Longitudinal retention of families in the assessment of a prevention program targeting adolescent alcohol and tobacco use: the utility of an ecological systems framework.

Stancliffe et al. (2007) followed families in a teen drug-prevention study for three years. They asked: who keeps coming back for follow-up visits?

The team looked at child traits, family life, and neighborhood factors. They wanted to know which ones predicted staying in the study.

Only family and child factors mattered. Things like parent age, child mood, and home routines predicted who stayed.

School support, church ties, or crime rates in the area had no effect. Community variables did not predict retention.

Capio et al. (2013) saw the same pattern in PCIT. Families with three or more risk factors dropped out sooner. Both studies say: look at family load, not zip code.

Poppes et al. (2010) looked at early-intervention dropouts. They found both family and concrete barriers like no car or lost Medi-Cal. Their mix of predictors seems to clash with J et al., but P et al. studied babies and toddlers. Younger families need rides; teen parents may not.

Schinke et al. (2006) ran a similar teen drug-prevention trial. They kept most families for four years, yet never tested which variables drove retention. J et al. fills that gap.

When you run a long parent group, pour effort into home-level supports. Offer flexible meeting times, childcare, or brief check-ins with parents. Do not spend energy on flyers to the whole neighborhood or school assemblies. Target the family unit and you will keep more participants and collect better data.