The relationship between polypharmacy and trajectories of cognitive decline in people with dementia: A large representative cohort study.

Polypharmacy, defined through the number of medications prescribed, has been linked to a range of adverse health outcomes in people with dementia. It is however unclear whether a numerical threshold of concurrently prescribed drugs is a suitable predictor for cognitive decline. We aimed to test associations between polypharmacy and both short-term (six months) and long-term (three years) cognitive trajectories in patients with incident dementia. Using data from a large mental health and dementia care database in South London, a cohort of 12,148 patients (mean age = 80.7 years, 61.1% female, mean MMSE = 18.6) clinically diagnosed with dementia was identified. We determined the number of medications prescribed at dementia diagnosis and defined two exposure groups: polypharmacy (5-9 medication) and excessive polypharmacy (≥10 medications), with 0-4 medications as reference group. All Mini Mental State Examination (MMSE) scores between one year before and three years after dementia diagnosis were ascertained. Effects of polypharmacy on cognitive decline were studied using Generalized Additive Models for Location, Scale and Shape and Linear Mixed Estimation Models. At the time of dementia diagnosis polypharmacy was present in 3503 (28.8%) patients and excessive polypharmacy in 1235 (10.2%) patients. In all three groups MMSE scores initially improved after dementia diagnosis and further decline was detected in the time interval from six months to three years after dementia diagnosis. No significant differences to the control group were found in relation to polypharmacy or excessive polypharmacy, neither in the initial cognitive improvement nor long-term decline. In conclusion, polypharmacy defined by the number of drugs does not appear to predict cognitive decline in a naturalistic cohort of patients with dementia. More sophisticated tools, considering appropriateness of prescribing and the clinical picture, might be better placed to evaluate cognitive outcomes in dementia and to make practice and research recommendations.