Can we do more to prevent readmissions after hospital discharge?
Comprehensive and highly-supportive discharge interventions – aimed at enhancing patients’ capacity for self-care – are more effective at reducing early hospital admissions (<30days) than interventions without these features.
Ward staff strive hard to prepare patients well for discharge and many patients transition safely from hospital to home. However, some patients contact you after discharge for advice about their care, medications, and potentially serious symptoms. You feel sure this advice has prevented readmissions and question whether more discharge support (such as telephone follow-up or home visits) is required to help patients transition home safely.
Among adult patients discharged home from hospital, does a highly supportive discharge intervention reduce early readmissions compared with standard discharge planning?
PubMed Clinical Queries (therapy, broad): hospital readmission AND discharge interventions.
Leppin, A. L., Gionfriddo, R.., Kessler, M., Brito, J. P., Mair, F. S., Gallacher, K., . . . Montori, V. M. (2014). Preventing 30-Day Hospital Readmissions: A Systematic Review and Meta-analysis of Randomised Trials. JAMA Intern Med. doi: 10.1001/jamainternmed.2014.1608
A systematic review testing the effects of discharge interventions to reduce early (<30 days) hospital
readmissions. The cumulative complexity model (CuCoM) was used to help identify features within each intervention that might explain their effect on readmission rate.
Inclusion criteria were:
Type of study was randomised controlled trials involving adults admitted from the community to an inpatient ward for at least 24 hours with a medical or surgical cause. Obstetric or psychiatric admissions, and discharges to skilled nursing facilities, were excluded;
Interventions focused on the hospital-to-home transition and were generalisable to contexts beyond a single patient diagnosis. Discharge activities occurring in the intervention but not the comparison group were coded by their unique activity (e.g. home visit, formal discharge planning, timely follow-up). Intervention features, such as the extent they affected patient workload or capacity to self-care, were also rated;
Outcomes were unplanned or all-cause readmission with or without out-of-hospital deaths at 30 days from discharge. Pre-specified sub-group analyses included patient and intervention characteristics, and publication year.
The Search Strategy involved searching electronic databases PubMed, Ovid MEDLINE, Ovid EMBASE, EBSCO CINAHL, and Scopus (1990 until April 1, 2013). Unpublished studies located by hand-searching bibliographies of included studies and recent reviews. Experts in the field contacted. Language of publication restricted to English and Spanish. Studies were selected by independent screening of titles/abstracts, and then full texts. A piloted, standardised form was used for risk of bias assessment and data extraction. Conflicts resolved by consensus. Reviewers were blinded to trial results systematically coded and rated discharge interventions (framework, definitions supplied); reviewer agreement checked and study authors confirmed fidelity. The risk of bias assessment was based on the Cochrane Collaboration tool. Studies obtaining readmission data from internal health systems records only were considered to have a high risk of bias. The publication bias was assessed as small, the effect unknown. Overall this was a high quality review of mainly good quality studies.
From the 1135 titles screened, 256 full texts were reviewed, and 47 trials met the criteria to be included in this review. Interventions occurred in both inpatient and outpatient settings, and varied in complexity but commonly involved activities of case management, patient education, home visits and self-management promotion. Meta-analysis of 42 trials found that highly supportive interventions reduced 30 day readmission rates by nearly 20 per cent (table) compared with baseline discharge activity: this was regardless of age (≥ 65 years or not), diagnosis (heart failure or not), and hospital ward (medical or not). More effective interventions were those rated to increase patients’ capacity for self-care, involved ≥ 5 unique activities, and involved ≥ 2 individuals in a structured and required role. Readmission rates were not associated with delivery setting (an inpatient and outpatient component, or not) or the effect on patients’ workload. Post hoc analysis showed increased effectiveness with interventions providing comprehensive support (table).
The significance of this review lies in its use of the CuCoM model to explore and highlight the underlying features of what makes an effective discharge intervention. This patient-centred model predicts that readmissions may be avoided by interventions that balance the patients’ workload (related to accessing care and adhering to complex care plans) with their capacity to look after themselves at home (determined by the quality and availability of resources).
The findings provide objective evidence of the benefits of a commonsense approach to effective discharge interventions – i.e. targeting identified barriers to self-care with comprehensive support. Understanding these barriers from the patient and caregivers’ perspective, especially for those most at risk (ethnic groups, complex conditions), is an essential step towards addressing this important healthcare issue.
Table. Results with 95% Confidence Intervals (CI)
within 30 days
|I2 = 31%*|
*low to moderate ** variable developed from subgroup analyses findings and CuCoM prediction
Cynthia Wensley RN MHSc is a PhD candidate who also works at the School of Nursing, The University of Auckland as an Honorary professional teaching fellow.