Prevalence of lifestyle characteristics in glucocorticoid users and non-users: A Danish population-based cross-sectional study

Abstract

Objectives Lifestyle may affect observed associations between glucocorticoid use and adverse events. This study aimed to investigate whether lifestyle differ according to use of systemic glucocorticoids. Design Population-based cross-sectional study. Setting The Central Denmark Region. Participants 30 245 adults (≥25 years of age) who participated in a questionnaire-based public health survey in 2010. Outcome measures Systemic glucocorticoid use was categorised as never use, current use (prescription redemption ≤90 days before completing the questionnaire), recent use (prescription redemption 91-365 days before completing the questionnaire), former use (prescription redemption >365 days before completing the questionnaire) and according to cumulative dose expressed in prednisolone equivalents (<100, 100-499, 500-999, 1000-1999, 2000-4999, ≥5000 mg). We computed the prevalence of lifestyle factors (body mass index, smoking, alcohol intake, physical activity and dietary habits) according to glucocorticoid use. We then estimated age-adjusted prevalence ratios (aPRs) and 95% CIs, comparing the categories of glucocorticoid users versus never users. All analyses were stratified by sex. Results Of the 30 245 participants (53% women, median age 53 years), 563 (1.9%) were current users, 885 (2.9%) were recent users, 3054 (10%) were former users and 25 743 (85%) were never users. Ever users of glucocorticoids had a slightly higher prevalence of obesity than never users (18% vs 14%, aPR=1.4, 95% CI 1.2 to 1.5 in women and 17% vs 15%, aPR=1.2, 95% CI 1.1 to 1.4 in men). In women, ever users of glucocorticoids had a slightly lower prevalence of high-risk alcohol consumption compared with never users (17% vs 20%, aPR=0.8, 95% CI 0.7 to 1.0). Smoking, diet and physical activity did not differ substantially according to use of glucocorticoids. Conclusion Our study provides a framework for quantifying potential uncontrolled confounding by lifestyle factors in studies of systemic glucocorticoids.