Sex Differences in Multimorbidity, Inappropriate Medication and Adverse Outcomes of Inpatient Care: MoPIM Cohort Study

Abstract

There is no published evidence on the possible differences in multimorbidity, inappropriate prescribing, and adverse outcomes of care, simultaneously, from a sex perspective in older patients. We aimed to identify those possible differences in patients hospitalized because of a chronic disease exacerbation. A multicenter, prospective cohort study of 740 older hospitalized patients (≥65 years) was designed, registering sociodemographic variables, frailty, Barthel index, chronic conditions (CCs), geriatric syndromes (GSs), polypharmacy, potentially inappropriate prescribing (PIP) according to STOPP/START criteria, and adverse drug reactions (ADRs). Outcomes were length of stay (LOS), discharge to nursing home, in-hospital mortality, cause of mortality, and existence of any ADR and its worst consequence. Bivariate analyses between sex and all variables were performed, and a network graph was created for each sex using CC and GS. A total of 740 patients were included (53.2% females, 53.5% ≥85 years old). Women presented higher prevalence of frailty, and more were living in a nursing home or alone, and had a higher percentage of PIP related to anxiolytics or pain management drugs. Moreover, they presented significant pairwise associations between CC, such as asthma, vertigo, thyroid diseases, osteoarticular diseases, and sleep disorders, and with GS, such as chronic pain, constipation, and anxiety/depression. No significant differences in immediate adverse outcomes of care were observed between men and women in the exacerbation episode.

Keywords: adverse drug reaction; in-hospital mortality; multimorbidity; network analysis; outcomes of care; potentially inappropriate medication; sex perspective.

Figure 1

Forest plot of significant clinical and sociodemographic variables. Odds ratios and their upper and lower 95% confidence intervals are shown. Reference category: men. CNS: central nervous system. COPD: chronic obstructive pulmonary disease. PIM: potentially inappropriate medication. PPO: potential prescribing omission.

Figure 2

Network visualization of chronic conditions and geriatric syndromes using the ForceAtlas2 graph layout algorithm. Data was filtered by >2% prevalence and by >2 pairwise O/E ratio. The main components of networks are nodes (chronic diseases/geriatric syndromes) and edges (coexistence of diseases or syndromes) that connect nodes in the network. Edge width is proportional to the prevalence of each chronic disease pair. Edge color is proportional to the Observed/Expected (O/E) ratio of the pair of diseases (a darker edge means a higher O/E ratio). The size of the nodes is proportional to the prevalence of the disease. The intensity of the color is proportional to the sex exclusivity. CNS: central nervous system; COPD: chronic obstructive pulmonary disease; DM: diabetes mellitus; ds: disease. Figure 2a includes male patients; Figure 2b includes female patients.