Associations of coprescribed medications for chronic comorbid conditions in very old adults with clinical dementia: a retrospective cohort study using insurance claims data.

Abstract

To assess the association of coprescribed medications for chronic comorbid conditions with clinical dementia in older adults, as indicated by the initiation of a new prescription of antidementia medication (NPADM). Retrospective enumeration cohort study. A Japanese city in Tokyo Metropolitan Area. A total of 42 024 adults aged ≥77 years residing in Kashiwa City, a suburban city of Tokyo Metropolitan Area, who did not have any prscription of antidementia medication from 1 April to 30 June 2012. The primary outcome was NPADM during follow-up period until 31 March 2015 (35 months). Subjects were categorised into four age groups: group 1 (77-81 years), group 2 (82-86 years), group 3 (87-91 years) and group 4 (≥92 years). In addition to age and sex, 14 sets of medications prescribed during the initial background period (from 1 April 2012 and 31 June 2012) were used as covariates in a Cox proportional hazard model. In a follow-up period of 1 345 457 person-months (mean=32.0±7.5 months, median 35 months), NPADM occurred in 2365 subjects. NPADM incidence at 12 months was 1.9%±0.1% (group 1: 0.9%±0.1%, group 2: 2.1%±0.1%, group 3: 3.2%±0.2% and group 4: 3.6%±0.3%; p<0.0001). In addition to older age and female sex, use of the following medications was significantly associated with NPADM: statins (HR: 0.82, 95% CI 0.73 to 0.92; p=0.001), antihypertensives (HR: 0.80, 95% CI 0.71 to 0.85; p<0.0001), non-steroidal bronchodilators (HR: 0.72, 95% CI 0.58 to 0.88; p=0.002), antidepressants (HR: 1.79, 95% CI 1.47 to 2.18; p<0.0001), poststroke medications (HR: 1.45, 95% CI 1.16 to 1.82; p=0.002), insulin (HR: 1.34, 95% CI 1.01 to 1.78; p=0.046) and antineoplastics (HR: 1.12, 95% CI 1.01 to 1.24; p=0.035). This retrospective cohort study identified the associations of coprescribed medications for chronic comorbid conditions with NPADM in older adults. These findings would be helpful in understanding the current clinical practice for dementia in real-world setting and potentially contribute to healthcare policymaking. UMIN000039040.