Frontiers in Pharmacology · Jun 15, 2026
Lessons from the STOPPCog criteria: prevalence of potentially inappropriate medications for cognition and their association with a cognitive status - a cross-sectional study
By Szoszkiewicz M, Szafarkiewicz Z, Szkudelski F et al.
INTRODUCTION: Potentially inappropriate medications (PIMs), are common among residents of long-term care institutions (LTCIs) and may contribute to cognitive impairment (CIm). STOPPCog criteria, developed in 2025, support the identification of PIMs relevant to cognition and accuracy of dementia treatment. OBJECTIVES: The study aimed to identify PIMs according to the STOPPCog criteria among residents of long-term care institutions in Poland, and to assess the association between CIm and exposure to specific drug classes. PATIENTS AND METHODS: This is a multicenter cross-sectional study in four LTCIs. Medication lists were reviewed using STOPPCog criteria and anticholinergic burden scales. Cognitive status was assessed with the Mini-Mental State Examination (MMSE), and CIm was defined using a conventional cut-off. We compared exposure frequencies between residents with and without CIm and explored factors associated with MMSE using regression models adjusted for key demographic and clinical variables and facility effects. RESULTS: Of 204 residents, 133 (65.2%) had at least one STOPPCog-listed medication on their medication list. Drugs with anticholinergic burden (ACB) (54.9%), especially one subgroup - antipsychotics (39.2%), were most frequently used within the study sample. Antipsychotics were the only drug class significantly associated with CIm independently of sex, age, comorbidity burden, and facility (p < 0.001). The majority of residents with a previous diagnosis of dementia in their medical records and consequently dementia-specific pharmacotherapy had severe CIm. CONCLUSION: There is a high prevalence of PIMs among LTCIs residents. Antipsychotic deprescribing and early diagnosis of CIm should be further evaluated in future prospective studies and clinical practice.
doi.org/10.3389/fphar.2026.1814603Open accessPubMed 42375615Populations