A Wolk1, S Rössner. 1. Cancer Epidemiology Unit, Uppsala, Sweden.
Abstract
OBJECTIVES: To study the effects of changes in smoking and exercise habits on body weighty in Sweden. DESIGN: Analysis of factors affecting relative body weight by construction of regression equations using two models: one adjusted for age, education, socioeconomic group, geographical region and nationality, the other further adjusted for cigarette smoking and leisure-time physical activity. SETTING: Cross-sectional interview data derived from the Undersökning om LevnadsFŏrhållanden' (Living Conditions) survey 1980-81 and 1989-89. SUBJECTS: Nationally representative samples of Swedish men and women aged 16-84. MAIN OUTCOME MEASURES: Changes in body-mass index (BMI) in Sweden during the 1980s. RESULTS: Using the first model, an increase in the BMI of 0.23 kg/m2 in men and 0.17 kg/m2 in women, from 1980 to 1989, was found. When controlling for smoking and physical activity, these values were 0.20 and 0.18 kg/m2, respectively. (P at least < 0.003). CONCLUSIONS: These findings indicate that the reported decrease in smoking and changes in physical activity during the 1980s only marginally explain the BMI increase in the adult Swedish population. The health programmes launched during the 1980s seem to have little beneficial effects on body-weight development, in particular in younger generations of both sexes.
OBJECTIVES: To study the effects of changes in smoking and exercise habits on body weighty in Sweden. DESIGN: Analysis of factors affecting relative body weight by construction of regression equations using two models: one adjusted for age, education, socioeconomic group, geographical region and nationality, the other further adjusted for cigarette smoking and leisure-time physical activity. SETTING: Cross-sectional interview data derived from the Undersökning om LevnadsFŏrhållanden' (Living Conditions) survey 1980-81 and 1989-89. SUBJECTS: Nationally representative samples of Swedish men and women aged 16-84. MAIN OUTCOME MEASURES: Changes in body-mass index (BMI) in Sweden during the 1980s. RESULTS: Using the first model, an increase in the BMI of 0.23 kg/m2 in men and 0.17 kg/m2 in women, from 1980 to 1989, was found. When controlling for smoking and physical activity, these values were 0.20 and 0.18 kg/m2, respectively. (P at least < 0.003). CONCLUSIONS: These findings indicate that the reported decrease in smoking and changes in physical activity during the 1980s only marginally explain the BMI increase in the adult Swedish population. The health programmes launched during the 1980s seem to have little beneficial effects on body-weight development, in particular in younger generations of both sexes.