Sensitivity analysis of weight reduction results of an observational cohort study in overweight and obese children and adolescents in Germany: the evakuj study.

In the German EvAKuJ observational cohort study, changes in the body mass index standard deviation score (BMI-SDS) of overweight and obese children and adolescents as primary outcome of multimodal (short, inpatient or long, outpatient) weight-loss interventions are difficult to interpret. Published intention-to-treat (ITT) and per protocol data obtained at the end of the intervention (T1), one year (T2), and two years (T3) after its end were used for sensitivity analysis of treatment success rates. The odds ratio and the number needed to treat (NNT) for BMI-SDS reduction of at least -0.2 (successful treatment) and at least -0.5 (good treatment success) were related to spontaneous BMI-SDS reduction rates in a hypothetical control group (control event rate, CER). At T1, treatment seems to be effective up to a CER of 10% in inpatients and of 5% in outpatients. ITT analysis, compromised by a loss to follow-up of 81 to 90% (inpatients) and 57 to 66% (outpatients), indicated that treatment may become less effective at a CER above 1% in inpatients (e.g., successful treatment at T2: NNT=106, at T3: NNT=51), and above 5% in outpatients (successful treatment at T2: NNT=7, at T3: NNT=8; good treatment success at T2 and T3: NNT=25). Positive short-term effects of inpatient treatment of overweight and obese children and adolescents may not be maintained in the long term. Long-term effectiveness of outpatient treatment may depend on age and the degree of overweight.