Overactive bladder (OAB) and its symptoms are associated with an increased risk of experiencing falls and/or fractures, particularly among older adults (≥65 years of age). Exposure to anticholinergic medications, referred to as anticholinergic burden, has also been shown to increase the risk of falls and fractures among patients with other conditions (including, for example, neurologic impairments and cardiovascular disease), but its impact in an OAB population has not previously been studied. This study aimed to estimate the association between time varying anticholinergic burden and the risk of subsequent falls and fractures in individuals with OAB.

A retrospective cohort study was performed using the Truven MarketScan United States claims databases. Adults aged ≥18 years were included if they were identified as having OAB based on the presence of International Classification of Diseases (9th Edition) (ICD-9) codes specific for OAB, or having claims for OAB-specific medications, between January 2007 and December 2015; and if they had a minimum of 12 months of medical and pharmaceutical coverage in the year prior to index date. Both newly-diagnosed (defined as the absence of ICD codes or dispensations related to OAB in the 12 months prior to the first identified OAB-related code or dispensation) and prevalent cases of OAB were included in the study cohort. A 15% sample of the cohort was randomly selected for computational feasibility (i.e. ‘OAB cohort’). Within the OAB cohort, the relationship between anticholinergic burden and a subsequent fall or fracture that was sufficiently severe to require medical attention was assessed.
Anticholinergic burden was assessed serially and longitudinally using a novel total cumulative burden measure that estimates anticholinergic exposure over a predefined time period and considers anticholinergic load of medications based on the Anticholinergic Cognitive Burden scale. Cumulative anticholinergic exposure was calculated over the 12 months prior to index date (i.e. baseline) and updated at six-month intervals over the study period; at each interval, anticholinergic exposure was calculated over the 12 months prior. Scores were categorized according to no burden (score of zero), low burden (1 to 89), medium burden (90 to 499), and high burden (500+). 
A composite outcome of a fall or fracture was defined using a validated list of ICD-9 and Healthcare Common Procedure Coding System/Current Procedural Terminology codes.(1) The relationship between the time to the first fall or fracture and time-varying anticholinergic burden was assessed using the Andersen-Gill counting style formulation of an extended Cox proportional hazards model, adjusting for age, sex, and the presence of baseline comorbidities associated with high anticholinergic burden. Sensitivity analyses were conducted applying the Cox model in subgroups of individuals ≥65 years, as well as among those with presence of comorbidities associated with high anticholinergic burden. As a sensitivity analysis, this association was investigated via a marginal structural model (MSM) with sequential inverse-probability weighting to allow for the control of time-varying covariates that may be related to both anticholinergic medication use as well as risk of falls and fractures.

The OAB cohort consisted of 154,432 individuals identified as having OAB during the study period; 33% of those had claims for OAB medications in the year prior to index date (inclusive), the remainder were eligible exclusively based on ICD-9 codes. At baseline, the mean age was 56 years, 67.9% were female, with the majority of OAB cases newly-diagnosed (69.1%). Mean anticholinergic burden at baseline was classified as moderate (266.7, with a standard deviation [SD] of 486.5); more than 60% had either no or low anticholinergic burden, and 19.1% had high burden. Burden was considerably higher among prevalent cases relative to those newly-diagnosed at baseline (522.2 [613.3] vs. 152.6 [363.6]), and among older adults (≥65 years) relative to younger individuals (433.8 [570.3] vs.213.8 [443.9]). Within the OAB cohort, there were 1,178 individuals who experienced falls and 4,603 individuals who experienced fractures in the year before baseline. The rate of falls, fractures, and falls or fractures over the study period was 1.8, 3.3, and 5.0 per 100 patient years, respectively. After adjusting for important confounders, estimates from the Cox model (Table 1) indicated that the risk of experiencing a fall or fracture increased with increasing anticholinergic burden. The increased risk of falls or fractures ranged from 1.23 (1.17-1.30) for low vs. no burden, to 1.38 (1.32-1.44) for high vs no burden. There was also a strong and non-linear association between the occurrence of falls or fractures and increasing age. The impact of increasing anticholinergic burden was present, but less pronounced among older adults. The presence of cardiovascular disease at baseline, as well as endocrine, nutritional, and metabolic disorders, was also associated with a greater probability of experiencing falls or fractures. Results from the MSM showed a similar trend as with the base case Cox model, although the magnitude of the association was lower, as expected, as this type of model is better at controlling for confounding factors.

Higher levels of anticholinergic burden were associated with a higher risk of experiencing a fall or fracture among individuals with OAB, both overall and among older adults with a higher baseline level of anticholinergic burden.

The study findings suggest that individuals with OAB with at least some level of cumulative anticholinergic burden were at higher risk of experiencing falls and fractures. These results were consistent across all conducted analyses and stratifications, including subgroup analyses in highly burdened subpopulations. The results of this study contribute evidence to inform the appropriate use of anticholinergic medications, particularly in older OAB patients with multifaceted comorbidity.

Darkow T, Fontes CL, Williamson TE. Costs associated with the management of overactive bladder and related comorbidities. Pharmacotherapy. 2005;25(4):511-9.