Risk factors for nonelective hospital readmissions

Published models that predict hospital readmission: a critical appraisal

INTRODUCTION

The number of readmission risk prediction models available has increased rapidly, and these models are used extensively for health decision-making. Unfortunately, readmission models can be subject to flaws in their development and validation, as well as limitations in their clinical usefulness.

OBJECTIVE

To critically appraise readmission models in the published literature using Delphi-based recommendations for their development and validation.

METHODS

We used the modified Delphi process to create Critical Appraisal of Models that Predict Readmission (CAMPR), which lists expert recommendations focused on development and validation of readmission models. Guided by CAMPR, two researchers independently appraised published readmission models in two recent systematic reviews and concurrently extracted data to generate reference lists of eligibility criteria and risk factors.

RESULTS

We found that published models (n=81) followed 6.8 recommendations (45%) on average. Many models had weaknesses in their development, including failure to internally validate (12%), failure to account for readmission at other institutions (93%), failure to account for missing data (68%), failure to discuss data preprocessing (67%) and failure to state the model's eligibility criteria (33%).

CONCLUSIONS

The high prevalence of weaknesses in model development identified in the published literature is concerning, as these weaknesses are known to compromise predictive validity. CAMPR may support researchers, clinicians and administrators to identify and prevent future weaknesses in model development.

Physical Function Does Not Predict Care Assessment Need Score in Older Veterans

Objective: The Veterans Health Administration's Care Assessment Need (CAN) score is a statistical model, aimed to predict high-risk patients. We were interested in determining if a relationship existed between physical function and CAN scores. Method: Seventy-four older (71 ± 1 years) male Veterans underwent assessment of CAN score and subjective (Short Form-36 [SF-36]) and objective (self-selected walking speed, four square step test, short physical performance battery) assessment of physical function. Results: Approximately 25% of participants self-reported limitations performing lower intensity activities, while 70% to 90% reported limitations with more strenuous activities. When compared with cut points indicative of functional limitations, 35% to 65% of participants had limitations for each of the objective measures. Any measure of subjective or objective physical function did not predict CAN score. Conclusion: These data indicate that the addition of a physical function assessment may complement the CAN score in the identification of high-risk patients.

The Impact of Disability and Social Determinants of Health on Condition-Specific Readmissions beyond Medicare Risk Adjustments: A Cohort Study

BACKGROUND

Readmission rates after pneumonia, heart failure, and acute myocardial infarction hospitalizations are risk-adjusted for age, gender, and medical comorbidities and used to penalize hospitals.

OBJECTIVE

To assess the impact of disability and social determinants of health on condition-specific readmissions beyond current risk adjustment.

DESIGN, SETTING, AND PARTICIPANTS

Retrospective cohort study of Medicare patients using 1) linked Health and Retirement Study-Medicare claims data (HRS-CMS) and 2) Healthcare Cost and Utilization Project State Inpatient Databases (Florida, Washington) linked with ZIP Code-level measures from the Census American Community Survey (ACS-HCUP). Multilevel logistic regression models assessed the impact of disability and selected social determinants of health on readmission beyond current risk adjustment.

MAIN MEASURES

Outcomes measured were readmissions ≤30 days after hospitalizations for pneumonia, heart failure, or acute myocardial infarction. HRS-CMS models included disability measures (activities of daily living [ADL] limitations, cognitive impairment, nursing home residence, home healthcare use) and social determinants of health (spouse, children, wealth, Medicaid, race). ACS-HCUP model measures were ZIP Code-percentage of residents ≥65 years of age with ADL difficulty, spouse, income, Medicaid, and patient-level and hospital-level race.

KEY RESULTS

For pneumonia, ≥3 ADL difficulties (OR 1.61, CI 1.079-2.391) and prior home healthcare needs (OR 1.68, CI 1.204-2.355) increased readmission in HRS-CMS models (N = 1631); ADL difficulties (OR 1.20, CI 1.063-1.352) and 'other' race (OR 1.14, CI 1.001-1.301) increased readmission in ACS-HCUP models (N = 27,297). For heart failure, children (OR 0.66, CI 0.437-0.984) and wealth (OR 0.53, CI 0.349-0.787) lowered readmission in HRS-CMS models (N = 2068), while black (OR 1.17, CI 1.056-1.292) and 'other' race (OR 1.14, CI 1.036-1.260) increased readmission in ACS-HCUP models (N = 37,612). For acute myocardial infarction, nursing home status (OR 4.04, CI 1.212-13.440) increased readmission in HRS-CMS models (N = 833); 'other' patient-level race (OR 1.18, CI 1.012-1.385) and hospital-level race (OR 1.06, CI 1.001-1.125) increased readmission in ACS-HCUP models (N = 17,496).

CONCLUSIONS

Disability and social determinants of health influence readmission risk when added to the current Medicare risk adjustment models, but the effect varies by condition.

Predicting 30-Day Readmissions in an Asian Population: Building a Predictive Model by Incorporating Markers of Hospitalization Severity

Background

To reduce readmissions, it may be cost-effective to consider risk stratification, with targeting intervention programs to patients at high risk of readmissions. In this study, we aimed to derive and validate a prediction model including several novel markers of hospitalization severity, and compare the model with the LACE index (Length of stay, Acuity of admission, Charlson comorbidity index, Emergency department visits in past 6 months), an established risk stratification tool.

Method

This was a retrospective cohort study of all patients ≥ 21 years of age, who were admitted to a tertiary hospital in Singapore from January 1, 2013 through May 31, 2015. Data were extracted from the hospital’s electronic health records. The outcome was defined as unplanned readmissions within 30 days of discharge from the index hospitalization. Candidate predictive variables were broadly grouped into five categories: Patient demographics, social determinants of health, past healthcare utilization, medical comorbidities, and markers of hospitalization severity. Multivariable logistic regression was used to predict the outcome, and receiver operating characteristic analysis was performed to compare our model with the LACE index.

Results

74,102 cases were enrolled for analysis. Of these, 11,492 patient cases (15.5%) were readmitted within 30 days of discharge. A total of fifteen predictive variables were strongly associated with the risk of 30-day readmissions, including number of emergency department visits in the past 6 months, Charlson Comorbidity Index, markers of hospitalization severity such as ‘requiring inpatient dialysis during index admission, and ‘treatment with intravenous furosemide 40 milligrams or more’ during index admission. Our predictive model outperformed the LACE index by achieving larger area under the curve values: 0.78 (95% confidence interval [CI]: 0.77–0.79) versus 0.70 (95% CI: 0.69–0.71).

Conclusion

Several factors are important for the risk of 30-day readmissions, including proxy markers of hospitalization severity.

Lower handgrip strength at discharge from acute care hospitals is associated with 30-day readmission: A prospective cohort study

BACKGROUND

Malnutrition at admission, using various parameters, is associated with 30-day readmission. However, the association between 30-day readmission and nutritional parameters at discharge has not been studied.

METHOD

From a large cohort study (n = 1022), 413 patients with a length of stay of ≥7 days who had information on readmission and discharge location were included into the analysis. Their nutritional status at discharge was assessed by subjective global assessment, body mass index, albumin, nutritional risk index and handgrip strength. Data on demography, diagnoses and Charlson comorbidity index (CCI) were also collected. Missing data was handled using multiple imputations by chained equations. Association of nutrition related measures with 30 day readmission was tested in logistic regression models.

RESULTS

Of the 413 patients, 86 (20.8%) were readmitted within 30 days. The proportion of readmitted patients was higher for medical (42.2%) versus surgical patients (25.6%) (p = 0.005) and disease severity was higher in the readmission group with (median (q1, q3) CCI of 3 (2, 6) versus 2(1, 4) for no readmission (p = 0.009). Among the nutritional parameters assessed at discharge, only handgrip strength was significantly associated with 30-day readmission both in unadjusted and adjusted models. Stronger handgrip was associated with decreased chances for readmission where adjusted OR (95% CI) per unit increase were 0.95 (0.92, 0.99). Handgrip strength was not associated with disease severity assessed by CCI (p = 0.14) but was significantly associated with SGA (SGA A and B significantly different from SGA C: both p-values <0.001) after adjusting for age and gender.

CONCLUSION

Lower handgrip at discharge was associated with 30-day readmission. This assessment may be useful to detect patients at risk of readmission to better individualize discharge planning including nutrition care.

Association of impaired functional status at hospital discharge and subsequent rehospitalization

OBJECTIVE

To determine whether functional status near the time of discharge from acute care hospitalization is associated with acute care readmission.

PATIENTS AND METHODS

Retrospective cohort study of 9405 consecutive patients admitted from an acute care hospital to an inpatient rehabilitation facility between July 1, 2006 and December 31, 2012. Patients' functional status at admission to the rehabilitation facility was assessed by the Functional Independence Measure (FIM) score, and divided into low, middle, or high functional status. The main outcome was readmission to an acute care hospital within 30 days of acute care discharge (for all patients and by subgroup according to diagnostic group: medical, orthopedic, or neurologic).

RESULTS

There were 1182 (13%) readmissions. FIM score was significantly associated with readmission, with adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for low and middle versus high FIM score category of 3.0 (2.5-3.6; P < 0.001) and 1.5 (95% CI: 1.3-1.8; P < 0.001), respectively. This relationship between FIM score and readmission held across diagnostic category. Medical patients with low functional status had the highest readmission rate (OR: 29%; 95% CI: 25%-32%) and an adjusted OR for readmission of 3.2 (95% CI: 2.4-4.3, P < 0.001) compared to medical patients with high FIM scores.

CONCLUSIONS AND RELEVANCE

For patients admitted to an acute inpatient rehabilitation facility, functional status near the time of discharge from an acute care hospital is strongly associated with acute care readmission, particularly for medical patients with greater functional impairments. Reducing functional status decline during acute care hospitalization may be an important strategy to lower readmissions.

Functional status impairment is associated with unplanned readmissions

OBJECTIVE

To determine whether functional status on admission to a Comprehensive Integrated Inpatient Rehabilitation Program (CIIRP) is associated with unplanned readmission to acute care.

DESIGN

Retrospective cohort study.

SETTING

Academic hospital-based CIIRP.

PARTICIPANTS

Consecutive patients (N=1515) admitted to a CIIRP between January 2009 and June 2012.

INTERVENTIONS

Patients' functional status, the primary exposure variable, was assessed using tertiles of the total FIM score at CIIRP admission, with secondary analyses using the FIM motor and cognitive domains. A propensity score, consisting of 25 relevant clinical and demographic variables, was used to adjust for confounding in the analysis.

MAIN OUTCOME MEASURES

Readmission to acute care was categorized as (1) readmission before planned discharge from the CIIRP, (2) readmission within 30 days of discharge from the CIIRP, and (3) total readmissions from both groups, with total readmissions being the a priori primary outcome.

RESULTS

Among the 1515 patients, there were 347 total readmissions. Total readmissions were significantly associated with FIM scores, with adjusted odds ratios (AORs) and 95% confidence intervals (CIs) for the lowest and middle FIM tertiles versus the highest tertile (AOR=2.6; 95% CI, 1.9-3.7; P<.001 and AOR=1.7; 95% CI, 1.2-2.4; P=.002, respectively). There were similar findings for secondary analyses of readmission before planned discharge from the CIIRP (AOR=3.5; 95% CI, 2.2-5.8; P<.001 and AOR=2.1; 95% CI, 1.3-3.5l P=.002, respectively), and a weaker association for readmissions after discharge from the CIIRP (AOR=1.6; 95% CI, 1.0-2.4; P=.047 and AOR=1.3; 95% CI, 0.8-1.9; P=.28, respectively). The FIM motor domain score was more strongly associated with readmissions than the FIM cognitive score.

CONCLUSIONS

Functional status on admission to the CIIRP is strongly associated with readmission to acute care, particularly for motor aspects of functional status and readmission before planned discharge from the CIIRP. Efforts to reduce hospital readmissions should consider patient functional status as an important and potentially modifiable risk factor.

Influence of a transitional care clinic on subsequent 30-day hospitalizations and emergency department visits in individuals discharged from a skilled nursing facility

OBJECTIVES

To evaluate an intervention to improve care transitions at the time of skilled nursing facility (SNF) discharge.

DESIGN

Natural experiment using a pre-post design.

SETTING

Veterans Affairs hospital, community SNF, and outpatient clinic.

PARTICIPANTS

The pre-intervention group comprised 134 individuals discharged to the community from posthospitalization SNF care, and the intervention group was 217 individuals who received a postdischarge clinic (PDC) intervention at SNF discharge after receiving posthospitalization care at the SNF.

INTERVENTION

This study is a natural experiment using a pre-post design. The intervention was a one-time visit to a PDC before SNF discharge, where an advanced nurse practitioner conducted medication reconciliation, ordered medical supplies and equipment and home health services if needed, provided individual and caregiver education, and communicated the information to the individual's primary outpatient care provider through electronic medical records.

MEASUREMENTS

The pre-PDC and PDC intervention groups were compared on various measures of hospital utilization within 30 days of the SNF discharge (number of rehospitalizations, acute care inpatient days, and emergency department (ED) visits).

RESULTS

Although there was a 23% rehospitalization rate in the pre-PDC group, participants in the PDC intervention group had a 14% rehospitalization rate within 30 days of SNF discharge (P = .02). Those who received the PDC intervention had significantly fewer acute care inpatient days during the 30-day follow-up (P < .001). Although the difference in the number of ED visits between the two groups was not statistically significant, the number of ED visits per 1,000 patient follow-up days during the 30-day interval was significantly lower in the PDC intervention group (P = .03).

CONCLUSION

Comprehensive care coordination at the time of SNF discharge can reduce postdischarge hospital use in settings with shared electronic records.

Risk prediction models for hospital readmission: a systematic review

CONTEXT

Predicting hospital readmission risk is of great interest to identify which patients would benefit most from care transition interventions, as well as to risk-adjust readmission rates for the purposes of hospital comparison.

OBJECTIVE

To summarize validated readmission risk prediction models, describe their performance, and assess suitability for clinical or administrative use.

DATA SOURCES AND STUDY SELECTION

The databases of MEDLINE, CINAHL, and the Cochrane Library were searched from inception through March 2011, the EMBASE database was searched through August 2011, and hand searches were performed of the retrieved reference lists. Dual review was conducted to identify studies published in the English language of prediction models tested with medical patients in both derivation and validation cohorts.

DATA EXTRACTION

Data were extracted on the population, setting, sample size, follow-up interval, readmission rate, model discrimination and calibration, type of data used, and timing of data collection.

DATA SYNTHESIS

Of 7843 citations reviewed, 30 studies of 26 unique models met the inclusion criteria. The most common outcome used was 30-day readmission; only 1 model specifically addressed preventable readmissions. Fourteen models that relied on retrospective administrative data could be potentially used to risk-adjust readmission rates for hospital comparison; of these, 9 were tested in large US populations and had poor discriminative ability (c statistic range: 0.55-0.65). Seven models could potentially be used to identify high-risk patients for intervention early during a hospitalization (c statistic range: 0.56-0.72), and 5 could be used at hospital discharge (c statistic range: 0.68-0.83). Six studies compared different models in the same population and 2 of these found that functional and social variables improved model discrimination. Although most models incorporated variables for medical comorbidity and use of prior medical services, few examined variables associated with overall health and function, illness severity, or social determinants of health.

CONCLUSIONS

Most current readmission risk prediction models that were designed for either comparative or clinical purposes perform poorly. Although in certain settings such models may prove useful, efforts to improve their performance are needed as use becomes more widespread.

Hospital readmission in general medicine patients: a prediction model

BACKGROUND

Previous studies of hospital readmission have focused on specific conditions or populations and generated complex prediction models.

OBJECTIVE

To identify predictors of early hospital readmission in a diverse patient population and derive and validate a simple model for identifying patients at high readmission risk.

DESIGN

Prospective observational cohort study.

PATIENTS

Participants encompassed 10,946 patients discharged home from general medicine services at six academic medical centers and were randomly divided into derivation (n = 7,287) and validation (n = 3,659) cohorts.

MEASUREMENTS

We identified readmissions from administrative data and 30-day post-discharge telephone follow-up. Patient-level factors were grouped into four categories: sociodemographic factors, social support, health condition, and healthcare utilization. We performed logistic regression analysis to identify significant predictors of unplanned readmission within 30 days of discharge and developed a scoring system for estimating readmission risk.

RESULTS

Approximately 17.5% of patients were readmitted in each cohort. Among patients in the derivation cohort, seven factors emerged as significant predictors of early readmission: insurance status, marital status, having a regular physician, Charlson comorbidity index, SF12 physical component score, >or=1 admission(s) within the last year, and current length of stay >2 days. A cumulative risk score of >or=25 points identified 5% of patients with a readmission risk of approximately 30% in each cohort. Model discrimination was fair with a c-statistic of 0.65 and 0.61 for the derivation and validation cohorts, respectively.

CONCLUSIONS

Select patient characteristics easily available shortly after admission can be used to identify a subset of patients at elevated risk of early readmission. This information may guide the efficient use of interventions to prevent readmission.

Risk Factors for Nonelective Hospitalization in Frail and Older Adult, Inner-City Outpatients

PURPOSE

In our study, we sought to improve the accuracy of predicting the risk of hospitalization and to identify older, inner-city patients who could be targeted for preventive interventions.

DESIGN AND METHODS

Participants (56% were African American) in a randomized trial were from a primary care practice and included 1,041 patients living in the inner city who were either > or = 75 years of age or were > or = 50 years of age with severe disease. As a secondary analysis, we assessed patient characteristics at baseline involving five domains of health, including utilization and satisfaction. We followed participants for 12 months and recorded the occurrence of nonelective hospitalization within the study period. We developed a multivariate model using logistic regression to predict this outcome.

RESULTS

The following patient characteristics independently predicted an increased risk for nonelective hospitalization: having the diagnosis of congestive heart failure, diabetes mellitus, or anemia; and having more medications prescribed, having a lower body mass index, and having more emergency department visits during the previous year. Better physical functioning reduced the risk of hospitalization.

IMPLICATIONS

Moderate accuracy of a prediction model (0.73) was observed. In addition to focusing on patients with chronic disease, helping them maintain physical functioning may help reduce nonelective hospitalization.

Effect of discharge summary availability during post-discharge visits on hospital readmission

OBJECTIVE

To determine if the delivery of hospital discharge summaries to follow-up physicians decreases the risk of hospital readmission.

SUBJECTS

Eight hundred eighty-eight patients discharged from a single hospital following treatment for an acute medical illness.

SETTING

Teaching hospital in a universal health-care system.

DESIGN

We determined the date that each patient's discharge summary was printed and the physicians to whom it was sent. Summary receipt was confirmed by survey and phoning each physician's office. Each patient's hospital chart was reviewed to determine their acute and chronic medical conditions as well as their course in hospital. Using population-based administrative databases, all post-hospitalization visits were identified. For each of these visits, we determined whether the summary was available.

MAIN OUTCOME MEASURES

Time to nonelective hospital readmission during 3 months following discharge.

RESULTS

The discharge summary was available for only 568 of 4,639 outpatient visits (12.2%). Overall, 240 (27.0%) of patients were urgently readmitted to hospital. After adjusting for significant patient and hospitalization factors, we found a trend toward a decreased risk of readmission for patients who were seen in follow-up by a physician who had received a summary (relative risk 0.74, 95% confidence interval 0.50 to 1.11).

CONCLUSIONS

The risk of rehospitalization may decrease when patients are assessed following discharge by physicians who have received the discharge summary. Further research is required to determine if better continuity of patient information improves patient outcomes.

Predicting non-elective hospital readmissions: a multi-site study. Department of Veterans Affairs Cooperative Study Group on Primary Care and Readmissions

OBJECTIVE

To determine clinical and patient-centered factors predicting non-elective hospital readmissions.

DESIGN

Secondary analysis from a randomized clinical trial.

CLINICAL SETTING

Nine VA medical centers.

PARTICIPANTS

Patients discharged from the medical service with diabetes mellitus, congestive heart failure, and/or chronic obstructive pulmonary disease (COPD).

MAIN OUTCOME MEASUREMENT

Non-elective readmission within 90 days.

RESULTS

Of 1378 patients discharged, 23.3% were readmitted. After controlling for hospital and intervention status, risk of readmission was increased if the patient had more hospitalizations and emergency room visits in the prior 6 months, higher blood urea nitrogen, lower mental health function, a diagnosis of COPD, and increased satisfaction with access to emergency care assessed on the index hospitalization.

CONCLUSIONS

Both clinical and patient-centered factors identifiable at discharge are related to non-elective readmission. These factors identify high-risk patients and provide guidance for future interventions. The relationship of patient satisfaction measures to readmission deserves further study.

Factors associated with unplanned hospital readmission among patients 65 years of age and older in a Medicare managed care plan

PURPOSE

Unplanned hospital readmission within 30 days of discharge is considered a "sentinel event" for poor quality. Patients at high risk for this adverse event could be targeted for interventions designed to reduce their risk of readmission. The purpose of this study was to identify patient characteristics and risk factors at discharge associated with unplanned readmission within 30 days of hospital discharge.

SUBJECTS AND METHODS

We performed a matched case-control study among patients in a Medicare managed care plan who had been admitted to an academic hospital. The cases were patients aged 65 years or older who were urgently or emergently readmitted to the hospital within 30 days of discharge. One control patient who was not readmitted within 30 days was matched to each case by principal diagnosis. The medical records of the first admission of the cases and the admission of the controls underwent review (blinded to case-control status) to determine the patient's baseline demographic characteristics, comorbid conditions, previous health care utilization, and functional status. The records were also reviewed to assess risk factors on discharge, including clinical instability, inability to ambulate and feed, mental status changes, number of discharge medications, and discharge disposition.

RESULTS

Five factors were independently associated (P < 0.05) with unplanned readmission within 30 days. These included four baseline patient characteristics: age 80 years or older [odds ratio = 1.8; 95% confidence interval (CI), 1.02-3.2], previous admission within 30 days (odds ratio = 2.3; 95% CI, 1.2-4.6), five or more medical comorbidities (odds ratio = 2.6; 95% CI, 1.5-4.7), and history of depression (odds ratio = 3.2; 95% CI, 1.4-7.9); and one discharge factor: lack of documented patient or family education (odds ratio = 2.3; 95% CI, 1.2-4.5).

CONCLUSIONS

If validated, these factors may identify patients at high risk of readmission. They suggest that interventions, such as improved discharge education programs, may reduce unplanned readmission.