Getting Active Mindfully: Rationale and Case Illustration of a Group Mind-body and Activity Program for Chronic Pain

Abstract

Chronic pain is associated with substantial decreases in physical and emotional health. Psychosocial and physical restoration interventions, although potentially helpful, typically show small-to-moderate improvements that are limited to the short term, and often exhibit problematic adherence. Here, we present GetActive-Fitbit, a novel 10-week group program that integrates mind-body skills, pain coping and gradual increases in activity reinforced by a commercially available digital monitoring device (Fitbit). We illustrate the program among a group of 4 adults with heterogeneous chronic pain. We also highlight pre to post-program improvements in physical function (objective, performance-based and self-report), emotional function (depression and anxiety) and other relevant outcomes targeted by the program (e.g., pain intensity, catastrophizing, mindfulness, coping, kinesiophobia, emotional support, social isolation, pain resilience, program satisfaction and impression of change). Group participants’ experiences suggest that GetActive-Fitbit is credible, useful, and shows potential to improve physical and emotional function among this challenging population.

Clinical trial number: NCT03412916.

Introduction

Chronic pain is highly prevalent (Turk and Okifuji 2001), costly (Schappert 2006) and difficult to treat (Weisberg and Clavel 1999). In recent decades, the conceptualization of chronic pain and its treatment has consistently broadened from a strictly biological focus to one that stresses cognitive, affective and social factors. Treatments for chronic pain, thus, increasingly incorporate psychosocial elements. Cognitive behavioral therapy for chronic pain typically emphasizes psychoeducation about chronic pain and coping skills training (Keefe 1996; Ehde et al. 2014), while mindfulness and acceptance-based practices typically emphasize pain acceptance and engagement in activities that are value driven, even if these are painful (Mccracken and Vowles 2014). Such psychosocial interventions improve outcomes, though effect sizes are generally small and fade over time (Andersson et al. 2012; Anheyer et al. 2017; Hilton et al. 2017; Jackson et al. 2019). A similar approach to increasing activity, regardless of pain, has been historically employed by functional restoration programs, which, although efficacious (Hurwitz et al. 2005; Kroll, 2015; Geneen et al. 2017), pose problems with engagement and adherence (Peek et al. 2016; Meade et al. 2019) due to anxiety about pain, fear of causing damage (Crombez et al. 2012; Vranceanu et al. 2014a, b), lack of social support and the absence of specific goal setting (Meade et al. 2019).

A multimodal program focused on: (1) increasing activity through quota-based pacing (e.g., increasing activity gradually based on an objectively measured criterion not contingent on pain), linked to activities valued by the patient; (2) targeting barriers to engagement and adherence; and (3) objectively monitoring and reinforcing increases in activity with a digital monitoring device, may be the most effective and efficient way to improve both emotional and physical function outcomes. Tracking and increasing activity may be particularly important for patients with chronic pain given that monitoring is an important component of behavioral change (Compernolle et al. 2019) and that increasing activity can reduce disability and healthcare costs as well as improve quality of life and productivity in this population (Pincus et al. 2002; Jackson et al. 2019). Over the past decade, commercially available Bluetooth 4.0 digital monitoring devices such as Fitbit (Fitbit, Inc., San Francisco, CA) provide a user-friendly strategy for patients to track their activity and receive real time feedback. The recent use of these devices in populations with chronic illness has shown that they are feasible and useful in tracking physical activity (Phan and Mobbs 2016; Block et al. 2017; Chandrasekar et al. 2018; Maijers et al. 2018; Richeson and Croteau 2018; Haglin et al. 2019).

In patients with chronic pain, Fitbit was found to yield high adherence in some (Feehan et al. 2014; Kulich et al. 2015) but not all (Amorim et al. 2019; Jamison et al. 2017) studies, particularly when worn on the wrist rather than being clipped-on (Jamison et al. 2017). However, two recent studies testing physical activity and healthy lifestyle programs with the aid of a Fitbit failed to find significant increases in step-count in the groups receiving the Fitbit, or significant step-count differences compared to the control groups (i.e., standard care with information on physical activity (Amorim et al. 2019), or use of a pedometer instead of a Fitbit (Gordon and Bloxham, 2017)). A more holistic and integrative approach to increasing activity with the use of a Fitbit, while teaching specific skills that may aid in bypassing barriers to engagement in physical activity and coping with the challenges associated with chronic pain, may benefit this population.

Our team has developed the first multimodal, mind body and activity program aided by a Fitbit, for adults with heterogeneous chronic pain (GetActive-Fitbit; (Greenberg et al. 2019, 2020)) aimed at improving emotional and physical function. The program includes several novel elements, including specific tailoring of the program to the needs of patients with chronic pain via qualitative interviews, the combination of mind-body and pain management skills with systematic increases in physical activity, and being the first pain trial to implement recent guidelines from the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT; (Taylor et al. 2016; Gewandter et al. 2019)) and the International Classification of Functioning, Disability and Health (ICF; (World Health Organization 2001)) regarding comprehensive assessment of physical function using objective, performance-based and self-report measures. We used the Relaxation Response Resiliency Program (Park et al. 2013), a well-established evidence-based (Vranceanu et al. 2014a, b; Gonzalez et al. 2016; Gonzalez et al. 2018) mind-body program as the starting point. Based on qualitative feedback, we tailored this program to address issues pertinent to patients with chronic pain, as well as skills focused on quota-based pacing linked to activities valued by the patient and reinforced by a Fitbit. In an open pilot study, we showed that the 8-week program (GetActive-Fitbit; (Greenberg et al. 2019)) was feasible, acceptable, and showed signals of improvement in physical activity (self-report and number of steps), emotional and physical (self-report and 6 min walk test) function and coping measures. Following participants’ qualitative feedback from exit interviews, we further revised and refined the intervention and lengthened it to 10-weeks. An RCT of the refined program further supported its feasibility and promise in improving physical and emotional outcomes (Greenberg et al. 2020). Here, we detail the program structure and illustrate intervention delivery and outcomes for a group of 4 adults with heterogeneous chronic pain. We showcase delivery of skills, setting up gradual increases in step-goals, adherence to home practice, Fitbit wear and activity goals, feasibility markers and improvement in targeted outcomes (e.g., physical and emotional function, pain and related coping).

Methods

Participants

We report on four patients with heterogeneous chronic pain who enrolled in the GetActive-Fitbit program. Eva was a woman in her fifties with chronic migraines and face-pain for the past decade. Rachel was a woman in her thirties with chronic back and knee pain after an accident several years prior. Daria was a woman in her fifties with fibromyalgia for the past decade. Isabella was woman in her forties with chronic upper back, neck and hand pain for the past 6 months. Participants were referred from the Pain Clinic at Massachusetts General Hospital and were medically cleared for participation. They did not have untreated severe psychiatric illnesses, current substance use or active suicidality and denied any medical illness expected to get worse in the next 6 months. They denied engagement in yoga or mindfulness practice in the past 3 months. They scored in the sedentary category on the International Physical Activity Questionnaire Short Form (IPAQ-SF; (Craig et al. 2003)). They did not use a Fitbit in the past 3 months and were willing to use one for the duration of the program. The study was approved by the Institutional Review Board at Massachusetts General Hospital and registered as a clinicaltrials.gov study under NCT03412916.

Materials

Physical function

Physical function was comprehensively assessed before and after the program via objective, performance based and patient-reported measures, per IMMPACT (Taylor et al. 2016; Gewandter et al. 2019) and ICF (World Health Organization 2001) guidelines. Objective physical function was assessed via a GT3X ActiGraph accelerometer (Cain et al. 2013) fastened via an elastic band over the right hip for 7 days prior to and after the program to assess their average number of daily steps. We used a minimum clinically important difference (MCID) (Copay et al. 2007) of 800 steps for ActiGraph (Motl et al. 2013). MCID refers to the smallest difference in score in a domain of interest, which is perceived by patients as beneficial and would mandate a change in patients’ management in the absence of excessive side effects or costs (Jaeschke et al. 1989). MCIDs are often assessed via comparison of an outcome to an external measure, although multiple methods are available to calculate an MCID (Rai et al. 2015). Performance-based physical function was assessed via the 6-min walk test ((Redelmeier et al. 1997); MCID = 54 meters (Redelmeier et al. 1997)), which measures the distance in meters participants walk over 6 min. Self-reported physical function was assessed via 3 different questionnaires to be able to compare their respective sensitivity in preparation for a future RCT; the PROMIS Physical Function, v.1.2.8b ((Stone et al. 2016); MCID = 5.48 (Selvarajah et al. 2016)), which assessed the ability to perform physical tasks, ranging from daily self-care activities to complex tasks (higher scores reflect higher levels of function); the World Health Organization Disability Assessment Schedule (WHODAS) 2.0 (Garin et al. 2010) which assesses functional difficulties in life domains such as getting around, self-care, and social activities (higher scored indicate greater disability); and the self-reported Physical Activity Scale for Individuals with Physical Disabilities (PASIPD) (Washburn et al. 2002), which assessed participants’ engagement in leisure, household and work-related activities (higher scores indicate more activity).

Emotional Function

To measure depression and anxiety, we used the PROMIS depression (v1.08b; 2015) and anxiety (v1.08a; 2015) scales (MCIDs = 5.19 and 4.28, respectively (Selvarajah et al. 2016)). Both are 8-item questionnaires assessing the frequency of symptoms over the past week (higher scores indicate higher anxiety and depression, respectively).

Other Patient-Reported Assessments

We measured pain at rest and during activity using the 0–10 Numerical Rating Scale (NRS; (Farrar et al. 2001; Rodriguez 2001); MCID = 1 (Salaffi et al. 2004)), mindfulness with the Cognitive and Affective Mindfulness Scale-Revised (CAMS-R; (Feldman et al. 2007)), catastrophic thinking about pain with the Pain Catastrophizing Scale (PCS; (Sullivan et al. 1995)), fear of pain with the Tampa Kinesiophobia Scale (TKS; (Woby et al. 2005) MCID = 6 (Monticone et al. 2017)), general healthy coping skills (e.g., relaxation, use of social support and adaptative thinking) with the Measures of Current Status-A (MOCS; (Carver 2006)), social isolation with the PROMIS Social Isolation Short Form (4a; 2015), emotional support with the PROMIS Emotional Support (4a; 2015), pain resilience with the Pain Resilience Scale (PRS; (Slepian et al. 2016)) and program satisfaction with the Client Satisfaction Questionnaire (CSQ-3; (Attkisson and Zwick 1982)). We used the Modified Patient Global Impression of Change (MPGIC; (Geisser et al. 2010)) to assess perceived improvement in levels of pain, physical activity, physical function, emotional function, pain resiliency and the degree to which the Fitbit helped increase their activity. Higher scores on all measures indicate higher levels of the described construct.

Procedure

Participants traveled to the clinic for the baseline assessment (~2 h), which included a detailed description of the study and obtainment of written informed consent. Participants completed study questionnaires with research staff present to answer any questions and check questionnaires for completion. A research assistant performed individual walk tests using a predetermined protocol (American Thoractic Society 2002). At the end of the assessment session, participants received a GT3X ActiGraph to wear daily for 1-week, detailed wear instructions indicating the ActiGraph should be worn in all waking hours except when in water (e.g., bathing or swimming), and a wear time log to specify daily on and off times as well as activities they engaged in. Participants returned the ActiGraphs and the logs a week later and participated in Session 1 of the group program. The research assistant downloaded the ActiGraph data as the clinician led the first group session, prepared individual Fitbit accounts and step goals for all participants using predetermined criteria (see Determining weekly step goals below), and paired each participant’s phone with the Fitbit. During each of the subsequent group visits, the research assistant manually programed steps goals directly into each participant’s Fitbit using the same criteria. All testing procedures were repeated a week after the final session.

Development and Adaptation of GetActive-Fitbit

We followed the NIH stage model for behavioral intervention development (National Center for Complementary and Integrative Health 2017) and ORBIT model for developing behavioral treatments for chronic diseases (Czajkowski et al. 2015) to iteratively develop the GetActive-Fitbit program with the goal of improving physical and emotional function (depression and anxiety). The program was adapted from the Relaxation Response Resiliency Program (3RP), an 8-week program for coping with chronic stress for medical patients (Park et al. 2013; Vranceanu et al. 2014a, b; Vranceanu et al. 2016, 2013; Gonzalez et al. 2018; Zale et al. 2018). We added specific pain and activity-related skills (see Skills and Content below), partially based on available cognitive-behavioral and mind-body interventions for pain (Keefe 1996; Mccracken and Vowles 2014). The initial iteration of the GetActive-Fitbit program (8 weekly group sessions) was informed by focus groups with 22 adults with chronic pain. Exit interviews from an open pilot (Greenberg et al. 2019) informed further adaptations (e.g., increasing program length to 10 weeks to better absorb materials, omitting information on health behaviors such as sleep and nutrition, which some participants indicated was distracting), leading to the final version of the program which we now describe (see also Greenberg et al. 2020).

Skills and Content

The GetActive-Fitbit program retains the core skills of the 3RP, which are now adapted to directly address management of chronic pain: (1) relaxation response (RR) skills such as deep breathing, body scan, mindful awareness exercises and MINIs (brief versions of the long exercises); and (2) coping skills such as adaptive thinking and problem solving; and (3) positive-psychology-based skills like gratitude, laughter and acceptance. GetActive-Fitbit also includes new skills and exercises borrowed from cognitive behavioral therapy and adapted for pain and increased physical activity such as: (1) debunking “myths” about chronic pain; (2) understanding the Disability Spiral and the Pain Cycle; (3) identifying own “false pain alarm”, designed to help participants distinguish harmful pain (e.g., acute pain, new injury) from non-harmful pain (e.g., pain that is part of chronic pain); (4) engaging in quota based activity pacing (i.e., non-pain-contingent); (5) pairing increased activity with valued-based and pleasurable activities of daily living; and (6) using SMART goal setting for increased number of steps, paired with activities of daily living. All sessions were audio-recorded. Program structure is detailed in Table 1. The program was led by a clinical psychologist with 16 years of experience in mindfulness and meditation-based approaches.

Table 1.

Program structure

	Skills	Rationale
Session 1	1. Debunking Pain Myths 2. The Disability Spiral 3. Deep breathing 4. Fitbit use 5. Gratitude and appreciations	1. Challenging beliefs such as “hurt always means harm” and “pain means you need to stop” 2. Understanding how decreased activity leads to deconditioning, negative emotions, avoidance and increases in pain and disability 3. Experientially introduce the relaxation response 4. Outline the use of Fitbit to track activity, outline plan to wear, charge and sync it. 5. Develop the capacity to be more attuned to positive experiences and cultivate social connectedness
Session 2	1. Quota-based activity pacing 2. Choosing meaningful activities 3. SMART goals 4. Activity barriers	1. Prevent doing “too much too soon”, gradually increase activity non-contingent on pain levels 2. Increase motivation and bypass challenges associated with engagement in physical activity 3. Optimize goal setting by identifying goals that are Specific, Measurable, Attainable, Relevant and Time-based 4. Targeting factors that can get in the way of increasing activity
Session 3	1. Stress warning signals 2. Single pointed focus on the breath 3. Body scan 4. MINIs	1. Facilitate early recognition of the activation of the stress response 2. Down-regulation of the stress response via activation of the relaxation response 3. Identifying and releasing stress and tension in the body 4. Learn short versions of the relaxation practices that can be applied throughout the day
Session 4	1. Mindful awareness 2. Pain meditation 3. Walking meditation for pain	1. Increase awareness and reduce reactivity 2. Decrease reactivity to pain, develop a more accurate perception of pain and facilitate tolerance and acceptance 3. Generalize the quality of mindful awareness to various activities
Session 5	1. Identifying types of social support 2. Identifying social reactions to pain 3. The pain cycle	1. Evaluate available support system, identify social needs 2. Promote understanding of the interpersonal manifestations of chronic pain 3. Develop an integrative view of how pain and disability are maintained via thoughts, emotions, avoidance and social isolation
Session 6	1. Identifying pleasant activities 2. Identifying negative automatic thoughts (NATs)	1. Encourage motivation and activity, boosting mood 2. Recognize negative thoughts as they arise and identify “thoughts distortions” (e.g., black and white thinking, catastrophizing, ‘should statements’, etc.)
Session 7	1. Guided imagery 2. Adaptive thinking 3. Stop, breathe, reflect, choose	1. Decrease anxiety, instill a sense of calmness and confidence 2. Introduce strategies to create adaptive perspectives such as reframing NATs, using positive emotions to generate adaptive perspectives and acceptance 3. Managing stress by taking a step back, initiating the relaxation response, reflecting on any thought distortions, and consciously choosing an appropriate response rather than reacting automatically
Session 8	1. Loving kindness meditation 2. Optimistic storytelling 3. Identifying relaxation signals 4. Getting back on track	1. Cultivate compassion, empathy and social connectedness 2. Override the pessimistic perspective typically accompanying chronic pain 3. Cultivate awareness of the relaxation-response as participants gradually gain mastery over it 4. Anticipate future life circumstances that may get in the way of maintaining levels of physical activity and RR practice and cultivate a plan to get back to these activities
Session 9	1. Problem solving and acceptance 2. Mindfulness of thoughts 3. Empathy 4. Contemplation	1. Identifying appropriate adaptive coping strategies based on one’s level of control over the situation 2. Develop an accepting and non-judgmental attitude towards thoughts 3. Cultivate mindful listening and improve one’s relationship with themselves and others 4. Elicit the RR by identifying sources of inspiration, appreciation and compassion, and focusing the mind on positive intent
Session 10	1. Humor and laughter 2. Staying resilient	1. Promoting resiliency, positive mood and adaptive coping via identifying strategies to facilitate laughter 2. Review of all skills, identify those which were most helpful and cultivate a plan to continue their implementation

Determining Weekly Step Goals

We followed a protocol designed to allow for flexible tailoring of activity to each individual participant. In session 1, participants were given a Fitbit programmed with a step goal determined by the average number of steps measured by the ActiGraph during the baseline week. For all subsequent sessions, participants who met their step goal had the option to increase their goal by 10% or stay at the same goal. Participants who did not meet their weekly step goal repeated their goal. Participants who did not meet their step goal for two consecutive weeks, chose whether to repeat their step goal or set a new goal which was 10% higher than their average step-count. Participants’ weekly Fitbit average step-count was calculated in 5-day blocks, excluding the days of highest and lowest step-counts (Greenberg et al. 2019). Participants’ Fitbit wear-time was inferred based on heart-rate data availability. Studies vary in wear-time requirements between 2 and 24 h/day (McGrath et al. 2017). We chose a relatively low minimum wear-time of 7-h/day for both Fitbit and ActiGraph, given the prevalence of sleep disturbances and increased time spent in bed among patients with chronic pain (Morin et al. 1998; Spitzer and Broadman 2010). Fitabase, an online research platform enabling the collection and analysis of Fitbit data while maintaining participant confidentiality (Franzen-Castle et al. 2017), was used to download Fitbit data for all participants. A staff member calculated participants’ upcoming step goals prior to the session, programmed this number directly into each individual’s Fitbit through their own account on the Fitbit website. The staff member then provided the study clinician with the data. At the end of the study, participants retained the Fitbit and gained control over their own Fitbit account.

Results

Participants attended an average of 8.75/10 sessions and handed in an average of 5.75/9 homework logs. Participants wore the ActiGraph for an average of 4.25 days at baseline and 5 days at the post-program assessment. All participants wore their Fitbit at least 6 days a week for the entirety of the program. As a group, throughout the program, participants exhibited increases in their number of daily steps on ActiGraph (+877 steps; above the MCID (800 steps)), and Fitbit (+~700 steps; Table 2). They also improved distance walked in 6-min (+58.25 meters; above MCID (54 meters)) as well as self-reported physical activity (+4.67 points), physical function (+2.92 on physical function measured via PROMIS; - 13.02 on disability measured via WHODAS) and emotional function (-8.52 on PROMIS depression; above MCID; -4.07 on PROMIS anxiety; slightly below MCID; Table 3; see Table 4 for other self-reported outcomes). All four participants rated their impression of change in physical activity, physical function, resiliency and from use of Fitbit as “much” or “very much” improved (Table 5). Impression of change in emotional function was rated as “much improved” or “minimally improved” (two participants each). Participants were highly satisfied with the program (average 11/12; Table 5).

Table 2.

Quantitative outcomes of physical activity

ID	Physical Function (PROMIS)			Disability (WHODAS)			Physical Activity (PASIPD)			Actigraph steps			6-min Walk Test (m)			Fitbit Steps
	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ
Isabella	42.4	48.9	+6.5	35.42	14.71	−20.71*	8.1	12.64	+4.54	4853.16	8516.6	+3663.44*	412	478	+66*	7642.2	7453.67	−188.53
Daria	36.0	38.6	+2.6	42.19	28.91	−13.28	16.5	20.18	+3.68	7786	5846.2	−1939.8	386	399	+ 13	8924.25	7863	−1061.25
Rachel	43.8	46.2	+2.4	6.94	5.55	−1.39	8.3	18.28	+9.98	5269	5635.2	+366.2	265	366	+ 101*	6452.2	7885.8	1433.6*
Eva	29.4	29.6	+.2	59.68	43.75	−15.93*	0.95	1.45	+0.5	4266.5	5686	+ 1419.5*	366	419	+53*	3775	6386.33	2611.33*
Mean change			2.92			−13.02			4.67			877.33*			58.25*			698.78*

For measures that have established MCIDs, changes above the MCID are indicated with a*.

Table 3.

Quantitative self-report outcomes of pain and emotional function

ID	Depression (PROMIS)			Anxiety (PROMIS)			Pain at rest			Pain with activity			Pain catastrophizing (PCS)			Pain resilience (PRS)
	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ
Isabella	70.3	61.7	−8.6*	66.8	62.5	−4.3	7	6	−1*	7	4	−3*	38	8	−30*	15	25	+10*
Daria	61.7	61.7	0	57.6	62.5	+4.9	6	4	−2*	9	7	−2*	36	29	−7*	29	33	+4*
Rachel	48.8	37.1	−11.7*	50.6	43.2	−7.4*	4	0	−4*	7	3	−4*	9	6	−3	42	46	+4
Eva	68.4	54.6	−13.8*	65.7	56.2	−9.5*	8	7	−1*	10	7	−3*	37	21	−16*	44	40	−4
Mean change			−8.52*			−4.07			−2			−3			−13.75			3.5

Note: For measures that have established MCIDs, changes above the MCID are indicated with a*

Table 4.

Self-report outcomes of coping, and physical and social functioning

ID	Mindfulness (CAMS-R)			Kinesiophobia (TKS)			Coping (MOCS-A)			Emotional Support (PROMIS)			Social Isolation (PROMIS)
	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ	T1	T2	Δ
Isabella	19	28	+9*	34	25	−9	11	24	+13*	56.2	56.2	0	56.2	53.9	−2.3
Daria	28	26	−2	40	40	0	15	32	+17*	40.5	42.1	+1.6	56.2	56.2	0
Rachel	39	38	−1	36	26	−10*	37	34	−3	56.2	62	+5.8	34.8	34.8	0
Eva	29	28	−1	53	49	−4	42	43	+1	42.1	62	+19.9*	62.9	49.9	−13*
Mean change			1.25			−5.75			7		6.82				−3.82

For measures that have established MCIDs, changes above the MCID are indicated with a *.

Table 5.

Program satisfaction (range 1–12) and impression of change (range 1 (“very much improved”)-7(very much worse”); post-test only).

ID	Satisfaction with program (CSQ-3)	Impression of change in pain	Impression of change in physical activity	Impression of change in physical function	Impression of change in emotional function	Impression of change in resiliency	Impression of change from use of Fitbit
Isabella	12	2	1	2	3	2	1
daria	11	2	2	2	3	2	1
Rachel	12	2	2	4	2	1	1
Eva	9	2	2	2	2	2	2
Mean	11	2	1.75	2.5	2.5	1.75	1.33

Below, we detail the process and participants’ experiences of the program, as well as the change in outcome measures for each participant.

Eva

Eva started the program with high levels of pain catastrophizing (“I swear the pain is never going to go away, I’m not going to get better, it’s definitely a brain tumor!”) and struggled to perform even basic functional tasks (“with this chronic illness you feel so defeated, even waking up is a problem, and I think that I can’t even do the dishes so my life is worthless, I am worthless, and I can’t do it”). She described a tendency to “wallow in self-pity” and feel like she is “in the abyss.” She was also skeptical about the potential benefits of the relaxation skills (“if you are in pain and someone tells you to breathe you just want to hit them in the face with an axe!”). As the program progressed, she was able to utilize relaxation skills daily and reported significantly benefitting from them (“the deep breathing definitely helps my pain more than the medication sometimes”). Despite initially being ambivalent about the Fitbit, she increasingly reported relying on her Fitbit to engage in physical activity (“I walk around the house with my phone, I track when I hit 250 hourly steps, saying ‘I can do it! I can do it!’”) Eva also reported changes in her relationships throughout the program (“Just a few weeks ago I was isolated, I didn’t want to call my friends. I don’t know if it’s the breathing or gratitude, but now I’m reaching out.”). By the end of the program, Eva indicated that for the first time in over two years, she has gone an entire week without taking any of her migraine medications, and had an increase sense of control “I now know the things I have to do, whereas prior to this I basically just buried myself in sorrow.”)

Eva’s reported experiences throughout the program were largely supported by the quantitative outcome measures (Table 2). She exhibited increases above the MCID for both ActiGraph (over 1400 steps) and Fitbit (over 2500 steps). Her improvement in the 6-min walk test were just at the point of the MCID. She exhibited clinically meaningful improvements in pain catastrophizing (reduced from a score of 37 (88th percentile) to 21 (53rd percentile; (Sullivan et al. 1995))), physical function as measured by WHODAS (reduced close to one standard deviation in scores (Garin et al. 2010)), as well as PROMIS measures of emotional support (increased 2 SD), social isolation (reduced over 1 SD), and reductions in depression, anxiety and pain during rest and activity (above the MCID). She also self-reported small improvements in kinesiophobia and coping. There were little or no improvements on physical function as measured by PROMIS (+.2) and mindfulness (−1).

Isabella

Isabella’s primary focus during the program was on getting back to physical activities such as biking and yoga, which she used to engage in before the onset of her chronic pain, but has been wary of trying again due to fear of exacerbating her pain. She indicated that giving up these activities has taken a significant toll on her mood and sense of self-worth, and that she frequently worries about her future in light of her pain (“if this is my pain now, what is going to happen 20 years from now?”) During the course of the program, Isabella started biking again and ended up biking to work daily. She also resumed attending yoga classes, frequently walking rather than taking the bus and choosing to take the staircase rather than an elevator at her workplace. She reported occasionally using the relaxation MINIs, although struggled to significantly incorporate longer meditation practices into her daily routine, which she described as hectic, due family and work-related responsibilities. By the end of the program, Isabella sharply increased her ActiGraph step-count (> 3600 step increase; above MCID) and exhibited increases in the 6-min walk test (+ 66 meters; above MCID), although her Fitbit step-count slightly decreased. She exhibited reductions in anxiety (− 4.3 points; just below MCID), depression (− 8.6; above MCID), pain at rest and with activity (− 1 and − 3 respectively; above MCID) and substantial improvements in disability (WHODAS; reduced over one SD (Garin et al. 2010)), pain catastrophizing (reduced from a score of 38 (90th percentile) to 8 (19th percentile) (Sullivan et al. 1995)), pain resilience (close to an increase of 1 SD (Slepian et al. 2016)), mindfulness (+ 9 points) and coping (+ 13 points). She exhibited some improvement in the PROMIS measure physical function (+ 6.5 points), little improvement in the PROMIS measure of social isolation (− 2.3 points) and no improvement in the PROMIS measure of emotional support, in which she scored above the population standardized mean (M = 50) at baseline.

Rachel

Following an accident years ago, Rachel gave up going to the gym and other forms of physical activity, which she reported led to self-criticism, sadness and hopelessness. She started the program with several barriers to engaging in physical activity, including having to spend much of her time at home with her frequently sick child as well as feeling anxious about negative effects of becoming more active. She exhibited a tendency to catastrophize her symptoms and reported thoughts such as “the doctors are lying, people don’t survive this, I’m going to die!”. She also had a pessimistic view of her pain and the treatments offered to her, which caused her to give up on treatments before giving them a fair try (“there is no point, nothing works”). During the program, and with the support of the group, Rachel was able to find ways to increase her activity even when at home via joint activities with her child such as dancing together. Although initially Rachel reported that relaxation exercises such as the body scan triggered anxiety about her health, with continuous practice and incorporating loving kindness meditation in which particular emphasis is placed on awareness of the heart, Rachel reported learning a new and more positive association with the sensations of her heart (“I don’t usually like to feel my heart cause I get anxious”… “but now it’s like my heart is saying ‘I’m here, I help to keep you alive!’”).

Rachel exhibited increases in the 6-min walk test (+ 101 meters; above MCID) and an increase of more than 1400 steps on the Fitbit (above MCID), although her increase in ActiGraph step-count was more modest (+ 366 steps). She experienced the largest reductions in reported pain following the program (a four-point decrease in the numerical rating scale for both pain during activity and during rest; above MCID). She reported her pain during rest as 0 at post-test. She also exhibited reductions in fear of movement on the kinesiophobia scale (− 10 points; above MCID), depression (− 11.7 points; above MCID), anxiety (− 7.4; above MCID), emotional support (+ 5.8), pain catastrophizing (− 3) as well as some increases in self-reported physical activity (+ 9.98 points). Little or no improvement was noted in measures that she endorsed the least amount of symptoms at baseline compared to the other group members such as mindfulness, social isolation and physical function.

Daria

Daria was the quietest group member and didn’t often share her experiences with the group. She also indicated that her tendency to isolate herself from others is a common response to stress. Throughout the program, Daria faced several unanticipated stressful experiences which negatively affected her motivation and mood. First, one of her parents’ illness worsened, which led to quarrels with other family members around the parent’s care. She also noted increased financial strain due to expenses related to her parent’s care. Second, she needed to undergo a medical procedure, which caused her significant anxiety. Lastly, she felt her medical providers did not take her pain seriously, which led to feuds with her medical care staff. Daria made efforts to increase her physical activity, primarily by dancing to music. However, she struggled to keep up with most of her step-goals, primarily in the second half of the program (see Fig. 1), when she experienced most of the stressful events described above. Daria’s post-program step-count decreased on ActiGraph (~1940 steps) and Fitbit (~1000 steps). She exhibited little improvement in the 6-min walk test (+ 13 m), which did not reach the MCID. Her anxiety levels somewhat increased (+ 4.9 points). However, she exhibited clinically meaningful decreases in pain (2-point decrease on the pain numerical-scale at rest and during activity; above MCID) as well as improvements in pain catastrophizing (reduced from a score of 36 (87th percentile) to 29 (72nd percentile), pain resilience (+ 4 points) and coping (+ 17 points). Little or no improvement was observed in PROMIS physical function (+ 2.6 points), kinesiophobia (0 points), depression (0 points), emotional support (+ 1.6 points) or isolation (0 points).

Fig. 1.

Weekly averaged Fitbit steps over the GetActive-Fitbit program

Discussion

Here, we described the rationale and structure of GetActive-Fitbit, the first multi-modal, group-based mind-body and activity program aided by a Fitbit device and tailored specifically for patients with heterogeneous chronic pain. We also described the delivery and impact of the program in a group of four patients with long standing heterogeneous chronic pain. Attendance and adherence to ActiGraph and Fitbit was high. All participants reported high satisfaction with their participation. All participants rated their impression of change in pain, physical activity, physical function, resiliency and from the use of the Fitbit as “much” or “very much” improved. As a group, participants experienced some improvement on most outcome measures. Three out of four participants exhibited clinically meaningful improvements in performance-based physical activity (6-min walk test) and emotional function (depression and anxiety). Two participants exhibited clinically meaningful increases in objectively measured physical activity (ActiGraph). All four participants exhibited clinically meaningful reductions in pain at rest and with activity. Participants exhibited more prominent improvements in physical function measured by the WHODAS compared to the PROMIS measure of physical function, suggesting that the WHODAS, which captures difficulties in a broader range of a physical functions (e.g., understanding and communicating, getting around, self-care, getting along with people, life activities and participation in society), may better capture the effects of the intervention. While most other measures don’t have established MCIDs in this population, we also observed improvement in several other self-report measures and identified areas of improvement.

Although participants largely improved in the primary outcomes of GetActive-Fitbit - physical and emotional function as well as in pain, there was variability in improvement in the secondary outcomes (e.g., one participant improved in mindfulness, three in kinesiophobia, coping, and pain resiliency, all improved in pain catastrophizing, all to varying degrees). This may be in part due to the fact that GetActive-Fitbit program is a multimodal intervention that teaches a wide range of skills aimed at helping participants improve their physical and emotional function. The breadth of the intervention may be particularly advantageous for patients with chronic pain, due in part to the heterogeneity of needs and challenges confronted by this population, as well as the different preferences expressed during the qualitative focus groups that informed the development of GetActive-Fitbit (Greenberg et al. 2019). The range of skills taught and the program’s multimodal structure provide a platform enabling participants to choose and utilize the program components which best fit their needs.

Although heterogeneous, patients were unified by a tendency to catastrophize their pain and be hypervigilant to bodily sensations. This was evident in their description of their experiences throughout the program as well as their high pain catastrophizing scores at baseline. The fear of exacerbating pain also manifested in an initial hesitance to engage in activities and even attend sessions when faced with high levels of pain. This supports the fear avoidance model of chronic pain (Crombez et al. 2012; Wideman et al. 2013) and emphasizes the importance of teaching skills to address catastrophizing and fear avoidance in addition to gradually helping participants increase their activity. This may also highlight the potential benefit of specific focus in GetActive-Fitbit on recognizing and challenging thought distortions such as catastrophizing, distinguishing the false from the true pain-alarm and utilizing mind-body skills to elicit relaxation, coupled with gradual increases in step-count and engagement in meaningful activities. This may explain why prior Fitbit-only interventions that did not employ additional coping skills showed little or no significant improvement in outcomes in patients with chronic pain (Gordon and Bloxham 2017; Amorim et al. 2019).

Some participants articulated difficulty in finding the time and opportunity to practice the relaxation skills at home and expressed a need for a continuous “coaching” presence between sessions. The audio recordings, setting specific goals for relaxation practice and the “MINIs,” which offered opportunities for brief practices throughout the day seemed helpful in facilitating implementation of independent practice. Practicing at home emerged a factor conducive to having these skills accessible when participants need them most such as during pain flare-ups. This supports recent findings emphasizing the importance of home-practice in mind-body interventions (Parsons et al. 2017; Greenberg et al. 2018).

Participants indicated that the Fitbit device helped them stay motivated and increase step-count and physical activity. They utilized the device to monitor their progress, and, despite some ambivalence towards the hourly reminders to move, indicated they were helpful motivators to walk around. Reaching their step-goals reportedly instilled a sense of achievement, accomplishment, and progress which participants experienced as rewarding. The value of the Fitbit in increasing activity was also evident in the Modified Patient Global Impression of Change (MPGIC), where participants rated the impression of change from use of the Fitbit as higher than the impression of change in any other domain (average 6.8/7). These findings, thus, provide preliminary support for the usefulness of Fitbit to increase activity among patients with chronic pain.

There was some discrepancy between the ActiGraph and Fitbit-based step-count. This is in line with previous findings, mostly indicating a higher step-count on Fitbit, which may stem from an overestimation (Chu et al. 2017; Reid et al. 2017). The discrepancy between Fitbit and ActiGraph step-count may have been further widened by the prolonged and more continuous wear time of Fitbit compared to ActiGraph. Despite the aforementioned overestimation of steps, Fitbit offers advantages as an activity reinforcer over ActiGraph and other digital monitoring devices in the sense that it is more comfortable to wear and does not require frequent removal (e.g., during sleep). This may ultimately increase adherence and improve data quality (Chu et al. 2017).

For some participants such as Daria, engagement and motivation declined with the accumulation of stressful life events over the course of the program. Although she continued attending the sessions and meaningfully participating, she grew distracted and limited her involvement in some of the program activities (e.g., handing in only 3/9 homework logs, the lowest of all participants). This may exemplify the impact of stress and negative life events on treatment engagement and motivation (Stetson et al. 1997; Kalichman and Grebler 2010; Walders-Abramson et al. 2014). Openly addressing possible barriers and formulating a plan to “get back on track” early in the program as well as capitalizing on motivational interviewing techniques such as having participants reiterate their reasons for wanting to learn and practice these skills may aid in maintaining engagement and motivation levels (Rollnick and Miller 1995).

Implementation of GetActive-Fitbit with this group also provided an important insight regarding step-goal setting to be carried forward to future iterations of this program. Participants’ progression of weekly Fitbit step-count indicated a general increase until approximately the end of week 6, after which step-counts tended to plateau or decline. This may be due to factors such as a fade in the novelty of the program and the Fitbit device as well as the difficulty of keeping up with further increases after steps have already continuously increased for several weeks, particularly considering participants’ relative sedentariness at baseline. This may indicate that in the final few weeks of the program, it may be beneficial to emphasize sustaining the gains made rather than continuously increasing step-goals.

The current study is a case illustration and inherently bears the limitations of case reports, including a lack of ability to generalize findings or establish cause-effect relationships (Nissen and Wynn 2014). Rather, this study intends to illustrate the rationale and delivery of the GetActive-Fitbit program, without drawing conclusions about the specific efficacy or effectiveness of the program. We recently published data from a pilot randomized controlled trial further supporting the feasibility of this program as well as that of a similar program, GetActive, without the use of a Fitbit (Greenberg et al. 2020). Given these encouraging results, we intend to proceed to a trial testing multisite feasibility and fidelity of training/implementation, in preparation for a future efficacy trial of GetActive and GetActive-Fitbit compared to a control group of health education.

Conclusion

We illustrated the delivery of a novel multimodal program which combines mind-body exercises, CBT principles, positive psychology skills and targeting physical activity with the use of a Fitbit device tailored specifically for patients with chronic pain. We illustrate and showcase pre to post-program changes in study outcomes and intervention targets. The group’s experiences as well as their self-reported and objective data suggest that GetActive-Fitbit is credible, useful, and shows potential in aiding patients increase their activity, improve their emotional and physical function and promote adaptive coping with the challenges this population faces.