Patterns of polypharmacy before diagnosis of dementia: a data-driven, retrospective, population-based study with primary care electronic health records

Background Polypharmacy presents a serious and significant public health challenge as average lifespans increase. The objective of this study was to identify patterns of polypharmacy before dementia diagnosis, their prevalence, and possible complications. Methods We used electronic health records in the Welsh general practitioners’ database from 1990 to 2015, via the Secure Anonymised Information Linkage Databank. Cohort selection, based on a previously validated phenotype, was done with the use of Read Codes to identify any dementia diagnosis within the dataset. No age restriction was applied, but complete data for each patient was required. Analysis was stratified by sex and age, but not dementia type. Medications were identified by Read Codes, and split into four 5-year sub-periods. Factor analysis was used for patients taking at least three medicines in each period. Findings We identified 33 451 patients with a dementia diagnosis from more than 42 million rows of health records. 22 011 (65·8%) patients in the cohort were female, with mean age at dementia diagnosis of 72·75 years (SD 13·59). In sub-period 1 (0–5 years pre-diagnosis), 22 157 (82·2%) of 26 968 patients were taking three or more medications, and three clusters were identified: medicines for urinary or respiratory infections, arthritis or rheumatism, and heart disease (17 947 [66·5%] patients); medicines for urinary or respiratory infections, arthritis or rheumatism, heart disease, and depression (5938 [22·0%]); and medicines for osteoarthritis (701 [2·6%]). By contrast, in sub-period 4 (15–20 years pre-diagnosis), 421 (5·5%) of 7660 patients were taking three or more medications, and five patterns emerged: medicines for infections, arthropathy, and cardiovascular disease (421 [5·5%] patients); medicines for anxiety and acute respiratory infection (184 [2·4%]); medicines for acute respiratory infection and cardiovascular disease (161 [2·1%]); medicines for dermatological disease (23 [0·3%]); and medicines for asthma and other acute respiratory infections (four [0·1%]). 23 891 (71·4%) patients in all periods took medications for arthritis, heart disease, and infections, with or without medications for mental health. Interpretation The earlier before dementia diagnosis, the greater the number of medicine clusters are observed, with lower prevalence of each cluster. Understanding how dementia medicines interact and cause complications with other groups of medicines is fundamental. These patterns can inform safe prescribing practices before diagnosis in selecting medicines with a low anticholinergic burden to minimise their effect on cognitive impairments. Further analysis of these types of polypharmacy among people without dementia will be done in the future. Funding Health Data Research UK (NIWA1).