Background: For the conservative treatment of Overactive Bladder (OAB) symptoms, the National Institute for Health and Care Excellence (NICE) in the UK currently recommends a course of supervised pelvic floor muscle training, behavioural therapy, anticholinergic medication, sacral nerve stimulation, and more recently, botulinum toxin type A (BoNTA) and Mirabegron. Given the large number of interventions and relatively few primary trials, network meta-analyses (NMAs) produce considerable uncertainty in the estimated treatment effects and consequently, there is little evidence of the most clinically effective intervention. Objectives: To evaluate the use of hierarchical NMAs incorporating ordering constraints on increasing doses in order to identify the most effective intervention for the treatment of OAB symptoms. Methods: Using Bayesian Markov Chain Monte Carlo methods, we apply a 3-level hierarchical NMA that accounts for both the correlation between treatments within the same class, as well as the residual between-study heterogeneity. We further extend this model to incorporate ordering constraints on increasing doses of the same intervention. We apply the methods to a dataset obtained from a systematic literature review of randomised controlled trials evaluating interventions for OAB syndrome. The primary outcomes of interest were mean change from baseline for voiding, urgency, and incontinence episodes. Results: The dataset includes 78 trials comparing 39 interventions that can be further categorised into 10 classes of interventions, including placebo. For voiding, and urgency episodes, BoNTA 200u was the most effective intervention with estimated mean reduction of -2.24 (95% CrI: -2.95, -1.48), and -2.6 (95% CrI: -3.46, -1.7) episodes relative to placebo, respectively. BoNTA 300u was the most effective intervention for reducing incontinence episodes with an estimated mean reduction of -1.81 (95% CrI: -2.39, -1.33) episodes relative to placebo. Conclusions: Use of hierarchical NMAs, incorporating ordering constraints, increases the precision in the effect estimates but maintains the interpretability of individual interventions. BoNTA was found to be the most effective intervention for reducing symptoms of OAB.
Owen, R. K., Tincello, D. G., Bujkiewicz, S., & Abrams, K. (2014). Hierarchical Network Meta-Analysis Incorporating Ordering Constraints on Increasing Doses of Interventions - Application to Overactive Bladder Syndrome. Value in Health, 17(7), A543. https://doi.org/10.1016/j.jval.2014.08.1755