Improvements in pain interference among geographically diverse adults with neurofibromatosis: Results from a fully powered randomized controlled trial

Abstract

Background

Pain is prevalent among adults with neurofibromatoses (NF) and hinders quality of life. Pain management for NF is predominantly pharmacological and often ineffective. Psychosocial treatments improve pain outcomes in other chronic illness populations but have not been developed and tested in fully powered efficacy trials among adults with NF. Using data from a fully powered randomized clinical trial of an 8-week mind–body program (Relaxation Response Resiliency Program for NF [3RP-NF]) versus a health education control (HEP-NF), we examined (1) improvements in pain intensity and pain interference, and (2) mechanisms underlying improvements.

Methods

Participants (N = 210, M_age = 42.6, 73.4% female) were randomized to 3RP-NF versus HEP-NF. They completed measures of pain intensity, pain interference, and putative mechanisms (eg, mindfulness, coping, and social support) at baseline, post-intervention, 6-month, and 12-month follow-ups.

Results

There was a statistically significant change in pain interference over time F(3, 537.06) = 7.21, P < .001, but not pain intensity. Neither group (3RP-NF vs. HEP-NF) nor the group-by-time interaction predicted change in pain interference. While the group-by-time interaction was not statistically significant across all time points, planned post-hoc analyses probing the interaction at specific time points revealed a statistically significant decrease in pain interference from baseline to post-intervention (P < .001), which was sustained (ie, no subsequent change) from post-intervention through 6-month (P = 1.00) and 12-month follow-ups (P = 1.00) in the 3RP-NF group. The HEP-NF group had no significant changes in pain intensity or interference over time. The association between group (3RP-NF) and decreased pain interference from baseline to post-intervention was fully mediated by change in coping over the same period.

Conclusions

Participation in the 3RP-NF is associated with sustained improvement in pain interference. Improvement occurred through increased coping.

Trial Registration

ClinicalTrials.gov Identifier: NCT03406208

Trial Registration URL

https://clinicaltrials.gov/study/NCT03406208

Neurofibromatoses (NF; including NF1, NF2-related schwannomatosis [(NF2), previously called neurofibromatosis type 2], and other forms of schwannomatosis) are heterogeneous, incurable syndromes characterized by the development of nerve sheath tumors.¹ Pain is highly prevalent across the NF diagnostic spectrum, with some estimates indicating that approximately 64%–70% of individuals with NF1^2,3 and 74% of individuals with NF2 experience pain.^3,4 Pain is also a core symptom of schwannomatosis,⁵ though prior work has indicated that the experience of pain may differ by NF diagnostic subtype.⁶ Despite being less associated with NF1 and NF2 historically, more recent evidence indicates that pain is present across each of the NF diagnoses and contributes to poorer quality of life.⁷ Potential causes of pain in NF include tumor growth or removal, headaches, neuropathy, and other illness-related complications.^3,8,9 NF-related pain can significantly interfere with daily functioning and quality of life.^10,11 Importantly, patients with NF have identified pain as an important treatment target and outcome when considering joining NF-specific treatment trials.⁷ Consequently, there is a critical need for pain-focused interventions in this population.

Current NF treatments are primarily limited to surgical interventions and palliative approaches,¹² and patients with NF have reported that they view pharmacotherapies (eg, opioids) as having little to no effect on their pain.^4,13 Complementary, nonpharmacological treatments for pain (eg, yoga, Acceptance and Commitment Therapy, Cognitive–Behavioral Therapy) have been explored in pilot trials and demonstrate promising effects on pain outcomes among individuals with NF1.^12,14,15 However, these studies have primarily assessed feasibility, are small in scale, and focused exclusively on individuals with NF1 despite the prevalence of pain across NF subtypes. Our prior work developing the Relaxation Response Resiliency Program for Neurofibromatosis (3RP-NF)^16–20 aims to address this gap. The 3RP-NF is a virtual mind–body intervention developed to enhance quality of life, resiliency, and illness-related coping, and to reduce distress among adults with NF.¹⁶ This intervention was intentionally designed to account for the shared psychosocial difficulties and heterogeneity across the 3 NF subtypes¹⁷ and is delivered in a group format to foster social support. Importantly, 3RP-NF targets specific psychosocial resiliency factors strongly associated with favorable pain-related outcomes (eg, mindfulness, shifting maladaptive thinking), but these factors remain underexplored among adults with NF who experience pain.

Several psychosocial resiliency factors have been associated with improved pain outcomes (eg, decreased pain intensity and pain interference) among adults with chronic pain,^21,22 such as mindfulness, coping (eg, relaxation, disputing maladaptive thoughts), and social support. Our prior work suggests that interventions targeting these psychosocial factors can improve short-term quality of life and overall resilience among patients with NF,^18,20 but the effects on pain among this population are unclear. Similar interventions have evinced decreases in pain intensity (ie, subjective assessment of the magnitude of experienced pain) and pain interference (ie, the extent to which pain hinders engagement with/participation in physical, cognitive, emotional, recreational, and social activities).²² Comparable effects among individuals with NF may improve engagement with recreational and work-related activities. Adaptations of the 3RP-NF program targeting other medical populations (eg, individuals with chronic musculoskeletal pain) have been found to significantly improve pain,²³ and our NF-specific program has demonstrated significant improvements in coping outcomes among adults.²⁴ 3RP-NF was specifically developed and pilot-tested for individuals across the NF diagnostic spectrum¹⁸ and thus may be well-suited for reducing pain intensity and pain interference among this heterogenous population.

Given that very few pilot studies and no fully powered efficacy psychosocial randomized clinical trials (RCTs) have focused on adults with NF, there is a clear need for targeted psychosocial interventions for pain among adults across the NF diagnostic spectrum. The present study aimed to examine the effects of 3RP-NF (versus a health education program control group [HEP-NF]) on pain intensity and pain interference across three time periods (post-intervention, 6-month follow-up, and 12-month follow-up). We hypothesized that participants randomized to 3RP-NF versus an educational control would demonstrate improvements in pain intensity and pain interference at post-intervention and that these improvements would be maintained at both 6-month and 12-month follow-ups (Hypothesis 1). Additionally, we tested mindfulness, coping skills, and social support as mediators of the effect of the treatment group on changes in pain intensity and pain interference. We hypothesized that changes in these psychosocial factors would mediate the association between group and pain outcomes at each time point (Hypothesis 2).

Methods

Patient Population and Data Source

Study procedures were approved by the Institutional Review Board at the Massachusetts General Hospital. We assessed 371 patients with NF for eligibility after recruiting via email listserv advertisements through the NF registry within the Children’s Tumor Foundation and NF centers worldwide. Two hundred twenty-eight patients were deemed eligible and randomized. Inclusion criteria were (1) diagnosis of NF1, NF2, or schwannomatosis by a medical professional, (2) ≥18 years of age, (3) ability to provide informed consent, (4) English literacy equivalent to a sixth-grade reading level or above, (5) self-reported stress and difficulties coping with NF symptoms, and (6) score ≥6 on the Perceived Stress Scale-4 items.²⁵ Exclusion criteria were (1) severe psychopathology (eg, schizophrenia or active substance use) that would interfere with study procedures, or (2) being unwilling or unable to participate in virtual group sessions. Within the present study, only participants who endorsed any pain at baseline were included. Individuals had to endorse a pain rating of at least a 1 (low pain) out of 10 (worst pain imaginable) during their baseline visit to be included in the present analyses, allowing for a broad range of pain experiences to be captured. Participants were enrolled between October 2017 and January 2021. Participants joined from a variety of geographic locations (including Australia, Canada, the United Kingdom, and the United States)¹⁶ and sessions were scheduled to accommodate optimal times across several time zones. The final follow-up outcome assessment was completed in February 2022.¹⁶

Participants provided written informed consent for all procedures and did not receive compensation for participation. The study was advertised as a Stress Management Program to maintain the single-blind study design. All participants met virtually with study staff to review consent procedures, inclusion/exclusion criteria, and the safety protocol. Assessments were completed at baseline, post-intervention, 6-month, and 12-month follow-up via REDCap.²⁶

Active and Control Intervention Conditions

A detailed description of both the active and control conditions, along with intervention contents, has been published previously.¹⁶ Each condition, 3RP-NF and Health Enhancement Program-NF (HEP-NF), included eight 90-min group sessions conducted via Zoom videoconferencing. Three clinical psychology doctoral trainees specializing in mind–body medicine served as study clinicians; they were trained and supervised by a senior clinical psychologist with expertise in mind–body interventions and NF who developed both programs.^18,27

Active Intervention

The 3RP-NF, adapted from the original 3RP,²⁸ is a comprehensive, multimodal group treatment designed to increase quality of life and coping with NF symptoms. This program has 3 core components: (1) Relaxation Response elicitation, which teaches relaxation and mindfulness techniques (eg, deep breathing, mindful awareness); (2) Appraisal and Coping, which teaches adaptive coping skills to manage stress and medical symptoms such as pain and disease burden (eg, identifying and challenging negative thoughts); and (3) Growth Enhancement, which teaches acceptance, problem-solving, and positive psychology skills (eg, positive storytelling, guided imagery) to increase resiliency in the face of new symptom development and uncertainty of prognosis.¹⁹

Control Condition

The HEP-NF was adapted from HEP (an established active control for mind–body interventions)²⁷ and NF-specific materials from the Children’s Tumor Foundation website. The structure and timing of this program were matched to that of 3RP-NF. Each module provides educational information on NF and overall health, including stress, symptoms, sleep, nutrition, exercise, communication, and healthcare management.

Measures

Pain Intensity.—

The Graded Chronic Pain Scale—Characteristic Pain Intensity (CPI) subscale (GCPS-CPI)²⁹ is a reliable and valid method of assessing global pain intensity across a range of chronically painful conditions.³⁰ Participants were asked to rate their pain “right now,” their “worst” pain in the past week, and their pain “on average” during the past week on separate scales ranging from 0 to 10, with higher total scores (range = 0–30) indicating more intense pain (3 items total). We only computed a total pain intensity score for participants who completed all 3 items. An average of the 3 items was multiplied by 10 to yield a total score in a similar range as the full GCPS. Consistent with prior work,⁶ we adapted the GCPS to measure worst pain and average pain within the past week to ensure that we captured pain symptoms proximal to other measured variables (α = 0.915).

Pain Interference.—

The 8-item PROMIS v1.0—Pain Interference—Short Form 8a measures self-reported consequences of pain on daily activities over the past 7 days using a 5-point Likert scale (1 = “Not at All”; 5 = “Very Much”; PROMIS—PI8).³¹ Items are summed to create a total raw score, which is converted to a T-score (ie, standardized score; M = 50, SD = 10). We computed a total pain interference score for each participant if they completed at least 75% of the items. This total score equaled the average score across completed items multiplied by 8 (ie, the total number of items; α = 0.976).

Mindfulness.—

The Cognitive and Affective Mindfulness Scale—Revised (CAMS-R) measures the degree to which individuals are mindful of and experience their thoughts and feelings in the present moment using 12 items rated on a 4-point Likert scale (1 = “rarely/not at all”; 4 = “almost always”).³² Higher scores represent greater mindfulness (range = 12–48; α = 0.828).

Coping.—

The Measure of Current Status-A (MOCS-A)³³ is a 13-item measure assessing perceived ability in using relaxation, recognizing tension and stress-inducing situations, assertiveness, and ability to restructure thoughts and choose appropriate coping responses using 13 items rated on a 5-point Likert scale (0 = “I cannot do this at all”;4 = “I can do this extremely well”). Higher scores represent greater perceived coping ability (α = 0.886).

Social Support.—

The 19-item Medical Outcomes Study Social Support Survey (MOS) assesses perceived availability of social support across 4 domains: emotional, tangible, affectionate, and positive social interaction. Items are rated on a 5-point Likert scale (1 = “none of the time”; 5 = “all of the time”), with higher scores representing greater perceived social support (α = 0.964).³⁴

Sociodemographic Information.—

Participants self-reported their age, sex, race, ethnicity, education, marital status, and NF diagnostic subtype (NF1, NF2, and SCHWN).

Analytic Strategy

Data analysis was conducted in SPSS 28.³⁵ Because our analyses examined the association between 3RP-NF and (decreased) pain over time, we excluded participants (n = 14) who denied pain at baseline (ie, GCPS pain intensity score of 0). Next, we ran bivariate correlations, chi-square tests, and 1-way analyses of variance (ANOVAs) to assess whether demographic and clinical variables were evenly distributed between groups (ie, confirming successful randomization). Analyses revealed no statistically significant differences between groups, so we did not include any covariates for our hypothesis tests. For all primary analyses, continuous variables were standardized as z-scores to account for differences in scaling.

To test Hypothesis 1 (the effect of 3RP-NF on pain), we constructed linear mixed models with group (HEP-NF = 0; 3RP = 1), time (baseline = T1, post-intervention = T2, 6-month follow-up = T3, 12-month follow-up = T4), and the group × time interaction predicting pain intensity and pain interference in respective models. We included random intercepts with restricted maximum likelihood (REML) estimation and computed effect size using partial eta-squared. We included planned contrasts in the linear mixed models with Bonferroni corrections to examine change in pain intensity and interference for 3RP-NF and HEP-NF from each time point to the next (ie, T1 to T2, T2 to T3, and T3 to T4).

To test Hypothesis 2, we examined the mediating effects of changes in mindfulness, coping, and social support on change in pain intensity and pain interference in separate models using the SPSS PROCESS macro, Version 4.3.³⁶ We a priori planned to conduct mediation analysis solely on pain outcomes that changed significantly over time based on the Hypothesis 1 analysis. Specifically, we estimated the effect of group on each proposed mediator (path a), the effects of each proposed mediator on pain outcomes (path b), the indirect effect of group on pain outcomes via each proposed mediator (path a × b), the total effect of group on pain outcomes, excluding the mediator (path c), and the direct effect of group on pain outcomes, including the mediator (path c’). We assessed the magnitude of effects using standardized beta coefficients. We evaluated the reliability of indirect effects using bootstrapped standard errors and 95% confidence intervals (CIs; resampling rate = 10 000), consistent with the literature.^37–39

To circumvent bias due to attrition, we conducted intention-to-treat analyses for both hypotheses whereby all participants who were randomized to a treatment group were analyzed as part of that group, regardless of how many study time points they completed. When reporting change in variables over time within linear mixed models, we reversed the signs for coefficients such that an increase from T1 to T2 is represented by a positive coefficient and a decrease from T1 to T2 is represented by a negative coefficient (for ease of interpretation).

Results

Of the 228 participants randomized to 3RP-NF or HEP, 4 did not complete the intervention and 14 were excluded from the current analyses due to no reported pain at baseline. Thus, 210 participants were analyzed in our hypothesis tests. The sample consisted of 73.4% women, and the mean age was 42.59 (SD = 14.5). Participants had 14.41 years of education on average. Participants in the present sample reported experiencing moderate pain intensity “right now” as measured by the GCPS (M_{pain rating} = 4.19, SD = 2.78, 95% CI: 3.81–4.56), a moderate–severe “worst pain” on average across the past week (M_{pain rating} = 6.57, SD = 2.62, 95% CI: 6.21–6.93), and moderate pain “on average” over the past week M_{pain rating} = 4.74, SD = 2.49, 95% CI: 4.39–5.07). See Table 1 for a complete report of demographics by treatment group, and Table 2 for a report of mean scores by treatment group of all pain variables and hypothesized mediators at each time point.

Table 1.

Demographics and Clinical Characteristics (n = 210)

Characteristic	3RP-NF (n = 105)	HEP-NF (n = 105)
NF type, n (%)
NF1	76 (72.4)	74 (70.5)
NF2	15 (14.3)	16 (15.2)
Schwannomatosis	14 (13.3)	15 (14.3)
Age, mean (SD)	42.8 (14.1)	42.4 (15.0)
Gender, n (%)
Female	74 (70.5)	79 (75.2)
Male	31 (29.5)	26 (24.8)
Race, n (%)
American Indian/Alaskan Native	0 (0.00)	1 (1.0)
Asian	3 (2.9)	4 (3.8)
Black/African American	4 (3.8)	3 (2.9)
White/Caucasian	88 (83.8)	90 (85.7)
More than one race	8 (7.6)	4 (3.8)
Choose not to answer	2 (1.9)	3 (2.9)
Ethnicity, n (%)
Hispanic or Latino/Latina	4 (3.8)	8 (7.6)
Not Hispanic or Latino/Latina	101 (96.2)	93 (88.6)
Choose not to answer	0 (0.00)	4 (3.8)
Marital status, n (%)
Married	42 (40.0)	47 (44.8)
Living with someone in a committed relationship	7 (6.7)	8 (7.6)
Single	44 (41.9)	42 (38.1)
Separated	1 (1.0)	1 (1.0)
Divorced	10 (9.5)	5 (4.8)
Widowed	1 (1.0)	2 (1.9)
Choose not to answer	0 (0.0)	2 (1.9)
Years of education, mean (SD)	14.4 (3.47)	14.4 (4.37)
Learning disability, n (%)
Yes, I was diagnosed with one	24 (22.9)	26 (24.8)
I think so, no formal diagnosis	22 (21.0)	17 (16.2)
No	45 (42.9)	53 (50.5)
I don’t know	13 (12.4)	8 (7.6)
Choose not to answer	1 (1.0)	1 (1.0)
Baseline pain intensitya, n (%)
Low	33 (31.4)	34 (32.4)
Moderate	56 (53.3)	58 (55.2)
High	16 (12.4)	13 (15.2)

3RP-NF = Relaxation Response Resiliency Program for NF; HEP-NF = health education control for NF; NF = neurofibromatosis.

^aAs measured by the Graded Chronic Pain Scale—characteristic pain intensity (GCPS-CPI) subscale average past-week pain rating.

Table 2.

Mean Scores of All Measures at All Time Points (n = 210).

	Baseline		Post-test		6 Months		12 Months
	3RP-NF	HEP-NF	3RP-NF	HEP-NF	3RP-NF	HEP-NF	3RP-NF	HEP-NF
	n = 105	n = 105	n = 98	n = 101	n = 88	n = 95	n = 81	n = 87
Pain intensity	23.08 (10.10)	22.05 (10.08)	20.04 (9.68)	21.17 (10.22)	20.25 (10.72)	20.67 (10.11)	20.63 (9.99)	21.90 (9.63)
Pain interference	52.54 (23.44)	50.76 (25.37)	47.96 (26.47)	48.55 (25.66)	48.76 (26.10)	48.71 (26.31)	48.41 (25.70)	50.76 (25.37)
Social support	3.53 (1.02)	3.63 (0.97)	3.87 (0.98)	3.91 (0.89)	3.95 (1.05)	3.88 (0.93)	3.91 (1.03)	3.88 (0.85)
Coping	1.85 (0.76)	1.83 (0.70)	2.53 (0.72)	2.26 (0.77)	2.44 (0.78)	2.16 (0.79)	2.34 (0.84)	2.21 (0.73)
Mindfulness	28.72 (6.46)	30.31 (6.30)	32.42 (6.06)	33.37 (6.14)	32.42 (6.13)	33.37 (6.76)	31.85 (6.61)	32.01 (6.09)

Social support = Medical Outcomes Study Social Support Survey (MOS; range: 1–5, higher scores indicate greater perceived social support); Coping = Measure of Current Status-A (MOCS-A: range: 1–5, higher scores represent greater perceived coping ability); Mindfulness = Cognitive and Affective Mindfulness Scale—Revised (CAMS-R; range: 12–48, high score indicates greater mindfulness); Pain interference = PROMIS v1.0—Pain Interference—Short Form 8a (range: 0–100, higher total score indicates more pain interference); Pain intensity = The Graded Chronic Pain Scale—Characteristic Pain Intensity (GCPS-CPI) subscale (range: 0–30, higher score indicates more intense pain). Standard deviations are reported in parentheses.

Effects of 3RP-NF and HEP on Pain Intensity and Pain Interference Over Time

Pain Intensity.—

Linear mixed modeling results for Hypothesis 1 revealed no statistically significant change in pain intensity across time points, F(3, 542.90) = 1.75, P = .155. Pain intensity ratings did not differ by group (3RP-NF vs. HEP-NF), F(1, 217.14) = 0.001, P = .977, and the group-by-time interaction was not significant, F(3, 542.90) = 1.25, P = .292 (ie, pain intensity did not change over time for 1 study group more than the other; Figure 1).

Figure 1.

Graph of pain intensity and pain interference over time between 3RP-NF and HEP-NF groups. The Y-axes for both plots are truncated to better visualize changes and differences between groups. The pain interference scale (PROMIS-PI) ranges from 0 to 100, with scores representing T-scores (mean = 50, SD = 10). The pain intensity scale (GCPS) ranges from 0 to 30. Error bars represent the standard error of the means of pain intensity and interference. 3RP-NF = Relaxation Response Resiliency Program for NF; GCPS = Graded Chronic Pain Scale; HEP-NF = health education control for NF; NF = neurofibromatoses.

Pain Interference.—

In contrast, there was a statistically significant change in pain interference across time points, F(3, 537.06) = 7.21, P < .001, η_p² = 0.039 (Figure 1). There was no statistically significant difference in pain interference by group, F(1, 219.04) = 0.20, P = .659, η_p² = 0.001, and the group-by-time interaction did not predict change in pain interference when considering all time points, F(3, 537.06) = 2.60, P = .052, η_p² = 0.014. However, planned post-hoc analyses of this group-by-time interaction revealed that pain interference significantly decreased from baseline to post-intervention (estimate = -3.10, SE = 0.76, P < .001) and was sustained (ie, no subsequent changes) from post-intervention to 6-month follow-up (estimate = −0.35, SE = 0.80, P = 1.00) and from 6-month to 12-month follow-up (estimate = 0.65, SE = 0.83, P = 1.00) in the 3RP-NF group. The HEP-NF group had no significant changes in pain interference over time.

Testing Mediation Through Mindfulness, Coping, and Social Support

Per the results of the post-hoc analyses earlier, our mediation analyses focused on potential mechanisms of change in pain interference from baseline to post-intervention. We investigated whether changes in mindfulness, coping, and social support from baseline to post-intervention mediated the effects of the 3RP intervention (ie, “group”) on pain interference. We computed change scores (T2–T1) for the outcome and mediators. Because group was coded as 3RP-NF = 1 and HEP-NF = 0, positive coefficients for group indicate an association between 3RP-NF and the outcome while negative coefficients indicate an association between HEP-NF and the outcome.

Testing Changes in Mindfulness as a Mediator in the Association Between Intervention Group and Pain Outcomes at Each Time Point.—

We examined whether improvements in mindfulness explained the effect of the intervention on pain outcomes over time. In the model including mindfulness as a mediator, group (3RP-NF) was not associated with change in mindfulness (path a: β = 0.10, SE = 0.14, 95% CI = [−0.18, 0.39], P = .466). Change in mindfulness was associated with change in pain interference (path b: β = −0.29, SE = 0.07, 95% CI = [−0.43, −0.16], P < . 001). Neither the total effect (ie, the effect of group on pain interference, factoring out mindfulness; path c: β = −0.28, SE = 0.14, 95% CI = [−0.56, 0.00], P = .052) nor the direct effect (ie, the effect of group on pain interference, including the mindfulness mediation path; path c’: β = −0.25, SE = 0.14, 95% CI = [−0.52, 0.02], P = .071) of group on change in pain interference were statistically significant. There was no indirect effect of group on change in pain interference through change in mindfulness (path a × b: β = −0.03, SE_Boot = 0.04, 95% CI_Boot = [−0.12, 0.05]), suggesting that change in mindfulness did not mediate the effect of group on change in pain interference (ie, changes in mindfulness were not the reason that the intervention influenced pain interference over time).

Testing Changes in Coping as a Mediator in the Association Between Intervention Group and Pain Outcomes at Each Time Point.—

We examined whether improvements in coping skills explained the effect of the intervention on pain outcomes over time. In the model including coping as a mediator, group (3RP-NF) was associated with increased coping (path a: β = 0.36, SE = 0.14, 95% CI = [0.08, 0.64], P = .012). Increased coping was associated with decreased pain interference (path b: β = −0.27, SE = 0.07, 95% CI = [−0.40, −0.13], P < .001). There was a statistically significant total effect of group on change in pain interference, factoring out coping as a mediator (path c: β = −0.30, SE = 0.14, 95% CI = [−0.58, −0.02], P = .033). The direct effect of group on change in pain interference, after accounting for coping as a mediator, was not statistically significant (path c’: β = −0.21, SE = 0.14, 95% CI = [−0.48, 0.07], P = .139). The indirect effect of group on change in pain interference through change in coping was statistically significant (path a × b: β = −0.10, SE_Boot = 0.05, 95% CI_Boot = [−0.21, −0.02]). The significant indirect effect with a nonsignificant direct effect indicates that change in coping fully mediated the association between group and change in pain interference (ie, the intervention may have helped patients develop better coping strategies, which in turn led to reductions in pain interference at different time points; Figure 2).

Figure 2.

Mediation model testing the indirect effect of group (3RP-NF versus HEP) on change in pain interference from baseline (T1) to post-intervention (T2) via T1–T2 change in coping. Group was coded as HEP = 0 and 3RP-NF = 1. 3RP-NF = Relaxation Response Resiliency Program for NF; HEP-NF = health education control for NF; NF = neurofibromatoses.

Testing Changes in Social Support as a Mediator in the Association Between Intervention Group and Pain Outcomes at Each Time Point.—

We examined whether improvements in social support explained the effect of the intervention on pain outcomes over time. In the model including social support as a mediator, group (3RP-NF) was not associated with change in social support (path a: β = 0.15, SE = 0.14, 95% CI = [−0.14, 0.43], P = .314). Increased social support was associated with decreased pain interference (path b: β = −0.18, SE = 0.07, 95% CI = [−0.32, −0.04], P = .010). The total effect of group on change in pain interference, factoring out the social support mediator, was statistically significant (path c: β = −0.28, SE = 0.14, 95% CI = [−0.57, −0.0001], P = .049). The direct effect of group on change in pain interference, accounting for the meditator, was not (path c’: β = 0.26, SE = 0.14, 95% CI = [−0.54, 0.02], P = .072). The indirect effect of group on change in pain interference through the social support mediation path was not statistically significant (path a × b: β = −0.03, SE_Boot = 0.03, 95% CI_Boot = [−0.11, 0.02]), suggesting that change in social support did not mediate the effect of group on change in pain interference (ie, changes in social support were not the reason that the intervention influenced pain interference over time).

Discussion

This is the first fully powered study to test the effects of a mind–body intervention (3RP-NF) on pain intensity and interference across a long-term follow-up, as well as the largest psychosocial efficacy trial in adults with NF to date. Consistent with our hypotheses, we found that individuals randomized to 3RP-NF experienced statistically significant reductions in pain interference from baseline to post-intervention, and these reductions were sustained through 6- and 12-month follow-ups. That is, individuals assigned to receive a mind–body intervention showed significant decreases in the extent to which pain interfered with their daily life, and these improvements were maintained throughout the follow-up period. Contrary to our hypotheses, neither group (3RP-NF vs. HEP-NF) nor the group-by-time interaction predicted change in pain interference (ie, the amount of change over time was overall similar for both groups). While the group-by-time interaction was not statistically significant across all time points, analyses probing the interaction at specific time points revealed a statistically significant decrease in pain interference from baseline to post-intervention (ie, patients in the 3RP-NF group demonstrated improvement in pain interference), which was sustained (ie, maintained with no subsequent change) from post-intervention through 6- and 12-month follow-ups among the 3RP-NF group. The HEP-NF group had no significant changes in pain intensity or interference over time. Clinically, this suggests that integrating mind–body techniques may provide long-term benefits for managing pain interference among patients with NF.

Although pain interference decreased among individuals in the 3RP-NF group, pain intensity remained unchanged from baseline through follow-up. Mind–body interventions tend to focus on improving coping, promoting self-efficacy, and engaging in valued activities alongside pain, thereby reducing the extent to which pain interferes with daily functioning (rather than reducing pain itself). Indeed, positive psychological components (eg, guided imagery, positive storytelling, humor as a coping skill) and cognitive–behavioral concepts (eg, appraisal of stress and medical symptoms) may have been more effective in changing the way participants perceived the impact of their pain and how they responded to it, while the pain itself (ie, pain intensity) remained stable. In chronic illnesses like NF, pain may be unavoidable, but mind–body interventions may modulate patients’ emotional experience of pain, which can make it less disruptive. Still, the 3RP-NF program could potentially be enhanced by including mind–body skills that more directly address pain intensity, such as pain-specific relaxation techniques.⁴⁰

Change in coping was the only factor that fully mediated the reduction in pain interference observed for the 3RP-NF group. This aligns with research emphasizing the importance of coping-focused treatments in managing chronic pain.⁴¹ Notably, the coping skills taught in 3RP-NF were not pain-specific, but they still facilitated change in pain interference. To boost effectiveness, clinicians should adopt pain-focused coping skills to help reduce the negative emotional components of pain, enhance pain self-efficacy, and painless disruptive to daily life. Future iterations of 3RP-NF could benefit from including pain-specific coping skills to bolster pain-related coping and decrease pain interference, perhaps targeted at coping with types of pain experienced most frequently by individuals with NF (eg, headaches, gastrointestinal pain).⁴ Understanding whether coping skills are more impactful for patients with greater intensity may also be valuable. The present study was not powered for such an analysis, and baseline pain intensity (low vs. medium vs. high) was evenly distributed between intervention and control groups, though this remains an important area for future inquiry. Interestingly, social support did not mediate change in pain interference for the 3RP-NF group. Prior work has highlighted social support as a resiliency factor associated with better adjustment to chronic pain by promoting adaptive coping skills.^41,42 One possible explanation for this is that the measure used to assess social support in the present study does not assess social support skills use, but rather focuses on the presence of a social support network influenced by factors outside of skills use.

Although receiving the 3RP-NF intervention was not associated with change in mindfulness compared to the control condition, increased mindfulness was associated with decreased pain interference from baseline to post-intervention. Our findings related to mindfulness are generally consistent with prior work in chronic pain.^41,43,44 Previous work comparing the performance of mindfulness-based stress reduction and a pain education control for chronic pain similarly found that mindfulness changed equivalently across both conditions and was equally associated with functional outcomes across the 2 groups.⁴³ It has been suggested that mindfulness may have a greater effect on pain-related psychosocial factors, such as pain catastrophizing (eg, a negative cognitive and emotional response to pain that involves a tendency to ruminate about pain, a propensity to magnify the threat value of pain, and feelings of helplessness), than pain interference or pain intensity.⁴¹ Indeed, prior work indicates that mindfulness-based interventions may exert a larger effect on pain catastrophizing in the context of high pain intensity.^41,44,45 Although participants in the present sample reported experiencing moderate levels of pain intensity, the intervention may exert a stronger influence on mindfulness outcomes among patient populations with consistently higher levels of pain intensity. The impact of 3RP-NF on factors such as pain catastrophizing or pain-related anxiety should be explored in future work. Lastly, given the inclusion of individuals across the NF diagnostic spectrum within the present study, future work may benefit from exploring more granular pain-related outcomes that may reveal differences resulting from the potential heterogeneity of the sample.

The present study has several important strengths. This includes a rigorous methodological design, multidimensional assessment of pain and resiliency factors, the use of live video to increase study access, low attrition rate, and a 12-month follow-up period. Additionally, 3RP-NF is tailored to address the needs of an underserved, high-need clinical population, and the present study is the first fully powered examination of pain characteristics and psychosocial outcomes across the NF diagnostic spectrum. This study also has some limitations. First, although we recruited individuals with NF internationally, the sample was primarily comprised of White, highly educated women. Individuals interested in receiving psychosocial support may have been overrepresented in our sample, and future trials should employ targeted recruitment methods (eg, community-based methods or peer referrals) to increase diversity. Additionally, given that the larger RCT was not focused on pain outcomes, we did not collect data regarding alternative pain treatments patients received during the course of the study (eg, medications). Future iterations of the 3RP-NF program should include a more thorough evaluation of pain treatments received during the course of the study. Despite the virtual format of our study providing geographic diversity, excluding individuals who were unable to join via live video may have precluded individuals without adequate technology or internet access from participating in the study. Recent adaptations of 3RP-NF have made it accessible as a web-based platform (NF-Web)⁴⁶ which may increase accessibility further. Given that the present study recruited participants using an email listserv, participants in the present study may have higher technological literacy than other populations. Future iterations of the 3RP-NF program should focus on in-person recruitment and delivery of the 3RP-NF program to better target populations that may have difficulty accessing virtually delivered intervention studies.

Our results indicate that the iteratively developed mind–body 3RP-NF group program yields durable reductions in pain interference. Although pain is highly prevalent among individuals with NF,¹² our findings suggest that pain interference is malleable through psychosocial intervention. Our highly tailored 3RP-NF program may be an effective nonpharmacological approach for managing NF-related pain. Coping skills may represent a promising mechanism by which mind–body interventions reduce pain interference in this population.

Funding

This work was supported by the U.S. Department of Defense Grant # W81XWH-17-1-0121 awarded to A.M.V.

Conflict of interest statement

None declared.