Association between depressive symptoms and use of HMG-CoA reductase inhibitors (statins), corticosteroids and histamine H2 receptor antagonists in community-dwelling older persons -: Cross-sectional analysis of a population-based cohort

Background: Strong evidence supporting the existence of causal associations of drug use with depression is often lacking from empirical studies. Limited evidence suggests an inverse association of use of HMG-CoA reductase inhibitors (statins) with depressive symptoms. Sparse and sometimes conflicting data in this respect have also been reported for corticosteroids, histamine H-2 receptor antagonists and antihypertensive agents. Objective: To identify associations between use of different drug classes and depressive symptoms in community-dwelling older patients and to determine whether any such associations varied between young-old and old-old patients. Method: We analysed cross-sectional data from a population-based cohort of 2804 older adults aged >= 55 years, in whom specific drug use was systematically identified from self-reports and physical inspection of medication labels. Depressive symptoms were determined using the 15-item Geriatric Depression Scale (GDS), with a GDS score >= 5 defining the presence of depressive symptoms. Results: Multivariate analysis controlling for the confounding effects of risk factors and correlates of depression revealed significant associations between depressive symptoms and use of statins (odds ratio [OR] 0.71; 95% Cl 0.52, 0.97), histamine H2 receptor antagonists for patients aged <65 years (OR 4.67; 95% Cl 1.59, 13.74) and, in participants aged >= 65 years, systemic corticosteroids (OR 4.02; 95% Cl 1.12, 14.42). No convincing associations between the presence of depressive symptoms and use of anti hypertensives, anti parkinsonism drugs, NSAIDs and tranquilizers/hypnotic/sedatives were demonstrated. Conclusion: The evidence in this study suggests a possible protective effect of statin use on depressive symptoms, and affirms a positive association between depressive symptoms and corticosteroid use, especially in the older elderly. Use of H-2 receptor antagonists was also associated with depressive symptoms, probably through their link with underlying dyspeptic syndromes.