PREDICTIVE MODELING OF HOSPITAL READMISSION RATES USING ELECTRONIC MEDICAL RECORD-WIDE MACHINE LEARNING: A CASE-STUDY USING MOUNT SINAI HEART FAILURE COHORT

Background: Heart failure (HF) hospitalization rates have remained high in the past 10 years. Numerous studies have shown significant improvement in HF readmission rates when pharmacists or pharmacy residents conduct postdischarge telephone calls. Objective: The purpose of this retrospective review of a pilot program was to evaluate the impact of pharmacy student–driven postdischarge phone calls on 30- and 90-day hospital readmission rates in patients recently discharged with HF. Methods: A retrospective manual chart review was conducted for all patients who received a telephone call from the pharmacy students. The primary endpoint compared historical readmissions, 30 and 90 days prior to hospital discharge, with 30 and 90 days post discharge readmissions. For the secondary endpoints, historical and postdischarge 30-day and 90-day readmission rates were compared for patients with a primary diagnosis of HF and for patients with a secondary diagnosis of HF. Descriptive statistics were calculated in the form of means and standard deviations for continuous variables and frequencies and percentages for categorical variables. Results: Statistically significant decrease was observed for both the 30-day ( P = .006) and 90-day ( P = .007) readmission periods. Prior to the pharmacy students’ phone calls, the overall group of 131 patients had historical readmission rates of 24.43% within 30 days and 38.17% within 90 days after hospital discharge. After the postdischarge phone calls, the readmission rates decreased to 11.45%, for 30 days, and 22.90%, for 90 days. Conclusion: Postdischarge phone calls, specifically made by pharmacy students, demonstrated a positive impact on reducing HF-associated hospital readmissions, adding to the growing body of evidence of different methods of pharmacy interventions and highlighting the clinical impact pharmacy students may have in transition of care services.

Download Full-text

Abstract 14996: Healthcare Resource Utilization and Costs in Congestive Heart Failure 90-day Episodes Within the Bundled Payment for Care Initiative (BPCI) Model 2, 2016-2018

Circulation ◽

10.1161/circ.142.suppl_3.14996 ◽

2020 ◽

Vol 142 (Suppl_3) ◽

Author(s):

Xian Shen ◽

Gabriel Sullivan ◽

Mark Adelsberg ◽

Martins Francis ◽

Taylor T Schwartz ◽

...

Keyword(s):

Heart Failure ◽

Congestive Heart Failure ◽

Hospital Stay ◽

Hospital Readmission ◽

Hospital Readmissions ◽

Angiotensin Receptor ◽

Post Discharge ◽

Readmission Rates ◽

Hospital Readmission Rates ◽

Initial Hospital

Introduction: Congestive heart failure (HF) is the fourth most commonly selected clinical episode among Model 2 participants of the Medicare Bundled Payments for Care Improvement (BPCI) Initiative. This study describes utilization of pharmacologic therapies, hospital readmission rates, and HF episode costs within the BPCI framework. Methods: The 100% sample of Medicare FFS enrollment/claims were used to identify acute hospital stays with a MS-DRG 291/292/293 between 1JAN2016 and 31DEC2018. A HF episode consisted of the initial hospital stay and all Part A & B covered services up to 90-days post-discharge. Prescription fills for angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), or angiotensin receptor-neprilysin inhibitors (ARNI) during the 90 days post-discharge were captured. Rates of all-cause and HF readmissions were reported per 10,000 episodes during the 30-, 60-, and 90-days post-discharge period. Total episode costs were defined as the sum of Medicare payments for the initial hospital stay plus all Part A & B covered medical services in the 90-day post-discharge. Results: The sample included 634,307 HF episodes. Patients received ARNIs in 3%, ACEIs/ARBs in 45%, and neither in 52% of the episodes, respectively. All-cause hospital readmission rates were 2,503, 4,465, and 6,368 per 10,000 episodes during the 30-, 60-, and 90-day periods. The 30-, 60-, and 90-day HF readmission rates were 958, 1,696, and 2,394 per 10,000 episodes. Total mean 90-day episode cost was $20,122, of which $8,002 was attributable to hospital readmissions. Conclusions: Hospital readmissions are frequent for HF patients and contribute a notable proportion of overall HF BPCI episode costs. BPCI participants may consider improving utilization of guideline directed medical therapies for HF, including ACEIs/ARBs and ARNI, as a strategy for reducing hospital readmissions and associated costs.

Download Full-text

437 Hospital readmission rates in out-of-hospital mechanical circulatory support patients justify this treatment option in end-stage heart failure patients

European Journal of Heart Failure Supplements ◽

10.1016/s1567-4215(04)90313-x ◽

2004 ◽

Vol 3 (1) ◽

pp. 108

Author(s):

A ELBANAYOSY

Keyword(s):

Heart Failure ◽

Treatment Option ◽

Hospital Readmission ◽

Mechanical Circulatory Support ◽

Circulatory Support ◽

Readmission Rates ◽

Heart Failure Patients ◽

Hospital Readmission Rates ◽

End Stage Heart Failure ◽

End Stage

Download Full-text

Comprehensive electronic medical record implementation levels not associated with 30-day all-cause readmissions within Medicare beneficiaries with heart failure.

Applied Clinical Informatics ◽

10.4338/aci-2014-01-ra-0008 ◽

2014 ◽

Vol 05 (03) ◽

pp. 670-684 ◽

Cited By ~ 6

Author(s):

P. Marken ◽

Y. Zhong ◽

S. D. Simon ◽

W. Ketcherside ◽

M. E. Patterson

Keyword(s):

Heart Failure ◽

Electronic Health Record ◽

Medical Record ◽

Electronic Medical Record ◽

Incidence Rates ◽

Health Record ◽

Medicare Beneficiaries ◽

Readmission Rates ◽

Heart Failure Patients ◽

Electronic Health

SummaryBackground: Regulatory standards for 30-day readmissions incentivize hospitals to improve quality of care. Implementing comprehensive electronic health record systems potentially decreases readmission rates by improving medication reconciliation at discharge, demonstrating the additional benefits of inpatient EHRs beyond improved safety and decreased errors.Objective: To compare 30-day all-cause readmission incidence rates within Medicare fee-for-service with heart failure discharged from hospitals with full implementation levels of comprehensive EHR systems versus those without.Methods: This retrospective cohort study uses data from the American Hospital Association Health IT survey and Medicare Part A claims to measure associations between hospital EHR implementation levels and beneficiary readmissions. Multivariable Cox regressions estimate the hazard ratio of 30-day all-cause readmissions within beneficiaries discharged from hospitals implementing comprehensive EHRs versus those without, controlling for beneficiary health status and hospital organizational factors. Propensity scores are used to account for selection bias.Results: The proportion of heart failure patients with 30-day all-cause readmissions was 30%, 29%, and 32% for those discharged from hospitals with full, some, and no comprehensive EHR systems. Heart failure patients discharged from hospitals with fully implemented comprehensive EHRs compared to those with no comprehensive EHR systems had equivalent 30-day readmission incidence rates (HR = 0.97, 95% CI 0.73 – 1.3)Conclusions: Implementation of comprehensive electronic health record systems does not necessarily improve a hospital’s ability to decrease 30-day readmission rates. Improving the efficiency of post-acute care will require more coordination of information systems between inpatient and ambulatory providers.Citation: Patterson ME, Marken P, Zhong Y, Simon SD, Ketcherside W. Comprehensive electronic medical record implementation levels not associated with 30-day all-cause readmissions within Medicare beneficiaries with heart failure. Appl Clin Inf 2014; 5: 670–684http://dx.doi.org/10.4338/ACI-2014-01-RA-0008

Download Full-text

Do Hemoglobin and Creatinine Clearance Affect Hospital Readmission Rates from a Skilled Nursing Facility Heart Failure Rehabilitation Unit?

Journal of the American Medical Directors Association ◽

10.1016/j.jamda.2007.12.004 ◽

2008 ◽

Vol 9 (3) ◽

pp. 194-198 ◽

Cited By ~ 9

Author(s):

Umesh Tamhane ◽

John Voytas ◽

Rabeea Aboufakher ◽

Michael Maddens

Keyword(s):

Heart Failure ◽

Creatinine Clearance ◽

Hospital Readmission ◽

Skilled Nursing Facility ◽

Readmission Rates ◽

Rehabilitation Unit ◽

Nursing Facility ◽

Skilled Nursing ◽

Hospital Readmission Rates

Download Full-text

Abstract 136: The Relationship of Changing Hospital Readmission Rates and Mortality Rates After Hospitalization for Heart Failure, Acute Myocardial Infarction, and Pneumonia

Circulation Cardiovascular Quality and Outcomes ◽

10.1161/circoutcomes.10.suppl_3.136 ◽

2017 ◽

Vol 10 (suppl_3) ◽

Author(s):

Kumar Dharmarajan ◽

Yongfei Wang ◽

Susannah Bernheim ◽

Zhenqiu Lin ◽

Leora Horwitz ◽

...

Keyword(s):

Myocardial Infarction ◽

Heart Failure ◽

Acute Myocardial Infarction ◽

Hospital Readmission ◽

Mortality Rates ◽

Hospital Length Of Stay ◽

Readmission Rates ◽

Hospital Readmission Rates ◽

Monthly Changes ◽

Monthly Change

Background: It is unknown if financial pressures to reduce hospital readmission rates following passage of the Affordable Care Act (ACA) have had the unintended effect of increasing mortality rates after hospitalization. We therefore examined correlations between paired changes in hospital 30-day readmission rates and 30-day mortality rates among Medicare fee-for-service beneficiaries hospitalized with heart failure (HF), acute myocardial infarction (AMI), or pneumonia from 2008 to 2014. Methods: We used linear regression to calculate monthly changes in hospitals’ 30-day risk-adjusted readmission rates (RARRs) and 30-day risk-adjusted mortality rates (RAMRs) after discharge for HF, AMI, and pneumonia from 2008 to 2014. Adjustment was made for patient age, sex, comorbidities, hospital length of stay, and season. We then examined the correlation of hospitals’ paired monthly changes in 30-day RARRs and monthly changes in 30-day RAMRs after discharge. Results: From 2008 to 2014, we identified 2,962,554, 1,229,939, and 2,544,530 hospitalizations for HF, AMI, and pneumonia at 5,016, 4,772, and 5,057 hospitals, respectively. Hospital 30-day RARRs declined for all three conditions from 2008 to 2014; the monthly change in RARRs was -0.053 (95% CI -0.055, -0.051) for HF, -0.044 (95% CI -0.047, -0.041) for AMI, and -0.033 (95% CI -0.035, -0.031) for pneumonia. In contrast, the monthly change in hospital 30-day RAMRs after discharge varied by admitting condition and was 0.008 (95% CI 0.007, 0.010) for HF, -0.003 (95% CI -0.006, -0.001) for AMI, and 0.001 (95% CI -0.001, 0.003) for pneumonia. The correlation between monthly changes in hospitals’ 30-day RARRs and 30-day RAMRs after discharge was 0.060 for HF (p<0.001), 0.059 for AMI (p=0.003), and 0.106 for pneumonia (p<0.001). Representative data showing the poor correlation in hospitals’ paired monthly changes in 30-day RARRs and 30-day RAMRs for AMI is shown in the Figure. Conclusion: Changes in hospital readmission rates for HF, AMI, and pneumonia were poorly correlated with changes in mortality rates after hospitalization between 2008 and 2014. These findings suggest that financial incentives to improve hospitals’ readmission performance have not increased mortality after hospitalization.

Download Full-text