Predicting weight gain in adults – results from a multi‐center cohort approach

Abstract

Listen

Translate article

The aim of the study was to develop a risk score for substantial weight gain among German adults. Substantial weight gain was defined as gaining ≥10% of weight between baseline and end of follow‐up. Information on 15 variables (age, sex, baseline weight, education, sport smoking, alcohol, sleep disorders, fruits and vegetable, meat, fish, whole‐grain, cake and cookies, chocolate, soft drinks) from 32,204 participants of five population‐based German cohort studies was used for the prediction model. All participants had a follow‐up time of more than 6 years. The cohort‐specific background survival probabilities at 5 years were estimated based on the average value of each predictor over all individuals of the five cohorts and subsequently meta‐analytically combined. The meta‐analysed beta‐coefficients of the predictors were transformed into score points. Cross‐validation was used to evaluate the score's discriminatory accuracy. On average, study participants gained 299g to 429g weight per year, depending on the cohort. With the 15 variables, a cross‐validated c index (95% CI) was 0.71 (0.67 to 0.75) obtained. Baseline age and weight most strongly determined risk. A cut‐off value of ≥475 score points yielded a sensitivity of 71% and a specificity of 63%. The corresponding positive and negative predictive values were 10.4% and 97.6%, respectively. The study showed that substantial weight gain can moderately be predicted. In view of the low prediction ability, the rate of false positives could be high if a high sensitivity is aspired. Nevertheless, the proposed risk score could support health care providers in a proper decision‐making and also facilitate an efficient selection of subjects into intervention trials.Support or Funding Information“Kompetenznetz Adipositas (Competence Network Obesity)” funded by the Federal Ministry of Education and Research (FKZ: 01GI1121B).