Background: Long-term outcome after lifestyle interventions in obese children is largely unknown but important to improving intervention.
Objective: The aim was to identify predictors of long-term changes in body mass index (BMI) after lifestyle intervention.
Design: Annual changes in the BMI SD score (BMI-SDS) over 5 y in 663 obese children (aged 4-16 y) motivated to participate in an outpatient lifestyle intervention were analyzed. Child-specific longitudinal curves based on multilevel growth curve models (MLMs) over 5 y were estimated depending on patient characteristics (age and sex).
Results: The mean decrease in BMI-SDS was 0.36 (95% CI: 0.33, 0.39) at the end of the 1-y intervention and 0.46 (95% CI: 0.36, 0.55) 4 y after the intervention. Change in BMI-SDS in the intervention period predicted long-term outcome after 5 y (P < 0.001). MLMs identified age but not sex as a predictor of the outcome: the youngest children (<8 y) at the onset of the intervention had the greatest decrease in BMI-SDS over 5 y, and the oldest children (>13 y) had the least decrease in BMI-SDS (P < 0.05). Whereas there was a larger reduction in BMI-SDS during the intervention in children aged 8-10 y than in children aged 11-12 y, long-term decrease in BMI-SDS was greater in 11-12-y-old children (P < 0.001).
Conclusions: Younger age was associated with the best long-term outcome after participation in the lifestyle intervention, which supports the need for early intervention in childhood obesity. Children aged 8-10 y may need modified intervention, because BMI-SDS increased more in the older children in the long term. However, mean BMI-SDS was significantly lower 4 y after the end of the intervention than at baseline in all age groups. This study was registered at clinicaltrials.gov as NCT00435734.