Anticholinergic Burden and Associated Healthcare Resource Utilization in Older Adults with Overactive Bladder

Pharmacological options for overactive bladder (OAB) include bladder anticholinergic agents and beta-3 agonists (e.g., mirabegron) [1‐4]. Bladder anticholinergic agents are the most widely used OAB drugs; they are effective, but can cause anticholinergic adverse effects, including dry mouth, blurred vision, constipation and somnolence [5]. These off-target effects are influenced by the drug’s selectivity for different muscarinic receptor subtypes and its disposition (e.g., propensity to cross the blood-brain barrier) [6]. Many commonly prescribed medications (tricyclic antidepressants, certain antihistamines, etc.) also have off-target anticholinergic effects [2, 7].

Older individuals, who constitute a high proportion of the OAB population [8‐10], may be more vulnerable to off-target anticholinergic effects [11]. Older patients have higher levels of comorbidity and polypharmacy, and are more likely to receive multiple drugs with anticholinergic activity [11]. Indeed, an estimated 20–50% of older people are routinely exposed to medicines with anticholinergic activity [12]. Furthermore, older individuals may have age-related decline in cholinergic function, increased blood-brain barrier permeability and altered drug pharmacokinetics, which may increase their sensitivity to anticholinergic effects [12]. Anticholinergic burden is the cumulative effect of multiple medications with varying degrees of anticholinergic activity [12] and can be measured using a range of scales, such as the Anticholinergic Drug Scale and Anticholinergic Activity Scale (general scales) and the Anticholinergic Cognitive Burden [ACB] scale and Anticholinergic Effect on Cognition [AEC] scale (cognitive scales) [13‐15].

According to US, European and international urinary incontinence/OAB guidelines, the cumulative effects of anticholinergic drugs on cognition should be considered when prescribing treatment [1‐3, 16]. The American Geriatric Society also recommends keeping the number of anticholinergic drugs to a minimum in older people and avoiding some drugs with strong anticholinergic properties [7]. Cumulative anticholinergic burden has been shown to increase healthcare resource utilization (HRU) in a diverse ambulatory population aged ≥ 65 years [17]. Furthermore, studies in patients with OAB have shown that ACB scores were higher than in those without OAB (and in about 60% of patients, ≥ 80% of the score was attributable exclusively to bladder anticholinergics) [18] and that higher anticholinergic burden was associated with higher rates of falls and fractures [19]. However, there is a lack of data on the effect of anticholinergic burden on HRU and related costs in patients with OAB, despite the frequent use of bladder anticholinergic agents. The aim of the current study was therefore to evaluate the association between cumulative anticholinergic burden and HRU in older patients with OAB.

This is the first study to evaluate the association between cumulative anticholinergic burden, HRU and costs in older patients with OAB. The results suggest that an increase in anticholinergic burden was generally associated with increased HRU and increased costs after adjusting for significant confounders.

In the current study population, the mean (SD) ACB score was 1.0 (1.1) and approximately 10% of patients had a mean total ACB score of ≥ 3 during the 3-year pre-index period. Despite this, 95.7% were starting first-line bladder anticholinergic treatment for OAB at the index date. This is perhaps not surprising as the only pharmacological non-anticholinergic treatment for OAB (mirabegron) was not launched until 2013. Initiation of bladder anticholinergic treatment at the index date could add significantly to their underlying anticholinergic burden. Indeed, post-index mean total daily ACB scores were approximately 1.5 points higher versus pre-index scores (and 48.7% had a mean score ≥ 3) in patients starting a bladder anticholinergic agent. In comparison, post-index scores in patients starting mirabegron were only approximately 0.1 points higher than pre-index scores (and 17.2% had a mean score ≥ 3). Interpretation of these results is limited by the relatively small number of mirabegron-treated patients; however, as mirabegron is not an anticholinergic, it provides one way to avoid adding to anticholinergic burden in the older OAB population. Studies evaluating mirabegron in OAB in older patients have shown that it is effective and has a favourable tolerability profile compared with bladder anticholinergic agents, especially with respect to rates of dry mouth [23‐29]. Before prescribing a treatment for OAB, it is also important that clinicians assess patients’ existing medications with a view to reducing the anticholinergic burden if possible.

The results obtained are generally consistent with a previous study from a general population, which showed that a 1-point increase in cumulative anticholinergic burden (also assessed using the ACB score) was associated with a significant increase in HRU [17]. The current study extends these findings by looking specifically at older patients with OAB and showing that costs generally rise with increasing cumulative anticholinergic burden.

The study does have some limitations inherent to its design. Use of the UK CPRD database means that the extent to which the results can be extrapolated to countries outside the UK is unclear, as there may be differences in licensed/commonly used drugs with anticholinergic properties, prescribing practices and guidelines, and funding arrangements. One of the other limitations of using the CPRD database is that it provides drug prescription data only. Information on whether drugs are taken as prescribed is not available, and it does not include information on over-the-counter (OTC) medication use. A future study in which adherence with anticholinergic medication and OTC medication use are taken into account may add value, but at the expense of patient numbers. It would also be interesting to assess the extent to which anticholinergic burden is related to the target and off-target effects of drugs, and explore the qualitative impact of cumulative anticholinergic burden on quality of life.

Another limitation of the current study is the potential for survivor bias, as it was based on patients with data available for at least 2 years after OAB diagnosis/first OAB prescription. However, this was addressed by one of the sensitivity analyses, which included data regardless of the duration of follow-up, and generated similar results to the main analysis. There may have also been additional factors (e.g., body mass index, smoking status, etc.) contributing to resource use that were not adjusted for in the analyses. Comorbid conditions were not adjusted for per se, as the definitions used in this study did not reflect comorbidities coding in UK general practice, but the number of co-medications was used as a proxy. Finally, the current analysis did not investigate the contribution of post-index anticholinergic burden on HRU, although adjustment was made for the unweighted ACB score on the index date. The results obtained for the association between anticholinergic burden and HRU and costs in the pre-index period should be interpreted with caution as it is not known whether ACB exposure occurred before or after the HRU events.

Despite the limitations highlighted above, use of the UK CPRD database does have benefits. It includes data from 35 million people in the UK and provides a longitudinal dataset that is representative of the UK population. It is widely used internationally for epidemiological research across a broad range of health outcomes, producing over 1000 studies published in peer-reviewed journals [20]. The results therefore provide a useful indicator of the association between cumulative anticholinergic burden and HRU and costs in patients with OAB. The results also provide longitudinal data for clinicians, which is more useful than cross-sectional information, particularly in relation to cumulative anticholinergic burden.

In the current study, the primary analysis was based on calculations using the ACB scale. This is the most widely used scale in studies quantifying anticholinergic burden and has been found to be well-suited to analyzing administrative data [30], although there is no gold standard. We therefore performed a sensitivity analysis using the AEC scale [31] and this provided similar results to the ACB. The AEC scale is relevant to the current study population, as the scale was based on a list of drugs commonly used in older patients in the UK. It also downgrades the scores of many drugs (including some bladder anticholinergic agents) if they do not penetrate the CNS or there is no clinical evidence for cognitive effects. Indeed, only 5.0% of patients had a pre-index score of 3 or more using the AEC compared with 9.9% using the ACB, and a score of 1 or more was seen in 26.3% and 56.5% of patients, respectively. It should be noted that both the ACB and AEC focus on cognition and may not accurately quantify anticholinergic burden in the periphery, where effects on dry mouth, constipation, etc. also contribute to HRU.

The results of the current study show that, at the point of OAB diagnosis or initiation of first-line treatment, patients aged ≥ 65 years already have some underlying anticholinergic burden. Higher anticholinergic burden before initiating OAB drugs is associated with higher HRU and costs. As such, clinicians and decision makers should evaluate overall anticholinergic burden for patients > 65 years with OAB, including the use of alternative treatment options to bladder anticholinergic agents, to help reduce anticholinergic burden and associated resource use.