VA providers in 2017 and is available through the Internet Explorer browser on VA computers with VA intranet access. The goal of Voogle is to reduce HCP cognitive load by reducing the time and effort needed to seek relevant information for the care of a specific patient.

Natural Language Processing

Linguistic analysis of text seeking to understand its meaning constitutes a rapidly expanding field, with current heavy emphasis on the role of artificial intelligence and machine learning.¹ Advances in processing both structured data and free-text notes in the health care domain is in its infancy, despite the investment of considerable resources. Undoubtedly, advances in this arena will dramatically change provider cognitive work in the next decades.

VistA is coded in MUMPS (Massachusetts General Hospital Utility Multi-Programming System, also known as M), which has been in use for more than 50 years. Voogle employs iKnow, a novel natural language processing (NLP) application that resides in Caché (Intersystems, Boston, MA), the vendor-supported MUMPS infrastructure VistA uses to perform text analysis. iKnow does not attempt to interpret the meaning of text as do other common NLP applications, but instead relies on the expert user to interpret the meaning of the analyzed text. iKnow initially divides sentences into relations (usually verbs) and concepts, and then generates an index of these entities. The efficiency of iKnow results in very rapid indexing—often several thousand notes (not an uncommon number) can be indexed in 20 to 30 seconds. iKnow responds to a user query by searching for specific terms or similar terms within the indexed text, and then displays these terms within the original source documents, similar to well-known commercial search engines. Structured data are indexed by the iFind program simultaneously with free-text indexing (Figure 1).

Security

Maintaining high levels of security of Health Insurance Portability and Accountability (HIPAA)-compliant information in an online application such as Voogle is critical to ensure trust of veterans and HCPs. All patient data accessed by Voogle reside within the secure firewall-protected VINCI environment. All moving information is protected with high-level encryption protocols (transport layer security [TLS]), and data at rest are also encrypted. As the application is online, no data are stored on the accessing device. Voogle uses a secure Microsoft Windows logon using VA Active Directory coupled with VistA authorization to regulate who can see the data and use the application. All access is audited, not only for “sensitive patients,” but also for specific data types. Users are reminded of this Voogle attribute on the home screen.

Accessing Voogle

Voogle is available on the VA intranet to all authorized users at https://voogle.vha.med.va.gov/voogle. To assure high-level security the application can only be accessed with the Internet Explorer browser using established user identification protocols to avoid unauthorized access or duplicative log-in tasks.

Indexing

Indexing is user-driven and is required prior to patient selection and term query. The user is prompted for a patient identifier and a date range. The CDW unique patient identifier is used for all internal processing. However, a social security number look-up table is incorporated to facilitate patient selection. The date field defaults to 3 years but can be extended to approximately the year 2000.

Queries

Entering the patient name in Lastname, Firstname (no space) format will yield a list of indexed patients. All access is audited in order to deter unauthorized queries. Data from a demonstration patient are displayed in Figures 2, 3, 4, 5,
and 6.

Structured Data Searches

Structured data categories that contain the queried term, as well as a term count, are displayed after the “Structured Data” toggle is selected (Figure 2). After the desired category (Figure 2: “Outpatient Rx”) is selected, Voogle accesses the data file and displays it as a grid (medication list, Figure 3). Filter and sort functions enable display of specific medications, drug classes, or date ranges (Figure 4).

Display of Terms Within Text Notes

Selecting a term from the drop-down list (Figure 5) opens a grid with the term highlighted in a snippet of text (Figure 6). Opening the document displays the context of the term, along with negation terms (ie, not, denies, no, etc) in red font if present. Voogle, unlike other NLP tools that attempt to interpret medical notes, relies on interpretation by the HCP user. Duplicate note fragments will be displayed in multiple notes, often across multiple screens, vividly demonstrating the pervasive use of the copy-and-paste text-entry strategy. Voogle satisfies 2 of the 4 recommendations of the recent report on copy-and-paste by Tsou and colleagues.⁹ The Voogle text display grid identifies copy-and-pasted text as well as establishes the provenance of the text (by sorting on the date column). Text can be copied from Voogle into an active Computerized Patient Record System (CPRS) note if needed for active patient care. Reindexing the following day and then repeating the search will demonstrate the newly copied-and-pasted text appended to the sequence.

Limitations

Voogle is unable to access all VA patient data currently. There are a dozen or so clinical domains that are indexed by Voogle that include prescriptions, problem lists, health factors, and others. More domains can be added with minimal effort and would then be available for fast search. The most critical deficiency is its inability to access, index, or query text reports stored as images within VistA Imaging. This includes nearly all reports from outside HCPs, emergency department visits or discharge summaries from unlinked hospitals, anesthesia reports, intensive care unit flow sheets, electrocardiograms, as well as numerous other text reports such as pulmonary function reports or sleep studies. Information that is transcribed by the provider into VistA as text (as in the case presented) is available within the CDW and can be found and displayed by Voogle search.

Voogle requires that the user initiates the indexing process prior to initiating the search process. Although Voogle defaults to 3 years prior to the current date, the user can specify a start date extending to close to the year 2000. The volume of data flowing into the CDW precludes automatic indexing of all patient data, as well as automatic updating of previously indexed data. We have explored the feasibility of queueing scheduled appointments for the following day, and although the strategy shows some promise, avoiding conflict with user-requested on-demand indexing remains challenging.

The current VA network architecture updates the CDW every night, resulting in up to a 24-hour delay in data availability. However, this delay should be reduced to several minutes after implementation of real-time data feeds accompanying the coming transition to a new EHR platform.

Conclusions

The recent introduction of the Joint Legacy Viewer (JLV) to the VA EHR desktop has enhanced the breadth of patient-specific information available to any VHA clinician, with recent enhancements providing access to some community care notes from outside HCPs. Voogle builds on this capability by enabling rapid search of text notes and structured data from multiple VA sites, over an extended time frame, and perhaps entered by hundreds of authors, as demonstrated in the case example. Formal usability and workload studies have not been performed; however, anecdotal reports indicate the application dramatically reduces the time required to search for critical information needed for care of complex patients who have been treated in multiple different VA hospitals and clinics.

The Voogle paradigm of leveraging patient information stored within a large enterprise-wide data warehouse through NLP techniques may be applicable to other systems as well, and warrants exploration. We believe that replacing traditional data search paradigms that require knowledge of data structure with a true query-based paradigm is a potential game changer for health information systems. Ultimately this strategy may help provide an antidote for the information chaos impacting HCP cognition. Moreover, reducing HCP cognitive load and time on task may lessen overall health care costs, reduce provider burn-out, and improve the quality of care received by patients.

Near real-time data feeds and adding additional clinical domains will potentially provide other benefits to patient care. For example, the authors plan to investigate whether sampling incoming data may assist with behind-the-scenes continuous monitoring of indicators of patient status to facilitate early warning of impending physiologic collapse.¹⁰ Other possible applications could include real-time scans for biosurveillance or other population screening requirements.

Acknowledgments
The authors express their sincere appreciation to Leslie DeYoung for documentation and Justin Wilson who constructed much of the graphical user interface for the Voogle application and design. Without their expertise, passion, and commitment the application would not be available as it is now.

A 50-year-old man with acute myeloid leukemia (AML) with a complex karyotype was admitted to the hospital with several days of dull, left-sided abdominal pain. His most recent bone marrow biopsy showed 30% blasts, and immunophenotyping was suggestive of persistent AML (CD13+, CD34+, CD117+, CD33+, CD7+, MPO–). He was on treatment with venetoclax and cytarabine after induction therapy had failed.

On admission, his heart rate was 101 beats per minute and his blood pressure was 122/85 mm Hg. Abdominal examination revealed mild distention, hepatomegaly, and previously known massive splenomegaly, with the splenic tip extending to the umbilicus, and mild tenderness.

Results of laboratory testing revealed persistent pancytopenia:

• Hemoglobin level 6.8 g/dL (reference range 13.0–17.0)
• Total white blood cell count 0.8 × 10⁹/L (4.5–11.0)
• Platelet count 8 × 10⁹/L (150–400).

The next day, he developed severe, acute-onset left-sided abdominal pain. A check of vital signs showed worsening sinus tachycardia at 132 beats per minute and a drop in blood pressure to 90/56 mm Hg. He had worsening diffuse abdominal tenderness with sluggish bowel sounds. His hemoglobin concentration was 6.4 g/dL and platelet count 12 × 10⁹/L.

He received supportive transfusions of blood products. Surgical exploration was deemed risky, given his overall condition and severe thrombocytopenia. Splenic angiography showed no evidence of pseudoaneurysm or focal contrast extravasation. He underwent empiric embolization of the midsplenic artery, after which his hemodynamic status stabilized. He died 4 weeks later of acute respiratory failure from pneumonia.

SPLENIC RUPTURE IN AML

Atraumatic splenic rupture is rare but potentially life-threatening, especially if the diagnosis is delayed. Conditions that can cause splenomegaly and predispose to rupture include infection (infectious mononucleosis, malaria), malignant hematologic disorders (leukemia, lymphoma), other neoplasms, and amyloidosis.¹

The literature includes a few reports of splenic rupture in patients with AML.^2–4 The proposed mechanisms include bleeding from infarction sites or tumor foci, dysregulated hemostasis, and leukostasis.

The classic presentation of splenic rupture is acute-onset left-sided abdominal pain associated with hypotension and decreasing hemoglobin levels. CT of the abdomen is confirmatory, and resuscitation with crystalloids and blood products is a vital initial step in management. Choice of treatment depends on the patient’s surgical risk and hemodynamic status; options include conservative medical management, splenic artery embolization, and exploratory laparotomy.

In patients with AML and splenomegaly presenting with acute abdominal pain, clinicians need to be aware of this potential hematologic emergency.

A 50-year-old man with acute myeloid leukemia (AML) with a complex karyotype was admitted to the hospital with several days of dull, left-sided abdominal pain. His most recent bone marrow biopsy showed 30% blasts, and immunophenotyping was suggestive of persistent AML (CD13+, CD34+, CD117+, CD33+, CD7+, MPO–). He was on treatment with venetoclax and cytarabine after induction therapy had failed.

On admission, his heart rate was 101 beats per minute and his blood pressure was 122/85 mm Hg. Abdominal examination revealed mild distention, hepatomegaly, and previously known massive splenomegaly, with the splenic tip extending to the umbilicus, and mild tenderness.

Results of laboratory testing revealed persistent pancytopenia:

• Hemoglobin level 6.8 g/dL (reference range 13.0–17.0)
• Total white blood cell count 0.8 × 10⁹/L (4.5–11.0)
• Platelet count 8 × 10⁹/L (150–400).

The next day, he developed severe, acute-onset left-sided abdominal pain. A check of vital signs showed worsening sinus tachycardia at 132 beats per minute and a drop in blood pressure to 90/56 mm Hg. He had worsening diffuse abdominal tenderness with sluggish bowel sounds. His hemoglobin concentration was 6.4 g/dL and platelet count 12 × 10⁹/L.

He received supportive transfusions of blood products. Surgical exploration was deemed risky, given his overall condition and severe thrombocytopenia. Splenic angiography showed no evidence of pseudoaneurysm or focal contrast extravasation. He underwent empiric embolization of the midsplenic artery, after which his hemodynamic status stabilized. He died 4 weeks later of acute respiratory failure from pneumonia.

SPLENIC RUPTURE IN AML

Atraumatic splenic rupture is rare but potentially life-threatening, especially if the diagnosis is delayed. Conditions that can cause splenomegaly and predispose to rupture include infection (infectious mononucleosis, malaria), malignant hematologic disorders (leukemia, lymphoma), other neoplasms, and amyloidosis.¹

The literature includes a few reports of splenic rupture in patients with AML.^2–4 The proposed mechanisms include bleeding from infarction sites or tumor foci, dysregulated hemostasis, and leukostasis.

The classic presentation of splenic rupture is acute-onset left-sided abdominal pain associated with hypotension and decreasing hemoglobin levels. CT of the abdomen is confirmatory, and resuscitation with crystalloids and blood products is a vital initial step in management. Choice of treatment depends on the patient’s surgical risk and hemodynamic status; options include conservative medical management, splenic artery embolization, and exploratory laparotomy.

In patients with AML and splenomegaly presenting with acute abdominal pain, clinicians need to be aware of this potential hematologic emergency.

A 50-year-old man with acute myeloid leukemia (AML) with a complex karyotype was admitted to the hospital with several days of dull, left-sided abdominal pain. His most recent bone marrow biopsy showed 30% blasts, and immunophenotyping was suggestive of persistent AML (CD13+, CD34+, CD117+, CD33+, CD7+, MPO–). He was on treatment with venetoclax and cytarabine after induction therapy had failed.

On admission, his heart rate was 101 beats per minute and his blood pressure was 122/85 mm Hg. Abdominal examination revealed mild distention, hepatomegaly, and previously known massive splenomegaly, with the splenic tip extending to the umbilicus, and mild tenderness.

Results of laboratory testing revealed persistent pancytopenia:

• Hemoglobin level 6.8 g/dL (reference range 13.0–17.0)
• Total white blood cell count 0.8 × 10⁹/L (4.5–11.0)
• Platelet count 8 × 10⁹/L (150–400).

The next day, he developed severe, acute-onset left-sided abdominal pain. A check of vital signs showed worsening sinus tachycardia at 132 beats per minute and a drop in blood pressure to 90/56 mm Hg. He had worsening diffuse abdominal tenderness with sluggish bowel sounds. His hemoglobin concentration was 6.4 g/dL and platelet count 12 × 10⁹/L.

He received supportive transfusions of blood products. Surgical exploration was deemed risky, given his overall condition and severe thrombocytopenia. Splenic angiography showed no evidence of pseudoaneurysm or focal contrast extravasation. He underwent empiric embolization of the midsplenic artery, after which his hemodynamic status stabilized. He died 4 weeks later of acute respiratory failure from pneumonia.

SPLENIC RUPTURE IN AML

Atraumatic splenic rupture is rare but potentially life-threatening, especially if the diagnosis is delayed. Conditions that can cause splenomegaly and predispose to rupture include infection (infectious mononucleosis, malaria), malignant hematologic disorders (leukemia, lymphoma), other neoplasms, and amyloidosis.¹

The literature includes a few reports of splenic rupture in patients with AML.^2–4 The proposed mechanisms include bleeding from infarction sites or tumor foci, dysregulated hemostasis, and leukostasis.

The classic presentation of splenic rupture is acute-onset left-sided abdominal pain associated with hypotension and decreasing hemoglobin levels. CT of the abdomen is confirmatory, and resuscitation with crystalloids and blood products is a vital initial step in management. Choice of treatment depends on the patient’s surgical risk and hemodynamic status; options include conservative medical management, splenic artery embolization, and exploratory laparotomy.

In patients with AML and splenomegaly presenting with acute abdominal pain, clinicians need to be aware of this potential hematologic emergency.

A morbidly obese 54-year-old woman presented to the emergency department after experiencing generalized abdominal pain for 3 days. She rated the pain as 5 on a scale of 10 and described it as dull, cramping, waxing and waning, not radiating, and not relieved with changes of position—in fact, not alleviated by anything she had tried. Her pain was associated with nausea and 1 episode of vomiting. She also experienced constipation before the onset of pain.

She denied recent trauma, recent travel, diarrhea, fevers, weakness, shortness of breath, chest pain, other muscle pains, or recent changes in diet. She also denied having this pain in the past. She said she had unintentionally lost some weight but was not certain how much. She denied tobacco, alcohol, or illicit drug use. She had no history of surgery.

Her medical history included hypertension, anemia, and uterine fibroids. Her current medications included losartan, hydrochlorothiazide, and albuterol. She had no family history of significant disease.

INITIAL EVALUATION AND MANAGEMENT

On admission, her temperature was 97.8°F (36.6°C), heart rate 100 beats per minute, blood pressure 136/64 mm Hg, respiratory rate 18 breaths per minute, oxygen saturation 97% on room air, weight 130.6 kg, and body mass index 35 kg/m².

She was alert and oriented to person, place, and time. She was in mild discomfort but no distress. Her lungs were clear to auscultation, with no wheezing or crackles. Heart rate and rhythm were regular, with no extra heart sounds or murmurs. Bowel sounds were normal in all 4 quadrants, with tenderness to palpation of the epigastric area, but with no guarding or rebound tenderness.

Laboratory test results

Notable results of blood testing at presentation were as follows:

• Hemoglobin 8.2 g/dL (reference range 12.3–15.3)
• Hematocrit 26% (41–50)
• Mean corpuscular volume 107 fL (80–100)
• Blood urea nitrogen 33 mg/dL (8–21); 6 months earlier it was 16
• Serum creatinine 3.6 mg/dL (0.58–0.96); 6 months earlier, it was 0.75
• Albumin 3.3 g/dL (3.5–5)
• Calcium 18.4 mg/dL (8.4–10.2); 6 months earlier, it was 9.6
• Corrected calcium 19 mg/dL.

Findings on imaging, electrocardiography

Chest radiography showed no acute cardiopulmonary abnormalities. Abdominal computed tomography without contrast showed no abnormalities within the pancreas and no evidence of inflammation or obstruction. Electrocardiography showed sinus tachycardia.

DIFFERENTIAL DIAGNOSIS

1. Which is the most likely cause of this patient’s symptoms?

• Primary hyperparathyroidism
• Malignancy
• Her drug therapy
• Familial hypercalcemic hypocalciuria

In total, her laboratory results were consistent with macrocytic anemia, severe hypercalcemia, and acute kidney injury, and she had generalized symptoms.

Primary hyperparathyroidism

A main cause of hypercalcemia is primary hyperparathyroidism, and this needs to be ruled out. Benign adenomas are the most common cause of primary hyperparathyroidism, and a risk factor for benign adenoma is exposure to therapeutic levels of radiation.³

In hyperparathyroidism, there is an increased secretion of parathyroid hormone (PTH), which has multiple effects including increased reabsorption of calcium from the urine, increased excretion of phosphate, and increased expression of 1,25-hydroxyvitamin D hydroxylase to activate vitamin D. PTH also stimulates osteoclasts to increase their expression of receptor activator of nuclear factor kappa B ligand (RANKL), which has a downstream effect on osteoclast precursors to cause bone reabsorption.³

Inherited primary hyperparathyroidism tends to present at a younger age, with multiple overactive parathyroid glands.³ Given our patient’s age, inherited primary hyparathyroidism is thus less likely.

The probability that malignancy is causing the hypercalcemia increases with calcium levels greater than 13 mg/dL. Epidemiologically, in hospitalized patients with hypercalcemia, the source tends to be malignancy.⁴ Typically, patients who develop hypercalcemia from malignancy have a worse prognosis.⁵

Solid tumors and leukemias can cause hypercalcemia. The mechanisms include humoral factors secreted by the malignancy, local osteolysis due to tumor invasion of bone, and excessive absorption of calcium due to excess vitamin D produced by malignancies.⁵ The cancers that most frequently cause an increase in calcium resorption are lung cancer, renal cancer, breast cancer, and multiple myeloma.¹

Solid tumors with no bone metastasis and non-Hodgkin lymphoma that release PTH-related protein (PTHrP) cause humoral hypercalcemia in malignancy. The patient is typically in an advanced stage of disease. PTHrP increases serum calcium levels by decreasing the kidney’s ability to excrete calcium and by increasing bone turnover. It has no effect on intestinal absorption because of its inability to stimulate activated vitamin D₃. Thus, the increase in systemic calcium comes directly from breakdown of bone and inability to excrete the excess.

PTHrP has a unique role in breast cancer: it is released locally in areas where cancer cells have metastasized to bone, but it does not cause a systemic effect. Bone resorption occurs in areas of metastasis and results from an increase in expression of RANKL and RANK in osteoclasts in response to the effects of PTHrP, leading to an increase in the production of osteoclastic cells.¹

Tamoxifen, an endocrine therapy often used in breast cancer, also causes a release of bone-reabsorbing factors from tumor cells, which can partially contribute to hypercal­cemia.⁵

Myeloma cells secrete RANKL, which stimulates osteoclastic activity, and they also  release interleukin 6 (IL-6) and activating macrophage inflammatory protein alpha. Serum testing usually shows low or normal intact PTH, PTHrP, and 1,25-dihydroxyvitamin D.¹

Patients with multiple myeloma have a worse prognosis if they have a high red blood cell distribution width, a condition shown to correlate with malnutrition, leading to deficiencies in vitamin B₁₂ and to poor response to treatment.⁶ Up to 14% of patients with multiple myeloma have vitamin B₁₂ deficiency.⁷

Our patient’s recent weight loss and severe hypercalcemia raise suspicion of malignancy. Further, her obesity makes proper routine breast examination difficult and thus increases the chance of undiagnosed breast cancer.⁸ Her decrease in renal function and her anemia complicated by hypercalcemia also raise suspicion of multiple myeloma.

Hypercalcemia due to drug therapy

Thiazide diuretics, lithium, teriparatide, and vitamin A in excessive amounts can raise the serum calcium concentration.⁵ Our patient was taking a thiazide for hypertension, but her extremely high calcium level places drug-induced hypercalcemia as the sole cause lower on the differential list.

Familial hypercalcemic hypocalciuria

Familial hypercalcemic hypocalciuria is a rare autosomal-dominant cause of hypercalcemia in which the ability of the body (and especially the kidneys) to sense levels of calcium is impaired, leading to a decrease in excretion of calcium in the urine.³ Very high calcium levels are rare in hypercalcemic hypocalciuria.³ In our patient with a corrected calcium concentration of nearly 19 mg/dL, familial hypercalcemic hypocalciuria is very unlikely to be the cause of the hypercalcemia.

WHAT ARE THE NEXT STEPS IN THE WORKUP?

As hypercalcemia has been confirmed, the intact PTH level should be checked to determine whether the patient’s condition is PTH-mediated. If the PTH level is in the upper range of normal or is minimally elevated, primary hyperparathyroidism is likely. Elevated PTH confirms primary hyperparathyroidism. A low-normal or low intact PTH confirms a non-PTH-mediated process, and once this is confirmed, PTHrP levels should be checked. An elevated PTHrP suggests humoral hypercalcemia of malignancy. Serum protein electrophoresis, urine protein electrophoresis, and a serum light chain assay should be performed to rule out multiple myeloma.

Vitamin D toxicity is associated with high concentrations of 1,25-dihydroxyvitamin D and 25-hydroxyvitamin D metabolites. These levels should be checked in this patient.

Other disorders that cause hypercalcemia are vitamin A toxicity and hyperthyroidism, so vitamin A and thyroid-stimulating hormone levels should also be checked.⁵

CASE CONTINUED

After further questioning, the patient said that she had had lower back pain about 1 to 2 weeks before coming to the emergency room; her primary care doctor had said the pain was likely from muscle strain. The pain had almost resolved but was still present.

The results of further laboratory testing were as follows:

• Serum PTH 11 pg/mL (15–65)
• PTHrP 3.4 pmol/L (< 2.0)
• Protein electrophoresis showed a monoclonal (M) spike of 0.2 g/dL (0)
• Activated vitamin D < 5 ng/mL (19.9–79.3)
• Vitamin A 7.2 mg/dL (33.1–100)
• Vitamin B₁₂ 194 pg/mL (239–931)
• Thyroid-stimulating hormone 1.21 mIU/ L (0.47–4.68
• Free thyroxine 1.27 ng/dL (0.78–2.19)
• Iron 103 µg/dL (37–170)
• Total iron-binding capacity 335 µg/dL (265–497)
• Transferrin 248 mg/dL (206–381)
• Ferritin 66 ng/mL (11.1–264)
• Urine protein (random) 100 mg/dL (0–20)
• Urine microalbumin (random) 5.9 mg/dL (0–1.6)
• Urine creatinine clearance 88.5 mL/min (88–128)
• Urine albumin-creatinine ratio 66.66 mg/g (< 30).

A nuclear bone scan showed increased bone uptake in the hip and both shoulders, consistent with arthritis, and increased activity in 2 of the lower left ribs, associated with rib fractures secondary to lytic lesions. A skeletal survey at a later date showed multiple well-circumscribed “punched-out” lytic lesions in both forearms and both femurs.

2. What should be the next step in this patient’s management?

• Intravenous (IV) fluids
• Calcitonin
• Bisphosphonate treatment
• Denosumab
• Hemodialysis

Initial treatment of severe hypercalcemia includes the following:

Start IV isotonic fluids at a rate of 150 mL/h (if the patient is making urine) to maintain urine output at more than 100 mL/h. Closely monitor urine output.

Give calcitonin 4 IU/kg in combination with IV fluids to reduce calcium levels within the first 12 to 48 hours of treatment.

Give a bisphosphonate, eg, zoledronic acid 4 mg over 15 minutes, or pamidronate 60 to 90 mg over 2 hours. Zoledronic acid is preferred in malignancy-induced hypercal­cemia because it is more potent. Doses should be adjusted in patients with renal failure.

Give denosumab if hypercalcemia is refractory to bisphosphonates, or when bisphosphonates cannot be used in renal failure.⁹

Hemodialysis is performed in patients who have significant neurologic symptoms irrespective of acute renal insufficiency.

Our patient was started on 0.9% sodium chloride at a rate of 150 mL/h for severe hypercalcemia. Zoledronic acid 4 mg IV was given once. These measures lowered her calcium level and lessened her acute kidney injury.

ADDITIONAL FINDINGS

Urine testing was positive for Bence Jones protein. Immune electrophoresis, performed because of suspicion of multiple myeloma, showed an elevated level of kappa light chains at 806.7 mg/dL (0.33–1.94) and normal lambda light chains at 0.62 mg/dL (0.57–2.63). The immunoglobulin G level was low at 496 mg/dL (610–1,660). In patients with severe hypercalcemia, these results point to a diagnosis of malignancy. Bone marrow aspiration study showed greater than 10% plasma cells, confirming multiple myeloma.

MULTIPLE MYELOMA

The diagnosis of multiple myeloma is based in part on the presence of 10% or more of clonal bone marrow plasma cells¹⁰ and of specific end-organ damage (anemia, hypercalcemia, renal insufficiency, or bone lesions).⁹

Bone marrow clonality can be shown by the ratio of kappa to lambda light chains as  detected with immunohistochemistry, immunofluorescence, or flow cytometry.11 The normal ratio is 0.26 to 1.65 for a patient with normal kidney function. In this patient, however, the ratio was 1,301.08 (806.67 kappa to 0.62 lambda), which was extremely out of range. The patient’s bone marrow biopsy results revealed the presence of 15% clonal bone marrow plasma cells.

Multiple myeloma causes osteolytic lesions through increased activation of osteoclast activating factor that stimulates the growth of osteoclast precursors. At the same time, it inhibits osteoblast formation via multiple pathways, including the action of sclerostin.¹¹ Our patient had lytic lesions in 2 left lower ribs and in both forearms and femurs.

Hypercalcemia in multiple myeloma is attributed to 2 main factors: bone breakdown and macrophage overactivation. Multiple myeloma cells increase the release of macrophage inflammatory protein 1-alpha and tumor necrosis factor, which are inflammatory proteins that cause an increase in macrophages, which cause an increase in calcitriol.¹¹ As noted, our patient’s calcium level at presentation was 18.4 mg/dL uncorrected and 18.96 mg/dL corrected.

Cast nephropathy can occur in the distal tubules from the increased free light chains circulating and combining with Tamm-Horsfall protein, which in turn causes obstruction and local inflammation,¹² leading to a rise in creatinine levels and resulting in acute kidney injury,¹² as in our patient.

TREATMENT CONSIDERATIONS IN MULTIPLE MYELOMA

Our patient was referred to an oncologist for management.

In the management of multiple myeloma, the patient’s quality of life needs to be considered. With the development of new agents to combat the damages of the osteolytic effects, there is hope for improving quality of life.^13,14 New agents under study include anabolic agents such as antisclerostin and anti-Dickkopf-1, which promote osteoblastogenesis, leading to bone formation, with the possibility of repairing existing damage.¹⁵

TAKE-HOME POINTS

• If hypercalcemia is mild to moderate, consider primary hyperparathyroidism.
• Identify patients with severe symptoms of hypercalcemia such as volume depletion, acute kidney injury, arrhythmia, or seizures.
• Confirm severe cases of hypercalcemia and treat severe cases effectively.
• Severe hypercalcemia may need further investigation into a potential underlying malignancy.

A morbidly obese 54-year-old woman presented to the emergency department after experiencing generalized abdominal pain for 3 days. She rated the pain as 5 on a scale of 10 and described it as dull, cramping, waxing and waning, not radiating, and not relieved with changes of position—in fact, not alleviated by anything she had tried. Her pain was associated with nausea and 1 episode of vomiting. She also experienced constipation before the onset of pain.

She denied recent trauma, recent travel, diarrhea, fevers, weakness, shortness of breath, chest pain, other muscle pains, or recent changes in diet. She also denied having this pain in the past. She said she had unintentionally lost some weight but was not certain how much. She denied tobacco, alcohol, or illicit drug use. She had no history of surgery.

Her medical history included hypertension, anemia, and uterine fibroids. Her current medications included losartan, hydrochlorothiazide, and albuterol. She had no family history of significant disease.

INITIAL EVALUATION AND MANAGEMENT

On admission, her temperature was 97.8°F (36.6°C), heart rate 100 beats per minute, blood pressure 136/64 mm Hg, respiratory rate 18 breaths per minute, oxygen saturation 97% on room air, weight 130.6 kg, and body mass index 35 kg/m².

She was alert and oriented to person, place, and time. She was in mild discomfort but no distress. Her lungs were clear to auscultation, with no wheezing or crackles. Heart rate and rhythm were regular, with no extra heart sounds or murmurs. Bowel sounds were normal in all 4 quadrants, with tenderness to palpation of the epigastric area, but with no guarding or rebound tenderness.

Laboratory test results

Notable results of blood testing at presentation were as follows:

• Hemoglobin 8.2 g/dL (reference range 12.3–15.3)
• Hematocrit 26% (41–50)
• Mean corpuscular volume 107 fL (80–100)
• Blood urea nitrogen 33 mg/dL (8–21); 6 months earlier it was 16
• Serum creatinine 3.6 mg/dL (0.58–0.96); 6 months earlier, it was 0.75
• Albumin 3.3 g/dL (3.5–5)
• Calcium 18.4 mg/dL (8.4–10.2); 6 months earlier, it was 9.6
• Corrected calcium 19 mg/dL.

Findings on imaging, electrocardiography

Chest radiography showed no acute cardiopulmonary abnormalities. Abdominal computed tomography without contrast showed no abnormalities within the pancreas and no evidence of inflammation or obstruction. Electrocardiography showed sinus tachycardia.

DIFFERENTIAL DIAGNOSIS

1. Which is the most likely cause of this patient’s symptoms?

• Primary hyperparathyroidism
• Malignancy
• Her drug therapy
• Familial hypercalcemic hypocalciuria

In total, her laboratory results were consistent with macrocytic anemia, severe hypercalcemia, and acute kidney injury, and she had generalized symptoms.

Primary hyperparathyroidism

A main cause of hypercalcemia is primary hyperparathyroidism, and this needs to be ruled out. Benign adenomas are the most common cause of primary hyperparathyroidism, and a risk factor for benign adenoma is exposure to therapeutic levels of radiation.³

In hyperparathyroidism, there is an increased secretion of parathyroid hormone (PTH), which has multiple effects including increased reabsorption of calcium from the urine, increased excretion of phosphate, and increased expression of 1,25-hydroxyvitamin D hydroxylase to activate vitamin D. PTH also stimulates osteoclasts to increase their expression of receptor activator of nuclear factor kappa B ligand (RANKL), which has a downstream effect on osteoclast precursors to cause bone reabsorption.³

Inherited primary hyperparathyroidism tends to present at a younger age, with multiple overactive parathyroid glands.³ Given our patient’s age, inherited primary hyparathyroidism is thus less likely.

The probability that malignancy is causing the hypercalcemia increases with calcium levels greater than 13 mg/dL. Epidemiologically, in hospitalized patients with hypercalcemia, the source tends to be malignancy.⁴ Typically, patients who develop hypercalcemia from malignancy have a worse prognosis.⁵

Solid tumors and leukemias can cause hypercalcemia. The mechanisms include humoral factors secreted by the malignancy, local osteolysis due to tumor invasion of bone, and excessive absorption of calcium due to excess vitamin D produced by malignancies.⁵ The cancers that most frequently cause an increase in calcium resorption are lung cancer, renal cancer, breast cancer, and multiple myeloma.¹

Solid tumors with no bone metastasis and non-Hodgkin lymphoma that release PTH-related protein (PTHrP) cause humoral hypercalcemia in malignancy. The patient is typically in an advanced stage of disease. PTHrP increases serum calcium levels by decreasing the kidney’s ability to excrete calcium and by increasing bone turnover. It has no effect on intestinal absorption because of its inability to stimulate activated vitamin D₃. Thus, the increase in systemic calcium comes directly from breakdown of bone and inability to excrete the excess.

PTHrP has a unique role in breast cancer: it is released locally in areas where cancer cells have metastasized to bone, but it does not cause a systemic effect. Bone resorption occurs in areas of metastasis and results from an increase in expression of RANKL and RANK in osteoclasts in response to the effects of PTHrP, leading to an increase in the production of osteoclastic cells.¹

Tamoxifen, an endocrine therapy often used in breast cancer, also causes a release of bone-reabsorbing factors from tumor cells, which can partially contribute to hypercal­cemia.⁵

Myeloma cells secrete RANKL, which stimulates osteoclastic activity, and they also  release interleukin 6 (IL-6) and activating macrophage inflammatory protein alpha. Serum testing usually shows low or normal intact PTH, PTHrP, and 1,25-dihydroxyvitamin D.¹

Patients with multiple myeloma have a worse prognosis if they have a high red blood cell distribution width, a condition shown to correlate with malnutrition, leading to deficiencies in vitamin B₁₂ and to poor response to treatment.⁶ Up to 14% of patients with multiple myeloma have vitamin B₁₂ deficiency.⁷

Our patient’s recent weight loss and severe hypercalcemia raise suspicion of malignancy. Further, her obesity makes proper routine breast examination difficult and thus increases the chance of undiagnosed breast cancer.⁸ Her decrease in renal function and her anemia complicated by hypercalcemia also raise suspicion of multiple myeloma.

Hypercalcemia due to drug therapy

Thiazide diuretics, lithium, teriparatide, and vitamin A in excessive amounts can raise the serum calcium concentration.⁵ Our patient was taking a thiazide for hypertension, but her extremely high calcium level places drug-induced hypercalcemia as the sole cause lower on the differential list.

Familial hypercalcemic hypocalciuria

Familial hypercalcemic hypocalciuria is a rare autosomal-dominant cause of hypercalcemia in which the ability of the body (and especially the kidneys) to sense levels of calcium is impaired, leading to a decrease in excretion of calcium in the urine.³ Very high calcium levels are rare in hypercalcemic hypocalciuria.³ In our patient with a corrected calcium concentration of nearly 19 mg/dL, familial hypercalcemic hypocalciuria is very unlikely to be the cause of the hypercalcemia.

WHAT ARE THE NEXT STEPS IN THE WORKUP?

As hypercalcemia has been confirmed, the intact PTH level should be checked to determine whether the patient’s condition is PTH-mediated. If the PTH level is in the upper range of normal or is minimally elevated, primary hyperparathyroidism is likely. Elevated PTH confirms primary hyperparathyroidism. A low-normal or low intact PTH confirms a non-PTH-mediated process, and once this is confirmed, PTHrP levels should be checked. An elevated PTHrP suggests humoral hypercalcemia of malignancy. Serum protein electrophoresis, urine protein electrophoresis, and a serum light chain assay should be performed to rule out multiple myeloma.

Vitamin D toxicity is associated with high concentrations of 1,25-dihydroxyvitamin D and 25-hydroxyvitamin D metabolites. These levels should be checked in this patient.

Other disorders that cause hypercalcemia are vitamin A toxicity and hyperthyroidism, so vitamin A and thyroid-stimulating hormone levels should also be checked.⁵

CASE CONTINUED

After further questioning, the patient said that she had had lower back pain about 1 to 2 weeks before coming to the emergency room; her primary care doctor had said the pain was likely from muscle strain. The pain had almost resolved but was still present.

The results of further laboratory testing were as follows:

• Serum PTH 11 pg/mL (15–65)
• PTHrP 3.4 pmol/L (< 2.0)
• Protein electrophoresis showed a monoclonal (M) spike of 0.2 g/dL (0)
• Activated vitamin D < 5 ng/mL (19.9–79.3)
• Vitamin A 7.2 mg/dL (33.1–100)
• Vitamin B₁₂ 194 pg/mL (239–931)
• Thyroid-stimulating hormone 1.21 mIU/ L (0.47–4.68
• Free thyroxine 1.27 ng/dL (0.78–2.19)
• Iron 103 µg/dL (37–170)
• Total iron-binding capacity 335 µg/dL (265–497)
• Transferrin 248 mg/dL (206–381)
• Ferritin 66 ng/mL (11.1–264)
• Urine protein (random) 100 mg/dL (0–20)
• Urine microalbumin (random) 5.9 mg/dL (0–1.6)
• Urine creatinine clearance 88.5 mL/min (88–128)
• Urine albumin-creatinine ratio 66.66 mg/g (< 30).

A nuclear bone scan showed increased bone uptake in the hip and both shoulders, consistent with arthritis, and increased activity in 2 of the lower left ribs, associated with rib fractures secondary to lytic lesions. A skeletal survey at a later date showed multiple well-circumscribed “punched-out” lytic lesions in both forearms and both femurs.

2. What should be the next step in this patient’s management?

• Intravenous (IV) fluids
• Calcitonin
• Bisphosphonate treatment
• Denosumab
• Hemodialysis

Initial treatment of severe hypercalcemia includes the following:

Start IV isotonic fluids at a rate of 150 mL/h (if the patient is making urine) to maintain urine output at more than 100 mL/h. Closely monitor urine output.

Give calcitonin 4 IU/kg in combination with IV fluids to reduce calcium levels within the first 12 to 48 hours of treatment.

Give a bisphosphonate, eg, zoledronic acid 4 mg over 15 minutes, or pamidronate 60 to 90 mg over 2 hours. Zoledronic acid is preferred in malignancy-induced hypercal­cemia because it is more potent. Doses should be adjusted in patients with renal failure.

Give denosumab if hypercalcemia is refractory to bisphosphonates, or when bisphosphonates cannot be used in renal failure.⁹

Hemodialysis is performed in patients who have significant neurologic symptoms irrespective of acute renal insufficiency.

Our patient was started on 0.9% sodium chloride at a rate of 150 mL/h for severe hypercalcemia. Zoledronic acid 4 mg IV was given once. These measures lowered her calcium level and lessened her acute kidney injury.

ADDITIONAL FINDINGS

Urine testing was positive for Bence Jones protein. Immune electrophoresis, performed because of suspicion of multiple myeloma, showed an elevated level of kappa light chains at 806.7 mg/dL (0.33–1.94) and normal lambda light chains at 0.62 mg/dL (0.57–2.63). The immunoglobulin G level was low at 496 mg/dL (610–1,660). In patients with severe hypercalcemia, these results point to a diagnosis of malignancy. Bone marrow aspiration study showed greater than 10% plasma cells, confirming multiple myeloma.

MULTIPLE MYELOMA

The diagnosis of multiple myeloma is based in part on the presence of 10% or more of clonal bone marrow plasma cells¹⁰ and of specific end-organ damage (anemia, hypercalcemia, renal insufficiency, or bone lesions).⁹

Bone marrow clonality can be shown by the ratio of kappa to lambda light chains as  detected with immunohistochemistry, immunofluorescence, or flow cytometry.11 The normal ratio is 0.26 to 1.65 for a patient with normal kidney function. In this patient, however, the ratio was 1,301.08 (806.67 kappa to 0.62 lambda), which was extremely out of range. The patient’s bone marrow biopsy results revealed the presence of 15% clonal bone marrow plasma cells.

Multiple myeloma causes osteolytic lesions through increased activation of osteoclast activating factor that stimulates the growth of osteoclast precursors. At the same time, it inhibits osteoblast formation via multiple pathways, including the action of sclerostin.¹¹ Our patient had lytic lesions in 2 left lower ribs and in both forearms and femurs.

Hypercalcemia in multiple myeloma is attributed to 2 main factors: bone breakdown and macrophage overactivation. Multiple myeloma cells increase the release of macrophage inflammatory protein 1-alpha and tumor necrosis factor, which are inflammatory proteins that cause an increase in macrophages, which cause an increase in calcitriol.¹¹ As noted, our patient’s calcium level at presentation was 18.4 mg/dL uncorrected and 18.96 mg/dL corrected.

Cast nephropathy can occur in the distal tubules from the increased free light chains circulating and combining with Tamm-Horsfall protein, which in turn causes obstruction and local inflammation,¹² leading to a rise in creatinine levels and resulting in acute kidney injury,¹² as in our patient.

TREATMENT CONSIDERATIONS IN MULTIPLE MYELOMA

Our patient was referred to an oncologist for management.

In the management of multiple myeloma, the patient’s quality of life needs to be considered. With the development of new agents to combat the damages of the osteolytic effects, there is hope for improving quality of life.^13,14 New agents under study include anabolic agents such as antisclerostin and anti-Dickkopf-1, which promote osteoblastogenesis, leading to bone formation, with the possibility of repairing existing damage.¹⁵

TAKE-HOME POINTS

• If hypercalcemia is mild to moderate, consider primary hyperparathyroidism.
• Identify patients with severe symptoms of hypercalcemia such as volume depletion, acute kidney injury, arrhythmia, or seizures.
• Confirm severe cases of hypercalcemia and treat severe cases effectively.
• Severe hypercalcemia may need further investigation into a potential underlying malignancy.

A morbidly obese 54-year-old woman presented to the emergency department after experiencing generalized abdominal pain for 3 days. She rated the pain as 5 on a scale of 10 and described it as dull, cramping, waxing and waning, not radiating, and not relieved with changes of position—in fact, not alleviated by anything she had tried. Her pain was associated with nausea and 1 episode of vomiting. She also experienced constipation before the onset of pain.

She denied recent trauma, recent travel, diarrhea, fevers, weakness, shortness of breath, chest pain, other muscle pains, or recent changes in diet. She also denied having this pain in the past. She said she had unintentionally lost some weight but was not certain how much. She denied tobacco, alcohol, or illicit drug use. She had no history of surgery.

Her medical history included hypertension, anemia, and uterine fibroids. Her current medications included losartan, hydrochlorothiazide, and albuterol. She had no family history of significant disease.

INITIAL EVALUATION AND MANAGEMENT

On admission, her temperature was 97.8°F (36.6°C), heart rate 100 beats per minute, blood pressure 136/64 mm Hg, respiratory rate 18 breaths per minute, oxygen saturation 97% on room air, weight 130.6 kg, and body mass index 35 kg/m².

She was alert and oriented to person, place, and time. She was in mild discomfort but no distress. Her lungs were clear to auscultation, with no wheezing or crackles. Heart rate and rhythm were regular, with no extra heart sounds or murmurs. Bowel sounds were normal in all 4 quadrants, with tenderness to palpation of the epigastric area, but with no guarding or rebound tenderness.

Laboratory test results

Notable results of blood testing at presentation were as follows:

• Hemoglobin 8.2 g/dL (reference range 12.3–15.3)
• Hematocrit 26% (41–50)
• Mean corpuscular volume 107 fL (80–100)
• Blood urea nitrogen 33 mg/dL (8–21); 6 months earlier it was 16
• Serum creatinine 3.6 mg/dL (0.58–0.96); 6 months earlier, it was 0.75
• Albumin 3.3 g/dL (3.5–5)
• Calcium 18.4 mg/dL (8.4–10.2); 6 months earlier, it was 9.6
• Corrected calcium 19 mg/dL.

Findings on imaging, electrocardiography

Chest radiography showed no acute cardiopulmonary abnormalities. Abdominal computed tomography without contrast showed no abnormalities within the pancreas and no evidence of inflammation or obstruction. Electrocardiography showed sinus tachycardia.

DIFFERENTIAL DIAGNOSIS

1. Which is the most likely cause of this patient’s symptoms?

• Primary hyperparathyroidism
• Malignancy
• Her drug therapy
• Familial hypercalcemic hypocalciuria

In total, her laboratory results were consistent with macrocytic anemia, severe hypercalcemia, and acute kidney injury, and she had generalized symptoms.

Primary hyperparathyroidism

A main cause of hypercalcemia is primary hyperparathyroidism, and this needs to be ruled out. Benign adenomas are the most common cause of primary hyperparathyroidism, and a risk factor for benign adenoma is exposure to therapeutic levels of radiation.³

In hyperparathyroidism, there is an increased secretion of parathyroid hormone (PTH), which has multiple effects including increased reabsorption of calcium from the urine, increased excretion of phosphate, and increased expression of 1,25-hydroxyvitamin D hydroxylase to activate vitamin D. PTH also stimulates osteoclasts to increase their expression of receptor activator of nuclear factor kappa B ligand (RANKL), which has a downstream effect on osteoclast precursors to cause bone reabsorption.³

Inherited primary hyperparathyroidism tends to present at a younger age, with multiple overactive parathyroid glands.³ Given our patient’s age, inherited primary hyparathyroidism is thus less likely.

The probability that malignancy is causing the hypercalcemia increases with calcium levels greater than 13 mg/dL. Epidemiologically, in hospitalized patients with hypercalcemia, the source tends to be malignancy.⁴ Typically, patients who develop hypercalcemia from malignancy have a worse prognosis.⁵

Solid tumors and leukemias can cause hypercalcemia. The mechanisms include humoral factors secreted by the malignancy, local osteolysis due to tumor invasion of bone, and excessive absorption of calcium due to excess vitamin D produced by malignancies.⁵ The cancers that most frequently cause an increase in calcium resorption are lung cancer, renal cancer, breast cancer, and multiple myeloma.¹

Solid tumors with no bone metastasis and non-Hodgkin lymphoma that release PTH-related protein (PTHrP) cause humoral hypercalcemia in malignancy. The patient is typically in an advanced stage of disease. PTHrP increases serum calcium levels by decreasing the kidney’s ability to excrete calcium and by increasing bone turnover. It has no effect on intestinal absorption because of its inability to stimulate activated vitamin D₃. Thus, the increase in systemic calcium comes directly from breakdown of bone and inability to excrete the excess.

PTHrP has a unique role in breast cancer: it is released locally in areas where cancer cells have metastasized to bone, but it does not cause a systemic effect. Bone resorption occurs in areas of metastasis and results from an increase in expression of RANKL and RANK in osteoclasts in response to the effects of PTHrP, leading to an increase in the production of osteoclastic cells.¹

Tamoxifen, an endocrine therapy often used in breast cancer, also causes a release of bone-reabsorbing factors from tumor cells, which can partially contribute to hypercal­cemia.⁵

Myeloma cells secrete RANKL, which stimulates osteoclastic activity, and they also  release interleukin 6 (IL-6) and activating macrophage inflammatory protein alpha. Serum testing usually shows low or normal intact PTH, PTHrP, and 1,25-dihydroxyvitamin D.¹

Patients with multiple myeloma have a worse prognosis if they have a high red blood cell distribution width, a condition shown to correlate with malnutrition, leading to deficiencies in vitamin B₁₂ and to poor response to treatment.⁶ Up to 14% of patients with multiple myeloma have vitamin B₁₂ deficiency.⁷

Our patient’s recent weight loss and severe hypercalcemia raise suspicion of malignancy. Further, her obesity makes proper routine breast examination difficult and thus increases the chance of undiagnosed breast cancer.⁸ Her decrease in renal function and her anemia complicated by hypercalcemia also raise suspicion of multiple myeloma.

Hypercalcemia due to drug therapy

Thiazide diuretics, lithium, teriparatide, and vitamin A in excessive amounts can raise the serum calcium concentration.⁵ Our patient was taking a thiazide for hypertension, but her extremely high calcium level places drug-induced hypercalcemia as the sole cause lower on the differential list.

Familial hypercalcemic hypocalciuria

Familial hypercalcemic hypocalciuria is a rare autosomal-dominant cause of hypercalcemia in which the ability of the body (and especially the kidneys) to sense levels of calcium is impaired, leading to a decrease in excretion of calcium in the urine.³ Very high calcium levels are rare in hypercalcemic hypocalciuria.³ In our patient with a corrected calcium concentration of nearly 19 mg/dL, familial hypercalcemic hypocalciuria is very unlikely to be the cause of the hypercalcemia.

WHAT ARE THE NEXT STEPS IN THE WORKUP?

As hypercalcemia has been confirmed, the intact PTH level should be checked to determine whether the patient’s condition is PTH-mediated. If the PTH level is in the upper range of normal or is minimally elevated, primary hyperparathyroidism is likely. Elevated PTH confirms primary hyperparathyroidism. A low-normal or low intact PTH confirms a non-PTH-mediated process, and once this is confirmed, PTHrP levels should be checked. An elevated PTHrP suggests humoral hypercalcemia of malignancy. Serum protein electrophoresis, urine protein electrophoresis, and a serum light chain assay should be performed to rule out multiple myeloma.

Vitamin D toxicity is associated with high concentrations of 1,25-dihydroxyvitamin D and 25-hydroxyvitamin D metabolites. These levels should be checked in this patient.

Other disorders that cause hypercalcemia are vitamin A toxicity and hyperthyroidism, so vitamin A and thyroid-stimulating hormone levels should also be checked.⁵

CASE CONTINUED

After further questioning, the patient said that she had had lower back pain about 1 to 2 weeks before coming to the emergency room; her primary care doctor had said the pain was likely from muscle strain. The pain had almost resolved but was still present.

The results of further laboratory testing were as follows:

• Serum PTH 11 pg/mL (15–65)
• PTHrP 3.4 pmol/L (< 2.0)
• Protein electrophoresis showed a monoclonal (M) spike of 0.2 g/dL (0)
• Activated vitamin D < 5 ng/mL (19.9–79.3)
• Vitamin A 7.2 mg/dL (33.1–100)
• Vitamin B₁₂ 194 pg/mL (239–931)
• Thyroid-stimulating hormone 1.21 mIU/ L (0.47–4.68
• Free thyroxine 1.27 ng/dL (0.78–2.19)
• Iron 103 µg/dL (37–170)
• Total iron-binding capacity 335 µg/dL (265–497)
• Transferrin 248 mg/dL (206–381)
• Ferritin 66 ng/mL (11.1–264)
• Urine protein (random) 100 mg/dL (0–20)
• Urine microalbumin (random) 5.9 mg/dL (0–1.6)
• Urine creatinine clearance 88.5 mL/min (88–128)
• Urine albumin-creatinine ratio 66.66 mg/g (< 30).

A nuclear bone scan showed increased bone uptake in the hip and both shoulders, consistent with arthritis, and increased activity in 2 of the lower left ribs, associated with rib fractures secondary to lytic lesions. A skeletal survey at a later date showed multiple well-circumscribed “punched-out” lytic lesions in both forearms and both femurs.

2. What should be the next step in this patient’s management?

• Intravenous (IV) fluids
• Calcitonin
• Bisphosphonate treatment
• Denosumab
• Hemodialysis

Initial treatment of severe hypercalcemia includes the following:

Start IV isotonic fluids at a rate of 150 mL/h (if the patient is making urine) to maintain urine output at more than 100 mL/h. Closely monitor urine output.

Give calcitonin 4 IU/kg in combination with IV fluids to reduce calcium levels within the first 12 to 48 hours of treatment.

Give a bisphosphonate, eg, zoledronic acid 4 mg over 15 minutes, or pamidronate 60 to 90 mg over 2 hours. Zoledronic acid is preferred in malignancy-induced hypercal­cemia because it is more potent. Doses should be adjusted in patients with renal failure.

Give denosumab if hypercalcemia is refractory to bisphosphonates, or when bisphosphonates cannot be used in renal failure.⁹

Hemodialysis is performed in patients who have significant neurologic symptoms irrespective of acute renal insufficiency.

Our patient was started on 0.9% sodium chloride at a rate of 150 mL/h for severe hypercalcemia. Zoledronic acid 4 mg IV was given once. These measures lowered her calcium level and lessened her acute kidney injury.

ADDITIONAL FINDINGS

Urine testing was positive for Bence Jones protein. Immune electrophoresis, performed because of suspicion of multiple myeloma, showed an elevated level of kappa light chains at 806.7 mg/dL (0.33–1.94) and normal lambda light chains at 0.62 mg/dL (0.57–2.63). The immunoglobulin G level was low at 496 mg/dL (610–1,660). In patients with severe hypercalcemia, these results point to a diagnosis of malignancy. Bone marrow aspiration study showed greater than 10% plasma cells, confirming multiple myeloma.

MULTIPLE MYELOMA

The diagnosis of multiple myeloma is based in part on the presence of 10% or more of clonal bone marrow plasma cells¹⁰ and of specific end-organ damage (anemia, hypercalcemia, renal insufficiency, or bone lesions).⁹

Bone marrow clonality can be shown by the ratio of kappa to lambda light chains as  detected with immunohistochemistry, immunofluorescence, or flow cytometry.11 The normal ratio is 0.26 to 1.65 for a patient with normal kidney function. In this patient, however, the ratio was 1,301.08 (806.67 kappa to 0.62 lambda), which was extremely out of range. The patient’s bone marrow biopsy results revealed the presence of 15% clonal bone marrow plasma cells.

Multiple myeloma causes osteolytic lesions through increased activation of osteoclast activating factor that stimulates the growth of osteoclast precursors. At the same time, it inhibits osteoblast formation via multiple pathways, including the action of sclerostin.¹¹ Our patient had lytic lesions in 2 left lower ribs and in both forearms and femurs.

Hypercalcemia in multiple myeloma is attributed to 2 main factors: bone breakdown and macrophage overactivation. Multiple myeloma cells increase the release of macrophage inflammatory protein 1-alpha and tumor necrosis factor, which are inflammatory proteins that cause an increase in macrophages, which cause an increase in calcitriol.¹¹ As noted, our patient’s calcium level at presentation was 18.4 mg/dL uncorrected and 18.96 mg/dL corrected.

Cast nephropathy can occur in the distal tubules from the increased free light chains circulating and combining with Tamm-Horsfall protein, which in turn causes obstruction and local inflammation,¹² leading to a rise in creatinine levels and resulting in acute kidney injury,¹² as in our patient.

TREATMENT CONSIDERATIONS IN MULTIPLE MYELOMA

Our patient was referred to an oncologist for management.

In the management of multiple myeloma, the patient’s quality of life needs to be considered. With the development of new agents to combat the damages of the osteolytic effects, there is hope for improving quality of life.^13,14 New agents under study include anabolic agents such as antisclerostin and anti-Dickkopf-1, which promote osteoblastogenesis, leading to bone formation, with the possibility of repairing existing damage.¹⁵

TAKE-HOME POINTS

• If hypercalcemia is mild to moderate, consider primary hyperparathyroidism.
• Identify patients with severe symptoms of hypercalcemia such as volume depletion, acute kidney injury, arrhythmia, or seizures.
• Confirm severe cases of hypercalcemia and treat severe cases effectively.
• Severe hypercalcemia may need further investigation into a potential underlying malignancy.

Despite guidelines recommending low-dose oral glucocorticoids over high-dose intravenous (IV) glucocorticoids for inpatient management of acute exacerbations of chronic obstructive pulmonary disease (COPD), we have observed that most patients still receive high-dose IV therapy before being transitioned to low-dose oral therapy at discharge. Clinical inertia undoubtedly plays a significant role in the slow adoption of new recommendations, but in this era of evidence-based practice, the unfortunate lack of data supporting low over high steroid doses for acute exacerbations of COPD also contributes to hesitancy of physicians.

A SIGNIFICANT AND GROWING BURDEN

COPD is one of the most common pulmonary conditions managed by hospitalists today, and by the year 2030, it is predicted to become the third leading cause of death worldwide.¹

COPD is also a significant economic burden, costing $50 billion to manage in the United States, most of that from the cost of lengthy hospital stays.² COPD patients have 1 to 2 exacerbations per year.³ Bacterial and viral infections are responsible for most exacerbations, and 15% to 20% are from air pollution and other environmental causes of airway inflammation.³

CHALLENGES TO CHANGING PRACTICE

Glucocorticoids are the gold standard for treatment of acute exacerbations of COPD. It is well-documented that compared with placebo, glucocorticoids reduce mortality risk, length of hospital stay, and exacerbation recurrence after 1 month.⁴ And while high-dose IV steroid therapy has been the standard approach, oral administration has been found to be noninferior to IV administration with regard to treatment and length of hospital stay.⁵

While adverse effects are more common at higher doses, the optimal dose and duration of systemic glucocorticoid therapy for acute exacerbations of COPD are still largely at the discretion of the physician. The 2019 report of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recommends low doses (40 mg) for no more than 5 to 7 days for exacerbations, based on reports that showed no worse outcomes with low-dose oral than with high-dose IV therapy.^6,7 (In the 2010 study by Lindenauer et al,⁷ 92% of nearly 80,000 patients received high-dose IV steroids, reflecting standard practice at that time.) However, the GOLD guidelines do not address mortality rates, length of stay, or readmission rates for either approach, as they are devised to direct treatment in patients with stable mild to advanced COPD, not exacerbations.

Mortality rates

Aksoy et al⁸ established that, compared with placebo, low-dose steroids improved mortality rates in a subset of patients with acute exacerbations, specifically those with eosinophilic exacerbations. This study followed the 2013 Reduction in the Use of Corticosteroids in Exacerbated COPD (REDUCE) trial, which showed mortality rates were not lower with 14 days of low-dose prednisone treatment than with 5 days.9

Length of hospital stay

With regard to length of hospital stay, in 2011 Wang et al¹⁰ found no statistically significant difference between high- and low-dose steroid treatment.However, the REDUCE trial found that low-dose steroids shortened the median length of stay by 1 day compared with placebo.⁹

Hospital readmission rates

The REDUCE trial found no statistically significant difference in readmission rates when comparing 5 days of low-dose treatment vs 14 days.⁹ However, Aksoy et al⁸ found that readmission rates were significantly lower with low-dose treatment than with placebo.No study has yet examined readmission rates with high-dose vs low-dose steroid treatment.

What does the evidence tell us?

Low-dose oral glucocorticoid treatment shows definitive benefits in terms of lower mortality rates, shorter hospital length of stay, and lower readmission rates vs placebo in the treatment of acute exacerbations of COPD. Furthermore, a 14-day course is no better than 5 days in terms of mortality rates. And low-dose glucocorticoid treatment shows reduced mortality rates in addition to similar hospital length of stay when compared to high-dose glucocorticoid treatment.

Together, these findings lend credibility to the current GOLD recommendations. However, we have observed that in sharp contrast to the leading clinical guidelines, most patients hospitalized for acute exacerbations of COPD are still treated initially with high-dose IV corticosteroids. Why?

Obstacles that perpetuate the use of high-dose over low-dose treatment include lack of knowledge of glucocorticoid pharmacokinetics among clinicians, use of outdated order sets, and the reflex notion that more of a drug is more efficacious in its desired effect. In addition, administrative obstacles include using high-dose IV steroids to justify an inpatient stay or continued hospitalization.

COUNTERING THE OBSTACLES: THE HOSPITALIST’S ROLE

To counter these obstacles, we propose standardization of inpatient treatment of acute exacerbations of COPD to include initial low-dose steroid treatment in accordance with the most recent GOLD guidelines.⁶ This would benefit the patient by reducing undesirable effects of high-dose steroids, and at the same time reduce the economic burden of managing COPD exacerbations. Considering the large number of hospitalizations for COPD exacerbation each year, hospitalists can play a large role in this effort by routinely incorporating the low-dose steroid recommendation into their clinical practice.

Despite guidelines recommending low-dose oral glucocorticoids over high-dose intravenous (IV) glucocorticoids for inpatient management of acute exacerbations of chronic obstructive pulmonary disease (COPD), we have observed that most patients still receive high-dose IV therapy before being transitioned to low-dose oral therapy at discharge. Clinical inertia undoubtedly plays a significant role in the slow adoption of new recommendations, but in this era of evidence-based practice, the unfortunate lack of data supporting low over high steroid doses for acute exacerbations of COPD also contributes to hesitancy of physicians.

A SIGNIFICANT AND GROWING BURDEN

COPD is one of the most common pulmonary conditions managed by hospitalists today, and by the year 2030, it is predicted to become the third leading cause of death worldwide.¹

COPD is also a significant economic burden, costing $50 billion to manage in the United States, most of that from the cost of lengthy hospital stays.² COPD patients have 1 to 2 exacerbations per year.³ Bacterial and viral infections are responsible for most exacerbations, and 15% to 20% are from air pollution and other environmental causes of airway inflammation.³

CHALLENGES TO CHANGING PRACTICE

Glucocorticoids are the gold standard for treatment of acute exacerbations of COPD. It is well-documented that compared with placebo, glucocorticoids reduce mortality risk, length of hospital stay, and exacerbation recurrence after 1 month.⁴ And while high-dose IV steroid therapy has been the standard approach, oral administration has been found to be noninferior to IV administration with regard to treatment and length of hospital stay.⁵

While adverse effects are more common at higher doses, the optimal dose and duration of systemic glucocorticoid therapy for acute exacerbations of COPD are still largely at the discretion of the physician. The 2019 report of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recommends low doses (40 mg) for no more than 5 to 7 days for exacerbations, based on reports that showed no worse outcomes with low-dose oral than with high-dose IV therapy.^6,7 (In the 2010 study by Lindenauer et al,⁷ 92% of nearly 80,000 patients received high-dose IV steroids, reflecting standard practice at that time.) However, the GOLD guidelines do not address mortality rates, length of stay, or readmission rates for either approach, as they are devised to direct treatment in patients with stable mild to advanced COPD, not exacerbations.

Mortality rates

Aksoy et al⁸ established that, compared with placebo, low-dose steroids improved mortality rates in a subset of patients with acute exacerbations, specifically those with eosinophilic exacerbations. This study followed the 2013 Reduction in the Use of Corticosteroids in Exacerbated COPD (REDUCE) trial, which showed mortality rates were not lower with 14 days of low-dose prednisone treatment than with 5 days.9

Length of hospital stay

With regard to length of hospital stay, in 2011 Wang et al¹⁰ found no statistically significant difference between high- and low-dose steroid treatment.However, the REDUCE trial found that low-dose steroids shortened the median length of stay by 1 day compared with placebo.⁹

Hospital readmission rates

The REDUCE trial found no statistically significant difference in readmission rates when comparing 5 days of low-dose treatment vs 14 days.⁹ However, Aksoy et al⁸ found that readmission rates were significantly lower with low-dose treatment than with placebo.No study has yet examined readmission rates with high-dose vs low-dose steroid treatment.

What does the evidence tell us?

Low-dose oral glucocorticoid treatment shows definitive benefits in terms of lower mortality rates, shorter hospital length of stay, and lower readmission rates vs placebo in the treatment of acute exacerbations of COPD. Furthermore, a 14-day course is no better than 5 days in terms of mortality rates. And low-dose glucocorticoid treatment shows reduced mortality rates in addition to similar hospital length of stay when compared to high-dose glucocorticoid treatment.

Together, these findings lend credibility to the current GOLD recommendations. However, we have observed that in sharp contrast to the leading clinical guidelines, most patients hospitalized for acute exacerbations of COPD are still treated initially with high-dose IV corticosteroids. Why?

Obstacles that perpetuate the use of high-dose over low-dose treatment include lack of knowledge of glucocorticoid pharmacokinetics among clinicians, use of outdated order sets, and the reflex notion that more of a drug is more efficacious in its desired effect. In addition, administrative obstacles include using high-dose IV steroids to justify an inpatient stay or continued hospitalization.

COUNTERING THE OBSTACLES: THE HOSPITALIST’S ROLE

To counter these obstacles, we propose standardization of inpatient treatment of acute exacerbations of COPD to include initial low-dose steroid treatment in accordance with the most recent GOLD guidelines.⁶ This would benefit the patient by reducing undesirable effects of high-dose steroids, and at the same time reduce the economic burden of managing COPD exacerbations. Considering the large number of hospitalizations for COPD exacerbation each year, hospitalists can play a large role in this effort by routinely incorporating the low-dose steroid recommendation into their clinical practice.

Despite guidelines recommending low-dose oral glucocorticoids over high-dose intravenous (IV) glucocorticoids for inpatient management of acute exacerbations of chronic obstructive pulmonary disease (COPD), we have observed that most patients still receive high-dose IV therapy before being transitioned to low-dose oral therapy at discharge. Clinical inertia undoubtedly plays a significant role in the slow adoption of new recommendations, but in this era of evidence-based practice, the unfortunate lack of data supporting low over high steroid doses for acute exacerbations of COPD also contributes to hesitancy of physicians.

A SIGNIFICANT AND GROWING BURDEN

COPD is one of the most common pulmonary conditions managed by hospitalists today, and by the year 2030, it is predicted to become the third leading cause of death worldwide.¹

COPD is also a significant economic burden, costing $50 billion to manage in the United States, most of that from the cost of lengthy hospital stays.² COPD patients have 1 to 2 exacerbations per year.³ Bacterial and viral infections are responsible for most exacerbations, and 15% to 20% are from air pollution and other environmental causes of airway inflammation.³

CHALLENGES TO CHANGING PRACTICE

Glucocorticoids are the gold standard for treatment of acute exacerbations of COPD. It is well-documented that compared with placebo, glucocorticoids reduce mortality risk, length of hospital stay, and exacerbation recurrence after 1 month.⁴ And while high-dose IV steroid therapy has been the standard approach, oral administration has been found to be noninferior to IV administration with regard to treatment and length of hospital stay.⁵

While adverse effects are more common at higher doses, the optimal dose and duration of systemic glucocorticoid therapy for acute exacerbations of COPD are still largely at the discretion of the physician. The 2019 report of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recommends low doses (40 mg) for no more than 5 to 7 days for exacerbations, based on reports that showed no worse outcomes with low-dose oral than with high-dose IV therapy.^6,7 (In the 2010 study by Lindenauer et al,⁷ 92% of nearly 80,000 patients received high-dose IV steroids, reflecting standard practice at that time.) However, the GOLD guidelines do not address mortality rates, length of stay, or readmission rates for either approach, as they are devised to direct treatment in patients with stable mild to advanced COPD, not exacerbations.

Mortality rates

Aksoy et al⁸ established that, compared with placebo, low-dose steroids improved mortality rates in a subset of patients with acute exacerbations, specifically those with eosinophilic exacerbations. This study followed the 2013 Reduction in the Use of Corticosteroids in Exacerbated COPD (REDUCE) trial, which showed mortality rates were not lower with 14 days of low-dose prednisone treatment than with 5 days.9

Length of hospital stay

With regard to length of hospital stay, in 2011 Wang et al¹⁰ found no statistically significant difference between high- and low-dose steroid treatment.However, the REDUCE trial found that low-dose steroids shortened the median length of stay by 1 day compared with placebo.⁹

Hospital readmission rates

The REDUCE trial found no statistically significant difference in readmission rates when comparing 5 days of low-dose treatment vs 14 days.⁹ However, Aksoy et al⁸ found that readmission rates were significantly lower with low-dose treatment than with placebo.No study has yet examined readmission rates with high-dose vs low-dose steroid treatment.

What does the evidence tell us?

Low-dose oral glucocorticoid treatment shows definitive benefits in terms of lower mortality rates, shorter hospital length of stay, and lower readmission rates vs placebo in the treatment of acute exacerbations of COPD. Furthermore, a 14-day course is no better than 5 days in terms of mortality rates. And low-dose glucocorticoid treatment shows reduced mortality rates in addition to similar hospital length of stay when compared to high-dose glucocorticoid treatment.

Together, these findings lend credibility to the current GOLD recommendations. However, we have observed that in sharp contrast to the leading clinical guidelines, most patients hospitalized for acute exacerbations of COPD are still treated initially with high-dose IV corticosteroids. Why?

Obstacles that perpetuate the use of high-dose over low-dose treatment include lack of knowledge of glucocorticoid pharmacokinetics among clinicians, use of outdated order sets, and the reflex notion that more of a drug is more efficacious in its desired effect. In addition, administrative obstacles include using high-dose IV steroids to justify an inpatient stay or continued hospitalization.

COUNTERING THE OBSTACLES: THE HOSPITALIST’S ROLE

To counter these obstacles, we propose standardization of inpatient treatment of acute exacerbations of COPD to include initial low-dose steroid treatment in accordance with the most recent GOLD guidelines.⁶ This would benefit the patient by reducing undesirable effects of high-dose steroids, and at the same time reduce the economic burden of managing COPD exacerbations. Considering the large number of hospitalizations for COPD exacerbation each year, hospitalists can play a large role in this effort by routinely incorporating the low-dose steroid recommendation into their clinical practice.

“The most valuable of all talents is that of never using two words when one will do.“
—Thomas Jefferson

Attendings, residents, and medical students identify education as a top purpose of team rounds.¹ Learners report being dissatisfied with teaching on rounds most of the time.² Time with learners is a finite resource that has become even more precious with increasing clinical demands and work hour restrictions.³ Attendings report insufficient time to teach on rounds, and often neglect teaching because of time constraints.⁴ What can we do to in the face of this conflict between time and teaching?

One approach to this problem is what we call “ultra-brief, deliberate teaching sessions.” These sessions, or UBDTs, led by clinicians, create dedicated time for teaching on service. UBDTs ideally occur before team rounds because, in our experience, this is when the team is most unified and focused. Our sessions are time-limited (5 minutes or less) and designed so they are applicable to clinical scenarios the team is actively facing. Other learners can also lead these sessions with faculty coaching. Sessions of germane size and scope include: (1) Focus on a single clinical question from the previous day; (2) Discuss Choosing Wisely^® recommendations from a single specialty; (3) Provide a concise cognitive framework for a diagnostic or treatment dilemma (eg, draw a simple algorithm to evaluate causes of hyponatremia); (4) Review one image or electrocardiogram; (5) Present one case-based multiple-choice question; (6) Prime the team with a structured approach to a difficult conversation (eg, opioid discussions, goals of care).

As an example, if our team orders intravenous antihypertensives overnight, a UBDT session on asymptomatic hypertension would occur. The first minute may involve a discussion on the definition of hypertensive emergency versus asymptomatic hypertension. Next, we spend one minute asking learners the common causes of inpatient hypertension (eg, missed medications, pain, anxiety, withdrawal), highlighting that this warrants a bedside assessment. For two minutes, we next discuss the management options for asymptomatic hypertension with an emphasis on the avoidance of intravenous antihypertensives, tying this back to our current patient. Questions are welcomed, and a one-page summary of the major points and references is distributed during or after the talk. A repository of common topics and summaries may be a useful faculty development resource to be shared.

We have found UBDTs to be easy to implement for a variety of clinician educators. Because they are so brief and focused, they are also fun to create and share among teaching faculty. Importantly, these sessions should not delay clinical work. To ensure the avoidance of this trap, don’t select a topic that is too large or involves complex clinical reasoning, exceeds 5 minutes, or lead a UBDT session in a distracting environment or without preparation.

While we have not found a way to slow down time, UBDT sessions prior to the start of rounds can prioritize teaching, ensure the delivery of important content, and engage learners without significantly delaying clinical work. We invite you to try one!

Acknowledgments

The authors thank John Ragsdale, MD, MS for his leadership and support for UBDTs.

Disclosures

We have no relevant conflicts of interest to report. No payment or services from a third party were received for any aspect of this submitted work. We have no financial relationships with entities in the bio-medical arena that could be perceived to influence, or that give the appearance of potentially influencing, what was written in this submitted work.

“The most valuable of all talents is that of never using two words when one will do.“
—Thomas Jefferson

Attendings, residents, and medical students identify education as a top purpose of team rounds.¹ Learners report being dissatisfied with teaching on rounds most of the time.² Time with learners is a finite resource that has become even more precious with increasing clinical demands and work hour restrictions.³ Attendings report insufficient time to teach on rounds, and often neglect teaching because of time constraints.⁴ What can we do to in the face of this conflict between time and teaching?

One approach to this problem is what we call “ultra-brief, deliberate teaching sessions.” These sessions, or UBDTs, led by clinicians, create dedicated time for teaching on service. UBDTs ideally occur before team rounds because, in our experience, this is when the team is most unified and focused. Our sessions are time-limited (5 minutes or less) and designed so they are applicable to clinical scenarios the team is actively facing. Other learners can also lead these sessions with faculty coaching. Sessions of germane size and scope include: (1) Focus on a single clinical question from the previous day; (2) Discuss Choosing Wisely^® recommendations from a single specialty; (3) Provide a concise cognitive framework for a diagnostic or treatment dilemma (eg, draw a simple algorithm to evaluate causes of hyponatremia); (4) Review one image or electrocardiogram; (5) Present one case-based multiple-choice question; (6) Prime the team with a structured approach to a difficult conversation (eg, opioid discussions, goals of care).

As an example, if our team orders intravenous antihypertensives overnight, a UBDT session on asymptomatic hypertension would occur. The first minute may involve a discussion on the definition of hypertensive emergency versus asymptomatic hypertension. Next, we spend one minute asking learners the common causes of inpatient hypertension (eg, missed medications, pain, anxiety, withdrawal), highlighting that this warrants a bedside assessment. For two minutes, we next discuss the management options for asymptomatic hypertension with an emphasis on the avoidance of intravenous antihypertensives, tying this back to our current patient. Questions are welcomed, and a one-page summary of the major points and references is distributed during or after the talk. A repository of common topics and summaries may be a useful faculty development resource to be shared.

We have found UBDTs to be easy to implement for a variety of clinician educators. Because they are so brief and focused, they are also fun to create and share among teaching faculty. Importantly, these sessions should not delay clinical work. To ensure the avoidance of this trap, don’t select a topic that is too large or involves complex clinical reasoning, exceeds 5 minutes, or lead a UBDT session in a distracting environment or without preparation.

While we have not found a way to slow down time, UBDT sessions prior to the start of rounds can prioritize teaching, ensure the delivery of important content, and engage learners without significantly delaying clinical work. We invite you to try one!

Acknowledgments

The authors thank John Ragsdale, MD, MS for his leadership and support for UBDTs.

Disclosures

We have no relevant conflicts of interest to report. No payment or services from a third party were received for any aspect of this submitted work. We have no financial relationships with entities in the bio-medical arena that could be perceived to influence, or that give the appearance of potentially influencing, what was written in this submitted work.

“The most valuable of all talents is that of never using two words when one will do.“
—Thomas Jefferson

Attendings, residents, and medical students identify education as a top purpose of team rounds.¹ Learners report being dissatisfied with teaching on rounds most of the time.² Time with learners is a finite resource that has become even more precious with increasing clinical demands and work hour restrictions.³ Attendings report insufficient time to teach on rounds, and often neglect teaching because of time constraints.⁴ What can we do to in the face of this conflict between time and teaching?

One approach to this problem is what we call “ultra-brief, deliberate teaching sessions.” These sessions, or UBDTs, led by clinicians, create dedicated time for teaching on service. UBDTs ideally occur before team rounds because, in our experience, this is when the team is most unified and focused. Our sessions are time-limited (5 minutes or less) and designed so they are applicable to clinical scenarios the team is actively facing. Other learners can also lead these sessions with faculty coaching. Sessions of germane size and scope include: (1) Focus on a single clinical question from the previous day; (2) Discuss Choosing Wisely^® recommendations from a single specialty; (3) Provide a concise cognitive framework for a diagnostic or treatment dilemma (eg, draw a simple algorithm to evaluate causes of hyponatremia); (4) Review one image or electrocardiogram; (5) Present one case-based multiple-choice question; (6) Prime the team with a structured approach to a difficult conversation (eg, opioid discussions, goals of care).

As an example, if our team orders intravenous antihypertensives overnight, a UBDT session on asymptomatic hypertension would occur. The first minute may involve a discussion on the definition of hypertensive emergency versus asymptomatic hypertension. Next, we spend one minute asking learners the common causes of inpatient hypertension (eg, missed medications, pain, anxiety, withdrawal), highlighting that this warrants a bedside assessment. For two minutes, we next discuss the management options for asymptomatic hypertension with an emphasis on the avoidance of intravenous antihypertensives, tying this back to our current patient. Questions are welcomed, and a one-page summary of the major points and references is distributed during or after the talk. A repository of common topics and summaries may be a useful faculty development resource to be shared.

We have found UBDTs to be easy to implement for a variety of clinician educators. Because they are so brief and focused, they are also fun to create and share among teaching faculty. Importantly, these sessions should not delay clinical work. To ensure the avoidance of this trap, don’t select a topic that is too large or involves complex clinical reasoning, exceeds 5 minutes, or lead a UBDT session in a distracting environment or without preparation.

While we have not found a way to slow down time, UBDT sessions prior to the start of rounds can prioritize teaching, ensure the delivery of important content, and engage learners without significantly delaying clinical work. We invite you to try one!

Acknowledgments

The authors thank John Ragsdale, MD, MS for his leadership and support for UBDTs.

Disclosures

We have no relevant conflicts of interest to report. No payment or services from a third party were received for any aspect of this submitted work. We have no financial relationships with entities in the bio-medical arena that could be perceived to influence, or that give the appearance of potentially influencing, what was written in this submitted work.

Hospital medicine has grown dramatically over the past 20 years.^1,2 A recent survey regarding hospitalists’ clinical roles showed an expansion to triaging emergency department (ED) medical admissions and transfers from outside hospitals.³ From the hospitalist perspective, triaging involves the evaluation of patients for potential admission.⁴ With scrutiny on ED metrics, such as wait times (https://www.medicare.gov/hospitalcompare/search.html), health system administrators have heightened expectations for efficient patient flow, which increasingly falls to hospitalists.^5-7

Despite the growth in hospitalists’ triagist activities, there has been little formal assessment of their role. We hypothesized that this role differs from inpatient care in significant ways.^6-8 We sought to describe the triagist role in adult academic inpatient medicine settings to understand the responsibilities and skill set required.

Ten academic medical center (AMC) sites were recruited from Research Committee session attendees at the 2014 Society of Hospital Medicine national meeting and the 2014 Society of General Internal Medicine southern regional meeting. The AMCs were geographically diverse: three Western, two Midwestern, two Southern, one Northeastern, and two Southeastern. Site representatives were identified and completed a web-based questionnaire about their AMC (see Appendix 1 for the information collected). Clarifications regarding survey responses were performed via conference calls between the authors (STV, ESW) and site representatives.

Hospitalist Survey

In January 2018, surveys were sent to 583 physicians who worked as triagists. Participants received an anonymous 28-item RedCap survey by e-mail and were sent up to five reminder e-mails over six weeks (see Appendix 2 for the questions analyzed in this paper). Respondents were given the option to be entered in a gift card drawing.

Demographic information and individual workflow/practices were obtained. A 5-point Likert scale (strongly disagree – strongly agree) was used to assess hospitalists’ concurrence with current providers (eg, ED, clinic providers) regarding the management and whether patients must meet the utilization management (UM) criteria for admission. Time estimates used 5% increments and were categorized into four frequency categories based on the local modes provided in responses: Seldom (0%-10%), Occasional (15%-35%), Half-the-Time (40%-60%), and Frequently (65%-100%). Free text responses on effective/ineffective triagist qualities were elicited. Responses were included for analysis if at least 70% of questions were completed.

Data Analysis

Quantitative

Descriptive statistics were calculated for each variable. The Kruskal-Wallis test was used to evaluate differences across AMCs in the time spent on in-person evaluation and communication. Weighting, based on the ratio of hospitalists to survey respondents at each AMC, was used to calculate the average institutional percentages across the study sample.

Qualitative

Responses to open-ended questions were analyzed using thematic analysis.⁹ Three independent reviewers (STV, JC, ESW) read, analyzed, and grouped the responses by codes. Codes were then assessed for overlap and grouped into themes by one reviewer (STV). A table of themes with supporting quotes and the number of mentions was subsequently developed by all three reviewers. Similar themes were combined to create domains. The domains were reviewed by the steering committee members to create a consensus description (Appendix 3).

The University of Texas Health San Antonio’s Institutional Review Board and participating institutions approved the study as exempt.

Site Characteristics

Representatives from 10 AMCs reported data on a range of one to four hospitals for a total of 22 hospitals. The median reported that the number of medical patients admitted in a 24-hour period was 31-40 (range, 11-20 to >50). The median group size of hospitalists was 41-50 (range, 0-10 to >70).

The survey response rate was 40% (n = 235), ranging from 9%-70% between institutions. Self-identified female hospitalists accounted for 52% of respondents. Four percent were 25-29 years old, 66% were 30-39 years old, 24% were 40-49 years old, and 6% were ≥50 years old. The average clinical time spent as a triagist was 16%.

Description of Triagist Activities

The activities identified by the majority of respondents across all sites included transferring patients within the hospital (73%), and assessing/approving patient transfers from outside hospitals and clinics (82%). Internal transfer activities reported by >50% of respondents included allocating patients within the hospital or bed capacity coordination, assessing intensive care unit transfers, assigning ED admissions, and consulting other services. The ED accounted for an average of 55% of calls received. Respondents also reported being involved with the documentation related to these activities.

Similarities and Differences across AMCs

Two AMCs did not have a dedicated triagist; instead, physicians supervised residents and advanced practice providers. Among the eight sites with triagists, triaging was predominantly done by faculty physicians contacted via pagers. At seven of these sites, 100% of hospitalists worked as triagists. The triage service was covered by faculty physicians from 8-24 hours per day.

Bed boards and transfer centers staffed by registered nurses, nurse coordinators, house supervisors, or physicians were common support systems, though this infrastructure was organized differently across institutions. A UM review before admission was performed at three institutions 24 hours/day. The remaining institutions reviewed patients retrospectively.

Twenty-eight percent of hospitalists across all sites “Disagreed” or “Strongly disagreed” that a patient must meet UM criteria for admission. Forty-two percent had “Frequent” different opinions regarding patient management than the consulting provider.

Triagist and current provider communication practices varied widely across AMCs (Figure). There was significant variability in verbal communication (P = .02), with >70% of respondents at two AMCs reporting verbal communication at least half the time, but <30% reporting this frequency at two other AMCs. Respondents reported variable use of electronic communication (ie, notes/orders in the electronic health record) across AMCs (P < .0001). Half of the hospitalists use it “Seldom”, 20% use it “Occasionally”, and 23% use it “Frequently”.

Differences within AMCs

Variability within AMCs was greatest for the rate of verbal communication practices, with a typical interquartile range (IQR) of 20% to 90% among the hospitalists within a given AMC and for the rate of electronic communication with a typical IQR of 0% to 50%. For other survey questions, the IQR was typically 15 to 20 percentage points.

Thematic Analysis

We received 207 and 203 responses (88% and 86%, respectively) to the open-ended questions “What qualities does an effective triagist have?’ and ‘What qualities make a triagist ineffective?” We identified 22 themes for effective and ineffective qualities, which were grouped into seven domains (Table). All themes had at least three mentions by respondents. The three most frequently mentioned themes, communication skills, efficiency, and systems knowledge, had greater than 60 mentions.

Our study of the triagist role at 10 AMCs describes critical triagist functions and identifies key findings across and within AMCs. Twenty-eight percent of hospitalists reported admitting patients even when the patient did not meet the admission criteria, consistent with previous research demonstrating the influence of factors other than clinical disease severity on triage decisions.¹⁰ However, preventable admissions remain a hospital-level quality metric.^11,12 Triagists must often balance each patient’s circumstances with the complexities of the system. Juggling the competing demands of the system while providing patient-centered care can be challenging and may explain why attending physicians are more frequently filling this role.¹³

Local context/culture is likely to play a role in the variation across sites; however, compensation for the time spent may also be a factor. If triage activities are not reimbursable, this could lead to less documentation and a lower likelihood that patients are evaluated in person.¹⁴ This reason may also explain why all hospitalists were required to serve as a triagist at most sites.

Currently, no consensus definition of the triagist role has been developed. Our results demonstrate that this role is heterogeneous and grounded in the local healthcare system practices. We propose the following working definition of the triagist: a physician who assesses patients for admission, actively supporting the transition of the patient from the outpatient to the inpatient setting. A triagist should be equipped with a skill set that includes not only clinical knowledge but also emphasizes systems knowledge, awareness of others’ goals, efficiency, an ability to communicate effectively, and the knowledge of UM. We recommend that medical directors of hospitalist programs focus their attention on locally specific, systems-based skills development when orienting new hospitalists. The financial aspects of cost should be considered and delineated as well.

Our analysis is limited in several respects. Participant AMCs were not randomly chosen, but do represent a broad array of facility types, group size, and geographic regions. The low response rates at some AMCs may result in an inaccurate representation of those sites. Data was not obtained on hospitalists that did not respond to the survey; therefore, nonresponse bias may affect outcomes. This research used self-report rather than direct observation, which could be subject to recall and social desirability bias. Finally, our results may not be generalizable to nonacademic institutions.

The hospitalist role as triagist at AMCs emphasizes communication, organizational skills, efficiency, systems-based practice, and UM knowledge. Although we found significant variation across and within AMCs, internal transfer activities were common across programs. Hospitalist programs should focus on systems-based skills development to prepare hospitalists for the role. The skill set necessary for triagist responsibilities also has implications for internal medicine resident education.⁴ With increasing emphasis on value and system effectiveness in care delivery, further studies of the triagist role should be undertaken.

Acknowledgments

The TRIAGIST Collaborative Group consists of: Maralyssa Bann, MD, Andrew White, MD (University of Washington); Jagriti Chadha, MD (University of Kentucky); Joel Boggan, MD (Duke University); Sherwin Hsu, MD (UCLA); Jeff Liao, MD (Harvard Medical School); Tabatha Matthias, DO (University of Nebraska Medical Center); Tresa McNeal, MD (Scott and White Texas A&M); Roxana Naderi, MD, Khooshbu Shah, MD (University of Colorado); David Schmit, MD (University of Texas Health San Antonio); Manivannan Veerasamy, MD (Michigan State University).

Disclaimer

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs.

Hospital medicine has grown dramatically over the past 20 years.^1,2 A recent survey regarding hospitalists’ clinical roles showed an expansion to triaging emergency department (ED) medical admissions and transfers from outside hospitals.³ From the hospitalist perspective, triaging involves the evaluation of patients for potential admission.⁴ With scrutiny on ED metrics, such as wait times (https://www.medicare.gov/hospitalcompare/search.html), health system administrators have heightened expectations for efficient patient flow, which increasingly falls to hospitalists.^5-7

Despite the growth in hospitalists’ triagist activities, there has been little formal assessment of their role. We hypothesized that this role differs from inpatient care in significant ways.^6-8 We sought to describe the triagist role in adult academic inpatient medicine settings to understand the responsibilities and skill set required.

Ten academic medical center (AMC) sites were recruited from Research Committee session attendees at the 2014 Society of Hospital Medicine national meeting and the 2014 Society of General Internal Medicine southern regional meeting. The AMCs were geographically diverse: three Western, two Midwestern, two Southern, one Northeastern, and two Southeastern. Site representatives were identified and completed a web-based questionnaire about their AMC (see Appendix 1 for the information collected). Clarifications regarding survey responses were performed via conference calls between the authors (STV, ESW) and site representatives.

Hospitalist Survey

In January 2018, surveys were sent to 583 physicians who worked as triagists. Participants received an anonymous 28-item RedCap survey by e-mail and were sent up to five reminder e-mails over six weeks (see Appendix 2 for the questions analyzed in this paper). Respondents were given the option to be entered in a gift card drawing.

Demographic information and individual workflow/practices were obtained. A 5-point Likert scale (strongly disagree – strongly agree) was used to assess hospitalists’ concurrence with current providers (eg, ED, clinic providers) regarding the management and whether patients must meet the utilization management (UM) criteria for admission. Time estimates used 5% increments and were categorized into four frequency categories based on the local modes provided in responses: Seldom (0%-10%), Occasional (15%-35%), Half-the-Time (40%-60%), and Frequently (65%-100%). Free text responses on effective/ineffective triagist qualities were elicited. Responses were included for analysis if at least 70% of questions were completed.

Data Analysis

Quantitative

Descriptive statistics were calculated for each variable. The Kruskal-Wallis test was used to evaluate differences across AMCs in the time spent on in-person evaluation and communication. Weighting, based on the ratio of hospitalists to survey respondents at each AMC, was used to calculate the average institutional percentages across the study sample.

Qualitative

Responses to open-ended questions were analyzed using thematic analysis.⁹ Three independent reviewers (STV, JC, ESW) read, analyzed, and grouped the responses by codes. Codes were then assessed for overlap and grouped into themes by one reviewer (STV). A table of themes with supporting quotes and the number of mentions was subsequently developed by all three reviewers. Similar themes were combined to create domains. The domains were reviewed by the steering committee members to create a consensus description (Appendix 3).

The University of Texas Health San Antonio’s Institutional Review Board and participating institutions approved the study as exempt.

Site Characteristics

Representatives from 10 AMCs reported data on a range of one to four hospitals for a total of 22 hospitals. The median reported that the number of medical patients admitted in a 24-hour period was 31-40 (range, 11-20 to >50). The median group size of hospitalists was 41-50 (range, 0-10 to >70).

The survey response rate was 40% (n = 235), ranging from 9%-70% between institutions. Self-identified female hospitalists accounted for 52% of respondents. Four percent were 25-29 years old, 66% were 30-39 years old, 24% were 40-49 years old, and 6% were ≥50 years old. The average clinical time spent as a triagist was 16%.

Description of Triagist Activities

The activities identified by the majority of respondents across all sites included transferring patients within the hospital (73%), and assessing/approving patient transfers from outside hospitals and clinics (82%). Internal transfer activities reported by >50% of respondents included allocating patients within the hospital or bed capacity coordination, assessing intensive care unit transfers, assigning ED admissions, and consulting other services. The ED accounted for an average of 55% of calls received. Respondents also reported being involved with the documentation related to these activities.

Similarities and Differences across AMCs

Two AMCs did not have a dedicated triagist; instead, physicians supervised residents and advanced practice providers. Among the eight sites with triagists, triaging was predominantly done by faculty physicians contacted via pagers. At seven of these sites, 100% of hospitalists worked as triagists. The triage service was covered by faculty physicians from 8-24 hours per day.

Bed boards and transfer centers staffed by registered nurses, nurse coordinators, house supervisors, or physicians were common support systems, though this infrastructure was organized differently across institutions. A UM review before admission was performed at three institutions 24 hours/day. The remaining institutions reviewed patients retrospectively.

Twenty-eight percent of hospitalists across all sites “Disagreed” or “Strongly disagreed” that a patient must meet UM criteria for admission. Forty-two percent had “Frequent” different opinions regarding patient management than the consulting provider.

Triagist and current provider communication practices varied widely across AMCs (Figure). There was significant variability in verbal communication (P = .02), with >70% of respondents at two AMCs reporting verbal communication at least half the time, but <30% reporting this frequency at two other AMCs. Respondents reported variable use of electronic communication (ie, notes/orders in the electronic health record) across AMCs (P < .0001). Half of the hospitalists use it “Seldom”, 20% use it “Occasionally”, and 23% use it “Frequently”.

Differences within AMCs

Variability within AMCs was greatest for the rate of verbal communication practices, with a typical interquartile range (IQR) of 20% to 90% among the hospitalists within a given AMC and for the rate of electronic communication with a typical IQR of 0% to 50%. For other survey questions, the IQR was typically 15 to 20 percentage points.

Thematic Analysis

We received 207 and 203 responses (88% and 86%, respectively) to the open-ended questions “What qualities does an effective triagist have?’ and ‘What qualities make a triagist ineffective?” We identified 22 themes for effective and ineffective qualities, which were grouped into seven domains (Table). All themes had at least three mentions by respondents. The three most frequently mentioned themes, communication skills, efficiency, and systems knowledge, had greater than 60 mentions.

Our study of the triagist role at 10 AMCs describes critical triagist functions and identifies key findings across and within AMCs. Twenty-eight percent of hospitalists reported admitting patients even when the patient did not meet the admission criteria, consistent with previous research demonstrating the influence of factors other than clinical disease severity on triage decisions.¹⁰ However, preventable admissions remain a hospital-level quality metric.^11,12 Triagists must often balance each patient’s circumstances with the complexities of the system. Juggling the competing demands of the system while providing patient-centered care can be challenging and may explain why attending physicians are more frequently filling this role.¹³

Local context/culture is likely to play a role in the variation across sites; however, compensation for the time spent may also be a factor. If triage activities are not reimbursable, this could lead to less documentation and a lower likelihood that patients are evaluated in person.¹⁴ This reason may also explain why all hospitalists were required to serve as a triagist at most sites.

Currently, no consensus definition of the triagist role has been developed. Our results demonstrate that this role is heterogeneous and grounded in the local healthcare system practices. We propose the following working definition of the triagist: a physician who assesses patients for admission, actively supporting the transition of the patient from the outpatient to the inpatient setting. A triagist should be equipped with a skill set that includes not only clinical knowledge but also emphasizes systems knowledge, awareness of others’ goals, efficiency, an ability to communicate effectively, and the knowledge of UM. We recommend that medical directors of hospitalist programs focus their attention on locally specific, systems-based skills development when orienting new hospitalists. The financial aspects of cost should be considered and delineated as well.

Our analysis is limited in several respects. Participant AMCs were not randomly chosen, but do represent a broad array of facility types, group size, and geographic regions. The low response rates at some AMCs may result in an inaccurate representation of those sites. Data was not obtained on hospitalists that did not respond to the survey; therefore, nonresponse bias may affect outcomes. This research used self-report rather than direct observation, which could be subject to recall and social desirability bias. Finally, our results may not be generalizable to nonacademic institutions.

The hospitalist role as triagist at AMCs emphasizes communication, organizational skills, efficiency, systems-based practice, and UM knowledge. Although we found significant variation across and within AMCs, internal transfer activities were common across programs. Hospitalist programs should focus on systems-based skills development to prepare hospitalists for the role. The skill set necessary for triagist responsibilities also has implications for internal medicine resident education.⁴ With increasing emphasis on value and system effectiveness in care delivery, further studies of the triagist role should be undertaken.

Acknowledgments

The TRIAGIST Collaborative Group consists of: Maralyssa Bann, MD, Andrew White, MD (University of Washington); Jagriti Chadha, MD (University of Kentucky); Joel Boggan, MD (Duke University); Sherwin Hsu, MD (UCLA); Jeff Liao, MD (Harvard Medical School); Tabatha Matthias, DO (University of Nebraska Medical Center); Tresa McNeal, MD (Scott and White Texas A&M); Roxana Naderi, MD, Khooshbu Shah, MD (University of Colorado); David Schmit, MD (University of Texas Health San Antonio); Manivannan Veerasamy, MD (Michigan State University).

Disclaimer

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs.

Hospital medicine has grown dramatically over the past 20 years.^1,2 A recent survey regarding hospitalists’ clinical roles showed an expansion to triaging emergency department (ED) medical admissions and transfers from outside hospitals.³ From the hospitalist perspective, triaging involves the evaluation of patients for potential admission.⁴ With scrutiny on ED metrics, such as wait times (https://www.medicare.gov/hospitalcompare/search.html), health system administrators have heightened expectations for efficient patient flow, which increasingly falls to hospitalists.^5-7

Despite the growth in hospitalists’ triagist activities, there has been little formal assessment of their role. We hypothesized that this role differs from inpatient care in significant ways.^6-8 We sought to describe the triagist role in adult academic inpatient medicine settings to understand the responsibilities and skill set required.

Ten academic medical center (AMC) sites were recruited from Research Committee session attendees at the 2014 Society of Hospital Medicine national meeting and the 2014 Society of General Internal Medicine southern regional meeting. The AMCs were geographically diverse: three Western, two Midwestern, two Southern, one Northeastern, and two Southeastern. Site representatives were identified and completed a web-based questionnaire about their AMC (see Appendix 1 for the information collected). Clarifications regarding survey responses were performed via conference calls between the authors (STV, ESW) and site representatives.

Hospitalist Survey

In January 2018, surveys were sent to 583 physicians who worked as triagists. Participants received an anonymous 28-item RedCap survey by e-mail and were sent up to five reminder e-mails over six weeks (see Appendix 2 for the questions analyzed in this paper). Respondents were given the option to be entered in a gift card drawing.

Demographic information and individual workflow/practices were obtained. A 5-point Likert scale (strongly disagree – strongly agree) was used to assess hospitalists’ concurrence with current providers (eg, ED, clinic providers) regarding the management and whether patients must meet the utilization management (UM) criteria for admission. Time estimates used 5% increments and were categorized into four frequency categories based on the local modes provided in responses: Seldom (0%-10%), Occasional (15%-35%), Half-the-Time (40%-60%), and Frequently (65%-100%). Free text responses on effective/ineffective triagist qualities were elicited. Responses were included for analysis if at least 70% of questions were completed.

Data Analysis

Quantitative

Descriptive statistics were calculated for each variable. The Kruskal-Wallis test was used to evaluate differences across AMCs in the time spent on in-person evaluation and communication. Weighting, based on the ratio of hospitalists to survey respondents at each AMC, was used to calculate the average institutional percentages across the study sample.

Qualitative

Responses to open-ended questions were analyzed using thematic analysis.⁹ Three independent reviewers (STV, JC, ESW) read, analyzed, and grouped the responses by codes. Codes were then assessed for overlap and grouped into themes by one reviewer (STV). A table of themes with supporting quotes and the number of mentions was subsequently developed by all three reviewers. Similar themes were combined to create domains. The domains were reviewed by the steering committee members to create a consensus description (Appendix 3).

The University of Texas Health San Antonio’s Institutional Review Board and participating institutions approved the study as exempt.

Site Characteristics

Representatives from 10 AMCs reported data on a range of one to four hospitals for a total of 22 hospitals. The median reported that the number of medical patients admitted in a 24-hour period was 31-40 (range, 11-20 to >50). The median group size of hospitalists was 41-50 (range, 0-10 to >70).

The survey response rate was 40% (n = 235), ranging from 9%-70% between institutions. Self-identified female hospitalists accounted for 52% of respondents. Four percent were 25-29 years old, 66% were 30-39 years old, 24% were 40-49 years old, and 6% were ≥50 years old. The average clinical time spent as a triagist was 16%.

Description of Triagist Activities

The activities identified by the majority of respondents across all sites included transferring patients within the hospital (73%), and assessing/approving patient transfers from outside hospitals and clinics (82%). Internal transfer activities reported by >50% of respondents included allocating patients within the hospital or bed capacity coordination, assessing intensive care unit transfers, assigning ED admissions, and consulting other services. The ED accounted for an average of 55% of calls received. Respondents also reported being involved with the documentation related to these activities.

Similarities and Differences across AMCs

Two AMCs did not have a dedicated triagist; instead, physicians supervised residents and advanced practice providers. Among the eight sites with triagists, triaging was predominantly done by faculty physicians contacted via pagers. At seven of these sites, 100% of hospitalists worked as triagists. The triage service was covered by faculty physicians from 8-24 hours per day.

Bed boards and transfer centers staffed by registered nurses, nurse coordinators, house supervisors, or physicians were common support systems, though this infrastructure was organized differently across institutions. A UM review before admission was performed at three institutions 24 hours/day. The remaining institutions reviewed patients retrospectively.

Twenty-eight percent of hospitalists across all sites “Disagreed” or “Strongly disagreed” that a patient must meet UM criteria for admission. Forty-two percent had “Frequent” different opinions regarding patient management than the consulting provider.

Triagist and current provider communication practices varied widely across AMCs (Figure). There was significant variability in verbal communication (P = .02), with >70% of respondents at two AMCs reporting verbal communication at least half the time, but <30% reporting this frequency at two other AMCs. Respondents reported variable use of electronic communication (ie, notes/orders in the electronic health record) across AMCs (P < .0001). Half of the hospitalists use it “Seldom”, 20% use it “Occasionally”, and 23% use it “Frequently”.

Differences within AMCs

Variability within AMCs was greatest for the rate of verbal communication practices, with a typical interquartile range (IQR) of 20% to 90% among the hospitalists within a given AMC and for the rate of electronic communication with a typical IQR of 0% to 50%. For other survey questions, the IQR was typically 15 to 20 percentage points.

Thematic Analysis

We received 207 and 203 responses (88% and 86%, respectively) to the open-ended questions “What qualities does an effective triagist have?’ and ‘What qualities make a triagist ineffective?” We identified 22 themes for effective and ineffective qualities, which were grouped into seven domains (Table). All themes had at least three mentions by respondents. The three most frequently mentioned themes, communication skills, efficiency, and systems knowledge, had greater than 60 mentions.

Our study of the triagist role at 10 AMCs describes critical triagist functions and identifies key findings across and within AMCs. Twenty-eight percent of hospitalists reported admitting patients even when the patient did not meet the admission criteria, consistent with previous research demonstrating the influence of factors other than clinical disease severity on triage decisions.¹⁰ However, preventable admissions remain a hospital-level quality metric.^11,12 Triagists must often balance each patient’s circumstances with the complexities of the system. Juggling the competing demands of the system while providing patient-centered care can be challenging and may explain why attending physicians are more frequently filling this role.¹³

Local context/culture is likely to play a role in the variation across sites; however, compensation for the time spent may also be a factor. If triage activities are not reimbursable, this could lead to less documentation and a lower likelihood that patients are evaluated in person.¹⁴ This reason may also explain why all hospitalists were required to serve as a triagist at most sites.

Currently, no consensus definition of the triagist role has been developed. Our results demonstrate that this role is heterogeneous and grounded in the local healthcare system practices. We propose the following working definition of the triagist: a physician who assesses patients for admission, actively supporting the transition of the patient from the outpatient to the inpatient setting. A triagist should be equipped with a skill set that includes not only clinical knowledge but also emphasizes systems knowledge, awareness of others’ goals, efficiency, an ability to communicate effectively, and the knowledge of UM. We recommend that medical directors of hospitalist programs focus their attention on locally specific, systems-based skills development when orienting new hospitalists. The financial aspects of cost should be considered and delineated as well.

Our analysis is limited in several respects. Participant AMCs were not randomly chosen, but do represent a broad array of facility types, group size, and geographic regions. The low response rates at some AMCs may result in an inaccurate representation of those sites. Data was not obtained on hospitalists that did not respond to the survey; therefore, nonresponse bias may affect outcomes. This research used self-report rather than direct observation, which could be subject to recall and social desirability bias. Finally, our results may not be generalizable to nonacademic institutions.

The hospitalist role as triagist at AMCs emphasizes communication, organizational skills, efficiency, systems-based practice, and UM knowledge. Although we found significant variation across and within AMCs, internal transfer activities were common across programs. Hospitalist programs should focus on systems-based skills development to prepare hospitalists for the role. The skill set necessary for triagist responsibilities also has implications for internal medicine resident education.⁴ With increasing emphasis on value and system effectiveness in care delivery, further studies of the triagist role should be undertaken.

Acknowledgments

The TRIAGIST Collaborative Group consists of: Maralyssa Bann, MD, Andrew White, MD (University of Washington); Jagriti Chadha, MD (University of Kentucky); Joel Boggan, MD (Duke University); Sherwin Hsu, MD (UCLA); Jeff Liao, MD (Harvard Medical School); Tabatha Matthias, DO (University of Nebraska Medical Center); Tresa McNeal, MD (Scott and White Texas A&M); Roxana Naderi, MD, Khooshbu Shah, MD (University of Colorado); David Schmit, MD (University of Texas Health San Antonio); Manivannan Veerasamy, MD (Michigan State University).

Disclaimer

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs.

Acute pancreatitis (AP) is the most common gastrointestinal discharge diagnosis in the United States, with a mortality rate of 1%-5%.¹ Recent data demonstrate increasing AP-related admissions, making AP management of utmost importance to hospitalists.¹ The American Gastroenterological Association (AGA) guideline specifically addresses AP management in the initial 48-72 hours of admission, during which management decisions can alter disease course and length of stay. AP requires two of the following three criteria for diagnosis: characteristic abdominal pain, elevation of lipase or amylase ≥3 times the upper limit of normal, and/or radiographic evidence of pancreatitis on cross-sectional imaging. The guideline provides eight recommendations, which we consolidated to highlight practice changing recommendations: fluids, nutrition, management of the most common causes, and prophylactic antibiotics.^2,3

KEY RECOMMENDATIONS FOR THE HOSPITALIST

Fluids

Recommendation 1. In patients with AP, use goal-directed isotonic crystalloids for fluid management (conditional recommendation, very low-quality evidence).

The guideline emphasizes goal-directed fluid management despite low-quality, heterogeneous evidence and does not recommend Ringer’s lactate over normal saline. “Goal-directed” fluid management involves the use of crystalloid infusions titrated to improve physiologic and biochemical markers, but no target volume is specified by the guideline. Frequent reassessments should look for signs of volume overload, the primary risk of harm with fluid therapy. Despite failure to reduce mortality or morbidities such as pancreatic necrosis or persistent multi-organ failure, the AGA cites the mortality benefit of goal-directed therapy in sepsis as justification for this approach in AP, given the similar physiologic abnormalities.

Nutrition

Recommendation 2. Begin feeding early in patients with AP regardless of predicted severity. If oral nutrition is not tolerated, enteral feeding with either a nasogastric or nasojejunal tube is preferred to parenteral nutrition (strong recommendation, moderate-quality evidence).

Early feeding (ie, within 24 hours) is recommended regardless of AP severity. This represents a change from prior practices of bowel rest, theorized to prevent continued stimulation of an inflamed pancreas. Although early feeding has not been linked to improved mortality, it has demonstrated lower rates of multi-organ failure and infected pancreatic necrosis, possibly due to maintenance of the gut mucosal barrier and reduced bacterial translocation. When oral feeding is not tolerated, enteral nutrition is preferred over parenteral nutrition due to less risks. The preferred dietary composition guidance for patients with persistent pain or ileus is not addressed.

Management of the Most Common Causes of AP in Adults

Recommendation 3. Patients with mild acute biliary pancreatitis should have cholecystectomy during the initial admission (strong recommendation, moderate-quality evidence).

All patients with suspected biliary pancreatitis should receive a surgical consultation for cholecystectomy during the index admission. At the time of the guideline release, only one trial was available to support the recommendation of early cholecystectomy; however, newer studies similarly support cholecystectomy during index admission by demonstrating reductions in composite outcomes of mortality and gallstone-related complications, readmission for pancreatitis, and other pancreatobiliary complications.⁴ A Cochrane review included in the guideline found no differences in complication rates even in patients with severe biliary pancreatitis. In the absence of cholangitis, urgent endoscopic retrograde cholangiography (ERCP) is not indicated as most stones causing biliary pancreatitis pass spontaneously.

Recommendation 4. In patients with acute alcoholic pancreatitis, brief alcohol intervention should occur during admission (strong recommendation, moderate-quality evidence).

Ongoing alcohol consumption is a risk factor for recurrent acute and chronic pancreatitis. Only one trial assessed the impact of inpatient alcohol cessation counseling on recurrent AP, noting a trend toward reduced readmissions.⁵ However, indirect evidence from similar interventions in ambulatory settings demonstrates reductions in alcohol intake, leading to the AGA recommendation for inpatients with alcohol-induced AP.³

Antibiotics

Recommendation 5. Avoid empiric antibiotics in patients with AP who otherwise lack an indication, regardless of predicted severity (conditional recommendation, low-quality evidence).

Since 2002, well performed trials have consistently failed to demonstrate improvement in outcomes such as multi-organ failure or length of stay with use of prophylactic antibiotics for AP, even severe AP and pancreatic necrosis. Therefore, the AGA recommends against prophylactic antibiotics in initial management of AP regardless of disease severity. Lack of blinding in the majority of trial designs conducted before 2002 contributed to the overall assessment of low-quality evidence. The guideline does not address acute biliary pancreatitis with cholangitis, for which antibiotics and ERCP for decompression are critical.

CRITIQUE

The AGA Institute supported this guideline development and employed the rigorous and standardized GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology. This approach allowed the guideline panel members to account not only for evidence quality, but also the benefits and harms of an intervention and resource utilization. None of the authors had any stated conflicts of interest.

The guideline heavily weighted results from randomized control trials, most of which excluded key populations cared for by hospitalists (eg, patients older than 75 years, with end-stage renal disease). Particular areas where this creates challenges for clinicians and patients alike include goal-directed fluid therapy and when to consider more invasive interventions such as ERCP and early cholecystectomy. For example, patients considered to be poor surgical candidates may benefit from ERCP with biliary sphincterotomy to reduce the risk of recurrent biliary pancreatitis.

Lack of specificity in the guidelines for goal-directed fluid management and enteral feeding regimens makes it challenging to standardize hospitalists’ approach to the early care of patients with AP. Interestingly, the 2013 American College of Gastroenterology (ACG) Guideline for the Management of AP included strong recommendations for the use of Ringer’s lactate and volume targets in the initial management of AP.⁶ Evidence supporting the use of Ringer’s lactate versus normal saline is based largely upon improved inflammatory markers, theoretical potentiation of pancreatic enzyme activation with hypercholemic metabolic acidosis, and small studies demonstrating trends toward improved mortality.⁷ The ACG guideline was released prior to mounting evidence suggesting that goal-directed fluid therapy in sepsis does not improve mortality versus usual care.⁸ The growing uncertainty regarding the efficacy of goal-directed fluids for septic shock, as well limitations of studies on AP, may contribute to the differences between the AGA and ACG recommendations.

Finally, as the guideline covers the initial therapeutic management of AP, no recommendations are made for diagnostic studies such as right upper quadrant ultrasound. This noninvasive and readily available test plays a critical role in evaluating for presence of gallstones and other potential etiologies of abdominal pain.

AREAS IN NEED OF FUTURE STUDY

Additional research is needed to better understand goal-directed fluid therapy with respect to the fluid type, amount, and target outcomes. Similarly, determining the optimal enteral feeding regimens for patients failing oral intake would help clinicians meet the recommendation for early nutrition. Finally, clarification on the roles and timing of endoscopic and surgical procedures for patients with severe biliary pancreatitis, as well as geriatric and medically complex populations, would help hospitalists advocate for a multidisciplinary approach to this common and often serious disease.

Disclosures

The authors have nothing to disclose.

Acute pancreatitis (AP) is the most common gastrointestinal discharge diagnosis in the United States, with a mortality rate of 1%-5%.¹ Recent data demonstrate increasing AP-related admissions, making AP management of utmost importance to hospitalists.¹ The American Gastroenterological Association (AGA) guideline specifically addresses AP management in the initial 48-72 hours of admission, during which management decisions can alter disease course and length of stay. AP requires two of the following three criteria for diagnosis: characteristic abdominal pain, elevation of lipase or amylase ≥3 times the upper limit of normal, and/or radiographic evidence of pancreatitis on cross-sectional imaging. The guideline provides eight recommendations, which we consolidated to highlight practice changing recommendations: fluids, nutrition, management of the most common causes, and prophylactic antibiotics.^2,3

KEY RECOMMENDATIONS FOR THE HOSPITALIST

Fluids

Recommendation 1. In patients with AP, use goal-directed isotonic crystalloids for fluid management (conditional recommendation, very low-quality evidence).

The guideline emphasizes goal-directed fluid management despite low-quality, heterogeneous evidence and does not recommend Ringer’s lactate over normal saline. “Goal-directed” fluid management involves the use of crystalloid infusions titrated to improve physiologic and biochemical markers, but no target volume is specified by the guideline. Frequent reassessments should look for signs of volume overload, the primary risk of harm with fluid therapy. Despite failure to reduce mortality or morbidities such as pancreatic necrosis or persistent multi-organ failure, the AGA cites the mortality benefit of goal-directed therapy in sepsis as justification for this approach in AP, given the similar physiologic abnormalities.

Nutrition

Recommendation 2. Begin feeding early in patients with AP regardless of predicted severity. If oral nutrition is not tolerated, enteral feeding with either a nasogastric or nasojejunal tube is preferred to parenteral nutrition (strong recommendation, moderate-quality evidence).

Early feeding (ie, within 24 hours) is recommended regardless of AP severity. This represents a change from prior practices of bowel rest, theorized to prevent continued stimulation of an inflamed pancreas. Although early feeding has not been linked to improved mortality, it has demonstrated lower rates of multi-organ failure and infected pancreatic necrosis, possibly due to maintenance of the gut mucosal barrier and reduced bacterial translocation. When oral feeding is not tolerated, enteral nutrition is preferred over parenteral nutrition due to less risks. The preferred dietary composition guidance for patients with persistent pain or ileus is not addressed.

Management of the Most Common Causes of AP in Adults

Recommendation 3. Patients with mild acute biliary pancreatitis should have cholecystectomy during the initial admission (strong recommendation, moderate-quality evidence).

All patients with suspected biliary pancreatitis should receive a surgical consultation for cholecystectomy during the index admission. At the time of the guideline release, only one trial was available to support the recommendation of early cholecystectomy; however, newer studies similarly support cholecystectomy during index admission by demonstrating reductions in composite outcomes of mortality and gallstone-related complications, readmission for pancreatitis, and other pancreatobiliary complications.⁴ A Cochrane review included in the guideline found no differences in complication rates even in patients with severe biliary pancreatitis. In the absence of cholangitis, urgent endoscopic retrograde cholangiography (ERCP) is not indicated as most stones causing biliary pancreatitis pass spontaneously.

Recommendation 4. In patients with acute alcoholic pancreatitis, brief alcohol intervention should occur during admission (strong recommendation, moderate-quality evidence).

Ongoing alcohol consumption is a risk factor for recurrent acute and chronic pancreatitis. Only one trial assessed the impact of inpatient alcohol cessation counseling on recurrent AP, noting a trend toward reduced readmissions.⁵ However, indirect evidence from similar interventions in ambulatory settings demonstrates reductions in alcohol intake, leading to the AGA recommendation for inpatients with alcohol-induced AP.³

Antibiotics

Recommendation 5. Avoid empiric antibiotics in patients with AP who otherwise lack an indication, regardless of predicted severity (conditional recommendation, low-quality evidence).

Since 2002, well performed trials have consistently failed to demonstrate improvement in outcomes such as multi-organ failure or length of stay with use of prophylactic antibiotics for AP, even severe AP and pancreatic necrosis. Therefore, the AGA recommends against prophylactic antibiotics in initial management of AP regardless of disease severity. Lack of blinding in the majority of trial designs conducted before 2002 contributed to the overall assessment of low-quality evidence. The guideline does not address acute biliary pancreatitis with cholangitis, for which antibiotics and ERCP for decompression are critical.

CRITIQUE

The AGA Institute supported this guideline development and employed the rigorous and standardized GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology. This approach allowed the guideline panel members to account not only for evidence quality, but also the benefits and harms of an intervention and resource utilization. None of the authors had any stated conflicts of interest.

The guideline heavily weighted results from randomized control trials, most of which excluded key populations cared for by hospitalists (eg, patients older than 75 years, with end-stage renal disease). Particular areas where this creates challenges for clinicians and patients alike include goal-directed fluid therapy and when to consider more invasive interventions such as ERCP and early cholecystectomy. For example, patients considered to be poor surgical candidates may benefit from ERCP with biliary sphincterotomy to reduce the risk of recurrent biliary pancreatitis.

Lack of specificity in the guidelines for goal-directed fluid management and enteral feeding regimens makes it challenging to standardize hospitalists’ approach to the early care of patients with AP. Interestingly, the 2013 American College of Gastroenterology (ACG) Guideline for the Management of AP included strong recommendations for the use of Ringer’s lactate and volume targets in the initial management of AP.⁶ Evidence supporting the use of Ringer’s lactate versus normal saline is based largely upon improved inflammatory markers, theoretical potentiation of pancreatic enzyme activation with hypercholemic metabolic acidosis, and small studies demonstrating trends toward improved mortality.⁷ The ACG guideline was released prior to mounting evidence suggesting that goal-directed fluid therapy in sepsis does not improve mortality versus usual care.⁸ The growing uncertainty regarding the efficacy of goal-directed fluids for septic shock, as well limitations of studies on AP, may contribute to the differences between the AGA and ACG recommendations.

Finally, as the guideline covers the initial therapeutic management of AP, no recommendations are made for diagnostic studies such as right upper quadrant ultrasound. This noninvasive and readily available test plays a critical role in evaluating for presence of gallstones and other potential etiologies of abdominal pain.

AREAS IN NEED OF FUTURE STUDY

Additional research is needed to better understand goal-directed fluid therapy with respect to the fluid type, amount, and target outcomes. Similarly, determining the optimal enteral feeding regimens for patients failing oral intake would help clinicians meet the recommendation for early nutrition. Finally, clarification on the roles and timing of endoscopic and surgical procedures for patients with severe biliary pancreatitis, as well as geriatric and medically complex populations, would help hospitalists advocate for a multidisciplinary approach to this common and often serious disease.

Disclosures

The authors have nothing to disclose.

Acute pancreatitis (AP) is the most common gastrointestinal discharge diagnosis in the United States, with a mortality rate of 1%-5%.¹ Recent data demonstrate increasing AP-related admissions, making AP management of utmost importance to hospitalists.¹ The American Gastroenterological Association (AGA) guideline specifically addresses AP management in the initial 48-72 hours of admission, during which management decisions can alter disease course and length of stay. AP requires two of the following three criteria for diagnosis: characteristic abdominal pain, elevation of lipase or amylase ≥3 times the upper limit of normal, and/or radiographic evidence of pancreatitis on cross-sectional imaging. The guideline provides eight recommendations, which we consolidated to highlight practice changing recommendations: fluids, nutrition, management of the most common causes, and prophylactic antibiotics.^2,3

KEY RECOMMENDATIONS FOR THE HOSPITALIST

Fluids

Recommendation 1. In patients with AP, use goal-directed isotonic crystalloids for fluid management (conditional recommendation, very low-quality evidence).

The guideline emphasizes goal-directed fluid management despite low-quality, heterogeneous evidence and does not recommend Ringer’s lactate over normal saline. “Goal-directed” fluid management involves the use of crystalloid infusions titrated to improve physiologic and biochemical markers, but no target volume is specified by the guideline. Frequent reassessments should look for signs of volume overload, the primary risk of harm with fluid therapy. Despite failure to reduce mortality or morbidities such as pancreatic necrosis or persistent multi-organ failure, the AGA cites the mortality benefit of goal-directed therapy in sepsis as justification for this approach in AP, given the similar physiologic abnormalities.

Nutrition

Recommendation 2. Begin feeding early in patients with AP regardless of predicted severity. If oral nutrition is not tolerated, enteral feeding with either a nasogastric or nasojejunal tube is preferred to parenteral nutrition (strong recommendation, moderate-quality evidence).

Early feeding (ie, within 24 hours) is recommended regardless of AP severity. This represents a change from prior practices of bowel rest, theorized to prevent continued stimulation of an inflamed pancreas. Although early feeding has not been linked to improved mortality, it has demonstrated lower rates of multi-organ failure and infected pancreatic necrosis, possibly due to maintenance of the gut mucosal barrier and reduced bacterial translocation. When oral feeding is not tolerated, enteral nutrition is preferred over parenteral nutrition due to less risks. The preferred dietary composition guidance for patients with persistent pain or ileus is not addressed.

Management of the Most Common Causes of AP in Adults

Recommendation 3. Patients with mild acute biliary pancreatitis should have cholecystectomy during the initial admission (strong recommendation, moderate-quality evidence).

All patients with suspected biliary pancreatitis should receive a surgical consultation for cholecystectomy during the index admission. At the time of the guideline release, only one trial was available to support the recommendation of early cholecystectomy; however, newer studies similarly support cholecystectomy during index admission by demonstrating reductions in composite outcomes of mortality and gallstone-related complications, readmission for pancreatitis, and other pancreatobiliary complications.⁴ A Cochrane review included in the guideline found no differences in complication rates even in patients with severe biliary pancreatitis. In the absence of cholangitis, urgent endoscopic retrograde cholangiography (ERCP) is not indicated as most stones causing biliary pancreatitis pass spontaneously.

Recommendation 4. In patients with acute alcoholic pancreatitis, brief alcohol intervention should occur during admission (strong recommendation, moderate-quality evidence).

Ongoing alcohol consumption is a risk factor for recurrent acute and chronic pancreatitis. Only one trial assessed the impact of inpatient alcohol cessation counseling on recurrent AP, noting a trend toward reduced readmissions.⁵ However, indirect evidence from similar interventions in ambulatory settings demonstrates reductions in alcohol intake, leading to the AGA recommendation for inpatients with alcohol-induced AP.³

Antibiotics

Recommendation 5. Avoid empiric antibiotics in patients with AP who otherwise lack an indication, regardless of predicted severity (conditional recommendation, low-quality evidence).

Since 2002, well performed trials have consistently failed to demonstrate improvement in outcomes such as multi-organ failure or length of stay with use of prophylactic antibiotics for AP, even severe AP and pancreatic necrosis. Therefore, the AGA recommends against prophylactic antibiotics in initial management of AP regardless of disease severity. Lack of blinding in the majority of trial designs conducted before 2002 contributed to the overall assessment of low-quality evidence. The guideline does not address acute biliary pancreatitis with cholangitis, for which antibiotics and ERCP for decompression are critical.

CRITIQUE

The AGA Institute supported this guideline development and employed the rigorous and standardized GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology. This approach allowed the guideline panel members to account not only for evidence quality, but also the benefits and harms of an intervention and resource utilization. None of the authors had any stated conflicts of interest.

The guideline heavily weighted results from randomized control trials, most of which excluded key populations cared for by hospitalists (eg, patients older than 75 years, with end-stage renal disease). Particular areas where this creates challenges for clinicians and patients alike include goal-directed fluid therapy and when to consider more invasive interventions such as ERCP and early cholecystectomy. For example, patients considered to be poor surgical candidates may benefit from ERCP with biliary sphincterotomy to reduce the risk of recurrent biliary pancreatitis.

Lack of specificity in the guidelines for goal-directed fluid management and enteral feeding regimens makes it challenging to standardize hospitalists’ approach to the early care of patients with AP. Interestingly, the 2013 American College of Gastroenterology (ACG) Guideline for the Management of AP included strong recommendations for the use of Ringer’s lactate and volume targets in the initial management of AP.⁶ Evidence supporting the use of Ringer’s lactate versus normal saline is based largely upon improved inflammatory markers, theoretical potentiation of pancreatic enzyme activation with hypercholemic metabolic acidosis, and small studies demonstrating trends toward improved mortality.⁷ The ACG guideline was released prior to mounting evidence suggesting that goal-directed fluid therapy in sepsis does not improve mortality versus usual care.⁸ The growing uncertainty regarding the efficacy of goal-directed fluids for septic shock, as well limitations of studies on AP, may contribute to the differences between the AGA and ACG recommendations.

Finally, as the guideline covers the initial therapeutic management of AP, no recommendations are made for diagnostic studies such as right upper quadrant ultrasound. This noninvasive and readily available test plays a critical role in evaluating for presence of gallstones and other potential etiologies of abdominal pain.

AREAS IN NEED OF FUTURE STUDY

Additional research is needed to better understand goal-directed fluid therapy with respect to the fluid type, amount, and target outcomes. Similarly, determining the optimal enteral feeding regimens for patients failing oral intake would help clinicians meet the recommendation for early nutrition. Finally, clarification on the roles and timing of endoscopic and surgical procedures for patients with severe biliary pancreatitis, as well as geriatric and medically complex populations, would help hospitalists advocate for a multidisciplinary approach to this common and often serious disease.

Disclosures

The authors have nothing to disclose.

Given the breadth and depth of patients cared for by hospital medicine providers, it is challenging to remain current with the literature. The authors critically appraised the literature from March 2018 to April 2019 for high-quality studies relevant to hospital medicine. Articles were selected based on methodologic rigor and likelihood to impact clinical practice. Thirty articles were selected by the presenting authors for the Hospital Medicine Updates at the 2019 Society of Hospital Medicine (CH, CM) and Society of General Internal Medicine Annual Meetings (BS, AB). After two sequential rounds of voting and group discussion to adjudicate voting discrepancies, the authors selected the 10 most impactful articles for this review. Each article is described below with the key points summarized in the Table.

Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). McDonald LC, et al. Clin Infect Dis. 2018;66(7):e1–e48.¹

Background. In the United States, approximately 500,000 Clostridioides difficile infections (CDI) occur annually with 15,000-30,000 deaths. CDI has become a marker of hospital quality and has been placed under numerous “pay for performance” metrics. The Infectious Diseases Society of America/Society of Healthcare Epidemiology of America updated their guidelines from 2010 regarding hospital surveillance, diagnostic testing, treatment, and infection precautions and control.

Findings. The panel included 14 multidisciplinary experts in epidemiology, diagnosis, infection control, and clinical management of adult and pediatric CDI. They used problem intervention comparison-outcome (PICO)-formatted, evidence-based questions. The selection of data and final recommendations were made in accordance with the GRADE criteria. A total of 35 recommendations were made.

Key clinical recommendations for hospitalists caring for adults: (1) Prescribe vancomycin or fidaxomicin over metronidazole for the initial treatment of CDI (strong recommendation, high quality of evidence); (2) Limit testing to the patients with unexplained new onset diarrhea, which is defined as greater than or equal to 3 unformed stools in 24 hours (weak recommendation, very low-quality evidence); (3) Avoid routine repeat testing within seven days, and only test asymptomatic patients for epidemiologic reasons (strong recommendation, moderate-quality evidence); (4) Minimize the frequency and duration of high-risk antibiotic therapy and the number of antibiotic agents prescribed (strong recommendation, moderate quality of evidence); (5) Discontinue therapy with the inciting antibiotic agent as soon as possible (strong recommendation, moderate quality of evidence).

Caveats. As with the clinical application of any guidelines, individual case adjustments may be required.

Mortality and Morbidity in Acutely Ill Adults Treated with Liberal versus Conservative Oxygen Therapy (IOTA): a Systematic Review and Meta-analysis. Chu DK, et al. Lancet. 2018;391(10131):1693-1705.²

Background. Supplemental oxygen is often given to acutely ill hospitalized adults, even when they are not hypoxic or dyspneic. The safety and efficacy of this practice is unknown.

Findings. This systematic review and meta-analysis evaluated 25 randomized controlled trials enrolling 16,037 patients. Patients presented with several conditions, including sepsis, critical illness, stroke, myocardial infarction, and emergency surgery. The fraction of inspired oxygen in the liberal arms varied from 30% to 100%. Most patients randomized to the conservative arm received no supplemental oxygen. Delivery of liberal oxygen to acutely ill adults was associated with increased in-hospital mortality (relative risk [RR]: 1.21; 95% CI: 1.03-1.43), 30-day mortality (RR: 1.14; 95% CI: 1.01-1.29), and 90-day mortality (RR: 1.10; 95% CI: 1.00-1.20). The results were believed to be of high quality and were robust across multiple sensitivity analyses. It seemed that the mortality began to increase when supplemental oxygen raised the peripheral oxygen saturation (Sp0₂) above a range of 94%-96%.

Caveats. Heterogeneity was observed in the study settings and oxygen delivery. In addition, the cause for increased mortality could not be determined.

Implications. In hospitalized acutely ill adults, “liberal” supplemental oxygen was associated with increased in-hospital and longer-term mortality. The study authors postulated that this finding resulted from the direct toxic effects of oxygen or that oxygen delivery may “mask” illness and lead to delays in diagnosis and treatment. A subsequent clinical practice guideline recommends (1) a target SpO₂ of less than 96% for patients receiving oxygen therapy; (2) a target SpO₂ range of 90%-94% seems appropriate for most hospitalized adults.³

Do Words Matter? Stigmatizing Language and the Transmission of Bias in the Medical Record. P Goddu A, et al. J Gen Intern Med. 2018;33(5):68-91.⁴

Background. Previous work has shown that clinician bias affects health outcomes, often worsening health disparities. It is unknown whether clinicians’ language in medical records biases other clinicians and whether this affects patients.

Findings. The investigators randomized medical students and residents in internal and emergency medicine at one academic medical center to review one of two vignettes in the format of notes on the same hypothetical patient with sickle cell disease (SCD) admitted with a pain crisis. One vignette contained stigmatizing language, and the other contained neutral language. The trainees exposed to the vignettes with stigmatizing language showed a more negative attitude toward the patient, as measured by a previously validated scale of attitudes toward patients with SCD (20.6 stigmatizing vs 25.6 neutral, with a total score range of 7-35 for the instrument; higher scores indicate more positive attitudes; P < .001). Furthermore, the intensity of pain treatment was assessed in the resident group and was less aggressive when residents were exposed to stigmatizing language (5.56 stigmatizing vs 6.22 neutral on a scale of 2-7, with higher scores indicating more aggressive pain treatment; P = .003).

Cautions. This research was a single-center study of residents and medical students in two departments. Additionally, the study used vignettes on a hypothetical patient so trainees in the study group might have witnessed stronger stigmatizing language than what is typically observed in an actual patients’ notes.

Implications. Stigmatizing language used in medical records possibly contributed to health disparities by negatively impacting other physicians’ biases and prescribing practices toward patients with SCD at an academic medical center. Clinicians should avoid stigmatizing language in medical records.

Catheter Ablation for Atrial Fibrillation with Heart Failure. Marrouche, NF et al. New Engl J Med. 2018;378:417-427.⁵

Background. Atrial fibrillation (AF) in patients with heart failure is associated with increased mortality and morbidity. Small-scale studies have suggested that ablation of AF may benefit patients with heart failure.

Findings. This multicenter trial included 398 patients with heart failure and symptomatic AF. Patients had New York Heart Association Class II-IV heart failure, an ejection fraction (EF) of 35% or less, and an internal cardiac defibrillator (ICD). Patients were randomized to either ablation or medical therapy. All enrolled patients either refused, failed, or showed poor tolerance to antiarrhythmic therapy for AF. The primary outcome was death from any cause or hospitalization for heart failure.

The composite endpoint occurred in 28.5% of the ablation group versus 44.6% of patients in the medical therapy group (hazard ratio [HR]: 0.62; 95% CI: 0.43-0.87). Fewer patients in the ablation group died (13% vs 25%; HR: 0.53; 95% CI: 0.32-0.86) or were hospitalized for heart failure (21% vs 36%; HR: 0.56; 95% CI: 0.37-0.83). The patients in the ablation group had higher EF increases above baseline and a greater proportion were in sinus rhythm at the 60-month follow-up visit.

Cautions. The trial was terminated early due to slow recruitment and lower than expected events. Over twice as many patients were lost to follow-up in the ablation group versus the medical therapy group, and by 60 months, AF recurred in 50% of patients who underwent ablation. The sample size was small, and the trial was unblinded.

Implications. Ablation should be considered for AF in patients with heart failure. Additional studies to evaluate ablation versus medical therapy for patients with heart failure and AF are underway.

Medication for Opioid Use Disorder after Nonfatal Opioid Overdose and Association with Mortality. Larochelle MR, et al. Ann Intern Med. 2018;169(3):137-145.⁶

Background. More than 70,000 Americans died of drug overdose in 2017; this number is higher than the deaths resulting from human immunodeficiency virus, car crash, or gun violence at their peaks.⁷ Methadone, buprenorphine, and naltrexone are approved by the Federal Drug Administration for the treatment of opioid use disorder (OUD). These medications increase treatment retention; methadone and buprenorphine have been associated with significant decreases in all-cause and overdose mortality.⁸ However, whether receipt of these medications following a nonfatal opioid overdose reduces mortality is unknown.

Findings. This retrospective cohort study included 17,568 opioid overdose survivors from the Massachusetts’s Public Health Dataset between 2012 and 2014. Only three in 10 of these patients received any medications for OUD over 12 months following overdose. All-cause mortality was 4.7 deaths (95% CI: 4.4-5.0 deaths) per 100 person-years. The relative risk for all-cause mortality was 53% lower with methadone (adjusted hazard ratio [aHR]: 0.47; 95% CI: 0.32-0.71) and 37% lower with buprenorphine (aHR: 0.63; 95% CI: 0.46-0.87).

Caveats. This cohort study may have missed confounders explaining why certain patients received medications for OUD. As a result, association cannot be interpreted as causation.

Implications. Methadone and buprenorphine are associated with a reduction in preventable deaths in patients with OUD who have survived an overdose. All patients with OUD should be considered for therapy.

Outcomes Associated with Apixaban Use in Patients with End-Stage Kidney Disease and Atrial Fibrillation in the United States. Siontis, KC, et al. Circulation. 2018;138:1519–1529.⁹

Background. Patients with end-stage kidney disease (ESKD) have poor outcomes when treated with warfarin for AF. These patients were excluded from clinical trials of direct oral anticoagulants. The goal of this study was to determine the outcomes of the use of apixaban in patients with ESKD and AF.

Findings. This retrospective cohort study included 25,523 Medicare patients with ESKD and AF on anticoagulants. A 3:1 propensity score match was performed between patients on warfarin and apixaban. Time without stroke/systemic embolism, bleeding (major, gastrointestinal, and intracranial), and death were assessed. A total of 2,351 patients were on apixaban, and 23,172 patients were on warfarin. No difference was observed in the risk of stroke/systemic embolism between apixaban and warfarin (HR 0.88; 95% CI: 0.69-1.12). Apixaban was associated with a lower risk of major bleeding (HR: 0.72; 95% CI: 0.59-0.87). Standard-dose apixaban (5 mg twice a day) was associated with lower risks of stroke/systemic embolism and death compared with reduced-dose apixaban (2.5 mg twice a day; n = 1,317; HR: 0.61; 95% CI: 0.37-0.98; P = .04 for stroke/systemic embolism; HR: 0.64; 95% CI: 0.45-0.92; P = .01 for death) or warfarin (HR: 0.64; 95% CI: 0.42-0.97; P = .04 for stroke/systemic embolism; HR: 0.63; 95% CI: 0.46-0.85; P = .003 for death).

Cautions. There may be unique patient factors that led providers to prescribe apixaban to patients with ESKD.

Implications. The use of standard-dose apixaban appears safe and potentially preferable in patients with ESKD and AF due to reductions in major bleeding, thromboembolism, and mortality risk compared with warfarin. Several additional studies are pending to evaluate the use and dose of apixaban in patients with ESKD and AF.

Outcomes Associated with De-escalating Therapy for Methicillin-Resistant Staphylococcus aureus in Culture-Negative Nosocomial Pneumonia. Cowley MC, et al. Chest. 2019;155(1):53-59.¹⁰

Background. Patients diagnosed with hospital-acquired pneumonia (HAP) are often treated empirically with broad-spectrum antibiotics. In many patients with HAP, cultures remain negative, and providers must decide if antibiotics can safely be narrowed. Specifically, the safety of deciding to “de-escalate” and discontinue the coverage for methicillin-resistant Staphylococcus aureus (MRSA) if cultures remain negative is unclear.

Findings. In this single-center retrospective cohort study, 279 patients who were (1) diagnosed with HAP and (2) had negative sputum cultures were enrolled. The patients in whom MRSA coverage was de-escalated by day four were compared with those with continued anti-MRSA coverage. No difference was observed between the two groups in terms of degree of illness or comorbidities. The patients who were de-escalated received five fewer days of anti-MRSA coverage than patients who were not. No difference was noted in the 28-day mortality between the two groups (de-escalation: 23% vs no de-escalation: 28%; 95% CI: −16.1%-6.5%). The incidence of acute kidney injury (AKI) was significantly lower in the de-escalation group (36% vs 50%; 95% CI: −26.9- 0.04), and the overall length of stay was five days shorter in the de-escalation group (95% CI: 0.1-6.4 days).

Caveats. Given the retrospective nature, unmeasured confounders may have impacted the decision to de-escalate anti-MRSA coverage. The observed lower risk of AKI in the de-escalation group may be due to the simultaneous de-escalation of anti-Pseudomonas antibiotic agents in addition to the de-escalation of anti-MRSA coverage, as opposed to de-escalation of the anti-MRSA coverage alone.

Implications. De-escalation of anti-MRSA coverage in patients with HAP with negative cultures is associated with fewer antibiotic days, less AKI, and possibly shorter length of stay.

Partial Oral versus Intravenous Antibiotic Treatment for Endocarditis (POET). Iversen K et al. New Engl J Med. 2019;380(5):415-424.¹¹

Background. Patients with left-sided infective endocarditis are typically treated with up to six weeks of intravenous (IV) antibiotics. The investigators studied the effectiveness and safety of switching to oral antibiotics after at least 10 days of IV therapy.

Findings. This randomized, multicenter, noninferiority trial at cardiac centers across Denmark included 400 adults with left-sided endocarditis who were clinically stable after at least 10 days of IV antibiotics. Half of the patients were randomized to continue IV therapy, whereas the other half was switched to oral antibiotics to complete the treatment course. Six months after therapy, no significant difference was observed between the two groups in terms of the primary composite outcomes, including all-cause mortality, unplanned cardiac surgery, embolic events, or relapse of bacteremia with the primary pathogen (IV-treated group: 12.1%; orally treated group: 9.0% [between-group difference: 3.1%; P = .40]).

Caveats. A total of 20% of the screened population (1,954 adults) was randomized, and about 1% (5/400) of patients used injection drugs. None of the patients had MRSA. Patients in the oral group were assessed two to three times per week as outpatients, which may not be feasible in most settings.

Implications. Switching to oral antibiotics after at least 10 days of IV therapy appears to be safe and effective in selected patients with left-sided endocarditis. However, this study largely excluded patients with injection drug use and/or MRSA infections.

Oral versus Intravenous Antibiotics for Bone and Joint Infection (OVIVA). Li HK, et al. New Engl J Med. 2019;380(5):425-436.¹²

Background. Most complex orthopedic infections are treated with several weeks of IV antibiotics. This study sought to determine whether oral antibiotics are noninferior to IV antibiotics for bone and joint infections.

Findings. This randomized, multicenter, noninferiority, open-label trial of 1,054 adults with bone and joint infections in the United Kingdom included patients with prosthetic joints, other indwelling joint hardware, and native joint infections. Within seven days of antibiotic medication or within seven days of surgery (if performed), the patients received either IV or oral antibiotics for six weeks with a primary endpoint of treatment failure one year after the study randomization. The choice and duration of antibiotic treatment were determined by the involved infectious disease physician. A majority (77%) of patients received greater than six weeks of therapy. Treatment failure was defined by clinical, microbiologic, or histologic criteria. Most enrolled patients were infected with Staphylococcus aureus, with 10% having methicillin-resistant S. aureus. Treatment failure was more frequent in the IV group than the oral group (14.6% vs 13.2%), and these findings were consistent across all subgroups. More patients discontinued treatment in the IV group than the oral group.

Cautions. This study included a heterogenous population of patients with bone and joint infections, with or without hardware, and with different species of bacteria. Patients with bacteremia, endocarditis, or another indication for IV therapy were excluded. Limited injection drug use history was available for the enrolled patients. Most patients had lower limb infections. Thus, these findings are less applicable to vertebral osteomyelitis. Additionally, the study offered no comparison of specific antibiotics.

Implications. With appropriate oversight from infectious disease specialists, targeted oral therapy may be appropriate for the treatment of osteomyelitis. This shift in practice likely requires more study before broad implementation.

Prognostic Accuracy of the HEART Score for Prediction of Major Adverse Cardiac Events in Patients Presenting with Chest Pain: A Systematic Review and Meta‐analysis. Fernando S, et al. Acad Emerg Med. 2019;26(2):140-151.¹³

Background. Chest pain accounts for over eight million emergency department (ED) visits yearly in the United States. Of those presenting with chest pain, 10%-20% will experience acute coronary syndrome (ACS) requiring further medical treatment. Given the fear of missing ACS, many low-risk patients are hospitalized. The American Heart Association has advocated using validated predictive scoring models to identify patients with chest pain who are at low risk for short-term major cardiovascular adverse event (MACE) for potential discharge without further testing. The authors evaluated the prognostic accuracy of higher risk scores to predict MACE in adult ED patients presenting with chest pain.

Findings. This study was a systematic review and meta-analysis of 30 prospective and retrospective studies evaluating the history–electrocardiogram–age–risk factors–troponin (HEART) score through May 1, 2018. Meta-analysis compared the sensitivity, specificity, positive likelihood ratios, negative likelihood ratios, and diagnostic odds ratios of the HEART score and the Thrombolysis in Myocardial Infarction (TIMI) score when reported. An intermediate HEART score of 4-6 had a sensitivity of 95.9% and a specificity of 44.6%. A high HEART score of greater than or equal to 7 had a sensitivity of 39.5% and a specificity of 95.0%. Similarly, a high TIMI score of great than or equal to 6 had a sensitivity of only 2.8% and a specificity of 99.6%. The authors concluded that a HEART score of greater than or equal to 4 best identifies patients at risk of MACE who need greater consideration for additional testing.

Caveats. This meta-analysis failed to assess the potential adverse effects of false positive downstream testing. Additionally, no study compared the HEART score with the experienced clinician gestalt, which has often been equivalent to decision rules.

Implication. A HEART score greater than or equal to 4 risk stratifies ED patients with chest pain requiring further consideration for evaluation versus those that can be discharged with low risk for short-term MACE.

Given the breadth and depth of patients cared for by hospital medicine providers, it is challenging to remain current with the literature. The authors critically appraised the literature from March 2018 to April 2019 for high-quality studies relevant to hospital medicine. Articles were selected based on methodologic rigor and likelihood to impact clinical practice. Thirty articles were selected by the presenting authors for the Hospital Medicine Updates at the 2019 Society of Hospital Medicine (CH, CM) and Society of General Internal Medicine Annual Meetings (BS, AB). After two sequential rounds of voting and group discussion to adjudicate voting discrepancies, the authors selected the 10 most impactful articles for this review. Each article is described below with the key points summarized in the Table.

Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). McDonald LC, et al. Clin Infect Dis. 2018;66(7):e1–e48.¹

Background. In the United States, approximately 500,000 Clostridioides difficile infections (CDI) occur annually with 15,000-30,000 deaths. CDI has become a marker of hospital quality and has been placed under numerous “pay for performance” metrics. The Infectious Diseases Society of America/Society of Healthcare Epidemiology of America updated their guidelines from 2010 regarding hospital surveillance, diagnostic testing, treatment, and infection precautions and control.

Findings. The panel included 14 multidisciplinary experts in epidemiology, diagnosis, infection control, and clinical management of adult and pediatric CDI. They used problem intervention comparison-outcome (PICO)-formatted, evidence-based questions. The selection of data and final recommendations were made in accordance with the GRADE criteria. A total of 35 recommendations were made.

Key clinical recommendations for hospitalists caring for adults: (1) Prescribe vancomycin or fidaxomicin over metronidazole for the initial treatment of CDI (strong recommendation, high quality of evidence); (2) Limit testing to the patients with unexplained new onset diarrhea, which is defined as greater than or equal to 3 unformed stools in 24 hours (weak recommendation, very low-quality evidence); (3) Avoid routine repeat testing within seven days, and only test asymptomatic patients for epidemiologic reasons (strong recommendation, moderate-quality evidence); (4) Minimize the frequency and duration of high-risk antibiotic therapy and the number of antibiotic agents prescribed (strong recommendation, moderate quality of evidence); (5) Discontinue therapy with the inciting antibiotic agent as soon as possible (strong recommendation, moderate quality of evidence).

Caveats. As with the clinical application of any guidelines, individual case adjustments may be required.

Mortality and Morbidity in Acutely Ill Adults Treated with Liberal versus Conservative Oxygen Therapy (IOTA): a Systematic Review and Meta-analysis. Chu DK, et al. Lancet. 2018;391(10131):1693-1705.²

Background. Supplemental oxygen is often given to acutely ill hospitalized adults, even when they are not hypoxic or dyspneic. The safety and efficacy of this practice is unknown.

Findings. This systematic review and meta-analysis evaluated 25 randomized controlled trials enrolling 16,037 patients. Patients presented with several conditions, including sepsis, critical illness, stroke, myocardial infarction, and emergency surgery. The fraction of inspired oxygen in the liberal arms varied from 30% to 100%. Most patients randomized to the conservative arm received no supplemental oxygen. Delivery of liberal oxygen to acutely ill adults was associated with increased in-hospital mortality (relative risk [RR]: 1.21; 95% CI: 1.03-1.43), 30-day mortality (RR: 1.14; 95% CI: 1.01-1.29), and 90-day mortality (RR: 1.10; 95% CI: 1.00-1.20). The results were believed to be of high quality and were robust across multiple sensitivity analyses. It seemed that the mortality began to increase when supplemental oxygen raised the peripheral oxygen saturation (Sp0₂) above a range of 94%-96%.

Caveats. Heterogeneity was observed in the study settings and oxygen delivery. In addition, the cause for increased mortality could not be determined.

Implications. In hospitalized acutely ill adults, “liberal” supplemental oxygen was associated with increased in-hospital and longer-term mortality. The study authors postulated that this finding resulted from the direct toxic effects of oxygen or that oxygen delivery may “mask” illness and lead to delays in diagnosis and treatment. A subsequent clinical practice guideline recommends (1) a target SpO₂ of less than 96% for patients receiving oxygen therapy; (2) a target SpO₂ range of 90%-94% seems appropriate for most hospitalized adults.³

Do Words Matter? Stigmatizing Language and the Transmission of Bias in the Medical Record. P Goddu A, et al. J Gen Intern Med. 2018;33(5):68-91.⁴

Background. Previous work has shown that clinician bias affects health outcomes, often worsening health disparities. It is unknown whether clinicians’ language in medical records biases other clinicians and whether this affects patients.

Findings. The investigators randomized medical students and residents in internal and emergency medicine at one academic medical center to review one of two vignettes in the format of notes on the same hypothetical patient with sickle cell disease (SCD) admitted with a pain crisis. One vignette contained stigmatizing language, and the other contained neutral language. The trainees exposed to the vignettes with stigmatizing language showed a more negative attitude toward the patient, as measured by a previously validated scale of attitudes toward patients with SCD (20.6 stigmatizing vs 25.6 neutral, with a total score range of 7-35 for the instrument; higher scores indicate more positive attitudes; P < .001). Furthermore, the intensity of pain treatment was assessed in the resident group and was less aggressive when residents were exposed to stigmatizing language (5.56 stigmatizing vs 6.22 neutral on a scale of 2-7, with higher scores indicating more aggressive pain treatment; P = .003).

Cautions. This research was a single-center study of residents and medical students in two departments. Additionally, the study used vignettes on a hypothetical patient so trainees in the study group might have witnessed stronger stigmatizing language than what is typically observed in an actual patients’ notes.

Implications. Stigmatizing language used in medical records possibly contributed to health disparities by negatively impacting other physicians’ biases and prescribing practices toward patients with SCD at an academic medical center. Clinicians should avoid stigmatizing language in medical records.

Catheter Ablation for Atrial Fibrillation with Heart Failure. Marrouche, NF et al. New Engl J Med. 2018;378:417-427.⁵

Background. Atrial fibrillation (AF) in patients with heart failure is associated with increased mortality and morbidity. Small-scale studies have suggested that ablation of AF may benefit patients with heart failure.

Findings. This multicenter trial included 398 patients with heart failure and symptomatic AF. Patients had New York Heart Association Class II-IV heart failure, an ejection fraction (EF) of 35% or less, and an internal cardiac defibrillator (ICD). Patients were randomized to either ablation or medical therapy. All enrolled patients either refused, failed, or showed poor tolerance to antiarrhythmic therapy for AF. The primary outcome was death from any cause or hospitalization for heart failure.

The composite endpoint occurred in 28.5% of the ablation group versus 44.6% of patients in the medical therapy group (hazard ratio [HR]: 0.62; 95% CI: 0.43-0.87). Fewer patients in the ablation group died (13% vs 25%; HR: 0.53; 95% CI: 0.32-0.86) or were hospitalized for heart failure (21% vs 36%; HR: 0.56; 95% CI: 0.37-0.83). The patients in the ablation group had higher EF increases above baseline and a greater proportion were in sinus rhythm at the 60-month follow-up visit.

Cautions. The trial was terminated early due to slow recruitment and lower than expected events. Over twice as many patients were lost to follow-up in the ablation group versus the medical therapy group, and by 60 months, AF recurred in 50% of patients who underwent ablation. The sample size was small, and the trial was unblinded.

Implications. Ablation should be considered for AF in patients with heart failure. Additional studies to evaluate ablation versus medical therapy for patients with heart failure and AF are underway.

Medication for Opioid Use Disorder after Nonfatal Opioid Overdose and Association with Mortality. Larochelle MR, et al. Ann Intern Med. 2018;169(3):137-145.⁶

Background. More than 70,000 Americans died of drug overdose in 2017; this number is higher than the deaths resulting from human immunodeficiency virus, car crash, or gun violence at their peaks.⁷ Methadone, buprenorphine, and naltrexone are approved by the Federal Drug Administration for the treatment of opioid use disorder (OUD). These medications increase treatment retention; methadone and buprenorphine have been associated with significant decreases in all-cause and overdose mortality.⁸ However, whether receipt of these medications following a nonfatal opioid overdose reduces mortality is unknown.

Findings. This retrospective cohort study included 17,568 opioid overdose survivors from the Massachusetts’s Public Health Dataset between 2012 and 2014. Only three in 10 of these patients received any medications for OUD over 12 months following overdose. All-cause mortality was 4.7 deaths (95% CI: 4.4-5.0 deaths) per 100 person-years. The relative risk for all-cause mortality was 53% lower with methadone (adjusted hazard ratio [aHR]: 0.47; 95% CI: 0.32-0.71) and 37% lower with buprenorphine (aHR: 0.63; 95% CI: 0.46-0.87).

Caveats. This cohort study may have missed confounders explaining why certain patients received medications for OUD. As a result, association cannot be interpreted as causation.

Implications. Methadone and buprenorphine are associated with a reduction in preventable deaths in patients with OUD who have survived an overdose. All patients with OUD should be considered for therapy.

Outcomes Associated with Apixaban Use in Patients with End-Stage Kidney Disease and Atrial Fibrillation in the United States. Siontis, KC, et al. Circulation. 2018;138:1519–1529.⁹

Background. Patients with end-stage kidney disease (ESKD) have poor outcomes when treated with warfarin for AF. These patients were excluded from clinical trials of direct oral anticoagulants. The goal of this study was to determine the outcomes of the use of apixaban in patients with ESKD and AF.

Findings. This retrospective cohort study included 25,523 Medicare patients with ESKD and AF on anticoagulants. A 3:1 propensity score match was performed between patients on warfarin and apixaban. Time without stroke/systemic embolism, bleeding (major, gastrointestinal, and intracranial), and death were assessed. A total of 2,351 patients were on apixaban, and 23,172 patients were on warfarin. No difference was observed in the risk of stroke/systemic embolism between apixaban and warfarin (HR 0.88; 95% CI: 0.69-1.12). Apixaban was associated with a lower risk of major bleeding (HR: 0.72; 95% CI: 0.59-0.87). Standard-dose apixaban (5 mg twice a day) was associated with lower risks of stroke/systemic embolism and death compared with reduced-dose apixaban (2.5 mg twice a day; n = 1,317; HR: 0.61; 95% CI: 0.37-0.98; P = .04 for stroke/systemic embolism; HR: 0.64; 95% CI: 0.45-0.92; P = .01 for death) or warfarin (HR: 0.64; 95% CI: 0.42-0.97; P = .04 for stroke/systemic embolism; HR: 0.63; 95% CI: 0.46-0.85; P = .003 for death).

Cautions. There may be unique patient factors that led providers to prescribe apixaban to patients with ESKD.

Implications. The use of standard-dose apixaban appears safe and potentially preferable in patients with ESKD and AF due to reductions in major bleeding, thromboembolism, and mortality risk compared with warfarin. Several additional studies are pending to evaluate the use and dose of apixaban in patients with ESKD and AF.

Outcomes Associated with De-escalating Therapy for Methicillin-Resistant Staphylococcus aureus in Culture-Negative Nosocomial Pneumonia. Cowley MC, et al. Chest. 2019;155(1):53-59.¹⁰

Background. Patients diagnosed with hospital-acquired pneumonia (HAP) are often treated empirically with broad-spectrum antibiotics. In many patients with HAP, cultures remain negative, and providers must decide if antibiotics can safely be narrowed. Specifically, the safety of deciding to “de-escalate” and discontinue the coverage for methicillin-resistant Staphylococcus aureus (MRSA) if cultures remain negative is unclear.

Findings. In this single-center retrospective cohort study, 279 patients who were (1) diagnosed with HAP and (2) had negative sputum cultures were enrolled. The patients in whom MRSA coverage was de-escalated by day four were compared with those with continued anti-MRSA coverage. No difference was observed between the two groups in terms of degree of illness or comorbidities. The patients who were de-escalated received five fewer days of anti-MRSA coverage than patients who were not. No difference was noted in the 28-day mortality between the two groups (de-escalation: 23% vs no de-escalation: 28%; 95% CI: −16.1%-6.5%). The incidence of acute kidney injury (AKI) was significantly lower in the de-escalation group (36% vs 50%; 95% CI: −26.9- 0.04), and the overall length of stay was five days shorter in the de-escalation group (95% CI: 0.1-6.4 days).

Caveats. Given the retrospective nature, unmeasured confounders may have impacted the decision to de-escalate anti-MRSA coverage. The observed lower risk of AKI in the de-escalation group may be due to the simultaneous de-escalation of anti-Pseudomonas antibiotic agents in addition to the de-escalation of anti-MRSA coverage, as opposed to de-escalation of the anti-MRSA coverage alone.

Implications. De-escalation of anti-MRSA coverage in patients with HAP with negative cultures is associated with fewer antibiotic days, less AKI, and possibly shorter length of stay.

Partial Oral versus Intravenous Antibiotic Treatment for Endocarditis (POET). Iversen K et al. New Engl J Med. 2019;380(5):415-424.¹¹

Background. Patients with left-sided infective endocarditis are typically treated with up to six weeks of intravenous (IV) antibiotics. The investigators studied the effectiveness and safety of switching to oral antibiotics after at least 10 days of IV therapy.

Findings. This randomized, multicenter, noninferiority trial at cardiac centers across Denmark included 400 adults with left-sided endocarditis who were clinically stable after at least 10 days of IV antibiotics. Half of the patients were randomized to continue IV therapy, whereas the other half was switched to oral antibiotics to complete the treatment course. Six months after therapy, no significant difference was observed between the two groups in terms of the primary composite outcomes, including all-cause mortality, unplanned cardiac surgery, embolic events, or relapse of bacteremia with the primary pathogen (IV-treated group: 12.1%; orally treated group: 9.0% [between-group difference: 3.1%; P = .40]).

Caveats. A total of 20% of the screened population (1,954 adults) was randomized, and about 1% (5/400) of patients used injection drugs. None of the patients had MRSA. Patients in the oral group were assessed two to three times per week as outpatients, which may not be feasible in most settings.

Implications. Switching to oral antibiotics after at least 10 days of IV therapy appears to be safe and effective in selected patients with left-sided endocarditis. However, this study largely excluded patients with injection drug use and/or MRSA infections.

Oral versus Intravenous Antibiotics for Bone and Joint Infection (OVIVA). Li HK, et al. New Engl J Med. 2019;380(5):425-436.¹²

Background. Most complex orthopedic infections are treated with several weeks of IV antibiotics. This study sought to determine whether oral antibiotics are noninferior to IV antibiotics for bone and joint infections.

Findings. This randomized, multicenter, noninferiority, open-label trial of 1,054 adults with bone and joint infections in the United Kingdom included patients with prosthetic joints, other indwelling joint hardware, and native joint infections. Within seven days of antibiotic medication or within seven days of surgery (if performed), the patients received either IV or oral antibiotics for six weeks with a primary endpoint of treatment failure one year after the study randomization. The choice and duration of antibiotic treatment were determined by the involved infectious disease physician. A majority (77%) of patients received greater than six weeks of therapy. Treatment failure was defined by clinical, microbiologic, or histologic criteria. Most enrolled patients were infected with Staphylococcus aureus, with 10% having methicillin-resistant S. aureus. Treatment failure was more frequent in the IV group than the oral group (14.6% vs 13.2%), and these findings were consistent across all subgroups. More patients discontinued treatment in the IV group than the oral group.

Cautions. This study included a heterogenous population of patients with bone and joint infections, with or without hardware, and with different species of bacteria. Patients with bacteremia, endocarditis, or another indication for IV therapy were excluded. Limited injection drug use history was available for the enrolled patients. Most patients had lower limb infections. Thus, these findings are less applicable to vertebral osteomyelitis. Additionally, the study offered no comparison of specific antibiotics.

Implications. With appropriate oversight from infectious disease specialists, targeted oral therapy may be appropriate for the treatment of osteomyelitis. This shift in practice likely requires more study before broad implementation.

Prognostic Accuracy of the HEART Score for Prediction of Major Adverse Cardiac Events in Patients Presenting with Chest Pain: A Systematic Review and Meta‐analysis. Fernando S, et al. Acad Emerg Med. 2019;26(2):140-151.¹³

Background. Chest pain accounts for over eight million emergency department (ED) visits yearly in the United States. Of those presenting with chest pain, 10%-20% will experience acute coronary syndrome (ACS) requiring further medical treatment. Given the fear of missing ACS, many low-risk patients are hospitalized. The American Heart Association has advocated using validated predictive scoring models to identify patients with chest pain who are at low risk for short-term major cardiovascular adverse event (MACE) for potential discharge without further testing. The authors evaluated the prognostic accuracy of higher risk scores to predict MACE in adult ED patients presenting with chest pain.

Findings. This study was a systematic review and meta-analysis of 30 prospective and retrospective studies evaluating the history–electrocardiogram–age–risk factors–troponin (HEART) score through May 1, 2018. Meta-analysis compared the sensitivity, specificity, positive likelihood ratios, negative likelihood ratios, and diagnostic odds ratios of the HEART score and the Thrombolysis in Myocardial Infarction (TIMI) score when reported. An intermediate HEART score of 4-6 had a sensitivity of 95.9% and a specificity of 44.6%. A high HEART score of greater than or equal to 7 had a sensitivity of 39.5% and a specificity of 95.0%. Similarly, a high TIMI score of great than or equal to 6 had a sensitivity of only 2.8% and a specificity of 99.6%. The authors concluded that a HEART score of greater than or equal to 4 best identifies patients at risk of MACE who need greater consideration for additional testing.

Caveats. This meta-analysis failed to assess the potential adverse effects of false positive downstream testing. Additionally, no study compared the HEART score with the experienced clinician gestalt, which has often been equivalent to decision rules.

Implication. A HEART score greater than or equal to 4 risk stratifies ED patients with chest pain requiring further consideration for evaluation versus those that can be discharged with low risk for short-term MACE.

Given the breadth and depth of patients cared for by hospital medicine providers, it is challenging to remain current with the literature. The authors critically appraised the literature from March 2018 to April 2019 for high-quality studies relevant to hospital medicine. Articles were selected based on methodologic rigor and likelihood to impact clinical practice. Thirty articles were selected by the presenting authors for the Hospital Medicine Updates at the 2019 Society of Hospital Medicine (CH, CM) and Society of General Internal Medicine Annual Meetings (BS, AB). After two sequential rounds of voting and group discussion to adjudicate voting discrepancies, the authors selected the 10 most impactful articles for this review. Each article is described below with the key points summarized in the Table.

Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). McDonald LC, et al. Clin Infect Dis. 2018;66(7):e1–e48.¹

Background. In the United States, approximately 500,000 Clostridioides difficile infections (CDI) occur annually with 15,000-30,000 deaths. CDI has become a marker of hospital quality and has been placed under numerous “pay for performance” metrics. The Infectious Diseases Society of America/Society of Healthcare Epidemiology of America updated their guidelines from 2010 regarding hospital surveillance, diagnostic testing, treatment, and infection precautions and control.

Findings. The panel included 14 multidisciplinary experts in epidemiology, diagnosis, infection control, and clinical management of adult and pediatric CDI. They used problem intervention comparison-outcome (PICO)-formatted, evidence-based questions. The selection of data and final recommendations were made in accordance with the GRADE criteria. A total of 35 recommendations were made.

Key clinical recommendations for hospitalists caring for adults: (1) Prescribe vancomycin or fidaxomicin over metronidazole for the initial treatment of CDI (strong recommendation, high quality of evidence); (2) Limit testing to the patients with unexplained new onset diarrhea, which is defined as greater than or equal to 3 unformed stools in 24 hours (weak recommendation, very low-quality evidence); (3) Avoid routine repeat testing within seven days, and only test asymptomatic patients for epidemiologic reasons (strong recommendation, moderate-quality evidence); (4) Minimize the frequency and duration of high-risk antibiotic therapy and the number of antibiotic agents prescribed (strong recommendation, moderate quality of evidence); (5) Discontinue therapy with the inciting antibiotic agent as soon as possible (strong recommendation, moderate quality of evidence).

Caveats. As with the clinical application of any guidelines, individual case adjustments may be required.

Mortality and Morbidity in Acutely Ill Adults Treated with Liberal versus Conservative Oxygen Therapy (IOTA): a Systematic Review and Meta-analysis. Chu DK, et al. Lancet. 2018;391(10131):1693-1705.²

Background. Supplemental oxygen is often given to acutely ill hospitalized adults, even when they are not hypoxic or dyspneic. The safety and efficacy of this practice is unknown.

Findings. This systematic review and meta-analysis evaluated 25 randomized controlled trials enrolling 16,037 patients. Patients presented with several conditions, including sepsis, critical illness, stroke, myocardial infarction, and emergency surgery. The fraction of inspired oxygen in the liberal arms varied from 30% to 100%. Most patients randomized to the conservative arm received no supplemental oxygen. Delivery of liberal oxygen to acutely ill adults was associated with increased in-hospital mortality (relative risk [RR]: 1.21; 95% CI: 1.03-1.43), 30-day mortality (RR: 1.14; 95% CI: 1.01-1.29), and 90-day mortality (RR: 1.10; 95% CI: 1.00-1.20). The results were believed to be of high quality and were robust across multiple sensitivity analyses. It seemed that the mortality began to increase when supplemental oxygen raised the peripheral oxygen saturation (Sp0₂) above a range of 94%-96%.

Caveats. Heterogeneity was observed in the study settings and oxygen delivery. In addition, the cause for increased mortality could not be determined.

Implications. In hospitalized acutely ill adults, “liberal” supplemental oxygen was associated with increased in-hospital and longer-term mortality. The study authors postulated that this finding resulted from the direct toxic effects of oxygen or that oxygen delivery may “mask” illness and lead to delays in diagnosis and treatment. A subsequent clinical practice guideline recommends (1) a target SpO₂ of less than 96% for patients receiving oxygen therapy; (2) a target SpO₂ range of 90%-94% seems appropriate for most hospitalized adults.³

Do Words Matter? Stigmatizing Language and the Transmission of Bias in the Medical Record. P Goddu A, et al. J Gen Intern Med. 2018;33(5):68-91.⁴

Background. Previous work has shown that clinician bias affects health outcomes, often worsening health disparities. It is unknown whether clinicians’ language in medical records biases other clinicians and whether this affects patients.

Findings. The investigators randomized medical students and residents in internal and emergency medicine at one academic medical center to review one of two vignettes in the format of notes on the same hypothetical patient with sickle cell disease (SCD) admitted with a pain crisis. One vignette contained stigmatizing language, and the other contained neutral language. The trainees exposed to the vignettes with stigmatizing language showed a more negative attitude toward the patient, as measured by a previously validated scale of attitudes toward patients with SCD (20.6 stigmatizing vs 25.6 neutral, with a total score range of 7-35 for the instrument; higher scores indicate more positive attitudes; P < .001). Furthermore, the intensity of pain treatment was assessed in the resident group and was less aggressive when residents were exposed to stigmatizing language (5.56 stigmatizing vs 6.22 neutral on a scale of 2-7, with higher scores indicating more aggressive pain treatment; P = .003).

Cautions. This research was a single-center study of residents and medical students in two departments. Additionally, the study used vignettes on a hypothetical patient so trainees in the study group might have witnessed stronger stigmatizing language than what is typically observed in an actual patients’ notes.

Implications. Stigmatizing language used in medical records possibly contributed to health disparities by negatively impacting other physicians’ biases and prescribing practices toward patients with SCD at an academic medical center. Clinicians should avoid stigmatizing language in medical records.

Catheter Ablation for Atrial Fibrillation with Heart Failure. Marrouche, NF et al. New Engl J Med. 2018;378:417-427.⁵

Background. Atrial fibrillation (AF) in patients with heart failure is associated with increased mortality and morbidity. Small-scale studies have suggested that ablation of AF may benefit patients with heart failure.

Findings. This multicenter trial included 398 patients with heart failure and symptomatic AF. Patients had New York Heart Association Class II-IV heart failure, an ejection fraction (EF) of 35% or less, and an internal cardiac defibrillator (ICD). Patients were randomized to either ablation or medical therapy. All enrolled patients either refused, failed, or showed poor tolerance to antiarrhythmic therapy for AF. The primary outcome was death from any cause or hospitalization for heart failure.

The composite endpoint occurred in 28.5% of the ablation group versus 44.6% of patients in the medical therapy group (hazard ratio [HR]: 0.62; 95% CI: 0.43-0.87). Fewer patients in the ablation group died (13% vs 25%; HR: 0.53; 95% CI: 0.32-0.86) or were hospitalized for heart failure (21% vs 36%; HR: 0.56; 95% CI: 0.37-0.83). The patients in the ablation group had higher EF increases above baseline and a greater proportion were in sinus rhythm at the 60-month follow-up visit.

Cautions. The trial was terminated early due to slow recruitment and lower than expected events. Over twice as many patients were lost to follow-up in the ablation group versus the medical therapy group, and by 60 months, AF recurred in 50% of patients who underwent ablation. The sample size was small, and the trial was unblinded.

Implications. Ablation should be considered for AF in patients with heart failure. Additional studies to evaluate ablation versus medical therapy for patients with heart failure and AF are underway.

Medication for Opioid Use Disorder after Nonfatal Opioid Overdose and Association with Mortality. Larochelle MR, et al. Ann Intern Med. 2018;169(3):137-145.⁶

Background. More than 70,000 Americans died of drug overdose in 2017; this number is higher than the deaths resulting from human immunodeficiency virus, car crash, or gun violence at their peaks.⁷ Methadone, buprenorphine, and naltrexone are approved by the Federal Drug Administration for the treatment of opioid use disorder (OUD). These medications increase treatment retention; methadone and buprenorphine have been associated with significant decreases in all-cause and overdose mortality.⁸ However, whether receipt of these medications following a nonfatal opioid overdose reduces mortality is unknown.

Findings. This retrospective cohort study included 17,568 opioid overdose survivors from the Massachusetts’s Public Health Dataset between 2012 and 2014. Only three in 10 of these patients received any medications for OUD over 12 months following overdose. All-cause mortality was 4.7 deaths (95% CI: 4.4-5.0 deaths) per 100 person-years. The relative risk for all-cause mortality was 53% lower with methadone (adjusted hazard ratio [aHR]: 0.47; 95% CI: 0.32-0.71) and 37% lower with buprenorphine (aHR: 0.63; 95% CI: 0.46-0.87).

Caveats. This cohort study may have missed confounders explaining why certain patients received medications for OUD. As a result, association cannot be interpreted as causation.

Implications. Methadone and buprenorphine are associated with a reduction in preventable deaths in patients with OUD who have survived an overdose. All patients with OUD should be considered for therapy.

Outcomes Associated with Apixaban Use in Patients with End-Stage Kidney Disease and Atrial Fibrillation in the United States. Siontis, KC, et al. Circulation. 2018;138:1519–1529.⁹

Background. Patients with end-stage kidney disease (ESKD) have poor outcomes when treated with warfarin for AF. These patients were excluded from clinical trials of direct oral anticoagulants. The goal of this study was to determine the outcomes of the use of apixaban in patients with ESKD and AF.

Findings. This retrospective cohort study included 25,523 Medicare patients with ESKD and AF on anticoagulants. A 3:1 propensity score match was performed between patients on warfarin and apixaban. Time without stroke/systemic embolism, bleeding (major, gastrointestinal, and intracranial), and death were assessed. A total of 2,351 patients were on apixaban, and 23,172 patients were on warfarin. No difference was observed in the risk of stroke/systemic embolism between apixaban and warfarin (HR 0.88; 95% CI: 0.69-1.12). Apixaban was associated with a lower risk of major bleeding (HR: 0.72; 95% CI: 0.59-0.87). Standard-dose apixaban (5 mg twice a day) was associated with lower risks of stroke/systemic embolism and death compared with reduced-dose apixaban (2.5 mg twice a day; n = 1,317; HR: 0.61; 95% CI: 0.37-0.98; P = .04 for stroke/systemic embolism; HR: 0.64; 95% CI: 0.45-0.92; P = .01 for death) or warfarin (HR: 0.64; 95% CI: 0.42-0.97; P = .04 for stroke/systemic embolism; HR: 0.63; 95% CI: 0.46-0.85; P = .003 for death).

Cautions. There may be unique patient factors that led providers to prescribe apixaban to patients with ESKD.

Implications. The use of standard-dose apixaban appears safe and potentially preferable in patients with ESKD and AF due to reductions in major bleeding, thromboembolism, and mortality risk compared with warfarin. Several additional studies are pending to evaluate the use and dose of apixaban in patients with ESKD and AF.

Outcomes Associated with De-escalating Therapy for Methicillin-Resistant Staphylococcus aureus in Culture-Negative Nosocomial Pneumonia. Cowley MC, et al. Chest. 2019;155(1):53-59.¹⁰

Background. Patients diagnosed with hospital-acquired pneumonia (HAP) are often treated empirically with broad-spectrum antibiotics. In many patients with HAP, cultures remain negative, and providers must decide if antibiotics can safely be narrowed. Specifically, the safety of deciding to “de-escalate” and discontinue the coverage for methicillin-resistant Staphylococcus aureus (MRSA) if cultures remain negative is unclear.

Findings. In this single-center retrospective cohort study, 279 patients who were (1) diagnosed with HAP and (2) had negative sputum cultures were enrolled. The patients in whom MRSA coverage was de-escalated by day four were compared with those with continued anti-MRSA coverage. No difference was observed between the two groups in terms of degree of illness or comorbidities. The patients who were de-escalated received five fewer days of anti-MRSA coverage than patients who were not. No difference was noted in the 28-day mortality between the two groups (de-escalation: 23% vs no de-escalation: 28%; 95% CI: −16.1%-6.5%). The incidence of acute kidney injury (AKI) was significantly lower in the de-escalation group (36% vs 50%; 95% CI: −26.9- 0.04), and the overall length of stay was five days shorter in the de-escalation group (95% CI: 0.1-6.4 days).

Caveats. Given the retrospective nature, unmeasured confounders may have impacted the decision to de-escalate anti-MRSA coverage. The observed lower risk of AKI in the de-escalation group may be due to the simultaneous de-escalation of anti-Pseudomonas antibiotic agents in addition to the de-escalation of anti-MRSA coverage, as opposed to de-escalation of the anti-MRSA coverage alone.

Implications. De-escalation of anti-MRSA coverage in patients with HAP with negative cultures is associated with fewer antibiotic days, less AKI, and possibly shorter length of stay.

Partial Oral versus Intravenous Antibiotic Treatment for Endocarditis (POET). Iversen K et al. New Engl J Med. 2019;380(5):415-424.¹¹

Background. Patients with left-sided infective endocarditis are typically treated with up to six weeks of intravenous (IV) antibiotics. The investigators studied the effectiveness and safety of switching to oral antibiotics after at least 10 days of IV therapy.

Findings. This randomized, multicenter, noninferiority trial at cardiac centers across Denmark included 400 adults with left-sided endocarditis who were clinically stable after at least 10 days of IV antibiotics. Half of the patients were randomized to continue IV therapy, whereas the other half was switched to oral antibiotics to complete the treatment course. Six months after therapy, no significant difference was observed between the two groups in terms of the primary composite outcomes, including all-cause mortality, unplanned cardiac surgery, embolic events, or relapse of bacteremia with the primary pathogen (IV-treated group: 12.1%; orally treated group: 9.0% [between-group difference: 3.1%; P = .40]).

Caveats. A total of 20% of the screened population (1,954 adults) was randomized, and about 1% (5/400) of patients used injection drugs. None of the patients had MRSA. Patients in the oral group were assessed two to three times per week as outpatients, which may not be feasible in most settings.

Implications. Switching to oral antibiotics after at least 10 days of IV therapy appears to be safe and effective in selected patients with left-sided endocarditis. However, this study largely excluded patients with injection drug use and/or MRSA infections.

Oral versus Intravenous Antibiotics for Bone and Joint Infection (OVIVA). Li HK, et al. New Engl J Med. 2019;380(5):425-436.¹²

Background. Most complex orthopedic infections are treated with several weeks of IV antibiotics. This study sought to determine whether oral antibiotics are noninferior to IV antibiotics for bone and joint infections.

Findings. This randomized, multicenter, noninferiority, open-label trial of 1,054 adults with bone and joint infections in the United Kingdom included patients with prosthetic joints, other indwelling joint hardware, and native joint infections. Within seven days of antibiotic medication or within seven days of surgery (if performed), the patients received either IV or oral antibiotics for six weeks with a primary endpoint of treatment failure one year after the study randomization. The choice and duration of antibiotic treatment were determined by the involved infectious disease physician. A majority (77%) of patients received greater than six weeks of therapy. Treatment failure was defined by clinical, microbiologic, or histologic criteria. Most enrolled patients were infected with Staphylococcus aureus, with 10% having methicillin-resistant S. aureus. Treatment failure was more frequent in the IV group than the oral group (14.6% vs 13.2%), and these findings were consistent across all subgroups. More patients discontinued treatment in the IV group than the oral group.

Cautions. This study included a heterogenous population of patients with bone and joint infections, with or without hardware, and with different species of bacteria. Patients with bacteremia, endocarditis, or another indication for IV therapy were excluded. Limited injection drug use history was available for the enrolled patients. Most patients had lower limb infections. Thus, these findings are less applicable to vertebral osteomyelitis. Additionally, the study offered no comparison of specific antibiotics.

Implications. With appropriate oversight from infectious disease specialists, targeted oral therapy may be appropriate for the treatment of osteomyelitis. This shift in practice likely requires more study before broad implementation.

Prognostic Accuracy of the HEART Score for Prediction of Major Adverse Cardiac Events in Patients Presenting with Chest Pain: A Systematic Review and Meta‐analysis. Fernando S, et al. Acad Emerg Med. 2019;26(2):140-151.¹³

Background. Chest pain accounts for over eight million emergency department (ED) visits yearly in the United States. Of those presenting with chest pain, 10%-20% will experience acute coronary syndrome (ACS) requiring further medical treatment. Given the fear of missing ACS, many low-risk patients are hospitalized. The American Heart Association has advocated using validated predictive scoring models to identify patients with chest pain who are at low risk for short-term major cardiovascular adverse event (MACE) for potential discharge without further testing. The authors evaluated the prognostic accuracy of higher risk scores to predict MACE in adult ED patients presenting with chest pain.

Findings. This study was a systematic review and meta-analysis of 30 prospective and retrospective studies evaluating the history–electrocardiogram–age–risk factors–troponin (HEART) score through May 1, 2018. Meta-analysis compared the sensitivity, specificity, positive likelihood ratios, negative likelihood ratios, and diagnostic odds ratios of the HEART score and the Thrombolysis in Myocardial Infarction (TIMI) score when reported. An intermediate HEART score of 4-6 had a sensitivity of 95.9% and a specificity of 44.6%. A high HEART score of greater than or equal to 7 had a sensitivity of 39.5% and a specificity of 95.0%. Similarly, a high TIMI score of great than or equal to 6 had a sensitivity of only 2.8% and a specificity of 99.6%. The authors concluded that a HEART score of greater than or equal to 4 best identifies patients at risk of MACE who need greater consideration for additional testing.

Caveats. This meta-analysis failed to assess the potential adverse effects of false positive downstream testing. Additionally, no study compared the HEART score with the experienced clinician gestalt, which has often been equivalent to decision rules.

Implication. A HEART score greater than or equal to 4 risk stratifies ED patients with chest pain requiring further consideration for evaluation versus those that can be discharged with low risk for short-term MACE.

As the VA moves toward increased utilization of non-VA care, it is crucial to understand and address the challenges of transitional care faced by dual-use veterans to provide high-quality care that improves healthcare outcomes.^7,11,12 The VA implemented a shift in policy from the Veterans Access, Choice, and Accountability Act of 2014 (Public Law 113-146; “Choice Act”) to the VA Maintaining Internal Systems and Strengthening Integrated Outside Networks (MISSION) Act beginning June 6, 2019.^13,14 Under the MISSION Act, veterans have more ways to access healthcare within the VA’s network and through approved non-VA medical providers in the community known as “community care providers.”¹⁵ This shift expanded the existing VA Choice Act of 2014, where the program allowed those veterans who are unable to schedule an appointment within 30 days of their preferred date or the clinically appropriate date, or on the basis of their place of residence, to elect to receive care from eligible non-VA healthcare entities or providers.^14,15 These efforts to better serve veterans by increasing non-VA care might present added care coordination challenges for patients and their providers when they seek care in the VA.

High-quality transitional care prevents poor outcomes such as hospital readmissions.^16-18 When communication and coordination across healthcare delivery systems are lacking, patients and their families often find themselves at risk for adverse events.^19,20 Past research shows that patients have fewer adverse events when they receive comprehensive postdischarge care, including instructions on medications and self-care, symptom recognition and management, and reminders to attend follow-up appointments.^17,21,22 Although researchers have identified the components of effective transitional care,²³ barriers persist. The communication and collaboration needed to provide coordinated care across healthcare delivery systems are difficult due to the lack of standardized approaches between systems.²⁴ Consequently, follow-up care may be delayed or missed altogether. To our knowledge, there is no published research identifying transitional care challenges for clinicians, staff, and veterans in transitioning from non-VA hospitals to a VA primary care setting.

Study Design

We conducted a qualitative assessment within the VA Eastern Colorado Health Care System, an urban tertiary medical center, as well as urban and rural non-VA hospitals used by veterans. Semi-structured interview guides informed by the practical robust implementation and sustainability (PRISM) model, the Lean approach, and the Ideal Transitions of Care Bridge were used.^25-27 We explored the PRISM domains such as recipient’s characteristics, the interaction with the external environment, and the implementation and sustainability infrastructure to inform the design and implementation of the intervention.²⁵ The Lean approach included methods to optimize processes by maximizing efficiency and minimizing waste.²⁶ The Ideal Transitions of Care Bridge was used to identify the domains in transitions of care such as discharge planning, communication of information, and care coordination.²⁷

Setting and Participants

We identified the top 10 non-VA hospitals serving the most urban and rural veterans in 2015 using VA administrative data. Purposive sampling was used to ensure that urban and rural non-VA hospitals and different roles within these hospitals were represented. VA clinicians and staff were selected from the Denver VA Medical Center, a tertiary hospital within the Eastern Colorado Health Care System and one VA Community-Based Outpatient Clinic (CBOC) that primarily serves rural veterans. The Denver VA Medical Center has three clinics staffed by Patient Aligned Care Teams (PACTs), a model built on the concept of Patient-Centered Medical Home.²⁸ Hospital leadership were initially approached for permission to recruit their staff and to be involved as key informants, and all agreed. To ensure representativeness, diversity of roles was recruited, including PACT primary care physicians, nurses, and other staff members such as medical assistants and administrators. Veterans were approached for sampling if they were discharged from a non-VA hospital during June–September 2015 and used the VA for primary care. This was to ensure that they remembered the process they went through postdischarge at the time of the interview.

Data Collection and Analysis

The evaluation team members (RA, EL, and MM) conducted the interviews from November 2015 to July 2016. Clinicians, staff, and veterans were asked semi-structured questions about their experiences and their role in transitioning VA patients across systems (see Appendix for interview guides). Veterans were asked to describe their experience and satisfaction with the current postdischarge transition process. We stopped the interviews when we reached data saturation.²⁹

Interviews were audio-recorded, transcribed verbatim, and validated (transcribed interviews were double-checked against recording) to ensure data quality and accuracy. Coding was guided by a conventional content analysis technique^{30, 31} using a deductive and inductive coding approach.³¹ The deductive coding approach was drawn from the Ideal Transitions of Care Bridge and PRISM domains. ^32,33 Two evaluation team members (RA and EL) defined the initial code book by independently coding the first three interviews, worked to clarify the meanings of emergent codes, and came to a consensus when disagreements occurred. Next, a priori codes were added by team members to include the PRISM domains. These PRISM domains included the implementation and sustainability infrastructure, the external environment, the characteristics of intervention recipients, and the organizational and patient perspectives of an intervention.

Additional emergent codes were added to the code book and agreed upon by team members (RA, EL, and MM). Consistent with previously used methods, consensus building was achieved by identifying and resolving differences by discussing with team members (RA, EL, MM, CB, and RB).²⁹ Codes were examined and organized into themes by team members.^29,34-36 This process was continued until no new themes were identified. Results were reviewed by all evaluation team members to assess thoroughness and comprehensiveness.^34,35 In addition, team members triangulated the findings with VA and non-VA participants to ensure validity and reduce researcher bias.^29,37

We conducted a total of 70 interviews with 23 VA and 29 non-VA hospital clinicians and staff and 18 veterans (Table 1). Overall, we found that there was no standardized process for transitioning veterans across healthcare delivery systems. Participants reported that transitions were often inefficient when non-VA hospitals could not (1) identify patients as veterans and notify VA primary care of discharge; (2) transfer non-VA hospital medical records to VA primary care; (3) obtain follow-up care appointments with VA primary care; and (4) write VA formulary medications for veterans to fill at VA pharmacies. In addition, participants discussed about facilitators and suggestions to overcome these inefficiencies and improve transitional care (Table2). We mapped the identified barriers as well as the suggestions for improvement to the PRISM and the Ideal Transitions of Care Bridge domains (Table 3).

Unable to Identify Patients as Veterans and Notify VA Primary Care of Discharge

VA and non-VA participants reported difficulty in communicating about veterans’ hospitalizations and discharge follow-up needs across systems. Non-VA clinicians referenced difficulty in identifying patients as veterans to communicate with VA, except in instances where the VA is a payor, while VA providers described feeling largely uninformed of the veterans non-VA hospitalization. For non-VA clinicians, the lack of a systematic method for veteran identification often left them to inadvertently identify veteran status by asking about their primary care clinicians and insurance and even through an offhanded comment made by the veteran. If a veteran was identified, non-VA clinicians described being uncertain about the best way to notify VA primary care of the patient’s impending discharge. Veterans described instances of the non-VA hospital knowing their veteran status upon admission, but accounts varied on whether the non-VA hospital notified the VA primary care of their hospitalization (Table 2, Theme 1).

Unable to Transfer Non-VA Hospital Medical Records to VA Primary Care

VA clinicians discussed about the challenges associated with obtaining the veteran’s medical record from the non-VA hospitals, and when it was received, it was often incomplete information and significantly delayed. They described relying on the veteran’s description of the care received, which was not complete or accurate information needed to make clinical judgment or coordinate follow-up care. Non-VA clinicians mentioned about trying several methods for transferring the medical record to VA primary care, including discharge summary via electronic system and sometimes solely relying on patients to deliver discharge paperwork to their primary care clinicians. In instances where non-VA hospitals sent discharge paperwork to VA, there was no way for non-VA hospitals to verify whether the faxed electronic medical record was received by the VA hospital. Most of the veterans discussed receiving written postdischarge instructions to take to their VA primary care clinicians; however, they were unsure whether the VA primary care received their medical record or any other information from the non-VA hospital (Table 2, Theme 2).

Unable to Obtain Follow-Up Care Appointments with VA Primary Care

All participants described how difficult it was to obtain a follow-up appointment for veterans with VA primary care. This often resulted in delayed follow-up care. VA clinicians also shared that a non-VA hospitalization can be the impetus for a veteran to seek care at the VA for the very first time. Once eligibility is determined, the veteran is assigned a VA primary care clinician. This process may take up to six weeks, and in the meantime, the veteran is scheduled in VA urgent care for immediate postdischarge care. This lag in primary care assignment creates delayed and fragmented care (Table 2, Theme 3).

Non-VA clinicians, administrators, and staff also discussed the difficulties in scheduling follow-up care with VA primary care. Although discharge paperwork instructed patients to see their VA clinicians, there was no process in place for non-VA clinicians to confirm whether the follow-up care was received due to lack of bilateral communication. In addition, veterans discussed the inefficiencies in scheduling follow-up appointments with VA clinicians where attempts to follow-up with primary care clinicians took eight weeks or more. Several veterans described walking into the clinic without an appointment asking to be seen postdischarge or utilizing the VA emergency department for follow-up care after discharge from a non-VA hospital. Veterans admitted utilizing the VA emergency department for nonemergent reasons such as filling their prescriptions because they are unable to see a VA PCP in a timely manner (Table 2, Theme 3).

Unable to Write VA Formulary Medications for Veterans to Fill at VA Pharmacies

All participants described the difficulties in obtaining medications at VA pharmacies when prescribed by the non-VA hospital clinicians. VA clinicians often had to reassess, and rewrite prescriptions written by clinicians, causing delays. Moreover, rural VA clinicians described lack of VA pharmacies in their locations, where veterans had to mail order medications, causing further delays in needed medications. Non-VA clinicians echoed these frustrations. They noted that veterans were confused about their VA pharmacy benefits as well as the need for the non-VA clinicians to follow VA formulary guidelines. Veterans expressed that it was especially challenging to physically go to the VA pharmacy to pick up medications after discharge due to lack of transportation, limited VA pharmacy hours, and long wait times. Several veterans discussed paying for their prescriptions out of pocket even though they had VA pharmacy benefits because it was more convenient to use the non-VA pharmacy. In other instances, veterans discussed going to a VA emergency department and waiting for hours to have their non-VA clinician prescription rewritten by a VA clinician (Table 2, Theme 4).

Facilitators of the Current Transition Process

Several participants provided examples of when transitional care communication between systems occurred seamlessly. VA staff and veterans noted that the VA increased the availability of urgent care appointments, which allowed for timelier postacute care follow-up appointments. Non-VA hospital clinicians also noted the availability of additional appointment slots but stated that they did not learn about these additional appointments directly from the VA. Instead, they learned of these through medical residents caring for patients at both VA and non-VA hospitals. One VA CBOC designated two nurses to care for walk-in veterans for their postdischarge follow-up needs. Some VA participants also noted that the VA Call Center Nurses occasionally called veterans upon discharge to schedule a follow-up appointment and facilitated timely care.

Participants from a VA CBOC discussed being part of a Community Transitions Consortium aimed at identifying high-utilizing patients (veteran and nonveteran) and improving communication across systems. The consortium members discussed each facility’s transition-of-care process, described having access to local non-VA hospital medical records and a backline phone number at the non-VA hospitals to coordinate transitional care. This allowed the VA clinicians to learn about non-VA hospital processes and veteran needs.

Suggestions for Improving the Transitional Care Process

VA and non-VA clinicians suggested hiring a VA liaison, preferably with a clinical background to facilitate care coordination across healthcare systems. They recommended that this person work closely with VA primary care, strengthen the relationship with non-VA hospitals, and help veterans learn more about the transition-of-care processes. Topics discussed for veteran education included how to (1) access their primary care team; (2) alert VA of non-VA hospitalization and the billing process; (3) recognize symptoms and manage care; and (4) obtain follow-up care appointments. Furthermore, they suggested that the liaison would help facilitate the transfer of medical records between VA and non-VA hospitals. Other suggestions included allowing veterans to fill prescriptions at non-VA pharmacies and dedicating a phone line for non-VA clinicians to speak to VA clinicians and staff.

Veterans agreed that improvements to the current process should include an efficient system for obtaining medications and the ability to schedule timely follow-up appointments. Furthermore, veterans wanted education about the VA transition-of-care process following a non-VA hospitalization, including payment and VA notification processes (Table 2, Theme 5).

Participants described the current transitional care process as inefficient with specific barriers that have negative consequences on patient care and clinician and staff work processes. They described difficulties in obtaining medications prescribed by non-VA clinicians from VA pharmacies, delays in follow-up appointments at the VA, and lack of bilateral communication between systems and medical record transfer. Participants also provided concrete suggestions to improving the current process, including a care coordinator with clinical background. These findings are important in the context of VA increasing veteran access to care in the community.

Despite an increasing emphasis on veteran access to non-VA care as a result of the VA strategic goals and several new programs,^7,12,13 there has not been a close examination of the current transition-of-care process from non-VA hospitals to VA primary care. Several studies have shown that the period following a hospitalization is especially vulnerable and associated with adverse events such as readmission, high cost, and death.^12,31,32 Our findings agree with previous research that identified medical record transfer across systems as one of the most challenging issues contributing to deficits in communication between care teams.³³ In addition, our study brought into focus the significant challenges faced by veterans in obtaining medications post non-VA hospital discharge. Addressing these key barriers in transitional care will improve the quality, safety, and value of healthcare in the current transition process.^38,39

Based on our findings, our participants’ concern in transitional care can be addressed in various ways. First, as veterans are increasingly receiving care in the community, identifying their veteran status early on in the non-VA hospital setting could help in improved, real time communication with the VA. This could be done by updating patient intake forms to ask patients whether they are veterans or not. Second, VA policy-level changes should work to provide veterans access to non-VA pharmacy benefits equivalent to the access patients are receiving for hospital, specialty, and outpatient care. Third, patient and provider satisfaction for dual-use veterans should be examined closely. Although participants expressed frustration with the overall transitions of care from non-VA hospitals to VA primary care setting, influence of this on the Quadruple Aim-improving patient outcomes, experience, and reducing clinician and staff burnout should be examined closely.⁴⁰ Fourth, evidence-based interventions such as nurse-led transitional care programs that have proven helpful in reducing adverse outcomes in both VA and non-VA settings will be useful to implement.^41-45 Such programs could be located in the VA, and a care coordinator role could help facilitate transitional care needs for veterans by working with multiple non-VA hospitals.

The limitations of this study are that the perspectives shared by these participants may not represent all VA and non-VA hospitals as well as veterans’ experiences with transition of care. In addition, the study was conducted in one state and the findings may not be applicable to other healthcare systems. However, our study highlighted the consistent challenges of receiving care across VA and other hospital systems. Two strengths of this study are that it was conducted by multidisciplinary research team members with expertise in qualitative research, clinical care, and implementation science and that we obtained convergent information from VA, non-VA, and veteran participants.

Our current transition-of-care process has several shortcomings. There was a clear agreement on barriers, facilitators, and suggestions for improving the current transitions-of-care process among VA and non-VA hospital participants, as well as from veterans who experienced transitions across different delivery systems. Transitioning veterans to VA primary care following a non-VA hospitalization is a crucial first step for improving care for veterans and reducing adverse outcomes such as avoidable hospital readmissions and death.

These results describe the inefficiencies experienced by patients, clinicians, and staff and their suggestions to alleviate these barriers for optimal continuum of care. To avoid frustration and inefficiencies, the increased emphasis of providing non-VA care for veterans should consider the challenges experienced in transitional care and the opportunities for increased coordination of care.

As the VA moves toward increased utilization of non-VA care, it is crucial to understand and address the challenges of transitional care faced by dual-use veterans to provide high-quality care that improves healthcare outcomes.^7,11,12 The VA implemented a shift in policy from the Veterans Access, Choice, and Accountability Act of 2014 (Public Law 113-146; “Choice Act”) to the VA Maintaining Internal Systems and Strengthening Integrated Outside Networks (MISSION) Act beginning June 6, 2019.^13,14 Under the MISSION Act, veterans have more ways to access healthcare within the VA’s network and through approved non-VA medical providers in the community known as “community care providers.”¹⁵ This shift expanded the existing VA Choice Act of 2014, where the program allowed those veterans who are unable to schedule an appointment within 30 days of their preferred date or the clinically appropriate date, or on the basis of their place of residence, to elect to receive care from eligible non-VA healthcare entities or providers.^14,15 These efforts to better serve veterans by increasing non-VA care might present added care coordination challenges for patients and their providers when they seek care in the VA.

High-quality transitional care prevents poor outcomes such as hospital readmissions.^16-18 When communication and coordination across healthcare delivery systems are lacking, patients and their families often find themselves at risk for adverse events.^19,20 Past research shows that patients have fewer adverse events when they receive comprehensive postdischarge care, including instructions on medications and self-care, symptom recognition and management, and reminders to attend follow-up appointments.^17,21,22 Although researchers have identified the components of effective transitional care,²³ barriers persist. The communication and collaboration needed to provide coordinated care across healthcare delivery systems are difficult due to the lack of standardized approaches between systems.²⁴ Consequently, follow-up care may be delayed or missed altogether. To our knowledge, there is no published research identifying transitional care challenges for clinicians, staff, and veterans in transitioning from non-VA hospitals to a VA primary care setting.

Study Design

We conducted a qualitative assessment within the VA Eastern Colorado Health Care System, an urban tertiary medical center, as well as urban and rural non-VA hospitals used by veterans. Semi-structured interview guides informed by the practical robust implementation and sustainability (PRISM) model, the Lean approach, and the Ideal Transitions of Care Bridge were used.^25-27 We explored the PRISM domains such as recipient’s characteristics, the interaction with the external environment, and the implementation and sustainability infrastructure to inform the design and implementation of the intervention.²⁵ The Lean approach included methods to optimize processes by maximizing efficiency and minimizing waste.²⁶ The Ideal Transitions of Care Bridge was used to identify the domains in transitions of care such as discharge planning, communication of information, and care coordination.²⁷

Setting and Participants

We identified the top 10 non-VA hospitals serving the most urban and rural veterans in 2015 using VA administrative data. Purposive sampling was used to ensure that urban and rural non-VA hospitals and different roles within these hospitals were represented. VA clinicians and staff were selected from the Denver VA Medical Center, a tertiary hospital within the Eastern Colorado Health Care System and one VA Community-Based Outpatient Clinic (CBOC) that primarily serves rural veterans. The Denver VA Medical Center has three clinics staffed by Patient Aligned Care Teams (PACTs), a model built on the concept of Patient-Centered Medical Home.²⁸ Hospital leadership were initially approached for permission to recruit their staff and to be involved as key informants, and all agreed. To ensure representativeness, diversity of roles was recruited, including PACT primary care physicians, nurses, and other staff members such as medical assistants and administrators. Veterans were approached for sampling if they were discharged from a non-VA hospital during June–September 2015 and used the VA for primary care. This was to ensure that they remembered the process they went through postdischarge at the time of the interview.

Data Collection and Analysis

The evaluation team members (RA, EL, and MM) conducted the interviews from November 2015 to July 2016. Clinicians, staff, and veterans were asked semi-structured questions about their experiences and their role in transitioning VA patients across systems (see Appendix for interview guides). Veterans were asked to describe their experience and satisfaction with the current postdischarge transition process. We stopped the interviews when we reached data saturation.²⁹

Interviews were audio-recorded, transcribed verbatim, and validated (transcribed interviews were double-checked against recording) to ensure data quality and accuracy. Coding was guided by a conventional content analysis technique^{30, 31} using a deductive and inductive coding approach.³¹ The deductive coding approach was drawn from the Ideal Transitions of Care Bridge and PRISM domains. ^32,33 Two evaluation team members (RA and EL) defined the initial code book by independently coding the first three interviews, worked to clarify the meanings of emergent codes, and came to a consensus when disagreements occurred. Next, a priori codes were added by team members to include the PRISM domains. These PRISM domains included the implementation and sustainability infrastructure, the external environment, the characteristics of intervention recipients, and the organizational and patient perspectives of an intervention.

Additional emergent codes were added to the code book and agreed upon by team members (RA, EL, and MM). Consistent with previously used methods, consensus building was achieved by identifying and resolving differences by discussing with team members (RA, EL, MM, CB, and RB).²⁹ Codes were examined and organized into themes by team members.^29,34-36 This process was continued until no new themes were identified. Results were reviewed by all evaluation team members to assess thoroughness and comprehensiveness.^34,35 In addition, team members triangulated the findings with VA and non-VA participants to ensure validity and reduce researcher bias.^29,37

We conducted a total of 70 interviews with 23 VA and 29 non-VA hospital clinicians and staff and 18 veterans (Table 1). Overall, we found that there was no standardized process for transitioning veterans across healthcare delivery systems. Participants reported that transitions were often inefficient when non-VA hospitals could not (1) identify patients as veterans and notify VA primary care of discharge; (2) transfer non-VA hospital medical records to VA primary care; (3) obtain follow-up care appointments with VA primary care; and (4) write VA formulary medications for veterans to fill at VA pharmacies. In addition, participants discussed about facilitators and suggestions to overcome these inefficiencies and improve transitional care (Table2). We mapped the identified barriers as well as the suggestions for improvement to the PRISM and the Ideal Transitions of Care Bridge domains (Table 3).

Unable to Identify Patients as Veterans and Notify VA Primary Care of Discharge

VA and non-VA participants reported difficulty in communicating about veterans’ hospitalizations and discharge follow-up needs across systems. Non-VA clinicians referenced difficulty in identifying patients as veterans to communicate with VA, except in instances where the VA is a payor, while VA providers described feeling largely uninformed of the veterans non-VA hospitalization. For non-VA clinicians, the lack of a systematic method for veteran identification often left them to inadvertently identify veteran status by asking about their primary care clinicians and insurance and even through an offhanded comment made by the veteran. If a veteran was identified, non-VA clinicians described being uncertain about the best way to notify VA primary care of the patient’s impending discharge. Veterans described instances of the non-VA hospital knowing their veteran status upon admission, but accounts varied on whether the non-VA hospital notified the VA primary care of their hospitalization (Table 2, Theme 1).

Unable to Transfer Non-VA Hospital Medical Records to VA Primary Care

VA clinicians discussed about the challenges associated with obtaining the veteran’s medical record from the non-VA hospitals, and when it was received, it was often incomplete information and significantly delayed. They described relying on the veteran’s description of the care received, which was not complete or accurate information needed to make clinical judgment or coordinate follow-up care. Non-VA clinicians mentioned about trying several methods for transferring the medical record to VA primary care, including discharge summary via electronic system and sometimes solely relying on patients to deliver discharge paperwork to their primary care clinicians. In instances where non-VA hospitals sent discharge paperwork to VA, there was no way for non-VA hospitals to verify whether the faxed electronic medical record was received by the VA hospital. Most of the veterans discussed receiving written postdischarge instructions to take to their VA primary care clinicians; however, they were unsure whether the VA primary care received their medical record or any other information from the non-VA hospital (Table 2, Theme 2).

Unable to Obtain Follow-Up Care Appointments with VA Primary Care

All participants described how difficult it was to obtain a follow-up appointment for veterans with VA primary care. This often resulted in delayed follow-up care. VA clinicians also shared that a non-VA hospitalization can be the impetus for a veteran to seek care at the VA for the very first time. Once eligibility is determined, the veteran is assigned a VA primary care clinician. This process may take up to six weeks, and in the meantime, the veteran is scheduled in VA urgent care for immediate postdischarge care. This lag in primary care assignment creates delayed and fragmented care (Table 2, Theme 3).

Non-VA clinicians, administrators, and staff also discussed the difficulties in scheduling follow-up care with VA primary care. Although discharge paperwork instructed patients to see their VA clinicians, there was no process in place for non-VA clinicians to confirm whether the follow-up care was received due to lack of bilateral communication. In addition, veterans discussed the inefficiencies in scheduling follow-up appointments with VA clinicians where attempts to follow-up with primary care clinicians took eight weeks or more. Several veterans described walking into the clinic without an appointment asking to be seen postdischarge or utilizing the VA emergency department for follow-up care after discharge from a non-VA hospital. Veterans admitted utilizing the VA emergency department for nonemergent reasons such as filling their prescriptions because they are unable to see a VA PCP in a timely manner (Table 2, Theme 3).

Unable to Write VA Formulary Medications for Veterans to Fill at VA Pharmacies

All participants described the difficulties in obtaining medications at VA pharmacies when prescribed by the non-VA hospital clinicians. VA clinicians often had to reassess, and rewrite prescriptions written by clinicians, causing delays. Moreover, rural VA clinicians described lack of VA pharmacies in their locations, where veterans had to mail order medications, causing further delays in needed medications. Non-VA clinicians echoed these frustrations. They noted that veterans were confused about their VA pharmacy benefits as well as the need for the non-VA clinicians to follow VA formulary guidelines. Veterans expressed that it was especially challenging to physically go to the VA pharmacy to pick up medications after discharge due to lack of transportation, limited VA pharmacy hours, and long wait times. Several veterans discussed paying for their prescriptions out of pocket even though they had VA pharmacy benefits because it was more convenient to use the non-VA pharmacy. In other instances, veterans discussed going to a VA emergency department and waiting for hours to have their non-VA clinician prescription rewritten by a VA clinician (Table 2, Theme 4).

Facilitators of the Current Transition Process

Several participants provided examples of when transitional care communication between systems occurred seamlessly. VA staff and veterans noted that the VA increased the availability of urgent care appointments, which allowed for timelier postacute care follow-up appointments. Non-VA hospital clinicians also noted the availability of additional appointment slots but stated that they did not learn about these additional appointments directly from the VA. Instead, they learned of these through medical residents caring for patients at both VA and non-VA hospitals. One VA CBOC designated two nurses to care for walk-in veterans for their postdischarge follow-up needs. Some VA participants also noted that the VA Call Center Nurses occasionally called veterans upon discharge to schedule a follow-up appointment and facilitated timely care.

Participants from a VA CBOC discussed being part of a Community Transitions Consortium aimed at identifying high-utilizing patients (veteran and nonveteran) and improving communication across systems. The consortium members discussed each facility’s transition-of-care process, described having access to local non-VA hospital medical records and a backline phone number at the non-VA hospitals to coordinate transitional care. This allowed the VA clinicians to learn about non-VA hospital processes and veteran needs.

Suggestions for Improving the Transitional Care Process

VA and non-VA clinicians suggested hiring a VA liaison, preferably with a clinical background to facilitate care coordination across healthcare systems. They recommended that this person work closely with VA primary care, strengthen the relationship with non-VA hospitals, and help veterans learn more about the transition-of-care processes. Topics discussed for veteran education included how to (1) access their primary care team; (2) alert VA of non-VA hospitalization and the billing process; (3) recognize symptoms and manage care; and (4) obtain follow-up care appointments. Furthermore, they suggested that the liaison would help facilitate the transfer of medical records between VA and non-VA hospitals. Other suggestions included allowing veterans to fill prescriptions at non-VA pharmacies and dedicating a phone line for non-VA clinicians to speak to VA clinicians and staff.

Veterans agreed that improvements to the current process should include an efficient system for obtaining medications and the ability to schedule timely follow-up appointments. Furthermore, veterans wanted education about the VA transition-of-care process following a non-VA hospitalization, including payment and VA notification processes (Table 2, Theme 5).

Participants described the current transitional care process as inefficient with specific barriers that have negative consequences on patient care and clinician and staff work processes. They described difficulties in obtaining medications prescribed by non-VA clinicians from VA pharmacies, delays in follow-up appointments at the VA, and lack of bilateral communication between systems and medical record transfer. Participants also provided concrete suggestions to improving the current process, including a care coordinator with clinical background. These findings are important in the context of VA increasing veteran access to care in the community.

Despite an increasing emphasis on veteran access to non-VA care as a result of the VA strategic goals and several new programs,^7,12,13 there has not been a close examination of the current transition-of-care process from non-VA hospitals to VA primary care. Several studies have shown that the period following a hospitalization is especially vulnerable and associated with adverse events such as readmission, high cost, and death.^12,31,32 Our findings agree with previous research that identified medical record transfer across systems as one of the most challenging issues contributing to deficits in communication between care teams.³³ In addition, our study brought into focus the significant challenges faced by veterans in obtaining medications post non-VA hospital discharge. Addressing these key barriers in transitional care will improve the quality, safety, and value of healthcare in the current transition process.^38,39

Based on our findings, our participants’ concern in transitional care can be addressed in various ways. First, as veterans are increasingly receiving care in the community, identifying their veteran status early on in the non-VA hospital setting could help in improved, real time communication with the VA. This could be done by updating patient intake forms to ask patients whether they are veterans or not. Second, VA policy-level changes should work to provide veterans access to non-VA pharmacy benefits equivalent to the access patients are receiving for hospital, specialty, and outpatient care. Third, patient and provider satisfaction for dual-use veterans should be examined closely. Although participants expressed frustration with the overall transitions of care from non-VA hospitals to VA primary care setting, influence of this on the Quadruple Aim-improving patient outcomes, experience, and reducing clinician and staff burnout should be examined closely.⁴⁰ Fourth, evidence-based interventions such as nurse-led transitional care programs that have proven helpful in reducing adverse outcomes in both VA and non-VA settings will be useful to implement.^41-45 Such programs could be located in the VA, and a care coordinator role could help facilitate transitional care needs for veterans by working with multiple non-VA hospitals.

The limitations of this study are that the perspectives shared by these participants may not represent all VA and non-VA hospitals as well as veterans’ experiences with transition of care. In addition, the study was conducted in one state and the findings may not be applicable to other healthcare systems. However, our study highlighted the consistent challenges of receiving care across VA and other hospital systems. Two strengths of this study are that it was conducted by multidisciplinary research team members with expertise in qualitative research, clinical care, and implementation science and that we obtained convergent information from VA, non-VA, and veteran participants.

Our current transition-of-care process has several shortcomings. There was a clear agreement on barriers, facilitators, and suggestions for improving the current transitions-of-care process among VA and non-VA hospital participants, as well as from veterans who experienced transitions across different delivery systems. Transitioning veterans to VA primary care following a non-VA hospitalization is a crucial first step for improving care for veterans and reducing adverse outcomes such as avoidable hospital readmissions and death.

These results describe the inefficiencies experienced by patients, clinicians, and staff and their suggestions to alleviate these barriers for optimal continuum of care. To avoid frustration and inefficiencies, the increased emphasis of providing non-VA care for veterans should consider the challenges experienced in transitional care and the opportunities for increased coordination of care.

As the VA moves toward increased utilization of non-VA care, it is crucial to understand and address the challenges of transitional care faced by dual-use veterans to provide high-quality care that improves healthcare outcomes.^7,11,12 The VA implemented a shift in policy from the Veterans Access, Choice, and Accountability Act of 2014 (Public Law 113-146; “Choice Act”) to the VA Maintaining Internal Systems and Strengthening Integrated Outside Networks (MISSION) Act beginning June 6, 2019.^13,14 Under the MISSION Act, veterans have more ways to access healthcare within the VA’s network and through approved non-VA medical providers in the community known as “community care providers.”¹⁵ This shift expanded the existing VA Choice Act of 2014, where the program allowed those veterans who are unable to schedule an appointment within 30 days of their preferred date or the clinically appropriate date, or on the basis of their place of residence, to elect to receive care from eligible non-VA healthcare entities or providers.^14,15 These efforts to better serve veterans by increasing non-VA care might present added care coordination challenges for patients and their providers when they seek care in the VA.

High-quality transitional care prevents poor outcomes such as hospital readmissions.^16-18 When communication and coordination across healthcare delivery systems are lacking, patients and their families often find themselves at risk for adverse events.^19,20 Past research shows that patients have fewer adverse events when they receive comprehensive postdischarge care, including instructions on medications and self-care, symptom recognition and management, and reminders to attend follow-up appointments.^17,21,22 Although researchers have identified the components of effective transitional care,²³ barriers persist. The communication and collaboration needed to provide coordinated care across healthcare delivery systems are difficult due to the lack of standardized approaches between systems.²⁴ Consequently, follow-up care may be delayed or missed altogether. To our knowledge, there is no published research identifying transitional care challenges for clinicians, staff, and veterans in transitioning from non-VA hospitals to a VA primary care setting.

Study Design

We conducted a qualitative assessment within the VA Eastern Colorado Health Care System, an urban tertiary medical center, as well as urban and rural non-VA hospitals used by veterans. Semi-structured interview guides informed by the practical robust implementation and sustainability (PRISM) model, the Lean approach, and the Ideal Transitions of Care Bridge were used.^25-27 We explored the PRISM domains such as recipient’s characteristics, the interaction with the external environment, and the implementation and sustainability infrastructure to inform the design and implementation of the intervention.²⁵ The Lean approach included methods to optimize processes by maximizing efficiency and minimizing waste.²⁶ The Ideal Transitions of Care Bridge was used to identify the domains in transitions of care such as discharge planning, communication of information, and care coordination.²⁷

Setting and Participants

We identified the top 10 non-VA hospitals serving the most urban and rural veterans in 2015 using VA administrative data. Purposive sampling was used to ensure that urban and rural non-VA hospitals and different roles within these hospitals were represented. VA clinicians and staff were selected from the Denver VA Medical Center, a tertiary hospital within the Eastern Colorado Health Care System and one VA Community-Based Outpatient Clinic (CBOC) that primarily serves rural veterans. The Denver VA Medical Center has three clinics staffed by Patient Aligned Care Teams (PACTs), a model built on the concept of Patient-Centered Medical Home.²⁸ Hospital leadership were initially approached for permission to recruit their staff and to be involved as key informants, and all agreed. To ensure representativeness, diversity of roles was recruited, including PACT primary care physicians, nurses, and other staff members such as medical assistants and administrators. Veterans were approached for sampling if they were discharged from a non-VA hospital during June–September 2015 and used the VA for primary care. This was to ensure that they remembered the process they went through postdischarge at the time of the interview.

Data Collection and Analysis

The evaluation team members (RA, EL, and MM) conducted the interviews from November 2015 to July 2016. Clinicians, staff, and veterans were asked semi-structured questions about their experiences and their role in transitioning VA patients across systems (see Appendix for interview guides). Veterans were asked to describe their experience and satisfaction with the current postdischarge transition process. We stopped the interviews when we reached data saturation.²⁹

Interviews were audio-recorded, transcribed verbatim, and validated (transcribed interviews were double-checked against recording) to ensure data quality and accuracy. Coding was guided by a conventional content analysis technique^{30, 31} using a deductive and inductive coding approach.³¹ The deductive coding approach was drawn from the Ideal Transitions of Care Bridge and PRISM domains. ^32,33 Two evaluation team members (RA and EL) defined the initial code book by independently coding the first three interviews, worked to clarify the meanings of emergent codes, and came to a consensus when disagreements occurred. Next, a priori codes were added by team members to include the PRISM domains. These PRISM domains included the implementation and sustainability infrastructure, the external environment, the characteristics of intervention recipients, and the organizational and patient perspectives of an intervention.

Additional emergent codes were added to the code book and agreed upon by team members (RA, EL, and MM). Consistent with previously used methods, consensus building was achieved by identifying and resolving differences by discussing with team members (RA, EL, MM, CB, and RB).²⁹ Codes were examined and organized into themes by team members.^29,34-36 This process was continued until no new themes were identified. Results were reviewed by all evaluation team members to assess thoroughness and comprehensiveness.^34,35 In addition, team members triangulated the findings with VA and non-VA participants to ensure validity and reduce researcher bias.^29,37

We conducted a total of 70 interviews with 23 VA and 29 non-VA hospital clinicians and staff and 18 veterans (Table 1). Overall, we found that there was no standardized process for transitioning veterans across healthcare delivery systems. Participants reported that transitions were often inefficient when non-VA hospitals could not (1) identify patients as veterans and notify VA primary care of discharge; (2) transfer non-VA hospital medical records to VA primary care; (3) obtain follow-up care appointments with VA primary care; and (4) write VA formulary medications for veterans to fill at VA pharmacies. In addition, participants discussed about facilitators and suggestions to overcome these inefficiencies and improve transitional care (Table2). We mapped the identified barriers as well as the suggestions for improvement to the PRISM and the Ideal Transitions of Care Bridge domains (Table 3).

Unable to Identify Patients as Veterans and Notify VA Primary Care of Discharge

VA and non-VA participants reported difficulty in communicating about veterans’ hospitalizations and discharge follow-up needs across systems. Non-VA clinicians referenced difficulty in identifying patients as veterans to communicate with VA, except in instances where the VA is a payor, while VA providers described feeling largely uninformed of the veterans non-VA hospitalization. For non-VA clinicians, the lack of a systematic method for veteran identification often left them to inadvertently identify veteran status by asking about their primary care clinicians and insurance and even through an offhanded comment made by the veteran. If a veteran was identified, non-VA clinicians described being uncertain about the best way to notify VA primary care of the patient’s impending discharge. Veterans described instances of the non-VA hospital knowing their veteran status upon admission, but accounts varied on whether the non-VA hospital notified the VA primary care of their hospitalization (Table 2, Theme 1).

Unable to Transfer Non-VA Hospital Medical Records to VA Primary Care

VA clinicians discussed about the challenges associated with obtaining the veteran’s medical record from the non-VA hospitals, and when it was received, it was often incomplete information and significantly delayed. They described relying on the veteran’s description of the care received, which was not complete or accurate information needed to make clinical judgment or coordinate follow-up care. Non-VA clinicians mentioned about trying several methods for transferring the medical record to VA primary care, including discharge summary via electronic system and sometimes solely relying on patients to deliver discharge paperwork to their primary care clinicians. In instances where non-VA hospitals sent discharge paperwork to VA, there was no way for non-VA hospitals to verify whether the faxed electronic medical record was received by the VA hospital. Most of the veterans discussed receiving written postdischarge instructions to take to their VA primary care clinicians; however, they were unsure whether the VA primary care received their medical record or any other information from the non-VA hospital (Table 2, Theme 2).

Unable to Obtain Follow-Up Care Appointments with VA Primary Care

All participants described how difficult it was to obtain a follow-up appointment for veterans with VA primary care. This often resulted in delayed follow-up care. VA clinicians also shared that a non-VA hospitalization can be the impetus for a veteran to seek care at the VA for the very first time. Once eligibility is determined, the veteran is assigned a VA primary care clinician. This process may take up to six weeks, and in the meantime, the veteran is scheduled in VA urgent care for immediate postdischarge care. This lag in primary care assignment creates delayed and fragmented care (Table 2, Theme 3).

Non-VA clinicians, administrators, and staff also discussed the difficulties in scheduling follow-up care with VA primary care. Although discharge paperwork instructed patients to see their VA clinicians, there was no process in place for non-VA clinicians to confirm whether the follow-up care was received due to lack of bilateral communication. In addition, veterans discussed the inefficiencies in scheduling follow-up appointments with VA clinicians where attempts to follow-up with primary care clinicians took eight weeks or more. Several veterans described walking into the clinic without an appointment asking to be seen postdischarge or utilizing the VA emergency department for follow-up care after discharge from a non-VA hospital. Veterans admitted utilizing the VA emergency department for nonemergent reasons such as filling their prescriptions because they are unable to see a VA PCP in a timely manner (Table 2, Theme 3).

Unable to Write VA Formulary Medications for Veterans to Fill at VA Pharmacies

All participants described the difficulties in obtaining medications at VA pharmacies when prescribed by the non-VA hospital clinicians. VA clinicians often had to reassess, and rewrite prescriptions written by clinicians, causing delays. Moreover, rural VA clinicians described lack of VA pharmacies in their locations, where veterans had to mail order medications, causing further delays in needed medications. Non-VA clinicians echoed these frustrations. They noted that veterans were confused about their VA pharmacy benefits as well as the need for the non-VA clinicians to follow VA formulary guidelines. Veterans expressed that it was especially challenging to physically go to the VA pharmacy to pick up medications after discharge due to lack of transportation, limited VA pharmacy hours, and long wait times. Several veterans discussed paying for their prescriptions out of pocket even though they had VA pharmacy benefits because it was more convenient to use the non-VA pharmacy. In other instances, veterans discussed going to a VA emergency department and waiting for hours to have their non-VA clinician prescription rewritten by a VA clinician (Table 2, Theme 4).

Facilitators of the Current Transition Process

Several participants provided examples of when transitional care communication between systems occurred seamlessly. VA staff and veterans noted that the VA increased the availability of urgent care appointments, which allowed for timelier postacute care follow-up appointments. Non-VA hospital clinicians also noted the availability of additional appointment slots but stated that they did not learn about these additional appointments directly from the VA. Instead, they learned of these through medical residents caring for patients at both VA and non-VA hospitals. One VA CBOC designated two nurses to care for walk-in veterans for their postdischarge follow-up needs. Some VA participants also noted that the VA Call Center Nurses occasionally called veterans upon discharge to schedule a follow-up appointment and facilitated timely care.

Participants from a VA CBOC discussed being part of a Community Transitions Consortium aimed at identifying high-utilizing patients (veteran and nonveteran) and improving communication across systems. The consortium members discussed each facility’s transition-of-care process, described having access to local non-VA hospital medical records and a backline phone number at the non-VA hospitals to coordinate transitional care. This allowed the VA clinicians to learn about non-VA hospital processes and veteran needs.

Suggestions for Improving the Transitional Care Process

VA and non-VA clinicians suggested hiring a VA liaison, preferably with a clinical background to facilitate care coordination across healthcare systems. They recommended that this person work closely with VA primary care, strengthen the relationship with non-VA hospitals, and help veterans learn more about the transition-of-care processes. Topics discussed for veteran education included how to (1) access their primary care team; (2) alert VA of non-VA hospitalization and the billing process; (3) recognize symptoms and manage care; and (4) obtain follow-up care appointments. Furthermore, they suggested that the liaison would help facilitate the transfer of medical records between VA and non-VA hospitals. Other suggestions included allowing veterans to fill prescriptions at non-VA pharmacies and dedicating a phone line for non-VA clinicians to speak to VA clinicians and staff.

Veterans agreed that improvements to the current process should include an efficient system for obtaining medications and the ability to schedule timely follow-up appointments. Furthermore, veterans wanted education about the VA transition-of-care process following a non-VA hospitalization, including payment and VA notification processes (Table 2, Theme 5).

Participants described the current transitional care process as inefficient with specific barriers that have negative consequences on patient care and clinician and staff work processes. They described difficulties in obtaining medications prescribed by non-VA clinicians from VA pharmacies, delays in follow-up appointments at the VA, and lack of bilateral communication between systems and medical record transfer. Participants also provided concrete suggestions to improving the current process, including a care coordinator with clinical background. These findings are important in the context of VA increasing veteran access to care in the community.

Despite an increasing emphasis on veteran access to non-VA care as a result of the VA strategic goals and several new programs,^7,12,13 there has not been a close examination of the current transition-of-care process from non-VA hospitals to VA primary care. Several studies have shown that the period following a hospitalization is especially vulnerable and associated with adverse events such as readmission, high cost, and death.^12,31,32 Our findings agree with previous research that identified medical record transfer across systems as one of the most challenging issues contributing to deficits in communication between care teams.³³ In addition, our study brought into focus the significant challenges faced by veterans in obtaining medications post non-VA hospital discharge. Addressing these key barriers in transitional care will improve the quality, safety, and value of healthcare in the current transition process.^38,39

Based on our findings, our participants’ concern in transitional care can be addressed in various ways. First, as veterans are increasingly receiving care in the community, identifying their veteran status early on in the non-VA hospital setting could help in improved, real time communication with the VA. This could be done by updating patient intake forms to ask patients whether they are veterans or not. Second, VA policy-level changes should work to provide veterans access to non-VA pharmacy benefits equivalent to the access patients are receiving for hospital, specialty, and outpatient care. Third, patient and provider satisfaction for dual-use veterans should be examined closely. Although participants expressed frustration with the overall transitions of care from non-VA hospitals to VA primary care setting, influence of this on the Quadruple Aim-improving patient outcomes, experience, and reducing clinician and staff burnout should be examined closely.⁴⁰ Fourth, evidence-based interventions such as nurse-led transitional care programs that have proven helpful in reducing adverse outcomes in both VA and non-VA settings will be useful to implement.^41-45 Such programs could be located in the VA, and a care coordinator role could help facilitate transitional care needs for veterans by working with multiple non-VA hospitals.

The limitations of this study are that the perspectives shared by these participants may not represent all VA and non-VA hospitals as well as veterans’ experiences with transition of care. In addition, the study was conducted in one state and the findings may not be applicable to other healthcare systems. However, our study highlighted the consistent challenges of receiving care across VA and other hospital systems. Two strengths of this study are that it was conducted by multidisciplinary research team members with expertise in qualitative research, clinical care, and implementation science and that we obtained convergent information from VA, non-VA, and veteran participants.

Our current transition-of-care process has several shortcomings. There was a clear agreement on barriers, facilitators, and suggestions for improving the current transitions-of-care process among VA and non-VA hospital participants, as well as from veterans who experienced transitions across different delivery systems. Transitioning veterans to VA primary care following a non-VA hospitalization is a crucial first step for improving care for veterans and reducing adverse outcomes such as avoidable hospital readmissions and death.

These results describe the inefficiencies experienced by patients, clinicians, and staff and their suggestions to alleviate these barriers for optimal continuum of care. To avoid frustration and inefficiencies, the increased emphasis of providing non-VA care for veterans should consider the challenges experienced in transitional care and the opportunities for increased coordination of care.