Association between inappropriate medications and comprehensive geriatric assessment outcomes: a cross-sectional...

Different criteria were developed to identify drugs with a negative benefit/risk ratio in older adults, so-called potentially inappropriate medication (PIM). We investigated the correlation between the number of PIMs identified using PRISCUS 2.0, STOPP version 3, and Beers criteria and the functional and nutritional status of Turkish geriatric patients, as assessed by comprehensive geriatric assessment (CGA) tools.

This retrospective cross-sectional analysis was conducted in geriatric outpatients over the past 10 months. Data were collected on age, clinical conditions, medication, and CGA test results. To identify independent factors associated with PIM use and to adjust for potential confounding, logistic regression analyses were performed. PIM use was analysed as a binary outcome (no PIM versus at least one PIM), and no imputation was performed. Two logistic regression models were fitted in 362 patients with complete data, without (Model 1) and with (Model 2) the number of drugs.

Among 397 participants, 58.7% (n = 233) of patients received at least one PIM when combining all three criteria (PIM-ALL). Multivariable analyses showed that the number of medications and functional status (Katz Activities of Daily Living) were most strongly associated with PIM use. In the fully adjusted model (Model 2, n = 362), each additional medication was associated with 55% higher odds of PIM use (OR 1.55, 95% CI 1.40-1.73), while a higher Katz ADL score was associated with lower odds (OR 0.73, 95% CI 0.60-0.87).

There is relevant variability among PIM tools and in their association with CGA test results. Advanced, systematic, and easily implementable methodologies are required to address these shortcomings.

This study does not involve a clinical trial and thus was not registered.