A Novel Couple-Based Intervention for Chronic Pain and Relationship Distress: A Pilot Study

Abstract

Chronic pain contributes to psychological and relationship distress in individuals with pain as well as their partners. Prior pain interventions have addressed this important social context by engaging partners in treatment; however, partners have not been considered co-participants who can benefit directly from therapy, but rather incorporated as pain management coaches or guides. This manuscript assesses the feasibility, acceptability, and preliminary outcomes of a novel intervention which targets both partners and focuses on improving well-being in couples in which one or both partners experiences chronic pain and relationship distress. Fifteen couples participated in Mindful Living and Relating, a 6-session in-person intervention, and completed baseline and post-treatment outcome measures. Both quantitative and qualitative methods were used to evaluate participants’ engagement in and experiences of the intervention, as well as preliminary outcomes. Results suggest that couples were engaged in, and reported satisfaction with, the treatment. Participants who completed the therapy (N = 28; 14 couples) reported reductions in depressive symptoms and improvements in relationship satisfaction and partner responsiveness, and individuals with pain reported reductions in pain interference. In post-treatment interviews, couples reported their preference for couple therapy over individual therapy for pain and relationship distress. Although the conduct of the therapy was feasible for couples who enrolled in the trial, initial recruitment difficulties suggested feasibility challenges. Recommendations are made for researchers who are interested in designing psychological interventions to improve quality of life in the context of chronic illness.

According to the biopsychosocial model, chronic pain is a result of the interplay between physiological, psychological, and social factors (Gatchel, Peng, Peters, Fuchs, & Turk, 2007). The social impact of chronic pain is so significant that individuals with chronic pain who experience interpersonal distress are more likely to drop out of individual pain interventions, and those who complete such interventions have poorer outcomes than those without interpersonal distress (Carmody, 2001). In response, the social context of pain has received increased attention as a point of intervention. Some psychological pain treatments include family members, particularly spouses or romantic partners, as coaches to support pain management goals (Keefe et al., 1996; Lewandowski, Morris, Draucker, & Risko, 2007); however, trials have not included romantic partners as targets of treatment. Additionally, there is little research on the feasibility of pain interventions involving partners and couples. Therefore, the purpose of this study is to examine the feasibility, acceptability, and safety data of a novel couple-based intervention for chronic pain (Cano et al., 2018). Preliminary data examining the intervention’s efficacy are also presented.

The Impact of Chronic Pain within Intimate Relationships

Interpersonal dynamics, communication, and pain-related symptoms impact people with pain and their partners. For example, feeling emotionally supported and validated by one’s partner can positively influence the physical and psychological well-being of an individual with chronic pain, just as conflict, criticism, and hostility can negatively affect health behaviors, daily functioning, and illness adjustment (Burns, Johnson, Mahoney, Devine, & Pawl, 1996; Burns et al., 2013; Cano & Williams, 2010; Martire & Schulz, 2007). Couples with pre-existing maladaptive relational patterns may also have even more difficulty managing the stressors brought on by chronic health conditions, compared to couples who effectively communicate and relate (Rolland, 2018). Additionally, partners’ individual attitudes and behaviors affect the pain intensity, self-management of symptoms, and illness-related decision-making of those with pain (Leonard & Cano, 2006; Martire, Schulz, Helgeson, Small, & Saghafi, 2010). Even well-intended partners can unintentionally enable maladaptive health behaviors (Fischer, Baucom, & Cohen, 2016).

Research also indicates that a partner’s chronic illness can be a source of stress and an unavoidable strain on the family unit due to uncertainties about the future (Cano, Corley, Clark, & Martinez, 2018; Rolland, 2018). Specifically, chronic pain contributes to decreased relationship satisfaction, and an individual’s pain severity is related to the partner’s feelings of perceived helplessness, psychological distress, and depressed mood (Geisser, Cano, & Leonard, 2005; Leonard & Cano, 2006; Leonard, Cano, & Johansen, 2006). Pain catastrophizing has also been linked to partners’ caregiver burden and relationship dissatisfaction, which can reinforce the cycle of maladaptive coping strategies and relationship distress (DeLongis, Holtzman, Puterman, & Lam, 2010). Fortunately, interventions that focus on the relationship, in addition to individual symptoms, have been shown to be more effective in improving the health of family members than standard treatment (Hartmann, Bazner, Wild, Eisler, & Herzog, 2010). Using this family systems theoretical approach to treat chronic health conditions is essential, as it expands the traditional medical model that narrowly focuses on the individual patient (Rolland, 2018). This model can help couples move towards conceptualizing chronic pain as a “we” problem, which has been shown to optimize couple functioning (Skerrett, 1998).

Existing Interventions for Couples with Chronic Pain

Given the importance of intimate relationships in chronic pain, some interventions have included partners in treatment (see Martire et al., 2003; Ramke, Sharpe, & Newton-John, 2016). For example, Keefe and colleagues (1996) developed spouse-assisted coping skills training (CST), which focuses on mutual goal setting and skill building for both partners, to reinforce pain management strategies. The individual with pain, however, remains the focus of treatment, while the partner acts as a coach or facilitator of change. As a result, the partner’s well-being and the couple’s relationship history and communication dynamics are not addressed. This patient-focused model differs from our model, which focuses on the needs of both partners. This distinction may explain why partner-assisted interventions for chronic health conditions, such as spouse-assisted CST, have demonstrated little or no incremental benefit above individual interventions in improving marital functioning and reducing pain and depressive symptoms (Keefe et al., 1996; Martire et al., 2010; Ramke et al., 2016). Spouse-assisted interventions may be best suited for couples that are functioning well, as relationship distress and communication are not treatment targets (Martire et al., 2010). In couples with interpersonal distress, actively engaging romantic partners in treatment as co-participants rather than coaches may be preferable (Tankha, Kerns, & Cano, 2018). Furthermore, the emotional components of illness management can be more effectively addressed when a partner is involved in treatment, compared to individual treatment (Martire et al., 2003). Yet, little is known about whether such an approach would be acceptable or feasible to couples who have been negatively impacted by the effects of chronic pain, though it is important to note that a family systems approach has been shown to be acceptable and to optimize family functioning when one member has been diagnosed with other chronic illnesses (Skerrett, 1998).

The Current Intervention and Study

Because of the critical role that relationships play in chronic pain conditions, focused attention on the dyad is crucial in order to improve clinical outcomes for individuals with chronic pain. Therefore, we developed a novel, integrative, couple-based intervention for people with comorbid chronic pain and relationship distress that was initially described in a case study (Cano et al., 2018). Mindful Living and Relating is composed of evidence-based components rooted in mindfulness and ACT that aim to improve quality of life by targeting psychological flexibility (i.e., adapting behaviors or shifting perspectives to remain aligned with one’s values; Kashdan & Rottenberg, 2010) and relational flexibility (i.e., acceptance, validation, and emotional disclosure in the presence of another; Cano et al., 2018). Individual mindfulness-based interventions for chronic pain reduce pain and improve overall quality of life, fatigue, and depression; however, such interventions have not yet incorporated partners (Hilton et al., 2017; Kabat-Zinn, 1982; Wicksell, Olsson, & Hayes, 2010; Zautra et al., 2008). Additionally, individual Acceptance and Commitment Therapy (ACT) improves pain acceptance, which is associated with improved psychological distress, social and physical functioning, and pain intensity (McCracken & Gutierrez-Martinez, 2011; Scott, Hann, & McCracken, 2016). Mindful Living and Relating builds on this research and suggests that mindfulness- and acceptance-based interventions for couples with pain may offer more robust results than individual therapy and lead to improvements in relationships and intimacy—in turn, these qualities can offer protection from the impact of distress, such as periods of heightened physical pain (Karremans, Schellekens, & Kappen, 2017). More generally, individual mindfulness- and acceptance-based interventions will likely not address the relationship distress associated with chronic pain, and improvements for individuals receiving these treatments may be hindered or blocked if they are in relationships that are hostile, unsupportive, or otherwise distressed (DeLongis et al., 2010).

Because Mindful Living and Relating showed promise in a case study, we offered it to a sample of couples and explored the intervention’s acceptability, feasibility, and preliminary effectiveness. Addressing these aspects before testing the efficacy via a randomized controlled trial is crucial, as interventions need to be modified in response to feasibility and acceptability data—this subsequently increases the probability that the intervention will be effective (Bowen et al., 2009). Therefore, the goals of this study were to examine the feasibility and acceptability of a 6-week in-person intervention for couples with chronic pain and relationship distress by gathering participants’ comprehensive feedback, including opinions on the content, format, and delivery of the intervention, and to evaluate preliminary outcome data. To accomplish these goals, quantitative and qualitative methods were employed. The incorporation of qualitative methods provides the opportunity to provide context to quantitative analyses in order to convey the first-hand perspectives of study participants. In this respect, multiple methods are essential at early stages of treatment development to ensure that refinements are truly aligned with the experience of patients (Zhang, 2014).

Methods

Participants

This treatment development study was approved by the Institutional Review Board of Wayne State University and registered with clinicaltrials.gov prior to recruitment (NCT02316288). We used various recruitment strategies, including advertisements on our university website, laboratory website, newspaper advertisements, and flyers and brochures in pain clinics. Potential participants telephoned and were screened by a research assistant. Inclusion criteria were that: 1) both partners were at least 21 years old and married or cohabitating for at least two years; 2) one or both partners had a chronic pain condition (at least 3 months of musculoskeletal or visceral pain, headache, various types of arthritis, or other types of non-malignant pain); 3) one or both partners reported pain interference scores of 3 or greater (on a 0 to 10 scale) in at least one out of the seven interference domains from the Brief Pain Inventory - Short Form (Cleeland, 1991); and 4) at least one partner reported relationship distress, as indicated by a score of 25 or less on the 7-item Dyadic Adjustment Scale (DAS-7 (Sharpley & Rogers, 1984). Exclusion criteria were either partner reporting: (1) suicidal or homicidal ideation, (2) current psychotic symptoms, (3) symptoms of dementia or cognitive impairment, (4) malignancies (e.g., cancer) within the prior five years, or (5) current domestic violence within the relationship (measured by the Hurt, Insult, Threaten, and Scream, a 4-item screening tool (Sherin, Sinacore, Li, Zitter, & Shakil, 1998).

The study generated the interest of 118 individuals (72.5% female) who called to obtain more information. Data from those who called (M_age = 48.77, SD = 15.27; range: 18–85) indicated that 82.28% of the callers experienced chronic pain and 34.57% reported that their partners also had pain. The callers with chronic pain had pain for a mean of 11.64 years (SD = 10.72) and reported that their partners had pain for 12.57 years (SD = 11.92). Of those who were interested, 45 couples were ineligible: 17 couples were not married or living together for one year, 11 couples did not meet the relationship satisfaction cut-off (i.e., they were relatively satisfied in their relationships), 5 couples were under age 21, and the remaining were ineligible due to medical or mental health reasons.

Twenty-three couples enrolled in this treatment feasibility and acceptability study (N = 46 individuals, M_age = 52.86 years, SD = 14.0, range: 22–79 years). Couples were randomized to the intervention or an education control condition; the control condition was not part of the current manuscript. The object of randomization was not to compare treatment outcomes at this early stage, but to ensure that representative groups of couples were assigned to each group for development purposes. A 3:1 randomization strategy was chosen to oversample couples in the intervention group since the purpose of the study was to develop, refine, and test the feasibility and acceptability of the new intervention. This resulted in an allocation of 17 couples in the experimental group and 6 couples in the education-control group. Fifteen of the 17 couples assigned to the experimental group began the intervention, and no reasons were given by the other two couples for not entering the intervention phase (See document, Supplemental File 1, CONSORT diagram).

Thirteen out of 15 couples were heterosexual and two were female same-sex couples, with seven couples reporting having attended couples therapy in the past. The sample had some racial diversity (72.4% Caucasian, 20.7% African American, 3.4% American Indian/Alaskan Native, 3.4% Mixed, and 3.3% did not respond). Almost a third of participants (36.7%, n = 11) reported a history of one or more psychiatric diagnoses, including depression (n = 8), anxiety disorders (n = 3), and bipolar disorder (n = 2). A majority of individuals (60%; n = 18) reported additional medical problems, including high blood pressure, hypothyroidism, diabetes, high cholesterol, and asthma.

In eight couples, both partners reported chronic pain, whereas only one partner reported pain in the remaining seven couples. For all the couples in which only one partner reported pain, the individual in pain was female. For all participants with pain (n = 23), the average pain duration was 17.93 years (SD = 14.47), and 70% (n = 21) reported that they had received a pain-related diagnosis (e.g., arthritis, spinal stenosis, fibromyalgia, osteoarthritis, compression fracture, myofascial pain syndrome, degenerative disc disease, disc herniation, low back pain, chronic migraine).

Intervention

The goals of Mindful Living and Relating were to improve quality of life and relationship distress by enhancing psychological and relational flexibility within the couple. Individual couples met for six weekly 1.5-hour sessions delivered by one of five master’s-level clinicians, including one master’s level social worker and four advanced graduate students pursuing their doctoral degrees in clinical psychology. All clinicians had training and experience in mindfulness-based techniques, Cognitive-Behavioral Therapy, and Acceptance and Commitment Therapy. A manual was utilized by therapists to promote treatment fidelity, the sessions were audio-recorded, and therapists received weekly supervision from the study’s principle investigator. During each therapy session, one mindfulness-based activity and one to two value-based activities were introduced, each of which guided that week’s skill practice or written homework assignments. (See Table 1 and Cano et al., 2018 for further details on weekly session topics). Briefly, mindfulness activities included individual and conjoint exercises designed to promote nonjudgmental awareness to the present moment. Values-based exercises were designed to reacquaint couples with their values as individuals and as a couple and to align their daily activities with these values.

Table 1.

Overview of session activities and homework assignments (Adapted from Cano et al., 2018).

Session	In-Session Activities	Assigned Homework
1	Mindfulness-based: Breathing meditation Values-based: Anniversary exercise	• 1a: Practice breathing meditation • 1b: Complete anniversary exercise
2	Mindfulness-based: Body scan meditation Values-based: Developing values statements from anniversary exercise and introduction to the values compass	• 2a: Practice body scan meditation • 2b: Complete values statements • 2c: Complete values compass
3	Mindfulness-based: Leaves on a stream meditation (Hayes, Strosahl, & Wilson, 1999) Values-based: Developing goals that align with values	• 3a: Practice leaves on a stream meditation • 3b: Complete goals development
4	Mindfulness-based  • Psychological Flexibility: Mindful handholding meditation  • Relational Flexibility: Mindful listening and communication skills Values-based:  • Psychological Flexibility: Select one values-based action goal to complete (from Session 3)  • Relational Flexibility: Identifying and disclosing positive emotions about one’s partner while the partner practices listening and acceptance of emotions.	• 4a: Practice mindful handholding meditation • 4b: Complete a values-based action • 4c: Practice identifying and disclosing positive emotions about partner (to partner)
5	Mindfulness-based  • Psychological Flexibility: Loving-kindness meditation  • Relational Flexibility: Mindful listening and communication skills Values-based:  • Psychological Flexibility: Select one values-based action goal to complete (from Session 3)  • Relational Flexibility: Identifying and disclosing negative emotions about pain and stress while partner mindfully listens	• 5a: Practice loving-kindness meditation • 5b: Complete a values-based action • 5c: Each partner was instructed to record his/her negative feelings about pain or stress for three days and then begin daily sharing on the fourth day.
6	Mindfulness-based: No new skills; review of Session 5’s activity Values-based: No new skills; review of Session 5’s activity Wrap-up: Develop an individualized plan for continued practice and brainstorm solutions to barriers	• Follow individualized plan

Primary goals of the first session were to establish rapport, offer a rationale for the intervention, and learn about the couple’s relationship using a semi-structured interview. An adapted version of the Oral History Interview (Buehlman, Gottman, & Katz, 1992) was used to gather relationship history and target the couple’s experience with chronic pain (e.g., “How do you think your pain/your partner’s pain impacts your relationship?”). The second half of this session introduced a mindful breathing meditation and an anniversary exercise, which was aimed at helping couples clarify their values. Sessions 2–6 began with practice of the prior week’s mindfulness activity and a discussion about the couple’s experience with their practice throughout the week. With the therapist’s help, the couple not only reflected on the benefits of practicing the skills, but also identified obstacles they encountered and brainstormed solutions. Values-based assignment(s) were also reviewed. The remaining part of each session consisted of introducing new mindfulness- and values-based activities and assigning homework.

Measures

Multiple self-report measures were administered at baseline, after each weekly session, and immediately post-invention. Outcomes focused on the intervention’s feasibility, acceptability, safety, and efficacy, and the following measures are clustered in correspondence with the study aims.

Feasibility Measures

Recruitment and retention.

Data were collected during recruitment, including the number of individuals who were screened and deemed eligible/ineligible, and how many of those eligible enrolled or declined. The number of participants who withdrew from treatment and reasons for withdrawing were recorded.

Homework feasibility.

Select items from the Homework Rating Scale (HRS; Kazantzis, Deane, & Ronan, 2004), and one additional item added for this project, were used to evaluate the feasibility of the weekly homework assignment(s). One HRS was administered for each homework assigned, which ranged from one to three assignments weekly. Five items representing five domains of feasibility, which were selected based on themes found in prior feasibility research (e.g., Kashikar-Zuck et al., 2016), included: difficulty (“How difficult was the assignment?”), barriers to completion (“How much did obstacles interfere with the assignment?”), comprehension (“How well did you understand what to do?”), specificity (“How specific were the guidelines on how to do the assignment?”), and rationale (“How well did you understand the reason for doing the assignment?”). One additional item was added to measure adherence [“In the past seven days, how many times did you practice the homework outside of the session?” (for mindfulness exercises), or “How much of the assignment were you able to do?” (for written assignments)]. On this subscale, the possible range of scores was 0 to 24, with a higher score indicating higher ratings of feasibility. For the purposes of the current study, the median score of 12 was used as a margin of feasibility.

Intervention Adherence.

After each session, therapists completed weekly checklists which recorded whether participants attended the session and if they completed the prior week’s homework assignment(s). These items were used to evaluate participants’ adherence to the intervention.

Acceptability Measures

Treatment Credibility.

The 6-item Credibility/Expectancy Questionnaire (Devilly & Borkovec, 2000) was completed by participants after treatment session 1 to measure how much they believed that the therapy would help them.

Satisfaction with Treatment.

After completing the intervention, participants completed the 21-item Satisfaction with Treatment survey, a measure developed for this project in order to assess participant satisfaction with the overall intervention. The measure also assessed participants’ agreement with statements about the intervention, including: I would recommended this therapy to others; I would have preferred individual sessions instead of sessions with my partner; The sessions helped me to communicate with my loved ones; The sessions helped me to learn more about how to cope with pain; and The sessions led to a change in my relationships or the way I interact with others.

Participant Engagement.

Therapists completed the Therapist Weekly Feedback Form post-session in which they rated the participants’ engagement in that day’s session using a scale of 0 (“not at all engaged”?) to 4 (“a lot engaged”). This measure was developed for the original study.

Intention to Complete Homework Assignments.

Using a 1-item scale developed for the original study (“Do you plan to complete the therapist’s assigned homework?”), participants rated their intention to complete homework assignments after each session on a 5-point scale.

Homework Acceptability.

Select items from the Homework Rating Scale (HRS; Kazantzis et al., 2004) were used to evaluate the acceptability of the weekly homework assignment(s). One HRS was administered for each homework assigned, which ranged from one to three assignments weekly. Using themes found in prior acceptability research (e.g., Kashikar-Zuck et al., 2016), five HRS items were deemed appropriate to evaluate the acceptability of the weekly homework assignments including: quality of work (“How well did you do the assignment?”), match with therapy goals (“How well did the assignment match your therapy goals?”), satisfaction (“How much did you enjoy the assignment?”), mastery (“How much did the assignment help you to gain control over your problems?”) and progress (“Did the assignment help with your progress in therapy?”). On this subscale, the possible range of scores was 0 to 20, with a higher score indicating higher ratings of acceptability. For the purposes of the current study, the median score of 10 was used as a margin of acceptability.

Safety Measure

Adverse Events.

The occurrence of individual adverse events was assessed at each therapy session (study visits 2–7) and at post-intervention (study visit 8).

Combined Acceptability and Feasibility Measures

Semi-Structured Interviews.

The 15 couples in the treatment group who began the intervention were asked to give feedback on their experiences via a 17-question semi-structured interview following treatment (see document, Supplemental File 2, which lists interview questions), and 14 couples agreed to participate in the interview. An interviewer who was not the couple’s therapist and who did not conduct the baseline assessments led the face-to-face interviews in the research laboratory. Interview questions targeted the feasibility and acceptability of the intervention (e.g., satisfaction with sessions and homework assignments, what they found most and least helpful, etc.) and were designed as open-ended questions to give the couples flexibility in their responses. The interviews were digitally video-recorded and transcribed verbatim.

Clinical Outcome Measures

The following measures were administered at baseline and post-treatment.

Depressive symptoms.

Depressive symptoms were measured using the Beck Depression Inventory-II (BDI-II; Beck, Steer, Ball, & Ranieri, 1996; Beck, Steer, & Brown, 1996), a 21-item scale that assesses specific depression symptoms including affective (e.g., hopelessness, irritability) and physical (e.g., fatigue) symptoms. Scores can range from 0 to 63, with higher scores indicating more severe levels of depressive symptoms. The BDI-II has been widely used in studies and has strong psychometric properties, including high internal consistency and concurrent validity with other measures of depression (Beck, Steer, & Brown, 1996). In the current sample, the BDI-II had high internal consistency at baseline (α = .92) and post-intervention (α = .89).

Pain-Related Outcomes.

The Multidimensional Pain Inventory (MPI; Kerns, Turk, & Rudy, 1985) was used to assess pain-related interference, perceived support, pain severity, affective distress, level of involvement in general activities (e.g., household chores, social activities), and individuals’ perceptions of how their partners respond to pain expressions (i.e., solicitous or instrumental support behaviors, distracting, or negative responses). The MPI has demonstrated strong psychometric properties including high internal consistency, stability, and convergent validity with other pain-outcome measures (Kerns et al., 1985). In the current sample, the MPI subscales (pain severity, interference, general activities, and partner responses to pain expressions) had high internal consistency at baseline (pain severity/interference α = .89; general activities α = .86; partner responses to pain expressions α = .72) and post-intervention (pain severity/interference α = .97; general activities α = .88; partner responses to pain expressions α = .88).

Relationship satisfaction.

Relationship satisfaction was measured using the Dyadic Adjustment Scale (DAS; Spanier, 1976). This 32-item scale is designed to measure the quality of romantic relationship adjustment, including cohesion, affectional expression, consensus, and satisfaction. Scores can range from 0 to 151, with higher scores indicating higher levels of relationship adjustment and satisfaction. For reference, the average DAS scores of married couples is 114.8 (SD = 17.8) and divorced couples is 70.7 (SD = 23.8; Spanier, 1976). The DAS has demonstrated content validity, concurrent validity with other marital adjustment scales, and high internal consistency (Spanier, 1976). In the current sample, the DAS had high internal consistency at baseline (α = .92) and post-intervention (α = .89).

Partner responsiveness.

Partner responsiveness was measured by the Perceived Partner Responsiveness Scale (PPRS; Reis, Clark, & Holmes, 2004), a 17-item scale that asks individuals to rate the extent to which their partners are responsive to their needs. Scores range from 17 to 158, with higher scores signifying higher levels of perceived partner responsiveness. The PPRS has demonstrated high internal consistency, a unidimensional construct via exploratory factor analysis, and concurrent validity with other measures of relationship satisfaction, trust, empathy, and emotional support (Reis et al., 2004). In the current sample, the PPRS had high internal consistency at baseline (α = .97) and post-intervention (α = .94).

Perceived Improvement.

Participants completed the Patient Global Impression of Change scale (PGIC; Guy, 1976) at the end of each session and at post-intervention. This one-item scale asks participants to rate their perceived change since the intervention began from 1 (“very much worse”) to 7 (“very much improved;” Dworkin et al., 2008). This measure assesses global improvement, not improvement in any particular outcome.

Statistical Analyses

Quantitative Analyses.

Descriptive statistics were used to present data on feasibility, acceptability, and safety. In addition, efficacy was tested by examining changes in depressive symptoms, relationship satisfaction, and partner responsiveness from baseline to post-intervention for both partners, and changes in pain-related symptoms were analyzed for all individuals with pain. Only data from individuals who began the therapy sessions were evaluated (15 couples). T-tests accounting for non-independence of observations across time and within couples were conducted on the psychological and relational distress measures using R 3.6.0 (Core Team, 2019). For pain measures (i.e., pain severity, interference, and partner responses to pain expressions), repeated measures t-tests were conducted without accounting for possible non-independence because the sample included couples in which one or both partners reported pain. Due to the difficulty of quantifying effect sizes in multilevel approaches and to provide consistency of reporting across psychological, relational, and pain outcomes, we calculate Cohen’s d (dividing the change by the baseline SD) on all participants to provide an estimate of the magnitude of the intervention’s effect on these variables.

Independent samples t-tests were also used to evaluate whether those with and without pain had significantly different responses to the intervention (i.e., change scores were calculated and compared between these two groups). In addition to the presented descriptive and inferential statistics (see Table 2), the work of Dworkin et al. (2008) was used to denote the clinical importance of participants’ improvement (i.e., Substantial, Clinically Important, Moderately Important, or Minimally Important improvements).

Table 2.

Treatment efficacy

Construct	Baseline M (SD)	Post-Intervention M (SD)	df	coefficient	p	d (size of effect)
Depression1,5	15.70 (11.77)	9.67 (6.95)	43	−6.033	.007	.51 (medium)
Relationship Satisfaction2,5	100.83 (16.10)	105.07 (14.30)	43	4.23	.120	.26 (small)
Partner Responsiveness3,5	101.57 (29.83)	110.57 (24.18)	43	9.00	.049	.30 (small-medium)
Pain Interference4,6	3.75 (1.30)	3.21 (1.36)	22	3.43	.002	.42 (small-medium)
Pain Severity4,6	3.36 (1.44)	3.19 (1.34)	22	.84	.413	.12 (very small)
Perceived Support4,6	4.20 (1.15)	4.52 (.89)	22	−1.34	.195	.28 (small)
Negative Partner Responses4,6	1.32 (1.05)	1.13 (1.13)	22	1.00	.326	.18 (small)
Solicitous Partner Responses4,6	3.56 (1.29)	3.88 (1.18)	22	−1.46	.158	.25 (small)
Distracting Partner Responses4,6	1.91 (1.24)	3.08 (.80)	22	−3.82	.001	.94 (large)

¹Beck Depression Inventory-II (Beck et al., 1996)

²Dyadic Adjustment Scale (Spanier, 1976)

³Perceived Partner Responsiveness Scale (Reis et al., 2004)

⁴Multidimensional Pain Inventory (Kerns et al., 1985)

⁵T-tests accounting for non-independence of observations within couples

⁶T-tests with only one partner (i.e., the identified partner with pain)

Qualitative Analyses.

Fourteen semi-structured interviews (one couple did not participate in the interview due to scheduling conflicts) were analyzed using thematic analysis, a flexible qualitative method that is used to identify patterns or themes while giving the researcher a rich description of the data (Braun & Clarke, 2006). Analysis began with a close reading of all 14 transcripts, and then a two-phase analysis was completed. Each analysis phase consisted of identifying codes (i.e., patterns of interview content) and then collapsing the codes into broader themes. Phase one utilized a theoretical analysis, which identified interview content related to the original research questions including feasibility, acceptability, and safety. Phase two utilized an inductive analysis, a broader analytic process consisting of re-reading the interviews and evaluating them for underlying concepts beyond the content identified during phase one. Direct quotations from the interview data are included to support the coded themes across both phases, and names and other identifying information have been removed to protect participant anonymity and confidentiality.

Results

Quantitative and qualitative results are presented concurrently to give a comprehensive view of the intervention’s feasibility and acceptability, as the qualitative themes closely aligned with the data from the quantitative measures used to evaluate these constructs. Twelve codes were identified during closed coding of the post-intervention interviews, as they related to the initial research question of the intervention’s feasibility and acceptability. In addition, quantitative results regarding preliminary outcomes are reported. Finally, a relationship-building theme (consisting of four codes) emerged during open coding, and it is explained independently to highlight this significant pattern of content as it pertains to feasibility and acceptability (see document, Supplemental File 3, which illustrates the themes and included codes).

Feasibility

Recruitment.

Seventy-three (61.86%) of the 118 couples who responded to the study were eligible. Fifty eligible couples (68.49%) expressed disinterest in the study. The majority of couples who were disinterested (68%) did not disclose a reason for their disinterest, but those who did reported that they were too busy and could not find time in their schedule (26%), were located too far from the lab (4%), or did not wish to be video recorded (2%). The following analyses are based on the 15 couples who completed both baseline and post-intervention surveys.

Adherence.

Descriptive data from the Therapy Weekly Checklist indicate that couples who began the therapy sessions were adherent to treatment, that is, 14 out of 15 couples (93.33%) attended at least 75% of sessions and completed at least one homework assignment per session. The majority of couples felt that six sessions was “just right,” with one participant commenting that she was “very amazed and astonished” at how much she and her husband were able to accomplish in only six weeks. On the other hand, some couples thought more sessions (e.g., 8 or 12 weeks) would be even more beneficial, as they thought there was not enough time for the “heavy work” required of the intervention. “You can’t overdo a good thing,” one participant shared.

Homework Feasibility.

Overall, 5 weeks of homework assignments yielded a feasibility score of 16.82 (SD = 1.93), indicating that participants found the homework to be feasbile (range: 15.33–18.92). One woman appreciated that the assignments gave her “something solid to hold onto and to focus on.” On occasion, participants reported barriers to completing the assignments, such as busy schedules and heightened physical pain, which negatively impacted homework feasibility; however, one participant explained how she overcame the pain barrier: “And so that’s the way I had to look at it: Is this harder than the pain? No, the pain is harder than this and at the end of the day doing the homework…is going to benefit me.”

Evaluation of individual assignments (See Table 1 for description of assignments) indicated that participants found the goals development exercise (Week 3) to be most feasible (M = 18.92, SD = 2.23), while they rated the anniversary exercise (Week 1) least feasible (M = 15.33, SD = 3.86).

Therapy Session Feasibility.

Eight participants explained that although the treatment rationale was explained during the first session, they had difficulty comprehending the aims of the intervention. Some described the content of the sessions as “ambiguous” at times and found themselves questioning what they were supposed to be doing and feeling lost: “It would be like reading a book and not having a…table of contents,” one man said. Another participant stated that she was not sure what the treatment consisted of or the rationale of certain skill trainings, while another acknowledged that she felt confused before Session 4 and did not understand how everything was going to fit together. Her partner elaborated, saying:

“It was like a puzzle when we first got it, and we didn’t understand what the puzzle was supposed to look like. We didn’t understand how to put the puzzle together. We didn’t have the front box [to look at]. It wasn’t ‘til later in there when the puzzle started coming together, where we could see how it was all kind of fitting together.”

A few couples recommended offering clearer objectives at the beginning of treatment, as well as detailed rationale for each skills training.

Acceptability

The acceptability theme that emerged from the interviews is defined as the participants’ satisfaction with the interventions’ content (and delivery of content), sessions, and assignments, as well as participants’ perceptions of the treatment’s credibility and positive effects the intervention had on their individual and relationship wellbeing.

Treatment Credibility.

Treatment credibility data were missing for five couples due to administrative error. Existing data (n = 20) indicated that, after the first therapy session, which included a treatment rationale, participants believed the intervention to be a credible form of treatment. Specifically, on a 9-point scale (1 to 9, with higher scores indicating higher levels of agreement), they perceived it to be logical (M = 7.55, SD = 2.04), believed it would be successful in reducing pain symptoms (M = 6.65, SD = 2.01), and endorsed confidence in recommending the treatment to others with similar problems (M = 7.0, SD = 2.15). Additionally, they predicted that they would see about a 45% improvement in their “pain symptoms” by the end of therapy, which appears to be defined by participants as specifically “reduction in physical pain severity.” However, because this intervention’s primary focus was not solely to reduce pain severity, but rather improve relationships and overall pain adjustment, a misalignment between patient and treatment expectations was detected. As one participant described, “I thought [the intervention] would kind of be between me and the pain, rather than be between me and [my wife].” Others expressed disappointment, making comments such as, “I anticipated some good techniques to eradicate the pain, but I never really got [them].” This perceived misalignment implies that some participants assumed the intervention was more physically-focused rather than relationship- and psychologically-focused. One participant recommended offering an explanation that the intervention is “not to cure your pain,” and in fact may not change pain levels at all, whereas others offered reasonable expectations of the intervention’s effects on pain. One woman said that the intervention “doesn’t take away the pain but it just gives you a different concept about how you can work to feel better about yourself,” and others supported this notion while adding that working together as a couple helped them feel overarching improvements in well-being. As one woman explained, “It’s a relationship therapy between us: you and I and chronic pain.”

Satisfaction with Treatment.

Participants reported that they were, on average, satisfied with the therapy sessions and activities (M = 7.65, SD = 1.52; 9-point scale, with 9 being completely satisfied). On average, participants also agreed they gained new knowledge during the intervention (M = 3.85, SD = .91; 5-point scale, with 5 being “strongly agree”). Many participants acknowledged that they liked all the sessions and that they were all helpful, and the sessions were often described as “relaxing.” One woman elaborated, saying that although life may have been stressful prior to coming in for the session, “by the time we left, everything felt calm and relaxed.” Additionally, participants reported that they would not prefer individual sessions over couple sessions (M = 1.77, SD = 1.07; 5-point scale, with 5 meaning they strongly agree with a preference for individual sessions), that they would pay for the treatment (M = 3.87, SD = 1.11; 5-point scale, with 5 meaning they strongly agree that they would pay for the intervention), and that they would recommend the treatment to others (M = 4.33, SD = 0.96; 5-point scale, with 5 meaning they strongly agree that they would recommend the treatment). In fact, one participant discussed during her interview that she had already recommended the treatment to someone she knows, saying that if someone is “just finding out about pain, [the intervention] probably would prevent them having marital problems.”

Furthermore, participants overall believed that the treatment fit with their spiritual, religious, and cultural practices. Couples praised the flexible, “non-denominational” approach of the program, saying they were able to modify skills to align with their views. For example, one man incorporated prayer into the mindfulness exercises, whereas another woman noticed that “if I had been a Buddhist, I could have made it work… if I was a Catholic it could still work,” and said that “nobody asked us to do anything outside of what we felt comfortable with.”

Participant Engagement.

After each therapy session, therapists completed the Therapist Weekly Feedback form in which they rated the participants’ engagement in the session and willingness to learn experiential mindfulness skills, explore values and goals, and learn communication strategies. They rated these areas on a scale from 0 to 4, with a rating of 4 being the most engaged and willing. Across all six therapy sessions, the average engagement rating was 3.33 (SD = .38) and the average willingness rating was 3.20 (SD = .73), indicating an overall engagement in the sessions and willingness to learn new techniques to improve relationships and manage pain and distress.

A few couples discussed their hesitation to engage in the therapy initially. One wife said she had to convince her husband to attend, but once he began, he realized it was “not so bad.” He elaborated, saying he felt engaged in the sessions because “they approached it from a pain standpoint not, ‘you’re broken let me fix you.’” Conversely, one participant acknowledged that she was “desperate for help” and “distraught,” and was convinced that there was nothing that could help her; therefore, she looked forward to coming to therapy and found herself very engaged, as it was something different than her weekly physican appointments.

Intention to Complete Homework Assignments.

After each session, participants rated their intention to complete that week’s homework assignment. On a 5-point scale, with 5 being confident that they will complete the assignment, the average intention rating overall was 4.80 (SD = .41). One participant explained that she “knew [the homework] is going to work [and] that it’s going to be effective… I knew that this homework is going to get me to the hurdle, get me through, get me there.”

Homework Acceptability.

Overall, five weeks of homework assignments yielded an acceptability score of 10.65 (SD = 1.93), meaning participants found the homework to be acceptable (range: 8.58–12.18). Couples described the assignments as “enlightening,” “beneficial,” and “grounding,” one participant believed that it improved her quality of life, and others appreciated how some exercises (e.g., mindful handholding) “brought us together” for quiet time after a busy day and helped them feel connected. Additionally, a majority of participants described the skill practice assignments (i.e., the mindfulness-based exercises) as “relaxing.” A quantitative evaluation of individual assignments indicate that participants found the assignments to be increasingly acceptable over the course of the intervention. One participant reinforced this outcome, saying “as you get more into it, it got more interesting.” The loving-kindness meditation (Week 5) had the highest acceptability rating (M = 12.71, SD = 4.02), with one participant describing it as the exercise that “put me on the road to a change in behavior.” One husband shared that he incorporated his religious beliefs into the meditations by integrating prayers for his wife, and this flexibility made the exercises more enjoyable for him.

Additionally, the values-based activities (i.e., completing a values compass and valued-based actions) and the emotional disclosure exercises were frequently mentioned during the interview as having good acceptability. Participants said that focusing on their values was “enlightening,” as it “helps you know what’s really important in your life. Because a lot of times you’re in pain but you’re just not thinking about it…. this helps you think about the other people in your life.” Participants also shared their enjoyment of the emotional disclosure exercises in which they were instructed to share negative feelings about pain and/or stress to their partners. One participant said she enjoyed this, as her “partner actually listened to it and didn’t try to fix it but just actually listened to it.”

In contrast, participants rated the anniversary exercise (Week 1) as the least acceptable assignment (M = 6.92, SD = 4.35). Further evaluation of this exercise indicated that, although participants believed they were able to satisfactorily complete the assignment and it seemed to match their therapy goals, the majority of participants reported only moderate to no enjoyment of the assignment. Additionally, the majority of participants reported that completing the assignment did not help them make meaningful progress in therapy, which may be related to participants’ aforementioned confusion about this first assignment not being specifically related to pain and being unsure as to why they were completing it. A full review of acceptability scores of individual homework assignments can be seen in Figure 1.

Figure 1:

Homework feasibility and acceptability ratings (see Table 1 for description of homework assignments)

Safety

There were no adverse events reported during the study, indicating that Mindful Living and Relating is a safe intervention to implement.

Preliminary Outcomes

After completing the intervention, participants rated their perceived improvement of their “overall status” using the Patient Global Impression of Change (PGIC) scale, a 7-point scale (1 to 7) ranging from “very much worse” to “very much improved.” In the overall sample, participants rated their improvement between minimally and much improved (M = 5.48, SD = 0.79). This rating was similar across those with and without pain (M_pain = 5.50, M_NoPain = 5.43). Specifically, more than half of participants (n = 16) rated their improvement as “much improved” (a Moderately Important improvement (Dworkin et al., 2008), one participant rated it as “very much improved” (a Substantial improvement (Dworkin et al., 2008), eight participants rated it as “minimally changed” (a Minimally Important improvement (Dworkin et al., 2008), and four participants reported “no change.” No one who completed the intervention reported that global status became worse over the course of treatment.

Some interview data appeared inconsistent with participant’s ratings of perceived global improvement. For example, one participant without pain acknowledged that he “found that some of the techniques…to deal with emotions, stress—it had a really positive effect on me.” His wife elaborated, saying their son recognized that his father “wasn’t as cranky or explosive” as he was prior to beginning the intervention. Despite his verbal reports of the intervention’s impact on his life, he rated his “overall status” as unchanged on the PGIC scale. This rating coincides with his unchanged BDI score of 0 and his DAS score remaining over 100; however, the “positive effect” he described during the interview was not captured by the quantitative measures.

In contrast, another participant with pain discussed his similar positive response to the intervention, saying that he noticed how participating in mindfulness activities after a long, frustrating, and busy day made him “come out feeling a hell of a lot better”; he rated his change in overall status as “much improved,” which supports his BDI score decrease of 11 points. Additionally, it is worth noting that participant ratings on the PGIC scale shared 6–10% of their variance with symptom specific measures: depressive symptoms (r = .329), pain severity (r = .246), or relationship satisfaction (r = .264), suggesting that the PGIC scale may be measuring an unknown construct that participants define independently. However, the magnitude of these correlations suggests small to medium effects.

Furthermore, despite the aforementioned misalignment between the intervention’s goals and pain relief expectations that some participants experienced, others discussed how the meditations were helpful in alleviating pain-related symptoms, so much so that they thought the activities could be an alternative to pain medications. One participant said that some of the exercises “might take your mind off the pain” instead of “constantly taking mind altering drugs.” Her husband also believed that if people tried this treatment approach they may be able to decrease pain medication usage. Comments on the intervention’s activities being effective in decreasing pain-related symptoms were supported by significant and marginally Clinically Important improvements (Dworkin et al., 2008) in daily pain-interference (i.e., physical functioning); however, there were no significant reductions in pain severity (see Table 2).

Finally, one woman with pain gave a descriptive overview of how the intervention gave her encouragement during a difficult time in her life when pain and anxiety were at their highest:

I felt like I’m sliding down a mountain, a cliff, or a rock and…my hands are bleeding my fingernails are broken off and I’m not making it, I’m sliding, I’m falling off I can’t help it…But [the intervention] helped me. I was so grateful and thankful and the main thing that helped me was the communication. I still have a lot of pain, but my breathing exercises help me to reduce some of it…It’s given me encouragement not to give up, and I’m not at that cliff level now. I’m resting nicely on the plateau, I still got rock climbing to do but I’m resting right now.

Her husband supported her and agreed, recognizing that his wife, through this intervention, learned how to cope differently with the pain and that she learned new “tools.” She acknowledged that she can “still survive and have a good life, and a happy life…now I have hope.” Her hope is also observed in her Clinically Important improvement in emotional functioning (Dworkin et al., 2008), as her BDI score decreased from a 53 (“severe depression) to 16 (“mild depression”). This improvement was aligned with Clinically Important improvements in emotional functioning (Dworkin et al., 2008) in the overall sample (as measured by the BDI; see Table 2). Those with pain (n = 23) had higher levels of depressive symptoms at baseline (M = 17.91, SD = 12.27) than those without pain (n = 7; M = 8.43, SD = 6.11), which may have contributed to the larger change scores (Pain M_change = 7.00; No-pain M_change = 2.86). However, the overall average depressive symptoms scores for those with and without pain were within normal limits at post-intervention (Pain M_BDI = 10.91, SD = 6.84; No-pain M_BDI = 5.57, SD = 6.02). Furthermore, individuals with pain reported improvements in perceived control over their life (d = .20), a small effect.

Relationship Rebuilding

During the open coding process, significant patterns of emotions, communication, and emotional intimacy and quality time emerged across the majority of interviews as primary aspects of a larger relationship rebuilding theme. This theme is defined as the transition couples made during the intervention, from relationships where communication struggled and partners felt disconnected from one another, to a place where partners felt reconnected and relationships were rebuilding. The ways in which participants spoke about feeling intimately reconnected after the intervention provides meaningful information that was not captured by the quantitative measures and which inform intervention acceptability and feasibility.

The feedback couples offered about changes in their relationship were unprompted; couples found ways to incorporate this topic into the discussion without being directly asked about their relationship. This speaks to the level of importance that couples place on their relationship, as when they were given the chance to speak openly, they wanted to talk about it. As one goal of Mindful Living and Relating was to improve relationship satisfaction, these data also suggest that the intervention was successful, impacting relationships by revising how partners relate to one another.

Several themes emerged including being emotional expression, communication skills, and emotional intimacy and quality time together. Couples learned to view things from the other’s point of view and got “in touch with…inner feelings of yourself and your partner.” Additionally, one woman shared that she and her husband learned to “put words to our feelings toward one another,” as she realized before treatment that she was simply using the same words to express different feelings. Others even acknowledged the role that working on emotions with their partners played in improving pain-related symptoms.

Couples acknowledged that they have always understood how important communication is within relationships but admitted to having difficulty communicating with their partner prior to the intervention; however, they felt empowered to improve communication after the completing treatment because it gave them appropriate techniques to not only “cure” arguments but also “to express your love and compassion to the other person” and confide in one another. One woman with pain explained that “[the intervention] helped us work out our problems” and it “was very beneficial… to get [my husband] to where I need him to open up and talk to me…I feel like it’s done a world of good for us as a couple.” The vast majority (11 of 14) couples who were interviewed credited the intervention for increasing and improving communication within their relationships, including learning how to take their partners’ perspective instead of being so inner-focused. Furthermore, couples appreciated that the intervention’s activities afforded them the opportunity to relate to each other in positive ways, while learning to express how their partners’ actions affected their own mood and emotions. These improvements in partner responsiveness were significant in the overall sample (see Table 2). Additionally, the “tools” they received helped partners respond differently to pain expressions, as there were small effects on decreases in partners’ negative responses (d = .18), large effects on increases in distracting responses (d = .94), and small to medium effects on increases in solicitous responses (d = .25). Individuals with pain also reported small to medium effects increases in perceived support (d = .28).

Finally, couples spoke about how intimacy had been lacking in their relationships and knew it was something they needed to focus on but were not sure how to do so. One woman shared how the intervention impacted her relationship:

The intimacy has been missing from our relationship for a long time. So, that was one of the serious things that I felt like we needed to focus on, work on. I didn’t expect it to happen from this, but it did and thank goodness…it’s a definite benefit to me.

Furthermore, many couples shared their appreciation for the intervention, reminding them how to enjoy spending time with each other again. Some also acknowledged how their pain has impeded on quality time, which has caused feelings of burden and resentment in themselves and their partners. Fortunately, a number of couples noticed that the format of treatment itself helped couples relearn how to be with each other; for example, one woman commented that she had been through individual therapy before, but the current one was “better because I have a partner with me and we’re both on the same page.”

Despite the majority of couples sharing how much the intervention positively impacted their relationship and the positive increases in perceived partner responsiveness, no significant changes on the DAS were observed (see Table 2). There were no significant differences in changes in relationship satisfaction between those with and without pain. One participant offered an explanation to these findings, based on advice she received from her Mindful Living and Relating therapist:

You can’t expect a person to change completely in one step…the change has to occur in baby steps and I realized that I was being much too hard on [my husband] in my expectations. And not in the sense of lowering my expectations but just not being so disappointed when things didn’t keep going smoothly…it’s a progression of change that you need to work on. You need to be working at it and it will get better as time goes along if both people want it to get better.

Discussion

Mindful Living and Relating is a novel intervention that positively impacted couples experiencing chronic pain and relationship distress. Improvements were observed in symptoms of depression, relationship satisfaction, partner responsiveness, and pain-related interference. Furthermore, this intervention was acceptable to its users. The results of this study reinforce the idea that the needs of individuals in pain and their partners should be considered when designing pain-related interventions, especially in situations where there is relationship distress (Leonard et al., 2006). Including the partner as an active participant and a focus of treatment, rather than the role of a pain management coach or guide, may result in more robust changes for both partners, as it is known that family members benefit from interventions when they are considered a target in treatment (Martire & Schulz, 2007). Without both partners working to improve their distress, communication, and relationship satisfaction, improvements such as the ones observed in this study would not likely be plausible.

Unique to the current study was the incorporation of post-treatment semi-structured interviews, which provides valuable information for intervention development. Couples reported that they found the intervention meaningful and that relationship-building activities contributed, in part, to their treatment gains. It is apparent from interview data that not only did couples change the way they relate to each other, make substantial improvements in how they communicate, and reconnect after feeling distant, but many couples effectively received the skills they needed during treatment to feel confident to continue to work on their relationship after study completion. Couples described an impactful shift in their relationships that was not fully captured by the quantitative measures used in the original study; this observation highlights the importance of utilizing qualitative measures when creating interventions, testing their feasibility, and evaluating their effects. Whereas many intervention developers likely collect qualitative data to inform their treatments, it is recommended that they also analyze and report these findings to inform the pain intervention literature (regardless of whether significant others are included).

With regard to feasibility, it is impressive that, with one exception due to events external to the therapy, all couples who began the therapy sessions were adherent to treatment and completed all six sessions. This indicates that couples’ in-session experiences were positive and engaging enough to retain their interest. Additionally, participants found the session and homework material to be rationale and comprehensible. Although some participants encountered barriers to completing assignments, most participants were able to overcome obstacles with the support of their partners and therapists. Recruitment difficulties speak to challenges inherent in conducting in-person interventions for couples with chronic illnesses, which has been observed in other couple-based treatments for health conditions (see Baucom et al., 2009).

Furthermore, participants’ perceptions of the acceptability of homework assignments increased over the course of the intervention, which also suggests a positive user experience. As participants progressed in the treatment, continued to learn new information, and overcame barriers to completing assignments (physical, psychological, or logistical), they realized how impactful the treatment was—this is even true for those who were resistant to beginning the intervention. Introducing potentially stress- and pain-relieving activity in the first session may lead to a more satisfactory first week of treatment.

The preliminary efficacy data suggest that Mindful Living and Relating has the potential to improve both individual and relationship outcomes. Decreases in depression were observed in the overall sample, and reduced pain interference was observed in those participants with pain. Couples also reported that the intervention positively impacted their relationships (as observed in the interview data) and perceived partner responsiveness (as shown by significant increases on the PPRS), and individuals with pain reported significant increases in distracting responses from partners during pain expressions (as measured by the MPI). Yet, relationship satisfaction, as assessed by the DAS, did not change by engaging in the treatment. There are at least two possibilities for this finding. It is possible that the DAS is not sensitive to the types of relationship changes experienced by couples, that the limited range in DAS scores increased the difficulty to detect changes, or that it is easier for couples to identify specific changes (as they did during interviews) that do not immediately translate to global perceptions of change.

The next step in this research is to test the efficacy of the intervention against an active control group, and implement this intervention with a larger sample in order to test the effects of moderating variables, including pretreatment relationship satisfaction and distress.

Despite these positive effects, limitations of the therapy as conducted in this trial should be addressed in a future, improved version. The majority of couples were satisfied with the intervention and its impact on their symptoms and relationships; yet, a few couples expressed disappointment with the misalignment between their initial expectations that the intervention will alleviate their pain and the outcomes they experienced (i.e., non-significant reductions in pain severity). The treatment rationale handout given to participants explained that “the aim of this intervention is to teach couples how to cope with pain…” and “…your therapist will share different skills that can help you manage pain…;” however, some participants may have interpreted “manage” as “reduce” or “eliminate” pain. As prior research has shown that pretreatment expectations are associated with positive outcomes, including reductions in pain severity (Cormier, Lavigne, Choiniere, & Rainville, 2016), it is important to set expectations about specific outcomes and to use language and concepts that convey the treatment rationale clearly. Qualitative analyses suggest that adjusting specific content may lead to even greater effects for individuals with pain, including modifying the aims of the intervention, adjusting the delivery of the treatment rationale, and focusing on collaborating with individuals to set clear and attainable goals. Furthermore, the variance in ratings on the Patient Global Impression of Change (Guy, 1976) scale indicate that this specific scale used to measure perceived improvement may not be clear; that is, “overall status” may mean different things for different people, leading them to focus on individual changes that were important to them while completing the measure. Specifying the specific domain may be helpful in obtaining information on perceptions of outcomes, especially for interventions that target both individual and relationship outcomes.

In addition, treatment fidelity was not assessed in a formal manner (for example, a trained independent observer rating the therapist’s adherence to the treatment manual and skill level; see Heapy et al., 2017). Therefore, it remains possible that there was variation in treatment effects within and between therapists. Finally, as this was a treatment development study with a small sample, the preliminary outcome findings must be replicated in a larger study and comparisons with an appropriate control conditions are necessary before this treatment can be more widely adopted. Nevertheless, this study obtained comprehensive feedback from participants on a novel couple-based treatment for pain and interpersonal distress that led to observable positive impacts on individual and relationship functioning. Future directions include exploring specific relationship building activities that can positively impact couples with pain and support their adaptive pain management behaviors, and these activities may be adapted for non-coupled individuals who are cared for by their adult children.

Recruitment difficulties also presented unforeseen challenges during the original study that are important to address. Of the couples who were interested and eligible, only 32% enrolled (23 out of 73 couples) and 68% expressed disinterest (50 out of 73 couples), and we have thoughts on why these numbers were low. First, some interested couples were excluded because their relationship satisfaction was too high to qualify for the study, implying that they are functioning well interpersonally (as measured by the DAS), despite the presence of chronic pain; therefore, future studies may lower this relationship satisfaction threshold to see if couples with chronic pain who are in less distress can still benefit from this novel treatment. Couples who do not report any relationship distress may be better suited for spouse-assisted treatments, as discussed by Keefe et al. (1996), as couples who are functioning well interpersonally may benefit from such treatments (Martire et al., 2010). Furthermore, many interested couples were excluded because they were not married or living together for one year, so future couple-based pain research may choose to include any cohabitating couples who are interested in dedicating their time to improving their relationship, individual wellness, and pain-related symptoms. Some couples may have also not been interested in or convinced of a couples-based format to alleviate symptoms of chronic pain, which highlights the importance of giving a comprehensive treatment rationale in the earliest stages of recruitment. Finally, many interested couples disclosed that they were unable to make the time commitment, as finding time in two partners’ schedules to dedicate to treatment was difficult—it is unclear if this was due to the 1.5 hour weekly time commitment for 6 weeks, compared to traditional 50 minute therapy sessions, or if any in-person treatment for two partners would have been difficult to accommodate. Given these challenges in recruitment, it may be useful to explore alternative intervention delivery formats. Future research may investigate the efficacy of a 2-session relationship intervention for chronic pain, similar to the 2phase format of the Marriage Checkup, where a comprehensive assessment is followed by a motivational feedback and brief skill-building session in order to highlight the couple’s strengths and areas for improvement (Morrill et al., 2011). Alternatively, online platforms have been shown to be the preferred methods for individuals seeking improvements in their relationships (Georgia & Doss, 2013; Nowlan, Roddy, & Doss, 2017). Online interventions for both chronic pain and relationship distress (although none have been conducted on comorbid pain and relationship distress) have been shown to be efficacious in improving individual (pain severity and interference, depression, anxiety, quality of life, perceived health) and relationship (confidence and satisfaction) factors (see Doss et al., 2016; Ruehlman, Karoly, & Enders, 2012). Researchers may also explore utilizing social media platforms where there is already a sizeable presence of individuals with chronic pain seeking support, advice, and guidance, though the effectiveness of such platforms is currently unknown. These alternative methods have the potential to reduce barriers for not only couples, but for health care delivery systems.

In sum, the current study highlights the potential value of utilizing couple-based approaches for treating chronic pain. Including the partner and intimate relationship as targets of intervention means two significant and influential aspects of an individual’s daily life are considered, rather than ignored, as happens in most treatments. This couple-based intervention for pain was acceptable for participants, and the dyadic approach led to positive outcomes in both the individual with pain and the partner, which were observed through both the qualitative and quantitative analyses. As some of these reported benefits were not detected from the quantitative analyses, these findings also have implications for pain intervention research. It is recommended that pain intervention studies incorporate qualitative methods so that participant experiences can be used to illuminate quantitative outcomes and modify the intervention for future use, which may lead to increased efficacy for its participants. Using couple-based approaches and collecting qualitative data afford clinical researchers with key processes and knowledge to make sustainable improvements in the lives of individuals experiencing chronic pain and the people most important to them.