Supplementary Material for: Sex Differences in Characteristics Associated with Potentially Inappropriate Medication Use and Associations with Functional Capacity in Older Participants of the Berlin Aging Study II

Introduction: Medication safety is a vital aim in older adults’ pharmacotherapy. Increased morbidity and vulnerability require particularly careful prescribing. Beneath avoiding unnecessary polypharmacy and prescribing omissions, physicians have to be aware of potentially inappropriate medications (PIMs) and related outcomes to optimize older adults’ drug therapy, and to reduce adverse drug events. Objective: The aim of this study was to identify participants characteristics associated with PIM use and associations of PIM use with functional capacity with a focus on sex differences. Methods: Multivariable logistic regression analyses of cross-sectional Berlin Aging Study II (BASE-II) data (N = 1,382, median age 69 years, interquartile range 67–71, 51.3% women) were performed with PIM classification according to the EU(7)-PIM list. Results: In the overall study population, higher education was associated with lower odds of PIM use (odds ratio [OR] 0.93, confidence interval [CI] 95% 0.87–0.99, p = 0.017). Falls (OR 1.53, CI 95% 1.08–2.17, p = 0.016), frailty/prefrailty (OR 1.68, 1.17–2.41, p = 0.005), and depression (OR 2.12, CI 95% 1.32–3.41, p = 0.002) were associated with increased odds of PIM use. A better nutritional status was associated with lower odds of PIM use (OR 0.88, CI 95% 0.81–0.97, p = 0.008). In the sex-stratified analysis, higher education was associated with lower odds of PIM use in men (OR 0.90, CI 95% 0.82–0.99, p = 0.032). Frailty/prefrailty was associated with increased odds of PIM use in men (OR 2.04, CI 95% 1.18–3.54, p = 0.011) and a better nutritional status was associated with lower odds of PIM use in men (OR 0.83, CI 95% 0.72–0.96, p = 0.011). Falls in the past 12 months were related to an increased prevalence of PIM use in women (OR 1.74, CI 95% 1.10–2.75, p = 0.019). Depression was associated with a higher prevalence of PIM use in both men (OR 2.74, CI 95% 1.20–6.24, p = 0.016) and women (OR 2.06, CI 95% 1.14–3.71, p = 0.017). We did not detect sex differences regarding the overall use of drugs with anticholinergic effects, but more men than women used PIMs referring to the cardiovascular system (p = 0.036), while more women than men used PIMs referring to the genitourinary system and sex hormones (p < 0.001). Conclusion: We found similarities, but also differences between men and women as to the associations between PIM use and participants’ characteristics and functional capacity assessments. The association of lower education with PIM use may suggest that physicians’ prescribing behavior is modified by patient education, a relationship that could evolve from more critical attitudes of educated patients towards medication use. We conclude that sex differences in associations of PIM use with functional capacities might be partly attributable to sex differences in drug classes used, but not with regard to anticholinergics, as these are used to a similar extent in men and women in the cohort studied here.