Cluster Randomized Controlled Trial of Intervention to Increase Participant Retention and Completed Home Visits in the Nurse-Family Partnership

We conducted a cluster-based randomized controlled trial of an intervention designed to improve participant retention in community replication sites of the Nurse-Family Partnership (NFP). We registered 26 sites and randomized them into three groups: retention intervention (RI, N = 9), delayed RI (DRI, N = 6), or control (C, NFP as usual, N = 11). The RI consisted of training nurses to give more explicit control over the frequency of visits and content of the program to the parent participants. Two of the sites assigned to the RI, two assigned to the DRI, and two out of four nurses in one other site assigned to the DRI chose not to participate in the intervention. Primary analyses (intention to treat) contrasted changes in participant retention and completed visits (the primary outcomes) in the two intervention groups (RI and DRI) compared to control sites, focusing on differences in performance among baseline cohorts compared to cohorts enrolled during the first year during which the retention intervention was implemented. Compared to baseline, retention declined in the control sites over time but stayed the same in the RI and DRI sites (p value for interaction = 0.099). Compared to baseline, the number of completed home visits declined over time in the control sites but did not in the RI and DRI sites, producing a significant treatment difference in change in mean completed home visits over time (2.71 visits, SE = 1.164, p = 0.020). The intervention offset a decline over time in retention and completed home visits found in the control group during the time covered by this trial. Quantitative and qualitative evaluation of the intervention indicated that improvements are needed to promote its uptake.