Hospitalists have been suggested to offer rational and efficient medical care through specialized knowledge about inpatient care services.1 The goal that hospitalists will use care resources more efficiently presumes hospitalists' accurate knowledge about charges and costs.

Data regarding physicians' awareness of care charges and its impact upon care is limited. An international meta‐analysis of clinicians' awareness of pharmaceutical prices demonstrated poor accuracy of physicians' estimates of charges, but effects of increasing their knowledge remained unexamined.2 Continuous exposure to education and alerts about charges have been demonstrated to diminish physicians' unnecessary use of specific laboratory assays in a single teaching hospital, in a pediatric emergency department, and an outpatient primary care system; test use declined when physicians were alerted to test charges at the point‐of‐care without negative impact upon clinical outcomes; when the notices ceased, utilization climbed back towards baseline levels. Specific to inpatient care, a single‐center study evaluated the impact of price‐alerts upon laboratory and imaging use, but showed no effects.36

Applicability of these existing data for contemporary hospitalists are limited, and most data were collected before hospitalists developed as an organized focus of practice. A review of existing literature revealed no published data demonstrating hospitalists' higher expert awareness of charges generated by inpatient care. Published comparisons of the care expense generated by hospitalists' care versus that of general internists or academic teams have shown minimal and inconsistent effects.79 Those data showing reduced costs from hospitalists were associated with small length‐of‐stay reductions, rather than more expert resource utilization.7 We measured the accuracy and precision of hospitalist's estimates of charges associated with services commonly used in inpatient care.

Setting

Two community‐based private, academic‐affiliated hospitals operated by a not‐for‐profit health system in Washington State, comprising together 895 inpatient beds. The questionnaire instrument was approved by the governing Institutional Review Board (IRB).

Methods

A list of true charges for 14 services, procedures, tests, and physician charges commonly ordered by adult medicine hospitalists was acquired directly from the responsible departments of a multi‐hospital system operated by a single non‐profit entity using a unified chargemaster. Specifically, we acquired the charge that a hypothetical self‐paying patient would receive for each service, excluding any adjustments exercised by other payer sources. The list of charges was reported to the organization's financial officers for affirmation. Physician charges were standardized to geographically‐adjusted Medicare charges obtained directly from the American Medical Assocation's online Common Procedural Terminology tool.10 A cross‐section of hospitalists (n = 25) was surveyed from a private hospitalist group and an academically‐affiliated hospitalist service. Hospitalists included US and international medical graduates, new‐graduates from residency training and clinicians with a range of prior experiences in academic centers, government hospital systems, and private primary care. Respondents were asked to estimate to the nearest dollar the billing charge that a hypothetical self‐pay patient would receive for each care item. Direct data collection was arranged through the groups' medical directors and occurred in the hospitalist groups' regular business meetings. The design gathered no data on individual characteristics of respondents such as domestic or international education, sex, age, or time in practice, nor from which practice‐group a given respondent originated, to affirm to participants that their responses could not be used to imply performance measures or quality profiles.

Findings

Hospitalists tended to rank the expense of items in essentially the correct order, as reflected by the rough trend of rising estimates compared to true‐charges (Figure 1). Of note, we did not ask hospitalists to discern the different charges of 2 appropriate competing clinical choices, but asked for estimated prices of diverse services. The range of respondents' estimates about each item was broad. The mean‐value of hospitalists' estimates for each care item was less revealing than the range and diversity of estimates about each care item. Accuracy of hospitalists' estimates of charges was poor. Only 10.8% of hospitalists' estimates were within 10% of the actual unadjusted charge, 17.8% within 20% of that charge, and 24.8% were within a 30% margin of accuracy. Summary results are presented (Table 1). Pearson's r correlation value between the unadjusted charges and the estimates made by hospitalists was 0.548, a coefficient of determination equal to 0.300. Thus, the true charges list we obtained had only a low‐grade association with hospitalists' estimates (Figure 1). Hospitalists' estimates about the charges of relatively‐expensive items (abdominal computed tomography [CT]) overlapped with their estimates about the least‐expensive items (such as a urine culture ). Inter‐hospitalist agreement about charges associated with each care item was also low; estimates for each care item charge varied over logarithmic orders of magnitude.

Figure 1

Discussion

To date, hospitalist programs have shown little impact upon care costs when compared with other inpatient care staffing models. One limiting factor may be the opacity of medical care pricing. Patients have been demonstrated to have little access to knowledge of what care will cost them and complex barriers prevent them from gaining pricing information.11, 12 Hospitalists may be conjectured to serve as expert sources on the costs and values of medical care services on behalf of inpatients, but our observations suggest that hospitalists' actual knowledge of patient‐charges is lacking. The opacity of US medical prices to patients appears to extend to hospital‐care physicians as well. We observe that no widespread mechanism exists by which hospitalists would be well‐positioned to become informed about the actual charges their patients receive. Unadjusted chargemaster lists are generally restricted information, and would be difficult to access outside of participation in the charge‐notifications used in the existing studies cited above.

The inquiry was specifically limited to how closely hospitalists' estimates of the unadjusted charges for some commonly‐ordered items compare to the actual unadjusted chargemaster at their own institutions. We did not assess the hospitalists' perceptions about the accuracy of their estimates, nor the impact of specific hospitalist characteristics upon accuracy. Our sample's representation of the larger national population of hospital physicians is not established, but engenders no expectation that these clinicians' charge‐awareness is substantively different from that of hospitalists in most other institutions. It is not known what specific clinician or practice‐setting characteristics will direct charge‐awareness, or will influence the impact of charge‐awareness upon clinical practices.

The range of estimates different hospitalists made about the same care items in the same facilities was very broad, which argues that respondents did not estimate charges based upon a different knowledge base of which the investigators were unaware. This is important because our use of unadjusted charges to self‐pay patients as true prices is necessarily somewhat arbitrary. Chargemaster price may not reflect the institution's cost of performing the service, the different prices paid for a single service by different payer sources, nor reflect services' true value based upon outcomes. However, recognizing that these actual prices are somewhat artificial, the use of these prices suffices for the current inquiry, and does not negate our findings of hospitalists' low accuracy and low agreement. Also noteworthy, an unadjusted chargemaster can often represent the charges received by those uninsured US patients for whom payer‐source adjustment is inaccessible, and informs downstream accounting such as the value of unpaid care a hospital delivers annually.

The most immediate matter for examination among hospitalists is what effects increased charge‐awareness may exert upon clinical decisions and practice processes. It appears that the premise that hospitalists' exercise expert knowledge of costs is likely not valid; but it is unknown whether accurate charge‐awareness among hospitalists will improve cost‐reductions by hospitalists. In some payer‐arrangements, accurate charge‐awareness might engender reduced care quantity, rather than increased efficiency or quality. The impact of upgrading hospitalists' knowledge about the charges and costs they generate, and the most effective method to do so, is worthy of investigation; based upon this initial data we encourage and are undertaking a larger‐scale study and exploration of the effects of enhanced hospitalist charge‐awareness.

Conclusion

Hospitalists have low awareness of the charges associated with inpatient care. The opacity of hospital care pricing to patient populations extends also to hospitalist physicians. Hospitalists likely do not improve cost‐efficiency through expert knowledge of services' costs to patients. Education and reminder systems to apprise hospitalists of charges should be examined as possible tools to optimize the use of inpatient care resources.

Hospitalists have been suggested to offer rational and efficient medical care through specialized knowledge about inpatient care services.1 The goal that hospitalists will use care resources more efficiently presumes hospitalists' accurate knowledge about charges and costs.

Data regarding physicians' awareness of care charges and its impact upon care is limited. An international meta‐analysis of clinicians' awareness of pharmaceutical prices demonstrated poor accuracy of physicians' estimates of charges, but effects of increasing their knowledge remained unexamined.2 Continuous exposure to education and alerts about charges have been demonstrated to diminish physicians' unnecessary use of specific laboratory assays in a single teaching hospital, in a pediatric emergency department, and an outpatient primary care system; test use declined when physicians were alerted to test charges at the point‐of‐care without negative impact upon clinical outcomes; when the notices ceased, utilization climbed back towards baseline levels. Specific to inpatient care, a single‐center study evaluated the impact of price‐alerts upon laboratory and imaging use, but showed no effects.36

Applicability of these existing data for contemporary hospitalists are limited, and most data were collected before hospitalists developed as an organized focus of practice. A review of existing literature revealed no published data demonstrating hospitalists' higher expert awareness of charges generated by inpatient care. Published comparisons of the care expense generated by hospitalists' care versus that of general internists or academic teams have shown minimal and inconsistent effects.79 Those data showing reduced costs from hospitalists were associated with small length‐of‐stay reductions, rather than more expert resource utilization.7 We measured the accuracy and precision of hospitalist's estimates of charges associated with services commonly used in inpatient care.

Setting

Two community‐based private, academic‐affiliated hospitals operated by a not‐for‐profit health system in Washington State, comprising together 895 inpatient beds. The questionnaire instrument was approved by the governing Institutional Review Board (IRB).

Methods

A list of true charges for 14 services, procedures, tests, and physician charges commonly ordered by adult medicine hospitalists was acquired directly from the responsible departments of a multi‐hospital system operated by a single non‐profit entity using a unified chargemaster. Specifically, we acquired the charge that a hypothetical self‐paying patient would receive for each service, excluding any adjustments exercised by other payer sources. The list of charges was reported to the organization's financial officers for affirmation. Physician charges were standardized to geographically‐adjusted Medicare charges obtained directly from the American Medical Assocation's online Common Procedural Terminology tool.10 A cross‐section of hospitalists (n = 25) was surveyed from a private hospitalist group and an academically‐affiliated hospitalist service. Hospitalists included US and international medical graduates, new‐graduates from residency training and clinicians with a range of prior experiences in academic centers, government hospital systems, and private primary care. Respondents were asked to estimate to the nearest dollar the billing charge that a hypothetical self‐pay patient would receive for each care item. Direct data collection was arranged through the groups' medical directors and occurred in the hospitalist groups' regular business meetings. The design gathered no data on individual characteristics of respondents such as domestic or international education, sex, age, or time in practice, nor from which practice‐group a given respondent originated, to affirm to participants that their responses could not be used to imply performance measures or quality profiles.

Findings

Hospitalists tended to rank the expense of items in essentially the correct order, as reflected by the rough trend of rising estimates compared to true‐charges (Figure 1). Of note, we did not ask hospitalists to discern the different charges of 2 appropriate competing clinical choices, but asked for estimated prices of diverse services. The range of respondents' estimates about each item was broad. The mean‐value of hospitalists' estimates for each care item was less revealing than the range and diversity of estimates about each care item. Accuracy of hospitalists' estimates of charges was poor. Only 10.8% of hospitalists' estimates were within 10% of the actual unadjusted charge, 17.8% within 20% of that charge, and 24.8% were within a 30% margin of accuracy. Summary results are presented (Table 1). Pearson's r correlation value between the unadjusted charges and the estimates made by hospitalists was 0.548, a coefficient of determination equal to 0.300. Thus, the true charges list we obtained had only a low‐grade association with hospitalists' estimates (Figure 1). Hospitalists' estimates about the charges of relatively‐expensive items (abdominal computed tomography [CT]) overlapped with their estimates about the least‐expensive items (such as a urine culture ). Inter‐hospitalist agreement about charges associated with each care item was also low; estimates for each care item charge varied over logarithmic orders of magnitude.

Figure 1

Discussion

To date, hospitalist programs have shown little impact upon care costs when compared with other inpatient care staffing models. One limiting factor may be the opacity of medical care pricing. Patients have been demonstrated to have little access to knowledge of what care will cost them and complex barriers prevent them from gaining pricing information.11, 12 Hospitalists may be conjectured to serve as expert sources on the costs and values of medical care services on behalf of inpatients, but our observations suggest that hospitalists' actual knowledge of patient‐charges is lacking. The opacity of US medical prices to patients appears to extend to hospital‐care physicians as well. We observe that no widespread mechanism exists by which hospitalists would be well‐positioned to become informed about the actual charges their patients receive. Unadjusted chargemaster lists are generally restricted information, and would be difficult to access outside of participation in the charge‐notifications used in the existing studies cited above.

The inquiry was specifically limited to how closely hospitalists' estimates of the unadjusted charges for some commonly‐ordered items compare to the actual unadjusted chargemaster at their own institutions. We did not assess the hospitalists' perceptions about the accuracy of their estimates, nor the impact of specific hospitalist characteristics upon accuracy. Our sample's representation of the larger national population of hospital physicians is not established, but engenders no expectation that these clinicians' charge‐awareness is substantively different from that of hospitalists in most other institutions. It is not known what specific clinician or practice‐setting characteristics will direct charge‐awareness, or will influence the impact of charge‐awareness upon clinical practices.

The range of estimates different hospitalists made about the same care items in the same facilities was very broad, which argues that respondents did not estimate charges based upon a different knowledge base of which the investigators were unaware. This is important because our use of unadjusted charges to self‐pay patients as true prices is necessarily somewhat arbitrary. Chargemaster price may not reflect the institution's cost of performing the service, the different prices paid for a single service by different payer sources, nor reflect services' true value based upon outcomes. However, recognizing that these actual prices are somewhat artificial, the use of these prices suffices for the current inquiry, and does not negate our findings of hospitalists' low accuracy and low agreement. Also noteworthy, an unadjusted chargemaster can often represent the charges received by those uninsured US patients for whom payer‐source adjustment is inaccessible, and informs downstream accounting such as the value of unpaid care a hospital delivers annually.

The most immediate matter for examination among hospitalists is what effects increased charge‐awareness may exert upon clinical decisions and practice processes. It appears that the premise that hospitalists' exercise expert knowledge of costs is likely not valid; but it is unknown whether accurate charge‐awareness among hospitalists will improve cost‐reductions by hospitalists. In some payer‐arrangements, accurate charge‐awareness might engender reduced care quantity, rather than increased efficiency or quality. The impact of upgrading hospitalists' knowledge about the charges and costs they generate, and the most effective method to do so, is worthy of investigation; based upon this initial data we encourage and are undertaking a larger‐scale study and exploration of the effects of enhanced hospitalist charge‐awareness.

Conclusion

Hospitalists have low awareness of the charges associated with inpatient care. The opacity of hospital care pricing to patient populations extends also to hospitalist physicians. Hospitalists likely do not improve cost‐efficiency through expert knowledge of services' costs to patients. Education and reminder systems to apprise hospitalists of charges should be examined as possible tools to optimize the use of inpatient care resources.

Hospitalists have been suggested to offer rational and efficient medical care through specialized knowledge about inpatient care services.1 The goal that hospitalists will use care resources more efficiently presumes hospitalists' accurate knowledge about charges and costs.

Data regarding physicians' awareness of care charges and its impact upon care is limited. An international meta‐analysis of clinicians' awareness of pharmaceutical prices demonstrated poor accuracy of physicians' estimates of charges, but effects of increasing their knowledge remained unexamined.2 Continuous exposure to education and alerts about charges have been demonstrated to diminish physicians' unnecessary use of specific laboratory assays in a single teaching hospital, in a pediatric emergency department, and an outpatient primary care system; test use declined when physicians were alerted to test charges at the point‐of‐care without negative impact upon clinical outcomes; when the notices ceased, utilization climbed back towards baseline levels. Specific to inpatient care, a single‐center study evaluated the impact of price‐alerts upon laboratory and imaging use, but showed no effects.36

Applicability of these existing data for contemporary hospitalists are limited, and most data were collected before hospitalists developed as an organized focus of practice. A review of existing literature revealed no published data demonstrating hospitalists' higher expert awareness of charges generated by inpatient care. Published comparisons of the care expense generated by hospitalists' care versus that of general internists or academic teams have shown minimal and inconsistent effects.79 Those data showing reduced costs from hospitalists were associated with small length‐of‐stay reductions, rather than more expert resource utilization.7 We measured the accuracy and precision of hospitalist's estimates of charges associated with services commonly used in inpatient care.

Setting

Two community‐based private, academic‐affiliated hospitals operated by a not‐for‐profit health system in Washington State, comprising together 895 inpatient beds. The questionnaire instrument was approved by the governing Institutional Review Board (IRB).

Methods

A list of true charges for 14 services, procedures, tests, and physician charges commonly ordered by adult medicine hospitalists was acquired directly from the responsible departments of a multi‐hospital system operated by a single non‐profit entity using a unified chargemaster. Specifically, we acquired the charge that a hypothetical self‐paying patient would receive for each service, excluding any adjustments exercised by other payer sources. The list of charges was reported to the organization's financial officers for affirmation. Physician charges were standardized to geographically‐adjusted Medicare charges obtained directly from the American Medical Assocation's online Common Procedural Terminology tool.10 A cross‐section of hospitalists (n = 25) was surveyed from a private hospitalist group and an academically‐affiliated hospitalist service. Hospitalists included US and international medical graduates, new‐graduates from residency training and clinicians with a range of prior experiences in academic centers, government hospital systems, and private primary care. Respondents were asked to estimate to the nearest dollar the billing charge that a hypothetical self‐pay patient would receive for each care item. Direct data collection was arranged through the groups' medical directors and occurred in the hospitalist groups' regular business meetings. The design gathered no data on individual characteristics of respondents such as domestic or international education, sex, age, or time in practice, nor from which practice‐group a given respondent originated, to affirm to participants that their responses could not be used to imply performance measures or quality profiles.

Findings

Hospitalists tended to rank the expense of items in essentially the correct order, as reflected by the rough trend of rising estimates compared to true‐charges (Figure 1). Of note, we did not ask hospitalists to discern the different charges of 2 appropriate competing clinical choices, but asked for estimated prices of diverse services. The range of respondents' estimates about each item was broad. The mean‐value of hospitalists' estimates for each care item was less revealing than the range and diversity of estimates about each care item. Accuracy of hospitalists' estimates of charges was poor. Only 10.8% of hospitalists' estimates were within 10% of the actual unadjusted charge, 17.8% within 20% of that charge, and 24.8% were within a 30% margin of accuracy. Summary results are presented (Table 1). Pearson's r correlation value between the unadjusted charges and the estimates made by hospitalists was 0.548, a coefficient of determination equal to 0.300. Thus, the true charges list we obtained had only a low‐grade association with hospitalists' estimates (Figure 1). Hospitalists' estimates about the charges of relatively‐expensive items (abdominal computed tomography [CT]) overlapped with their estimates about the least‐expensive items (such as a urine culture ). Inter‐hospitalist agreement about charges associated with each care item was also low; estimates for each care item charge varied over logarithmic orders of magnitude.

Figure 1

Discussion

To date, hospitalist programs have shown little impact upon care costs when compared with other inpatient care staffing models. One limiting factor may be the opacity of medical care pricing. Patients have been demonstrated to have little access to knowledge of what care will cost them and complex barriers prevent them from gaining pricing information.11, 12 Hospitalists may be conjectured to serve as expert sources on the costs and values of medical care services on behalf of inpatients, but our observations suggest that hospitalists' actual knowledge of patient‐charges is lacking. The opacity of US medical prices to patients appears to extend to hospital‐care physicians as well. We observe that no widespread mechanism exists by which hospitalists would be well‐positioned to become informed about the actual charges their patients receive. Unadjusted chargemaster lists are generally restricted information, and would be difficult to access outside of participation in the charge‐notifications used in the existing studies cited above.

The inquiry was specifically limited to how closely hospitalists' estimates of the unadjusted charges for some commonly‐ordered items compare to the actual unadjusted chargemaster at their own institutions. We did not assess the hospitalists' perceptions about the accuracy of their estimates, nor the impact of specific hospitalist characteristics upon accuracy. Our sample's representation of the larger national population of hospital physicians is not established, but engenders no expectation that these clinicians' charge‐awareness is substantively different from that of hospitalists in most other institutions. It is not known what specific clinician or practice‐setting characteristics will direct charge‐awareness, or will influence the impact of charge‐awareness upon clinical practices.

The range of estimates different hospitalists made about the same care items in the same facilities was very broad, which argues that respondents did not estimate charges based upon a different knowledge base of which the investigators were unaware. This is important because our use of unadjusted charges to self‐pay patients as true prices is necessarily somewhat arbitrary. Chargemaster price may not reflect the institution's cost of performing the service, the different prices paid for a single service by different payer sources, nor reflect services' true value based upon outcomes. However, recognizing that these actual prices are somewhat artificial, the use of these prices suffices for the current inquiry, and does not negate our findings of hospitalists' low accuracy and low agreement. Also noteworthy, an unadjusted chargemaster can often represent the charges received by those uninsured US patients for whom payer‐source adjustment is inaccessible, and informs downstream accounting such as the value of unpaid care a hospital delivers annually.

The most immediate matter for examination among hospitalists is what effects increased charge‐awareness may exert upon clinical decisions and practice processes. It appears that the premise that hospitalists' exercise expert knowledge of costs is likely not valid; but it is unknown whether accurate charge‐awareness among hospitalists will improve cost‐reductions by hospitalists. In some payer‐arrangements, accurate charge‐awareness might engender reduced care quantity, rather than increased efficiency or quality. The impact of upgrading hospitalists' knowledge about the charges and costs they generate, and the most effective method to do so, is worthy of investigation; based upon this initial data we encourage and are undertaking a larger‐scale study and exploration of the effects of enhanced hospitalist charge‐awareness.

Conclusion

Hospitalists have low awareness of the charges associated with inpatient care. The opacity of hospital care pricing to patient populations extends also to hospitalist physicians. Hospitalists likely do not improve cost‐efficiency through expert knowledge of services' costs to patients. Education and reminder systems to apprise hospitalists of charges should be examined as possible tools to optimize the use of inpatient care resources.

The demand for physician talent is intensifying as the US healthcare system confronts an unprecedented confluence of demographic pressures. Not only will 78 million retiring baby‐boomers require significant healthcare resources, but tens of thousands of practicing physicians will themselves reach retirement age within the next decade.1 At the same time, factors like the large increase in the percentage of female physicians (who are more likely to work part time), the growth of nonpractice opportunities for MDs, and generational demands for greater worklife balance are creating a major supply‐demand mismatch within the physician workforce.2 In this demographic atmosphere, the ability to recruit and retain physician leaders confers tremendous value to healthcare enterprisesboth public and private. Recruiting and retaining strategies already weigh heavily in the most palpable shortage areas, like primary care, but the system faces widespread unmet demand for a variety of specialist and generalist practitioners.36

This article does not address the public policy implications of the upcoming physician shortage, recognition of which will lead to the largest increase in new medical school slots in decades.7 Rather, we set out to illustrate how successful nonmedical businesses are embracing a thoughtful, systematic approach to retaining talent, based on the philosophy that keeping and engaging valued employees is more efficient than recruiting and orienting replacements. We posit that the innovations used by progressive companies could apply to recruitment and retention challenges confronting medicine.

How Industry Approaches the Talent Vacuum: Talent Facilitation

Leaders outside of medicine have long acknowledged that changing demographics and a global economy are driving unprecedented employee turnover.8 In confronting a talent vacuum, forward‐thinking managers have prioritized retaining key talent (rather than hiring anew) in planning for the future.9, 10 Doing so begins with attempts to understand the relationship between workers and the workplace, with a particular emphasis on appreciating workers' priorities. Indeed, new executive positions with titles like Chief Learning Officer and Chief Experience Officer are appearing as companies realize a need for focused expertise beyond traditional human resource departments. These companies understand that offering higher salaries is not the only retention strategyand often not even the most effective one.

The Four Actions of Talent Facilitation

The talent facilitation process centers on four actions: attract, engage, develop, and retain. None of these actions can stand alone, and all should be present, to some degree, at all stages of a worker's tenure. To attract or engage an employee or practice partner is not a 1‐time hook, but a constant and dynamic process.

Importantly, the concepts addressed here are not specific to 1 type of corporate system, size, or management level. Although the early business focus had been on upper‐level and executive talent within large corporate settings, there is an increasing recognition that a dedicated talent strategy is useful wherever recruiting and retaining talented people is important (and where is it not?).

The ideas presented here may seem most applicable to leaders of large physician corporations, hospital‐owned physician groups, or large integrated healthcare systems (such as Kaiser) that employ physicians. However, we also believe that the ideas apply across‐the‐board in medicine, including entities such as small, private, physician‐owned groups. We argue that regardless of the exact practice structure, a limited pool of resources must be dedicated to the attraction and retention of talented partners or employees, or to the cost of replacing those people if they pursue other opportunities (or the cost of inefficient and disengaged physicians). While an integrated health system may have the resources and scale to hire a Chief Experience Officer, we do not anticipate that a 5‐partner private practice would. Rather, we point to examples to illustrate the talent facilitation paradigm as a tool to systematically frame the allocation of those resources. Undoubtedly, the specific shape of a thoughtful talent facilitation effort will vary when applied in a large urban academic medical center vs. an integrated healthcare system vs. a small physician private practice, but the basic principles remain the same.

Attract

Increasingly, companies approach talented prospects with dedicated marketing campaigns to convey the value of a work environment.11 Silicon Valley employee lounges with free massages and foosball tables are the iconic example of attraction, but the concept runs deeper. Today's workers seek access to state‐of‐the‐art ideas and technology and often want to be part of a larger vision. Many seek opportunities to integrate their own professional and personal aspirations into a particular job description.

Hospital executives have long recognized the importance of attracting physicians to their facility (after all, the physicians draw patients and thus generate revenue). The traditional approach has surrounded perks, from comfortable doctor's lounges to the latest in surgical technology. But, the stakes seem greater now than before, and successful talent facilitation strategies are going beyond the tried and true.

Clearly, physicians seek financially stable practice settings with historical success. But they may also seek evidence of a defined strategic plan focused on more than mere profitability. Physicians may gravitate to practice environments that endorse progressive movements like the No One Dies Alone campaign.12 Similarly, recognition of movements beyond healthcarea commitment to Leadership in Energy and Environmental Design (LEED) (Green Building Rating System; US Green Building Council [USGBC]; http://www.usgbc.org) designation,13 for examplemay help attract talented staff physicians or enhance a hospital system's ability to facilitate partnerships with desirable physician groups. Within small private practices, this may take the form of a collaborative dedication to physician wellness and burnout prevention, or in a shared commitment to some form of local or international service.

The current recruitment campaign of California's prison healthcare system offers an unlikely source of inspiration. The prison system was placed in receivership to address a shortage of competent physician staff and other inadequacies. A central feature of the campaign is an attractive starting salary and good benefits. But, the campaign does not rely on money alone. For example, the campaign's website (http://changingprisonhealthcare.org) promises that prison physicians will create the standard for correctional healthcare and join an historic effort to make a difference in people's lives (Figure 1). The campaign thus rebrands what might be seen as an unpalatable job into a legitimate and noble career option. Of course, the long‐term success of the campaign will rely on a genuine commitment to the campaign's ideals. But the point is this: if a prison system can deploy innovative attraction strategies, so can healthcare leaders at all levels.

Figure 1

Lessons From Industry

Doctors frequently assume that the challenges and obstacles confronted in healthcare are unique to medicine. But, for every phrase like When I started practice, I decided how long office visits were, not the insurance company, or Young doctors just don't want to work as hard, there is a parallel utterance in the greater business world. Luckily, there are now examples of the healthcare world learning lessons from business. For example, innovators in medical quality improvement found value in the experience of other industries.20 Airlines and automakers have long honed systems for error prevention and possess expertise that may curb errors in the hospital.2123

We suspect that the ideas and practice of talent facilitation have already made their way into some medical settings. A Google search reveals multiple opportunities for hospital‐based talent managers, and websites advertise the availability of talent consultants ready to lend their expertise to the medical world. In the arena of academic medicine, the University of California, San Francisco (UCSF) Division of Hospital Medicine put some of the ideas of talent facilitation into practice over the past year, in part in response to an increasingly competitive market for academic hospitalists.24 Leaders introduced a formal faculty development program that links junior faculty with mentors and facilitates early and frequent feedback across hierarchical boundaries.25 These more intentional mentoring efforts were accompanied by a seminar series aimed at the needs of new faculty members, a research incubator program, divisional grand rounds, and other web‐based and in‐person forums for sharing best practices and innovations. Less formal social events have also been promoted. Importantly, these sweeping strategies seek to encompass the needs of both teaching and nonteaching hospitalists within UCSF.26

Clearly, an academic hospitalist group with 45 faculty physicians has unique characteristics that inform the specifics of its talent facilitation strategy. The interventions discussed above are meant to represent examples of the types of strategies that may be utilized by physician groups once a decision is made to focus on talent management. Undoubtedly, the shape of such efforts will vary in diverse practice settings, but physician leaders have much to gain through further exploration of where these core principles already exist within medicine and where they may be more effectively deployed. By examining how multinational businesses are systematically applying the concepts of talent facilitation to address a global talent shortage, the doctoring profession might again take an outside hint to help inform its future.

The demand for physician talent is intensifying as the US healthcare system confronts an unprecedented confluence of demographic pressures. Not only will 78 million retiring baby‐boomers require significant healthcare resources, but tens of thousands of practicing physicians will themselves reach retirement age within the next decade.1 At the same time, factors like the large increase in the percentage of female physicians (who are more likely to work part time), the growth of nonpractice opportunities for MDs, and generational demands for greater worklife balance are creating a major supply‐demand mismatch within the physician workforce.2 In this demographic atmosphere, the ability to recruit and retain physician leaders confers tremendous value to healthcare enterprisesboth public and private. Recruiting and retaining strategies already weigh heavily in the most palpable shortage areas, like primary care, but the system faces widespread unmet demand for a variety of specialist and generalist practitioners.36

This article does not address the public policy implications of the upcoming physician shortage, recognition of which will lead to the largest increase in new medical school slots in decades.7 Rather, we set out to illustrate how successful nonmedical businesses are embracing a thoughtful, systematic approach to retaining talent, based on the philosophy that keeping and engaging valued employees is more efficient than recruiting and orienting replacements. We posit that the innovations used by progressive companies could apply to recruitment and retention challenges confronting medicine.

How Industry Approaches the Talent Vacuum: Talent Facilitation

Leaders outside of medicine have long acknowledged that changing demographics and a global economy are driving unprecedented employee turnover.8 In confronting a talent vacuum, forward‐thinking managers have prioritized retaining key talent (rather than hiring anew) in planning for the future.9, 10 Doing so begins with attempts to understand the relationship between workers and the workplace, with a particular emphasis on appreciating workers' priorities. Indeed, new executive positions with titles like Chief Learning Officer and Chief Experience Officer are appearing as companies realize a need for focused expertise beyond traditional human resource departments. These companies understand that offering higher salaries is not the only retention strategyand often not even the most effective one.

The Four Actions of Talent Facilitation

The talent facilitation process centers on four actions: attract, engage, develop, and retain. None of these actions can stand alone, and all should be present, to some degree, at all stages of a worker's tenure. To attract or engage an employee or practice partner is not a 1‐time hook, but a constant and dynamic process.

Importantly, the concepts addressed here are not specific to 1 type of corporate system, size, or management level. Although the early business focus had been on upper‐level and executive talent within large corporate settings, there is an increasing recognition that a dedicated talent strategy is useful wherever recruiting and retaining talented people is important (and where is it not?).

The ideas presented here may seem most applicable to leaders of large physician corporations, hospital‐owned physician groups, or large integrated healthcare systems (such as Kaiser) that employ physicians. However, we also believe that the ideas apply across‐the‐board in medicine, including entities such as small, private, physician‐owned groups. We argue that regardless of the exact practice structure, a limited pool of resources must be dedicated to the attraction and retention of talented partners or employees, or to the cost of replacing those people if they pursue other opportunities (or the cost of inefficient and disengaged physicians). While an integrated health system may have the resources and scale to hire a Chief Experience Officer, we do not anticipate that a 5‐partner private practice would. Rather, we point to examples to illustrate the talent facilitation paradigm as a tool to systematically frame the allocation of those resources. Undoubtedly, the specific shape of a thoughtful talent facilitation effort will vary when applied in a large urban academic medical center vs. an integrated healthcare system vs. a small physician private practice, but the basic principles remain the same.

Attract

Increasingly, companies approach talented prospects with dedicated marketing campaigns to convey the value of a work environment.11 Silicon Valley employee lounges with free massages and foosball tables are the iconic example of attraction, but the concept runs deeper. Today's workers seek access to state‐of‐the‐art ideas and technology and often want to be part of a larger vision. Many seek opportunities to integrate their own professional and personal aspirations into a particular job description.

Hospital executives have long recognized the importance of attracting physicians to their facility (after all, the physicians draw patients and thus generate revenue). The traditional approach has surrounded perks, from comfortable doctor's lounges to the latest in surgical technology. But, the stakes seem greater now than before, and successful talent facilitation strategies are going beyond the tried and true.

Clearly, physicians seek financially stable practice settings with historical success. But they may also seek evidence of a defined strategic plan focused on more than mere profitability. Physicians may gravitate to practice environments that endorse progressive movements like the No One Dies Alone campaign.12 Similarly, recognition of movements beyond healthcarea commitment to Leadership in Energy and Environmental Design (LEED) (Green Building Rating System; US Green Building Council [USGBC]; http://www.usgbc.org) designation,13 for examplemay help attract talented staff physicians or enhance a hospital system's ability to facilitate partnerships with desirable physician groups. Within small private practices, this may take the form of a collaborative dedication to physician wellness and burnout prevention, or in a shared commitment to some form of local or international service.

The current recruitment campaign of California's prison healthcare system offers an unlikely source of inspiration. The prison system was placed in receivership to address a shortage of competent physician staff and other inadequacies. A central feature of the campaign is an attractive starting salary and good benefits. But, the campaign does not rely on money alone. For example, the campaign's website (http://changingprisonhealthcare.org) promises that prison physicians will create the standard for correctional healthcare and join an historic effort to make a difference in people's lives (Figure 1). The campaign thus rebrands what might be seen as an unpalatable job into a legitimate and noble career option. Of course, the long‐term success of the campaign will rely on a genuine commitment to the campaign's ideals. But the point is this: if a prison system can deploy innovative attraction strategies, so can healthcare leaders at all levels.

Figure 1

Lessons From Industry

Doctors frequently assume that the challenges and obstacles confronted in healthcare are unique to medicine. But, for every phrase like When I started practice, I decided how long office visits were, not the insurance company, or Young doctors just don't want to work as hard, there is a parallel utterance in the greater business world. Luckily, there are now examples of the healthcare world learning lessons from business. For example, innovators in medical quality improvement found value in the experience of other industries.20 Airlines and automakers have long honed systems for error prevention and possess expertise that may curb errors in the hospital.2123

We suspect that the ideas and practice of talent facilitation have already made their way into some medical settings. A Google search reveals multiple opportunities for hospital‐based talent managers, and websites advertise the availability of talent consultants ready to lend their expertise to the medical world. In the arena of academic medicine, the University of California, San Francisco (UCSF) Division of Hospital Medicine put some of the ideas of talent facilitation into practice over the past year, in part in response to an increasingly competitive market for academic hospitalists.24 Leaders introduced a formal faculty development program that links junior faculty with mentors and facilitates early and frequent feedback across hierarchical boundaries.25 These more intentional mentoring efforts were accompanied by a seminar series aimed at the needs of new faculty members, a research incubator program, divisional grand rounds, and other web‐based and in‐person forums for sharing best practices and innovations. Less formal social events have also been promoted. Importantly, these sweeping strategies seek to encompass the needs of both teaching and nonteaching hospitalists within UCSF.26

Clearly, an academic hospitalist group with 45 faculty physicians has unique characteristics that inform the specifics of its talent facilitation strategy. The interventions discussed above are meant to represent examples of the types of strategies that may be utilized by physician groups once a decision is made to focus on talent management. Undoubtedly, the shape of such efforts will vary in diverse practice settings, but physician leaders have much to gain through further exploration of where these core principles already exist within medicine and where they may be more effectively deployed. By examining how multinational businesses are systematically applying the concepts of talent facilitation to address a global talent shortage, the doctoring profession might again take an outside hint to help inform its future.

The demand for physician talent is intensifying as the US healthcare system confronts an unprecedented confluence of demographic pressures. Not only will 78 million retiring baby‐boomers require significant healthcare resources, but tens of thousands of practicing physicians will themselves reach retirement age within the next decade.1 At the same time, factors like the large increase in the percentage of female physicians (who are more likely to work part time), the growth of nonpractice opportunities for MDs, and generational demands for greater worklife balance are creating a major supply‐demand mismatch within the physician workforce.2 In this demographic atmosphere, the ability to recruit and retain physician leaders confers tremendous value to healthcare enterprisesboth public and private. Recruiting and retaining strategies already weigh heavily in the most palpable shortage areas, like primary care, but the system faces widespread unmet demand for a variety of specialist and generalist practitioners.36

This article does not address the public policy implications of the upcoming physician shortage, recognition of which will lead to the largest increase in new medical school slots in decades.7 Rather, we set out to illustrate how successful nonmedical businesses are embracing a thoughtful, systematic approach to retaining talent, based on the philosophy that keeping and engaging valued employees is more efficient than recruiting and orienting replacements. We posit that the innovations used by progressive companies could apply to recruitment and retention challenges confronting medicine.

How Industry Approaches the Talent Vacuum: Talent Facilitation

Leaders outside of medicine have long acknowledged that changing demographics and a global economy are driving unprecedented employee turnover.8 In confronting a talent vacuum, forward‐thinking managers have prioritized retaining key talent (rather than hiring anew) in planning for the future.9, 10 Doing so begins with attempts to understand the relationship between workers and the workplace, with a particular emphasis on appreciating workers' priorities. Indeed, new executive positions with titles like Chief Learning Officer and Chief Experience Officer are appearing as companies realize a need for focused expertise beyond traditional human resource departments. These companies understand that offering higher salaries is not the only retention strategyand often not even the most effective one.

The Four Actions of Talent Facilitation

The talent facilitation process centers on four actions: attract, engage, develop, and retain. None of these actions can stand alone, and all should be present, to some degree, at all stages of a worker's tenure. To attract or engage an employee or practice partner is not a 1‐time hook, but a constant and dynamic process.

Importantly, the concepts addressed here are not specific to 1 type of corporate system, size, or management level. Although the early business focus had been on upper‐level and executive talent within large corporate settings, there is an increasing recognition that a dedicated talent strategy is useful wherever recruiting and retaining talented people is important (and where is it not?).

The ideas presented here may seem most applicable to leaders of large physician corporations, hospital‐owned physician groups, or large integrated healthcare systems (such as Kaiser) that employ physicians. However, we also believe that the ideas apply across‐the‐board in medicine, including entities such as small, private, physician‐owned groups. We argue that regardless of the exact practice structure, a limited pool of resources must be dedicated to the attraction and retention of talented partners or employees, or to the cost of replacing those people if they pursue other opportunities (or the cost of inefficient and disengaged physicians). While an integrated health system may have the resources and scale to hire a Chief Experience Officer, we do not anticipate that a 5‐partner private practice would. Rather, we point to examples to illustrate the talent facilitation paradigm as a tool to systematically frame the allocation of those resources. Undoubtedly, the specific shape of a thoughtful talent facilitation effort will vary when applied in a large urban academic medical center vs. an integrated healthcare system vs. a small physician private practice, but the basic principles remain the same.

Attract

Increasingly, companies approach talented prospects with dedicated marketing campaigns to convey the value of a work environment.11 Silicon Valley employee lounges with free massages and foosball tables are the iconic example of attraction, but the concept runs deeper. Today's workers seek access to state‐of‐the‐art ideas and technology and often want to be part of a larger vision. Many seek opportunities to integrate their own professional and personal aspirations into a particular job description.

Hospital executives have long recognized the importance of attracting physicians to their facility (after all, the physicians draw patients and thus generate revenue). The traditional approach has surrounded perks, from comfortable doctor's lounges to the latest in surgical technology. But, the stakes seem greater now than before, and successful talent facilitation strategies are going beyond the tried and true.

Clearly, physicians seek financially stable practice settings with historical success. But they may also seek evidence of a defined strategic plan focused on more than mere profitability. Physicians may gravitate to practice environments that endorse progressive movements like the No One Dies Alone campaign.12 Similarly, recognition of movements beyond healthcarea commitment to Leadership in Energy and Environmental Design (LEED) (Green Building Rating System; US Green Building Council [USGBC]; http://www.usgbc.org) designation,13 for examplemay help attract talented staff physicians or enhance a hospital system's ability to facilitate partnerships with desirable physician groups. Within small private practices, this may take the form of a collaborative dedication to physician wellness and burnout prevention, or in a shared commitment to some form of local or international service.

The current recruitment campaign of California's prison healthcare system offers an unlikely source of inspiration. The prison system was placed in receivership to address a shortage of competent physician staff and other inadequacies. A central feature of the campaign is an attractive starting salary and good benefits. But, the campaign does not rely on money alone. For example, the campaign's website (http://changingprisonhealthcare.org) promises that prison physicians will create the standard for correctional healthcare and join an historic effort to make a difference in people's lives (Figure 1). The campaign thus rebrands what might be seen as an unpalatable job into a legitimate and noble career option. Of course, the long‐term success of the campaign will rely on a genuine commitment to the campaign's ideals. But the point is this: if a prison system can deploy innovative attraction strategies, so can healthcare leaders at all levels.

Figure 1

Lessons From Industry

Doctors frequently assume that the challenges and obstacles confronted in healthcare are unique to medicine. But, for every phrase like When I started practice, I decided how long office visits were, not the insurance company, or Young doctors just don't want to work as hard, there is a parallel utterance in the greater business world. Luckily, there are now examples of the healthcare world learning lessons from business. For example, innovators in medical quality improvement found value in the experience of other industries.20 Airlines and automakers have long honed systems for error prevention and possess expertise that may curb errors in the hospital.2123

We suspect that the ideas and practice of talent facilitation have already made their way into some medical settings. A Google search reveals multiple opportunities for hospital‐based talent managers, and websites advertise the availability of talent consultants ready to lend their expertise to the medical world. In the arena of academic medicine, the University of California, San Francisco (UCSF) Division of Hospital Medicine put some of the ideas of talent facilitation into practice over the past year, in part in response to an increasingly competitive market for academic hospitalists.24 Leaders introduced a formal faculty development program that links junior faculty with mentors and facilitates early and frequent feedback across hierarchical boundaries.25 These more intentional mentoring efforts were accompanied by a seminar series aimed at the needs of new faculty members, a research incubator program, divisional grand rounds, and other web‐based and in‐person forums for sharing best practices and innovations. Less formal social events have also been promoted. Importantly, these sweeping strategies seek to encompass the needs of both teaching and nonteaching hospitalists within UCSF.26

Clearly, an academic hospitalist group with 45 faculty physicians has unique characteristics that inform the specifics of its talent facilitation strategy. The interventions discussed above are meant to represent examples of the types of strategies that may be utilized by physician groups once a decision is made to focus on talent management. Undoubtedly, the shape of such efforts will vary in diverse practice settings, but physician leaders have much to gain through further exploration of where these core principles already exist within medicine and where they may be more effectively deployed. By examining how multinational businesses are systematically applying the concepts of talent facilitation to address a global talent shortage, the doctoring profession might again take an outside hint to help inform its future.

Acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are responsible for more than 600,000 hospitalizations annually, resulting in direct costs of over $20 billion.1 Bacterial infections appear responsible for 50% of such exacerbations,25 and current COPD guidelines recommend treatment with antibiotics for patients with severe exacerbations or a change in sputum.1^,69 These recommendations are based on a number of small randomized trials, most of which were conducted more than 20 years ago using narrow spectrum antibiotics that are no longer commonly used.10 Only 4 studies, totaling 321 subjects, included hospitalized patients, and most studies excluded patients who required steroids. Because no clinical trials have compared different antibiotic regimens for AECOPD, existing guidelines offer a range of treatment options, including amoxicillin‐clavulonate, macrolides, quinolones, cephalosporins, aminopenicillins, and tetracyclines.

Among hospitalized patients, macrolides and quinolones appear to be the most frequently prescribed antibiotics.11 Both are available in oral formulations, have excellent bioavailability, and are administered once daily. In addition to their antimicrobial activity, macrolides are believed to have antiinflammatory effects, which could be especially advantageous in AECOPD.1214 In trials of chronic bronchitis, however, fluoroquinolones have been shown to reduce the risk of recurrent exacerbation when compared to macrolides.15 The wide variation that has been observed in antibiotic selection for patients hospitalized for AECOPD suggests a high degree of uncertainty among clinicians about the benefits of different treatment options.11 Given the limited evidence from randomized trials, we sought to evaluate the comparative effectiveness of macrolides and quinolones among a large, representative sample of patients hospitalized with AECOPD.

Subjects and Methods

Results

Of 26,248 AECOPD patients treated with antibiotics, 19,608 patients met the inclusion criteria; of these, 6139 (31%) were treated initially with a macrolide; the median age was 70 years; 60% were female; and 78% were white. A total of 86% of patients had a primary diagnosis of obstructive chronic bronchitis with acute exacerbation, and 6% had respiratory failure. The most common comorbidities were hypertension, diabetes, and congestive heart failure. Twenty‐two percent had been admitted at least once in the preceding 12 months. Treatment failure occurred in 7.7% of patients, and 1.3% died in the hospital. Mean length of stay was 4.8 days. Hospital prescribing rates for macrolides varied from 0% to 100%, with a mean of 33% and an interquartile range of 14% to 46% (Supporting Appendix Figure 1).

Compared to patients receiving macrolides, those receiving quinolones were older, more likely to have respiratory failure, to be cared for by a pulmonologist, and to have an admission in the previous year (Table 1). They were also more likely to be treated with bronchodilators, methylxanthines, steroids, diuretics, and noninvasive positive pressure ventilation, and to have an arterial blood gas, but less likely to receive concomitant treatment with a cephalosporin (11% vs. 57%). With the exception of cephalosporin treatment, these differences were small, but due to the large sample were statistically significant. Comorbidities were similar in both groups. Patients in the quinolone group were also more likely to experience treatment failure (8.1% vs. 6.8%), death (1.5% vs. 1.0%), and antibiotic‐associated diarrhea (1.1% vs. 0.5%).

In the unadjusted analysis, compared to patients receiving quinolones, those treated with macrolides were less likely to experience treatment failure (OR, 0.83; 95% CI, 0.740.94) (Table 2). Adjusting for all patient, hospital, and physician covariates, including the propensity for treatment with macrolides, increased the OR to 0.89 and the results were no longer significant (95% CI, 0.781.01). Propensity matching successfully balanced all measured covariates except for the use of short‐acting bronchodilators and additional antibiotics (Table 1). In the propensity‐matched sample (Figure 1), quinolone‐treated patients were more likely to experience antibiotic‐associated diarrhea (1.2% vs. 0.6%; P = 0.0003) and late mechanical ventilation (1.3% vs. 0.8%; P = 0.02). There were no differences in adjusted cost or length of stay between the 2 groups. The results of the grouped treatment analysis, substituting the hospital's specific rate of macrolide use in place of the actual treatment that each patient received suggested that the 2 antibiotics were associated with similar rates of treatment failure. The OR for a 100% hospital rate of macrolide treatment vs. a 0% rate was 1.01 (95% CI, 0.751.35).

Figure 1

Discussion

In this large observational study conducted at 375 hospitals, we took advantage of a natural experiment in which antibiotic prescribing patterns varied widely across hospitals to compare the effectiveness of 2 common antibiotic regimens for AECOPD. Treatment with macrolides and quinolones were associated with a similar risk of treatment failure, costs, and length of stay; however, patients treated with macrolides were less likely to experience late mechanical ventilation or treatment for antibiotic‐associated diarrhea.

Despite broad consensus in COPD guidelines that patients with severe acute exacerbations should receive antibiotics, there is little agreement about the preferred empiric agent. Controversy exists regarding antibiotics' comparative effectiveness, and even over which pathogens cause COPD exacerbations. Given the frequency of hospitalization for AECOPD, understanding the comparative effectiveness of treatments in this setting could have important implications for health outcomes and costs. Unfortunately, most antibiotic studies in AECOPD were conducted >20 years ago, using antibiotics that rarely appeared in our sample.26 Consequently, clinical practice guidelines offer conflicting recommendations. For example, the National Institute for Clinical Excellence recommends empirical treatment with an aminopenicillin, a macrolide, or a tetracycline,8 while the American Thoracic Society recommends amoxicillin‐clavulonate or a fluoroquinolone.9

As might be expected in light of so much uncertainty, we found wide variation in prescribing patterns across hospitals. Overall, approximately one‐third of patients received a macrolide and two‐thirds a quinolone. Both regimens provide adequate coverage of H. influenza, S. pneumoniae, and M. catarrhalis, and conform to at least 1 COPD guideline. Nevertheless, patients receiving macrolides often received a cephalosporin as well; this pattern of treatment suggests that antibiotic selection is likely to have been influenced more by guidelines for the treatment of community‐acquired pneumonia than COPD.

Previous studies comparing antibiotic effectiveness suffer from shortcomings that limit their application to patients hospitalized with AECOPD. First, they enrolled patients with chronic bronchitis, and included patients without obstructive lung disease, and most studies included patients as young as 18 years old. Second, many either did not include treatment with steroids or excluded patients receiving more than 10 mg of prednisone daily. Third, almost all enrolled only ambulatory patients.

While there are no studies comparing quinolones and macrolides in patients hospitalized for AECOPD, a meta‐analysis comparing quinolones, macrolides, and amoxicillin‐clavulonate identified 19 trials of ambulatory patients with chronic bronchitis. That study found that all 3 drugs had similar efficacy initially, but that quinolones resulted in the fewest relapses over a 26‐week period.27 Macrolides and quinolones had similar rates of adverse effects. In contrast, we did not find a difference in treatment failure, cost, or length of stay, but did find a higher rate of diarrhea associated with quinolones. Others have also documented an association between fluoroquinolones and C. difficile diarrhea.2830 This trend, first noted in 2001, is of particular concern because the fluoroquinolone‐resistant strains appear to be hypervirulent and have been associated with nosocomial epidemics.3134

Our study has several limitations. First, its observational design leaves open the possibility of selection bias. For this reason we analyzed our data in several ways, including using a grouped treatment approach, an adaptation of the instrumental variable technique, and accepted only those differences which were consistent across all models. Second, our study used claims data, and therefore we could not directly adjust for physiological measures of severity. However, the highly detailed nature of the data allowed us to adjust for numerous tests and treatments that reflected the clinician's assessment of the patient's severity, as well as the number of prior COPD admissions. Third, we cannot exclude the possibility that some patients may have had concurrent pneumonia without an ICD‐9 code. We think that the number would be small because reimbursement for pneumonia is generally higher than for COPD, so hospitals have an incentive to code pneumonia as the principal diagnosis when present. Finally, we compared initial antibiotics only. More than one‐quarter of our patients received an additional antibiotic before discharge. In particular, patients receiving macrolides were often prescribed a concomitant cephalosporin. We do not know to what extent these additional antibiotics may have affected the outcomes.

Despite the large number of patients hospitalized annually for AECOPD, there are no randomized trials comparing different antibiotics in this population. Studies comparing antibiotics in chronic bronchitis can offer little guidance, since they have primarily focused on proving equivalence between existing antibiotics and newer, more expensive formulations.35 Because many of the patients enrolled in such trials do not benefit from antibiotics at all, either because they do not have COPD or because their exacerbation is not caused by bacteria, it is relatively easy to prove equivalence. Given that AECOPD is one of the leading causes of hospitalization in the United States, large, randomized trials comparing the effectiveness of different antibiotics should be a high priority. In the meantime, macrolides (often given together with cephalosporins) and quinolones appear to be equally effective initial antibiotic choices; considering antibiotic‐associated diarrhea, macrolides appear to be the safer of the 2.

Acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are responsible for more than 600,000 hospitalizations annually, resulting in direct costs of over $20 billion.1 Bacterial infections appear responsible for 50% of such exacerbations,25 and current COPD guidelines recommend treatment with antibiotics for patients with severe exacerbations or a change in sputum.1^,69 These recommendations are based on a number of small randomized trials, most of which were conducted more than 20 years ago using narrow spectrum antibiotics that are no longer commonly used.10 Only 4 studies, totaling 321 subjects, included hospitalized patients, and most studies excluded patients who required steroids. Because no clinical trials have compared different antibiotic regimens for AECOPD, existing guidelines offer a range of treatment options, including amoxicillin‐clavulonate, macrolides, quinolones, cephalosporins, aminopenicillins, and tetracyclines.

Among hospitalized patients, macrolides and quinolones appear to be the most frequently prescribed antibiotics.11 Both are available in oral formulations, have excellent bioavailability, and are administered once daily. In addition to their antimicrobial activity, macrolides are believed to have antiinflammatory effects, which could be especially advantageous in AECOPD.1214 In trials of chronic bronchitis, however, fluoroquinolones have been shown to reduce the risk of recurrent exacerbation when compared to macrolides.15 The wide variation that has been observed in antibiotic selection for patients hospitalized for AECOPD suggests a high degree of uncertainty among clinicians about the benefits of different treatment options.11 Given the limited evidence from randomized trials, we sought to evaluate the comparative effectiveness of macrolides and quinolones among a large, representative sample of patients hospitalized with AECOPD.

Subjects and Methods

Results

Of 26,248 AECOPD patients treated with antibiotics, 19,608 patients met the inclusion criteria; of these, 6139 (31%) were treated initially with a macrolide; the median age was 70 years; 60% were female; and 78% were white. A total of 86% of patients had a primary diagnosis of obstructive chronic bronchitis with acute exacerbation, and 6% had respiratory failure. The most common comorbidities were hypertension, diabetes, and congestive heart failure. Twenty‐two percent had been admitted at least once in the preceding 12 months. Treatment failure occurred in 7.7% of patients, and 1.3% died in the hospital. Mean length of stay was 4.8 days. Hospital prescribing rates for macrolides varied from 0% to 100%, with a mean of 33% and an interquartile range of 14% to 46% (Supporting Appendix Figure 1).

Compared to patients receiving macrolides, those receiving quinolones were older, more likely to have respiratory failure, to be cared for by a pulmonologist, and to have an admission in the previous year (Table 1). They were also more likely to be treated with bronchodilators, methylxanthines, steroids, diuretics, and noninvasive positive pressure ventilation, and to have an arterial blood gas, but less likely to receive concomitant treatment with a cephalosporin (11% vs. 57%). With the exception of cephalosporin treatment, these differences were small, but due to the large sample were statistically significant. Comorbidities were similar in both groups. Patients in the quinolone group were also more likely to experience treatment failure (8.1% vs. 6.8%), death (1.5% vs. 1.0%), and antibiotic‐associated diarrhea (1.1% vs. 0.5%).

In the unadjusted analysis, compared to patients receiving quinolones, those treated with macrolides were less likely to experience treatment failure (OR, 0.83; 95% CI, 0.740.94) (Table 2). Adjusting for all patient, hospital, and physician covariates, including the propensity for treatment with macrolides, increased the OR to 0.89 and the results were no longer significant (95% CI, 0.781.01). Propensity matching successfully balanced all measured covariates except for the use of short‐acting bronchodilators and additional antibiotics (Table 1). In the propensity‐matched sample (Figure 1), quinolone‐treated patients were more likely to experience antibiotic‐associated diarrhea (1.2% vs. 0.6%; P = 0.0003) and late mechanical ventilation (1.3% vs. 0.8%; P = 0.02). There were no differences in adjusted cost or length of stay between the 2 groups. The results of the grouped treatment analysis, substituting the hospital's specific rate of macrolide use in place of the actual treatment that each patient received suggested that the 2 antibiotics were associated with similar rates of treatment failure. The OR for a 100% hospital rate of macrolide treatment vs. a 0% rate was 1.01 (95% CI, 0.751.35).

Figure 1

Discussion

In this large observational study conducted at 375 hospitals, we took advantage of a natural experiment in which antibiotic prescribing patterns varied widely across hospitals to compare the effectiveness of 2 common antibiotic regimens for AECOPD. Treatment with macrolides and quinolones were associated with a similar risk of treatment failure, costs, and length of stay; however, patients treated with macrolides were less likely to experience late mechanical ventilation or treatment for antibiotic‐associated diarrhea.

Despite broad consensus in COPD guidelines that patients with severe acute exacerbations should receive antibiotics, there is little agreement about the preferred empiric agent. Controversy exists regarding antibiotics' comparative effectiveness, and even over which pathogens cause COPD exacerbations. Given the frequency of hospitalization for AECOPD, understanding the comparative effectiveness of treatments in this setting could have important implications for health outcomes and costs. Unfortunately, most antibiotic studies in AECOPD were conducted >20 years ago, using antibiotics that rarely appeared in our sample.26 Consequently, clinical practice guidelines offer conflicting recommendations. For example, the National Institute for Clinical Excellence recommends empirical treatment with an aminopenicillin, a macrolide, or a tetracycline,8 while the American Thoracic Society recommends amoxicillin‐clavulonate or a fluoroquinolone.9

As might be expected in light of so much uncertainty, we found wide variation in prescribing patterns across hospitals. Overall, approximately one‐third of patients received a macrolide and two‐thirds a quinolone. Both regimens provide adequate coverage of H. influenza, S. pneumoniae, and M. catarrhalis, and conform to at least 1 COPD guideline. Nevertheless, patients receiving macrolides often received a cephalosporin as well; this pattern of treatment suggests that antibiotic selection is likely to have been influenced more by guidelines for the treatment of community‐acquired pneumonia than COPD.

Previous studies comparing antibiotic effectiveness suffer from shortcomings that limit their application to patients hospitalized with AECOPD. First, they enrolled patients with chronic bronchitis, and included patients without obstructive lung disease, and most studies included patients as young as 18 years old. Second, many either did not include treatment with steroids or excluded patients receiving more than 10 mg of prednisone daily. Third, almost all enrolled only ambulatory patients.

While there are no studies comparing quinolones and macrolides in patients hospitalized for AECOPD, a meta‐analysis comparing quinolones, macrolides, and amoxicillin‐clavulonate identified 19 trials of ambulatory patients with chronic bronchitis. That study found that all 3 drugs had similar efficacy initially, but that quinolones resulted in the fewest relapses over a 26‐week period.27 Macrolides and quinolones had similar rates of adverse effects. In contrast, we did not find a difference in treatment failure, cost, or length of stay, but did find a higher rate of diarrhea associated with quinolones. Others have also documented an association between fluoroquinolones and C. difficile diarrhea.2830 This trend, first noted in 2001, is of particular concern because the fluoroquinolone‐resistant strains appear to be hypervirulent and have been associated with nosocomial epidemics.3134

Our study has several limitations. First, its observational design leaves open the possibility of selection bias. For this reason we analyzed our data in several ways, including using a grouped treatment approach, an adaptation of the instrumental variable technique, and accepted only those differences which were consistent across all models. Second, our study used claims data, and therefore we could not directly adjust for physiological measures of severity. However, the highly detailed nature of the data allowed us to adjust for numerous tests and treatments that reflected the clinician's assessment of the patient's severity, as well as the number of prior COPD admissions. Third, we cannot exclude the possibility that some patients may have had concurrent pneumonia without an ICD‐9 code. We think that the number would be small because reimbursement for pneumonia is generally higher than for COPD, so hospitals have an incentive to code pneumonia as the principal diagnosis when present. Finally, we compared initial antibiotics only. More than one‐quarter of our patients received an additional antibiotic before discharge. In particular, patients receiving macrolides were often prescribed a concomitant cephalosporin. We do not know to what extent these additional antibiotics may have affected the outcomes.

Despite the large number of patients hospitalized annually for AECOPD, there are no randomized trials comparing different antibiotics in this population. Studies comparing antibiotics in chronic bronchitis can offer little guidance, since they have primarily focused on proving equivalence between existing antibiotics and newer, more expensive formulations.35 Because many of the patients enrolled in such trials do not benefit from antibiotics at all, either because they do not have COPD or because their exacerbation is not caused by bacteria, it is relatively easy to prove equivalence. Given that AECOPD is one of the leading causes of hospitalization in the United States, large, randomized trials comparing the effectiveness of different antibiotics should be a high priority. In the meantime, macrolides (often given together with cephalosporins) and quinolones appear to be equally effective initial antibiotic choices; considering antibiotic‐associated diarrhea, macrolides appear to be the safer of the 2.

Acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are responsible for more than 600,000 hospitalizations annually, resulting in direct costs of over $20 billion.1 Bacterial infections appear responsible for 50% of such exacerbations,25 and current COPD guidelines recommend treatment with antibiotics for patients with severe exacerbations or a change in sputum.1^,69 These recommendations are based on a number of small randomized trials, most of which were conducted more than 20 years ago using narrow spectrum antibiotics that are no longer commonly used.10 Only 4 studies, totaling 321 subjects, included hospitalized patients, and most studies excluded patients who required steroids. Because no clinical trials have compared different antibiotic regimens for AECOPD, existing guidelines offer a range of treatment options, including amoxicillin‐clavulonate, macrolides, quinolones, cephalosporins, aminopenicillins, and tetracyclines.

Among hospitalized patients, macrolides and quinolones appear to be the most frequently prescribed antibiotics.11 Both are available in oral formulations, have excellent bioavailability, and are administered once daily. In addition to their antimicrobial activity, macrolides are believed to have antiinflammatory effects, which could be especially advantageous in AECOPD.1214 In trials of chronic bronchitis, however, fluoroquinolones have been shown to reduce the risk of recurrent exacerbation when compared to macrolides.15 The wide variation that has been observed in antibiotic selection for patients hospitalized for AECOPD suggests a high degree of uncertainty among clinicians about the benefits of different treatment options.11 Given the limited evidence from randomized trials, we sought to evaluate the comparative effectiveness of macrolides and quinolones among a large, representative sample of patients hospitalized with AECOPD.

Subjects and Methods

Results

Of 26,248 AECOPD patients treated with antibiotics, 19,608 patients met the inclusion criteria; of these, 6139 (31%) were treated initially with a macrolide; the median age was 70 years; 60% were female; and 78% were white. A total of 86% of patients had a primary diagnosis of obstructive chronic bronchitis with acute exacerbation, and 6% had respiratory failure. The most common comorbidities were hypertension, diabetes, and congestive heart failure. Twenty‐two percent had been admitted at least once in the preceding 12 months. Treatment failure occurred in 7.7% of patients, and 1.3% died in the hospital. Mean length of stay was 4.8 days. Hospital prescribing rates for macrolides varied from 0% to 100%, with a mean of 33% and an interquartile range of 14% to 46% (Supporting Appendix Figure 1).

Compared to patients receiving macrolides, those receiving quinolones were older, more likely to have respiratory failure, to be cared for by a pulmonologist, and to have an admission in the previous year (Table 1). They were also more likely to be treated with bronchodilators, methylxanthines, steroids, diuretics, and noninvasive positive pressure ventilation, and to have an arterial blood gas, but less likely to receive concomitant treatment with a cephalosporin (11% vs. 57%). With the exception of cephalosporin treatment, these differences were small, but due to the large sample were statistically significant. Comorbidities were similar in both groups. Patients in the quinolone group were also more likely to experience treatment failure (8.1% vs. 6.8%), death (1.5% vs. 1.0%), and antibiotic‐associated diarrhea (1.1% vs. 0.5%).

In the unadjusted analysis, compared to patients receiving quinolones, those treated with macrolides were less likely to experience treatment failure (OR, 0.83; 95% CI, 0.740.94) (Table 2). Adjusting for all patient, hospital, and physician covariates, including the propensity for treatment with macrolides, increased the OR to 0.89 and the results were no longer significant (95% CI, 0.781.01). Propensity matching successfully balanced all measured covariates except for the use of short‐acting bronchodilators and additional antibiotics (Table 1). In the propensity‐matched sample (Figure 1), quinolone‐treated patients were more likely to experience antibiotic‐associated diarrhea (1.2% vs. 0.6%; P = 0.0003) and late mechanical ventilation (1.3% vs. 0.8%; P = 0.02). There were no differences in adjusted cost or length of stay between the 2 groups. The results of the grouped treatment analysis, substituting the hospital's specific rate of macrolide use in place of the actual treatment that each patient received suggested that the 2 antibiotics were associated with similar rates of treatment failure. The OR for a 100% hospital rate of macrolide treatment vs. a 0% rate was 1.01 (95% CI, 0.751.35).

Figure 1

Discussion

In this large observational study conducted at 375 hospitals, we took advantage of a natural experiment in which antibiotic prescribing patterns varied widely across hospitals to compare the effectiveness of 2 common antibiotic regimens for AECOPD. Treatment with macrolides and quinolones were associated with a similar risk of treatment failure, costs, and length of stay; however, patients treated with macrolides were less likely to experience late mechanical ventilation or treatment for antibiotic‐associated diarrhea.

Despite broad consensus in COPD guidelines that patients with severe acute exacerbations should receive antibiotics, there is little agreement about the preferred empiric agent. Controversy exists regarding antibiotics' comparative effectiveness, and even over which pathogens cause COPD exacerbations. Given the frequency of hospitalization for AECOPD, understanding the comparative effectiveness of treatments in this setting could have important implications for health outcomes and costs. Unfortunately, most antibiotic studies in AECOPD were conducted >20 years ago, using antibiotics that rarely appeared in our sample.26 Consequently, clinical practice guidelines offer conflicting recommendations. For example, the National Institute for Clinical Excellence recommends empirical treatment with an aminopenicillin, a macrolide, or a tetracycline,8 while the American Thoracic Society recommends amoxicillin‐clavulonate or a fluoroquinolone.9

As might be expected in light of so much uncertainty, we found wide variation in prescribing patterns across hospitals. Overall, approximately one‐third of patients received a macrolide and two‐thirds a quinolone. Both regimens provide adequate coverage of H. influenza, S. pneumoniae, and M. catarrhalis, and conform to at least 1 COPD guideline. Nevertheless, patients receiving macrolides often received a cephalosporin as well; this pattern of treatment suggests that antibiotic selection is likely to have been influenced more by guidelines for the treatment of community‐acquired pneumonia than COPD.

Previous studies comparing antibiotic effectiveness suffer from shortcomings that limit their application to patients hospitalized with AECOPD. First, they enrolled patients with chronic bronchitis, and included patients without obstructive lung disease, and most studies included patients as young as 18 years old. Second, many either did not include treatment with steroids or excluded patients receiving more than 10 mg of prednisone daily. Third, almost all enrolled only ambulatory patients.

While there are no studies comparing quinolones and macrolides in patients hospitalized for AECOPD, a meta‐analysis comparing quinolones, macrolides, and amoxicillin‐clavulonate identified 19 trials of ambulatory patients with chronic bronchitis. That study found that all 3 drugs had similar efficacy initially, but that quinolones resulted in the fewest relapses over a 26‐week period.27 Macrolides and quinolones had similar rates of adverse effects. In contrast, we did not find a difference in treatment failure, cost, or length of stay, but did find a higher rate of diarrhea associated with quinolones. Others have also documented an association between fluoroquinolones and C. difficile diarrhea.2830 This trend, first noted in 2001, is of particular concern because the fluoroquinolone‐resistant strains appear to be hypervirulent and have been associated with nosocomial epidemics.3134

Our study has several limitations. First, its observational design leaves open the possibility of selection bias. For this reason we analyzed our data in several ways, including using a grouped treatment approach, an adaptation of the instrumental variable technique, and accepted only those differences which were consistent across all models. Second, our study used claims data, and therefore we could not directly adjust for physiological measures of severity. However, the highly detailed nature of the data allowed us to adjust for numerous tests and treatments that reflected the clinician's assessment of the patient's severity, as well as the number of prior COPD admissions. Third, we cannot exclude the possibility that some patients may have had concurrent pneumonia without an ICD‐9 code. We think that the number would be small because reimbursement for pneumonia is generally higher than for COPD, so hospitals have an incentive to code pneumonia as the principal diagnosis when present. Finally, we compared initial antibiotics only. More than one‐quarter of our patients received an additional antibiotic before discharge. In particular, patients receiving macrolides were often prescribed a concomitant cephalosporin. We do not know to what extent these additional antibiotics may have affected the outcomes.

Despite the large number of patients hospitalized annually for AECOPD, there are no randomized trials comparing different antibiotics in this population. Studies comparing antibiotics in chronic bronchitis can offer little guidance, since they have primarily focused on proving equivalence between existing antibiotics and newer, more expensive formulations.35 Because many of the patients enrolled in such trials do not benefit from antibiotics at all, either because they do not have COPD or because their exacerbation is not caused by bacteria, it is relatively easy to prove equivalence. Given that AECOPD is one of the leading causes of hospitalization in the United States, large, randomized trials comparing the effectiveness of different antibiotics should be a high priority. In the meantime, macrolides (often given together with cephalosporins) and quinolones appear to be equally effective initial antibiotic choices; considering antibiotic‐associated diarrhea, macrolides appear to be the safer of the 2.

A previously healthy 25‐year‐old Guatemalan man presented to the emergency department with 1 day of fever, nausea, vomiting, and right lower quadrant abdominal pain. A computed tomography (CT) scan revealed acute appendicitis. The patient underwent an uncomplicated laparoscopic appendectomy and was discharged in stable condition after 48 hours.

Five days after the operation he returned to the emergency department with abdominal pain, nausea, vomiting, and lightheadedness. He was tachycardic, and his hemoglobin was 9.5 g/dL (normal, 13.3‐17.7 g/dL), decreased from 14.4 g/dL prior to his appendectomy. A CT scan showed intraperitoneal blood with active extravasation of contrast at the site of the appendectomy.

Additional laboratory testing revealed an activated partial thromboplastin time (aPTT) of 52 seconds (normal, <37 seconds) and protime (also prothrombin time [PT]) of 14 seconds (normal, <14.1 seconds). The platelet count was 449,000/L (normal, 150‐400,000/L) and the fibrinogen level was 337 mg/dL (normal, 170‐440 mg/dL). Crystalloid and packed red blood cells were administered. Since further laboratory evaluation of the prolonged aPTT was not immediately available, the patient was empirically treated with fresh frozen plasma (FFP), cryoprecipitate, and Factor VIII/von Willebrand factor concentrate. At laparotomy, bleeding was observed at the previous operative site, and 2 L of intraperitoneal blood was evacuated.

The next morning, Factor VIII and Factor IX (FIX) activities and the ristocetin cofactor study performed on specimens obtained immediately prior to the second operation were normal, but the FIX activity was 5% of normal. The diagnosis of FXI deficiency was made and 2 to 3 units of FFP (the amount necessary to maintain the patient's measured FXI activity near 20% of normal) were transfused daily. Nine days of FFP infusions were required to achieve complete wound hemostasis. The patient had no further bleeding episodes after discharge.

Upon further interviewing, the patient revealed that 2 months prior he sustained a small laceration on his arm that bled for a long time and that his brother had experienced prolonged bleeding after a dental extraction.

Commentary

Routine performance of preprocedural laboratory testing, and complete reliance on the results as a means of excluding a propensity to bleeding, may not only lead to excessive testing and delayed procedures, but also provides false reassurance because normal routine laboratory studies cannot be used to exclude some bleeding disorders (Table 1).

Most studies evaluating routine laboratory testing of hemostatic variables prior to invasive procedures come from patients undergoing elective general surgery. A 1988 study concluded that there is no benefit in the routine preoperative use of the PT, aPTT, platelet count, and bleeding time in the absence of clinical evidence of a hemostatic defect, as assessed by a patient questionnaire and a thorough physical examination.¹ A subsequent European, prospective, multicenter study confirmed that abnormalities of preoperative laboratory screening in the absence of a history of bleeding or clinical abnormality were not associated with worse surgical morbidity or mortality, compared to patients with normal screening laboratory studies.² A recent systematic review has also confirmed the poor positive predictive value of screening tests when used in isolation, and recommended a history‐based and physical exam‐based approach.³ Questionnaires have been shown to be particularly important tools for eliciting clinically significant bleeding disorders that may require revision of the surgical plan.^1,4

FXI is a serine protease whose activity is crucial for robust fibrin clot formation and inhibition of fibrinolysis at sites of vascular injury.⁵ FXI deficiency is an autosomal recessive disorder with an incidence of 1 per 1,000,000 in the general population, with a significantly higher incidence in the Ashkenazi Jewish population. While the risk of spontaneous hemorrhage is typically low, life‐threatening bleeding may occur after surgery or trauma. The severity of the measured FXI level deficiency does not always correlate with risk of bleeding. Periprocedural prophylaxis and treatment of bleeding aim to replace FXI to the low‐normal range by administering FXI concentrate, (not available in the United States) or FFP. Antifibrinolytic agents such as tranexamic acid or ϵ‐aminocaproic acid may be used adjunctively in cases of mucosal bleeding.⁵

In this case, preoperative screening, either using a questionnaire or careful history‐taking, would have identified the patient's personal and family history of bleeding and prompted appropriate preoperative coagulation testing, which could have exposed the hemostatic defect, allowing for modification of the perioperative medical plan.

In summary, preoperative bleeding evaluations should be performed routinely and should begin with a careful history (use of a questionnaire may be considered) and physical examination. Excessive bleeding after prior surgery, trauma, dental extractions, parturition, or circumcision; bleeding tendency in family members; current use of medications that may increase bleeding risk (such as anticoagulants or aspirin); and physical signs associated with bleeding should be assessed. If clinical details fail to expose a potential bleeding disorder, it is safe and cost‐effective¹ to proceed with surgery without performing additional laboratory testing. In contrast, any abnormality on the clinical assessment should trigger preoperative laboratory analysis of basic hemostatic parameters, which may prompt further testing or hematology consultation.

A previously healthy 25‐year‐old Guatemalan man presented to the emergency department with 1 day of fever, nausea, vomiting, and right lower quadrant abdominal pain. A computed tomography (CT) scan revealed acute appendicitis. The patient underwent an uncomplicated laparoscopic appendectomy and was discharged in stable condition after 48 hours.

Five days after the operation he returned to the emergency department with abdominal pain, nausea, vomiting, and lightheadedness. He was tachycardic, and his hemoglobin was 9.5 g/dL (normal, 13.3‐17.7 g/dL), decreased from 14.4 g/dL prior to his appendectomy. A CT scan showed intraperitoneal blood with active extravasation of contrast at the site of the appendectomy.

Additional laboratory testing revealed an activated partial thromboplastin time (aPTT) of 52 seconds (normal, <37 seconds) and protime (also prothrombin time [PT]) of 14 seconds (normal, <14.1 seconds). The platelet count was 449,000/L (normal, 150‐400,000/L) and the fibrinogen level was 337 mg/dL (normal, 170‐440 mg/dL). Crystalloid and packed red blood cells were administered. Since further laboratory evaluation of the prolonged aPTT was not immediately available, the patient was empirically treated with fresh frozen plasma (FFP), cryoprecipitate, and Factor VIII/von Willebrand factor concentrate. At laparotomy, bleeding was observed at the previous operative site, and 2 L of intraperitoneal blood was evacuated.

The next morning, Factor VIII and Factor IX (FIX) activities and the ristocetin cofactor study performed on specimens obtained immediately prior to the second operation were normal, but the FIX activity was 5% of normal. The diagnosis of FXI deficiency was made and 2 to 3 units of FFP (the amount necessary to maintain the patient's measured FXI activity near 20% of normal) were transfused daily. Nine days of FFP infusions were required to achieve complete wound hemostasis. The patient had no further bleeding episodes after discharge.

Upon further interviewing, the patient revealed that 2 months prior he sustained a small laceration on his arm that bled for a long time and that his brother had experienced prolonged bleeding after a dental extraction.

Commentary

Routine performance of preprocedural laboratory testing, and complete reliance on the results as a means of excluding a propensity to bleeding, may not only lead to excessive testing and delayed procedures, but also provides false reassurance because normal routine laboratory studies cannot be used to exclude some bleeding disorders (Table 1).

Most studies evaluating routine laboratory testing of hemostatic variables prior to invasive procedures come from patients undergoing elective general surgery. A 1988 study concluded that there is no benefit in the routine preoperative use of the PT, aPTT, platelet count, and bleeding time in the absence of clinical evidence of a hemostatic defect, as assessed by a patient questionnaire and a thorough physical examination.¹ A subsequent European, prospective, multicenter study confirmed that abnormalities of preoperative laboratory screening in the absence of a history of bleeding or clinical abnormality were not associated with worse surgical morbidity or mortality, compared to patients with normal screening laboratory studies.² A recent systematic review has also confirmed the poor positive predictive value of screening tests when used in isolation, and recommended a history‐based and physical exam‐based approach.³ Questionnaires have been shown to be particularly important tools for eliciting clinically significant bleeding disorders that may require revision of the surgical plan.^1,4

FXI is a serine protease whose activity is crucial for robust fibrin clot formation and inhibition of fibrinolysis at sites of vascular injury.⁵ FXI deficiency is an autosomal recessive disorder with an incidence of 1 per 1,000,000 in the general population, with a significantly higher incidence in the Ashkenazi Jewish population. While the risk of spontaneous hemorrhage is typically low, life‐threatening bleeding may occur after surgery or trauma. The severity of the measured FXI level deficiency does not always correlate with risk of bleeding. Periprocedural prophylaxis and treatment of bleeding aim to replace FXI to the low‐normal range by administering FXI concentrate, (not available in the United States) or FFP. Antifibrinolytic agents such as tranexamic acid or ϵ‐aminocaproic acid may be used adjunctively in cases of mucosal bleeding.⁵

In this case, preoperative screening, either using a questionnaire or careful history‐taking, would have identified the patient's personal and family history of bleeding and prompted appropriate preoperative coagulation testing, which could have exposed the hemostatic defect, allowing for modification of the perioperative medical plan.

In summary, preoperative bleeding evaluations should be performed routinely and should begin with a careful history (use of a questionnaire may be considered) and physical examination. Excessive bleeding after prior surgery, trauma, dental extractions, parturition, or circumcision; bleeding tendency in family members; current use of medications that may increase bleeding risk (such as anticoagulants or aspirin); and physical signs associated with bleeding should be assessed. If clinical details fail to expose a potential bleeding disorder, it is safe and cost‐effective¹ to proceed with surgery without performing additional laboratory testing. In contrast, any abnormality on the clinical assessment should trigger preoperative laboratory analysis of basic hemostatic parameters, which may prompt further testing or hematology consultation.

A previously healthy 25‐year‐old Guatemalan man presented to the emergency department with 1 day of fever, nausea, vomiting, and right lower quadrant abdominal pain. A computed tomography (CT) scan revealed acute appendicitis. The patient underwent an uncomplicated laparoscopic appendectomy and was discharged in stable condition after 48 hours.

Five days after the operation he returned to the emergency department with abdominal pain, nausea, vomiting, and lightheadedness. He was tachycardic, and his hemoglobin was 9.5 g/dL (normal, 13.3‐17.7 g/dL), decreased from 14.4 g/dL prior to his appendectomy. A CT scan showed intraperitoneal blood with active extravasation of contrast at the site of the appendectomy.

Additional laboratory testing revealed an activated partial thromboplastin time (aPTT) of 52 seconds (normal, <37 seconds) and protime (also prothrombin time [PT]) of 14 seconds (normal, <14.1 seconds). The platelet count was 449,000/L (normal, 150‐400,000/L) and the fibrinogen level was 337 mg/dL (normal, 170‐440 mg/dL). Crystalloid and packed red blood cells were administered. Since further laboratory evaluation of the prolonged aPTT was not immediately available, the patient was empirically treated with fresh frozen plasma (FFP), cryoprecipitate, and Factor VIII/von Willebrand factor concentrate. At laparotomy, bleeding was observed at the previous operative site, and 2 L of intraperitoneal blood was evacuated.

The next morning, Factor VIII and Factor IX (FIX) activities and the ristocetin cofactor study performed on specimens obtained immediately prior to the second operation were normal, but the FIX activity was 5% of normal. The diagnosis of FXI deficiency was made and 2 to 3 units of FFP (the amount necessary to maintain the patient's measured FXI activity near 20% of normal) were transfused daily. Nine days of FFP infusions were required to achieve complete wound hemostasis. The patient had no further bleeding episodes after discharge.

Upon further interviewing, the patient revealed that 2 months prior he sustained a small laceration on his arm that bled for a long time and that his brother had experienced prolonged bleeding after a dental extraction.

Commentary

Routine performance of preprocedural laboratory testing, and complete reliance on the results as a means of excluding a propensity to bleeding, may not only lead to excessive testing and delayed procedures, but also provides false reassurance because normal routine laboratory studies cannot be used to exclude some bleeding disorders (Table 1).

Most studies evaluating routine laboratory testing of hemostatic variables prior to invasive procedures come from patients undergoing elective general surgery. A 1988 study concluded that there is no benefit in the routine preoperative use of the PT, aPTT, platelet count, and bleeding time in the absence of clinical evidence of a hemostatic defect, as assessed by a patient questionnaire and a thorough physical examination.¹ A subsequent European, prospective, multicenter study confirmed that abnormalities of preoperative laboratory screening in the absence of a history of bleeding or clinical abnormality were not associated with worse surgical morbidity or mortality, compared to patients with normal screening laboratory studies.² A recent systematic review has also confirmed the poor positive predictive value of screening tests when used in isolation, and recommended a history‐based and physical exam‐based approach.³ Questionnaires have been shown to be particularly important tools for eliciting clinically significant bleeding disorders that may require revision of the surgical plan.^1,4

FXI is a serine protease whose activity is crucial for robust fibrin clot formation and inhibition of fibrinolysis at sites of vascular injury.⁵ FXI deficiency is an autosomal recessive disorder with an incidence of 1 per 1,000,000 in the general population, with a significantly higher incidence in the Ashkenazi Jewish population. While the risk of spontaneous hemorrhage is typically low, life‐threatening bleeding may occur after surgery or trauma. The severity of the measured FXI level deficiency does not always correlate with risk of bleeding. Periprocedural prophylaxis and treatment of bleeding aim to replace FXI to the low‐normal range by administering FXI concentrate, (not available in the United States) or FFP. Antifibrinolytic agents such as tranexamic acid or ϵ‐aminocaproic acid may be used adjunctively in cases of mucosal bleeding.⁵

In this case, preoperative screening, either using a questionnaire or careful history‐taking, would have identified the patient's personal and family history of bleeding and prompted appropriate preoperative coagulation testing, which could have exposed the hemostatic defect, allowing for modification of the perioperative medical plan.

In summary, preoperative bleeding evaluations should be performed routinely and should begin with a careful history (use of a questionnaire may be considered) and physical examination. Excessive bleeding after prior surgery, trauma, dental extractions, parturition, or circumcision; bleeding tendency in family members; current use of medications that may increase bleeding risk (such as anticoagulants or aspirin); and physical signs associated with bleeding should be assessed. If clinical details fail to expose a potential bleeding disorder, it is safe and cost‐effective¹ to proceed with surgery without performing additional laboratory testing. In contrast, any abnormality on the clinical assessment should trigger preoperative laboratory analysis of basic hemostatic parameters, which may prompt further testing or hematology consultation.

A 56‐year‐old male presented to the emergency department with a 2‐week history of increasing abdominal girth, nausea, vomiting, and lower extremity edema. His girlfriend had also noted a yellow tinge to his skin and eyes. His past medical history was significant for bipolar disorder, alcohol‐related seizures, and pneumonia. He had no allergies and denied medications prior to admission. Family history was negative for liver disease and social history was notable for ongoing tobacco use and alcohol dependence. He was afebrile with stable vital signs. Physical examination demonstrated an alert gentleman whose answers to questions required occasional factual correction by his partner. His abdomen was distended and nontender with prominent vasculature and shifting dullness. Lower extremity edema was symmetric and bilateral, rated as 2+. Scattered spider angiomata and a fine bilateral hand tremor without asterixis were also noted. Initial laboratory data demonstrated a white blood cell count of 13,900/L, hematocrit 37%, and platelet count 176,000/L. His sodium was 130 mg/dL, blood urea nitrogen (BUN) 1 mg/dL, and creatinine 0.7 mg/dL. International normalized ratio (INR) was 1.8, aspartate aminotransferase (AST) was 117 U/L, alanine aminotransferase (ALT) 33 U/L, alkaline phosphatase 191 U/L, total bilirubin 9.2 mg/dL, total protein 7.0 g/dL, and albumin 1.9 g/dL. Abdominal ultrasound revealed a diffusely hyperechoic liver with a large amount of ascites.

The patient was admitted with the diagnoses of presumed alcoholic hepatitis and end‐stage liver disease. Model for End‐Stage Liver Disease (MELD) score was 21 and discriminant function 16.8. Paracentesis demonstrated a serum‐ascites albumin gradient of >1.1 and no evidence of spontaneous bacterial peritonitis. Diuresis was initiated. He was placed on unfractionated heparin at a dose of 5000 units every 8 hours for deep venous thrombosis (DVT) prophylaxis. By hospital day 3, the patient's laboratory values had improved, yet his stay was prolonged by alcohol withdrawal requiring benzodiazepines, altered mental status presumed secondary to hepatic encephalopathy, acute renal failure, aspiration pneumonia, and persistent unexplained leukocytosis. He required medical restraints during this time given confusion and propensity to ambulate without assistance, yet sustained no falls or other known trauma in care delivered during this time.

Between days 14 and 17, the patient's hematocrit fell from 36% to 30%; vital signs remained stable. He underwent an uncomplicated, ultrasound‐guided therapeutic paracentesis, which yielded 1.4 L of straw‐colored fluid on the afternoon of day 17; the procedure was attempted only on the right side. On the morning of day 18, the patient's blood pressure dropped to 78/55 mmHg with a pulse of 123 beats per minute; he became pale and unresponsive. Physical examination was notable for somnolence and a tender, warm left flank mass, contralateral to his paracentesis site. No flank or periumbilical ecchymoses were identified. Complete blood count demonstrated a white blood count (WBC) of 22,970/L, hematocrit 16%, and platelet count 104,000/L. INR was 2.0, unchanged from the last check on day 10. Partial thromboplastin time was 41 seconds and fibrinogen was 293 mg/dL (normal 150‐400 mg/dL). Peripheral blood smear was negative for red cell fragments. Blood chemistries revealed a sodium of 134 mg/dL, bicarbonate 20 mEq/L, anion gap 7, BUN 24 mg/dL, and creatinine 1.6 mg/dL (up from 1.0 mg/dL the previous day). His venous lactate level was 4.6 mmol/L and arterial blood gas sampling on room air demonstrated a pH of 7.35, partial pressure of carbon dioxide (pCO₂) 29 mmHg, partial pressure of oxygen (pCO₂) 54 mmHg, and bicarbonate 15 mEq/L. A femoral introducer was placed for volume resuscitation and the patient was urgently transfused with packed red blood cells (PRBCs) and fresh‐frozen plasma (FFP) to correct his coagulopathy. Computed tomography of the abdomen revealed a large left retroperitoneal hematoma measuring 15 15 22 cm³ (Figure 1). Despite transfusion, his hematocrit continued to fall. Urgent angiography was performed, upon which he was found to have active bleeding from the left L3‐L5 lumbar arteries. These were successfully embolized. He required PRBCs and FFP transfusion only once following this procedure. Given a transient decrease in his urine output, his bladder pressures were followed closely for evidence of abdominal compartment syndrome, which did not develop. He was transferred from the intensive care unit (ICU) to the floor on day 20, where his physical exam and hematocrit remained stable and his delirium slowly cleared. He was ultimately discharged to a skilled nursing facility on day 33.

Figure 1

Discussion

Spontaneous retroperitoneal hematoma is a well‐recognized entity that may present with the classic triad of abdominal or groin pain, palpable abdominal or flank mass, and shock or lower extremity motor or sensory changes due to femoral nerve compression.1 Although classically described as skin findings associated with retroperitoneal hemorrhage, Cullen and Grey‐Turner signs are relatively late findings that may not develop in all patients.

Retroperitoneal hemorrhage is well‐recognized as a result of iatrogenic anticoagulation,1 but has been reported less commonly as a result of coagulopathy related to cirrhosis. Di Bisceglie and Richart2 describe 2 patients with MELD scores of 29 and 25, respectively, who developed spontaneous retroperitoneal and rectus muscle hemorrhage. Both had evidence of associated disseminated intravascular coagulation (DIC) and died. Even less common is spontaneous lumbar artery rupture, occurring rarely in the absence of trauma or instrumentation. One reported bleed developed in the context of systemic anticoagulation for a mechanical valve.3 Halak et al.4 relate a case in which the only known risk factor was chronic renal disease. Hama et al.5 describe a patient with a history of alcoholic liver cirrhosis and INR of 2.3 whose lumbar artery rupture was successfully managed with transcatheter arterial embolization.

It is difficult to ascertain the effect of prophylactic anticoagulation in development of this particular hemorrhage. Retroperitoneal bleeding is a very rare complication of pharmacologic prophylaxis for DVT reported with both low‐molecular weight and unfractionated heparins.6 There may be additive risk of prophylaxis in a cirrhotic patient with baseline elevated INR and thrombocytopenia, particularly in the context of renal failure.

Options in the management of spontaneous retroperitoneal hematoma include transarterial embolization, percutaneous decompression, and open surgery. Nonoperative management of these bleeds when possible may be preferable in cirrhotic patients, as their baseline liver disease renders them higher‐risk candidates for surgery. There are no randomized trials comparing these approaches.1

In summary, we report the case of a 56‐year‐old man with end‐stage liver disease and associated coagulopathy without evidence of DIC who survived to discharge with intravascular management of a spontaneous retroperitoneal hemorrhage from the lumbar arteries. To our knowledge, this is the second reported case of spontaneous retroperitoneal hemorrhage in a cirrhotic in which the lumbar arteries were implicated and the first in which multiple arteries were found to be bleeding simultaneously. Any hospitalized patient who develops abdominal pain, flank pain, or hemodynamic instability in the context of coagulopathy, regardless of cause, warrants evaluation for retroperitoneal bleed. Appropriate early management includes immediate resuscitation, intensive monitoring, urgent imaging, and surgical and interventional radiology consultation in order to prevent a fatal outcome.

A 56‐year‐old male presented to the emergency department with a 2‐week history of increasing abdominal girth, nausea, vomiting, and lower extremity edema. His girlfriend had also noted a yellow tinge to his skin and eyes. His past medical history was significant for bipolar disorder, alcohol‐related seizures, and pneumonia. He had no allergies and denied medications prior to admission. Family history was negative for liver disease and social history was notable for ongoing tobacco use and alcohol dependence. He was afebrile with stable vital signs. Physical examination demonstrated an alert gentleman whose answers to questions required occasional factual correction by his partner. His abdomen was distended and nontender with prominent vasculature and shifting dullness. Lower extremity edema was symmetric and bilateral, rated as 2+. Scattered spider angiomata and a fine bilateral hand tremor without asterixis were also noted. Initial laboratory data demonstrated a white blood cell count of 13,900/L, hematocrit 37%, and platelet count 176,000/L. His sodium was 130 mg/dL, blood urea nitrogen (BUN) 1 mg/dL, and creatinine 0.7 mg/dL. International normalized ratio (INR) was 1.8, aspartate aminotransferase (AST) was 117 U/L, alanine aminotransferase (ALT) 33 U/L, alkaline phosphatase 191 U/L, total bilirubin 9.2 mg/dL, total protein 7.0 g/dL, and albumin 1.9 g/dL. Abdominal ultrasound revealed a diffusely hyperechoic liver with a large amount of ascites.

The patient was admitted with the diagnoses of presumed alcoholic hepatitis and end‐stage liver disease. Model for End‐Stage Liver Disease (MELD) score was 21 and discriminant function 16.8. Paracentesis demonstrated a serum‐ascites albumin gradient of >1.1 and no evidence of spontaneous bacterial peritonitis. Diuresis was initiated. He was placed on unfractionated heparin at a dose of 5000 units every 8 hours for deep venous thrombosis (DVT) prophylaxis. By hospital day 3, the patient's laboratory values had improved, yet his stay was prolonged by alcohol withdrawal requiring benzodiazepines, altered mental status presumed secondary to hepatic encephalopathy, acute renal failure, aspiration pneumonia, and persistent unexplained leukocytosis. He required medical restraints during this time given confusion and propensity to ambulate without assistance, yet sustained no falls or other known trauma in care delivered during this time.

Between days 14 and 17, the patient's hematocrit fell from 36% to 30%; vital signs remained stable. He underwent an uncomplicated, ultrasound‐guided therapeutic paracentesis, which yielded 1.4 L of straw‐colored fluid on the afternoon of day 17; the procedure was attempted only on the right side. On the morning of day 18, the patient's blood pressure dropped to 78/55 mmHg with a pulse of 123 beats per minute; he became pale and unresponsive. Physical examination was notable for somnolence and a tender, warm left flank mass, contralateral to his paracentesis site. No flank or periumbilical ecchymoses were identified. Complete blood count demonstrated a white blood count (WBC) of 22,970/L, hematocrit 16%, and platelet count 104,000/L. INR was 2.0, unchanged from the last check on day 10. Partial thromboplastin time was 41 seconds and fibrinogen was 293 mg/dL (normal 150‐400 mg/dL). Peripheral blood smear was negative for red cell fragments. Blood chemistries revealed a sodium of 134 mg/dL, bicarbonate 20 mEq/L, anion gap 7, BUN 24 mg/dL, and creatinine 1.6 mg/dL (up from 1.0 mg/dL the previous day). His venous lactate level was 4.6 mmol/L and arterial blood gas sampling on room air demonstrated a pH of 7.35, partial pressure of carbon dioxide (pCO₂) 29 mmHg, partial pressure of oxygen (pCO₂) 54 mmHg, and bicarbonate 15 mEq/L. A femoral introducer was placed for volume resuscitation and the patient was urgently transfused with packed red blood cells (PRBCs) and fresh‐frozen plasma (FFP) to correct his coagulopathy. Computed tomography of the abdomen revealed a large left retroperitoneal hematoma measuring 15 15 22 cm³ (Figure 1). Despite transfusion, his hematocrit continued to fall. Urgent angiography was performed, upon which he was found to have active bleeding from the left L3‐L5 lumbar arteries. These were successfully embolized. He required PRBCs and FFP transfusion only once following this procedure. Given a transient decrease in his urine output, his bladder pressures were followed closely for evidence of abdominal compartment syndrome, which did not develop. He was transferred from the intensive care unit (ICU) to the floor on day 20, where his physical exam and hematocrit remained stable and his delirium slowly cleared. He was ultimately discharged to a skilled nursing facility on day 33.

Figure 1

Discussion

Spontaneous retroperitoneal hematoma is a well‐recognized entity that may present with the classic triad of abdominal or groin pain, palpable abdominal or flank mass, and shock or lower extremity motor or sensory changes due to femoral nerve compression.1 Although classically described as skin findings associated with retroperitoneal hemorrhage, Cullen and Grey‐Turner signs are relatively late findings that may not develop in all patients.

Retroperitoneal hemorrhage is well‐recognized as a result of iatrogenic anticoagulation,1 but has been reported less commonly as a result of coagulopathy related to cirrhosis. Di Bisceglie and Richart2 describe 2 patients with MELD scores of 29 and 25, respectively, who developed spontaneous retroperitoneal and rectus muscle hemorrhage. Both had evidence of associated disseminated intravascular coagulation (DIC) and died. Even less common is spontaneous lumbar artery rupture, occurring rarely in the absence of trauma or instrumentation. One reported bleed developed in the context of systemic anticoagulation for a mechanical valve.3 Halak et al.4 relate a case in which the only known risk factor was chronic renal disease. Hama et al.5 describe a patient with a history of alcoholic liver cirrhosis and INR of 2.3 whose lumbar artery rupture was successfully managed with transcatheter arterial embolization.

It is difficult to ascertain the effect of prophylactic anticoagulation in development of this particular hemorrhage. Retroperitoneal bleeding is a very rare complication of pharmacologic prophylaxis for DVT reported with both low‐molecular weight and unfractionated heparins.6 There may be additive risk of prophylaxis in a cirrhotic patient with baseline elevated INR and thrombocytopenia, particularly in the context of renal failure.

Options in the management of spontaneous retroperitoneal hematoma include transarterial embolization, percutaneous decompression, and open surgery. Nonoperative management of these bleeds when possible may be preferable in cirrhotic patients, as their baseline liver disease renders them higher‐risk candidates for surgery. There are no randomized trials comparing these approaches.1

In summary, we report the case of a 56‐year‐old man with end‐stage liver disease and associated coagulopathy without evidence of DIC who survived to discharge with intravascular management of a spontaneous retroperitoneal hemorrhage from the lumbar arteries. To our knowledge, this is the second reported case of spontaneous retroperitoneal hemorrhage in a cirrhotic in which the lumbar arteries were implicated and the first in which multiple arteries were found to be bleeding simultaneously. Any hospitalized patient who develops abdominal pain, flank pain, or hemodynamic instability in the context of coagulopathy, regardless of cause, warrants evaluation for retroperitoneal bleed. Appropriate early management includes immediate resuscitation, intensive monitoring, urgent imaging, and surgical and interventional radiology consultation in order to prevent a fatal outcome.

A 56‐year‐old male presented to the emergency department with a 2‐week history of increasing abdominal girth, nausea, vomiting, and lower extremity edema. His girlfriend had also noted a yellow tinge to his skin and eyes. His past medical history was significant for bipolar disorder, alcohol‐related seizures, and pneumonia. He had no allergies and denied medications prior to admission. Family history was negative for liver disease and social history was notable for ongoing tobacco use and alcohol dependence. He was afebrile with stable vital signs. Physical examination demonstrated an alert gentleman whose answers to questions required occasional factual correction by his partner. His abdomen was distended and nontender with prominent vasculature and shifting dullness. Lower extremity edema was symmetric and bilateral, rated as 2+. Scattered spider angiomata and a fine bilateral hand tremor without asterixis were also noted. Initial laboratory data demonstrated a white blood cell count of 13,900/L, hematocrit 37%, and platelet count 176,000/L. His sodium was 130 mg/dL, blood urea nitrogen (BUN) 1 mg/dL, and creatinine 0.7 mg/dL. International normalized ratio (INR) was 1.8, aspartate aminotransferase (AST) was 117 U/L, alanine aminotransferase (ALT) 33 U/L, alkaline phosphatase 191 U/L, total bilirubin 9.2 mg/dL, total protein 7.0 g/dL, and albumin 1.9 g/dL. Abdominal ultrasound revealed a diffusely hyperechoic liver with a large amount of ascites.

The patient was admitted with the diagnoses of presumed alcoholic hepatitis and end‐stage liver disease. Model for End‐Stage Liver Disease (MELD) score was 21 and discriminant function 16.8. Paracentesis demonstrated a serum‐ascites albumin gradient of >1.1 and no evidence of spontaneous bacterial peritonitis. Diuresis was initiated. He was placed on unfractionated heparin at a dose of 5000 units every 8 hours for deep venous thrombosis (DVT) prophylaxis. By hospital day 3, the patient's laboratory values had improved, yet his stay was prolonged by alcohol withdrawal requiring benzodiazepines, altered mental status presumed secondary to hepatic encephalopathy, acute renal failure, aspiration pneumonia, and persistent unexplained leukocytosis. He required medical restraints during this time given confusion and propensity to ambulate without assistance, yet sustained no falls or other known trauma in care delivered during this time.

Between days 14 and 17, the patient's hematocrit fell from 36% to 30%; vital signs remained stable. He underwent an uncomplicated, ultrasound‐guided therapeutic paracentesis, which yielded 1.4 L of straw‐colored fluid on the afternoon of day 17; the procedure was attempted only on the right side. On the morning of day 18, the patient's blood pressure dropped to 78/55 mmHg with a pulse of 123 beats per minute; he became pale and unresponsive. Physical examination was notable for somnolence and a tender, warm left flank mass, contralateral to his paracentesis site. No flank or periumbilical ecchymoses were identified. Complete blood count demonstrated a white blood count (WBC) of 22,970/L, hematocrit 16%, and platelet count 104,000/L. INR was 2.0, unchanged from the last check on day 10. Partial thromboplastin time was 41 seconds and fibrinogen was 293 mg/dL (normal 150‐400 mg/dL). Peripheral blood smear was negative for red cell fragments. Blood chemistries revealed a sodium of 134 mg/dL, bicarbonate 20 mEq/L, anion gap 7, BUN 24 mg/dL, and creatinine 1.6 mg/dL (up from 1.0 mg/dL the previous day). His venous lactate level was 4.6 mmol/L and arterial blood gas sampling on room air demonstrated a pH of 7.35, partial pressure of carbon dioxide (pCO₂) 29 mmHg, partial pressure of oxygen (pCO₂) 54 mmHg, and bicarbonate 15 mEq/L. A femoral introducer was placed for volume resuscitation and the patient was urgently transfused with packed red blood cells (PRBCs) and fresh‐frozen plasma (FFP) to correct his coagulopathy. Computed tomography of the abdomen revealed a large left retroperitoneal hematoma measuring 15 15 22 cm³ (Figure 1). Despite transfusion, his hematocrit continued to fall. Urgent angiography was performed, upon which he was found to have active bleeding from the left L3‐L5 lumbar arteries. These were successfully embolized. He required PRBCs and FFP transfusion only once following this procedure. Given a transient decrease in his urine output, his bladder pressures were followed closely for evidence of abdominal compartment syndrome, which did not develop. He was transferred from the intensive care unit (ICU) to the floor on day 20, where his physical exam and hematocrit remained stable and his delirium slowly cleared. He was ultimately discharged to a skilled nursing facility on day 33.

Figure 1

Discussion

Spontaneous retroperitoneal hematoma is a well‐recognized entity that may present with the classic triad of abdominal or groin pain, palpable abdominal or flank mass, and shock or lower extremity motor or sensory changes due to femoral nerve compression.1 Although classically described as skin findings associated with retroperitoneal hemorrhage, Cullen and Grey‐Turner signs are relatively late findings that may not develop in all patients.

Retroperitoneal hemorrhage is well‐recognized as a result of iatrogenic anticoagulation,1 but has been reported less commonly as a result of coagulopathy related to cirrhosis. Di Bisceglie and Richart2 describe 2 patients with MELD scores of 29 and 25, respectively, who developed spontaneous retroperitoneal and rectus muscle hemorrhage. Both had evidence of associated disseminated intravascular coagulation (DIC) and died. Even less common is spontaneous lumbar artery rupture, occurring rarely in the absence of trauma or instrumentation. One reported bleed developed in the context of systemic anticoagulation for a mechanical valve.3 Halak et al.4 relate a case in which the only known risk factor was chronic renal disease. Hama et al.5 describe a patient with a history of alcoholic liver cirrhosis and INR of 2.3 whose lumbar artery rupture was successfully managed with transcatheter arterial embolization.

It is difficult to ascertain the effect of prophylactic anticoagulation in development of this particular hemorrhage. Retroperitoneal bleeding is a very rare complication of pharmacologic prophylaxis for DVT reported with both low‐molecular weight and unfractionated heparins.6 There may be additive risk of prophylaxis in a cirrhotic patient with baseline elevated INR and thrombocytopenia, particularly in the context of renal failure.

Options in the management of spontaneous retroperitoneal hematoma include transarterial embolization, percutaneous decompression, and open surgery. Nonoperative management of these bleeds when possible may be preferable in cirrhotic patients, as their baseline liver disease renders them higher‐risk candidates for surgery. There are no randomized trials comparing these approaches.1

In summary, we report the case of a 56‐year‐old man with end‐stage liver disease and associated coagulopathy without evidence of DIC who survived to discharge with intravascular management of a spontaneous retroperitoneal hemorrhage from the lumbar arteries. To our knowledge, this is the second reported case of spontaneous retroperitoneal hemorrhage in a cirrhotic in which the lumbar arteries were implicated and the first in which multiple arteries were found to be bleeding simultaneously. Any hospitalized patient who develops abdominal pain, flank pain, or hemodynamic instability in the context of coagulopathy, regardless of cause, warrants evaluation for retroperitoneal bleed. Appropriate early management includes immediate resuscitation, intensive monitoring, urgent imaging, and surgical and interventional radiology consultation in order to prevent a fatal outcome.

A 49‐year‐old retired air‐force officer presented with productive cough lasting 1 week. He was in good health except for back pain after a motor vehicle accident, 2 years ago, for which he took unspecified pain medications. His condition rapidly deteriorated necessitating mechanical ventilation. Chest x‐ray showed multilobar pneumonia (Figure 1). Blood and sputum cultures grew Serratia marcescens. Chest computed tomography (CT)‐scan showed multilobar involvement with debris in his airways that was thought to be respiratory secretions (Figure 2, arrows). He remained intubated for 2 weeks before his clinical status improved enough to permit extubation. Immediately following extubation, he coughed up a ball of crumpled, plastic material. His condition subsequently improved dramatically with complete radiological resolution of his pneumonia. On analyzing the foreign body, it turned out to be a fenestrated fentanyl patch (Figure 3). The patient then reported that he often cut‐up used fentanyl patches and sucked on them for an extra high. In retrospect, the endobronchial debris was actually the patch straddling the carina and had prevented rapid recovery. Aspiration of unusual foreign bodies have been reported in the literature. This foreign body masqueraded as respiratory secretions on imaging preventing early detection. This clinical encounter brings to light yet another innovative, but potentially deadly, form of drug abuse. It also emphasizes the importance of early bronchoscopy in nonresolving pneumonias.

Figure 1

Figure 2

Figure 3

A 49‐year‐old retired air‐force officer presented with productive cough lasting 1 week. He was in good health except for back pain after a motor vehicle accident, 2 years ago, for which he took unspecified pain medications. His condition rapidly deteriorated necessitating mechanical ventilation. Chest x‐ray showed multilobar pneumonia (Figure 1). Blood and sputum cultures grew Serratia marcescens. Chest computed tomography (CT)‐scan showed multilobar involvement with debris in his airways that was thought to be respiratory secretions (Figure 2, arrows). He remained intubated for 2 weeks before his clinical status improved enough to permit extubation. Immediately following extubation, he coughed up a ball of crumpled, plastic material. His condition subsequently improved dramatically with complete radiological resolution of his pneumonia. On analyzing the foreign body, it turned out to be a fenestrated fentanyl patch (Figure 3). The patient then reported that he often cut‐up used fentanyl patches and sucked on them for an extra high. In retrospect, the endobronchial debris was actually the patch straddling the carina and had prevented rapid recovery. Aspiration of unusual foreign bodies have been reported in the literature. This foreign body masqueraded as respiratory secretions on imaging preventing early detection. This clinical encounter brings to light yet another innovative, but potentially deadly, form of drug abuse. It also emphasizes the importance of early bronchoscopy in nonresolving pneumonias.

Figure 1

Figure 2

Figure 3

A 49‐year‐old retired air‐force officer presented with productive cough lasting 1 week. He was in good health except for back pain after a motor vehicle accident, 2 years ago, for which he took unspecified pain medications. His condition rapidly deteriorated necessitating mechanical ventilation. Chest x‐ray showed multilobar pneumonia (Figure 1). Blood and sputum cultures grew Serratia marcescens. Chest computed tomography (CT)‐scan showed multilobar involvement with debris in his airways that was thought to be respiratory secretions (Figure 2, arrows). He remained intubated for 2 weeks before his clinical status improved enough to permit extubation. Immediately following extubation, he coughed up a ball of crumpled, plastic material. His condition subsequently improved dramatically with complete radiological resolution of his pneumonia. On analyzing the foreign body, it turned out to be a fenestrated fentanyl patch (Figure 3). The patient then reported that he often cut‐up used fentanyl patches and sucked on them for an extra high. In retrospect, the endobronchial debris was actually the patch straddling the carina and had prevented rapid recovery. Aspiration of unusual foreign bodies have been reported in the literature. This foreign body masqueraded as respiratory secretions on imaging preventing early detection. This clinical encounter brings to light yet another innovative, but potentially deadly, form of drug abuse. It also emphasizes the importance of early bronchoscopy in nonresolving pneumonias.

Figure 1

Figure 2

Figure 3

Forty‐five‐million Americans speak a language other than English and more than 19 million of these speak English less than very wellor are limited English proficient (LEP).1 The number of non‐English‐speaking and LEP people in the US has risen in recent decades, presenting a challenge to healthcare systems to provide high‐quality, patient‐centered care for these patients.2

For outpatients, language barriers are a fundamental contributor to gaps in health care. In the clinic setting, patients who do not speak English well have less access to a usual source of care and lower rates of physician visits and preventive services.36 Even when patients with language barriers do have access to care, they have poorer adherence, decreased comprehension of their diagnoses, decreased satisfaction with care, and increased medication complications.710

Few studies, however, have examined how language influences outcomes of hospital care. Compared to English‐speakers, patients who do not speak English well may experience longer lengths of stay,11 and have more adverse events while in the hospital.12 However, these previous studies have not investigated outcomes immediately post‐hospitalization, such as readmission rates and mortality, nor have they directly addressed the interaction between ethnicity and language.

To understand these questions, we analyzed data collected from a university‐based teaching hospital which cares for patients of diverse cultural and language backgrounds. Using these data, we examined how patients' primary language influenced hospital costs, length of stay (LOS), 30‐day readmission, and 30‐day mortality risk.

Patients and Methods

Results

Conclusion/Discussion

Our results indicate that language barriers may contribute to higher readmission rates for non‐English speakers, but that they have less impact on care efficiency or mortality. This finding of an association between language and readmission, without a similar association with efficiency, suggests a potentially communication‐critical step in care.20, 21 Patients with language barriers are more likely to experience adverse events, and those events are often caused by errors in communication.12 It is conceivable that higher readmission risk for Chinese and Spanish speakers in our study was, at least in part, due to gaps in communication that are present in all patient groups, and exacerbated by the presence of a language barrier. This barrier is likely present during hospitalization but magnified at discharge, limiting caregivers' ability to understand patients' needs for home care, while simultaneously limiting patients' understanding of the discharge plan. After discharge, it is also possible that non‐English speakers are less able to communicate their needs as they arise, ormore subtlyfeel less supported by a primarily English speaking healthcare system. As in other clinical arenas,22 it is quite possible that increased access to professional interpreters in the hospital setting, and particularly at the time of discharge, would enhance communication and outcomes for LEP patients. Our interpreter services data showed that the patients in our study had quite limited access to staff professional interpreters.

Our findings differ somewhat from those of John‐Baptiste et al.,11 who found that language barriers contributed to increased LOS for patients with cardiac and major surgical diagnoses. Our study's findings are akin to recent research suggesting that being a monolingual Spanish speaker or receiving interpreter services may not significantly impact LOS or cost of hospitalization,23 and that LOS and in‐hospital mortality do not differ for non‐English speakers and English speakers after acute myocardial infarction.24 These studies, along with our results, suggest that that care efficiency in the hospital may be driven much more by clinical acuity (eg, the need to respond rapidly to urgent clinical signs such as hypotension, fever and respiratory distress) than by adequacy of communication. For example, elderly LEP patients may be even more likely than English speakers to have vigilant family members at the bedside throughout their hospitalization due to their need for communication assistance; these family members can quickly alert hospital staff to concerning changes in the patient's condition.

Our results also suggest the possibility that language and ethnicity are not monolithic concepts, and that even within language and ethnic groups there are potential differences in care pattern. For example, not speaking English may be a surrogate marker for unmeasured factors such as social supports and access to care. Language is intimately associated with culture; it remains plausible that cultural differences between highly acculturated and less acculturated members of a given ethno‐cultural group may have contributed to our observed differences in readmission rates. Differences in culture and associated factors, such as social support or use of multiple hospital systems, may account for lack of higher readmission risk in Russian speakers, while Chinese and Spanish speakers had higher readmission risk.

In addition, our finding that English‐speaking Latinos had lower readmission risk than any other group may be more consistent with their clinical characteristicseg, younger age, fewer comorbiditiesthan with cultural factors. Our finding that African American patients had the highest readmission risk in our hospital was both surprising and concerning. Some of this increased risk may be explained by clinical characteristics, such as higher comorbidities and higher rates of diagnoses leading to frequent admissions (eg, sickle cell disease); however, the reasons for this disparity deserves further investigation.

Our study has limitations. First, our data are administrative, and lack information about patients' educational attainment, social support, acculturation, utilization of other hospital systems, and usual source of care. Despite this, we were able to account for many significant covariates that might contribute to readmission rates, including age, insurance status, gender, comorbidities, and admission to the intensive care unit.2528 Second, our information about patients' English language proficiency is limited. While direct assessments of English proficiency are more accurate ways to determine a patient's ability to communicate with health care providers in English,29 our language validation work conducted in preparation for this study suggests that most of our patients recorded as having a non‐English primary language (87%) also have a low score on a language acculturation scale.

Third, only 14% of our non‐English speaking subjects utilized professional staff interpreters, and we had no information on the use of professional telephonic interpreters, or ad hoc interpretersfamily members, non‐interpreter staff membersand their impact on our results. It is well‐documented that ad hoc interpreters are used frequently in healthcare, particularly in the hospital setting, and thus we can assume this to be true in our study.30, 31 As noted above, it is likely that the advocacy of family members and friends at the bedside helped to minimize potential differences in care efficiency for patients with language barriers. Finally, our study was performed at a single university based hospital and may not produce results which are applicable to other care settings.

Our findings point to several avenues for future research on language barriers and hospitalized patients. First, the field would benefit from an examination of the impact of easy access to professional interpreters during hospitalization on outcomes of hospital care, in particular on readmission rates. Second, there is need for development and assessment of best practices for creating a culture of professional interpreter utilization in the hospital among physicians and nursing staff. Third, investigation of the role of caregiver presence in the hospital room and how this might differ by patient culture, age and language ability may further elucidate some of the differences across language groups observed in our study. Lastly, a more granular investigation of clinician‐patient communication and the importance of interpersonal processes of care on both patient satisfaction and understanding of and adherence to discharge instructions could lead to the development of detailed interventions to enhance this communication and these outcomes as it has done for communication‐sensitive outcomes in the outpatient arena.3234

In summary, our study suggests that higher risk for readmission can be added to the unfortunate list of outcomes which are worsened due to language barriers, pointing to transition from the hospital as a potentially communication‐critical step in care which may be amenable to intervention. Our findings also suggest that this risk can vary even between groups of patients who do not speak English primarily. Whether and to what degree language and communication barriers aloneincluding access to professional interpreters and patient‐centered communicationduring hospitalization, or differences in caregiver social support both during and after hospitalization as well as access to care post‐hospitalization contribute to these findings is a worthy subject of future research.