Profiles of Risk Among Enrollees in the Strengthening Families Program: For Parents and Youth Ages 10-14

Introduction: Family-based preventive interventions for adolescent substance use have been effective in reducing substance use initiation and escalation in adolescence (Van Ryzin & Fosco, 2016). Yet, the reach of family-based programs often suffers, as low enrollment and program drop-out are common. Some have argued that families who are most likely to enroll in interventions are at lower risk and therefore have less need for the intervention (Bauman et al., 2001; Offord, 2000). Others have found that higher risk families are more likely to engage, as they may perceive more potential benefits of participation (Dumas et al., 2007; Spoth et al., 2000). Prior studies have predominantly used variable-centered approaches to indicate that different risk factors predict intervention engagement. Yet, individual predictors may cluster around specific types of families, leaving it unclear if there are underserved subgroups who are at elevated risk. We aimed to (1) identify subgroups of family risk in a sample offered a family-based intervention program, and (2) assess whether subgroup membership predicts the probability of enrollment and engagement in the intervention program. We used a person-centered approach to understand how family-level risk factors work together and go on to influence intervention engagement.

Method: Data were collected from 6,055 rural adolescents (Mean age = 11.8, 86% White, 50% female) in the fall of Grade 6, prior to a family-based intervention program. Youth reported on their deviant behavior and their experiences of parental monitoring, consistent discipline, harsh discipline, family conflict, family climate, and parental warmth in their families. Attendance data were used to identify the 1,003 families who enrolled in the family-based intervention program, Strengthening Families Program: For Parents and Youth Ages 10-14 (SFP 10-14).

Results: Four latent classes of family risk were identified: (1) Low Risk (41% prevalence); (2) Ambivalent (30%), characterized by inconsistency across indicators; (3) Harsh (16%), characterized by harsh discipline, family conflict, and elevated adolescent deviant behaviors; and (4) Poor Family Climate (13%), characterized by negative family climate and elevated adolescent deviant behaviors. Latent class membership was unassociated with both the probability of enrolling in SFP 10-14 and having high program attendance.

Conclusions: This study expands upon previous research on family engagement in prevention programs for adolescent substance use. Findings indicate that all groups were equally likely to enroll or engage in SFP 10-14, regardless of their risk type, implying that no families are underserved. Implications for program recruitment will be discussed.