Epilepsy estimates available for the first time for every state show that the disorder is widespread, with at least 3.4 million people affected, according to the Centers for Disease Control and Prevention.

The CDC data also show that the number of people with epilepsy is increasing, probably as a result of population growth. The number of affected adults went from 2.3 million in 2010 to 3 million in 2015, and the number of children with epilepsy rose from 450,000 in 2007 to 470,000 in 2015, CDC investigators reported (MMWR. 2017 Aug 11;66[31]:821-5).

[email protected]

Epilepsy estimates available for the first time for every state show that the disorder is widespread, with at least 3.4 million people affected, according to the Centers for Disease Control and Prevention.

The CDC data also show that the number of people with epilepsy is increasing, probably as a result of population growth. The number of affected adults went from 2.3 million in 2010 to 3 million in 2015, and the number of children with epilepsy rose from 450,000 in 2007 to 470,000 in 2015, CDC investigators reported (MMWR. 2017 Aug 11;66[31]:821-5).

[email protected]

Epilepsy estimates available for the first time for every state show that the disorder is widespread, with at least 3.4 million people affected, according to the Centers for Disease Control and Prevention.

The CDC data also show that the number of people with epilepsy is increasing, probably as a result of population growth. The number of affected adults went from 2.3 million in 2010 to 3 million in 2015, and the number of children with epilepsy rose from 450,000 in 2007 to 470,000 in 2015, CDC investigators reported (MMWR. 2017 Aug 11;66[31]:821-5).

[email protected]

BOSTON – Under the newly revised Bethesda System for Reporting Thyroid Cytology, slated for official release in October 2017, the six cytology-based diagnostic categories for thyroid lesions stay exactly the same as in the 10-year-old first edition, but some associated malignancy risks have changed.

Important changes include molecular testing to further assess malignancy risk in thyroid nodules and the introduction of lobectomy as a treatment option, “which really wasn’t an option 10 years ago,” in the first iteration of the Bethesda System (New York: Springer US, 2010), its coauthor Edmund S. Cibas, MD, said at the World Congress on Thyroid Cancer.

Mitchel L. Zoler/Frontline Medical News

[email protected]

On Twitter @mitchelzoler

BOSTON – Under the newly revised Bethesda System for Reporting Thyroid Cytology, slated for official release in October 2017, the six cytology-based diagnostic categories for thyroid lesions stay exactly the same as in the 10-year-old first edition, but some associated malignancy risks have changed.

Important changes include molecular testing to further assess malignancy risk in thyroid nodules and the introduction of lobectomy as a treatment option, “which really wasn’t an option 10 years ago,” in the first iteration of the Bethesda System (New York: Springer US, 2010), its coauthor Edmund S. Cibas, MD, said at the World Congress on Thyroid Cancer.

Mitchel L. Zoler/Frontline Medical News

[email protected]

On Twitter @mitchelzoler

BOSTON – Under the newly revised Bethesda System for Reporting Thyroid Cytology, slated for official release in October 2017, the six cytology-based diagnostic categories for thyroid lesions stay exactly the same as in the 10-year-old first edition, but some associated malignancy risks have changed.

Important changes include molecular testing to further assess malignancy risk in thyroid nodules and the introduction of lobectomy as a treatment option, “which really wasn’t an option 10 years ago,” in the first iteration of the Bethesda System (New York: Springer US, 2010), its coauthor Edmund S. Cibas, MD, said at the World Congress on Thyroid Cancer.

Mitchel L. Zoler/Frontline Medical News

[email protected]

On Twitter @mitchelzoler

ESTES PARK, COLO. – Everyone diagnosed with osteoporosis deserves a laboratory assessment to rule out unsuspected secondary causes, according to Sterling West, MD. And he’s got a doozy of a workup he recommends to primary care physicians as “incredibly cost effective.”

“With this workup you’ll identify 98% of abnormalities at a mean cost of $366 per diagnosis. That’s incredibly cost effective. You’re going to get a lot of information with actually not very much outlay at all,” he said at a conference on internal medicine sponsored by the University of Colorado.

Bruce Jancin/Frontline Medical News

[email protected]
 

ESTES PARK, COLO. – Everyone diagnosed with osteoporosis deserves a laboratory assessment to rule out unsuspected secondary causes, according to Sterling West, MD. And he’s got a doozy of a workup he recommends to primary care physicians as “incredibly cost effective.”

“With this workup you’ll identify 98% of abnormalities at a mean cost of $366 per diagnosis. That’s incredibly cost effective. You’re going to get a lot of information with actually not very much outlay at all,” he said at a conference on internal medicine sponsored by the University of Colorado.

Bruce Jancin/Frontline Medical News

[email protected]
 

ESTES PARK, COLO. – Everyone diagnosed with osteoporosis deserves a laboratory assessment to rule out unsuspected secondary causes, according to Sterling West, MD. And he’s got a doozy of a workup he recommends to primary care physicians as “incredibly cost effective.”

“With this workup you’ll identify 98% of abnormalities at a mean cost of $366 per diagnosis. That’s incredibly cost effective. You’re going to get a lot of information with actually not very much outlay at all,” he said at a conference on internal medicine sponsored by the University of Colorado.

Bruce Jancin/Frontline Medical News

[email protected]
 

PARIS – Five-year hemodynamic results of the first randomized trial of transcatheter versus surgical aortic valve replacement in low-surgical-risk patients with severe aortic stenosis showed continued superior valve performance in the TAVR group, Lars Sondergaard, MD, reported at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.

“The durability results are very encouraging. We can’t see that the TAVR patients are doing worse. So I think this is setting the scene to try to move forward in patients at low risk and also in younger patients,” declared Dr. Sondergaard, professor of cardiology at the University of Copenhagen.

He presented an update from the Nordic Aortic Valve Intervention (NOTION) trial, a prospective, multicenter, randomized, all-comers clinical trial in which 280 patients with symptomatic severe aortic stenosis at low surgical risk were assigned to surgical aortic valve replacement (SAVR) or to TAVR with the self-expanding CoreValve. Their mean age was 79 years, with an average Society of Thoracic Surgeons projected risk of mortality score of 3%. Eighty-two percent of participants had an STS score below 4%. Roughly 40% of TAVR patients got the first-generation CoreValve in the 26-mm size, 40% received the 29-mm version, and the rest got the 31-mm CoreValve.

Bruce Jancin/Frontline Medical News

[email protected]

PARIS – Five-year hemodynamic results of the first randomized trial of transcatheter versus surgical aortic valve replacement in low-surgical-risk patients with severe aortic stenosis showed continued superior valve performance in the TAVR group, Lars Sondergaard, MD, reported at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.

“The durability results are very encouraging. We can’t see that the TAVR patients are doing worse. So I think this is setting the scene to try to move forward in patients at low risk and also in younger patients,” declared Dr. Sondergaard, professor of cardiology at the University of Copenhagen.

He presented an update from the Nordic Aortic Valve Intervention (NOTION) trial, a prospective, multicenter, randomized, all-comers clinical trial in which 280 patients with symptomatic severe aortic stenosis at low surgical risk were assigned to surgical aortic valve replacement (SAVR) or to TAVR with the self-expanding CoreValve. Their mean age was 79 years, with an average Society of Thoracic Surgeons projected risk of mortality score of 3%. Eighty-two percent of participants had an STS score below 4%. Roughly 40% of TAVR patients got the first-generation CoreValve in the 26-mm size, 40% received the 29-mm version, and the rest got the 31-mm CoreValve.

Bruce Jancin/Frontline Medical News

[email protected]

PARIS – Five-year hemodynamic results of the first randomized trial of transcatheter versus surgical aortic valve replacement in low-surgical-risk patients with severe aortic stenosis showed continued superior valve performance in the TAVR group, Lars Sondergaard, MD, reported at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.

“The durability results are very encouraging. We can’t see that the TAVR patients are doing worse. So I think this is setting the scene to try to move forward in patients at low risk and also in younger patients,” declared Dr. Sondergaard, professor of cardiology at the University of Copenhagen.

He presented an update from the Nordic Aortic Valve Intervention (NOTION) trial, a prospective, multicenter, randomized, all-comers clinical trial in which 280 patients with symptomatic severe aortic stenosis at low surgical risk were assigned to surgical aortic valve replacement (SAVR) or to TAVR with the self-expanding CoreValve. Their mean age was 79 years, with an average Society of Thoracic Surgeons projected risk of mortality score of 3%. Eighty-two percent of participants had an STS score below 4%. Roughly 40% of TAVR patients got the first-generation CoreValve in the 26-mm size, 40% received the 29-mm version, and the rest got the 31-mm CoreValve.

Bruce Jancin/Frontline Medical News

[email protected]

Hospitals typically allocate beds based on historical patient volumes. If funding decreases, hospitals will usually try to maximize resource utilization by allocating beds to attain occupancies close to 100% for significant periods of time. This will invariably cause days in which hospital occupancy exceeds capacity, at which time critical entry points (such as the emergency department and operating room) will become blocked. This creates significant concerns over the patient quality of care.

Hospital administrators have very few options when hospital occupancy exceeds 100%. They could postpone admissions for “planned” cases, bring in additional staff to increase capacity, or instigate additional methods to increase hospital discharges such as expanding care resources in the community. All options are costly, bothersome, or cannot be actioned immediately. The need for these options could be minimized by enabling hospital administrators to make more informed decisions regarding hospital bed management by knowing the likely number of discharges in the next 24 hours.

Predicting the number of people who will be discharged in the next day can be approached in several ways. One approach would be to calculate each patient’s expected length of stay and then use the variation around that estimate to calculate each day’s discharge probability. Several studies have attempted to model hospital length of stay using a broad assortment of methodologies, but a mechanism to accurately predict this outcome has been elusive^1,2 (with Verburg et al.³ concluding in their study’s abstract that “…it is difficult to predict length of stay…”). A second approach would be to use survival analysis methods to generate each patient’s hazard of discharge over time, which could be directly converted to an expected daily risk of discharge. However, this approach is complicated by the concurrent need to include time-dependent covariates and consider the competing risk of death in hospital, which can complicate survival modeling.^4,5 A third approach would be the implementation of a longitudinal analysis using marginal models to predict the daily probability of discharge,⁶ but this method quickly overwhelms computer resources when large datasets are present.

In this study, we decided to use nonparametric models to predict the daily number of hospital discharges. We first identified patient groups with distinct discharge patterns. We then calculated the conditional daily discharge probability of patients in each of these groups. Finally, these conditional daily discharge probabilities were then summed for each hospital day to generate the expected number of discharges in the next 24 hours. This paper details the methods we used to create our model and the accuracy of its predictions.

Study Setting and Databases Used for Analysis

The study took place at The Ottawa Hospital, a 1000-bed teaching hospital with 3 campuses that is the primary referral center in our region. The study was approved by our local research ethics board.

The Patient Registry Database records the date and time of admission for each patient (defined as the moment that a patient’s admission request is registered in the patient registration) and discharge (defined as the time when the patient’s discharge from hospital was entered into the patient registration) for hospital encounters. Emergency department encounters were also identified in the Patient Registry Database along with admission service, patient age and sex, and patient location throughout the admission. The Laboratory Database records all laboratory studies and results on all patients at the hospital.

Study Cohort

We used the Patient Registry Database to identify all people aged 1 year or more who were admitted to the hospital between January 1, 2013, and December 31, 2015. This time frame was selected to (i) ensure that data were complete; and (ii) complete calendar years of data were available for both derivation (patient-days in 2013-2014) and validation (2015) cohorts. Patients who were observed in the emergency room without admission to hospital were not included.

Study Outcome

The study outcome was the number of patients discharged from the hospital each day. For the analysis, the reference point for each day was 1 second past midnight; therefore, values for time-dependent covariates up to and including midnight were used to predict the number of discharges in the next 24 hours.

Study Covariates

Baseline (ie, time-independent) covariates included patient age and sex, admission service, hospital campus, whether or not the patient was admitted from the emergency department (all determined from the Patient Registry Database), and the Laboratory-based Acute Physiological Score (LAPS). The latter, which was calculated with the Laboratory Database using results for 14 tests (arterial pH, PaCO2, PaO2, anion gap, hematocrit, total white blood cell count, serum albumin, total bilirubin, creatinine, urea nitrogen, glucose, sodium, bicarbonate, and troponin I) measured in the 24-hour time frame preceding hospitalization, was derived by Escobar and colleagues⁷ to measure severity of illness and was subsequently validated in our hospital.⁸ The independent association of each laboratory perturbation with risk of death in hospital is reflected by the number of points assigned to each lab value with the total LAPS being the sum of these values. Time-dependent covariates included weekday in hospital and whether or not patients were in the intensive care unit.

Analysis

We used 3 stages to create a model to predict the daily expected number of discharges: we identified discharge risk strata containing patients having similar discharge patterns using data from patients in the derivation cohort (first stage); then, we generated the preliminary probability of discharge by determining the daily discharge probability in each discharge risk strata (second stage); finally, we modified the probability from the second stage based on the weekday and admission service and summed these probabilities to create the expected number of discharges on a particular date (third stage).

The first stage identified discharge risk strata based on the covariates listed above. This was determined by using a survival tree approach⁹ with proportional hazard regression models to generate the “splits.” These models were offered all covariates listed in the Study Covariates section. Admission service was clustered within 4 departments (obstetrics/gynecology, psychiatry, surgery, and medicine) and day of week was “binarized” into weekday/weekend-holiday (because the use of categorical variables with large numbers of groups can “stunt” regression trees due to small numbers of patients—and, therefore, statistical power—in each subgroup). The proportional hazards model identified the covariate having the strongest association with time to discharge (based on the Wald X² value divided by the degrees of freedom). This variable was then used to split the cohort into subgroups (with continuous covariates being categorized into quartiles). The proportional hazards model was then repeated in each subgroup (with the previous splitting variable[s] excluded from the model). This process continued until no variable was associated with time to discharge with a P value less than .0001. This survival-tree was then used to cluster all patients into distinct discharge risk strata.

In the second stage, we generated the preliminary probability of discharge for a specific date. This was calculated by assigning all patients in hospital to their discharge risk strata (Appendix). We then measured the probability of discharge on each hospitalization day in all discharge risk strata using data from the previous 180 days (we only used the prior 180 days of data to account for temporal changes in hospital discharge patterns). For example, consider a 75-year-old patient on her third hospital day under obstetrics/gynecology on December 19, 2015 (a Saturday). This patient would be assigned to risk stratum #133 (Appendix A). We then measured the probability of discharge of all patients in this discharge risk stratum hospitalized in the previous 6 months (ie, between June 22, 2015, and December 18, 2015) on each hospital day. For risk stratum #133, the probability of discharge on hospital day 3 was 0.1111; therefore, our sample patient’s preliminary expected discharge probability was 0.1111.

To attain stable daily discharge probability estimates, a minimum of 50 patients per discharge risk stratum-hospitalization day combination was required. If there were less than 50 patients for a particular hospitalization day in a particular discharge risk stratum, we grouped hospitalization days in that risk stratum together until the minimum of 50 patients was collected.

The third (and final) stage accounted for the lack of granularity when we created the discharge risk strata in the first stage. As we mentioned above, admission service was clustered into 4 departments and the day of week was clustered into weekend/weekday. However, important variations in discharge probabilities could still exist within departments and between particular days of the week.¹⁰ Therefore, we created a correction factor to adjust the preliminary expected number of discharges based on the admission division and day of week. This correction factor used data from the 180 days prior to the analysis date within which the expected daily number of discharges was calculated (using the methods above). The correction factor was the relative difference between the observed and expected number of discharges within each division-day of week grouping.

For example, to calculate the correction factor for our sample patient presented above (75-year-old patient on hospital day 3 under gynecology on Saturday, December 19, 2015), we measured the observed number of discharges from gynecology on Saturdays between June 22, 2015, and December 18, 2015, (n = 206) and the expected number of discharges (n = 195.255) resulting in a correction factor of (observed-expected)/expected = (195.255-206)/195.206 = 0.05503. Therefore, the final expected discharge probability for our sample patient was 0.1111+0.1111*0.05503=0.1172. The expected number of discharges on a particular date was the preliminary expected number of discharges on that date (generated in the second stage) multiplied by the correction factor for the corresponding division-day or week group.

There were 192,859 admissions involving patients more than 1 year of age that spent at least part of their hospitalization between January 1, 2013, and December 31, 2015 (Table). Patients were middle-aged and slightly female predominant, with about half being admitted from the emergency department. Approximately 80% of admissions were to surgical or medical services. More than 95% of admissions ended with a discharge from the hospital with the remainder ending in a death. Almost 30% of hospitalization days occurred on weekends or holidays. Hospitalizations in the derivation (2013-2014) and validation (2015) group were essentially the same, except there was a slight drop in hospital length of stay (from a median of 4 days to 3 days) between the 2 periods.

Patient and hospital covariates importantly influenced the daily conditional probability of discharge (Figure 1). Patients admitted to the obstetrics/gynecology department were notably more likely to be discharged from hospital with no influence from the day of week. In contrast, the probability of discharge decreased notably on the weekends in the other departments. Patients on the ward were much more likely to be discharged than those in the intensive care unit, with increasing age associated with a decreased discharge likelihood in the former but not the latter patients. Finally, discharge probabilities varied only slightly between campuses at our hospital with discharge risk decreasing as severity of illness (as measured by LAPS) increased.

Knowing how many patients will soon be discharged from the hospital should greatly facilitate hospital planning. This study showed that the TEND model used simple patient and hospitalization covariates to accurately predict the number of patients who will be discharged from hospital in the next day.

We believe that this study has several notable findings. First, we think that using a nonparametric approach to predicting the daily number of discharges importantly increased accuracy. This approach allowed us to generate expected likelihoods based on actual discharge probabilities at our hospital in the most recent 6 months of hospitalization-days within patients having discharge patterns that were very similar to the patient in question (ie, discharge risk strata, Appendix A). This ensured that trends in hospitalization habits were accounted for without the need of a period variable in our model. In addition, the lack of parameters in the model will make it easier to transplant it to other hospitals. Second, we think that the accuracy of the predictions were remarkable given the relative “crudeness” of our predictors. By using relatively simple factors, the TEND model was able to output accurate predictions for the number of daily discharges (Figure 3).

Disclosure: CvW is supported by a University of Ottawa Department of Medicine Clinician Scientist Chair. The authors have no conflicts of interest

Hospitals typically allocate beds based on historical patient volumes. If funding decreases, hospitals will usually try to maximize resource utilization by allocating beds to attain occupancies close to 100% for significant periods of time. This will invariably cause days in which hospital occupancy exceeds capacity, at which time critical entry points (such as the emergency department and operating room) will become blocked. This creates significant concerns over the patient quality of care.

Hospital administrators have very few options when hospital occupancy exceeds 100%. They could postpone admissions for “planned” cases, bring in additional staff to increase capacity, or instigate additional methods to increase hospital discharges such as expanding care resources in the community. All options are costly, bothersome, or cannot be actioned immediately. The need for these options could be minimized by enabling hospital administrators to make more informed decisions regarding hospital bed management by knowing the likely number of discharges in the next 24 hours.

Predicting the number of people who will be discharged in the next day can be approached in several ways. One approach would be to calculate each patient’s expected length of stay and then use the variation around that estimate to calculate each day’s discharge probability. Several studies have attempted to model hospital length of stay using a broad assortment of methodologies, but a mechanism to accurately predict this outcome has been elusive^1,2 (with Verburg et al.³ concluding in their study’s abstract that “…it is difficult to predict length of stay…”). A second approach would be to use survival analysis methods to generate each patient’s hazard of discharge over time, which could be directly converted to an expected daily risk of discharge. However, this approach is complicated by the concurrent need to include time-dependent covariates and consider the competing risk of death in hospital, which can complicate survival modeling.^4,5 A third approach would be the implementation of a longitudinal analysis using marginal models to predict the daily probability of discharge,⁶ but this method quickly overwhelms computer resources when large datasets are present.

In this study, we decided to use nonparametric models to predict the daily number of hospital discharges. We first identified patient groups with distinct discharge patterns. We then calculated the conditional daily discharge probability of patients in each of these groups. Finally, these conditional daily discharge probabilities were then summed for each hospital day to generate the expected number of discharges in the next 24 hours. This paper details the methods we used to create our model and the accuracy of its predictions.

Study Setting and Databases Used for Analysis

The study took place at The Ottawa Hospital, a 1000-bed teaching hospital with 3 campuses that is the primary referral center in our region. The study was approved by our local research ethics board.

The Patient Registry Database records the date and time of admission for each patient (defined as the moment that a patient’s admission request is registered in the patient registration) and discharge (defined as the time when the patient’s discharge from hospital was entered into the patient registration) for hospital encounters. Emergency department encounters were also identified in the Patient Registry Database along with admission service, patient age and sex, and patient location throughout the admission. The Laboratory Database records all laboratory studies and results on all patients at the hospital.

Study Cohort

We used the Patient Registry Database to identify all people aged 1 year or more who were admitted to the hospital between January 1, 2013, and December 31, 2015. This time frame was selected to (i) ensure that data were complete; and (ii) complete calendar years of data were available for both derivation (patient-days in 2013-2014) and validation (2015) cohorts. Patients who were observed in the emergency room without admission to hospital were not included.

Study Outcome

The study outcome was the number of patients discharged from the hospital each day. For the analysis, the reference point for each day was 1 second past midnight; therefore, values for time-dependent covariates up to and including midnight were used to predict the number of discharges in the next 24 hours.

Study Covariates

Baseline (ie, time-independent) covariates included patient age and sex, admission service, hospital campus, whether or not the patient was admitted from the emergency department (all determined from the Patient Registry Database), and the Laboratory-based Acute Physiological Score (LAPS). The latter, which was calculated with the Laboratory Database using results for 14 tests (arterial pH, PaCO2, PaO2, anion gap, hematocrit, total white blood cell count, serum albumin, total bilirubin, creatinine, urea nitrogen, glucose, sodium, bicarbonate, and troponin I) measured in the 24-hour time frame preceding hospitalization, was derived by Escobar and colleagues⁷ to measure severity of illness and was subsequently validated in our hospital.⁸ The independent association of each laboratory perturbation with risk of death in hospital is reflected by the number of points assigned to each lab value with the total LAPS being the sum of these values. Time-dependent covariates included weekday in hospital and whether or not patients were in the intensive care unit.

Analysis

We used 3 stages to create a model to predict the daily expected number of discharges: we identified discharge risk strata containing patients having similar discharge patterns using data from patients in the derivation cohort (first stage); then, we generated the preliminary probability of discharge by determining the daily discharge probability in each discharge risk strata (second stage); finally, we modified the probability from the second stage based on the weekday and admission service and summed these probabilities to create the expected number of discharges on a particular date (third stage).

The first stage identified discharge risk strata based on the covariates listed above. This was determined by using a survival tree approach⁹ with proportional hazard regression models to generate the “splits.” These models were offered all covariates listed in the Study Covariates section. Admission service was clustered within 4 departments (obstetrics/gynecology, psychiatry, surgery, and medicine) and day of week was “binarized” into weekday/weekend-holiday (because the use of categorical variables with large numbers of groups can “stunt” regression trees due to small numbers of patients—and, therefore, statistical power—in each subgroup). The proportional hazards model identified the covariate having the strongest association with time to discharge (based on the Wald X² value divided by the degrees of freedom). This variable was then used to split the cohort into subgroups (with continuous covariates being categorized into quartiles). The proportional hazards model was then repeated in each subgroup (with the previous splitting variable[s] excluded from the model). This process continued until no variable was associated with time to discharge with a P value less than .0001. This survival-tree was then used to cluster all patients into distinct discharge risk strata.

In the second stage, we generated the preliminary probability of discharge for a specific date. This was calculated by assigning all patients in hospital to their discharge risk strata (Appendix). We then measured the probability of discharge on each hospitalization day in all discharge risk strata using data from the previous 180 days (we only used the prior 180 days of data to account for temporal changes in hospital discharge patterns). For example, consider a 75-year-old patient on her third hospital day under obstetrics/gynecology on December 19, 2015 (a Saturday). This patient would be assigned to risk stratum #133 (Appendix A). We then measured the probability of discharge of all patients in this discharge risk stratum hospitalized in the previous 6 months (ie, between June 22, 2015, and December 18, 2015) on each hospital day. For risk stratum #133, the probability of discharge on hospital day 3 was 0.1111; therefore, our sample patient’s preliminary expected discharge probability was 0.1111.

To attain stable daily discharge probability estimates, a minimum of 50 patients per discharge risk stratum-hospitalization day combination was required. If there were less than 50 patients for a particular hospitalization day in a particular discharge risk stratum, we grouped hospitalization days in that risk stratum together until the minimum of 50 patients was collected.

The third (and final) stage accounted for the lack of granularity when we created the discharge risk strata in the first stage. As we mentioned above, admission service was clustered into 4 departments and the day of week was clustered into weekend/weekday. However, important variations in discharge probabilities could still exist within departments and between particular days of the week.¹⁰ Therefore, we created a correction factor to adjust the preliminary expected number of discharges based on the admission division and day of week. This correction factor used data from the 180 days prior to the analysis date within which the expected daily number of discharges was calculated (using the methods above). The correction factor was the relative difference between the observed and expected number of discharges within each division-day of week grouping.

For example, to calculate the correction factor for our sample patient presented above (75-year-old patient on hospital day 3 under gynecology on Saturday, December 19, 2015), we measured the observed number of discharges from gynecology on Saturdays between June 22, 2015, and December 18, 2015, (n = 206) and the expected number of discharges (n = 195.255) resulting in a correction factor of (observed-expected)/expected = (195.255-206)/195.206 = 0.05503. Therefore, the final expected discharge probability for our sample patient was 0.1111+0.1111*0.05503=0.1172. The expected number of discharges on a particular date was the preliminary expected number of discharges on that date (generated in the second stage) multiplied by the correction factor for the corresponding division-day or week group.

There were 192,859 admissions involving patients more than 1 year of age that spent at least part of their hospitalization between January 1, 2013, and December 31, 2015 (Table). Patients were middle-aged and slightly female predominant, with about half being admitted from the emergency department. Approximately 80% of admissions were to surgical or medical services. More than 95% of admissions ended with a discharge from the hospital with the remainder ending in a death. Almost 30% of hospitalization days occurred on weekends or holidays. Hospitalizations in the derivation (2013-2014) and validation (2015) group were essentially the same, except there was a slight drop in hospital length of stay (from a median of 4 days to 3 days) between the 2 periods.

Patient and hospital covariates importantly influenced the daily conditional probability of discharge (Figure 1). Patients admitted to the obstetrics/gynecology department were notably more likely to be discharged from hospital with no influence from the day of week. In contrast, the probability of discharge decreased notably on the weekends in the other departments. Patients on the ward were much more likely to be discharged than those in the intensive care unit, with increasing age associated with a decreased discharge likelihood in the former but not the latter patients. Finally, discharge probabilities varied only slightly between campuses at our hospital with discharge risk decreasing as severity of illness (as measured by LAPS) increased.

Knowing how many patients will soon be discharged from the hospital should greatly facilitate hospital planning. This study showed that the TEND model used simple patient and hospitalization covariates to accurately predict the number of patients who will be discharged from hospital in the next day.

We believe that this study has several notable findings. First, we think that using a nonparametric approach to predicting the daily number of discharges importantly increased accuracy. This approach allowed us to generate expected likelihoods based on actual discharge probabilities at our hospital in the most recent 6 months of hospitalization-days within patients having discharge patterns that were very similar to the patient in question (ie, discharge risk strata, Appendix A). This ensured that trends in hospitalization habits were accounted for without the need of a period variable in our model. In addition, the lack of parameters in the model will make it easier to transplant it to other hospitals. Second, we think that the accuracy of the predictions were remarkable given the relative “crudeness” of our predictors. By using relatively simple factors, the TEND model was able to output accurate predictions for the number of daily discharges (Figure 3).

Disclosure: CvW is supported by a University of Ottawa Department of Medicine Clinician Scientist Chair. The authors have no conflicts of interest

Hospitals typically allocate beds based on historical patient volumes. If funding decreases, hospitals will usually try to maximize resource utilization by allocating beds to attain occupancies close to 100% for significant periods of time. This will invariably cause days in which hospital occupancy exceeds capacity, at which time critical entry points (such as the emergency department and operating room) will become blocked. This creates significant concerns over the patient quality of care.

Hospital administrators have very few options when hospital occupancy exceeds 100%. They could postpone admissions for “planned” cases, bring in additional staff to increase capacity, or instigate additional methods to increase hospital discharges such as expanding care resources in the community. All options are costly, bothersome, or cannot be actioned immediately. The need for these options could be minimized by enabling hospital administrators to make more informed decisions regarding hospital bed management by knowing the likely number of discharges in the next 24 hours.

Predicting the number of people who will be discharged in the next day can be approached in several ways. One approach would be to calculate each patient’s expected length of stay and then use the variation around that estimate to calculate each day’s discharge probability. Several studies have attempted to model hospital length of stay using a broad assortment of methodologies, but a mechanism to accurately predict this outcome has been elusive^1,2 (with Verburg et al.³ concluding in their study’s abstract that “…it is difficult to predict length of stay…”). A second approach would be to use survival analysis methods to generate each patient’s hazard of discharge over time, which could be directly converted to an expected daily risk of discharge. However, this approach is complicated by the concurrent need to include time-dependent covariates and consider the competing risk of death in hospital, which can complicate survival modeling.^4,5 A third approach would be the implementation of a longitudinal analysis using marginal models to predict the daily probability of discharge,⁶ but this method quickly overwhelms computer resources when large datasets are present.

In this study, we decided to use nonparametric models to predict the daily number of hospital discharges. We first identified patient groups with distinct discharge patterns. We then calculated the conditional daily discharge probability of patients in each of these groups. Finally, these conditional daily discharge probabilities were then summed for each hospital day to generate the expected number of discharges in the next 24 hours. This paper details the methods we used to create our model and the accuracy of its predictions.

Study Setting and Databases Used for Analysis

The study took place at The Ottawa Hospital, a 1000-bed teaching hospital with 3 campuses that is the primary referral center in our region. The study was approved by our local research ethics board.

The Patient Registry Database records the date and time of admission for each patient (defined as the moment that a patient’s admission request is registered in the patient registration) and discharge (defined as the time when the patient’s discharge from hospital was entered into the patient registration) for hospital encounters. Emergency department encounters were also identified in the Patient Registry Database along with admission service, patient age and sex, and patient location throughout the admission. The Laboratory Database records all laboratory studies and results on all patients at the hospital.

Study Cohort

We used the Patient Registry Database to identify all people aged 1 year or more who were admitted to the hospital between January 1, 2013, and December 31, 2015. This time frame was selected to (i) ensure that data were complete; and (ii) complete calendar years of data were available for both derivation (patient-days in 2013-2014) and validation (2015) cohorts. Patients who were observed in the emergency room without admission to hospital were not included.

Study Outcome

The study outcome was the number of patients discharged from the hospital each day. For the analysis, the reference point for each day was 1 second past midnight; therefore, values for time-dependent covariates up to and including midnight were used to predict the number of discharges in the next 24 hours.

Study Covariates

Baseline (ie, time-independent) covariates included patient age and sex, admission service, hospital campus, whether or not the patient was admitted from the emergency department (all determined from the Patient Registry Database), and the Laboratory-based Acute Physiological Score (LAPS). The latter, which was calculated with the Laboratory Database using results for 14 tests (arterial pH, PaCO2, PaO2, anion gap, hematocrit, total white blood cell count, serum albumin, total bilirubin, creatinine, urea nitrogen, glucose, sodium, bicarbonate, and troponin I) measured in the 24-hour time frame preceding hospitalization, was derived by Escobar and colleagues⁷ to measure severity of illness and was subsequently validated in our hospital.⁸ The independent association of each laboratory perturbation with risk of death in hospital is reflected by the number of points assigned to each lab value with the total LAPS being the sum of these values. Time-dependent covariates included weekday in hospital and whether or not patients were in the intensive care unit.

Analysis

We used 3 stages to create a model to predict the daily expected number of discharges: we identified discharge risk strata containing patients having similar discharge patterns using data from patients in the derivation cohort (first stage); then, we generated the preliminary probability of discharge by determining the daily discharge probability in each discharge risk strata (second stage); finally, we modified the probability from the second stage based on the weekday and admission service and summed these probabilities to create the expected number of discharges on a particular date (third stage).

The first stage identified discharge risk strata based on the covariates listed above. This was determined by using a survival tree approach⁹ with proportional hazard regression models to generate the “splits.” These models were offered all covariates listed in the Study Covariates section. Admission service was clustered within 4 departments (obstetrics/gynecology, psychiatry, surgery, and medicine) and day of week was “binarized” into weekday/weekend-holiday (because the use of categorical variables with large numbers of groups can “stunt” regression trees due to small numbers of patients—and, therefore, statistical power—in each subgroup). The proportional hazards model identified the covariate having the strongest association with time to discharge (based on the Wald X² value divided by the degrees of freedom). This variable was then used to split the cohort into subgroups (with continuous covariates being categorized into quartiles). The proportional hazards model was then repeated in each subgroup (with the previous splitting variable[s] excluded from the model). This process continued until no variable was associated with time to discharge with a P value less than .0001. This survival-tree was then used to cluster all patients into distinct discharge risk strata.

In the second stage, we generated the preliminary probability of discharge for a specific date. This was calculated by assigning all patients in hospital to their discharge risk strata (Appendix). We then measured the probability of discharge on each hospitalization day in all discharge risk strata using data from the previous 180 days (we only used the prior 180 days of data to account for temporal changes in hospital discharge patterns). For example, consider a 75-year-old patient on her third hospital day under obstetrics/gynecology on December 19, 2015 (a Saturday). This patient would be assigned to risk stratum #133 (Appendix A). We then measured the probability of discharge of all patients in this discharge risk stratum hospitalized in the previous 6 months (ie, between June 22, 2015, and December 18, 2015) on each hospital day. For risk stratum #133, the probability of discharge on hospital day 3 was 0.1111; therefore, our sample patient’s preliminary expected discharge probability was 0.1111.

To attain stable daily discharge probability estimates, a minimum of 50 patients per discharge risk stratum-hospitalization day combination was required. If there were less than 50 patients for a particular hospitalization day in a particular discharge risk stratum, we grouped hospitalization days in that risk stratum together until the minimum of 50 patients was collected.

The third (and final) stage accounted for the lack of granularity when we created the discharge risk strata in the first stage. As we mentioned above, admission service was clustered into 4 departments and the day of week was clustered into weekend/weekday. However, important variations in discharge probabilities could still exist within departments and between particular days of the week.¹⁰ Therefore, we created a correction factor to adjust the preliminary expected number of discharges based on the admission division and day of week. This correction factor used data from the 180 days prior to the analysis date within which the expected daily number of discharges was calculated (using the methods above). The correction factor was the relative difference between the observed and expected number of discharges within each division-day of week grouping.

For example, to calculate the correction factor for our sample patient presented above (75-year-old patient on hospital day 3 under gynecology on Saturday, December 19, 2015), we measured the observed number of discharges from gynecology on Saturdays between June 22, 2015, and December 18, 2015, (n = 206) and the expected number of discharges (n = 195.255) resulting in a correction factor of (observed-expected)/expected = (195.255-206)/195.206 = 0.05503. Therefore, the final expected discharge probability for our sample patient was 0.1111+0.1111*0.05503=0.1172. The expected number of discharges on a particular date was the preliminary expected number of discharges on that date (generated in the second stage) multiplied by the correction factor for the corresponding division-day or week group.

There were 192,859 admissions involving patients more than 1 year of age that spent at least part of their hospitalization between January 1, 2013, and December 31, 2015 (Table). Patients were middle-aged and slightly female predominant, with about half being admitted from the emergency department. Approximately 80% of admissions were to surgical or medical services. More than 95% of admissions ended with a discharge from the hospital with the remainder ending in a death. Almost 30% of hospitalization days occurred on weekends or holidays. Hospitalizations in the derivation (2013-2014) and validation (2015) group were essentially the same, except there was a slight drop in hospital length of stay (from a median of 4 days to 3 days) between the 2 periods.

Patient and hospital covariates importantly influenced the daily conditional probability of discharge (Figure 1). Patients admitted to the obstetrics/gynecology department were notably more likely to be discharged from hospital with no influence from the day of week. In contrast, the probability of discharge decreased notably on the weekends in the other departments. Patients on the ward were much more likely to be discharged than those in the intensive care unit, with increasing age associated with a decreased discharge likelihood in the former but not the latter patients. Finally, discharge probabilities varied only slightly between campuses at our hospital with discharge risk decreasing as severity of illness (as measured by LAPS) increased.

Knowing how many patients will soon be discharged from the hospital should greatly facilitate hospital planning. This study showed that the TEND model used simple patient and hospitalization covariates to accurately predict the number of patients who will be discharged from hospital in the next day.

We believe that this study has several notable findings. First, we think that using a nonparametric approach to predicting the daily number of discharges importantly increased accuracy. This approach allowed us to generate expected likelihoods based on actual discharge probabilities at our hospital in the most recent 6 months of hospitalization-days within patients having discharge patterns that were very similar to the patient in question (ie, discharge risk strata, Appendix A). This ensured that trends in hospitalization habits were accounted for without the need of a period variable in our model. In addition, the lack of parameters in the model will make it easier to transplant it to other hospitals. Second, we think that the accuracy of the predictions were remarkable given the relative “crudeness” of our predictors. By using relatively simple factors, the TEND model was able to output accurate predictions for the number of daily discharges (Figure 3).

Disclosure: CvW is supported by a University of Ottawa Department of Medicine Clinician Scientist Chair. The authors have no conflicts of interest

In early August, an international team of biologists reported injecting gene editing proteins into more than a hundred human embryos in Portland, Ore. The scale and success of such experimentation with human embryos is unprecedented in the United States. Given the highly experimental nature of fertility clinics in the United States and abroad, many suggest that these findings open the door to designer babies. A careful read of the report, however, indicates that the door is still quite closed, perhaps cracked open just a little.

The research team used a new method of cutting the genome, called CRISPR-Cas9. CRISPR utilizes two key components that the team combined in a test tube together: a Cas9 protein that can cut the DNA and a synthetic RNA that can guide the protein to cut a 20-letter sequence in the human genome specifically. In these experiments, the Cas9-RNA protein was designed to cut a pathogenic mutation in the MYBPC3 gene, which can cause hypertrophic cardiomyopathy. The research team could not obtain human zygotes with this mutation on both copies of the genome (a rare homozygous genotype). Such zygotes would have the most severe phenotype and be the most compelling test case for CRISPR. Instead, they focused on gene editing heterozygous human zygotes that have one normal maternal copy of the MYBPC3 gene and one pathogenic paternal copy. The heterozygous zygotes were produced by the research team via in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) using sperm donated by males carrying the pathogenic mutation (Nature. 2017 Aug 2. doi: 10.1038/nature23305).

When researchers injected the Cas9-RNA protein targeting the mutation into already fertilized zygotes, they found that 67% of the resulting embryos had two normal copies of the MYBPC3 gene. Without gene editing, approximately 50% of the embryos would have two normal copies, because the male sperm donor would produce equal numbers of sperm with normal and pathogenic genotypes. Thus, editing likely corrected only about 17% of the embryos that would have otherwise had one pathogenic paternal mutation. Thirty-six percent of embryos had additional mutations from imprecise gene editing. Further, some of the gene edits and additional mutations were mosaic, meaning that the resulting embryo harbored many different genotypes.

To overcome these challenges, the research team precisely controlled the timing of CRISPR injection to coincide with fertilization. With controlled timing, gene editing was restricted to only the paternal pathogenic mutation, resulting in 72% of all injected embryos having two normal copies of the gene in all cells without any mosaicism. Whole genome sequencing revealed no additional mutations above the detection limit of the assay. Finally, preimplantation development proceeded normally to the blastocyst stage, suggesting that the edited embryos have no functional deficits from the procedure.

A surprising finding was that new sequences could not be put into the embryo. The research team had coinjected a synthetic DNA template that differed from the normal maternal copy, but never saw this sequence incorporated into any embryo. Instead, the zygote utilized the maternal copy of the gene with the normal sequence as a template for repairing the DNA cut in the paternal copy produced by CRISPR. The biology behind this repair process is poorly understood and has not been previously reported with other human cell types. These observations suggest that we cannot easily “write” our genome. Instead, our vocabulary is limited to what is already within either the maternal or paternal copy of the genome. In other words, designer babies are not around the corner. While preimplantation genetic diagnosis (PGD) is still currently the safest way to avoid passing on autosomal dominant mutations, these new findings could enable correction of such mutations within IVF embryos, resulting in a larger pool of embryos for IVF clinics to work with.

Apart from these technical challenges, the National Academies has not given a green light to implant edited human embryos. Instead, the organization calls for several requirements to be met, including “broad societal consensus” on the need for this type of intervention. While it is not clear whether or how consensus could be achieved, it is clear that scientists, clinicians, and patients will need help from the rest of society for this research to have an impact clinically.

Dr. Saha is assistant professor of biomedical engineering at the Wisconsin Institute for Discovery at the University of Wisconsin, Madison. His lab works on gene editing of human cells. He has patent filings through the Wisconsin Alumni Research Foundation on gene editing inventions.

In early August, an international team of biologists reported injecting gene editing proteins into more than a hundred human embryos in Portland, Ore. The scale and success of such experimentation with human embryos is unprecedented in the United States. Given the highly experimental nature of fertility clinics in the United States and abroad, many suggest that these findings open the door to designer babies. A careful read of the report, however, indicates that the door is still quite closed, perhaps cracked open just a little.

The research team used a new method of cutting the genome, called CRISPR-Cas9. CRISPR utilizes two key components that the team combined in a test tube together: a Cas9 protein that can cut the DNA and a synthetic RNA that can guide the protein to cut a 20-letter sequence in the human genome specifically. In these experiments, the Cas9-RNA protein was designed to cut a pathogenic mutation in the MYBPC3 gene, which can cause hypertrophic cardiomyopathy. The research team could not obtain human zygotes with this mutation on both copies of the genome (a rare homozygous genotype). Such zygotes would have the most severe phenotype and be the most compelling test case for CRISPR. Instead, they focused on gene editing heterozygous human zygotes that have one normal maternal copy of the MYBPC3 gene and one pathogenic paternal copy. The heterozygous zygotes were produced by the research team via in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) using sperm donated by males carrying the pathogenic mutation (Nature. 2017 Aug 2. doi: 10.1038/nature23305).

When researchers injected the Cas9-RNA protein targeting the mutation into already fertilized zygotes, they found that 67% of the resulting embryos had two normal copies of the MYBPC3 gene. Without gene editing, approximately 50% of the embryos would have two normal copies, because the male sperm donor would produce equal numbers of sperm with normal and pathogenic genotypes. Thus, editing likely corrected only about 17% of the embryos that would have otherwise had one pathogenic paternal mutation. Thirty-six percent of embryos had additional mutations from imprecise gene editing. Further, some of the gene edits and additional mutations were mosaic, meaning that the resulting embryo harbored many different genotypes.

To overcome these challenges, the research team precisely controlled the timing of CRISPR injection to coincide with fertilization. With controlled timing, gene editing was restricted to only the paternal pathogenic mutation, resulting in 72% of all injected embryos having two normal copies of the gene in all cells without any mosaicism. Whole genome sequencing revealed no additional mutations above the detection limit of the assay. Finally, preimplantation development proceeded normally to the blastocyst stage, suggesting that the edited embryos have no functional deficits from the procedure.

A surprising finding was that new sequences could not be put into the embryo. The research team had coinjected a synthetic DNA template that differed from the normal maternal copy, but never saw this sequence incorporated into any embryo. Instead, the zygote utilized the maternal copy of the gene with the normal sequence as a template for repairing the DNA cut in the paternal copy produced by CRISPR. The biology behind this repair process is poorly understood and has not been previously reported with other human cell types. These observations suggest that we cannot easily “write” our genome. Instead, our vocabulary is limited to what is already within either the maternal or paternal copy of the genome. In other words, designer babies are not around the corner. While preimplantation genetic diagnosis (PGD) is still currently the safest way to avoid passing on autosomal dominant mutations, these new findings could enable correction of such mutations within IVF embryos, resulting in a larger pool of embryos for IVF clinics to work with.

Apart from these technical challenges, the National Academies has not given a green light to implant edited human embryos. Instead, the organization calls for several requirements to be met, including “broad societal consensus” on the need for this type of intervention. While it is not clear whether or how consensus could be achieved, it is clear that scientists, clinicians, and patients will need help from the rest of society for this research to have an impact clinically.

Dr. Saha is assistant professor of biomedical engineering at the Wisconsin Institute for Discovery at the University of Wisconsin, Madison. His lab works on gene editing of human cells. He has patent filings through the Wisconsin Alumni Research Foundation on gene editing inventions.

In early August, an international team of biologists reported injecting gene editing proteins into more than a hundred human embryos in Portland, Ore. The scale and success of such experimentation with human embryos is unprecedented in the United States. Given the highly experimental nature of fertility clinics in the United States and abroad, many suggest that these findings open the door to designer babies. A careful read of the report, however, indicates that the door is still quite closed, perhaps cracked open just a little.

The research team used a new method of cutting the genome, called CRISPR-Cas9. CRISPR utilizes two key components that the team combined in a test tube together: a Cas9 protein that can cut the DNA and a synthetic RNA that can guide the protein to cut a 20-letter sequence in the human genome specifically. In these experiments, the Cas9-RNA protein was designed to cut a pathogenic mutation in the MYBPC3 gene, which can cause hypertrophic cardiomyopathy. The research team could not obtain human zygotes with this mutation on both copies of the genome (a rare homozygous genotype). Such zygotes would have the most severe phenotype and be the most compelling test case for CRISPR. Instead, they focused on gene editing heterozygous human zygotes that have one normal maternal copy of the MYBPC3 gene and one pathogenic paternal copy. The heterozygous zygotes were produced by the research team via in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) using sperm donated by males carrying the pathogenic mutation (Nature. 2017 Aug 2. doi: 10.1038/nature23305).

When researchers injected the Cas9-RNA protein targeting the mutation into already fertilized zygotes, they found that 67% of the resulting embryos had two normal copies of the MYBPC3 gene. Without gene editing, approximately 50% of the embryos would have two normal copies, because the male sperm donor would produce equal numbers of sperm with normal and pathogenic genotypes. Thus, editing likely corrected only about 17% of the embryos that would have otherwise had one pathogenic paternal mutation. Thirty-six percent of embryos had additional mutations from imprecise gene editing. Further, some of the gene edits and additional mutations were mosaic, meaning that the resulting embryo harbored many different genotypes.

To overcome these challenges, the research team precisely controlled the timing of CRISPR injection to coincide with fertilization. With controlled timing, gene editing was restricted to only the paternal pathogenic mutation, resulting in 72% of all injected embryos having two normal copies of the gene in all cells without any mosaicism. Whole genome sequencing revealed no additional mutations above the detection limit of the assay. Finally, preimplantation development proceeded normally to the blastocyst stage, suggesting that the edited embryos have no functional deficits from the procedure.

A surprising finding was that new sequences could not be put into the embryo. The research team had coinjected a synthetic DNA template that differed from the normal maternal copy, but never saw this sequence incorporated into any embryo. Instead, the zygote utilized the maternal copy of the gene with the normal sequence as a template for repairing the DNA cut in the paternal copy produced by CRISPR. The biology behind this repair process is poorly understood and has not been previously reported with other human cell types. These observations suggest that we cannot easily “write” our genome. Instead, our vocabulary is limited to what is already within either the maternal or paternal copy of the genome. In other words, designer babies are not around the corner. While preimplantation genetic diagnosis (PGD) is still currently the safest way to avoid passing on autosomal dominant mutations, these new findings could enable correction of such mutations within IVF embryos, resulting in a larger pool of embryos for IVF clinics to work with.

Apart from these technical challenges, the National Academies has not given a green light to implant edited human embryos. Instead, the organization calls for several requirements to be met, including “broad societal consensus” on the need for this type of intervention. While it is not clear whether or how consensus could be achieved, it is clear that scientists, clinicians, and patients will need help from the rest of society for this research to have an impact clinically.

Dr. Saha is assistant professor of biomedical engineering at the Wisconsin Institute for Discovery at the University of Wisconsin, Madison. His lab works on gene editing of human cells. He has patent filings through the Wisconsin Alumni Research Foundation on gene editing inventions.

SAN DIEGO – Listen to the feet of your patients, and don’t just focus on glucose control as a way to combat diabetic neuropathy, according to Eva L. Feldman, MD, PhD.

Regular foot exams are crucial for patients with diabetes because of the high risk of diabetic neuropathy (DN), Dr. Feldman, professor of neurology and director of the Program for Neurology Research and Discovery at University of Michigan, Ann Arbor, said in a presentation at the annual scientific sessions of the American Diabetes Association.

copyright Jorge Salcedo/Thinkstock

SAN DIEGO – Listen to the feet of your patients, and don’t just focus on glucose control as a way to combat diabetic neuropathy, according to Eva L. Feldman, MD, PhD.

Regular foot exams are crucial for patients with diabetes because of the high risk of diabetic neuropathy (DN), Dr. Feldman, professor of neurology and director of the Program for Neurology Research and Discovery at University of Michigan, Ann Arbor, said in a presentation at the annual scientific sessions of the American Diabetes Association.

copyright Jorge Salcedo/Thinkstock

SAN DIEGO – Listen to the feet of your patients, and don’t just focus on glucose control as a way to combat diabetic neuropathy, according to Eva L. Feldman, MD, PhD.

Regular foot exams are crucial for patients with diabetes because of the high risk of diabetic neuropathy (DN), Dr. Feldman, professor of neurology and director of the Program for Neurology Research and Discovery at University of Michigan, Ann Arbor, said in a presentation at the annual scientific sessions of the American Diabetes Association.

copyright Jorge Salcedo/Thinkstock

The agenda for this year’s Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) annual meeting included sessions on juvenile psoriatic arthritis (PsA), the microbiome in psoriasis and PsA, and setting up longitudinal cohort studies. There was also a workshop to define the clinical fit and feasibility of PsA outcome measures for clinical trials and updates on several GRAPPA-associated projects.

The meeting, held July 13-15 in Amsterdam, opened with the annual trainee session. This year’s trainee session attracted more than 40 abstracts, of which 6 were selected for oral presentations on a variety of topics including outcome measure assessment, imaging, the microbiome in PsA, and T-cell subsets in synovial fluid.

Among the most intriguing sessions at the GRAPPA meeting was a series of three talks by Jose Scher, MD, Hok Bing Thio, MD, PhD, and Dirk Elewaut, PhD, that introduced the audience to the complexity of the micro-organisms that call the human host “home” and the potential role in the development of inflammatory conditions, in particular psoriasis and PsA. These talks touched on the potential uses of the microbiome of the gut, skin, and oral mucosa in predicting therapy response and modulating the immune system.

Dafna Gladman, MD, led a session on “how to set up a cohort.” In this session, speakers provided a road map for setting up a longitudinal cohort and discussed opportunities and challenges along the way. This session provided a foundation for a meeting that followed the annual meeting, the GRAPPA Research Collaborative Network meeting (held July 15-16). The RCN meeting aimed to develop a plan for a GRAPPA research network supporting collaborative research to identify biomarkers and outcomes in psoriasis and PsA. Speakers/panelists from academics and industry kicked off the meeting with a discussion of prior experiences in establishing international cohort studies. Subsequent sessions presented individual aspects of beginning a longitudinal cohort study for biomarkers, including methods for sample collection, regulatory processes, and data collected at potential sites; capturing and harmonizing clinical data for such studies; and policies for publication of RCN studies.

This year’s workshop was focused on the GRAPPA-Outcome Measures in Rheumatology (OMERACT) working group’s plan to develop a Core Outcome Measure Set, a collection of outcome measurement instruments to be used in randomized, controlled trials (RCTs) for PsA. During the plenary session, the team presented a process for the group to evaluate instruments, including assessment of match to the domain or concept of interest, feasibility, construct validity, and discrimination (including reliability and the ability to distinguish between two groups such as responders and nonresponders). Breakout groups then discussed one of the six tools being considered and voted on match to the domain of interest and feasibility for RCTs.

In a skin session, ongoing efforts to standardize and simplify the measurement of skin psoriasis in both the clinic and RCTs were presented. While the Psoriasis Area and Severity Index (PASI) is the current standard for measuring psoriasis severity and response to therapy in RCTs, the PASI is challenging for use in clinical practice. Joseph Merola, MD, presented data to support the use of the psoriasis Physician Global Assessment (PGA) x body surface area (BSA), a simpler measure than the PASI, as a potential substitute for the PASI. Updates from the International Dermatology Outcome Measures board included reporting of the psoriasis core domain set and a summary of the PsA symptoms working group’s efforts to identify patient-reported outcomes to identify and measure PsA among patients enrolled in psoriasis RCTs. April Armstrong, MD, discussed the National Psoriasis Foundation’s efforts to develop treat-to-target goals for psoriasis. Finally, Laura Coates, MBChB, PhD, presented a report on her efforts to examine a variety of psoriasis cut points for minimal disease activity and very low disease activity outcomes.
 

Dr. Ogdie is director of the Penn Psoriatic Arthritis Clinic at the University of Pennsylvania, Philadelphia, and is a member of the GRAPPA Steering Committee.

The agenda for this year’s Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) annual meeting included sessions on juvenile psoriatic arthritis (PsA), the microbiome in psoriasis and PsA, and setting up longitudinal cohort studies. There was also a workshop to define the clinical fit and feasibility of PsA outcome measures for clinical trials and updates on several GRAPPA-associated projects.

The meeting, held July 13-15 in Amsterdam, opened with the annual trainee session. This year’s trainee session attracted more than 40 abstracts, of which 6 were selected for oral presentations on a variety of topics including outcome measure assessment, imaging, the microbiome in PsA, and T-cell subsets in synovial fluid.

Among the most intriguing sessions at the GRAPPA meeting was a series of three talks by Jose Scher, MD, Hok Bing Thio, MD, PhD, and Dirk Elewaut, PhD, that introduced the audience to the complexity of the micro-organisms that call the human host “home” and the potential role in the development of inflammatory conditions, in particular psoriasis and PsA. These talks touched on the potential uses of the microbiome of the gut, skin, and oral mucosa in predicting therapy response and modulating the immune system.

Dafna Gladman, MD, led a session on “how to set up a cohort.” In this session, speakers provided a road map for setting up a longitudinal cohort and discussed opportunities and challenges along the way. This session provided a foundation for a meeting that followed the annual meeting, the GRAPPA Research Collaborative Network meeting (held July 15-16). The RCN meeting aimed to develop a plan for a GRAPPA research network supporting collaborative research to identify biomarkers and outcomes in psoriasis and PsA. Speakers/panelists from academics and industry kicked off the meeting with a discussion of prior experiences in establishing international cohort studies. Subsequent sessions presented individual aspects of beginning a longitudinal cohort study for biomarkers, including methods for sample collection, regulatory processes, and data collected at potential sites; capturing and harmonizing clinical data for such studies; and policies for publication of RCN studies.

This year’s workshop was focused on the GRAPPA-Outcome Measures in Rheumatology (OMERACT) working group’s plan to develop a Core Outcome Measure Set, a collection of outcome measurement instruments to be used in randomized, controlled trials (RCTs) for PsA. During the plenary session, the team presented a process for the group to evaluate instruments, including assessment of match to the domain or concept of interest, feasibility, construct validity, and discrimination (including reliability and the ability to distinguish between two groups such as responders and nonresponders). Breakout groups then discussed one of the six tools being considered and voted on match to the domain of interest and feasibility for RCTs.

In a skin session, ongoing efforts to standardize and simplify the measurement of skin psoriasis in both the clinic and RCTs were presented. While the Psoriasis Area and Severity Index (PASI) is the current standard for measuring psoriasis severity and response to therapy in RCTs, the PASI is challenging for use in clinical practice. Joseph Merola, MD, presented data to support the use of the psoriasis Physician Global Assessment (PGA) x body surface area (BSA), a simpler measure than the PASI, as a potential substitute for the PASI. Updates from the International Dermatology Outcome Measures board included reporting of the psoriasis core domain set and a summary of the PsA symptoms working group’s efforts to identify patient-reported outcomes to identify and measure PsA among patients enrolled in psoriasis RCTs. April Armstrong, MD, discussed the National Psoriasis Foundation’s efforts to develop treat-to-target goals for psoriasis. Finally, Laura Coates, MBChB, PhD, presented a report on her efforts to examine a variety of psoriasis cut points for minimal disease activity and very low disease activity outcomes.
 

Dr. Ogdie is director of the Penn Psoriatic Arthritis Clinic at the University of Pennsylvania, Philadelphia, and is a member of the GRAPPA Steering Committee.

The agenda for this year’s Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) annual meeting included sessions on juvenile psoriatic arthritis (PsA), the microbiome in psoriasis and PsA, and setting up longitudinal cohort studies. There was also a workshop to define the clinical fit and feasibility of PsA outcome measures for clinical trials and updates on several GRAPPA-associated projects.

The meeting, held July 13-15 in Amsterdam, opened with the annual trainee session. This year’s trainee session attracted more than 40 abstracts, of which 6 were selected for oral presentations on a variety of topics including outcome measure assessment, imaging, the microbiome in PsA, and T-cell subsets in synovial fluid.

Among the most intriguing sessions at the GRAPPA meeting was a series of three talks by Jose Scher, MD, Hok Bing Thio, MD, PhD, and Dirk Elewaut, PhD, that introduced the audience to the complexity of the micro-organisms that call the human host “home” and the potential role in the development of inflammatory conditions, in particular psoriasis and PsA. These talks touched on the potential uses of the microbiome of the gut, skin, and oral mucosa in predicting therapy response and modulating the immune system.

Dafna Gladman, MD, led a session on “how to set up a cohort.” In this session, speakers provided a road map for setting up a longitudinal cohort and discussed opportunities and challenges along the way. This session provided a foundation for a meeting that followed the annual meeting, the GRAPPA Research Collaborative Network meeting (held July 15-16). The RCN meeting aimed to develop a plan for a GRAPPA research network supporting collaborative research to identify biomarkers and outcomes in psoriasis and PsA. Speakers/panelists from academics and industry kicked off the meeting with a discussion of prior experiences in establishing international cohort studies. Subsequent sessions presented individual aspects of beginning a longitudinal cohort study for biomarkers, including methods for sample collection, regulatory processes, and data collected at potential sites; capturing and harmonizing clinical data for such studies; and policies for publication of RCN studies.

This year’s workshop was focused on the GRAPPA-Outcome Measures in Rheumatology (OMERACT) working group’s plan to develop a Core Outcome Measure Set, a collection of outcome measurement instruments to be used in randomized, controlled trials (RCTs) for PsA. During the plenary session, the team presented a process for the group to evaluate instruments, including assessment of match to the domain or concept of interest, feasibility, construct validity, and discrimination (including reliability and the ability to distinguish between two groups such as responders and nonresponders). Breakout groups then discussed one of the six tools being considered and voted on match to the domain of interest and feasibility for RCTs.

In a skin session, ongoing efforts to standardize and simplify the measurement of skin psoriasis in both the clinic and RCTs were presented. While the Psoriasis Area and Severity Index (PASI) is the current standard for measuring psoriasis severity and response to therapy in RCTs, the PASI is challenging for use in clinical practice. Joseph Merola, MD, presented data to support the use of the psoriasis Physician Global Assessment (PGA) x body surface area (BSA), a simpler measure than the PASI, as a potential substitute for the PASI. Updates from the International Dermatology Outcome Measures board included reporting of the psoriasis core domain set and a summary of the PsA symptoms working group’s efforts to identify patient-reported outcomes to identify and measure PsA among patients enrolled in psoriasis RCTs. April Armstrong, MD, discussed the National Psoriasis Foundation’s efforts to develop treat-to-target goals for psoriasis. Finally, Laura Coates, MBChB, PhD, presented a report on her efforts to examine a variety of psoriasis cut points for minimal disease activity and very low disease activity outcomes.
 

Dr. Ogdie is director of the Penn Psoriatic Arthritis Clinic at the University of Pennsylvania, Philadelphia, and is a member of the GRAPPA Steering Committee.

Editor’s Note: This column is a quarterly feature written by members of the Physicians in Training Committee. It aims to encourage and educate students, residents, and early career hospitalists.

One of the biggest challenges early career hospitalists, residents and medical students face in launching their first quality improvement (QI) project is knowing how and where to get started. QI can be highly rewarding, but it can also take valuable time and resources without guarantees of sustainable improvement. In this article, we outline 10 key factors to consider when starting a new project.
 

1. Frame your project so that it aligns with your hospital’s current goals

Choose a project with your hospital’s goals in mind. Securing resources such as health IT, financial, or staffing support will prove difficult unless you get buy-in from hospital leadership. If your project does not directly address hospital goals, frame the purpose to demonstrate that it still fits with leadership priorities. For example, though improving handoffs from daytime to nighttime providers may not be a specific goal, leadership should appreciate that this project is expected to improve patient safety.

2. Be SMART about goals

Many QI projects fail because the scope of the initial project is too large, unrealistic, or vague. Creating a clear and focused aim statement and keeping it “SMART” (Specific, Measurable, Achievable, Realistic, and Timely) will bring structure to the project.¹ “We will reduce Congestive Heart Failure readmissions on 5 medicine units at our hospital by 2.5% in 6 months” is an example of a SMART aim statement.

3. Involve the right people from the start

QI project disasters often start with the wrong team. Select members based on who is needed and not who is available. It is critical to include representatives or “champions” from each area that will be affected. People will buy into a new methodology much more quickly if they were engaged in its development or know that respected members in their area were involved.

4. Use a simple, systematic approach to guide improvement work

Various QI models exist and each offers a systematic approach for assessing and improving care services. The Model for Improvement developed by the Associates in Process Improvement² is a simple and powerful framework for quality improvement that asks three questions: (1) What are we trying to accomplish? (2) How will we know a change is an improvement? (3) What changes can we make that will result in improvement? The model incorporates Plan-Do-Study-Act (PDSA) cycles to test changes on a small scale.

5. Good projects start with good background data

6. Choose interventions that are high impact, low effort

People will more easily change if the change itself is easy. So consider the question “does this intervention add significant work?” The best interventions change a process without causing undue burden to the clinicians and staff involved.

7. If you can’t measure it, you can’t improve it

After implementation, collect enough data to know whether the changes made improved the process. Study outcome, process, and balancing measures. If possible, use data already being collected by your institution. While it is critical to have quantitative measures, qualitative data such as surveys and observations can also enrich understanding.

Example: Increasing early discharges in medical unit.

Outcome measure – This is the desired outcome that the project aims to improve. This may be the percentage of discharges before noon (DBN) or the average discharge time.

Process measure – This is a measure of a specific change made to improve the outcome metric. The discharge orders may need to be placed earlier in the electronic medical record to improve DBN. This average discharge order time is an example of a process measure.

Balance measure – This metric evaluates whether the intended outcome is leading to unintended consequences. For example, tracking the readmission rate is an important balance measure to assess whether improved DBN is associated with rushed discharges and possible unsafe transitions.

8. Communicate project goals and progress

9. Manage resistance to change

“People responsible for planning and implementing change often forget that while the first task of change management is to understand the destination and how to get there, the first task of transition management is to convince people to leave home.” – William Bridges

Inertia is powerful. We may consider our continuous performance improvement initiative as “the next big thing” but others may not share this enthusiasm. We therefore need to build a compelling reason for others to become engaged and accept major changes to work flow. Different strategies may be needed depending on your audience. Though for some, data and a rational analysis will be persuasive, for others the emotional argument will be the most motivating. Share personal anecdotes and use patient stories. In addition, let providers know “what’s in it for them.” Some may have a personal interest in your project or may need QI experience for career advancement; others might be motivated by the possibilities for scholarship arising from this work.

10. Make the work count twice

Consider QI as a scholarly initiative from the start to bring rigor to the project at all phases. Describe the project in an abstract or manuscript once improvements have been made. Publication is a great way to boost team morale and help make a business case for future improvement work. The Standards for Quality Improvement Reporting Excellence (SQUIRE) guidelines provide an excellent framework for designing and writing up an improvement project.⁴ The guidelines focus on why the project was started, what was done, what was found, and what the findings mean.

Driving change is challenging, and it is tempting to jump ahead to “fixing the problem.” But implementing a successful QI project requires intelligent direction, strategic planning, and skillful execution. It is our hope that following the above tips will help you develop the best possible ideas and approach implementation in a systematic way, ultimately leading to meaningful change.
 

Dr. Reyna is assistant professor in the division of hospital medicine and unit medical director at Mount Sinai Medical Center in New York City. She is a Certified Clinical Microsystems Coach. Dr. Burger is associate professor and associate program director, internal medicine residency, at Mount Sinai Beth Israel. He is on the faculty for the SGIM Annual Meeting Precourse on QI and is head of the high value care committee at the department of medicine at Mount Sinai Beth Israel. Dr. Cho is assistant professor and director of quality and safety in the division of hospital medicine at Mount Sinai. He is a senior fellow at the Lown Institute.

References

1. MacLeod L. Making SMART goals smarter. Physician Exec. 2012 Mar-Apr;38(2):68-70, 72.

2. Langley GL, Moen R, Nolan KM, Nolan TW, Norman CL, Provost LP. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance (2nd edition). San Francisco: Jossey-Bass Publishers; 2009.

3. Nelson EC, Batalden PB, Godfrey MM. Quality By Design: A Clinical Microsystems Approach. San Francisco, California: Jossey-Bass; 2007.

4. Ogrinc G, Davies L, Goodman D et.al. SQUIRE 2.0 (Standards for Quality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2015 Sep 14.

Editor’s Note: This column is a quarterly feature written by members of the Physicians in Training Committee. It aims to encourage and educate students, residents, and early career hospitalists.

One of the biggest challenges early career hospitalists, residents and medical students face in launching their first quality improvement (QI) project is knowing how and where to get started. QI can be highly rewarding, but it can also take valuable time and resources without guarantees of sustainable improvement. In this article, we outline 10 key factors to consider when starting a new project.
 

1. Frame your project so that it aligns with your hospital’s current goals

Choose a project with your hospital’s goals in mind. Securing resources such as health IT, financial, or staffing support will prove difficult unless you get buy-in from hospital leadership. If your project does not directly address hospital goals, frame the purpose to demonstrate that it still fits with leadership priorities. For example, though improving handoffs from daytime to nighttime providers may not be a specific goal, leadership should appreciate that this project is expected to improve patient safety.

2. Be SMART about goals

Many QI projects fail because the scope of the initial project is too large, unrealistic, or vague. Creating a clear and focused aim statement and keeping it “SMART” (Specific, Measurable, Achievable, Realistic, and Timely) will bring structure to the project.¹ “We will reduce Congestive Heart Failure readmissions on 5 medicine units at our hospital by 2.5% in 6 months” is an example of a SMART aim statement.

3. Involve the right people from the start

QI project disasters often start with the wrong team. Select members based on who is needed and not who is available. It is critical to include representatives or “champions” from each area that will be affected. People will buy into a new methodology much more quickly if they were engaged in its development or know that respected members in their area were involved.

4. Use a simple, systematic approach to guide improvement work

Various QI models exist and each offers a systematic approach for assessing and improving care services. The Model for Improvement developed by the Associates in Process Improvement² is a simple and powerful framework for quality improvement that asks three questions: (1) What are we trying to accomplish? (2) How will we know a change is an improvement? (3) What changes can we make that will result in improvement? The model incorporates Plan-Do-Study-Act (PDSA) cycles to test changes on a small scale.

5. Good projects start with good background data

6. Choose interventions that are high impact, low effort

People will more easily change if the change itself is easy. So consider the question “does this intervention add significant work?” The best interventions change a process without causing undue burden to the clinicians and staff involved.

7. If you can’t measure it, you can’t improve it

After implementation, collect enough data to know whether the changes made improved the process. Study outcome, process, and balancing measures. If possible, use data already being collected by your institution. While it is critical to have quantitative measures, qualitative data such as surveys and observations can also enrich understanding.

Example: Increasing early discharges in medical unit.

Outcome measure – This is the desired outcome that the project aims to improve. This may be the percentage of discharges before noon (DBN) or the average discharge time.

Process measure – This is a measure of a specific change made to improve the outcome metric. The discharge orders may need to be placed earlier in the electronic medical record to improve DBN. This average discharge order time is an example of a process measure.

Balance measure – This metric evaluates whether the intended outcome is leading to unintended consequences. For example, tracking the readmission rate is an important balance measure to assess whether improved DBN is associated with rushed discharges and possible unsafe transitions.

8. Communicate project goals and progress

9. Manage resistance to change

“People responsible for planning and implementing change often forget that while the first task of change management is to understand the destination and how to get there, the first task of transition management is to convince people to leave home.” – William Bridges

Inertia is powerful. We may consider our continuous performance improvement initiative as “the next big thing” but others may not share this enthusiasm. We therefore need to build a compelling reason for others to become engaged and accept major changes to work flow. Different strategies may be needed depending on your audience. Though for some, data and a rational analysis will be persuasive, for others the emotional argument will be the most motivating. Share personal anecdotes and use patient stories. In addition, let providers know “what’s in it for them.” Some may have a personal interest in your project or may need QI experience for career advancement; others might be motivated by the possibilities for scholarship arising from this work.

10. Make the work count twice

Consider QI as a scholarly initiative from the start to bring rigor to the project at all phases. Describe the project in an abstract or manuscript once improvements have been made. Publication is a great way to boost team morale and help make a business case for future improvement work. The Standards for Quality Improvement Reporting Excellence (SQUIRE) guidelines provide an excellent framework for designing and writing up an improvement project.⁴ The guidelines focus on why the project was started, what was done, what was found, and what the findings mean.

Driving change is challenging, and it is tempting to jump ahead to “fixing the problem.” But implementing a successful QI project requires intelligent direction, strategic planning, and skillful execution. It is our hope that following the above tips will help you develop the best possible ideas and approach implementation in a systematic way, ultimately leading to meaningful change.
 

Dr. Reyna is assistant professor in the division of hospital medicine and unit medical director at Mount Sinai Medical Center in New York City. She is a Certified Clinical Microsystems Coach. Dr. Burger is associate professor and associate program director, internal medicine residency, at Mount Sinai Beth Israel. He is on the faculty for the SGIM Annual Meeting Precourse on QI and is head of the high value care committee at the department of medicine at Mount Sinai Beth Israel. Dr. Cho is assistant professor and director of quality and safety in the division of hospital medicine at Mount Sinai. He is a senior fellow at the Lown Institute.

References

1. MacLeod L. Making SMART goals smarter. Physician Exec. 2012 Mar-Apr;38(2):68-70, 72.

2. Langley GL, Moen R, Nolan KM, Nolan TW, Norman CL, Provost LP. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance (2nd edition). San Francisco: Jossey-Bass Publishers; 2009.

3. Nelson EC, Batalden PB, Godfrey MM. Quality By Design: A Clinical Microsystems Approach. San Francisco, California: Jossey-Bass; 2007.

4. Ogrinc G, Davies L, Goodman D et.al. SQUIRE 2.0 (Standards for Quality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2015 Sep 14.

Editor’s Note: This column is a quarterly feature written by members of the Physicians in Training Committee. It aims to encourage and educate students, residents, and early career hospitalists.

One of the biggest challenges early career hospitalists, residents and medical students face in launching their first quality improvement (QI) project is knowing how and where to get started. QI can be highly rewarding, but it can also take valuable time and resources without guarantees of sustainable improvement. In this article, we outline 10 key factors to consider when starting a new project.
 

1. Frame your project so that it aligns with your hospital’s current goals

Choose a project with your hospital’s goals in mind. Securing resources such as health IT, financial, or staffing support will prove difficult unless you get buy-in from hospital leadership. If your project does not directly address hospital goals, frame the purpose to demonstrate that it still fits with leadership priorities. For example, though improving handoffs from daytime to nighttime providers may not be a specific goal, leadership should appreciate that this project is expected to improve patient safety.

2. Be SMART about goals

Many QI projects fail because the scope of the initial project is too large, unrealistic, or vague. Creating a clear and focused aim statement and keeping it “SMART” (Specific, Measurable, Achievable, Realistic, and Timely) will bring structure to the project.¹ “We will reduce Congestive Heart Failure readmissions on 5 medicine units at our hospital by 2.5% in 6 months” is an example of a SMART aim statement.

3. Involve the right people from the start

QI project disasters often start with the wrong team. Select members based on who is needed and not who is available. It is critical to include representatives or “champions” from each area that will be affected. People will buy into a new methodology much more quickly if they were engaged in its development or know that respected members in their area were involved.

4. Use a simple, systematic approach to guide improvement work

Various QI models exist and each offers a systematic approach for assessing and improving care services. The Model for Improvement developed by the Associates in Process Improvement² is a simple and powerful framework for quality improvement that asks three questions: (1) What are we trying to accomplish? (2) How will we know a change is an improvement? (3) What changes can we make that will result in improvement? The model incorporates Plan-Do-Study-Act (PDSA) cycles to test changes on a small scale.

5. Good projects start with good background data

6. Choose interventions that are high impact, low effort

People will more easily change if the change itself is easy. So consider the question “does this intervention add significant work?” The best interventions change a process without causing undue burden to the clinicians and staff involved.

7. If you can’t measure it, you can’t improve it

After implementation, collect enough data to know whether the changes made improved the process. Study outcome, process, and balancing measures. If possible, use data already being collected by your institution. While it is critical to have quantitative measures, qualitative data such as surveys and observations can also enrich understanding.

Example: Increasing early discharges in medical unit.

Outcome measure – This is the desired outcome that the project aims to improve. This may be the percentage of discharges before noon (DBN) or the average discharge time.

Process measure – This is a measure of a specific change made to improve the outcome metric. The discharge orders may need to be placed earlier in the electronic medical record to improve DBN. This average discharge order time is an example of a process measure.

Balance measure – This metric evaluates whether the intended outcome is leading to unintended consequences. For example, tracking the readmission rate is an important balance measure to assess whether improved DBN is associated with rushed discharges and possible unsafe transitions.

8. Communicate project goals and progress

9. Manage resistance to change

“People responsible for planning and implementing change often forget that while the first task of change management is to understand the destination and how to get there, the first task of transition management is to convince people to leave home.” – William Bridges

Inertia is powerful. We may consider our continuous performance improvement initiative as “the next big thing” but others may not share this enthusiasm. We therefore need to build a compelling reason for others to become engaged and accept major changes to work flow. Different strategies may be needed depending on your audience. Though for some, data and a rational analysis will be persuasive, for others the emotional argument will be the most motivating. Share personal anecdotes and use patient stories. In addition, let providers know “what’s in it for them.” Some may have a personal interest in your project or may need QI experience for career advancement; others might be motivated by the possibilities for scholarship arising from this work.

10. Make the work count twice

Consider QI as a scholarly initiative from the start to bring rigor to the project at all phases. Describe the project in an abstract or manuscript once improvements have been made. Publication is a great way to boost team morale and help make a business case for future improvement work. The Standards for Quality Improvement Reporting Excellence (SQUIRE) guidelines provide an excellent framework for designing and writing up an improvement project.⁴ The guidelines focus on why the project was started, what was done, what was found, and what the findings mean.

Driving change is challenging, and it is tempting to jump ahead to “fixing the problem.” But implementing a successful QI project requires intelligent direction, strategic planning, and skillful execution. It is our hope that following the above tips will help you develop the best possible ideas and approach implementation in a systematic way, ultimately leading to meaningful change.
 

Dr. Reyna is assistant professor in the division of hospital medicine and unit medical director at Mount Sinai Medical Center in New York City. She is a Certified Clinical Microsystems Coach. Dr. Burger is associate professor and associate program director, internal medicine residency, at Mount Sinai Beth Israel. He is on the faculty for the SGIM Annual Meeting Precourse on QI and is head of the high value care committee at the department of medicine at Mount Sinai Beth Israel. Dr. Cho is assistant professor and director of quality and safety in the division of hospital medicine at Mount Sinai. He is a senior fellow at the Lown Institute.

References

1. MacLeod L. Making SMART goals smarter. Physician Exec. 2012 Mar-Apr;38(2):68-70, 72.

2. Langley GL, Moen R, Nolan KM, Nolan TW, Norman CL, Provost LP. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance (2nd edition). San Francisco: Jossey-Bass Publishers; 2009.

3. Nelson EC, Batalden PB, Godfrey MM. Quality By Design: A Clinical Microsystems Approach. San Francisco, California: Jossey-Bass; 2007.

4. Ogrinc G, Davies L, Goodman D et.al. SQUIRE 2.0 (Standards for Quality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2015 Sep 14.

NEW YORK – In a plenary session at the American Academy of Dermatology summer meeting, the AAD president offered an upbeat view of the profession, likening his role in leading the 19,000-member organization to that of a dancer and comparing the specialty itself to “a bright star on the dance floor.”

The specialty, however, is facing an uncertain future. “As the music changes, so must the dance,” Henry Lim, MD, told attendees. “And so it is with American medicine today. Successfully transitioning, adapting to those changes, is especially challenging for all of medicine, including for our specialty.”

NEW YORK – In a plenary session at the American Academy of Dermatology summer meeting, the AAD president offered an upbeat view of the profession, likening his role in leading the 19,000-member organization to that of a dancer and comparing the specialty itself to “a bright star on the dance floor.”

The specialty, however, is facing an uncertain future. “As the music changes, so must the dance,” Henry Lim, MD, told attendees. “And so it is with American medicine today. Successfully transitioning, adapting to those changes, is especially challenging for all of medicine, including for our specialty.”

NEW YORK – In a plenary session at the American Academy of Dermatology summer meeting, the AAD president offered an upbeat view of the profession, likening his role in leading the 19,000-member organization to that of a dancer and comparing the specialty itself to “a bright star on the dance floor.”

The specialty, however, is facing an uncertain future. “As the music changes, so must the dance,” Henry Lim, MD, told attendees. “And so it is with American medicine today. Successfully transitioning, adapting to those changes, is especially challenging for all of medicine, including for our specialty.”