Overactive bladder (OAB) is a prevalent condition in the elderly that often results in a referral to a urologist. The use of anticholinergic medications for OAB is associated with potential adverse effects (such as cognitive changes) compared to OAB beta-3 agonist medications. Specific potential risks include delirium and falls or fractures with the use of anticholinergic medications in general. These outcomes are important as they are associated with institutionalisation and mortality in the elderly. To our knowledge, the relative risk of delirium and falls or fractures with the new initiation of OAB anticholinergics has not been specifically studied. Beta-3 agonist users are well-suited to act as a condition-specific reference group. Our objective was to determine if there was an elevated risk of either delirium or falls/fractures among patients who initiated anticholinergic OAB medications compared to beta-3 agonist OAB medications.

We conducted a retrospective cohort study using routinely collected, population-level data. The cohort was drawn from Canada’s largest province of Ontario. All residents of Ontario have access to a single, universal healthcare system that provides medication coverage for those over 65 years of age (approximately 2.2 million). The use of the data was authorized under Ontario’s Personal Health Information Protection Act, which does not require approval by a Research Ethics Board, or patient consent. 

Participants were all residents >66 years of age who newly initiated an OAB medication between January 2016-March 2020. We identified and classified people based on the type of OAB medication prescribed (beta-3 agonist, oxybutynin, or newer OAB anticholinergics [tolterodine, solifenacin, darifenacin, fesoterodine, trospium]). Oxybutynin was considered separately due to the increased evidence of detrimental cognitive impact compared to the other anticholinergics. Co-primary outcomes were evidence of a hospital visit with delirium, or for a fall/fracture.

Matching weights (an extension of propensity score weighting) were used to balance the three exposure groups based on the 82 measured indicators of baseline health, comorbidity, medication usage, and healthcare utilization. Matching weights were generalized to permit comparison between three groups. Propensity scores were estimated using multinomial logistic regression, where the model output contained predicted probabilities for all three exposure groups. The matching weights were then assigned as the minimum of the three predicted probabilities (propensity scores) divided by the predicted probability for the treatment they actually received.  We used matching weights to make the three exposure groups comparable across 82 relevant baseline characteristics. We examined both the risk during the first 30-days (logistic regression) and the risk during continuous usage (cox proportional hazards).

The primary cohort included 103,024 older adults who were newly dispensed a beta-3 agonist (54%, n=56,062), oxybutynin (13%, n=13,865), or a newer anticholinergic (32%, n=33,097). The median age was 76, and most were female. With matching weights all measured variables had standardized differences ≤5%. In the newer anticholinergic group, the most common initial OAB medications were fesoterodine (41%, 13,733/33,097) and solifenacin (30%, 10,016/33,097). In the oxybutynin group 88% (12,180/13,865) started at a dose of <15mg/day, and in the newer anticholinergic group, 72% (23,787/33,097) started at the lower labelled daily dose.

The incidence of delirium within the first 30 days after the initiation of an OAB medication was 0.31%, and the incidence of a fall/fracture was 1.07%. After weighting, the use of oxybutynin or newer anticholinergic medications was not associated with a significantly increased odds of a hospital presentation of delirium or a fall/fracture (table 1). There was no significant difference in the risk of these outcomes between high dose and low dose users of either the oxybutynin group, or the newer anticholinergic group.

The median (IQR) duration of continuous usage was 113 (30-380) days for mirabegron, 30 (28-72) days for oxybutynin, and 62 (30-239) days for the newer anticholinergics. There was no significant increased risk of delirium among oxybutynin users, and a slightly increased HR for delirium among users of newer anticholinergics (Table 2, HR 1.13, 95% CI 1.02-1.26). This equates to one extra case of delirium for every 246 person-years of newer anticholinergics use instead of a beta-3 agonist. There was a slightly increased HR for fall or fracture among oxybutynin users (Table 2, HR 1.13, 95% CI 1.02-1.24), and no significant increased risk with newer OAB anticholinergics. This equates to one extra fall/fracture for every 45 person-years of use of oxybutynin instead of a beta-3 agonist.

In this study we examined whether OAB anticholinergic medications are associated with an increased risk of delirium or falls/fractures in an elderly population. We found that in the first 30 days after initiating oxybutynin or one of the newer OAB anticholinergics, there was no significant increased risk of these outcomes. This suggests that the risk of these events is low with a 30-day trial of anticholinergic medications. When considered the period of continuous use of these medications, newer anticholinergics had a slightly increased risk of delirium (13%), and oxybutynin had a slightly increased risk of fall/facture (13%). These results are statistically significant however the absolute risk difference is small. It is important to note that the point estimate for newer anticholinergics and delirium was OR 0.92 in the 30-day analysis, and HR 1.13 in the continuous use analysis, suggesting the risk for delirium may increase with continued use. The point estimate for oxybutynin and fall/fracture is similar between the two analyses (OR 1.19 compared to HR 1.13), suggesting the lack of a significant association in the 30-day analysis may have been due to a lack of statistical power for a relatively rare outcome. The differential effects of oxybutynin versus newer anticholinergics may be due to more significant central nervous system effects that specifically mediate falls/fractures, and longer persistence with newer anticholinergic medications may facilitate the significantly increased risk of delirium. Our study supports the use of OAB beta-3 agonists as more appropriate first line oral therapy for patients at risk or concerned about the risk of delirium or falls/fractures.

The use of oxybutynin and newer OAB anticholinergics does not significantly increase the 30-day risk of delirium or fall/fracture compared to patients who use beta-3 agonists. However, over time the use of oxybutynin is associated with a significantly increased risk of fall/fracture, and the use of newer anticholinergics is associated with a significantly increased risk of delirium. These results should be considered by prescribers, and further evaluated in different datasets. The use of beta-3 agonists to treat OAB symptoms may minimize the risk of either of these outcomes compared to anticholinergic medications.