Implementing & Evaluating a Strengths-Based Social Work Intervention to Reduce Re-Hospitalizations: An RCT

Background and purpose:  The Patient Protection and Affordable Care Act (ACA) views the re-hospitalization of a patient within the first 30 days after initial discharge, as being a result of poor hospital performance and therefore not reimbursable by Medicare (Averill et al., 2009; Brown, 2009).  In response to the threat of losing funds and receiving penalties, hospitals are seeking solutions to reduce readmission rates for the Medicare population.  Multiple studies have been implemented and reveal the role of transitional care as promising (e.g., Hernandez et al, 2010), however results of studies relying on nurses for support at discharge have been mixed (Krumholz et al., 2002; Rau, 2012).  To address this systemic need, we conducted a randomized control trial to examine the impact of follow-up at discharge by MSW interns in one upstate New York hospital.  The intervention relied on a strengths-based model of social work practice and included one phone call and one home visit per Medicare-eligible patient post discharge. 

Methods:  The sample consisted of 107 unduplicated participants who were randomly assigned to either the intervention (n=63) or control (standard of care) (n=44) group. The average age of the participants was 71(±11) years with a range of 51 to 92 years. Fifty participants were women and 57 were men; the average LACE score of all the subjects was 9.6(±2.6) and there was an average of 1.2 (±1.3) emergency room visits per subject in the six months prior to study participation.  All participants in the experimental and control groups received typical and usual medical care. Those participants enrolled in the intervention group received the additional post-discharge services of the MSW intern per the research protocol.

Results:  There were a total of 12 patients across the groups who were readmitted before 30 days.  Four readmissions occurred in the intervention group (4/63) and 8 readmissions were in the control group (8/44).  Contingency analysis revealed that the odds of being readmitted without the intervention was 3.28 (95% C.I. = .92-11.67; p=.056); that is, the patients who did not receive the intervention were over three times more likely to be admitted within 30 days.  Logistic regression revealed that accounting for the effects of gender and age had little effect on the increased odds of admission among the standard of care group (OR=2.93, 95% C.I. = .81-10.67; p=.10).  Analysis of the patients who completed the social work home visit intervention revealed that the difference in the proportion of readmissions (8/44 or 18.2% standard of care Vs. 0/45 or 0.0% social worker contact) would be highly statistically significant (X²=8.99; p=.003).  These findings suggest that the social work home visit was a significant factor in avoiding 30-day hospital readmission.   

Conclusion and Implications:  This study shows that a time-efficient in-home intervention by masters-level social work interns practicing from an empowerment perspective may reduce hospital readmission rates.  Cost analyses reveal that if scaled up and factored in costs of hiring social workers, one hospital system could potentially save 7.5 million dollars annually.