Predictors of participant dropout at various stages of a pediatric lifestyle program.

OBJECTIVE: To evaluate baseline predictors of drop out at various stages in a lifestyle intervention for overweight and obese children. PATIENTS AND METHODS: Children and their families (N = 248) (aged 8-14 years) attended a lifestyle intervention. At baseline, we assessed anthropometric and demographic data, measured competence and behavioral problems, and family functioning. Dropout rates were analyzed at various stages in treatment with logistic regression analyses.
RESULTS: Children who had mothers of non-white descent, who had higher BMI SDS, who participated in fewer activities, who did not have breakfast regularly, and who did not live in families with a static adaptability structure were more likely to drop out between 0 and 12 months. Different characteristics predicted dropout at various stages of treatment: (1) having an ethnic minority status and being older predicted dropping out between 0 and 3 months; (2) having a nonwhite mother, participating in fewer activities, having higher delinquency scores, and not presenting the family as extremely positive predicted dropping out between 3 and 9 months; and (3) having a higher BMI SDS, having fewer social problems, and not living in families with a static adaptability structure predicted dropping out between 9 and 12 months of treatment.
CONCLUSIONS: The results indicate different characteristics predict dropping out from a pediatric lifestyle program at various stages in treatment. These findings highlight the need for tailored interventions that target different characteristics at various stages of treatment to reduce drop out rates.