Resilience among adults with chronic urologic pelvic pain syndrome: A flexible self-regulation perspective.

The current study was a secondary data analysis of baseline data of participants that have enrolled to date for an ongoing NIH-funded clinical trial for UCPPS 39 . Participants included 113 ( M age = 43.56, SD = 14.79, 81.4% female) individuals who were recruited for an NIH-funded behavioral pain intervention trial for chronic pelvic pain through community (e.g., targeted advertisement, word of mouth) and provider referrals (e.g., urology, urogynecology, pelvic floor therapy therapists, primary care clinics) in the greater Western New York area 39 . Participants completed self-administered forms on a study-issued tablet during an in-person baseline assessment that confirmed study eligibility. Inclusion criteria for the overall study included being between the ages of 18–70, a physician-confirmed diagnosis of IC/BPS or CP/CPPS, the presence of pelvic pain of at least moderate severity (≥3/10 NRS) for at least 6 months in duration, ability to understand and provide informed consent, and willingness to participate in all components of the clinical trial (for study full study protocol and procedures, see Lackner et al. 39 ). Exclusion criteria included the presence of a neurological condition affecting the bladder, presence of symptomatic urethral stricture, history of cystitis caused by tuberculosis or cancer-directed therapies, pelvic-related malignancy, unstable medical or psychiatric condition (where therapeutic demands to stabilize a comorbid condition would interfere with the ability to comply with study procedure/perceived vulnerability that would compromise participant safety), gross cognitive impairment (severe or obvious deficits in mental functioning that would significantly impacts a person’s ability to understand process, information, and make informed decision), involved in current psychotherapy for pelvic pain relief, and inability to ready or fluently speak English. All study procedures were approved by the Institutional Review Board at the University at Buffalo. Participants completed a self-report battery of questionnaires assessing sociodemographic variables (age, race/ethnicity, etc.), as well as medical symptoms (UCPPS and comorbidities across 12 body systems and diseases (e.g., skin, neurological, cardiovascular, cancer, etc.)). Further, a variable was created to uniquely quantify the number of chronic overlapping pain conditions (COPCs) for each participant. Participants were also asked if they had a physician-confirmed COPC diagnosis over the past 3 months. COPCs include fibromyalgia, irritable bowel syndrome, interstitial cystitis, chronic prostatitis, vulvodynia, migraine, chronic tension-type headache, temporomandibular disorder, chronic low back pain, chronic fatigue syndrome, and endometriosis. Pelvic pain and urinary severity were assessed using the Genitourinary Pain Index (GUPI) 41 and the Interstitial Cystitis Problems and Symptoms Index (ICPI) 42 . Separate pelvic pain (range 0–28; α = 0.785) and urinary (range 0 – 25; α = 0.793) symptom severity indices were calculated by summing respective items from the GUPI (10 pain, 1 urinary) and ICSI (2 pain, 3 urinary) 43 . This approach, which has been validated in the MAPP Research Network Pelvic Pain study 1 , 44 , is preferrable to using composite scores that include both pain and urinary symptoms in a single score (e.g., GUPI total score) which may not provide adequate sensitivity to detect change in these two distinct symptom domains. Quality of life was measured with the PROMIS Scale Global Physical and Mental Health scales 45 , and the CDC Health-related Quality of Life measure healthy days core module 46 . The PROMIS Scale Global Physical (α = 0.766; e.g., “To what extent are you able to carry out your everyday physical activities such as walking, climbing stairs, carrying groceries, or moving a chair?” ) and Mental Health (α = 0.788; e.g., “In general, how would you rate your mental health, including your mood and your ability to think?” ) scales 45 each include 2-item self-report measures assessing perceived physical and mental health impact, rated on a 5-point Likert scale from 1 ( Poor ) to 5 ( Excellent ), and creating a sum score (0–10). The CDC Healthy Days measure includes three items assessing “days” of poor mental health, physical health, and activity limitations (e.g., “For how many days during the past 30 days was your physical health not good” ), with the sum of responses to the number of poor mental and physical health days yielding an overall “unhealthy days” when physical or mental health was not good over the previous 30. Additionally, one item assessed global self-perceived health rated from 1 ( Excellent ) to 5 ( Poor ), with higher scores indicating worse self-rated health. Both PROMIS and Healthy Days are psychometrically sound and efficient (in terms of administration time) instruments whose HRQOL profile is sensitive to treatment effects and not confounded by personality variables that burden the interpretability of other health-related QOL measures 47 , 48 . These measure each capture distinct aspects of quality of life, with CDC Healthy Days anchored in burden of disease-specific quality of life impairment, and PROMIS focused on global physical and mental health quality of life, respectively 49 . Components of flexible self-regulation were assessed using the Context Sensitivity Index (CSI) 50 and the Coping Flexibility Scale-Revised (CFS-R) 51 . The CSI is a self-reported scenario-based measure that characterizes the ability to accurately perceive the presence or absence of cues across different situations (context sensitivity). The CSI presents participants with 6 hypothetical stressful scenarios and asks participants to rate their perception of cues (e.g., “A friend calls and asks you to do a favor for their partner, whom you don’t like. How much control do you have over what happens next?” ), including threat, controllability, urgency and cooperation, rated on a 7-point Likert scale from 1 ( Not at all ) to 7 ( Very Much ) 52 . The CSI generates a cue-present and cue-absent score that represents an individual’s ability to consider context (or not) 53 . Higher scores represent a stronger inclination to perceive the presence of contextual cues or insensitivity to the absence of contextual cues. The CFS-R assesses an individual’s ability to flexibly adapt and modify coping strategies based on contextual feedback (feedback monitoring). Specifically, the CFS-R is a 12-item self-report measure assessing three distinct components focused on the evaluation of ineffective coping with aversive stimuli, including the ability to abandon an ineffective coping strategy (coping abandonment: “I can stop using a failed coping strategy ”), identify and implement an alternative and more effective strategy (re-coping: “If the situation has not improved, I consider a different coping strategy” ), and the ability to monitor coping effectiveness and determine if one should repeat the coping abandonment and re-coping cycle (meta-coping: ”I know if a coping strategy has been successful or unsuccessful” ). Items are rated on a 4-point Likert scale from 1 ( Very applicable ) to 4 ( Not applicable ). The CSI and CFS-R have shown strong reliability and validity in associations with psychological health in the context of pain conditions 37 , 38 , 54 . The ENRICHD Social Support Instrument (ESSI), a 7-item self-report scale, was used to measure participants’ range of social support in their life 55 (e.g., “Is there someone available to you whom you can count on to listen to you when you need to talk?” ). Participants rated items related to structural, instrumental, and emotional support, rated on a 5-point Likert scale from 1 ( None of the time ) to 5 ( All of the time ). Higher scores indicated higher levels of social support (α = 0.901), and negatively correlates with depression, and positively correlated with social functioning in medical populations 56 . Pain Resilience Scale – Short Form 57 measures a person’s ability to self-regulate thoughts and emotions and behavioral persistence in the face of sustained or intense pain (e.g., “When faced with intense or prolonged pain…” ). Four items from the original scale were included in the short-form (items 2 - … I still work to accomplish my goals , 3 - … I push through it , 10 - … I keep a hopeful attitude , and 11 - … I don’t let it get me down )) 58 , and each item was rated on a 5-point Likert scale from 0 ( Not at all ) to 4 ( All the time ) with higher scores associated with less pain interference (α = 0.861). The Pain Resilience Scale has also been used to examine pain-specific resilience in heterogenous chronic pain populations 59 . Positive affect was assessed using the Positive and Negative Affect Schedule (PANAS) 60 . Its positive affect scale assesses the presence of a range of positive emotions over the past week measured on a 5-point Likert scale from 1 ( Very slightly or not at all ) to 5 ( Extremely ; α = 0.907); 10 items assess positive affect (e.g., “Indicate the extent you have felt excited over the past week” ), and these items have been associated with behavioral, emotional, and quality of life outcomes across clinical populations 61 , 62 . Pain catastrophizing was assessed using the Coping Strategies Questionnaire – 2 item version (CSQ-2) 63 . The CSQ-2 is a self-report measure assessing the frequency of negatively skewed pain appraisals (e.g., “It is terrible and I feel it is never going to get any better ”), rated on a 7-point Likert scale from 1 ( Never do ) to 7 ( Always do that ; α = 0.860), and is related to psychological aspects of pain among chronic pain samples 64 . Negative affect was assessed using the Positive and Negative Affect Schedule (PANAS) 60 . Its negative affect scale assesses the presence of a range of negative emotions over the past week, measured on a 5-point Likert scale from 1 ( Very slightly or not at all ) to 5 ( Extremely ; α = 0.873); 10 items assess negative affect (e.g., “Indicate the extent you have felt distressed over the past week” ), and these items have been associated with behavioral, emotional, and quality of life outcomes across clinical populations 61 , 62 . Sample characteristics were described with mean and SDs for continuous variables, and numbers and percentages for categorical variables. Multiple linear regression models examined the impact of aspects of context sensitivity (cue present and absent), components of coping flexibility (abandonment, re-coping, meta coping), pain resilience, positive affect, social support, catastrophizing, and negative affect on self-rated health, total number of unhealthy days, PROMIS global physical and mental health, and MAPP pain and urinary severity. Control variables included theoretically relevant covariates, including age 9 , sex 12 , 65 , and number of COPCs 66 that have been associated with UCPPS symptoms, pain, and quality of life outcomes in prior research. Data analyses were conducted with R. First, we examined descriptive statistics and correlations among study variables. Then, as preliminary analyses, we calculated multiple linear regressions including all protective and vulnerability factors as predictors to understand the unique contribution of each variable on outcomes, while accounting for shared variance between variables. Given that number of healthy days is a count variable with skewed distribution, a Poisson regression model was used for this variable (with Incident Rate Ratios reported). Further, Minimum Detectable Effect Size calculations 67 , based on the sample size and number of predictors included in the model, indicate a minimum detectable R 2 of 0.15 for all models. Given the potential of reduced power in detecting significant effects due to potentially high correlation between predictors and small sample size, as well as specifying a more narrowly defined set of variables associated with each outcome, we employed stochastic search variable selection (SVSS) 68 , 69 for all models except the Poisson model, which has been used in past chronic pain research 70 . SVSS is a Bayesian variable selection method that provides information regarding the relative importance of a specific predictor while also accounting for the uncertainty of other predictors in the model. Ten thousand regression models (with a 1000 sample burn-in period) were sampled, and the inclusion of the variable in the final model was based on the proportion of times the variable was selected in each of the regression models (marginal inclusion probability (MIP) ≥0.5), which increases power and decreases the potential for a false positive result 71 , 72 . To avoid missing any predictors deemed critical to understanding outcomes for chronic pelvic pain, models were conducted in a series of steps. First, following initial multiple linear regression analyses, any statistically significant predictors ( p < 0.05) were retained. Then, following the SVSS procedure, any variables with MIP values between 0.4 and 0.5 were retained to determine if these variables, when considered in the context of other significant variables, predicted outcomes. The final set of variables included in each model included significant variables from the initial model, variables with MIP values between 0.4 and 0.5, and variables with MIP values ≥0.5, which could be different for each model depending on model evaluation procedures as described above; variables were retained in the final model regardless of statistical significance. For the Poisson model, we utilized a forward and backward stepwise model selection procedure, based on the Akaike Information Criteria (AIC) that iteratively identifies optimal models, by balancing goodness-of-fit indices with model complexity, selecting a subset of predictors that achieve the lowest AIC 73 . R 2 values were included as an index of relative importance of each predictor in the final models.

Most participants self-identified as White (92.9%) and female (81.4%). In terms of overall health, approximately half of the sample (47.8%) reported 30 unhealthy physical and mental health days over the past month, with an overall sample average of 20.23 days over the past 30. Additionally, participants reported an average of 10.49 ( SD = 7.59) physician-diagnosed medical comorbidities and 2.81 ( SD = 1.63) chronic overlapping pain conditions. See Table 1 for participant and clinical characteristics. For full descriptive statistics for study variables, including means and standard deviations, as well as their correlations with all other study variables, see Table 2 . Additionally, examining bi-variate correlations between context sensitivity and coping flexibility variables with other protective and vulnerability factors revealed differential patterns. Specifically, context sensitivity – cue presence was not associated with any other protective or vulnerability factors, context sensitivity – cue absent was significantly associated with decreased negative affect ( r = −0.287). Coping flexibility variables were each significantly positively associated with pain-specific resilience (abandonment: r = 0.23, re-coping: r = 0.35, meta-coping: r = 0.31), positive affect (abandonment: r = 0.39, re-coping: r = 0.49, meta-coping: r = 0.46), negative affect (abandonment: r = −0.34, re-coping: r = −0.27), and social support (abandonment: r = 0.27, re-coping: r = 0.24, meta-coping: r = 0.34). Pain catastrophizing was neither significantly associated with context sensitivity nor coping flexibility variables. Regression analyses identified a differential pattern of predictors associated with pain and quality of life outcomes (see Tables 3A and 3B ). Context sensitivity (cue absent), coping abandonment, and meta coping were the components of flexible self-regulation associated with outcomes. Specifically, difficulties in determining the absence of contextual cues about a situation was associated with a greater number of unhealthy days ( IRR = 1.01, p = 0.002, sr 2 = 0.01), corresponding with a 4% increase in the rate of unhealthy days. For coping flexibility, greater ability to abandon ineffective coping strategies (coping abandonment) was associated with better physical health (PROMIS; b = 0.11, p = 0.03, sr 2 = 0.04) and fewer unhealthy days ( IRR = 0.97, p < 0.001, sr 2 = 0.03). Further, greater endorsement of engaging in meta-coping was associated with greater number of unhealthy days ( IRR = 1.04, p < 0.001, sr 2 = 0.03). Pain resilience was associated with increased PROMIS mental health scores ( b = 0.45, p = 0.02, sr 2 = 0.04). Positive affect was associated with better self-rated health ( b = −0.03, p = 0.03, sr 2 = 0.05) and better PROMIS mental health ( b = 0.08, p < 0.001, sr 2 = 0.13). Social support was not associated with any pain or quality of life outcomes. Pain catastrophizing was associated with increased pain severity ( b = 1.32, p < 0.001, sr 2 = 0.14), and greater number of unhealth days ( IRR = 1.03, p = 0.03, sr 2 = 0.01). Negative affect was associated with greater number of unhealthy days ( IRR = 1.03, p < 0.001, sr 2 = 0.09) as well as poorer PROMIS mental health ( b = − 0.06, p = 0.007, sr 2 = 0.08).

The current study examined components of flexible-self regulation, along with known resilience-related protective and vulnerability factors, relative to painful UCPPS symptoms and quality of life (QoL). Results showed that (1) context sensitivity was associated with number of unhealthy days, and (2) aspects of coping flexibility were associated with two measures of health-related quality of life: healthy days (number of unhealthy days) and PROMIS physical health. Additionally, results suggest a differential pattern of previously established protective and vulnerability factors on outcomes. With respect to UCPPS symptoms, catastrophizing was the only vulnerability factor associated with pelvic pain severity. No other protective factors corresponded with severity of pelvic pain or urinary symptoms. While important to the overall clinical presentation of UCPPS, it is important to contextualize predictors of symptoms in the broader UCPPS literature that indicates that symptom severity is not necessarily associated with symptom duration or impairment 74 , which largely impacts QoL. For QoL impairment, there was more variability. Among vulnerabilities, negative affect and pain catastrophizing were associated with more unhealthy days and mental health-related QoL impairment. Additionally, across the majority of measures, poor outcomes were associated with a greater number of chronic overlapping pain conditions, a group of co-aggregating nociplastic pain condtions 75 , 76 with a strong centralized pathophysiology. These data reinforce the heavy patient burden these disorders impose 77 . Among resilience factors, two components of the flexibility sequence --- context sensitivity and coping flexibility --- were, as predicted, associated with QoL indices. Cue-absent context sensitivity was associated with a greater number of unhealthy days, capturing appraisals of both physical and mental health. In other words, a diminished ability to perceive and decode accurately situational demands without contextual cues corresponded with worse QoL. These data dovetail with the broader literature that has found that an inability to read a situation and decode cues for the purpose of deploying a situationally adaptive coping response correspond with poorer mental health outcomes 78 – 81 characteristic of chronic pain disorders like UCPPS 82 . Of note, context sensitivity was not associated with PROMIS mental or physical health. While findings relating context sensitivity and number of unhealthy days should be considered preliminary, it is also important to consider differences in the measures used: number of unhealthy days primarily relates to appraisals of health and disease-specific burden (e.g., “days of bad health”) compared to the impact of emotions on health and limitations in social and recreational aspects of health. These important measurement differences warrant replication to provide additional confirmatory evidence. Positive affect was associated with better self-rated and mental health. Dispositional pain resilience, on the other hand, was only associated with pain interference and QOL impairment due to mental health problems. Additionally, coping abandonment (the ability to relinquish a coping response recognized as ineffective) was associated with fewer unhealthy days and better physical health, and to a lesser extent self-rated health, and meta coping (the ability to monitor and to provide corrective feedback on the adaptiveness of coping responses) was associated with greater number of unhealthy days. Data echo broader research that implicates meta coping as key component of cognitive flexibility, helping individuals terminate the use of a maladaptive coping strategy in favor of a more adaptive one 25 , 83 . Importantly, meta-coping is indirectly involved in coping and QoL outcomes, working in tandem with abandonment coping and re-coping 51 . Given the cross-sectional nature of the findings, it is possible to interpret a bi-directional relationship, such that greater number of unhealthy days is associated with more meta-coping, which would make intuitive sense by requiring adjustment of coping strategies with poorer appraisals of health. Leveraging this finding with coping abandonment being associated with fewer unhealthy days, it is likely that there is a synergistic, sequential processes at play that cannot adequately be captured in cross-sectional data; longitudinal replication is necessary to untangle the dynamic processes underlying flexible self-regulation and pain. By linking specific components of the flexibility sequence to pain-related outcomes, this study helps contemporize our understanding of resilience in individuals with chronic pelvic pain by drawing on coping and emotion regulation research findings that resilient outcomes occur though flexible self regulation 20 . Flexibility in general is critical to adapting to the multiple -- and oftentimes simultaneous -- physical and mental demands of chronic pain 84 , 85 . Applying a contextual framework has potential to arrive at a more fine-grained understanding of how and why resilient outcomes develop 35 , which is beyond the scope of trait-like processes such as optimism, hope, and meaning making 86 . While undoubtedly critical pieces of the “resilience puzzle”, their dispositional nature may mean they are most important as predisposing (vs maintaining) factors that “provid[e]” the motivation for flexible responding” 27 , 40 (see Figure 1 ). This suggest that these factors impact cognitive competencies critical to engaging in FSR and achieving resilient outcomes 22 . This study sheds light on operative components of FSR that may underlie resilient outcomes in a high-impact pain disorder. Flink et al 36 found contextually insensitivity to social cues characterized individuals with more severe pain problems. Kato found that coping flexibility corresponded with lower depressive symptoms in individuals with headache 37 and menstrual pain 38 . Neither study simultaneously assessed the relationships between multiple components of FSR (context sensitivity and coping flexibility) with resilience outcomes relative to other risk and protective factors in a clinical population. Further, both studies relied on less conservative statistical approaches susceptible to false positive results than the data-driven approach we adopted here. This study improves on prior pain research that has focused on single components of the flexibility sequence 36 – 38 . That said, we did not formally assess patients access to a wide range of strategies (repertoire) to meet situational demands, the third pillar of FSR. Its influence on pain outcomes alone and relative to other componential FSR abilities is an important area for further study. Clinically, these data have important implications. If FSR is an integral aspect of navigating the day-to-day burden of a pain disorder like UCPPS, there may be therapeutic benefit in targeting the dimensions of FSR for each patient by implementing contextually sensitive strategies. For some patients, the key to improving resilience may involve strengthening their ability to read and decode situational cues (e.g., context sensitivity), such as proactively reading the controllability of a stressor to deploy a situationally appropriate coping strategy. For others, the goal is teaching an adequate set of emotion- and problem-focused coping responses in their symptom self-management “toolbox” 87 . For others, it is learning how to monitor the effectiveness of their coping efforts and marshal a more effective alternative response when their first - or second or third - attempt falls short. The advantage of a situationally sensitive therapeutic approach, in principle, is that it calibrates patient-level resilience deficits to concrete remediation efforts that strengthens patients’ ability to self-manage symptom burden. Rather than deploying an achievement-oriented problem solving approach across all situations as conventional approaches precribe 88 , a contextually sensitive approach teaches the patient to make more objective appraisal of situational controllability, for example, as a way of deploying the most adaptive coping response the situation warrants 89 even when it involves acceptance or resignation 89 . This approach has conceptual advantages over more generic resilience enhancement approaches that presume that strategies (e.g., positivity, behavioral persistence) 21 are uniformly effective across contexts in which the individual interacts (e.g., health, job, relationships). Because the situational context is ever changing, no one strategy is uniformly adaptive or maladaptive across the board 27 . What is important, is being flexible as a situation unfolds 90 , which means increasing an individual’s capacity to respond to situational demands in a contextually appropriate, proactive and efficient manner. An FSR approach also recognizes that the therapeutic benefit of a behavioral approach is not defined simply by the application of prescribed strategies but also ones’ ability to monitor and receive corrective feedback about whether prescribed strategies work or an alternative coping response is called for (i.e., meta coping). An emphasis on meta-coping addresses a significant gap in behavioral pain treatment which, for the most part, provides little guidance for patients and providers about when, how, and why a prescribed strategy falls short. A contextual model emphasizing strategy-situation fit not only helps explain why (e.g., context insensitivity, coping inflexibility) some people struggle to navigate adverse contexts to achieve resilient outcomes but offers theory-driven, empirically grounded, practical guidance for how to remediate resilience deficits to reduce symptom burden. The current study is not without limitations. Its cross-sectional design precludes making causal inferences about the relationship between components of the flexibility sequence and resilience outcomes. Additionally, components of FSR were not consistently associated with assessments of QoL. Quality of life is a complex construct that includes general and disease specific concepts that largely capture the impact of disease on an individual. Examining the parameter estimates in the variable selection models does provide support for components of FSR being selected as “important” predictors in physical health-related quality of life metrics, and lack of significant findings could relate to power. Replication of the findings utilizing prospective research, including both short- and long-term assessment, will be important to assess the predictive impact of the FSR components on pain and QoL outcomes. Real-time monitoring using ecological momentary assessment has the potential to gain an idiographic perspective of diverse contexts that influence the resilience trajectory at the patient (vs. aggregate) level. Because our sample was primarily female and white, it may not be representative of the entire population of individuals with chronic pelvic pain. Additionally, all participants were treatment-seeking and therefore may not represent most people with chronic pelvic pain. Overall, the study builds upon the emerging emotion regulation and pain resilience literature by linking contextually sensitive cognitive processes to outcomes clinically meaningful to patients with high impact chronic pain disorders like UCPPS. Rather than framing resilience strictly as a dispositional quality, findings argue that it is multifactorial and informed by trait, dynamic, and contextual mechanisms with potential for generating personalized care that places patients’ unique needs, strengths, and vulnerabilities at the center of clinical decision making. If contextually sensitive concepts, like flexible self-regulation, have greater explanatory value, their derivative therapeutic strategies stand to be more robust by virtue of teaching patients a portable regimen of self-management skills whose benefit cuts across different situational contexts. This would represent a novel way of optimizing the therapeutic profile of most behavioral pain treatments 91 .