[The Costs of Overweight and Obesity: a Systematic Review]. / Die Kosten von Übergewicht und Adipositas in Deutschland ­ ein systematischer Literaturüberblick.

ABSTRACT

AIM OF THE STUDY The aim of this paper was to conduct a systematic review of cost-of-illness studies for overweight and adiposity in Germany. METHODS: We conducted a PubMed search to identify relevant studies. To increase comparability, all cost data were inflated to 2014 prices. For bottom-up studies, we additionally calculated relative cost-differences between normal weight and overweight, as well as adiposity. These relative differences were pooled using meta-analytical techniques and extrapolated to the German population. RESULTS: We identified 15 bottom-up studies, 6 top-down studies and 2 Markov simulations. On average, top-down studies reported direct costs of 7.9 billion Euros and indirect costs of 3.6 billion Euros. Due to between-study heterogeneity, we were not able to pool absolute costs reported in bottom-up studies. The pooled relative cost-differences for studies conducted with adult persons were +22% for the difference between normal weight and overweight and +53% for the difference between normal weight and adiposity. The corresponding effect sizes were 0.07 (- 0.05; 0.19) and 0.15 (0.02; 0.28), respectively. In studies conducted with children and adolescents, relative cost-differences were considerably smaller with corresponding effect sizes being close to zero. Extrapolation of relative pooled cost-differences from bottom-up studies to the German total population yielded direct and indirect excess costs of 22.2 billion Euros for overweight and 23.0 billion Euros for adiposity. CONCLUSION: We found substantial heterogeneity of cost findings between studies for top-down and bottom-up studies as well, which points to substantial uncertainty and strongly hampers clear statements about the costs of overweight and adiposity. Our findings imply that costs are underestimated by top-down studies, whereas overestimated by our extrapolation. The true costs presumably lie between the findings of both approaches.