The Moderating Effects of Group Cohesion On the Association Between Adherence and Attendance Among Mexican-American Mothers Enrolled in a Preventive Intervention

Introduction: Despite the efficacy of evidence-based programs (EBP), adoption of EBPs face many challenges. One significant challenge is provider resistance to adoption of EBPs (Nelson, Steele, & Mize, 2006). One concern voiced by providers is that adherence to program manuals is incompatible with positive therapeutic process, which is critical to participant engagement (Nelson et al., 2006). Despite provider concerns, very few studies have examined how these 2 dimensions of implementation- adherence and process quality-interact to influence EBPs. The current study addresses this gap. Among Mexican-origin mothers enrolled in an efficacious 9-session, group-based preventive intervention (Gonzales, et al., 2008), we tested the hypothesis that high levels of adherence are positively associated with participant engagement under conditions of high process quality but negatively associated with engagement when process quality is low. 

Methods:Process quality was operationalized as group cohesion, an important dimension of group process (Burlingame, McClendon, & Alonso; 2011), and assessed using mother-report (n=295) of an adaptation of the Moos Group Environment Scale (Moos, 1984). Adherence was assessed at the group (n=23) level and was the percent of session content delivered as prescribed in the program manual averaged across all sessions. We operationalized engagement as number of sessions attended. We used Hierarchical Linear Modeling (HLM 6.0; Raudenbush et al. 2004) to test the moderating effects of group cohesion (Level 1) on the association between adherence (Level 2) and attendance.

Results:  Group-level adherence interacted with group cohesiveness (b = -12.09, p< .05). Simple slopes were graphed using the HLM Model Graph function at 1SD above and below the group cohesion mean. Contrary to our hypothesis, increased adherence was associated with increases in attendance at low levels of group cohesion but with decreases in attendance when mothers’ perceptions of group cohesion were high. However, at high levels of group cohesion mothers attended more sessions than when cohesion was low, irrespective of adherence.

Conclusions:  Results highlight the importance of studying how multiple dimensions of implementation function conjointly to influence EBPs and indicate that balancing adherence and process quality in the delivery of EBPs is important. Findings also suggest that provider perspectives are critical towards the development and implementation of effective and sustainable EBPs and that provider training to deliver EBPs must focus not only on adherence but also on how content is delivered. Future research should examine the threshold level of fidelity required for program effects without compromising process quality.