Retention before and during the COVID-19 Pandemic in a Child Maltreatment Prevention Study

Background and Purpose: Child maltreatment has detrimental effects on life-course outcomes including developmental delays, physical and mental health issues, and maladaptive adult functioning. Community building and connection to services can have a protective impact for families at risk for child maltreatment. Project SURROuND is a hospital-based intervention aimed at reducing child maltreatment risk through identification and connection to community resources. Approximately a year into recruitment, COVID-19 required all in-person program activities to shift to virtual and phone contact, raising concerns about retention rate stability. The purpose of this analysis is to 1) measure retention across two phases of a child maltreatment prevention study before and during the COVID-19 pandemic, and 2) examine retention rates stratified by race/ethnicity and depression scores.
Methods: Data include birth record and clinical assessments from an ongoing longitudinal cohort study of mothers and newborns. Thus far, 241 women/infant dyads have been recruited from the antenatal and mother/baby units at a university-affiliated hospital in a Midwest urban area. Eligible participants were: 1) English speaking without a translator; 2) residents of three hospital-adjacent counties; and 3) not enrolled in certain other ongoing hospital research studies. Phase 1 consisted of permission to access medical records and the completion of five assessments. Phase 2 consisted of four assessments and quarterly follow-up phone calls. Analysis included crosstabulations and chi-square tests to measure retention 1) before and during the pandemic and 2) its relationship with both race/ethnicity and depression.
Results: Of the 241 Phase 1 participants, 160 enrolled into Phase 2 (66.39%), 7 withdrew (2.90%), 14 declined (5.81%), 49 were lost to follow-up (20.33%), 11 are in process (4.56%). 107 participants completed all Phase 2 assessments (67% of Phase 2). Pre-pandemic, 159 participants enrolled in Phase 1, and 66 participants enrolled in Phase 2. During the pandemic, an additional 82 participants enrolled in Phase 1 and 94 enrolled in Phase 2. Participants who enrolled in Phase 2 identified as Black (n=106), white (n=28), Latina (n=6), AI/AN (n=2), Asian (n=2), and more than one race (n=5). Chi-square analyses for Phase 2 found statistically significant, positive relationships between completion of Phase 2 assessments and high depression scores, χ2(1, N=164)=4.35, p=.04. There were no significant findings for Phase 2 retention or assessment completion and Phase 1 depression scores alone, depression in both phases, by race/ethnicity, or movement through the phases before/during the pandemic.
Conclusion and Implications: There were no discernable relationships between race/ethnicity, Phase 1 depression, and retention or assessment completion for Phase 2. Those who met the clinical threshold for depression in Phase 2 were more likely to complete Phase 2 assessments. This preliminary analysis indicates potential to use this model to engage at-risk families, in-person or virtually, longitudinally and through pandemic conditions at similar rates across racial/ethnic populations. This intervention could be especially efficacious in engaging mothers at risk for child maltreatment with depression. Findings also suggest that studies of such programs need not be paused if in-person interaction is infeasible. Future analyses should compare longer term outcomes by pre- or post-pandemic enrollment.