Introduction

Overactive bladder (OAB), an idiopathic symptom-complex defined by urinary urgency (often with frequency, nocturia and incontinence) [1] increases in prevalence with age and contributes to multiple comorbidities and decreased quality of life [2, 3]. Behavioral therapy, which includes pelvic floor muscle (PFM) exercise, bladder training to reduce voiding, and lifestyle advice (e.g., modifying fluid intake), is the first-line treatment for OAB [4,5,6]. Controlling or suppressing the sensation of urgency is a key part of behavioral therapy and usually includes a combination of PFM contraction, mental distraction and general relaxation [7].

The mode of action of urgency suppression techniques is widely attributed to reflexive inhibition of the detrusor muscle [8,9,10]. This is the key rationale given for PFM exercise for OAB [10]. Notably, there has been considerably less research investigating the veracity of hypotheses proposed for potential effectiveness of PFM exercise to treat OAB than there has been for SUI [11]. The PFM exercise component of behavioral therapy has been evaluated in two recent systematic reviews (between them including 16 studies), both of which concluded that while PFM exercise seems to be effective for OAB, heterogeneous training programs and generally small studies of poor quality made conclusions about training details difficult [11, 12].

Cognitive urgency control strategies, which are typically included in behavioral therapy, comprise mental distraction, such as focusing on a task unrelated to the bladder (e.g., counting backwards) and/or relaxation, such as picturing oneself in a calming environment; both are intended to suppress urgency by focusing attention away from bladder sensation [7]. However, mental distraction has been shown to impair effective PFM contraction, indicating that distraction could be counterproductive for controlling urinary urgency [13, 14]. Recently, mindfulness practices such as non-judgmental awareness of bladder sensation and responding to urinary urgency instead of reacting have been proposed for urgency management as alternatives to suppression via distraction [15].

While previous systematic reviews on behavioral therapy for OAB have focused on PFM exercise, to our knowledge the cognitive techniques included in behavioral therapy programs have not been evaluated. Therefore, the purposes of this systematic review are to (1) describe the cognitive components of behavioral therapy for OAB, (2) describe any rationale presented for cognitive components and trace the origin of this rationale, (3) evaluate the effectiveness of behavioral therapy, which includes a cognitive component on OAB symptoms, and (4) determine whether the specific impact of the cognitive component of programs can be evaluated.

Materials and methods

An electronic search was conducted on the PubMed, CINAHL, Web of Science, Cochrane and PEDro databases. The search strategy is presented in Table 1. Inclusion criteria were trials on OAB therapy in which at least one group received some form of behavioral therapy as part of their treatment, published in English, including neurologically intact, non-pregnant participants aged 18 or older. Papers could be single-arm or randomized controlled trials (RCTs). Studies published since 2002, when the current definition of OAB was published, were included. The search was conducted in April 2019 and was updated in January and June 2020. The searches were saved and updates were sent weekly to the chief researcher (BR). Outcomes of interest were symptoms of OAB (urgency, frequency, nocturia, UUI), quality of life (QoL) and treatment satisfaction. Titles and abstracts were scanned to identify studies that included behavioral therapy for OAB symptoms; papers were then scrutinized for any description of a cognitive component to behavioral therapy. An aspect of behavioral therapy was considered to be a ‘cognitive strategy’ if it involved analysis of or change to a thought process related to the sensation of urgency or employment of a specified thought process during an episode of urgency. If a paper indicated that the behavioral therapy intervention had been previously published, the previously published protocol was evaluated for the specific components of behavioral therapy. For systematic reviews, the primary studies were scanned for a description of a cognitive aspect to the behavioral therapy. Reference lists of included studies were searched for any further possible inclusions.

Table 1 Search strategy
Full size table

Inclusion decisions were made by the chief investigator (BR) who also completed data extraction, including study objectives, description of participants and treatment groups, outcome measures, results and specific details of behavioral therapy. A second researcher (RD) identified any rationale presented for cognitive strategies or any referenced statements about behavioral therapy wherein the references may have elaborated on a rationale for cognitive strategies. Using the same approach, references within referenced papers were followed until no new references were found. Methodological quality and risk of bias in RCTs were assessed by two authors (BG, KO) using the Physiotherapy Evidence Database (PEDro) scale, with the scores reviewed and discussed by the other two authors (BR and RD). The PEDro scale is a reliable 11-item scale for assessing internal validity of RCTs [16]. Scores range from 0 to 10, with scores < 4 indicating poor, 4–6 fair and 7–10 high quality. For cohort studies, the Newcastle-Ottawa Scale (NOS) [17] was used; authors BR and RD independently assessed the papers and then discussed results until consensus was reached. Scores on the NOS are categorized as good, fair or poor quality based on assessment of participant selection, comparability and outcomes.

Results

The database search yielded 519 articles; 11 additional papers were identified by reviewing references. Thirty papers included behavioral therapy in the intervention for OAB and were evaluated for any cognitive component in the behavioral therapy protocol; four papers published between 2009 and 2019 met this criterion. A September 2020 notification from a saved search resulted in one additional paper for inclusion, for a total of five papers. Figure 1 outlines the selection process.

Fig. 1
figure 1

Selection process

Full size image

Studies were heterogeneous and were analyzed descriptively (no meta-analysis). Table 2 presents the treatment groups and the description of the cognitive component of behavioral therapy. Three studies were RCTs including incontinent women (Sung [18], Kaya [19], Firinci [20]); Sung and colleagues evaluated whether adding behavioral therapy to sling surgery for UI was helpful, Kaya and colleagues investigated whether behavioral therapy enhanced the effects of PFMT on OAB symptoms, and Firinci and colleagues assessed bladder training, biofeedback and electrical stimulation in women with idiopathic OAB. The other two studies were single-arm prospective studies including men and women. Lee and colleagues [21] evaluated the effect of multidimensional behavioral therapy on OAB (including ‘OAB dry,’ i.e., urgency without accompanying incontinence), and Klutke and colleagues [22] evaluated the addition of behavioral therapy to medication in incontinent individuals who were dissatisfied after 8 weeks of medication alone. Only Lee [21] and Kultke [22] included specific measures of urgency as outcomes. Table 3 summarizes the study objectives, participants, outcome measures and results.

Table 2 Treatment groups and cognitive component of the treatment protocol
Full size table
Table 3 Study characteristics and results
Full size table

Regarding descriptions of the cognitive component of behavioral therapy, detail varied substantially among the five studies. Two (Sung [18], Klutke [22]) gave details of or reference to patient handouts which described exactly what participants might focus on/think about when trying to manage urgency episodes. The handouts themselves were unreferenced. The other three studies provided brief descriptions of the cognitive component of their behavioral therapy intervention in the methods. No study addressed cognitive components of behavioral therapy in the discussion. Four of the five studies suggested distraction (e.g., reciting a poem, counting backwards, making a to-do list) as a cognitive approach for controlling urgency (Kaya [19], Klutke [22], Sung [18], Firinci [20]), while Lee [21] suggested that patients ignore/stop thinking about urgency. Other cognitive components included focusing one’s attention on breathing (Klutke [22], Kaya [19], Firinci [20]) and making self-affirming statements such as “I can wait” or “I am in control” (Klutke, Kaya, Sung, Firinci)” [18,19,20, 22].

Regarding the rationale for cognitive components and reference tracing, of the five studies included in this review, only Lee [21] provided a reference for their cognitive strategy; this was to their previous work, which itself provided no further details. Two of the other four studies contained referenced statements about behavioral therapy (Klutke [22] and Newman [23]). Table 4 presents referenced statements from each of the five studies and tracing of the references given. Newman [23] (2018) presented the behavioral therapy used in Sung [18] (2019) and gave 11 references regarding behavioral therapy. Reference tracing for two generations (i.e., looking for references in these 11 papers as well as in papers that were included in their reference lists and one more time in subsequent reference lists of papers that again had relevant referenced statements) revealed another four references. Lee [21] provided two references, neither of which presented any further references, and Klutke [22] gave six references regarding behavioral therapy; two generations of reference tracing found no references beyond those already identified in the reference tracing of Newman [23]. In total, 20 references for cognitive components of treatment were identified between the five studies: 13 clinical trials (1981–2011), 6 literature reviews (1986–2009) and a book (1989) (Table 4).

Table 4 Reference tracing timeline
Full size table

Fourteen of the 20 references about behavioral therapy contained one or more of three common authors: Burgio (n = 9) [7, 9, 24,25,26,27,28,29,30], Newman (n = 3) [7, 31, 32] or Wyman (n = 4) [7, 33,34,35]. These authors could as such be considered pioneers in the field, whose ideas on behavioral therapy have been replicated since 1985 [9]. The earliest study identified in the reference tracing was by Fantl in 1981 [36]; this paper definitively stated that the condition being treated (‘detrusor instability’) by their ‘bladder retraining drill’ was ‘postulated’ to be cortical in nature with a loss of inhibition of reflexes being the underlying mechanism and therefore the aim of treatment being to strengthen cortical control. They did not, however, describe details of the intervention. By 1991, another paper by Fantl (with Wyman as a co-author) provided a brief rationale for behavioral therapy, focused on inhibition of detrusor contractions and the possibility of ‘unspecified psychological benefits’ [33]. Of note, however, none of the references traced in support of behavioral therapy actually detailed potential cognitive or psychological mechanisms for urgency sensation or behavioral therapy as a treatment, nor did they test these postulations. The mode of action of cognitive strategies to treat urgency therefore remains unclear. This lack of clarity about mechanisms was reiterated in a review by Burgio (2004) [28], and the latest review by all three authors, published in 2009 [7] added nothing further on this point. The 2009 review may be considered the culmination of their collective ideas and contains the bladder training handout that was used in Klutke 2009 [22] (Wyman and Burgio were also authors on this paper). These ideas were also evident in the behavioral therapy program presented by Newman 2018 [23]. Hence, cognitive aspects of behavioral therapy appear to reflect accepted wisdom handed down over time, and no research has been identified which has proposed and tested physiological/psychological mechanisms.

Across the five studies included in this systematic review, 1169 participants (92% women) were included, mean age 49–60 years (Table 3). One study (Sung) [18] included women with SUI as well as MUI and UUI, but data were presented separately for the different diagnoses, allowing for specific analysis of women with OAB symptoms (MUI/UUI groups). Regarding the effectiveness of behavioral therapy that includes a cognitive component, no pooling of outcomes data was possible due to study heterogeneity. However, all five studies showed statistically significant improvements in OAB symptoms measured by validated instruments and/or high patient satisfaction with treatment. For the two studies which measured urgency specifically, Lee [21] found a reduction in the frequency of rushing to the toilet (measured by ICIQ-OAB), and Klutke [22] found a reduction in the number of urgency-related voids reported in the bladder diary. In the two RCTs that examined a behavioral therapy treatment with a specifically described cognitive component versus a treatment without this, the groups receiving behavioral therapy with a cognitive component experienced better results [18, 19]. However, the specific impact of the cognitive component of OAB could not be evaluated because none of the studies manipulated this variable. Regarding study quality and risk of bias, Sung et al. and Firinci et al. received a rating of “high quality” and Kaya et al. were rated “fair quality” on the PEDro scale, while Lee et al. and Klutke et al. received “poor quality” ratings on the NOS.

Discussion

This systematic review assessed the cognitive aspects of behavioral therapy interventions for OAB presented in five studies. Our first objectives were to describe these cognitive components and any rationale presented. Mental distraction as a means to control urinary urgency was the most common strategy but with no rationale presented for the choice of distraction over any other cognitive strategy. Reference tracing revealed three key authors who have described cognitive strategies over time. Furthermore, the historical rationale for cognitive strategies has never been explored separately from the rationale for behavioral therapy as a whole, in which the most common focus is on the rationale for the pelvic floor muscle exercise component. The rationale for behavioral therapy centers on the idea that patients learn to inhibit detrusor contractions [28, 33]. Our third objective was to evaluate the effectiveness of these interventions. Behavioral therapy with a cognitive component was effective in reducing signs and symptoms of OAB; these favorable results are in agreement with previous systematic reviews that evaluated PFM exercise, another component of behavioral therapy, for symptoms of OAB [11, 12]. However, despite apparent support for behavioral therapy as first-line treatment for OAB, the literature does not allow verification of the relevance of the different aspects of behavioral therapy. Indeed, we were not able to meet our third objective, determining the specific impact of a cognitive aspect of behavioral therapy, as none of the five studies manipulated that particular variable.

Features of the included studies warrant caution in interpreting the results of this review. The studies were heterogeneous regarding their methodological quality, interventions and outcomes, with their only commonality being that they included a cognitive component. Additionally, the age range of participants included in the studies was narrow, and despite the fact that OAB affects men almost as commonly as women [37], men were under-represented, as was OAB dry.

Another limitation is that despite cognitive components of behavioral therapy being directed at urgency control, urgency was only directly assessed in two studies, either as episode counting or the frequency of a behavioral response (rushing to the toilet). No study assessed whether the sensory experience of urgency itself changed. It has been recognized that limitations in urgency measurement methods have hampered research efforts into understanding and treating this bothersome sensation [38, 39]. Direct study of patient experiences of urgency and brain imaging studies demonstrate that the sensory experience of urgency has an affective component and that the brain areas involved are similar to those activated with other adverse sensations such as pain, dyspnea, pruritis and urge to cough [39,40,41,42]. This suggests that cognitive (including affective) aspects of OAB may play a large role in symptom severity and therefore a cognitive approach to behavioral therapy may be a key part of altering both symptom perception and behavioral response.

A striking aspect of this systematic review process was how few studies provided any description of a cognitive component of behavioral therapy for OAB (5/31). It is possible that other authors have implemented a cognitive aspect as part of their intervention, but without describing it. In fact, some authors have reported inclusion of strategies to reduce urgency, but did not report what comprised these strategies [29, 43, 44]. Other authors have described “relaxation” as part of the behavioral therapy response to urgency, but have not specified whether this refers to mental or physical relaxation [45, 46]. As recognized by Frawley and colleagues, even a physical exercise program will generally contain cognitive and behavioral elements which are rarely well described in reports [47]. In the literature on urgency management, not only are the cognitive and behavioral aspects of pelvic floor muscle exercise not well described, but strategies which are primarily cognitive in nature are particularly poorly described. The rationale for cognitive strategies was never given, and reference tracing revealed that cognitive strategies are substantially a hand down of ‘accepted wisdom,’ largely developed by three pioneers in the field (Burgio [9], Newman [31] and Wyman [34]) with untested rationale based on learning to control detrusor overactivity (DO), with other hypotheses such as psychological changes remaining untested [28, 33]. Indeed, a recently published ICS consensus statement on bladder training [48] re-confirms that mechanisms underpinning bladder training as a whole remain poorly understood, and the only references provided for a brief descriptive section on urgency suppression strategies were two other reviews: the review from 2009 [7] by the three pioneer authors and another by Burgio in 2002 [38].

Of the different components of behavioral therapy for OAB, such as urgency control and fluid manipulation, PFM exercise for urgency control has received the most attention in the literature. However, the relative contribution of PFM contraction versus cognitive strategies to urgency control is not understood, and many questions remain as to the mechanisms by which each may be effective [28]. For both, the predominant rationale presented has been the inhibition of DO [10, 11, 28], but this remains untested and is, in fact, problematic. It is intuitively appealing and relates to early research demonstrating that pelvic floor contraction does indeed reduce detrusor pressure and increase urethral pressure [8, 9]. However, this theory does not address that less than half of patients with urgency have urodynamic DO [49, 50] and that DO has long been challenged as a central underpinning mechanism of urgency [51]. If PFM contraction or other urgency suppression techniques do not work by reducing DO then how do they work? Indeed, Colombo (1995) [52] demonstrated that with behavioral therapy, urinary incontinence improvement was not associated with urodynamic changes in DO. This idea was borne out by Griffiths and colleagues who used functional MRI to investigate brain activation in women with UUI [53]. Interestingly patients’ responses to PFM exercise were predicted by their brain activity on urgency prior to starting training; those with activity that is most similar to the presumed ‘neurotag’ of DO were less likely to respond to PFM exercise. This again suggests that ‘sensory urgency’ (i.e., urgency not involving DO) better responds to the pelvic floor muscle training aspect of behavioral training and calls into question the commonly quoted rationale of detrusor inhibition.

Alternative hypotheses for the mechanisms underlying the effectiveness of behavioral therapy components are possible. For example, is a pelvic floor muscle contraction purely a physical strategy that inhibits DO, or is a patient’s awareness of and confidence in this contraction key to its success (i.e., impacting the cognitive evaluation of sensation and affective response)? How successful is either suppression or distraction? No studies have actually tested this aspect of behavioral management specifically, so we do not know if patients actually use these strategies, let alone how successful they are. Recent research has indicated that distraction may be counterproductive in terms of increasing reaction times for pelvic floor muscle contractions [13, 14]. Could it be that there is a difference between true distraction (such as attending to an unexpected phone call) and ‘deliberate distraction’ (such as counting backwards from 100), and what are the mechanisms behind distraction and a reduction (or disappearance) of the sensation of urgency? In pain literature distraction and suppression have been compared, with distraction generally faring better than suppression, which can result in ‘rebound’ (an unwanted reappearance of the suppressed thought) [54]. Finally, is it possible that cognitive strategies other than distraction or suppression may be effective?

While ‘urgency suppression’ strategies have been the most commonly described cognitive strategies in clinical trials of behavioral therapy, other approaches have been reported which illustrate that psychological aspects of OAB are relevant. Mindfulness-based stress reduction (MBSR) as a stand-alone intervention (no PFM exercise and no bladder-related lifestyle advice) has been demonstrated to have a positive impact on OAB [55]. MBSR approaches emphasize non-judgmental awareness of sensory experiences and as such could be argued to be directly opposite in nature to ‘suppression.’ Even earlier than this, a trial of hypnotherapy was highly effective for a subset of patients again illustrating that psychological treatment approaches can be effective for OAB [56]. An adapted cognitive-behavioral approach for OAB has also been described [57]. Of note, this approach has been used for patients without DO, categorized as having ‘sensory urgency.’ The focus of this treatment was to have patients understand the relationship between their cognitive and affective responses to urgency and the behavioral response. The authors emphasized the importance of patients developing their understanding and skill in continence control via an understanding of cognitive and affective responses to sensation and also via learning PFM contraction. Thus, physical skill, cognitive understanding and affective responses were entwined. This concept of ‘learning’ continence control (including physical and cognitive components) also harks back to early descriptions of behavioral therapy, which describe a purpose of ‘training the will and mind’ [58, 59] or a ‘skill acquisition therapy’ [28].

The fact that many reports make a distinction between ‘sensory’ and ‘motor’ urgency (i.e., urgency with or without associated DO) raises the important question of ‘phenotypes’ of OAB. Griffith’s MRI study referred to above clearly demonstrated two different types of urgency as represented by brain activation patterns [53]. As a symptom complex rather than a diagnosis of pathology, OAB probably encompasses many different underlying mechanisms [60], and each may respond differently to different aspects of behavioral therapy. In addition to ‘sensory’ versus ‘motor’ subtypes of OAB, another common distinction has been OAB ‘wet’ versus ‘dry’ (i.e., with or without incontinence). Other potential subtypes of urgency sensation reported in the literature include varying degrees of suddenness [61], varying relationship to bladder volume and filling rates [62, 63] and variability in the quality of sensation (stronger normal urge versus a completely different sensation) [64]. Other than DO, other possible underlying mechanisms of urgency include altered urothelial/neural signaling [65], impaired bladder circulation [66, 67], structural abnormities (such as prolapse) [68], central sensitization, anxiety and chronic stress [69,70,71,72] and altered central sensory processing [41]. There may also be overlap in underlying mechanisms (e.g., chronic stress affects urothelial signaling) [65], and cortical changes seen with anxiety may converge causally with central changes associated with OAB sensation [73]. Better understanding of phenotypes is likely to result in better directed treatments [60]. This may also allow a more clearly elucidated rationale and sophisticated development of behavioral management strategies including urgency management strategies specifically.

Urgency is an adverse sensation that shares much in common with other adverse sensations such as pain, and much may be learned by comparing and indeed contrasting research into urgency management with research into the management of other adverse sensations. Concepts from research into chronic pain which may transfer well to the treatment of chronic urgency and provide a more sophisticated rationale for behavioral training techniques could include neurophysiological education [74], graded exposure [75], acceptance and commitment strategies [76] and visual illusions [77]. For example, just as patients’ pain experiences change when they better understand all the factors that contribute to the creation of pain as a sensory experience [78], so patients with urgency may notice differences if they learn to recognize and understand multiple contributing factors to urgency. Similarly, principles of graded exposure to painful activities may translate well to graded exposure to conditions that induce urgency. Strategies such as mindfulness and the psychological principles underpinning acceptance and commitment therapies teach patients to respond differently to sensory experiences. Finally, if the visual illusion of a fit, strong back might reduce someone’s back pain [77], could the visual illusion of a calm, relaxed bladder alter someone’s experience of bladder sensation?

Development of cognitive strategies for the treatment of urgency beyond the limited ideas presented in published literature, centered around the ideas of ‘suppression’ or ‘control,’ will be an important focus for future research. Future studies should (1) thoroughly describe cognitive components of behavioral therapy, (2) present a scientific rationale for cognitive components of behavioral therapy, (3) compare behavioral therapy with a cognitive component vs. without, (4) compare different types of cognitive interventions, such as distraction versus mindfulness, and (5) include a more representative range of participants. Further research may also elucidate which phenotypes of OAB respond to which treatments, including cognitive strategies.

Conclusion

In conclusion, the cognitive components of behavioral therapy for OAB are neither well described nor rationalized, but behavioral therapy that includes a cognitive component shows promise for OAB treatment. The relative impact of this part of behavioral therapy cannot be ascertained. Given the affective nature of OAB, the connection between OAB and mental health, and the promise shown by cognitive-based strategies in the treatment of other adverse sensations, increased focus on cognitive interventions for OAB may be warranted. Future studies should thoroughly describe and present a scientifically justifiable rationale for the cognitive components of behavioral therapy and should manipulate this variable so its specific impact can be ascertained.