This paper is in the following e-collection/theme issue:

Formative Evaluation of Digital Health Interventions (2410) Digital Pain Assessment and Management (164) Digital Mental Health Interventions, e-Mental Health and Cyberpsychology (1514) Chatbots and Conversational Agents (531)

Published on in Vol 7 (2023)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/51422, first published .
Behavioral Mechanisms That Mediate Mental and Physical Health Improvements in People With Chronic Pain Who Receive a Digital Health Intervention: Prospective Cohort Pilot Study

Behavioral Mechanisms That Mediate Mental and Physical Health Improvements in People With Chronic Pain Who Receive a Digital Health Intervention: Prospective Cohort Pilot Study

Behavioral Mechanisms That Mediate Mental and Physical Health Improvements in People With Chronic Pain Who Receive a Digital Health Intervention: Prospective Cohort Pilot Study

Authors of this article:

Abby L Cheng1 Author Orcid Image ;   Mansi Agarwal2 Author Orcid Image ;   Melissa A Armbrecht1 Author Orcid Image ;   Joanna Abraham3 Author Orcid Image ;   Ryan P Calfee4 Author Orcid Image ;   Charles W Goss2 Author Orcid Image
Article Authors Cited by Tweetations (2) Metrics

    Original Paper

    1Division of Physical Medicine and Rehabilitation, Department of Orthopaedic Surgery, Washington University School of Medicine, St Louis, MO, United States

    2Institute for Informatics, Data Science and Biostatistics, Washington University School of Medicine, St Louis, MO, United States

    3Department of Anesthesiology and Institute for Informatics, Washington University School of Medicine, St Louis, MO, United States

    4Division of Hand and Wrist, Department of Orthopaedic Surgery, Washington University School of Medicine, St Louis, MO, United States

    Corresponding Author:

    Abby L Cheng, MSc, MD

    Division of Physical Medicine and Rehabilitation

    Department of Orthopaedic Surgery

    Washington University School of Medicine

    Campus Box 8233

    660 South Euclid Avenue

    St Louis, MO, 63110

    United States

    Phone: 1 3147472823

    Fax:1 3147472597

    Email: chengal@wustl.edu


    Background: Preliminary evidence suggests that digital mental health intervention (Wysa for Chronic Pain) can improve mental and physical health in people with chronic musculoskeletal pain and coexisting symptoms of depression or anxiety. However, the behavioral mechanisms through which this intervention acts are not fully understood.

    Objective: The purpose of this study was to identify behavioral mechanisms that may mediate changes in mental and physical health associated with use of Wysa for Chronic Pain during orthopedic management of chronic musculoskeletal pain. We hypothesized that improved behavioral activation, pain acceptance, and sleep quality mediate improvements in self-reported mental and physical health.

    Methods: In this prospective cohort, pilot mediation analysis, adults with chronic (≥3 months) neck or back pain received the Wysa for Chronic Pain digital intervention, which uses a conversational agent and text-based access to human counselors to deliver cognitive behavioral therapy and related therapeutic content. Patient-reported outcomes and proposed mediators were collected at baseline and 1 month. The exposure of interest was participants’ engagement (ie, total interactions) with the digital intervention. Proposed mediators were assessed using the Behavioral Activation for Depression Scale–Short Form, Chronic Pain Acceptance Questionnaire, and Athens Insomnia Scale. Outcomes included Patient-Reported Outcomes Measurement Information System Anxiety, Depression, Pain Interference, and Physical Function scores. A mediation analysis was conducted using the Baron and Kenny method, adjusting for age, sex, and baseline mediators and outcome values. P<.20 was considered significant for this pilot study.

    Results: Among 30 patients (mean age 59, SD 14, years; 21 [70%] female), the mediation effect of behavioral activation on the relationship between increased intervention engagement and improved anxiety symptoms met predefined statistical significance thresholds (indirect effect –0.4, 80% CI –0.7 to –0.1; P=.13, 45% of the total effect). The direction of mediation effect was generally consistent with our hypothesis for all other proposed mediator or outcome relationships, as well.

    Conclusions: In a full-sized randomized controlled trial of patients with chronic musculoskeletal pain, behavioral activation, pain acceptance, and sleep quality may play an important role in mediating the relationship between use of a digital mental health intervention (Wysa for Chronic Pain) and improved mental and physical health.

    Trial Registration: ClinicalTrials.gov NCT05194722; https://clinicaltrials.gov/ct2/show/NCT05194722

    JMIR Form Res 2023;7:e51422

    doi:10.2196/51422

    Keywords

    digital mental health intervention; chronic musculoskeletal pain; anxiety; depression; pain interference; physical function; behavioral activation; pain acceptance; sleep quality; mediation analysis; behavioral mechanism; chronic pain; digital health intervention; mobile phone 


    Background

    Chronic pain commonly coexists with depression and anxiety, and these conditions are bidirectional risk factors for one another [1-5]. Fortunately, behavioral interventions such as cognitive behavioral therapy and sleep interventions can simultaneously improve symptoms related to mental health and chronic pain [6-9]. Three behavioral targets that may mediate these changes include: (1) behavioral activation, which is the process of identifying and engaging in activities that are meaningful and enjoyable, regardless of a person’s current symptom state [10-13]; (2) pain acceptance, which is “acknowledging that one has pain, giving up unproductive attempts to control pain, acting as if pain does not imply disability, and [committing] one’s efforts toward living a satisfying life despite pain” [14-19]; and (3) sleep quality, which includes a person’s pattern of falling asleep, staying asleep, total sleep duration, sleep architecture, and restorative effect of sleep [20-24]. Our preliminary work suggests that a digital mental health intervention (Wysa for Chronic Pain) which was designed to address these 3 behavioral targets can improve self-reported mental and physical health in patients who present for orthopedic management of chronic musculoskeletal pain and have coexisting symptoms of depression and anxiety [25,26]. However, mediation effects by these 3 behavioral mechanisms through which this intervention is proposed to act have not been formally tested. Before a full-sized randomized controlled trial is conducted to determine the effectiveness of this digital mental health intervention, a better understanding of the behavioral targets which it engages is needed.

    Goal of This Study

    The purpose of this pilot study was to identify whether there is preliminary evidence that the behavioral mechanisms of behavioral activation, pain acceptance, and sleep quality mediate the relationship between increased engagement with a digital mental health intervention (Wysa for Chronic Pain) and improvements in mental and physical health among patients who present for orthopedic management of chronic musculoskeletal pain. We hypothesized that improved behavioral activation, pain acceptance, and sleep quality would mediate the relationship between increased engagement with the digital mental health intervention and improved self-reported anxiety, depression, pain interference, and physical function at 1 month.


    Design

    This was a single-site, single-arm, prospective cohort pilot study that was conducted at a tertiary care academic medical center in the United States. Participants were enrolled from January 2022 through May 2022. This study was prospectively registered through ClinicalTrials.gov (NCT05194722). Participants in this study also participated in a broader qualitative investigation regarding whether and how to address mental health in the orthopedic care setting [27,28].

    Ethical Considerations

    Institutional review board approval (IRB #: 202110165) was obtained from he Washington University Institutional Review Board prior to participant recruitment. Participants provided written consent prior to participation, and they received remuneration for participation in the form of a US $40 gift card. To maintain participants’ privacy and confidentiality, all study data were stored in a secure electronic REDCap (Research Electronic Data Capture; Vanderbilt University) database [29,30].

    Participants

    Participants were recruited from patients who presented to an orthopedic spine specialist within the institution’s orthopedic department. Eligibility criteria were assessed via medical record review and confirmation with the patient. Potential participants were contacted via telephone prior to their upcoming clinic visit with the orthopedic spine specialist or in person on the day of their visit. To be eligible, patients had to be an adult (18 years or older) who presented for a new or return evaluation of chronic (≥3 months) neck or back pain. Patients were excluded if they presented for a routine postoperative visit, endorsed active mental health crisis (eg, suicidal or homicidal ideation or psychosis), or had cognitive impairment which would interfere with engaging in this study’s surveys or intervention. The treating spine specialists included 11 physical medicine and rehabilitation physicians (physiatrists) with subspecialty training in pain medicine or sports medicine, 4 subspecialty-trained orthopedic spine surgeons, and 1 nurse practitioner who worked closely with the orthopedic spine surgery service.

    Intervention

    Study participants received a 1-month subscription to Wysa for Chronic Pain, which is a multicomponent digital mental health intervention that uses an artificial intelligence–based conversational agent (ie, chatbot) and human “coaches” (counselors) with master’s degrees in psychology to deliver therapeutic content including cognitive behavioral therapy, cognitive restructuring, motivational interviewing, mindfulness training, deep breathing techniques, and sleep meditations to collectively improve users’ behavioral activation, pain acceptance, and sleep quality [31-35]. This intervention guides users through a week-by-week curriculum that uses behavioral activation and pain acceptance principles to encourage users to engage with things that bring joy, despite having pain. Features include daily check-ins, nightly sleep meditations, weekly progress reports, a progress roadmap, and the ability to unlock premium “reward” tool packs by engaging with the weekly reports. The intervention encourages users to engage with the stepwise curriculum at least 3 times weekly. For the purpose of this study, participants were encouraged to use the intervention as much as they felt was helpful. Participants also received usual orthopedic care during the course of this study, as prescribed by their orthopedic clinician. Usual orthopedic care included analgesic medication, physical therapy, and interventional spine procedures, as appropriate. No participant had spine surgery during this study’s follow-up period.

    Measures

    Overview

    Data sources included a combination of participants’ self-report, their electronic medical records, and time-stamped usage reports from the digital intervention company (Wysa Inc). At baseline, standard clinical care data were collected from the electronic medical record including participants’ age, sex, race, ethnicity, orthopedic diagnosis, and pain medications, and participants also self-reported their pain duration, general smartphone use patterns, and current use of mental health treatment (eg, medications, psychotherapy, and behavioral techniques). Proposed behavioral mechanisms of change and clinical mental and physical health outcomes of interest (measured by Patient-Reported Outcomes Measurement Information System [PROMIS]) were also collected via electronic survey at baseline and 1-month follow-up, and participants reported any changes in mental health management at 1-month follow-up. Baseline PROMIS measures were obtained from the electronic medical record because they were collected as part of usual orthopedic care. Follow-up PROMIS measures were collected 1 month later via email invitation to an electronic survey. Further, 1 month was selected as the follow-up time point of interest because patients who present for orthopedic care of chronic musculoskeletal pain generally expect to start noticing clinical improvements related to conservative management and recommended behavior changes within 1 month.

    Exposure—Intervention Engagement

    Quantitative usage of the digital mental health intervention (Wysa for Chronic Pain) was determined by time-stamped data for each of participants’ interactions with the intervention’s conversational agent (ie, chatbot), other digital tools, or human coach. The exposure of interest was defined as the participants’ total number of interactions with the intervention during the 1-month study period, including all digital and human coach interactions. These data were provided to this study’s team via random user identifiers which ensured that only this study’s team could link participants’ usage data with their identities and other study data. Personally identifiable information was not shared between Wysa Inc and this study’s team.

    Proposed Mediators—Behavioral Targets

    The proposed behavioral mechanisms of action were operationalized by 3 self-reported measures. Behavioral activation was quantified with the Behavioral Activation for Depression Scale–Short Form (BADS-SF), which is scored from 0 to 54, with higher scores being favorable and representing more activation [36]. Pain acceptance was quantified with the 8-item Chronic Pain Acceptance Questionnaire, which is scored from 0 to 48, with higher scores being favorable and representing more pain acceptance [37]. Sleep quality was quantified with the Athens Insomnia Scale, which inquires about participants’ ability to fall asleep, stay asleep, total sleep duration, and restorative nature of sleep. It is scored from 0 to 24, with lower scores being favorable [38,39].

    Outcomes—Mental and Physical Health

    The clinical outcomes of interest were changes in self-reported mental and physical health, as measured by the adult PROMIS computer adaptive test Anxiety (version 1.0), Depression (version 1.0), Pain Interference (version 1.1), and Physical Function (version 2.0) measures [40-43]. Scoring for each PROMIS measure is normalized to the general US population with a mean of 50 and SD of 10. Higher scores represent more of the domain being measured. Therefore, lower scores are favorable for PROMIS Anxiety, Depression, and Pain Interference, and higher scores are favorable for PROMIS Physical Function [44]. For reference, clinically meaningful PROMIS score changes in patients with conservatively managed chronic musculoskeletal pain have previously been defined as 3.0 points for Anxiety, 3.2 points for Depression, 2.0 points for Pain Interference, and 2.2 points for Physical Function [45-47].

    Statistical Analysis

    The primary study outcome was whether preliminary evidence exists for the hypothesized mediation relationships of the proposed mediators on the exposure-outcome relationship between intervention engagement and self-reported (mental and physical) health. Using the approach from Baron and Kenny [48], the following criteria were used to evaluate our mediation analyses: (1) the exposure is significantly associated with the mediator variable, (2) the exposure is significantly associated with the outcome (total effect), (3) the mediator is significantly associated with the outcome when the exposure is also included in the model, and (4) the effect of the exposure on the outcome is attenuated when the mediator is also included in the model. The exposure (ie, intervention engagement) was natural log transformed to reduce skewness in the data, with +1 added to account for zeroes. The analyses were adjusted for baseline values for the proposed mediators and the outcomes, as well as for age and sex since these are potential confounders of the relationship between the exposures and outcomes. In addition to the regression coefficients, we also report the percentage of the total effect accounted for by the exposure and mediator to facilitate interpretation. Since this is a pilot trial, we considered P<.20 as a “significant” effect and reported 80% CIs to align with this threshold [49,50]. The sample size for this pilot study was determined by the availability of resources. There were no missing data. Statistical analyses were performed using SAS (version 9.4; SAS Institute Inc).


    In total, 30 of 85 (35%) eligible patients were enrolled, and all 30/30 (100%) completed follow-up measures (Figure 1). The cohort had a mean age of 59 (SD 14) years, and 21/30 (70%) were female (Table 1). During the 1-month study period, 28/30 (93%) engaged with the digital mental health intervention at least once, and 9/30 (30%) engaged at least once with the human coach of the digital intervention (Table 2). The majority of participants (24/30, 80%) did not make any other mental health management changes during the follow-up period. On average for the entire cohort, the directionality of change for all proposed mediators and clinical outcomes was consistent with improvement (Table 3). Among the proposed mediators, the greatest change was in behavioral activation (BADS-SF mean change 4.5 points, 80% CI 2.9 to 6.1). Among the clinical outcomes, clinically meaningful changes were observed for PROMIS Anxiety (–3.2, –4.3 to –2.0 points) and PROMIS Pain Interference (–3.5, –4.8 to –2.2 points).

    The mediation effect of behavioral activation on the relationship between increased intervention engagement and improved anxiety symptoms met our statistical threshold for significance (indirect effect –0.4, 80% CI –0.7 to –0.1; P=.13; Figure 2). Behavioral activation mediated 45% of this relationship. Although not statistically significant, the mediation analyses also revealed that the direction of mediation effect was generally consistent with our hypothesis for all other mediator or outcome relationships (Table 4). Specifically, improvements in clinical outcomes (ie, PROMIS Anxiety, Depression, Pain Interference, and Physical Function) were associated with increased engagement with the digital mental health intervention (Wysa for Chronic Pain) and were partially mediated by improvement in our proposed mediators (ie, behavioral activation [BADS-SF], pain acceptance [Chronic Pain Acceptance Questionnaire], and sleep quality [Athens Insomnia Scale]), with the exception that pain acceptance did not demonstrate any mediation effect on the relationship between intervention engagement and pain interference or physical function.

    ‎
    Figure 1. Inclusion flowsheet of patients who sought orthopedic care for chronic neck or back pain between January 2022 and May 2022.
    Table 1. Baseline characteristics of 30 patients who presented to an orthopedic clinic for management of chronic neck or back pain between January 2022 and May 2022.
    Characteristicsn (%)
    Age (years), mean (SD; range)59 (14; 30 to 78)
    Sex

    		Female21 (70)

    		Male9 (30)
    Race

    		White18 (60)

    		Black10 (33)

    		Asian2 (7)
    Ethnicity

    		Hispanic1 (3)

    		Not Hispanic29 (97)
    Smartphone use

    		Independent downloading and using apps17 (57)

    		Need help with downloading and using apps6 (20)

    		Smartphone user but not for apps5 (17)

    		Never use a smartphone2 (7)
    Pain duration (years), mean (SD; range)8.6 (10.7; 0.5 to 40.0)
    Orthopedic diagnosis

    		Spinal stenosis11 (37)

    		Radiculopathy11 (37)

    		Facet arthropathy3 (10)

    		Spondylosis9 (30)

    		Scoliosis2 (7)

    		Myofascial pain3 (10)
    Current treatments for mental health

    		Medication9 (30)

    		Psychiatrist, psychologist, or counselor5 (17)

    		Support group1 (3)

    		Mental health app0 (0)

    		Self-management (meditation and mindfulness)14 (47)

    		Family support2 (7)

    		None10 (33)

    		Prefer not to answer1 (3)
    Medications taken for pain

    		Opioid4 (18)

    		Tramadol3 (14)

    		Neuropathic (eg, gabapentin)10 (46)

    		Muscle relaxant (eg, cyclobenzaprine)9 (41)

    		Oral corticosteroid taper4 (18)

    		Nonsteroidal anti-inflammatory13 (59)

    		Other (eg, topical analgesic)2 (9)
    Table 2. Engagement with mental health resources during the 1-month study period among 30 patients who presented to an orthopedic clinic for management of chronic neck or back pain between January 2022 and May 2022.
    Characteristicsn (%)
    Engagement with digital mental health intervention

    		Completed onboarding28 (93)

    		Engaged with human coach at least once9 (30)

    		Total interactionsa,b, median (IQR; range)19 (7 to 64; 1 to 382)

    		Total interactions with chatbota, median (IQR; range)14 (5 to 40; 1 to 205)

    		Total interactions with coacha, median (IQR; range)0 (0 to 1; 0 to 7)
    Other change in mental health management

    		Added a medication1 (3)

    		Started psychotherapy6 (20)

    		No change24 (80)

    an=28 for intervention interactions because 2 participants did not complete onboarding.

    bTotal interactions (ie, digital tools plus human coach sessions) with the digital mental health intervention is the exposure of interest for the mediation analyses.

    Table 3. Longitudinal changes in proposed mediators and clinical outcomes during the 1-month study period among 30 patients who presented to an orthopedic clinic for management of chronic neck or back pain between January 2022 and May 2022.
    VariableBaseline, mean (SD; range)One month, mean (SD; range)Mean change (80% CI)aP value
    Proposed mediators—Behavioral targets

    		BADS-SFb total score26.7 (8.3; 12.0 to 52.0)31.2 (8.9; 16.0 to 53.0)4.5 (2.9 to 6.1)<.001

    		CPAQ-8c total score21.2 (7.0; 5.0 to 33.0)23.7 (6.6; 12.0 to 36.0)2.5 (1.2 to 3.7).014

    		AISd total score11.4 (6.3; 3.0 to 24.0)9.5 (5.1; 3.0 to 18.0)–2.0 (–3.1 to –0.8).03
    Outcomes—PROMISe Scores

    		Anxiety59.1 (8.9; 38.3 to 73.3)56.0 (7.4; 35.9 to 67.2)–3.2 (–4.3 to –2.0)<.001

    		Depression52.9 (8.7; 34.2 to 69.5)51.6 (8.0; 34.2 to 64.0)–1.3 (–2.9 to 0.3).31

    		Pain Interference64.2 (7.2; 50.1 to 80.0)60.7 (6.9; 50.1 to 83.8)–3.5 (–4.8 to –2.2).002

    		Physical Function38.6 (7.2; 26.9 to 58.9)39.9 (5.6; 30.0 to 51.2)1.4 (–0.2 to 2.9).26

    aMean changes were calculated using paired t tests.

    bBADS-SF: Behavioral Activation for Depression Scale–Short Form.

    cCPAQ-8: Chronic Pain Acceptance Questionnaire.

    dAIS: Athens Insomnia Scale.

    ePROMIS: Patient-Reported Outcomes Measurement Information System.

    ‎
    Figure 2. Mediation effect of behavioral activation on the relationship between intervention engagement and anxiety symptoms over a 1-month time interval among 30 patients who presented to an orthopedic clinic for management of chronic neck or back pain between January 2022 and May 2022. “a” represents the slope estimate between the exposure (intervention engagement) and proposed mediator (behavioral activation). “b” represents the slope estimate of the proposed mediator (behavioral activation) in the full model that includes the exposure (intervention engagement) and all covariates. Parentheses depict 80% CIs. Covariates include age, sex, and baseline BADS-SF and PROMIS Anxiety scores. c’: total direct effect; c: total effect; BADS-SF: Behavioral Activation for Depression Scale–Short Form; PROMIS: Patient-Reported Outcomes Measurement Information System.
    Table 4. Total and constituent effects of digital mental health intervention engagement (exposure) on PROMIS score changes (outcomes) over a 1-month time interval among 30 patients who presented to an orthopedic clinic for management of chronic neck or back pain between January 2022 and May 2022.a,b
    OutcomeTotal effectDirect effectscIndirect effectsd
    PROMISe domainEstimate (80% CI)P valueEstimate (80% CI)P valueEstimate (80% CI)P value
    Proposed mediator: behavioral activation (BADS-SFf)

    		Anxiety–0.9 (–1.4 to –0.4).03g–0.5 (–1.0 to 0.0).16g–0.4 (–0.7 to –0.1).13g

    		Depression–0.7 (–1.5 to 0.2).32–0.1 (–0.9 to 0.7).90–0.6 (–1.0 to –0.1).12g

    		Pain interference0.1 (–0.6 to 0.9).830.4 (–0.3 to 1.1).47–0.3 (–0.6 to 0.1).35

    		Physical function0.5 (–0.3 to 1.2).410.3 (–0.5 to 1.0).630.2 (–0.1 to 0.5).32
    Proposed mediator: pain acceptance (CPAQ-8h)

    		Anxiety–0.8 (–1.4 to –0.2).08g–0.7 (–1.3 to –0.1).11g–0.1 (–0.2 to 0.1).49

    		Depression–0.5 (–1.4 to 0.4).50–0.4 (–1.3 to 0.5).59–0.1 (–0.3 to 0.1).53

    		Pain interference0.0 (–0.7 to 0.7).990.0 (–0.7 to 0.7).970.0 (–0.1 to 0.2).94

    		Physical function0.2 (–0.4 to 0.8).710.2 (–0.4 to 0.8).710.0 (0.0 to 0.0).95
    Proposed mediator: sleep quality (AISi)

    		Anxiety–0.8 (–1.3 to –0.3).06g–0.6 (–1.1 to –0.1).11g–0.1 (–0.4 to 0.1).36

    		Depression–0.5 (–1.3 to 0.4).49–0.2 (–1.1 to 0.6).73–0.2 (–0.5 to 0.1).32

    		Pain interference–0.2 (–0.9 to 0.5).750.0 (–0.6 to 0.6).99–0.2 (–0.6 to 0.2).59

    		Physical function0.4 (–0.3 to 1.1).410.2 (–0.4 to 0.9).640.2 (–0.1 to 0.5).38

    aFor a mediation effect to be considered statistically significant, both the total and indirect effects for a given mediator or outcome pair must meet statistical significance.

    bModels are adjusted for age, sex, and baseline scores for the relevant proposed mediator (Behavioral Activation for Depression Scale–Short Form [BADS-SF], Chronic Pain Acceptance Questionnaire [CPAQ-8], or Athens Insomnia Scale [AIS]) and Patient-Reported Outcomes Measurement Information System [PROMIS] domain (Anxiety, Depression, Pain Interference, or Physical Function).

    cDirect effects represent the effect attributed to the exposure (ie, intervention engagement).

    dIndirect effects represent the effect attributed to the proposed mediator (ie, BADS-SF, CPAQ-8, or AIS).

    ePROMIS: Patient-Reported Outcomes Measurement Information System.

    fBADS-SF: Behavioral Activation for Depression Scale–Short Form.

    gthese P values indicate P<.20.

    hCPAQ-8: Chronic Pain Acceptance Questionnaire.

    iAIS: Athens Insomnia Scale.


    Principal Results

    The purpose of this pilot trial was to identify whether there is preliminary evidence that the mechanisms of behavioral activation, pain acceptance, and sleep quality mediate the relationship between increased engagement with a digital mental health intervention (Wysa for Chronic Pain) and improvements in mental and physical health among patients who present for orthopedic management of chronic musculoskeletal pain. We found preliminary evidence that behavioral activation partially mediates the relationship between increased intervention engagement and improved anxiety symptoms (indirect effect –0.4, 80% CI –0.7 to –0.1; P=.13, 45% of the total effect). Our hypothesis was also somewhat supported in that the directionality of most of the mediation analyses suggested that improvements in self-reported mental and physical health outcomes were associated with increased intervention engagement and were partially mediated by improvement in our proposed mediators of behavioral activation, pain acceptance, and sleep quality. Nevertheless, the majority of our analyses did not meet our a priori specified statistical significance of P<.20.

    Interpretations

    Given the pilot nature and small sample size of this study cohort, our findings suggest that a larger trial is needed to confirm these preliminary results. Worth particular mention, however, is that mediation analyses for the outcome of depression were limited due to the modest relationship (ie, small total effect sizes) observed between intervention engagement and PROMIS Depression score changes. We suspect this is, in part, related to participants’ relatively minimal reported depression symptoms at baseline, which likely also contributed to minimal improvement in depression symptom severity at 1-month follow-up. The modest changes in depression symptom severity may also be related to this study’s short follow-up period of 1 month. Our previous pilot work related to this digital mental health intervention suggested that depression symptoms continued to improve between 1- and 2-month follow-ups [25].

    Comparison With Prior Work

    This study complements prior studies performed by Vranceanu’s team [51,52] which showed that pain catastrophizing, kinesiophobia, mindfulness, and pain resilience are mediators of mental and physical health in people with chronic pain. The behavioral target of pain acceptance that we investigated is related to these other mechanisms because pain acceptance addresses: (1) pain catastrophizing, which is an exaggerated cognitive or emotional response to actual or anticipated pain [51-53], and (2) kinesiophobia, which is the avoidance of activities or movements due to excessive anxiety that it will cause pain or injury [54]. Our study is different from Vranceanu’s because, in contrast to her study: (1) all participants in our study were actively seeking treatment for chronic musculoskeletal pain, and (2) our intervention was less intensive, yet intervention use was still preliminarily associated with changes in our hypothesized behavioral mechanisms and clinical outcomes.

    Limitations

    This study was limited by a small sample size, lack of comparison intervention arm, short follow-up duration of 1 month, and minimal reported depression symptoms at baseline. The small sample size contributed to large CIs, and this study evaluated multiple comparisons and was not powered to detect prespecified mediation effects. Therefore, although P values were reported to facilitate rapid interpretation of the data, they should not be interpreted as definitive evidence of statistical significance. The lack of comparison arm, short follow-up duration, and minimal reported depression symptoms at baseline may also have affected the magnitude of clinical improvements observed during this study’s period. Nevertheless, clinical improvements were made in some domains at this time point, and this pilot study generated the necessary preliminary evidence regarding mediation effects to appropriately plan for a fully powered randomized controlled trial. Another limitation that should be addressed in a rigorous clinical trial is that, in this study, mediation and outcome measures were only collected at 2 time points, so it was not possible to assess whether improvements in proposed behavioral mediators preceded improvements in clinical outcomes.

    Conclusions

    We found preliminary evidence that behavioral activation partially mediates the relationship between increased engagement with a digital mental health intervention (Wysa for Chronic Pain) and improved self-reported anxiety symptoms. There is also less strong but directionally consistent evidence suggestive of mediation by pain acceptance and sleep quality on the relationship between increased intervention engagement and improvements in mental and physical health, with the exception of lack of evidence for a mediating effect of pain acceptance on pain interference or physical function. In a full-sized randomized controlled trial of patients with chronic musculoskeletal pain, it will be worthwhile to consider behavioral activation, pain acceptance, and sleep quality as hypothesized mediators of the relationship between engagement with a digital mental health intervention (Wysa for Chronic Pain) and improved self-reported mental and physical health.

    Acknowledgments

    This study was supported by the National Institutes of Health (grants K23AR074520, P50MH122351, UL1TR002345, and P30CA091842) and the Doris Duke Charitable Foundation. Wysa Inc provided complimentary access to Wysa for Chronic Pain for all study participants and provided participants’ usage data to this study’s team at no charge. None of this study’s funders were involved in the design, conduct, analysis, or reporting of this study’s findings. Generative artificial intelligence (AI) was not used in any portion of the writing of this paper.

    Data Availability

    The data set used and analyzed during this study are available from the corresponding author on reasonable request.

    Authors' Contributions

    ALC conceived of this study’s design, oversaw all aspects of this study’s conduct, and contributed to writing this study paper. MAA was responsible for data acquisition. MA and CWG performed the statistical analysis and contributed to writing this study paper and generating tables and figures. JA and RPC contributed to this study’s design and interpretation of results. All authors contributed to the critical revision of the paper and approved the final paper.

    Conflicts of Interest

    None declared.

    1. Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: a literature review. Arch Intern Med. 2003;163(20):2433-2445. [FREE Full text] [CrossRef] [Medline]
    2. Fernandez M, Colodro-Conde L, Hartvigsen J, Ferreira ML, Refshauge KM, Pinheiro MB, et al. Chronic low back pain and the risk of depression or anxiety symptoms: insights from a longitudinal twin study. Spine J. 2017;17(7):905-912. [CrossRef] [Medline]
    3. Tunks ER, Crook J, Weir R. Epidemiology of chronic pain with psychological comorbidity: prevalence, risk, course, and prognosis. Can J Psychiatry. 2008;53(4):224-234. [CrossRef] [Medline]
    4. Demyttenaere K, Bruffaerts R, Lee S, Posada-Villa J, Kovess V, Angermeyer MC, et al. Mental disorders among persons with chronic back or neck pain: results from the World Mental Health surveys. Pain. 2007;129(3):332-342. [CrossRef] [Medline]
    5. Sareen J, Cox BJ, Clara I, Asmundson GJG. The relationship between anxiety disorders and physical disorders in the U.S. National Comorbidity survey. Depress Anxiety. 2005;21(4):193-202. [CrossRef] [Medline]
    6. Archer KR, Devin CJ, Vanston SW, Koyama T, Phillips SE, Mathis SL, et al. Cognitive-behavioral-based physical therapy for patients with chronic pain undergoing lumbar spine surgery: a randomized controlled trial. J Pain. 2016;17(1):76-89. [FREE Full text] [CrossRef] [Medline]
    7. la Cour P, Petersen M. Effects of mindfulness meditation on chronic pain: a randomized controlled trial. Pain Med. 2015;16(4):641-652. [FREE Full text] [CrossRef] [Medline]
    8. Kamper SJ, Apeldoorn AT, Chiarotto A, Smeets RJEM, Ostelo RWJG, Guzman J, et al. Multidisciplinary biopsychosocial rehabilitation for chronic low back pain: cochrane systematic review and meta-analysis. BMJ. 2015;350:h444. [FREE Full text] [CrossRef] [Medline]
    9. Baumeister H, Paganini S, Sander LB, Lin J, Schlicker S, Terhorst Y, et al. Effectiveness of a guided internet- and mobile-based intervention for patients with chronic back pain and depression (WARD-BP): a multicenter, pragmatic randomized controlled trial. Psychother Psychosom. 2021;90(4):255-268. [CrossRef] [Medline]
    10. Imai H, Yamada M, Inagaki M, Watanabe N, Chino B, Mantani A, et al. Behavioral activation contributed to the total reduction of depression symptoms in the smartphone-based cognitive behavioral therapy: a secondary analysis of a randomized, controlled trial. Innov Clin Neurosci. 2020;17(7-9):21-25. [FREE Full text] [Medline]
    11. Cuijpers P, van Straten A, Warmerdam L. Behavioral activation treatments of depression: a meta-analysis. Clin Psychol Rev. 2007;27(3):318-326. [CrossRef] [Medline]
    12. Jensen MP, Ehde DM, Day MA. The behavioral activation and inhibition systems: implications for understanding and treating chronic pain. J Pain. 2016;17(5):529.e1-529.e18. [FREE Full text] [CrossRef] [Medline]
    13. Wenzel A. Basic strategies of cognitive behavioral therapy. Psychiatr Clin North Am. 2017;40(4):597-609. [CrossRef] [Medline]
    14. Veehof MM, Trompetter HR, Bohlmeijer ET, Schreurs KMG. Acceptance- and mindfulness-based interventions for the treatment of chronic pain: a meta-analytic review. Cogn Behav Ther. 2016;45(1):5-31. [FREE Full text] [CrossRef] [Medline]
    15. Hoffman BM, Papas RK, Chatkoff DK, Kerns RD. Meta-analysis of psychological interventions for chronic low back pain. Health Psychol. 2007;26(1):1-9. [CrossRef] [Medline]
    16. Greenberg J, Popok PJ, Lin A, Kulich RJ, James P, Macklin EA, et al. A mind-body physical activity program for chronic pain with or without a digital monitoring device: proof-of-concept feasibility randomized controlled trial. JMIR Form Res. 2020;4(6):e18703. [FREE Full text] [CrossRef] [Medline]
    17. Trindade IA, Guiomar R, Carvalho SA, Duarte J, Lapa T, Menezes P, et al. Efficacy of online-based acceptance and commitment therapy for chronic pain: a systematic review and meta-analysis. J Pain. 2021;22(11):1328-1342. [FREE Full text] [CrossRef] [Medline]
    18. Rayner L, Hotopf M, Petkova H, Matcham F, Simpson A, McCracken LM. Depression in patients with chronic pain attending a specialised pain treatment centre: prevalence and impact on health care costs. Pain. 2016;157(7):1472-1479. [FREE Full text] [CrossRef] [Medline]
    19. McCracken LM. Learning to live with the pain: acceptance of pain predicts adjustment in persons with chronic pain. Pain. 1998;74(1):21-27. [CrossRef] [Medline]
    20. Cohen SP, Vase L, Hooten WM. Chronic pain: an update on burden, best practices, and new advances. Lancet. 2021;397(10289):2082-2097. [CrossRef] [Medline]
    21. Finan PH, Goodin BR, Smith MT. The association of sleep and pain: an update and a path forward. J Pain. 2013;14(12):1539-1552. [FREE Full text] [CrossRef] [Medline]
    22. Edwards RR, Grace E, Peterson S, Klick B, Haythornthwaite JA, Smith MT. Sleep continuity and architecture: associations with pain-inhibitory processes in patients with temporomandibular joint disorder. Eur J Pain. 2009;13(10):1043-1047. [FREE Full text] [CrossRef] [Medline]
    23. Sutton BC, Opp MR. Sleep fragmentation exacerbates mechanical hypersensitivity and alters subsequent sleep-wake behavior in a mouse model of musculoskeletal sensitization. Sleep. 2014;37(3):515-524. [FREE Full text] [CrossRef] [Medline]
    24. Moldofsky H, Scarisbrick P. Induction of neurasthenic musculoskeletal pain syndrome by selective sleep stage deprivation. Psychosom Med. 1976;38(1):35-44. [CrossRef] [Medline]
    25. Leo AJ, Schuelke MJ, Hunt DM, Metzler JP, Miller JP, Areán PA, et al. A digital mental health intervention in an orthopedic setting for patients with symptoms of depression and/or anxiety: feasibility prospective cohort study. JMIR Form Res. 2022;6(2):e34889. [FREE Full text] [CrossRef] [Medline]
    26. Leo AJ, Schuelke MJ, Hunt DM, Miller JP, Areán PA, Cheng AL. Digital mental health intervention plus usual care compared with usual care only and usual care plus in-person psychological counseling for orthopedic patients with symptoms of depression or anxiety: cohort study. JMIR Form Res. 2022;6(5):e36203. [FREE Full text] [CrossRef] [Medline]
    27. Cheng AL, Leo AJ, Calfee RP, Dy CJ, Armbrecht MA, Abraham J. What are orthopaedic patients' and clinical team members' perspectives regarding whether and how to address mental health in the orthopaedic care setting? A qualitative investigation of patients with neck or back pain. Clin Orthop Relat Res. 2023;481(7):1415-1429. [CrossRef] [Medline]
    28. Cheng AL, Leo AJ, Calfee RP, Dy CJ, Armbrecht MA, Abraham J. Multi-stakeholder perspectives regarding preferred modalities for mental health intervention delivered in the orthopedic clinic: a qualitative analysis. BMC Psychiatry. 2023;23(1):347. [FREE Full text] [CrossRef] [Medline]
    29. Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O'Neal L, et al. The REDCap consortium: building an international community of software platform partners. J Biomed Inform. 2019;95:103208. [FREE Full text] [CrossRef] [Medline]
    30. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research Electronic Data Capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-381. [FREE Full text] [CrossRef] [Medline]
    31. Wysa Ltd. 2021. URL: https://www.wysa.com/ [accessed 2021-03-10]
    32. Baldry S. Wysa receives FDA breakthrough device designation for AI-led mental health conversational agent. Business Wire. 2022. URL: https:/​/www.​businesswire.com/​news/​home/​20220512005084/​en/​Wysa-Receives-FDA-Breakthrough-Device-Designation-for-AI-led-Mental-Health-Conversational-Agent [accessed 2022-06-03]
    33. Beatty C, Malik T, Meheli S, Sinha C. Evaluating the therapeutic alliance with a free-text CBT conversational agent (Wysa): a mixed-methods study. Front Digit Health. 2022;4:847991. [FREE Full text] [CrossRef] [Medline]
    34. Malik T, Ambrose AJ, Sinha C. Evaluating user feedback for an artificial intelligence-enabled, cognitive behavioral therapy-based mental health app (Wysa): qualitative thematic analysis. JMIR Hum Factors. 2022;9(2):e35668. [FREE Full text] [CrossRef] [Medline]
    35. Meheli S, Sinha C, Kadaba M. Understanding people with chronic pain who use a cognitive behavioral therapy-based artificial intelligence mental health app (Wysa): mixed methods retrospective observational study. JMIR Hum Factors. 2022;9(2):e35671. [FREE Full text] [CrossRef] [Medline]
    36. Fuhr K, Hautzinger M, Krisch K, Berking M, Ebert DD. Validation of the Behavioral Activation for Depression Scale (BADS)-psychometric properties of the long and short form. Compr Psychiatry. 2016;66:209-218. [CrossRef] [Medline]
    37. Fish RA, McGuire B, Hogan M, Morrison TG, Stewart I. Validation of the Chronic Pain Acceptance Questionnaire (CPAQ) in an internet sample and development and preliminary validation of the CPAQ-8. Pain. 2010;149(3):435-443. [CrossRef] [Medline]
    38. Soldatos CR, Dikeos DG, Paparrigopoulos TJ. The diagnostic validity of the Athens insomnia scale. J Psychosom Res. 2003;55(3):263-267. [CrossRef] [Medline]
    39. Soldatos CR, Dikeos DG, Paparrigopoulos TJ. Athens insomnia scale: validation of an instrument based on ICD-10 criteria. J Psychosom Res. 2000;48(6):555-560. [CrossRef] [Medline]
    40. Depression: a brief guide to the PROMIS© depression instruments. PROMIS. 2019. URL: http://www.healthmeasures.net/images/PROMIS/manuals/PROMIS_Depression_Scoring_Manual.pdf [accessed 2021-06-02]
    41. Anxiety: a brief guide to the PROMIS® anxiety instruments. PROMIS. 2019. URL: http://www.healthmeasures.net/images/PROMIS/manuals/PROMIS_Anxiety_Scoring_Manual.pdf [accessed 2023-11-09]
    42. Pain interference: a brief guide to the PROMIS© pain interference instruments. PROMIS. 2020. URL: http://www.healthmeasures.net/images/PROMIS/manuals/PROMIS_Pain_Interference_Scoring_Manual.pdf [accessed 2023-11-09]
    43. Physical function: a brief guide to the PROMIS® physical function instruments. PROMIS. 2020. URL: http://www.healthmeasures.net/images/PROMIS/manuals/PROMIS_Physical_Function_Scoring_Manual.pdf [accessed 2023-11-09]
    44. Cella D, Yount S, Rothrock N, Gershon R, Cook K, Reeve B, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS): progress of an NIH roadmap cooperative group during its first two years. Med Care. 2007;45(5 Suppl 1):S3-S11. [FREE Full text] [CrossRef] [Medline]
    45. Lee AC, Driban JB, Price LL, Harvey WF, Rodday AM, Wang C. Responsiveness and minimally important differences for 4 patient-reported outcomes measurement information system short forms: physical function, pain interference, depression, and anxiety in knee osteoarthritis. J Pain. 2017;18(9):1096-1110. [FREE Full text] [CrossRef] [Medline]
    46. Kroenke K, Yu Z, Wu J, Kean J, Monahan PO. Operating characteristics of PROMIS four-item depression and anxiety scales in primary care patients with chronic pain. Pain Med. 2014;15(11):1892-1901. [FREE Full text] [CrossRef] [Medline]
    47. Chen CX, Kroenke K, Stump TE, Kean J, Carpenter JS, Krebs EE, et al. Estimating minimally important differences for the PROMIS pain interference scales: results from 3 randomized clinical trials. Pain. 2018;159(4):775-782. [FREE Full text] [CrossRef] [Medline]
    48. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173-1182. [CrossRef] [Medline]
    49. Lee EC, Whitehead AL, Jacques RM, Julious SA. The statistical interpretation of pilot trials: should significance thresholds be reconsidered? BMC Med Res Methodol. 2014;14:41. [FREE Full text] [CrossRef] [Medline]
    50. Kwak S. Are only p-values less than 0.05 significant? A p-value greater than 0.05 is also significant!. J Lipid Atheroscler. 2023;12(2):89-95. [FREE Full text] [CrossRef] [Medline]
    51. Greenberg J, Mace RA, Bannon SM, Kulich RJ, Vranceanu AM. Mind-body activity program for chronic pain: exploring mechanisms of improvement in patient-reported, performance-based and ambulatory physical function. J Pain Res. 2021;14:359-368. [FREE Full text] [CrossRef] [Medline]
    52. Grunberg VA, Mace RA, Bannon SM, Greenberg J, Bakhshaie J, Vranceanu AM. Mechanisms of change in depression and anxiety within a mind-body activity intervention for chronic pain. J Affect Disord. 2021;292:534-541. [FREE Full text] [CrossRef] [Medline]
    53. Quartana PJ, Campbell CM, Edwards RR. Pain catastrophizing: a critical review. Expert Rev Neurother. 2009;9(5):745-758. [FREE Full text] [CrossRef] [Medline]
    54. Wertli MM, Rasmussen-Barr E, Weiser S, Bachmann LM, Brunner F. The role of fear avoidance beliefs as a prognostic factor for outcome in patients with nonspecific low back pain: a systematic review. Spine J. 2014;14(5):816-836.e4. [CrossRef] [Medline]

    ‎
    BADS-SF: Behavioral Activation for Depression Scale–Short Form
    PROMIS: Patient-Reported Outcomes Measurement Information System
    REDCap: Research Electronic Data Capture

    Edited by A Mavragani; submitted 31.07.23; peer-reviewed by M Pritchard, P Watson; comments to author 01.11.23; revised version received 03.11.23; accepted 05.11.23; published 17.11.23.

    Copyright

    ©Abby L Cheng, Mansi Agarwal, Melissa A Armbrecht, Joanna Abraham, Ryan P Calfee, Charles W Goss. Originally published in JMIR Formative Research (https://formative.jmir.org), 17.11.2023.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.