Overactive bladder disease (OAB) is defined by the Standardization Subcommittee of the International Continence Society (ICS) as urinary urgency, with or without urinary incontinence, usually with frequency and nocturia, with no proven infection or other obvious pathology 1. As part of the National Overactive Bladder Evaluation (NOBLE) program: 16.5% of the USA population (16% of men and 16.9% of women) over 18 years of age had symptoms consistent with OAB 2. This translates into approximately 34 million symptomatic people in the USA alone. Of the 16.5% of the adult population with symptoms consistent with OAB, 37.2% were characterized as having OAB with urinary urge incontinence (UUI) 2. OAB significantly impacts heath related quality of life 3 and is associated with comorbidities such as increased risk of falls and fractures, increased urinary tract and skin infections, sleep disturbances, depression, and decreased sexual health 45. The economic impact of OAB is considerable given its estimated attributable cost of $12.02 billion in the USA for the year 2000 6.

Depending on severity and etiology, initial treatment of OAB is characterized by behavioral modification (timed voiding, bladder training, pelvic floor exercises, etc) and antimuscarinic drugs. Behavioral interventions for treating OAB such as bladder training are often the first line therapy for mild symptoms of urge, frequency, and/or incontinence. Bladder training involves patient education, scheduled voiding, and positive reinforcement. Other behavioral interventions include pelvic floor exercises (PFE) and biofeedback. PFEs are thought to inhibit spontaneous bladder contractions and also increase bladder outlet resistance to result in reduced leakage and increased voiding intervals. Biofeedback is a way of notifying the patient when certain physiologic events are occurring (e.g. unstable bladder contraction or proper PFE contraction).

In a Cochrane review on bladder training, treatment outcomes consisted of patient's global observations (perception of a cure or improvement in their incontinence), quantification of symptoms (adverse events, number of incontinence episodes, number of micturitions in day and night), and quality of life (Incontinence Impact Questionnaire and SF-36) 7. Outcomes in PFEs and biofeedback include the quantification of symptom domains, and urodynamic outcomes 8.

Most urinary incontinent patients initiate treatment with antimuscarinic therapy. Six antimuscarinic drugs are currently marketed worldwide for the treatment of OAB: oxybutynin, tolterodine, propiverine, trospium, darifenacin, and solifenacin 9. Each product has demonstrated efficacy in treating OAB symptoms but includes such common adverse events as dry mouth, constipation, headache, and blurred vision. In a review of antimuscarinic drugs versus placebo, Nabi and colleagues found that there was a lack of consistency in the types of outcomes reported by trialists 10. Heterogeneity in urinary incontinence and other clinical characteristics in various OAB trial patient populations may partly describe such differences. We attempted to reduce clinical characteristic heterogeneity by focusing on treatment success definitions in populations that have failed antimuscarinic therapy.

If patients fail oral drug (antimuscarinic) therapy, more invasive interventions are available. Approved implantable neurostimulator devices have been used with some success, sacral neuromodulation (SNM), but are costly 11. Surgical options such as augmentation cystoplasty (AC), and neobladder construction, or urinary diversion are available, but are limited to those that have failed the previous mentioned techniques due to elevated risk profile and complexity of the surgeries. Although still in phase II and phase III studies, botulinum toxin type A (BoNTA) is emerging as a potential treatment option for idiopathic and neurogenic OAB with urinary urge incontinence, respectively 12.

Clinicians, payers, and other health care decision makers all have interest in determining whether or not a patient management strategy is successful. In part due to patient heterogeneity, interventions that are indicated for the management of OAB with UUI lack standardization in the measurement of treatment response or treatment success. A consistent and validated treatment success metric would allow for more efficient use of healthcare resources, facilitating the clinician and patient to more transparently weigh the risk-benefit profile of treatment alternatives. A standardized success outcome would also permit more consistent value for money assessments across and beyond OAB interventions.

The objectives of this research are to review common treatment success outcomes used in both idiopathic and neurogenic OAB with UUI, focusing on the different antimuscarinic refractory intervention classes and to recommend a measure of treatment success for clinical research and health economic applications.

Deriving a definition of treatment success in any disease area poses many challenges, but we argue that the benefits outweigh the harms in undergoing such an exercise in UUI OAB. Possible limitations to a definition of treatment success in UUI OAB include: patient clinical heterogeneity (e.g. UUI OAB patients have varying levels of incontinence), patient preference heterogeneity (e.g. some patients having the same study outcomes may opt to continue treatment whereas others may opt for alternative interventions), and duration of treatment effect (e.g. if incontinence or other symptoms are included in the measure of success, then how often and over what time period should they be measured?). At the population level, a uniform treatment response outcome provides for the comparison across various UUI OAB interventions, gives researchers a clear and achievable target for purposes of trial design, and allows payers to attain better value for money on UUI OAB interventions for their insured population.

Payne and Kelleher recently evaluated three possible treatment response definitions that were informed from the ICS definition of OAB: (i) a reduction by half or more in all baseline symptoms (episodes of urgency, incontinence, nocturia, and voiding frequency per 24 h); (ii) a reduction by half or more in urgency and at least one other symptom; or (iii) resolution of urgency episodes and at least one other symptom 23. All three treatment response definitions were correlated with significant changes in the King's Health Questionnaire. The authors argued that clinical trials and the evidence base in OAB focus on single endpoints such as incontinence episodes or reduction in incontinence, but correctly point out that the ICS definition of OAB mainly targets urgency and secondarily lists the other symptoms of incontinence, nocturia, and increased voiding frequency. One of the main pitfalls in using any of these three proposed treatment response definitions is that there is yet to be a consensus on how to measure urgency in a reliable and valid way 24. Further, if one is tasked to gather the current and existing evidence across OAB interventions on treatment response, then retrospectively constructing a composite definition of success is nearly impossible.

Taking into account the currently available evidence, the most likely candidates for a definition of OAB with UUI treatment success are measures that focus on reductions in incontinence episodes. The achievement of full continence has been a common metric in BoNTA studies. But in the case of a dichotomous outcome of success (yes/no), full continence likely excludes many patients that are clinically benefiting from treatment. The question then becomes, what is an appropriate threshold in terms of incontinence reduction that resonates as a clinically meaningful improvement yet does not disregard too many patients that are benefiting? We sought out a definition of success that limits misclassification/residual confounding by setting a threshold that most appropriately includes those receiving meaningful clinical benefit while excluding those that experience a measurable benefit that is not clinically relevant.

We recommend that the appropriate threshold for defining OAB with UUI treatment success should be ≥ 50% reduction in incontinence episodes from baseline. We stress that the use of this definition may be sensitive to the baseline incontinence rate and therefore, both the baseline incontinence rate and the change in the rate should be communicated in research findings. In part, the recommendation definition is based on what is currently available and reported in the literature. The ≥ 50% reduction in incontinence episodes treatment response definition has been used across all reviewed intervention classes and is associated with generic and disease specific HRQoL measures. The average difference in the disease specific HRQoL instrument, I-QOL, change scores was higher than the minimal important difference (MID) reported for stress urinary incontinence of 6.3 17. Although stress urinary incontinence is not the study population, it is the only known population wherein a MID for the I-QOL has been reported. Additionally, Yalcin et al reported that stress urinary incontinence patients appear to recognize important clinical value at incontinence reductions of approximately 50% 25. If the MID for OAB with UUI was found to be similar to that in stress urinary incontinence, then the findings from the Schurch et al trial suggest that the recommended treatment success definition is associated with a meaningful improvement in a disease specific HRQoL instrument.

There are several limitations to the study. The primary aim of this research was to review the evidence on treatment success definitions reported for OAB with UUI refractory to oral antimuscarinics. We focused on three primary intervention classes in order to make our literature search tractable. By doing so, we did not summarize treatment success definitions for more rare interventions in this population such as percutaneous tibial nerve stimulation given the limited quantity of published evidence on such interventions. We did not include behavior modification interventions or antimuscarinics because of the anticipated large amount of patient characteristic heterogeneity in this broad population. Increased heterogeneity in UUI symptoms of a population could limit the generalizability of a uniform treatment success definition. The association of the recommended treatment success definition with disease specific and generic HRQoL measures was performed in a BoNTA compared to placebo trial of neurogenic incontinent patients. This same association may not hold for open label studies or those with SNM or AC interventions. Although the recommended treatment success definition of ≥ 50% reduction in incontinence episodes from baseline was the most common observed definition in the idiopathic OAB with UUI population, it should be further validated against other OAB outcomes.

We echo arguments put forth by Payne and Kelleher on the need for a standardized treatment success definition. Moving forward, a success definition like those proposed by Payne and Kelleher that includes many components of the underlying symptoms of OAB (namely urgency) makes clinical sense, but is currently not practical based on difficulties in measuring urgency. Based on the current evidence, we have proposed a definition of success for OAB with UUI that is currently gaining momentum in the literature and is supported by associations with clinically meaningful outcomes. The adoption of the ≥ 50% reduction in incontinence episodes as the treatment response definition for UUI OAB will provide for consistent and valuable outcomes research for both clinical and health economic applications.

JDC, RRD, and SDS are consultants to Allergan, Inc. JDC and SDS were compensated for their contributions to this manuscript through an unrestricted grant by Allergan, Inc. KSG and JHW are University of Washington fellows funded by Allergan, Inc. AR was a former employee of Allergan, Inc. The authors declare that they have no other competing interests to disclose.

SDS, JDC, and AR, designed and specified the research objectives. KSG and JDC carried out the literature searches and data extraction from the literature. JHW and RRD were responsible for the clinical interpretation of the findings. JDC drafted the manuscript. All coauthors were responsible for critically revising the manuscript. All coauthors have given final approval of this version for publication.