Andrew L. Masica, MD, MSCI

The Centers for Medicare & Medicaid Services (CMS) publicly reports all‐cause risk‐standardized readmission rates after acute‐care hospitalization for acute myocardial infarction, pneumonia, heart failure, total hip and knee arthroplasty, chronic obstructive pulmonary disease, stroke, and for patients hospital‐wide.[1, 2, 3, 4, 5] Ideally, these measures should capture unplanned readmissions that arise from acute clinical events requiring urgent rehospitalization. Planned readmissions, which are scheduled admissions usually involving nonurgent procedures, may not be a signal of quality of care. Including planned readmissions in readmission quality measures could create a disincentive to provide appropriate care to patients who are scheduled for elective or necessary procedures unrelated to the quality of the prior admission. Accordingly, under contract to the CMS, we were asked to develop an algorithm to identify planned readmissions. A version of this algorithm is now incorporated into all publicly reported readmission measures.

Given the widespread use of the planned readmission algorithm in public reporting and its implications for hospital quality measurement and evaluation, the objective of this study was to describe the development process, and to validate and refine the algorithm by reviewing charts of readmitted patients.

METHODS

Algorithm Development

To create a planned readmission algorithm, we first defined planned. We determined that readmissions for obstetrical delivery, maintenance chemotherapy, major organ transplant, and rehabilitation should always be considered planned in the sense that they are desired and/or inevitable, even if not specifically planned on a certain date. Apart from these specific types of readmissions, we defined planned readmissions as nonacute readmissions for scheduled procedures, because the vast majority of planned admissions are related to procedures. We also defined readmissions for acute illness or for complications of care as unplanned for the purposes of a quality measure. Even if such readmissions included a potentially planned procedure, because complications of care represent an important dimension of quality that should not be excluded from outcome measurement, these admissions should not be removed from the measure outcome. This definition of planned readmissions does not imply that all unplanned readmissions are unexpected or avoidable. However, it has proven very difficult to reliably define avoidable readmissions, even by expert review of charts, and we did not attempt to do so here.[6, 7]

In the second stage, we operationalized this definition into an algorithm. We used the Agency for Healthcare Research and Quality's Clinical Classification Software (CCS) codes to group thousands of individual procedure and diagnosis International Classification of Disease, Ninth Revision, Clinical Modification (ICD‐9‐CM) codes into clinically coherent, mutually exclusive procedure CCS categories and mutually exclusive diagnosis CCS categories, respectively. Clinicians on the investigative team reviewed the procedure categories to identify those that are commonly planned and that would require inpatient admission. We also reviewed the diagnosis categories to identify acute diagnoses unlikely to accompany elective procedures. We then created a flow diagram through which every readmission could be run to determine whether it was planned or unplanned based on our categorizations of procedures and diagnoses (Figure 1, and Supporting Information, Appendix A, in the online version of this article). This version of the algorithm (v1.0) was submitted to the National Quality Forum (NQF) as part of the hospital‐wide readmission measure. The measure (NQR #1789) received endorsement in April 2012.

Figure 1

In the third stage of development, we posted the algorithm for 2 public comment periods and recruited 27 outside experts to review and refine the algorithm following a standardized, structured process (see Supporting Information, Appendix B, in the online version of this article). Because the measures publicly report and hold hospitals accountable for unplanned readmission rates, we felt it most important that the algorithm include as few planned readmissions in the reported, unplanned outcome as possible (ie, have high negative predictive value). Therefore, in equivocal situations in which experts felt procedure categories were equally often planned or unplanned, we added those procedures to the potentially planned list. We also solicited feedback from hospitals on algorithm performance during a confidential test run of the hospital‐wide readmission measure in the fall of 2012. Based on all of this feedback, we made a number of changes to the algorithm, which was then identified as v2.1. Version 2.1 of the algorithm was submitted to the NQF as part of the endorsement process for the acute myocardial infarction and heart failure readmission measures and was endorsed by the NQF in January 2013. The algorithm (v2.1) is now applied, adapted if necessary, to all publicly reported readmission measures.[8]

RESULTS

DISCUSSION

We developed an algorithm based on administrative data that in its currently implemented form is very accurate at identifying unplanned readmissions, ensuring that readmissions included in publicly reported readmission measures are likely to be truly unplanned. However, nearly half of readmissions the algorithm classifies as planned are actually unplanned. That is, the algorithm is overcautious in excluding unplanned readmissions that could have counted as outcomes, particularly among admissions that include diagnostic cardiac catheterization or placement of cardiac devices (pacemakers, defibrillators). However, these errors only occur within the 7.8% of readmissions that are classified as planned and therefore do not affect overall readmission rates dramatically. A perfect algorithm would reclassify approximately half of these planned readmissions as unplanned, increasing the overall readmission rate by 0.6 percentage points.

On the other hand, the algorithm also only identifies approximately half of true planned readmissions as planned. Because the true prevalence of planned readmissions is low (approximately 9% of readmissions based on weighted chart review prevalence, or an absolute rate of 1.4%), this low sensitivity has a small effect on algorithm performance. Removing all true planned readmissions from the measure outcome would decrease the overall readmission rate by 0.8 percentage points, similar to the expected 0.6 percentage point increase that would result from better identifying unplanned readmissions; thus, a perfect algorithm would likely decrease the reported unplanned readmission rate by a net 0.2%. Overall, the existing algorithm appears to come close to the true prevalence of planned readmissions, despite inaccuracy on an individual‐case basis. The algorithm performed best at large hospitals, which are at greatest risk of being statistical outliers and of accruing penalties under the Hospital Readmissions Reduction Program.[10]

We identified several changes that marginally improved the performance of the algorithm by reducing the number of unplanned readmissions that are incorrectly removed from the measure, while avoiding the inappropriate inclusion of planned readmissions in the outcome. This revised algorithm, v3.0, was applied to public reporting of readmission rates at the end of 2014. Overall, implementing these changes increases the reported readmission rate very slightly. We also identified other procedures associated with high inaccuracy rates, removal of which would have larger impact on reporting rates, and which therefore merit further evaluation.

There are other potential methods of identifying planned readmissions. For instance, as of October 1, 2013, new administrative billing codes were created to allow hospitals to indicate that a patient was discharged with a planned acute‐care hospital inpatient readmission, without limitation as to when it will take place.[11] This code must be used at the time of the index admission to indicate that a future planned admission is expected, and was specified only to be used for neonates and patients with acute myocardial infarction. This approach, however, would omit planned readmissions that are not known to the initial discharging team, potentially missing planned readmissions. Conversely, some patients discharged with a plan for readmission may be unexpectedly readmitted for an unplanned reason. Given that the new codes were not available at the time we conducted the validation study, we were not able to determine how often the billing codes accurately identified planned readmissions. This would be an important area to consider for future study.

An alternative approach would be to create indicator codes to be applied at the time of readmission that would indicate whether that admission was planned or unplanned. Such a code would have the advantage of allowing each planned readmission to be flagged by the admitting clinicians at the time of admission rather than by an algorithm that inherently cannot be perfect. However, identifying planned readmissions at the time of readmission would also create opportunity for gaming and inconsistent application of definitions between hospitals; additional checks would need to be put in place to guard against these possibilities.

Our study has some limitations. We relied on the opinion of chart abstractors to determine whether a readmission was planned or unplanned; in a few cases, such as smoldering wounds that ultimately require surgical intervention, that determination is debatable. Abstractions were done at local institutions to minimize risks to patient privacy, and therefore we could not centrally verify determinations of planned status except by reviewing source of admission, dates of procedures, and narrative comments reported by the abstractors. Finally, we did not have sufficient volume of planned procedures to determine accuracy of the algorithm for less common procedure categories or individual procedures within categories.

In summary, we developed an algorithm to identify planned readmissions from administrative data that had high specificity and moderate sensitivity, and refined it based on chart validation. This algorithm is in use in public reporting of readmission measures to maximize the probability that the reported readmission rates represent truly unplanned readmissions.[12]

Disclosures: Financial supportThis work was performed under contract HHSM‐500‐2008‐0025I/HHSM‐500‐T0001, Modification No. 000008, titled Measure Instrument Development and Support, funded by the Centers for Medicare and Medicaid Services (CMS), an agency of the US Department of Health and Human Services. Drs. Horwitz and Ross are supported by the National Institute on Aging (K08 AG038336 and K08 AG032886, respectively) and by the American Federation for Aging Research through the Paul B. Beeson Career Development Award Program. Dr. Krumholz is supported by grant U01 HL105270‐05 (Center for Cardiovascular Outcomes Research at Yale University) from the National Heart, Lung, and Blood Institute. No funding source had any role in the study design; in the collection, analysis, and interpretation of data; or in the writing of the article. The CMS reviewed and approved the use of its data for this work and approved submission of the manuscript. All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that all authors have support from the CMS for the submitted work. In addition, Dr. Ross is a member of a scientific advisory board for FAIR Health Inc. Dr. Krumholz chairs a cardiac scientific advisory board for UnitedHealth and is the recipient of research agreements from Medtronic and Johnson & Johnson through Yale University, to develop methods of clinical trial data sharing. All other authors report no conflicts of interest.

The Centers for Medicare & Medicaid Services (CMS) publicly reports all‐cause risk‐standardized readmission rates after acute‐care hospitalization for acute myocardial infarction, pneumonia, heart failure, total hip and knee arthroplasty, chronic obstructive pulmonary disease, stroke, and for patients hospital‐wide.[1, 2, 3, 4, 5] Ideally, these measures should capture unplanned readmissions that arise from acute clinical events requiring urgent rehospitalization. Planned readmissions, which are scheduled admissions usually involving nonurgent procedures, may not be a signal of quality of care. Including planned readmissions in readmission quality measures could create a disincentive to provide appropriate care to patients who are scheduled for elective or necessary procedures unrelated to the quality of the prior admission. Accordingly, under contract to the CMS, we were asked to develop an algorithm to identify planned readmissions. A version of this algorithm is now incorporated into all publicly reported readmission measures.

Given the widespread use of the planned readmission algorithm in public reporting and its implications for hospital quality measurement and evaluation, the objective of this study was to describe the development process, and to validate and refine the algorithm by reviewing charts of readmitted patients.

METHODS

Algorithm Development

To create a planned readmission algorithm, we first defined planned. We determined that readmissions for obstetrical delivery, maintenance chemotherapy, major organ transplant, and rehabilitation should always be considered planned in the sense that they are desired and/or inevitable, even if not specifically planned on a certain date. Apart from these specific types of readmissions, we defined planned readmissions as nonacute readmissions for scheduled procedures, because the vast majority of planned admissions are related to procedures. We also defined readmissions for acute illness or for complications of care as unplanned for the purposes of a quality measure. Even if such readmissions included a potentially planned procedure, because complications of care represent an important dimension of quality that should not be excluded from outcome measurement, these admissions should not be removed from the measure outcome. This definition of planned readmissions does not imply that all unplanned readmissions are unexpected or avoidable. However, it has proven very difficult to reliably define avoidable readmissions, even by expert review of charts, and we did not attempt to do so here.[6, 7]

In the second stage, we operationalized this definition into an algorithm. We used the Agency for Healthcare Research and Quality's Clinical Classification Software (CCS) codes to group thousands of individual procedure and diagnosis International Classification of Disease, Ninth Revision, Clinical Modification (ICD‐9‐CM) codes into clinically coherent, mutually exclusive procedure CCS categories and mutually exclusive diagnosis CCS categories, respectively. Clinicians on the investigative team reviewed the procedure categories to identify those that are commonly planned and that would require inpatient admission. We also reviewed the diagnosis categories to identify acute diagnoses unlikely to accompany elective procedures. We then created a flow diagram through which every readmission could be run to determine whether it was planned or unplanned based on our categorizations of procedures and diagnoses (Figure 1, and Supporting Information, Appendix A, in the online version of this article). This version of the algorithm (v1.0) was submitted to the National Quality Forum (NQF) as part of the hospital‐wide readmission measure. The measure (NQR #1789) received endorsement in April 2012.

Figure 1

In the third stage of development, we posted the algorithm for 2 public comment periods and recruited 27 outside experts to review and refine the algorithm following a standardized, structured process (see Supporting Information, Appendix B, in the online version of this article). Because the measures publicly report and hold hospitals accountable for unplanned readmission rates, we felt it most important that the algorithm include as few planned readmissions in the reported, unplanned outcome as possible (ie, have high negative predictive value). Therefore, in equivocal situations in which experts felt procedure categories were equally often planned or unplanned, we added those procedures to the potentially planned list. We also solicited feedback from hospitals on algorithm performance during a confidential test run of the hospital‐wide readmission measure in the fall of 2012. Based on all of this feedback, we made a number of changes to the algorithm, which was then identified as v2.1. Version 2.1 of the algorithm was submitted to the NQF as part of the endorsement process for the acute myocardial infarction and heart failure readmission measures and was endorsed by the NQF in January 2013. The algorithm (v2.1) is now applied, adapted if necessary, to all publicly reported readmission measures.[8]

RESULTS

DISCUSSION

We developed an algorithm based on administrative data that in its currently implemented form is very accurate at identifying unplanned readmissions, ensuring that readmissions included in publicly reported readmission measures are likely to be truly unplanned. However, nearly half of readmissions the algorithm classifies as planned are actually unplanned. That is, the algorithm is overcautious in excluding unplanned readmissions that could have counted as outcomes, particularly among admissions that include diagnostic cardiac catheterization or placement of cardiac devices (pacemakers, defibrillators). However, these errors only occur within the 7.8% of readmissions that are classified as planned and therefore do not affect overall readmission rates dramatically. A perfect algorithm would reclassify approximately half of these planned readmissions as unplanned, increasing the overall readmission rate by 0.6 percentage points.

On the other hand, the algorithm also only identifies approximately half of true planned readmissions as planned. Because the true prevalence of planned readmissions is low (approximately 9% of readmissions based on weighted chart review prevalence, or an absolute rate of 1.4%), this low sensitivity has a small effect on algorithm performance. Removing all true planned readmissions from the measure outcome would decrease the overall readmission rate by 0.8 percentage points, similar to the expected 0.6 percentage point increase that would result from better identifying unplanned readmissions; thus, a perfect algorithm would likely decrease the reported unplanned readmission rate by a net 0.2%. Overall, the existing algorithm appears to come close to the true prevalence of planned readmissions, despite inaccuracy on an individual‐case basis. The algorithm performed best at large hospitals, which are at greatest risk of being statistical outliers and of accruing penalties under the Hospital Readmissions Reduction Program.[10]

We identified several changes that marginally improved the performance of the algorithm by reducing the number of unplanned readmissions that are incorrectly removed from the measure, while avoiding the inappropriate inclusion of planned readmissions in the outcome. This revised algorithm, v3.0, was applied to public reporting of readmission rates at the end of 2014. Overall, implementing these changes increases the reported readmission rate very slightly. We also identified other procedures associated with high inaccuracy rates, removal of which would have larger impact on reporting rates, and which therefore merit further evaluation.

There are other potential methods of identifying planned readmissions. For instance, as of October 1, 2013, new administrative billing codes were created to allow hospitals to indicate that a patient was discharged with a planned acute‐care hospital inpatient readmission, without limitation as to when it will take place.[11] This code must be used at the time of the index admission to indicate that a future planned admission is expected, and was specified only to be used for neonates and patients with acute myocardial infarction. This approach, however, would omit planned readmissions that are not known to the initial discharging team, potentially missing planned readmissions. Conversely, some patients discharged with a plan for readmission may be unexpectedly readmitted for an unplanned reason. Given that the new codes were not available at the time we conducted the validation study, we were not able to determine how often the billing codes accurately identified planned readmissions. This would be an important area to consider for future study.

An alternative approach would be to create indicator codes to be applied at the time of readmission that would indicate whether that admission was planned or unplanned. Such a code would have the advantage of allowing each planned readmission to be flagged by the admitting clinicians at the time of admission rather than by an algorithm that inherently cannot be perfect. However, identifying planned readmissions at the time of readmission would also create opportunity for gaming and inconsistent application of definitions between hospitals; additional checks would need to be put in place to guard against these possibilities.

Our study has some limitations. We relied on the opinion of chart abstractors to determine whether a readmission was planned or unplanned; in a few cases, such as smoldering wounds that ultimately require surgical intervention, that determination is debatable. Abstractions were done at local institutions to minimize risks to patient privacy, and therefore we could not centrally verify determinations of planned status except by reviewing source of admission, dates of procedures, and narrative comments reported by the abstractors. Finally, we did not have sufficient volume of planned procedures to determine accuracy of the algorithm for less common procedure categories or individual procedures within categories.

In summary, we developed an algorithm to identify planned readmissions from administrative data that had high specificity and moderate sensitivity, and refined it based on chart validation. This algorithm is in use in public reporting of readmission measures to maximize the probability that the reported readmission rates represent truly unplanned readmissions.[12]

Disclosures: Financial supportThis work was performed under contract HHSM‐500‐2008‐0025I/HHSM‐500‐T0001, Modification No. 000008, titled Measure Instrument Development and Support, funded by the Centers for Medicare and Medicaid Services (CMS), an agency of the US Department of Health and Human Services. Drs. Horwitz and Ross are supported by the National Institute on Aging (K08 AG038336 and K08 AG032886, respectively) and by the American Federation for Aging Research through the Paul B. Beeson Career Development Award Program. Dr. Krumholz is supported by grant U01 HL105270‐05 (Center for Cardiovascular Outcomes Research at Yale University) from the National Heart, Lung, and Blood Institute. No funding source had any role in the study design; in the collection, analysis, and interpretation of data; or in the writing of the article. The CMS reviewed and approved the use of its data for this work and approved submission of the manuscript. All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that all authors have support from the CMS for the submitted work. In addition, Dr. Ross is a member of a scientific advisory board for FAIR Health Inc. Dr. Krumholz chairs a cardiac scientific advisory board for UnitedHealth and is the recipient of research agreements from Medtronic and Johnson & Johnson through Yale University, to develop methods of clinical trial data sharing. All other authors report no conflicts of interest.

The Centers for Medicare & Medicaid Services (CMS) publicly reports all‐cause risk‐standardized readmission rates after acute‐care hospitalization for acute myocardial infarction, pneumonia, heart failure, total hip and knee arthroplasty, chronic obstructive pulmonary disease, stroke, and for patients hospital‐wide.[1, 2, 3, 4, 5] Ideally, these measures should capture unplanned readmissions that arise from acute clinical events requiring urgent rehospitalization. Planned readmissions, which are scheduled admissions usually involving nonurgent procedures, may not be a signal of quality of care. Including planned readmissions in readmission quality measures could create a disincentive to provide appropriate care to patients who are scheduled for elective or necessary procedures unrelated to the quality of the prior admission. Accordingly, under contract to the CMS, we were asked to develop an algorithm to identify planned readmissions. A version of this algorithm is now incorporated into all publicly reported readmission measures.

Given the widespread use of the planned readmission algorithm in public reporting and its implications for hospital quality measurement and evaluation, the objective of this study was to describe the development process, and to validate and refine the algorithm by reviewing charts of readmitted patients.

METHODS

Algorithm Development

To create a planned readmission algorithm, we first defined planned. We determined that readmissions for obstetrical delivery, maintenance chemotherapy, major organ transplant, and rehabilitation should always be considered planned in the sense that they are desired and/or inevitable, even if not specifically planned on a certain date. Apart from these specific types of readmissions, we defined planned readmissions as nonacute readmissions for scheduled procedures, because the vast majority of planned admissions are related to procedures. We also defined readmissions for acute illness or for complications of care as unplanned for the purposes of a quality measure. Even if such readmissions included a potentially planned procedure, because complications of care represent an important dimension of quality that should not be excluded from outcome measurement, these admissions should not be removed from the measure outcome. This definition of planned readmissions does not imply that all unplanned readmissions are unexpected or avoidable. However, it has proven very difficult to reliably define avoidable readmissions, even by expert review of charts, and we did not attempt to do so here.[6, 7]

In the second stage, we operationalized this definition into an algorithm. We used the Agency for Healthcare Research and Quality's Clinical Classification Software (CCS) codes to group thousands of individual procedure and diagnosis International Classification of Disease, Ninth Revision, Clinical Modification (ICD‐9‐CM) codes into clinically coherent, mutually exclusive procedure CCS categories and mutually exclusive diagnosis CCS categories, respectively. Clinicians on the investigative team reviewed the procedure categories to identify those that are commonly planned and that would require inpatient admission. We also reviewed the diagnosis categories to identify acute diagnoses unlikely to accompany elective procedures. We then created a flow diagram through which every readmission could be run to determine whether it was planned or unplanned based on our categorizations of procedures and diagnoses (Figure 1, and Supporting Information, Appendix A, in the online version of this article). This version of the algorithm (v1.0) was submitted to the National Quality Forum (NQF) as part of the hospital‐wide readmission measure. The measure (NQR #1789) received endorsement in April 2012.

Figure 1

In the third stage of development, we posted the algorithm for 2 public comment periods and recruited 27 outside experts to review and refine the algorithm following a standardized, structured process (see Supporting Information, Appendix B, in the online version of this article). Because the measures publicly report and hold hospitals accountable for unplanned readmission rates, we felt it most important that the algorithm include as few planned readmissions in the reported, unplanned outcome as possible (ie, have high negative predictive value). Therefore, in equivocal situations in which experts felt procedure categories were equally often planned or unplanned, we added those procedures to the potentially planned list. We also solicited feedback from hospitals on algorithm performance during a confidential test run of the hospital‐wide readmission measure in the fall of 2012. Based on all of this feedback, we made a number of changes to the algorithm, which was then identified as v2.1. Version 2.1 of the algorithm was submitted to the NQF as part of the endorsement process for the acute myocardial infarction and heart failure readmission measures and was endorsed by the NQF in January 2013. The algorithm (v2.1) is now applied, adapted if necessary, to all publicly reported readmission measures.[8]

RESULTS

DISCUSSION

We developed an algorithm based on administrative data that in its currently implemented form is very accurate at identifying unplanned readmissions, ensuring that readmissions included in publicly reported readmission measures are likely to be truly unplanned. However, nearly half of readmissions the algorithm classifies as planned are actually unplanned. That is, the algorithm is overcautious in excluding unplanned readmissions that could have counted as outcomes, particularly among admissions that include diagnostic cardiac catheterization or placement of cardiac devices (pacemakers, defibrillators). However, these errors only occur within the 7.8% of readmissions that are classified as planned and therefore do not affect overall readmission rates dramatically. A perfect algorithm would reclassify approximately half of these planned readmissions as unplanned, increasing the overall readmission rate by 0.6 percentage points.

On the other hand, the algorithm also only identifies approximately half of true planned readmissions as planned. Because the true prevalence of planned readmissions is low (approximately 9% of readmissions based on weighted chart review prevalence, or an absolute rate of 1.4%), this low sensitivity has a small effect on algorithm performance. Removing all true planned readmissions from the measure outcome would decrease the overall readmission rate by 0.8 percentage points, similar to the expected 0.6 percentage point increase that would result from better identifying unplanned readmissions; thus, a perfect algorithm would likely decrease the reported unplanned readmission rate by a net 0.2%. Overall, the existing algorithm appears to come close to the true prevalence of planned readmissions, despite inaccuracy on an individual‐case basis. The algorithm performed best at large hospitals, which are at greatest risk of being statistical outliers and of accruing penalties under the Hospital Readmissions Reduction Program.[10]

We identified several changes that marginally improved the performance of the algorithm by reducing the number of unplanned readmissions that are incorrectly removed from the measure, while avoiding the inappropriate inclusion of planned readmissions in the outcome. This revised algorithm, v3.0, was applied to public reporting of readmission rates at the end of 2014. Overall, implementing these changes increases the reported readmission rate very slightly. We also identified other procedures associated with high inaccuracy rates, removal of which would have larger impact on reporting rates, and which therefore merit further evaluation.

There are other potential methods of identifying planned readmissions. For instance, as of October 1, 2013, new administrative billing codes were created to allow hospitals to indicate that a patient was discharged with a planned acute‐care hospital inpatient readmission, without limitation as to when it will take place.[11] This code must be used at the time of the index admission to indicate that a future planned admission is expected, and was specified only to be used for neonates and patients with acute myocardial infarction. This approach, however, would omit planned readmissions that are not known to the initial discharging team, potentially missing planned readmissions. Conversely, some patients discharged with a plan for readmission may be unexpectedly readmitted for an unplanned reason. Given that the new codes were not available at the time we conducted the validation study, we were not able to determine how often the billing codes accurately identified planned readmissions. This would be an important area to consider for future study.

An alternative approach would be to create indicator codes to be applied at the time of readmission that would indicate whether that admission was planned or unplanned. Such a code would have the advantage of allowing each planned readmission to be flagged by the admitting clinicians at the time of admission rather than by an algorithm that inherently cannot be perfect. However, identifying planned readmissions at the time of readmission would also create opportunity for gaming and inconsistent application of definitions between hospitals; additional checks would need to be put in place to guard against these possibilities.

Our study has some limitations. We relied on the opinion of chart abstractors to determine whether a readmission was planned or unplanned; in a few cases, such as smoldering wounds that ultimately require surgical intervention, that determination is debatable. Abstractions were done at local institutions to minimize risks to patient privacy, and therefore we could not centrally verify determinations of planned status except by reviewing source of admission, dates of procedures, and narrative comments reported by the abstractors. Finally, we did not have sufficient volume of planned procedures to determine accuracy of the algorithm for less common procedure categories or individual procedures within categories.

In summary, we developed an algorithm to identify planned readmissions from administrative data that had high specificity and moderate sensitivity, and refined it based on chart validation. This algorithm is in use in public reporting of readmission measures to maximize the probability that the reported readmission rates represent truly unplanned readmissions.[12]

Disclosures: Financial supportThis work was performed under contract HHSM‐500‐2008‐0025I/HHSM‐500‐T0001, Modification No. 000008, titled Measure Instrument Development and Support, funded by the Centers for Medicare and Medicaid Services (CMS), an agency of the US Department of Health and Human Services. Drs. Horwitz and Ross are supported by the National Institute on Aging (K08 AG038336 and K08 AG032886, respectively) and by the American Federation for Aging Research through the Paul B. Beeson Career Development Award Program. Dr. Krumholz is supported by grant U01 HL105270‐05 (Center for Cardiovascular Outcomes Research at Yale University) from the National Heart, Lung, and Blood Institute. No funding source had any role in the study design; in the collection, analysis, and interpretation of data; or in the writing of the article. The CMS reviewed and approved the use of its data for this work and approved submission of the manuscript. All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that all authors have support from the CMS for the submitted work. In addition, Dr. Ross is a member of a scientific advisory board for FAIR Health Inc. Dr. Krumholz chairs a cardiac scientific advisory board for UnitedHealth and is the recipient of research agreements from Medtronic and Johnson & Johnson through Yale University, to develop methods of clinical trial data sharing. All other authors report no conflicts of interest.

Elderly patients (aged 65 years and older) consume a disproportionate amount of acute health care resources, composing up to 20% of emergency department (ED) visits,1, 2 having a 2‐fold to 5‐fold increase in likelihood of hospital admission,1 and frequently incurring lengths of hospital stay (LOS) approximately 15% higher than the national averages.3 In addition, they are at increased risk for hospital readmission in the 90‐day interval following hospital discharge.1, 4, 5 Specific risk factors for readmission include age above 80 years, discharge within the previous 30 days, the presence of 3 or more comorbid diagnoses, use of 5 or more prescription medications, difficulty with at least 1 activity of daily living (ADL), and lack of discharge education.6 These risk factors can translate into adverse drug events,79 exacerbations of chronic diseases,10 or functional decline4, 5 that can trigger ED visits or hospital readmission.

Hospital‐based care coordinationdefined as a multidisciplinary interaction between inpatients and providers that focuses on education, communication, and discharge planning with the primary aim of improving outcomeshas demonstrated inconsistent results as a mechanism to reduce LOS, postdischarge ED visits, or hospital readmission rates. While disease‐specific care coordination programs for congestive heart failure and chronic obstructive pulmonary disease have been effective in reducing rehospitalization rates,1015 the benefits of comprehensive care coordination for elderly general medical inpatients with a broader range of diagnoses are less clear. In a group of 750 elderly patients with 1 of 11 common inpatient diagnoses (such as stroke or hip fracture) likely to ultimately require a high level of home support, Coleman et al.16 found that a structured transitional care program centered on a personal coach decreased rehospitalization rates at 30 and 90 days. Preen et al.17 found improved patient involvement and perceived quality of life with care coordination focused on discharge planning, but no impact on LOS. Likewise, a recent meta‐analysis18 failed to demonstrate statistically significant differences in mortality, LOS, or readmission rates in hospitalized patients who received intensive care coordination versus usual care; however, variation in the components of the care coordination intervention and reported outcomes restricted the ability to pool data in this study.

Care coordination programs demonstrating efficacy in reducing health care utilization in elderly medical patients have generally included an outpatient transitional component with out‐of‐hospital postacute care visits by health care personnel such as a nurse, pharmacist, or physician.1923 These offsite interventions generate additional expenses and resource demands that may not be practical for smaller hospitals to implement.24, 25 In contrast, hospital‐based care coordination programs have clear ownership and thus may be more practical to disseminate. Individual elements of hospital‐based care coordination such as pharmacist counseling, discharge education, and telephone follow‐up have been shown to reduce ED visitation and readmission rates in high‐risk elderly patients. Less information is available regarding the impact of these interventions delivered in an aggregate bundle by hospital staff in the absence of bridging transitional visits.2629

The objective of this pilot study was to determine whether a supplemental elderly care bundle, targeted to high‐risk inpatients by hospital staff as an enhancement to existing care coordination, would affect postdischarge readmission and ED visit rates. The intervention was designed to capitalize on existing resources, and focused specifically on elderly inpatients who were hospitalized with diagnoses commonly encountered in a general medical unit and predisposed to recidivism.

Patients and Methods

Delivery of the Supplemental Care Bundle

Starting no later than 24 hours after enrollment and continuing up to 1 week following hospital discharge, intervention group patients received a targeted care bundle provided by 1 of 3 care coordinators (CCs) and 1 of 4 clinical pharmacists (CPs) working with the study team. The care bundle was designed as an intensive patient‐centered educational program that would augment BUMC's existing care coordination processes (delivered to all patients regardless of study participation); specific elements are displayed in Figure 1. Study CCs saw patients daily throughout their hospital stay, and instructed patients on specific health conditions, with an emphasis on optimizing home self‐care and contingency plans if problems arose. CP visits focused on medication reconciliation and education regarding any new agents started during the hospitalization. The personal health record (PHR) provided a tool to engage patients in self‐care, and as discussed by Coleman et al.,7, 16, 33 promoted information transfer from the hospital to outpatient settings. During the postdischarge phone call, CCs followed a basic script to confirm receipt of medical equipment, medications, home health arrangements, and scheduling of follow‐up appointments. They also used this contact as an opportunity to reinforce patient education on managing their conditions. CPs reviewed medication use (type, schedule, dose), and spoke with patients about any symptoms they may have experienced as medication side effects. If indicated based on their phone discussions, both CCs and CPs could recommend an action plan to the patient.

Figure 1

The study CCs and CPs were existing hospital staff and performed their research activities in addition to their usual duties. Study CCs were highly experienced (averaging 8 years of inpatient floor nursing plus 10 years as CCs) and all had advanced nursing certifications (ACM, BSN, or MSN). The CPs were upper‐level pharmacy residents completing their inpatient clinical rotations. Additional training for both study CCs and CPs was limited to a series of 3 meetings (each 45 minutes in duration) regarding the intent and delivery of the supplemental care bundle, including use of study forms.

At the time of the trial, the particular CCs and CPs chosen to deliver the supplemental care bundle had work assignments ensuring that crossover between intervention and usual care groups would not occur. For example, 1 of the study CCs normally covered a surgical floor such that her normal scope of responsibilities would not influence the medical patients in the study (their baseline care coordination was provided by nonstudy personnel). Medication reconciliation and medication education is generally performed by floor nursing staff rather than CPs at BUMC.

Results

The final sample size for this pilot was small, with 41 total patients (21 controls, 20 interventions). The main reason for enrollment failure of patients meeting study criteria was an inability to obtain informed consent. Sixty patients declined participation after being approached, and another 56 patients were unable to give their informed consent due to impairments (poor cognition, medication induced sedation, severity of illness) with lack of an available proxy to give written consent during the 72‐hour postadmission recruitment window. There were no statistically significant differences in the baseline characteristics of the intervention and control groups (Table 2). A similar proportion of patients (23% in the intervention, 15% in controls; P = 0.70) had preexisting diagnoses of dementia or depression. However, on APR‐DRG measures relating to acuity of illness and mortality risk, patients in the intervention group trended toward higher severity (Table 2). Likewise, although it was not a statistically significant difference, 13 of 20 patients in the intervention group were taking medications from 2 drug classes commonly implicated in adverse drug events (warfarin, insulin, diuretics, sedating agents) as part of their discharge medication regimen compared to 10 of 21 patients in the control group.

Study outcomes are displayed in Table 3. Mean LOS is reported as a descriptive finding; there was insufficient power to compare this outcome statistically between groups. The majority of patients were discharged to home. A similar proportion of patients in the intervention (20%) and control groups (22%) who had lived at home immediately prior to admission were discharged from the hospital to skilled care facilities (P = 0.87). The number of readmissions/ED visits (taken as a composite measure of unplanned healthcare utilization) within 30 days of discharge was lower in the intervention group; by 60 days, there was no longer a statistically significant difference in readmission/ED visit rates between groups. For those patients who had a readmission or ED visit following hospital discharge, the intervention group had a longer time interval to first event compared to controls (36.2 versus 15.7 days, P = 0.05). Of the patients discharged to skilled care, 1 in the intervention group (at 53 days) and 1 in the control group (at 16 days) had a readmission/ED visit event. Figure 2 shows time‐to‐first readmission or ED visit event curves at 30 and 60 days for both intervention and control groups. For patients who had a readmission/ED visit event, LOS for this episode was 2.2 2.1 days in controls and 3.7 2.1 days in the intervention group (insufficient power for statistical comparison). The study's small sample size prevented development of a meaningful regression model.

Figure 2

Resource utilization and the specifics of patient‐study personnel interaction associated with the intervention were tracked. Research assistants spent an average of 50 minutes daily screening charts for potential candidates. For the 20 patients who received the supplemental elderly care bundle, study CCs averaged 20 to 25 minutes per patient daily of additional time counseling patients and families, identifying and attending to discharge barriers, filling out documentation, and faxing the supplemental study discharge form to the patient's primary care physician. Any residual home care needs or issues unresolved at discharge were addressed with the patient in the 5 to 7 day follow‐up phone call. Similarly, study CPs expended approximately 20 minutes daily per patient providing medication education, reconciliation, and optimization of drug therapy. Study pharmacists recommended a change to the medication regimens of 10 patients in the intervention group; physicians acted upon these recommendations for 7 of the patients. The changes included dosage adjustment, discontinuation of medications due to possible drug interaction or duplication of drugs with the same pharmacologic effect, and addition of medications as indicated by patient condition or to reconcile with patients' at‐home medication regimens. Patients contacted via phone by the study pharmacist within 1 week after discharge were able to describe proper use of new medications started in the hospital and confirm that they obtained or had the means to obtain the prescribed drugs.

Discussion

This pilot study examined the effects of a supplemental care bundle involving patient education and discharge planning delivered by hospital‐based CCs and CPs on the rate of readmission/ED visitation in 41 elderly (70 years of age) patients. The study was not adequately powered to detect an impact of the intervention on index LOS. The care bundle did lead to significantly fewer readmissions or ED visits 30 days postdischarge and appeared to increase the time interval to first unplanned readmission or ED visit compared to usual care. This effect was no longer present at 60 days postdischarge. Resource allocations and scope of duties for CCs and CPs (an average of 20 minutes per patient per day) related to delivering the intervention were realistic for broader implementation in the hospitalized elderly population at high risk for readmission or ED visitation following discharge.

Length of stay for the initial hospitalization associated with the care bundle was an original outcome of interest to the study team. However, with the final enrollment of 41 patients and a power of 0.8, the between group difference would have needed to be 2.6 days to be statistically significant. It is likely that any change in LOS related to the care bundle would be much smaller, particularly since 2 key determinants of LOS, severity of illness and physician behavior, were beyond this patient education‐oriented intervention's scope of influence.3437 Furthermore, the diverse range of eligible diagnoses limited the study CCs' ability to reduce variability through use of clinical care pathways. One approach in leveraging an elderly care bundle to reduce LOS may be to focus on a specific disease that has well‐established inpatient benchmarks and treatment algorithms. For example, in patients with community‐acquired pneumonia, the use of care coordination in combination with standardized order sets decreased LOS without compromising safety, mainly by shortening the time from clinical stability to discharge.38

On separation of the readmission/ED visit outcome into 30 and 60 day postdischarge time frames, the intervention group had a lower rate of unplanned acute health care use within 30 days postdischarge; the difference between groups had dissipated by 60 days postdischarge. This convergence suggests that a hospital‐based intervention's influence is strongest closer to the time of the initial hospital stay, and wanes as more time has elapsed. Indeed, interventions that have successfully maintained lower readmission rates beyond 60 and 90 days postdischarge in a high‐risk elderly population (such as the program advocated by Coleman et al.16) have included a transitional care provider engaging patients during the hospitalization and performing subsequent visits to the home or nursing facility.33 An optimal intervention would capitalize on the hospital‐based staff's ability to improve short‐term readmission/ED visit rates while linking patients to longer‐term transitional care to extend these outcomes. Electronic health records could potentially facilitate these care transitions, beginning with an automated screening process for identification of high‐risk inpatients and continuing through postdischarge follow‐up. How to develop these resources in settings where outpatient practices are independent or only loosely affiliated with hospitals is an area for continued investigation.

In a group of elderly patients with multiple comorbidities and complex pharmacotherapy regimens, the study bundle component targeting medication management appears to be a high‐yield intervention to reduce unplanned health care utilization following hospital discharge. These patients are more susceptible to nonadherence and drug‐related adverse events, which may contribute to hospital readmission or ED visitation.7, 9, 39 Consistent with findings at other sites,28, 40 a heightened level of CP involvement in the care of high‐risk elderly patients may have helped reduce these undesirable outcomes. Of the 9 readmission/ED visit events in the control group, 3 were attributable to medication related complications (2 from sedatives, 1 from a diuretic). None of the readmission/ED visit events in intervention group patients stemmed from medication effects.

Correspondingly, the research CCs' provision of daily condition‐specific education, additional time to more thoroughly investigate discharge needs, engagement of patients' families as active partners in self‐care, and the use of a structured discharge form along with follow‐up phone calls may have better prepared patients to manage their health problems once released from the hospital.26, 28, 29 For example, 1 patient in the control group was readmitted less than 24 hours after initial discharge due to inability to perform self‐care at home. Given the study power issues described previously, data on LOS for the second hospitalization for patients who had a readmission event are difficult to interpret, but could suggest the occurrence of some shorter, preventable readmissions in the control group. Conversely, the readmission/ED visit events in intervention patients appeared to be associated with a specific medical condition (eg, failure of diabetic cellulitis to respond to appropriate outpatient treatment) rather than problems that would have been corrected with an educational/self‐management program such as this targeted care bundle.

This pilot study had several limitations. The main issue was a small patient sample size that was primarily due to an inability to obtain informed consent. Design of the study as a randomized controlled trial and plans to disseminate study findings beyond BHCS necessitated IRB approval rather than delivery of the supplemental care bundle as a quality improvement (QI) project. Placing QI initiatives under research regulations can lead to project delays, higher costs, and patient frustrations with the process.41, 42 This tension was evident during study screening and enrollment, as many patients who otherwise met criteria and would potentially benefit from the intervention were hesitant to participate in a research study or refused to sign a multipage consent document. The difficulties of enrolling elderly patients in clinical trials have been well‐described.43, 44 Further research involving a minimal‐risk, educational intervention such as this elderly care‐bundle would likely better fit under the category of expedited IRB review with waiver or modification of the informed consent process.45

Incomplete blinding could have potentially affected our results. At the study site, the team members delivering the care bundle were a regular part of the hospital staff (as opposed to external researchers), and it is not unusual for a CC or a pharmacist to enter a patient's room (eg, to confirm a drug allergy history). In view of this, the impact of imperfect blinding on 30‐day outcomes would likely be minimal. Furthermore, a floor staff perception that a specific patient was being taken care of by the study team resulting in a lower than usual level of care, would tend to bias the result of the intervention toward the null effect.

vThe study cohort did not have enough subjects to perform analyses (ie, modeling or examination of subgroups) beyond basic comparative findings. Issues such as preadmission living situation and the presence of depression or cognitive impairment (Mini‐Mental Status Exams were not performed on these patients) may potentially influence postdischarge recidivism; their effects can not be reliably ascertained from these data. Additionally, to prevent study personnel from engaging patients who would soon be going home, it was felt that the benefits from the care bundle would be recognized only if the intervention could be initiated within 72 hours of admission and delivered in full, a requirement that further reduced the enrollment pool. The intent of this pilot work was to guide future investigations surrounding hospital‐to‐home transitional care. The next phase of research in this area will need an enhanced sample size with more extensive baseline data collection so that potential confounding factors or outcomes in specific populations can be explored.

Another problem restricting applicability of study findings was the use of only 3 different CCs and 3 pharmacists on the research team to deliver the components of patient education, discharge planning, and medication counseling in the elderly care bundle. Personnel for the trial were chosen for their experience and interest in the area of care transitions. To distinguish the benefit of the elderly care bundle in general versus the expertise of these particular CCs and study pharmacists, a larger‐scale, multisite trial would be necessary. Lastly, due to resource constraints, patients who resided in long‐term care (either LTAC, SNFs, or nursing homes) prior to admission with anticipated return to those sites were not eligible for the study. Similar to the patients whose comorbidities or acute severity of illness prevented informed consent, this segment of the elderly population may have derived even more benefit from receipt of the elderly care bundle.10, 15, 46 Despite exclusion of this group (which would be expected to lessen the impact of the intervention), a difference in readmission/ED visits rates at 30 days following discharge was observed.

Conclusions

This pilot randomized clinical trial (RCT) evaluated the effects of a supplemental, aggregate care bundle centered on patient education, discharge planning, and medication counseling and reconciliation compared to usual care in a group of elderly patients at high risk of readmission or ED visitation following an index hospitalization. The intervention was designed to be reproducible and make use of existing hospital resources. Probably through facilitation of patient self‐care and home management, the elderly care bundle reduced the composite outcome of readmission/ED visits at 30 days postdischarge. By 60 days, this effect had waned, demonstrating the short‐term benefit of a hospital‐based educational intervention and stressing the need to incorporate additional outpatient transitional care support to sustain favorable outcomes. The study was not powered to detect small differences (which would be more likely than a change of multiple days) in length of index hospital stay related to the care bundle. There were important study limitations (primarily associated with small sample size), and this work should be viewed as hypothesis‐generating. Future trials should assess the impact of a standardized targeted care bundle delivered across multiple healthcare systems on a larger cohort of high‐risk elderly patients, including analysis of financial and personnel allocations relative to the benefits of the intervention.

Elderly patients (aged 65 years and older) consume a disproportionate amount of acute health care resources, composing up to 20% of emergency department (ED) visits,1, 2 having a 2‐fold to 5‐fold increase in likelihood of hospital admission,1 and frequently incurring lengths of hospital stay (LOS) approximately 15% higher than the national averages.3 In addition, they are at increased risk for hospital readmission in the 90‐day interval following hospital discharge.1, 4, 5 Specific risk factors for readmission include age above 80 years, discharge within the previous 30 days, the presence of 3 or more comorbid diagnoses, use of 5 or more prescription medications, difficulty with at least 1 activity of daily living (ADL), and lack of discharge education.6 These risk factors can translate into adverse drug events,79 exacerbations of chronic diseases,10 or functional decline4, 5 that can trigger ED visits or hospital readmission.

Hospital‐based care coordinationdefined as a multidisciplinary interaction between inpatients and providers that focuses on education, communication, and discharge planning with the primary aim of improving outcomeshas demonstrated inconsistent results as a mechanism to reduce LOS, postdischarge ED visits, or hospital readmission rates. While disease‐specific care coordination programs for congestive heart failure and chronic obstructive pulmonary disease have been effective in reducing rehospitalization rates,1015 the benefits of comprehensive care coordination for elderly general medical inpatients with a broader range of diagnoses are less clear. In a group of 750 elderly patients with 1 of 11 common inpatient diagnoses (such as stroke or hip fracture) likely to ultimately require a high level of home support, Coleman et al.16 found that a structured transitional care program centered on a personal coach decreased rehospitalization rates at 30 and 90 days. Preen et al.17 found improved patient involvement and perceived quality of life with care coordination focused on discharge planning, but no impact on LOS. Likewise, a recent meta‐analysis18 failed to demonstrate statistically significant differences in mortality, LOS, or readmission rates in hospitalized patients who received intensive care coordination versus usual care; however, variation in the components of the care coordination intervention and reported outcomes restricted the ability to pool data in this study.

Care coordination programs demonstrating efficacy in reducing health care utilization in elderly medical patients have generally included an outpatient transitional component with out‐of‐hospital postacute care visits by health care personnel such as a nurse, pharmacist, or physician.1923 These offsite interventions generate additional expenses and resource demands that may not be practical for smaller hospitals to implement.24, 25 In contrast, hospital‐based care coordination programs have clear ownership and thus may be more practical to disseminate. Individual elements of hospital‐based care coordination such as pharmacist counseling, discharge education, and telephone follow‐up have been shown to reduce ED visitation and readmission rates in high‐risk elderly patients. Less information is available regarding the impact of these interventions delivered in an aggregate bundle by hospital staff in the absence of bridging transitional visits.2629

The objective of this pilot study was to determine whether a supplemental elderly care bundle, targeted to high‐risk inpatients by hospital staff as an enhancement to existing care coordination, would affect postdischarge readmission and ED visit rates. The intervention was designed to capitalize on existing resources, and focused specifically on elderly inpatients who were hospitalized with diagnoses commonly encountered in a general medical unit and predisposed to recidivism.

Patients and Methods

Delivery of the Supplemental Care Bundle

Starting no later than 24 hours after enrollment and continuing up to 1 week following hospital discharge, intervention group patients received a targeted care bundle provided by 1 of 3 care coordinators (CCs) and 1 of 4 clinical pharmacists (CPs) working with the study team. The care bundle was designed as an intensive patient‐centered educational program that would augment BUMC's existing care coordination processes (delivered to all patients regardless of study participation); specific elements are displayed in Figure 1. Study CCs saw patients daily throughout their hospital stay, and instructed patients on specific health conditions, with an emphasis on optimizing home self‐care and contingency plans if problems arose. CP visits focused on medication reconciliation and education regarding any new agents started during the hospitalization. The personal health record (PHR) provided a tool to engage patients in self‐care, and as discussed by Coleman et al.,7, 16, 33 promoted information transfer from the hospital to outpatient settings. During the postdischarge phone call, CCs followed a basic script to confirm receipt of medical equipment, medications, home health arrangements, and scheduling of follow‐up appointments. They also used this contact as an opportunity to reinforce patient education on managing their conditions. CPs reviewed medication use (type, schedule, dose), and spoke with patients about any symptoms they may have experienced as medication side effects. If indicated based on their phone discussions, both CCs and CPs could recommend an action plan to the patient.

Figure 1

The study CCs and CPs were existing hospital staff and performed their research activities in addition to their usual duties. Study CCs were highly experienced (averaging 8 years of inpatient floor nursing plus 10 years as CCs) and all had advanced nursing certifications (ACM, BSN, or MSN). The CPs were upper‐level pharmacy residents completing their inpatient clinical rotations. Additional training for both study CCs and CPs was limited to a series of 3 meetings (each 45 minutes in duration) regarding the intent and delivery of the supplemental care bundle, including use of study forms.

At the time of the trial, the particular CCs and CPs chosen to deliver the supplemental care bundle had work assignments ensuring that crossover between intervention and usual care groups would not occur. For example, 1 of the study CCs normally covered a surgical floor such that her normal scope of responsibilities would not influence the medical patients in the study (their baseline care coordination was provided by nonstudy personnel). Medication reconciliation and medication education is generally performed by floor nursing staff rather than CPs at BUMC.

Results

The final sample size for this pilot was small, with 41 total patients (21 controls, 20 interventions). The main reason for enrollment failure of patients meeting study criteria was an inability to obtain informed consent. Sixty patients declined participation after being approached, and another 56 patients were unable to give their informed consent due to impairments (poor cognition, medication induced sedation, severity of illness) with lack of an available proxy to give written consent during the 72‐hour postadmission recruitment window. There were no statistically significant differences in the baseline characteristics of the intervention and control groups (Table 2). A similar proportion of patients (23% in the intervention, 15% in controls; P = 0.70) had preexisting diagnoses of dementia or depression. However, on APR‐DRG measures relating to acuity of illness and mortality risk, patients in the intervention group trended toward higher severity (Table 2). Likewise, although it was not a statistically significant difference, 13 of 20 patients in the intervention group were taking medications from 2 drug classes commonly implicated in adverse drug events (warfarin, insulin, diuretics, sedating agents) as part of their discharge medication regimen compared to 10 of 21 patients in the control group.

Study outcomes are displayed in Table 3. Mean LOS is reported as a descriptive finding; there was insufficient power to compare this outcome statistically between groups. The majority of patients were discharged to home. A similar proportion of patients in the intervention (20%) and control groups (22%) who had lived at home immediately prior to admission were discharged from the hospital to skilled care facilities (P = 0.87). The number of readmissions/ED visits (taken as a composite measure of unplanned healthcare utilization) within 30 days of discharge was lower in the intervention group; by 60 days, there was no longer a statistically significant difference in readmission/ED visit rates between groups. For those patients who had a readmission or ED visit following hospital discharge, the intervention group had a longer time interval to first event compared to controls (36.2 versus 15.7 days, P = 0.05). Of the patients discharged to skilled care, 1 in the intervention group (at 53 days) and 1 in the control group (at 16 days) had a readmission/ED visit event. Figure 2 shows time‐to‐first readmission or ED visit event curves at 30 and 60 days for both intervention and control groups. For patients who had a readmission/ED visit event, LOS for this episode was 2.2 2.1 days in controls and 3.7 2.1 days in the intervention group (insufficient power for statistical comparison). The study's small sample size prevented development of a meaningful regression model.

Figure 2

Resource utilization and the specifics of patient‐study personnel interaction associated with the intervention were tracked. Research assistants spent an average of 50 minutes daily screening charts for potential candidates. For the 20 patients who received the supplemental elderly care bundle, study CCs averaged 20 to 25 minutes per patient daily of additional time counseling patients and families, identifying and attending to discharge barriers, filling out documentation, and faxing the supplemental study discharge form to the patient's primary care physician. Any residual home care needs or issues unresolved at discharge were addressed with the patient in the 5 to 7 day follow‐up phone call. Similarly, study CPs expended approximately 20 minutes daily per patient providing medication education, reconciliation, and optimization of drug therapy. Study pharmacists recommended a change to the medication regimens of 10 patients in the intervention group; physicians acted upon these recommendations for 7 of the patients. The changes included dosage adjustment, discontinuation of medications due to possible drug interaction or duplication of drugs with the same pharmacologic effect, and addition of medications as indicated by patient condition or to reconcile with patients' at‐home medication regimens. Patients contacted via phone by the study pharmacist within 1 week after discharge were able to describe proper use of new medications started in the hospital and confirm that they obtained or had the means to obtain the prescribed drugs.

Discussion

This pilot study examined the effects of a supplemental care bundle involving patient education and discharge planning delivered by hospital‐based CCs and CPs on the rate of readmission/ED visitation in 41 elderly (70 years of age) patients. The study was not adequately powered to detect an impact of the intervention on index LOS. The care bundle did lead to significantly fewer readmissions or ED visits 30 days postdischarge and appeared to increase the time interval to first unplanned readmission or ED visit compared to usual care. This effect was no longer present at 60 days postdischarge. Resource allocations and scope of duties for CCs and CPs (an average of 20 minutes per patient per day) related to delivering the intervention were realistic for broader implementation in the hospitalized elderly population at high risk for readmission or ED visitation following discharge.

Length of stay for the initial hospitalization associated with the care bundle was an original outcome of interest to the study team. However, with the final enrollment of 41 patients and a power of 0.8, the between group difference would have needed to be 2.6 days to be statistically significant. It is likely that any change in LOS related to the care bundle would be much smaller, particularly since 2 key determinants of LOS, severity of illness and physician behavior, were beyond this patient education‐oriented intervention's scope of influence.3437 Furthermore, the diverse range of eligible diagnoses limited the study CCs' ability to reduce variability through use of clinical care pathways. One approach in leveraging an elderly care bundle to reduce LOS may be to focus on a specific disease that has well‐established inpatient benchmarks and treatment algorithms. For example, in patients with community‐acquired pneumonia, the use of care coordination in combination with standardized order sets decreased LOS without compromising safety, mainly by shortening the time from clinical stability to discharge.38

On separation of the readmission/ED visit outcome into 30 and 60 day postdischarge time frames, the intervention group had a lower rate of unplanned acute health care use within 30 days postdischarge; the difference between groups had dissipated by 60 days postdischarge. This convergence suggests that a hospital‐based intervention's influence is strongest closer to the time of the initial hospital stay, and wanes as more time has elapsed. Indeed, interventions that have successfully maintained lower readmission rates beyond 60 and 90 days postdischarge in a high‐risk elderly population (such as the program advocated by Coleman et al.16) have included a transitional care provider engaging patients during the hospitalization and performing subsequent visits to the home or nursing facility.33 An optimal intervention would capitalize on the hospital‐based staff's ability to improve short‐term readmission/ED visit rates while linking patients to longer‐term transitional care to extend these outcomes. Electronic health records could potentially facilitate these care transitions, beginning with an automated screening process for identification of high‐risk inpatients and continuing through postdischarge follow‐up. How to develop these resources in settings where outpatient practices are independent or only loosely affiliated with hospitals is an area for continued investigation.

In a group of elderly patients with multiple comorbidities and complex pharmacotherapy regimens, the study bundle component targeting medication management appears to be a high‐yield intervention to reduce unplanned health care utilization following hospital discharge. These patients are more susceptible to nonadherence and drug‐related adverse events, which may contribute to hospital readmission or ED visitation.7, 9, 39 Consistent with findings at other sites,28, 40 a heightened level of CP involvement in the care of high‐risk elderly patients may have helped reduce these undesirable outcomes. Of the 9 readmission/ED visit events in the control group, 3 were attributable to medication related complications (2 from sedatives, 1 from a diuretic). None of the readmission/ED visit events in intervention group patients stemmed from medication effects.

Correspondingly, the research CCs' provision of daily condition‐specific education, additional time to more thoroughly investigate discharge needs, engagement of patients' families as active partners in self‐care, and the use of a structured discharge form along with follow‐up phone calls may have better prepared patients to manage their health problems once released from the hospital.26, 28, 29 For example, 1 patient in the control group was readmitted less than 24 hours after initial discharge due to inability to perform self‐care at home. Given the study power issues described previously, data on LOS for the second hospitalization for patients who had a readmission event are difficult to interpret, but could suggest the occurrence of some shorter, preventable readmissions in the control group. Conversely, the readmission/ED visit events in intervention patients appeared to be associated with a specific medical condition (eg, failure of diabetic cellulitis to respond to appropriate outpatient treatment) rather than problems that would have been corrected with an educational/self‐management program such as this targeted care bundle.

This pilot study had several limitations. The main issue was a small patient sample size that was primarily due to an inability to obtain informed consent. Design of the study as a randomized controlled trial and plans to disseminate study findings beyond BHCS necessitated IRB approval rather than delivery of the supplemental care bundle as a quality improvement (QI) project. Placing QI initiatives under research regulations can lead to project delays, higher costs, and patient frustrations with the process.41, 42 This tension was evident during study screening and enrollment, as many patients who otherwise met criteria and would potentially benefit from the intervention were hesitant to participate in a research study or refused to sign a multipage consent document. The difficulties of enrolling elderly patients in clinical trials have been well‐described.43, 44 Further research involving a minimal‐risk, educational intervention such as this elderly care‐bundle would likely better fit under the category of expedited IRB review with waiver or modification of the informed consent process.45

Incomplete blinding could have potentially affected our results. At the study site, the team members delivering the care bundle were a regular part of the hospital staff (as opposed to external researchers), and it is not unusual for a CC or a pharmacist to enter a patient's room (eg, to confirm a drug allergy history). In view of this, the impact of imperfect blinding on 30‐day outcomes would likely be minimal. Furthermore, a floor staff perception that a specific patient was being taken care of by the study team resulting in a lower than usual level of care, would tend to bias the result of the intervention toward the null effect.

vThe study cohort did not have enough subjects to perform analyses (ie, modeling or examination of subgroups) beyond basic comparative findings. Issues such as preadmission living situation and the presence of depression or cognitive impairment (Mini‐Mental Status Exams were not performed on these patients) may potentially influence postdischarge recidivism; their effects can not be reliably ascertained from these data. Additionally, to prevent study personnel from engaging patients who would soon be going home, it was felt that the benefits from the care bundle would be recognized only if the intervention could be initiated within 72 hours of admission and delivered in full, a requirement that further reduced the enrollment pool. The intent of this pilot work was to guide future investigations surrounding hospital‐to‐home transitional care. The next phase of research in this area will need an enhanced sample size with more extensive baseline data collection so that potential confounding factors or outcomes in specific populations can be explored.

Another problem restricting applicability of study findings was the use of only 3 different CCs and 3 pharmacists on the research team to deliver the components of patient education, discharge planning, and medication counseling in the elderly care bundle. Personnel for the trial were chosen for their experience and interest in the area of care transitions. To distinguish the benefit of the elderly care bundle in general versus the expertise of these particular CCs and study pharmacists, a larger‐scale, multisite trial would be necessary. Lastly, due to resource constraints, patients who resided in long‐term care (either LTAC, SNFs, or nursing homes) prior to admission with anticipated return to those sites were not eligible for the study. Similar to the patients whose comorbidities or acute severity of illness prevented informed consent, this segment of the elderly population may have derived even more benefit from receipt of the elderly care bundle.10, 15, 46 Despite exclusion of this group (which would be expected to lessen the impact of the intervention), a difference in readmission/ED visits rates at 30 days following discharge was observed.

Conclusions

This pilot randomized clinical trial (RCT) evaluated the effects of a supplemental, aggregate care bundle centered on patient education, discharge planning, and medication counseling and reconciliation compared to usual care in a group of elderly patients at high risk of readmission or ED visitation following an index hospitalization. The intervention was designed to be reproducible and make use of existing hospital resources. Probably through facilitation of patient self‐care and home management, the elderly care bundle reduced the composite outcome of readmission/ED visits at 30 days postdischarge. By 60 days, this effect had waned, demonstrating the short‐term benefit of a hospital‐based educational intervention and stressing the need to incorporate additional outpatient transitional care support to sustain favorable outcomes. The study was not powered to detect small differences (which would be more likely than a change of multiple days) in length of index hospital stay related to the care bundle. There were important study limitations (primarily associated with small sample size), and this work should be viewed as hypothesis‐generating. Future trials should assess the impact of a standardized targeted care bundle delivered across multiple healthcare systems on a larger cohort of high‐risk elderly patients, including analysis of financial and personnel allocations relative to the benefits of the intervention.

Elderly patients (aged 65 years and older) consume a disproportionate amount of acute health care resources, composing up to 20% of emergency department (ED) visits,1, 2 having a 2‐fold to 5‐fold increase in likelihood of hospital admission,1 and frequently incurring lengths of hospital stay (LOS) approximately 15% higher than the national averages.3 In addition, they are at increased risk for hospital readmission in the 90‐day interval following hospital discharge.1, 4, 5 Specific risk factors for readmission include age above 80 years, discharge within the previous 30 days, the presence of 3 or more comorbid diagnoses, use of 5 or more prescription medications, difficulty with at least 1 activity of daily living (ADL), and lack of discharge education.6 These risk factors can translate into adverse drug events,79 exacerbations of chronic diseases,10 or functional decline4, 5 that can trigger ED visits or hospital readmission.

Hospital‐based care coordinationdefined as a multidisciplinary interaction between inpatients and providers that focuses on education, communication, and discharge planning with the primary aim of improving outcomeshas demonstrated inconsistent results as a mechanism to reduce LOS, postdischarge ED visits, or hospital readmission rates. While disease‐specific care coordination programs for congestive heart failure and chronic obstructive pulmonary disease have been effective in reducing rehospitalization rates,1015 the benefits of comprehensive care coordination for elderly general medical inpatients with a broader range of diagnoses are less clear. In a group of 750 elderly patients with 1 of 11 common inpatient diagnoses (such as stroke or hip fracture) likely to ultimately require a high level of home support, Coleman et al.16 found that a structured transitional care program centered on a personal coach decreased rehospitalization rates at 30 and 90 days. Preen et al.17 found improved patient involvement and perceived quality of life with care coordination focused on discharge planning, but no impact on LOS. Likewise, a recent meta‐analysis18 failed to demonstrate statistically significant differences in mortality, LOS, or readmission rates in hospitalized patients who received intensive care coordination versus usual care; however, variation in the components of the care coordination intervention and reported outcomes restricted the ability to pool data in this study.

Care coordination programs demonstrating efficacy in reducing health care utilization in elderly medical patients have generally included an outpatient transitional component with out‐of‐hospital postacute care visits by health care personnel such as a nurse, pharmacist, or physician.1923 These offsite interventions generate additional expenses and resource demands that may not be practical for smaller hospitals to implement.24, 25 In contrast, hospital‐based care coordination programs have clear ownership and thus may be more practical to disseminate. Individual elements of hospital‐based care coordination such as pharmacist counseling, discharge education, and telephone follow‐up have been shown to reduce ED visitation and readmission rates in high‐risk elderly patients. Less information is available regarding the impact of these interventions delivered in an aggregate bundle by hospital staff in the absence of bridging transitional visits.2629

The objective of this pilot study was to determine whether a supplemental elderly care bundle, targeted to high‐risk inpatients by hospital staff as an enhancement to existing care coordination, would affect postdischarge readmission and ED visit rates. The intervention was designed to capitalize on existing resources, and focused specifically on elderly inpatients who were hospitalized with diagnoses commonly encountered in a general medical unit and predisposed to recidivism.

Patients and Methods

Delivery of the Supplemental Care Bundle

Starting no later than 24 hours after enrollment and continuing up to 1 week following hospital discharge, intervention group patients received a targeted care bundle provided by 1 of 3 care coordinators (CCs) and 1 of 4 clinical pharmacists (CPs) working with the study team. The care bundle was designed as an intensive patient‐centered educational program that would augment BUMC's existing care coordination processes (delivered to all patients regardless of study participation); specific elements are displayed in Figure 1. Study CCs saw patients daily throughout their hospital stay, and instructed patients on specific health conditions, with an emphasis on optimizing home self‐care and contingency plans if problems arose. CP visits focused on medication reconciliation and education regarding any new agents started during the hospitalization. The personal health record (PHR) provided a tool to engage patients in self‐care, and as discussed by Coleman et al.,7, 16, 33 promoted information transfer from the hospital to outpatient settings. During the postdischarge phone call, CCs followed a basic script to confirm receipt of medical equipment, medications, home health arrangements, and scheduling of follow‐up appointments. They also used this contact as an opportunity to reinforce patient education on managing their conditions. CPs reviewed medication use (type, schedule, dose), and spoke with patients about any symptoms they may have experienced as medication side effects. If indicated based on their phone discussions, both CCs and CPs could recommend an action plan to the patient.

Figure 1

The study CCs and CPs were existing hospital staff and performed their research activities in addition to their usual duties. Study CCs were highly experienced (averaging 8 years of inpatient floor nursing plus 10 years as CCs) and all had advanced nursing certifications (ACM, BSN, or MSN). The CPs were upper‐level pharmacy residents completing their inpatient clinical rotations. Additional training for both study CCs and CPs was limited to a series of 3 meetings (each 45 minutes in duration) regarding the intent and delivery of the supplemental care bundle, including use of study forms.

At the time of the trial, the particular CCs and CPs chosen to deliver the supplemental care bundle had work assignments ensuring that crossover between intervention and usual care groups would not occur. For example, 1 of the study CCs normally covered a surgical floor such that her normal scope of responsibilities would not influence the medical patients in the study (their baseline care coordination was provided by nonstudy personnel). Medication reconciliation and medication education is generally performed by floor nursing staff rather than CPs at BUMC.

Results

The final sample size for this pilot was small, with 41 total patients (21 controls, 20 interventions). The main reason for enrollment failure of patients meeting study criteria was an inability to obtain informed consent. Sixty patients declined participation after being approached, and another 56 patients were unable to give their informed consent due to impairments (poor cognition, medication induced sedation, severity of illness) with lack of an available proxy to give written consent during the 72‐hour postadmission recruitment window. There were no statistically significant differences in the baseline characteristics of the intervention and control groups (Table 2). A similar proportion of patients (23% in the intervention, 15% in controls; P = 0.70) had preexisting diagnoses of dementia or depression. However, on APR‐DRG measures relating to acuity of illness and mortality risk, patients in the intervention group trended toward higher severity (Table 2). Likewise, although it was not a statistically significant difference, 13 of 20 patients in the intervention group were taking medications from 2 drug classes commonly implicated in adverse drug events (warfarin, insulin, diuretics, sedating agents) as part of their discharge medication regimen compared to 10 of 21 patients in the control group.

Study outcomes are displayed in Table 3. Mean LOS is reported as a descriptive finding; there was insufficient power to compare this outcome statistically between groups. The majority of patients were discharged to home. A similar proportion of patients in the intervention (20%) and control groups (22%) who had lived at home immediately prior to admission were discharged from the hospital to skilled care facilities (P = 0.87). The number of readmissions/ED visits (taken as a composite measure of unplanned healthcare utilization) within 30 days of discharge was lower in the intervention group; by 60 days, there was no longer a statistically significant difference in readmission/ED visit rates between groups. For those patients who had a readmission or ED visit following hospital discharge, the intervention group had a longer time interval to first event compared to controls (36.2 versus 15.7 days, P = 0.05). Of the patients discharged to skilled care, 1 in the intervention group (at 53 days) and 1 in the control group (at 16 days) had a readmission/ED visit event. Figure 2 shows time‐to‐first readmission or ED visit event curves at 30 and 60 days for both intervention and control groups. For patients who had a readmission/ED visit event, LOS for this episode was 2.2 2.1 days in controls and 3.7 2.1 days in the intervention group (insufficient power for statistical comparison). The study's small sample size prevented development of a meaningful regression model.

Figure 2

Resource utilization and the specifics of patient‐study personnel interaction associated with the intervention were tracked. Research assistants spent an average of 50 minutes daily screening charts for potential candidates. For the 20 patients who received the supplemental elderly care bundle, study CCs averaged 20 to 25 minutes per patient daily of additional time counseling patients and families, identifying and attending to discharge barriers, filling out documentation, and faxing the supplemental study discharge form to the patient's primary care physician. Any residual home care needs or issues unresolved at discharge were addressed with the patient in the 5 to 7 day follow‐up phone call. Similarly, study CPs expended approximately 20 minutes daily per patient providing medication education, reconciliation, and optimization of drug therapy. Study pharmacists recommended a change to the medication regimens of 10 patients in the intervention group; physicians acted upon these recommendations for 7 of the patients. The changes included dosage adjustment, discontinuation of medications due to possible drug interaction or duplication of drugs with the same pharmacologic effect, and addition of medications as indicated by patient condition or to reconcile with patients' at‐home medication regimens. Patients contacted via phone by the study pharmacist within 1 week after discharge were able to describe proper use of new medications started in the hospital and confirm that they obtained or had the means to obtain the prescribed drugs.

Discussion

This pilot study examined the effects of a supplemental care bundle involving patient education and discharge planning delivered by hospital‐based CCs and CPs on the rate of readmission/ED visitation in 41 elderly (70 years of age) patients. The study was not adequately powered to detect an impact of the intervention on index LOS. The care bundle did lead to significantly fewer readmissions or ED visits 30 days postdischarge and appeared to increase the time interval to first unplanned readmission or ED visit compared to usual care. This effect was no longer present at 60 days postdischarge. Resource allocations and scope of duties for CCs and CPs (an average of 20 minutes per patient per day) related to delivering the intervention were realistic for broader implementation in the hospitalized elderly population at high risk for readmission or ED visitation following discharge.

Length of stay for the initial hospitalization associated with the care bundle was an original outcome of interest to the study team. However, with the final enrollment of 41 patients and a power of 0.8, the between group difference would have needed to be 2.6 days to be statistically significant. It is likely that any change in LOS related to the care bundle would be much smaller, particularly since 2 key determinants of LOS, severity of illness and physician behavior, were beyond this patient education‐oriented intervention's scope of influence.3437 Furthermore, the diverse range of eligible diagnoses limited the study CCs' ability to reduce variability through use of clinical care pathways. One approach in leveraging an elderly care bundle to reduce LOS may be to focus on a specific disease that has well‐established inpatient benchmarks and treatment algorithms. For example, in patients with community‐acquired pneumonia, the use of care coordination in combination with standardized order sets decreased LOS without compromising safety, mainly by shortening the time from clinical stability to discharge.38

On separation of the readmission/ED visit outcome into 30 and 60 day postdischarge time frames, the intervention group had a lower rate of unplanned acute health care use within 30 days postdischarge; the difference between groups had dissipated by 60 days postdischarge. This convergence suggests that a hospital‐based intervention's influence is strongest closer to the time of the initial hospital stay, and wanes as more time has elapsed. Indeed, interventions that have successfully maintained lower readmission rates beyond 60 and 90 days postdischarge in a high‐risk elderly population (such as the program advocated by Coleman et al.16) have included a transitional care provider engaging patients during the hospitalization and performing subsequent visits to the home or nursing facility.33 An optimal intervention would capitalize on the hospital‐based staff's ability to improve short‐term readmission/ED visit rates while linking patients to longer‐term transitional care to extend these outcomes. Electronic health records could potentially facilitate these care transitions, beginning with an automated screening process for identification of high‐risk inpatients and continuing through postdischarge follow‐up. How to develop these resources in settings where outpatient practices are independent or only loosely affiliated with hospitals is an area for continued investigation.

In a group of elderly patients with multiple comorbidities and complex pharmacotherapy regimens, the study bundle component targeting medication management appears to be a high‐yield intervention to reduce unplanned health care utilization following hospital discharge. These patients are more susceptible to nonadherence and drug‐related adverse events, which may contribute to hospital readmission or ED visitation.7, 9, 39 Consistent with findings at other sites,28, 40 a heightened level of CP involvement in the care of high‐risk elderly patients may have helped reduce these undesirable outcomes. Of the 9 readmission/ED visit events in the control group, 3 were attributable to medication related complications (2 from sedatives, 1 from a diuretic). None of the readmission/ED visit events in intervention group patients stemmed from medication effects.

Correspondingly, the research CCs' provision of daily condition‐specific education, additional time to more thoroughly investigate discharge needs, engagement of patients' families as active partners in self‐care, and the use of a structured discharge form along with follow‐up phone calls may have better prepared patients to manage their health problems once released from the hospital.26, 28, 29 For example, 1 patient in the control group was readmitted less than 24 hours after initial discharge due to inability to perform self‐care at home. Given the study power issues described previously, data on LOS for the second hospitalization for patients who had a readmission event are difficult to interpret, but could suggest the occurrence of some shorter, preventable readmissions in the control group. Conversely, the readmission/ED visit events in intervention patients appeared to be associated with a specific medical condition (eg, failure of diabetic cellulitis to respond to appropriate outpatient treatment) rather than problems that would have been corrected with an educational/self‐management program such as this targeted care bundle.

This pilot study had several limitations. The main issue was a small patient sample size that was primarily due to an inability to obtain informed consent. Design of the study as a randomized controlled trial and plans to disseminate study findings beyond BHCS necessitated IRB approval rather than delivery of the supplemental care bundle as a quality improvement (QI) project. Placing QI initiatives under research regulations can lead to project delays, higher costs, and patient frustrations with the process.41, 42 This tension was evident during study screening and enrollment, as many patients who otherwise met criteria and would potentially benefit from the intervention were hesitant to participate in a research study or refused to sign a multipage consent document. The difficulties of enrolling elderly patients in clinical trials have been well‐described.43, 44 Further research involving a minimal‐risk, educational intervention such as this elderly care‐bundle would likely better fit under the category of expedited IRB review with waiver or modification of the informed consent process.45

Incomplete blinding could have potentially affected our results. At the study site, the team members delivering the care bundle were a regular part of the hospital staff (as opposed to external researchers), and it is not unusual for a CC or a pharmacist to enter a patient's room (eg, to confirm a drug allergy history). In view of this, the impact of imperfect blinding on 30‐day outcomes would likely be minimal. Furthermore, a floor staff perception that a specific patient was being taken care of by the study team resulting in a lower than usual level of care, would tend to bias the result of the intervention toward the null effect.

vThe study cohort did not have enough subjects to perform analyses (ie, modeling or examination of subgroups) beyond basic comparative findings. Issues such as preadmission living situation and the presence of depression or cognitive impairment (Mini‐Mental Status Exams were not performed on these patients) may potentially influence postdischarge recidivism; their effects can not be reliably ascertained from these data. Additionally, to prevent study personnel from engaging patients who would soon be going home, it was felt that the benefits from the care bundle would be recognized only if the intervention could be initiated within 72 hours of admission and delivered in full, a requirement that further reduced the enrollment pool. The intent of this pilot work was to guide future investigations surrounding hospital‐to‐home transitional care. The next phase of research in this area will need an enhanced sample size with more extensive baseline data collection so that potential confounding factors or outcomes in specific populations can be explored.

Another problem restricting applicability of study findings was the use of only 3 different CCs and 3 pharmacists on the research team to deliver the components of patient education, discharge planning, and medication counseling in the elderly care bundle. Personnel for the trial were chosen for their experience and interest in the area of care transitions. To distinguish the benefit of the elderly care bundle in general versus the expertise of these particular CCs and study pharmacists, a larger‐scale, multisite trial would be necessary. Lastly, due to resource constraints, patients who resided in long‐term care (either LTAC, SNFs, or nursing homes) prior to admission with anticipated return to those sites were not eligible for the study. Similar to the patients whose comorbidities or acute severity of illness prevented informed consent, this segment of the elderly population may have derived even more benefit from receipt of the elderly care bundle.10, 15, 46 Despite exclusion of this group (which would be expected to lessen the impact of the intervention), a difference in readmission/ED visits rates at 30 days following discharge was observed.

Conclusions

This pilot randomized clinical trial (RCT) evaluated the effects of a supplemental, aggregate care bundle centered on patient education, discharge planning, and medication counseling and reconciliation compared to usual care in a group of elderly patients at high risk of readmission or ED visitation following an index hospitalization. The intervention was designed to be reproducible and make use of existing hospital resources. Probably through facilitation of patient self‐care and home management, the elderly care bundle reduced the composite outcome of readmission/ED visits at 30 days postdischarge. By 60 days, this effect had waned, demonstrating the short‐term benefit of a hospital‐based educational intervention and stressing the need to incorporate additional outpatient transitional care support to sustain favorable outcomes. The study was not powered to detect small differences (which would be more likely than a change of multiple days) in length of index hospital stay related to the care bundle. There were important study limitations (primarily associated with small sample size), and this work should be viewed as hypothesis‐generating. Future trials should assess the impact of a standardized targeted care bundle delivered across multiple healthcare systems on a larger cohort of high‐risk elderly patients, including analysis of financial and personnel allocations relative to the benefits of the intervention.