[Direct medical costs of (severe) obesity: a bottom-up assessment of over- vs. normal-weight adults in the KORA-study region (Augsburg, Germany)]

Abstract

Aim of the study: To estimate and compare direct medical costs of illness of German adults in different BMI-groups and different degrees of obesity.

Methods: In a sub-sample (n = 947) of the KORA-Survey S4 1999/2001, a cross-sectional health survey of the adult population in the Augsburg region (Germany; age: 25-74), visits to physicians, receipt and purchase of drugs, and inpatient days in hospital were assessed over half a year. Body mass index (BMI in kg/m(2)) was assessed anthropometrically. Respondents in normal weight (18.5 < or = BMI < 25), pre-obese (25 < or = BMI < 30), moderately obese (class 1: 30 < or = BMI < 35), and severely obese (classes 2-3: BMI > or = 35) range were compared in their costs of illness via analyses of covariance and regression analyses based on generalized linear models. Physician visits and inpatient days were evaluated as recommended by the Working Group "Methods in Health Economic Evaluation", and drugs by actual costs. Sex, age, socio-economic status (Helmert-Index), sickness fund (statutory vs. private), and place of residence (Augsburg City vs. District of Augsburg or Aichach-Friedberg) were adjusted for.

Results: While respondents with moderate obesity statistically did not differ significantly in their direct medical costs from those in normal weight or pre-obese range (1,080.14 euro vs. 847.60 euro and 830.59 euro; for users of care: 1,215.55 euro vs. 993.18 euro and 1,003.23 euro [all estimates adjusted and per annum]), those with severe obesity had significantly higher costs (2,572.19 euro; for users of care: 2,964.87 euro). Sub-analyses for individual parameters of health care use revealed that this pattern is largely due to inpatient days in hospital and receipt/purchase of drugs only available on prescription.

Conclusions: On average, results indicate excess direct medical costs primarily in people with severe, and less with moderate obesity. In particular, they underline the need to distinguish moderate vs. severe obesity (classes 1 vs. 2-3) in health economics and health services research.