Developing and testing a comprehensive tool to assess family meetings: Empirical distinctions between high- and low-quality meetings.

BACKGROUND: The heterogeneity with regard to findings on family meetings (or conferences) suggests a need to better understand factors that influence family meetings. While earlier studies have explored frequency or timing of family meetings, little is known about how factors (such as what is said during meetings, how it is said, and by whom) influence family meeting quality. OBJECTIVES: (1) To develop an evaluation tool to assess family meetings (Phase 1); (2) to identify factors that influence meeting quality by evaluating 34 family meetings (Phase 2). MATERIALS AND METHODS: For Phase 1, methods included developing a framework, cognitive testing, and finalizing the evaluation tool. The tool consisted of Facilitator Characteristics (i.e., gender, experience, and specialty of the person leading the meeting), and 22 items across 6 Meeting Elements (i.e., Introductions, Information Exchanges, Decisions, Closings, Communication Styles, and Emotional Support) and sub-elements. For Phase 2, methods included training evaluators, assessing family meetings, and analyzing data. We used Spearman's rank-order correlations to calculate meeting quality. Qualitative techniques were used to analyze free-text. RESULTS: No Facilitator Characteristic had a significant correlation with meeting quality. Sub-elements related to communication style and emotional support most strongly correlated with high-quality family meetings, as well as whether "next steps" were outlined (89.66%) and whether "family understanding" was elicited (86.21%). We also found a significant and strong positive association between overall proportion scores and evaluators' ratings (rs=0.731, p<0.001). CONCLUSIONS: We filled a gap by developing an evaluation tool to assess family meetings, and we identified how what is said during meetings impacts quality.

What Social Bonds have the Greatest Influence on Patterns of Substance Use among Child-Welfare-Involved Youth?

Child-welfare-involved youth may lack protective social bonds that could reduce their risk of substance use. We investigated whether caregiver, school, or peer bonds predict distinct patterns of substance use among child-welfare-involved youth. The sample included 720 participants in the National Survey of Child and Adolescent Well-Being II. Latent class analysis (LCA) and the three-step approach to incorporate indicator variables onto the latent classes were used. We found the following classes: (1) severe polysubstance use; (2) moderate polysubstance use; and (3) low use. Youth bonded to primary caregivers were less likely to be severe polysubstance drug users, but caregiver bonds did not protect against moderate polysubstance use. School bonds protected against severe polysubstance and moderate polysubstance. Youth bonded to deviant peers were more likely to be in the severe polysubstance use and moderate polysubstance use classes. Interventions targeting child-welfare-involved youth need to account for social bonds' effect on substance use.