Body weight changes and outpatient medical care utilisation: Results of the MONICA/KORA cohorts S3/F3 and S4/F4

Abstract

Objectives: To test the effects of body weight maintenance, gain, and loss on health care utilisation in terms of outpatient visits to different kinds of physicians in the general adult population.

Methods: Self-reported utilisation data were collected within two population-based cohorts (baseline surveys: MONICA-S3 1994/95 and KORA-S4 1999/2001; follow-ups: KORA-F3 2004/05 and KORA-F4 2006/08) in the region of Augsburg, Germany, and were pooled for present purposes. N=5,147 adults (complete cases) aged 25 to 64 years at baseline participated. Number of visits to general practitioners (GPs), internists, and other specialists as well as the total number of physician visits at follow-up were compared across 10 groups defined by body mass index (BMI) category maintenance or change. Body weight and height were measured anthropometrically. Hierarchical generalized linear regression analyses with negative binomial distribution adjusted for sex, age, socioeconomic status (SES), survey, and the need factors incident diabetes and first cancer between baseline and follow-up were conducted.

Results: In fully adjusted models, compared to the group of participants that maintained normal weight from baseline to follow-up, the following groups had significantly higher GP utilisation rates: weight gain from normal weight (+36%), weight loss from preobesity (+39%), maintained preobesity (+34%), weight gain after preobesity (+43%), maintained moderate obesity (+48%), weight gain from moderate obesity (+107%), weight loss from severe obesity (+114%), and maintained severe obesity (+83%). Regarding internists, those maintaining moderate obesity reported +107% more visits; those with weight gain from moderate obesity reported +91%. The latter group also had +41% more consultations with other physicians. Across all physicians, mean number of visits were estimated at 7.8 per year for maintained normal weight, 9 for maintained preobesity, 11 for maintained moderate obesity, and 12 for maintained severe obesity. Among those with weight loss, the mean number of visits were 8.7, 10.6 and 10.8 for baseline preobesity, moderate obesity, and severe obesity, respectively. Finally, those with weight gain from normal weight and preobesity reported 9.4 and 9.3 visits, respectively, and those with baseline moderate and follow-up severe obesity reported 13.1 visits (the most overall). Women reported higher GP and other physician utilisation. While all utilisation rates increased with age, GP utilisation was lower in middle to high SES groups.

Conclusion: Compared to maintained normal weight over a 7- to 10-year period, maintained overweight, weight gain and weight loss are associated with higher outpatient physician utilisation in adults, especially after baseline obesity. These effects only partly became insignificant after inclusion of incident diabetes or first cancer into the model. Future research should further elucidate the associations between weight development and health care utilisation by BMI status and the mechanisms underlying these associations.

Keywords: body mass index; body weight maintenance, gain, and loss; cohort studies; obesity; outpatient physician utilisation.

Table 1. Sample description by baseline BMI categories

Table 2. Development of weight status from baseline to follow-up

Table 3. Impact of BMI development on number of visits to general practitioners (N=5,147)

Table 4. Impact of BMI development on number of visits to internists (N=5,147)

Table 5. Impact of BMI development on number of visits to other physicians (N=5,147)

Table 6. Impact of BMI development on number of visits to all physicians (N=5,147)

Figure 1. Adjusted mean outpatient utilisation of GPs by BMI development (with 95%-confidence intervals)

Figure 2. Adjusted mean utilisation of internists by BMI development (with 95%-confidence intervals)

Figure 3. Adjusted mean utilisation of other physicians by BMI development (with 95%-confidence intervals)

Figure 4. Adjusted mean utilisation of all physicians by BMI development (with 95%-confidence intervals)