Screening by social workers in medical patients with risk of post-acute care needs : a stepped wedge cluster randomized trial

Background : Elderly patients often need post-acute care after hospital discharge. Involvement of social workers can positively affect the discharge planning process. Aim : To investigate the effect of screening patients at risk for post-acute care needs by social workers on time with respect to social workers’ notification, length of stay and delays in discharge compared to usual care. Methods : Cluster randomized stepped wedge trial design for five clusters (wards) and two steps (control to intervention) was used. A total of 400 patients (200 per period) with high risk of post-acute care needs (defined as Post-Acute Care Discharge score, PACD ≥ 7) were included. Social workers performed a screening to decide about self-referral to their services (intervention period), which was compared to a highly structured standard SW notification by physicians and nurses (control period). A Generalized Estimating Equations model adjusted the clustering and baseline differences. Results : A total of 139 patients were referred to social services (intervention: n = 76; control: n = 63). Time to social workers’ notification was significantly shorter in the intervention period when adjusted for all the differences in baseline (Mdn 1.2 vs 1.7, Beta = -0.73, 95%-CI 1.39 to -0.09). Both the length of stay and the delayed discharge time in nights showed no significant differences (Mdn 10.0 vs 9.1, Beta = -0.12, 95%-CI 0.46 to .22 nights 95%-CI, resp. Mdn 0.0 vs 0.0, Beta = .1 1, 95%-CI -0.64 to 0.86). Conclusion : Screening speeded up social workers’ notification but did not accelerate the discharge processes. The screening by social workers might show process improvement in settings with less structured discharge planning