A tailored pain self-management intervention for patients with cirrhosis is acceptable and improves pain control

Abstract

Background:

Chronic pain is common and uniquely challenging to manage in people with cirrhosis. The purpose of this pilot study was to evaluate the feasibility and acceptability of a tailored pain self-management (PSM) intervention for people with cirrhosis and chronic pain.

Methods:

This single-arm, single-site pilot study recruited patients with a diagnosis of cirrhosis and chronic pain to a virtual, health coach–led, modular PSM intervention. After 6 weekly one-on-one sessions, patients were invited to attend 6 weekly group sessions. Outcomes were measured at baseline, 6 weeks, 12 weeks (end of intervention), and 24 weeks (maintenance). The primary outcomes were acceptability, defined by a 4/5 rating on the Treatment Acceptability Questionnaire, and feasibility, defined as the ability to retain patients in 80% of the intervention sessions. Secondary outcomes of pain and function were measured using the Pain intensity, Enjoyment of life, and General activity (PEG) scales, and a rating of percent improvement with the intervention.

Results:

Among 21 participants who started the intervention, 16 (76%) attended ≥80% of sessions. Acceptability thresholds were met at all time points. On a scale of 0–100, where 30 is considered to be a meaningful change, participants rated their improvement in pain symptoms an average of 50 ± 25. The average PEG score decreased from 6.1 at baseline to 5.3 at 24 weeks. Participants reported increases in PSM behaviors, including cognitive and stress-reduction methods, physical activity, healthy sleep behaviors, and changes in diet.

Conclusions:

This pilot identified a reduction in pain and an increase in PSM activities and demonstrated the acceptability of a health coach–led PSM intervention for adults with chronic pain and cirrhosis.

BACKGROUND

Chronic pain, defined as pain lasting over 3 months, is reported by 14% of the general adult US population. ¹ By comparison, 79%–86% of patients with cirrhosis have chronic pain, regardless of the cause of their cirrhosis, with increasing rates as cirrhosis progresses.^2,3 Several reasons may explain why patients with cirrhosis develop chronic pain ⁴ including: widespread chronic pain potentially mediated by a proinflammatory milieu, ⁵ liver-related chronic abdominal pain (from hepatic capsular stretch, fluid accumulation, or splenomegaly), and pain from other processes (eg, osteoarthritis). ⁶ Regardless of the etiology, chronic pain in cirrhosis is associated with mental health symptoms, pain-related disability, and increased healthcare utilization.^6,7 Despite the frequency and consequences of chronic pain for people with cirrhosis, this pain is undertreated, resulting in impaired quality of life.^5,6,8–10

Analgesia is challenging in the context of cirrhosis, prompting an interest in non-pharmacological approaches to pain management. Safety concerns with common analgesics (eg, high-dose acetaminophen and non-steroidal anti-inflammatory drugs) complicate pain management for patients with cirrhosis.^4,11 Likewise, among patients with cirrhosis, opioid medications are associated with hepatic fibrosis, ¹² neuroinflammation, ¹³ and hepatic encephalopathy, ¹⁴ as well as increased hospitalization,^7,15 and adverse post-transplant outcomes,^2,16 without effective pain relief. ³ Therefore, non-pharmacologic approaches may offer the only effective approach in this population. However, few evidence-based interventions have focused on chronic pain in individuals with cirrhosis.

Pain self-management (PSM) interventions, a widespread and proven evidence-based treatment, could be adapted to patients with cirrhosis. The US Department of Health and Human Services National Pain Strategy recommends tailoring PSM interventions to disproportionately impacted target populations due to their high potential for implementation and scalability. ¹⁷ PSM interventions are typically delivered in 12 weekly sessions. One appealing feature of PSM interventions is that they are highly manualized and therefore can be delivered by individuals who are not experts in behavioral treatment (eg, clinical psychologists) in any setting. PSM interventions address topics known to impact pain outcomes (eg, mood, physical activity) and use various theory-based behavioral approaches designed to improve an individual’s pain self-efficacy, such as goal setting and modeling adaptive PSM behaviors.

To address this gap, we previously used intervention mapping, an approach that combines the input of key informants (ie, patients, clinicians, and health care leaders) with theory-based behavioral approaches to adapt a PSM intervention for patients with cirrhosis. ¹⁸ LEAP, or Liver Education About Pain, ¹⁸ includes a combination of 6 one-on-one sessions with a health coach, followed by 6 optional group sessions. The intervention focuses on problem solving, skill building, and support, with specific modules that can be selected by participants to focus on their individual needs (eg, substance use, metabolic dysfunction–associated steatotic liver disease, and weight loss). This design was based on recommendations made by people with cirrhosis and their providers. ¹⁸ The purpose of this pilot study was to evaluate the feasibility, acceptability, and preliminary efficacy of a tailored PSM intervention for cirrhosis.

METHODS

Study design

This single-arm, single-site pilot study enrolled participants in a PSM intervention that was developed for people with cirrhosis. We initially designed the pilot to be a randomized controlled trial, but changed it to a single-arm study before study initiation, due to logistical challenges during the pandemic. The study was approved by the University of Pittsburgh Institutional Review Board (STUDY21080148) with a waiver to document informed consent and was conducted in accordance with the Declarations of Helsinki and Istanbul. The study was registered on ClinicalTrials.gov (Identifier: NCT05128578).

Intervention components

Using social cognitive theory and the Consolidated Framework for Implementation Research in the context of Intervention Mapping, we completed a 6-step, structured, partnered approach to develop LEAP, as outlined in a prior publication. ¹⁸ This 12-week, modular, manualized intervention was designed to be delivered by phone or HIPAA-compliant Zoom, by a certified health coach, including 6 one-on-one weekly sessions, during which participants selected from 9 potential modules, followed by 6 optional weekly group sessions with others who completed their individual sessions. Three groups were offered based on the timing of individual session completion.

Inclusion/exclusion

Inclusion criteria included (1) ≥18 years of age, (2) capability of providing informed consent, and (3) a chart-confirmed diagnosis of cirrhosis, (4) chronic pain (at least 3 mo), and (5) an average score ≥4 across the 3-item PEG.^19,20 Patients with a history of liver cancer, transplant, or who could not provide informed consent were excluded.

Recruitment

We recruited patients using fliers, mailings followed by calls, in-person clinic recruitment, and Pitt+Me, a service where potential participants can contact researchers (or opt in) for screening. For in-person visits, staff pre-screened charts to identify patients with cirrhosis and chronic pain and informed hepatologists of the list of potentially eligible patients who were scheduled for hepatology clinic visits. Hepatology clinicians and staff then referred interested patients to study staff, both in person or asynchronously. Eight participants were recruited using Pitt+Me.

Timing of data collection

The research coordinator administered online assessments to participants at the enrollment visit (baseline), 6 weeks (end of individual sessions), 12 weeks (end of intervention), and 24 weeks (3 mo post-intervention) to collect primary and secondary quantitative outcomes, outlined below.

Measures collected at baseline

The study coordinator used the electronic medical record (EMR) to identify the etiology of cirrhosis, complications of liver disease, and comorbidities, as well as to calculate the Model for End-stage Liver Disease (MELD) score, which combines common laboratory values into a score that estimates mortality in cirrhosis. ²¹ Participants self-reported demographic data, including age, sex, race, and living situation.

Primary outcomes

The primary outcomes of this pilot were intervention feasibility and acceptability, which were measured at all time points (baseline and 6, 12, and 24 wk). Based on prior published studies, the primary feasibility outcome was defined a priori as attendance at ≥80% of individual intervention sessions with the health coach (5 of 6 sessions), who reported attendance in a spreadsheet. ²² The primary acceptability outcome was measured using an adapted version of the treatment acceptability questionnaire. ²³ Acceptability was a priori defined as an average of 4/5 rating on questions of acceptability in Table 3, with the end of intervention being the primary endpoint (week 12).

TABLE 3.

Acceptability questions (n=13)

	6 wk	12 wk	3 mo post
How likely would you be to participate in PSM?	4.2±1.1	4.4±0.7	4.5±0.8
How logical is PSM to you?	3.9±1.3	4.0±1.0	3.9±0.9
How successful was the LEAP intervention in reducing your pain symptoms?	3.7±1.2	4.5±1.1	3.8±0.7
How confident would you be in recommending this treatment to a friend who experiences pain?	4.3±0.9	4.5±0.7	4.4±0.5
How enthusiastic would you be to participate in PSM at this point?	4.6±0.5	4.1±1.0	4.2±0.8

Abbreviations: LEAP, Liver Education About Pain; PSM, Pain Self-Management.

Exploratory outcomes

Exploratory outcomes were collected at baseline, 6 weeks, 12 weeks (end of intervention), and 6 months (3 mo post-intervention) unless otherwise stated. The validated PEG and Brief Pain Inventory short form (BPI) 4-item severity scale were used at all time points to assess pain and function.^24,25 The PROMIS anxiety short form, depression short form, and fatigue short form were used to assess emotional and physical function.^19,26–29 Participants reported improvement in pain using the Patient Global Impression of Change Scale. ³⁰ To assess the hypothesis that pain would be addressed by improving proximal outcomes, including self-efficacy and outcome expectations, and that this would lead to increased PSM behaviors, participants completed the Pain Self-Efficacy Questionnaire (PSEQ). ³¹ The PSEQ is a global measure of pain self-efficacy, with a summary score from 10 question scores, each with a range of 1 (not at all confident) to 6 (completely confident). Since the purpose of the intervention was to improve chronic pain and function in patients with cirrhosis, the health coach assessed participant pain severity and the impact of pain on function using the PEG at each visit.

Semi-structured interviews

Semi-structured interviews were used to understand the impacts of the intervention on participants’ experiences, behaviors, and symptoms. These were conducted at the end of the intervention (12 wk). Semi-structured interviews included questions in Supplemental Table S1, http://links.lww.com/HC9/C354. These interviews were conducted by a trained interviewer, were transcribed verbatim, and coded by 2 trained coders, using matrix coding by question using standard methods. ³²

Analysis

All participants who completed at least one session were reported upon using descriptive statistics (eg, mean and standard deviation). The primary outcomes focused on the feasibility and acceptability as defined above. Qualitative data were summarized by 2 trained qualitative team members and organized by question and theme.

RESULTS

Recruitment

Thirty patients completed consent documentation, and 24 completed the baseline survey (Figure 1). Among the 24 consented participants who completed the baseline survey, 21 completed at least one session. The 3 patients who withdrew before starting intervention activities were not included in the analyses.

FIGURE 1.

CONSORT flow diagram. Abbreviation: LEAP, Liver Education About Pain.

Baseline characteristics

The average age of 21 participants who completed at least one session was 55±13 (Table 1). Most of the participants were women (67%) and were non-Hispanic White (90%). The average baseline PEG was 6.3 ± 1.8. The characteristics of the patients who completed versus did not complete the intervention (defined as 5–6 sessions completed) are shown in Table 1. Those who completed the intervention were on average younger (56 vs. 63).

TABLE 1.

Characteristics of 21 enrolled participants with cirrhosis who started the intervention, overall and stratified by intervention completion

Characteristic	Total cohort (n=21)	Completed <5 intervention sessions (n=5)	Completed at least 5 intervention sessions (n=16)
Demographic characteristics
Age (mean, SD)	55±13	63±9	56±13
Sex, n (%)
Women	14 (67%)	3 (60%)	11 (69%)
Men	7 (33%)	2 (40%)	5 (31%)
Race and ethnicity, n (%)
Hispanic	1 (5%)	0	1 (6%)
Non-Hispanic Black	1 (5%)	1 (20%)	0
Non-Hispanic White	19 (90%)	4 (80%)	15 (94%)
Employment status, n (%)
Employed	3 (14%)	0	3 (19%)
Retired	9 (43%)	3 (60%)	7 (44%)
Unemployed	4 (19%)	1 (20%)	3 (19%)
Disabled	5 (24%)	1 (20%)	3 (19%)
Living situation, n (%)
With partner/spouse	13 (62%)	2 (40%)	10 (63%)
With other family members	3 (14%)	2 (40%)	3 (19%)
Alone	5 (24%)	1 (20%)	3 (19%)
Liver disease characteristics
Etiology of disease, n (%)
MASH	5 (24%)	1 (20%)	4 (25%)
Viral hepatitis	3 (14%)	2 (40%)	1 (6%)
Alcohol	4 (19%)	0	4 (25%)
Other/multiple	9 (43%)	2 (40%)	7 (44%)
MELD (mean, SD, range)	10±4 (range 6–23)	9±3	10±5
Decompensated cirrhosis, n (%)	8 (38%)	2 (40%)	6 (38%)
Pain and symptoms scales
PEG (Pain, Enjoyment of life, General function Scale)	6.3±1.8	6.8±0.8	6.1±2.0
B Pain Inventory: pain severity scale (0–10)	6.4±1.8	7.0±2.0	6.3±1.9
PROMIS depression (4 items, 5–20 range), raw score	9.7±3.7	9.8±3.7	9.6±3.8
Depression (PROMIS raw score ≥11, T score ≥60), n (%)	10 (48%)	2 (40%)	8 (50%)
PROMIS anxiety (4 items, 5–20 range)	10.3±3.6	9.8±2.4	10.4±4.0
Anxiety (PROMIS raw score ≥11, T score ≥60), n (%)	8 (38%)	3 (60%)	5 (31%)
PROMIS fatigue (4 items, 5–20 range)	15.0±3.5	15.0±3.1	15.1±3.7

^a Other/multiple includes: HCV/NASH, alcohol/MASH, and autoimmune hepatitis.

Note that for all symptom measures, higher scores indicate more symptoms.

Abbreviations: HCV, hepatitis C virus; MASH, metabolic dysfunction–associated steatohepatitis; MELD, Model for End-stage Liver Disease; PROMIS, Patient-Reported Outcomes Measurement Information System; SD, standard deviation.

One-on-one intervention sessions

All 21 participants completed the required introductory session about chronic pain and cirrhosis. They were then allowed to select 5 of the 8 remaining modules of interest or relevance to their situations. Of these, the most popular were stress management, sleeping better, MASH, and physical activity. Nearly half also worked with the health coach in at least one review or makeup session. Table 2 illustrates the sessions, their selection, focus, and feedback from participants.

TABLE 2.

Individual session participation and feedback ^a

Session type	Number (%) picked	Focus of the session	Participant quotes about individual sessions
Topic 1: Introduction to chronic pain and cirrhosis	21 (100)	Introduction to the relationship between brain and body and behavior in pain, goal setting, and the challenges of pain management in cirrhosis	It’s been very helpful. I think because it’s giving me a lot of different ways to look at it…and someone to help and talk to, and acknowledge your pain, but at the same time it gives you ways to control your pain emotionally and to think about your pain in other ways and keeping yourself active. [P121]
Topic 2: Physical activity and pain	12 (57)	How physical activity can reduce pain; safety of physical activity in cirrhosis	I use a lot of physical movements and exercise to help me manage my pain. I do yoga and meditation, along with some spinning and weight training. [P125]
Topic 3: Metabolic dysfunction– associated steatohepatitis (MASH), weight loss, and pain	13 (62)	Weight management, dietary approaches to pain management	I went down about 30 pounds… that was very helpful, and I still utilize the mindful eating. [P121]
Topic 4: Stress management	17 (81)	Breathing, managing stress, and coping	You know, for me, when those things aren’t working or the pain is at a higher level, I know that I need to start doing some things to help me manage the stress to help manage the pain level. [P125]
Topic 5: Sleeping better	15 (71)	How sleep influences pain, challenges of sleep management in cirrhosis	She made me aware of things like how the sleeping also plays such a big part… as long as I’m able to get a certain amount of sleep like without interruption. That makes a difference. It really does, and that helps. It kind of resets you a little bit and, in the visualization and I’ll say like deep breathing. [P120]
Topic 6: Thinking differently about pain	2 (10)	Cognitive approaches to managing pain	I’ve been trying things like meditation. The [mindfulness] when you tense up a muscle, relax it…and focusing on each point of my body, I learned it in the LEAP study, and it is helpful. Probably the most effective is actually the mindfulness, it’s more effective than the ibuprofen. [P128]
Topic 7: Talking with family, friends, and doctors	6 (29)	The role of seeking social support for coping	But for people that care about me, I haven’t shared a lot of my medical issues with. I had thought that I was…trying to be brave and strong. And during that session, I learned that when you really care about others, you’re depriving them of caring about you. So that hit me, that really hit me….You could be open …When it’s time to reach out, I need to do that. [P111]
Topic 8: Taking opioid medications	1 (5)	The dangers of opioids in the context of cirrhosis, the role of other analgesics, and pain self-management in cirrhosis	Now I’m starting to cope; I’m lessening my morphine usage, and I went to my pain doctor and I kind of said, ‘You know I think this is affecting my liver. I think this is, you know, all these things that I’m taking that I can somehow maybe…. I could try to do… without medical, without opiates’. [P115]
Topic 9: Coping without substances	5 (24)	The dangers of alcohol and drug use in the context of cirrhosis; replacing them with other coping strategies	I am prescribed Suboxone, but that’s not for pain. The one of the main reasons is that I used to have a problem with pills, and heroin … I guess a lot has changed. You know, I got a new job … I just feel better during the day because of, like cutting out the sugar, working on my sleep …definitely I’m more active now. [P118]
Catch-up or review sessions	7 (33)	Participants selected this option to consolidate skills and knowledge at the end of the 6 wk	n/a

^a Participants started with the introductory module and could select 5 options from among the other modules for their individual sessions.

Primary outcome: Feasibility

In terms of feasibility, the median number of individual sessions attended was 6 of 6, with a range of 1–6 sessions attended. In all, 76% of participants who started the intervention completed at least 5 of 6 individual sessions; 13 participants completed all follow-up surveys. There were no numerically appreciable differences in the baseline characteristics between patients who did and did not complete the surveys. In terms of optional group sessions, 11 participants opted to attend at least 1 group session, with a median number of group sessions attended of 2.5 (range of 0–6).

Primary outcome: Acceptability

At the end of the intervention, all ratings on acceptability items (Table 3) were >4/5; the average of acceptability questions at all time points was above the 4 out of 5 ratings threshold (Table 4A). Acceptability was also assessed using semi-structured interviews, as described below, and illustrated through quotes in Table 5.

TABLE 4.

Baseline and follow-up measures, among participants who started the intervention (n=21), and among those with full data (n=13)

Scales (mean ± SD)	Baseline for all participants (n=21)	Baseline for patients with complete measures (n=13)	6 wk (n=13)	12 wk (n=13)	3 mo post (n=13)
PEG (Pain, Enjoyment of Life, and General Activity Scale, 3 items rated 0–10, where a higher score indicates more symptoms)	6.3±1.8	6.1±1.5	5.5±2.3	5.8±2.1	5.3±1.7
Brief Pain Inventory severity scale (0–10, where 10 indicates the most pain)	6.4±1.8	6.3± 2.1	6.2±1.6	5.7±2.1	5.8±2.0
PROMIS depression (4 items, score range of 5–20, with a higher score indicating more symptoms)	9.7±3.7	9.8±3.7	10.8±4.6	9.6±3.3	8.9±3.6
PROMIS anxiety (4 items, score range of 5–20, with a higher score indicating more symptoms)	10.3±3.6	10.3±4.0	10.4±5.0	9.6±3.4	9.3±3.8
PROMIS fatigue (4 items, score range of 5–20, where a higher score indicates more symptoms)	15.0±3.5	15.5±3.7	15.6±4.0	15.0±3.5	14.5±3.6
Pain self-efficacy questionnaire (higher score indicates more self-efficacy)	34±16	32±17	31±11	26±13	29±11
Acceptability (mean of 5-item scale, rated 0–5, where a higher score indicates more acceptable)	n/a	n/a	4.1±0.9	4.1±1.0	4.2±0.5
Global rating of change (rating from 0 to 100, with a higher score indicating greater pain relief from the intervention)	n/a	n/a	65±15	43±31	51±26

Abbreviation: PROMIS, Patient-Reported Outcomes Measurement Information System.

TABLE 5.

Median change in scores from baseline (n=13)

Measure	6 wk (n=13)	12 wk (n=13)	3 mo post (n=13)
PEG (median, IQR)	−1.0 (−1.8, 0.1)	−0.5 (−1.1, −0.2)	−0.3 (−1.4, 0.6)
Brief Pain Inventory severity scale (median, IQR)	−0.7 (−1.8, 1.0)	−0.7 (−2.0, 1.0)	0.0 (−0.7, 0.7)
PROMIS depression (median, IQR)	0.0 (−1.5, 3.5)	0.0 (−2.0, 2.0)	−1.0 (−3.0, 1.0)
PROMIS anxiety (median, IQR)	0.0 (−2.0, 2.0)	0.0 (−2.0, 1.0)	−2.0 (−2.0, 0.0)
PROMIS fatigue (median, IQR)	0.0 (−1.5, 1.0)	−1.0 (−3.0, 0.0)	−1.0 (−2.0, 0.0)
Pain self-efficacy questionnaire (median, IQR)	0.0 (−5.5, 7.5)	−4.5 (−7.0, 1.3)	−5.0 (−8.0, 6.0)

Abbreviations: PEG, Pain intensity, Enjoyment of life, General activity; PROMIS, Patient-Reported Outcomes Measurement Information System.

Secondary outcomes: pain and symptom improvement

The baseline PEG scores of 6.1 decreased to 5.3 by the end of follow-up, among patients with both baseline and follow-up measures. Though the study was not designed or powered to assess efficacy, this translated to an average decrease of 0.8 points on the pain scale. The Brief Pain Inventory (BPI) severity scale similarly decreased from baseline 6.3 to 5.8 at 24 weeks. Patients were asked to rate the % improvement on a scale of 0%–100%. Responses to this global rating of change scale ranged from 10 to 90, with a mean of 50 ± 25 and a median of 50%. Correspondingly, the Global Impression of Change Scale (a 7-point Likert scale on which patients describe improvement in pain) was rated an average of 2.7, corresponding with “somewhat better” pain as rated overall by patients. The combination of these results indicates a clinically significant reduction in pain. The anxiety, fatigue, and depression scores are shown in Tables 4, 5, with approximately a 1-point decrease in scores across follow-up. The Pain Self-Efficacy Questionnaire (PSEQ) decreased from 32 to 29.

Mechanisms of change

While self-efficacy questionnaire results (Tables 4, 5) indicated that there was not a meaningful increase in self-efficacy with the intervention, there were several mechanisms for improvements described by the participants in qualitative interviews (Table 6). When asked about how or why the program was helpful, participants spoke about changes in behaviors, including stress-reduction methods and eating, activity, and sleep behavior changes. Patients likewise described thinking differently about pain and employing coping strategies (eg, exercise, seeking social support, mindfulness). The participants universally appreciated the support and accountability of the weekly health coach sessions. As 1 participant described, support and accountability were helpful:

“You know, it was helpful because I was talking to [the coach] every week. And you know, it’s almost like a therapy….It’s been a nice process you know, because you’re showing up every week … it’ll be so bad if you don’t do the things you say you’re going to do when you get in front of somebody. You’re face to face, so that’s a psychologically a great thing to have.”

TABLE 6.

Patient responses to how the intervention helped them

Patient ID	Summary of improvements	Full response
111	Improved sleep, increased activity, improved medical adherence	Trying to change some of my sleep habits and diet. My attitude towards sharing some of my information has changed, and I still struggle with that. I basically increased my activity; [the program] allowed me to try to be a little bit more active, rather than being sedentary. … Keep up a mindful and hopeful view. It helped me with my pain because, you know, I don’t have as much. …there is that voice that they’ve basically instilled, you know that you better do this or there’s going to be problems in the future you know. It kind of is a reminder or a warning sign on the wall that basically says that I need to be a little bit more compliant with my medical management.
115	Increased activity Increased social interaction	Activity levels increased…. By going out, exercising and getting some sun, reading more and you know, getting involved and looking forward to each week talking to somebody looks like therapy, basically. You don’t go back and do the same thing sitting in the house doing the same thing. That right there is just there’s no advancement. There’s no way to change things you know, and you got to change a good habit the only way you could, change that bad habit and you guys have produced some good habits
118	New job Improved diet Improved sleep Increased activity	Well, a lot, I guess a lot has changed. You know, I got a new job, been doing that for a couple of months now. But like I said, I just feel better during the day because of cutting out the sugar, working on my sleep, which I had a problem with sleeping for as long as I can remember, and it’s gotten a lot better. I actually sleep through the night before I would get up like every hour, no matter what, to go to the bathroom or something. But now I sleep through the night so, that’s been great…Definitely I’m more active now. …I was kind of in a slump, but I didn’t really leave the house much. I kind of just hung out and chilled, watched my YouTube, watched football stuff or whatever, you know? Now I’m always on the go. I had problems sleeping all my life and now I can actually sleep through the night for the most part, so it’s awesome.
118	Increased social connection Improved sleep Stress reduction Cognitive reframing	Having connections with family and friends is so important, and I didn’t realize that until I started going through this. Because I was always the independent type, had to do everything by myself, you know, and I always helped everybody else…she made me aware of things like how the sleeping also plays such a big part… as long as I’m able to get a certain amount of sleep like without interruption. That makes a difference. It really does, and that helps. It kind of resets you a little bit and, in the visualization and I’ll say like deep breathing…. It was something I look forward to. And it was something that I found comfort in. It made me look at some things differently and made me think of things that I really hadn’t thought about…I have an awareness now about so many different things. Because of the groups and because of this program and I really hope that you all can keep it because I think everyone will benefit from it.
120	Social support	For the first time in such a long time, I really felt like I was being listened to by somebody other than family. I think they would tune me out, not because they don’t care, but it’s just everyday thing with me.
121	Mindfulness Outlook Acceptance Increased activity	My pain levels have changed… my pain level is much less, but I don’t know if it’s because now I’m walking, I’m using [mindfulness] to go on my little vacation and take a deep breath, or you know spend a little longer time in my shower or you know? So, I’ve been given some tools, and along with those tools, you know… I think because you feel less pain or you can accept your pain, you’re able to do more like get out and try to do things that you know you weren’t doing before…. being more active like I was before …it gives you better outlook too. I think the best part is having an [coach] to talk. You know, and someone that understood that.
122	Coping skills	The impact of it all was learning how to deal with all this stuff. … I learned how to how to cope with things a little bit more just for myself, not for anybody else…I try to talk myself into when I’m hurting. Go away and just don’t let it stop me.
125	Stress reduction Improved diet	I think the biggest change for me was that it kind of pointed out to me that my stress level is high and because it’s been high for a long time because you know life it is affecting my health and my normal bag of tricks to deal with life stress has been exhausted, so it was time to go back to therapy. So, I think that was kind of the biggest take away for me, was that and, I have one of the things that we talked about was controlling salt intake in my diet. So that’s also been kind of one of the long-term takeaways is you know I’ve stopped adding extra salt to my diet. My pain level lately hasn’t been-it’s been a little bit lower than what it normally runs. I’m usually around four or five. So, my pain level has been significantly lower this summer.
127	Social support Improved diet	It’s worth the time to you know… talk to somebody about how you’re feeling. It gives you the opportunity to understand and there’s something going on and you don’t know if it’s from…my diet and looking at my health differently, the way I take care of myself and do things. It’s made me more aware of what I’m doing.
128	Mindfulness Improved sleep	I’m going to be honest; it was really helpful because that mindfulness has been a godsend. It’s very helpful. It helps me sleep so much better. [I] get to sleep so much, and I seem to stay asleep better when I do that than when I, if I just like, crash. So, I found it to be very helpful. It’s been very good. I’ve learned some very useful techniques.

Other themes were identified in the data, including participants finding the intervention to be validating, experiencing weight loss over the time of the intervention, and appreciating the group sessions.

Participants had experienced invalidation around experiences of pain

Several participants felt that the intervention was helpful by validating the pain experienced by cirrhosis. One participant said, “a lot of people keep on just telling you the liver has no pain.” [P121] Likewise, another said, “You know, most doctors will tell you that the liver does not give you pain. It does. And everybody in the group even said the same thing. My doctor said your liver won’t give you pain. It does.” [P122] Other participants similarly reflected that it was helpful to share experiences outside of the medical encounter.

Weight loss was an added benefit of participation

Participants noted that setting goals around eating differently and physical activity resulted in weight loss. As one participant said, “my main goal that I have right now is just to continue to lose weight.” [P122]; Another said, “And I’ve lost some weight which has been a help as well. And we kind of talked about that nutritional wise and things.” [P120] A third mentioned: “I have lost so much weight.” [P127] Weight loss in cirrhosis is complex, in that sarcopenia is concerning. One participant described healthy eating as it pertained to both their weight loss journey and the weight gain efforts of another participant in their group. “I started showing her some of the things that I’m using to lose weight. Because it’s extra protein which she needed… to gain the weight. [P121]

Group sessions were generally appreciated by participants who chose to attend

Not all participants opted to participate in the groups. However, those who did participate reported generally positive experiences. One participant said, “I like the group sessions learned a lot from some of them. It was also nice to see that there’s more of community there.” [P121] Similarly, another participant said,

I also really appreciated the group sessions. It’s helpful to me. You know, I don’t know a lot of people that have liver disease or have autoimmune diseases affecting their liver, so, it was helpful to me to hear from other people who have the same experience and some of the things that they were doing. That helped them. And also, some of the struggles that they were having, because it made me feel less kind of less alone than what I’m going through. And I got some good tips and ideas from the other people in the group too. So that component of it was pretty helpful for me. [P125]

Another valued the diversity of experiences among participants, saying, “I enjoyed it because we each have different stages, so we all had something different going on. [P127]” Yet there was a sense that the groups were not for everyone. One participant had ongoing concerns about privacy saying, “It was OK. I’m not going to lie. It kind of graded my nerve because of the HIPAA violation. Just because that’s what I do for work [P128].” That said, the participant went on to say, “but it was nice hearing from other people how they felt during the process.” [P128] Participants appreciated having the option for groups but not mandatory participation.

DISCUSSION

This study is the first to assess the impact of a tailored PSM approach designed to meet the needs of people with cirrhosis and chronic pain. We found that this was an acceptable and feasible intervention, with improvements in pain and symptoms over time that lasted even 3 months after the end of the intervention. While a full-scale trial would offer more data, this pilot suggests that LEAP is a new option for a complex group of patients, meeting a need identified by both patients and providers.

Participants experienced clinically significant improvements in pain and symptom ratings over time. These differences of ~1 point on the BPI pain score were similar to those found in the STOMP trial, which reported a reduction in BPI of 1.25 at the end of intervention and −0.62 points after 3 months post-intervention. ³³ Notably, these pain rating improvements with LEAP were sustained at 3 months post-intervention and among people who did not opt to participate in the group sessions. Likewise, self-rated pain improvement was as high as an average of 65% at the end of the individual sessions, well above the 30% difference that defines clinically meaningful change on this scale. While this pilot was not powered to detect efficacy, the results suggest that the intervention is worthy of further investigation.

Modular PSM interventions have been used for people living with HIV, cancer, and other disease populations. While the educational content may shift, the intervention structure, goals, and methods are consistent regardless of the modules that are selected (eg, setting goals, tracking goals, self-talk, problem solving), with a goal of improving self-regulation and self-efficacy (ie, the mechanism of change is consistent). Thus, while there is intentionally no one-size-fits-all PSM intervention, people with cirrhosis were engaged in designing the informational content of the modules. This flexibility, rather than requiring patients to learn about specific content, like weight loss or alcohol, that may be irrelevant, allows the intervention to be more scalable and adaptable across liver disease etiologies. Because patients selected the behavior changes and goals that were relevant to them, participants described that the intervention changed different behaviors, as was outlined in Table 6. These preliminary data did not support the hypothesized mechanism of increasing self-efficacy. However, it may be that patients overestimated their baseline self-efficacy, since they had not tried the self-management activities at baseline. There was a meaningful increase in physical activity and PSM behaviors, indicating that behavior changes were likely key to pain reductions. There is emerging recognition that behavioral pain management strategies are beneficial for complex populations with individualized needs. A recent study of 311 patients found positive pre–post effects in patients with a range of chronic diseases. ³⁴ Likewise, increasing evidence supports tailored programs for specific chronic illnesses. STOMP is a program for people living with HIV that has resulted in significantly reduced pain and improved function. ³³ In hepatology, a small trial found that mindfulness-based stress reduction improved depressive symptoms and quality of life in 20 patients with cirrhosis and depressive symptoms. ³⁵ However, a comprehensive, individualized program of PSM for people with cirrhosis had not been previously tested, despite patients identifying a clear need for this type of intervention. The fact that participants selected different modules and focused on different goals but universally described changes in PSM behaviors suggests that a one-size-fits-all approach or a single modality of intervention (eg, mindfulness) is unlikely to have as much success as choices of multiple modalities across the heterogeneous population of people with cirrhosis. For example, some participants with cirrhosis need to lose weight while others need to gain muscle. Likewise, mindfulness worked better for some of the participants, while others benefited more from changes in physical activity. Future work will need to build on the understanding of the mechanism of the intervention, should it prove efficacious in a larger trial.

Participants experienced several unintended positive consequences. Multiple participants described weight loss successes. Participants valued the opportunity to learn about cirrhosis. The health coach spent more time than expected discussing dietary issues such as low salt, high protein diet, across modules, which would be expected to help health outcomes in this population. Because weight management and sleep are both associated with reduced pain, many of the strategies that address pain also address clinical needs for patients with cirrhosis. Protein at bedtime is good for sleep, encephalopathy, and sarcopenia, for example. Likewise, since weight management can help with chronic pain, and behavioral approaches can help with weight loss, the goal setting, tracking, education, support, and accountability offered by the intervention were helpful for participants with MASLD who were working on weight loss. Thus, future studies should track the impacts of PSM interventions on outcomes other than pain and mental health, with an additional focus on weight loss and symptoms of cirrhosis.

The educational aspects of the intervention that were deemed to be helpful by participants can inform general hepatology practice. The health coach repeatedly heard participants say that cirrhosis should be treated more like diabetes, in terms of structured education and interventions at the time of diagnosis. While most clinicians offer education and an overview of cirrhosis to new patients, more structured approaches, particularly for patients with lower health literacy, are likely warranted. Participants felt validated when they learned that others with cirrhosis experienced pain, because they felt that health care team members had dismissed their pain symptoms. This indicates that, even if such interventions cannot be offered, there are opportunities for clinicians to validate pain associated with liver disease and listen to their patients.

The group sessions were a novel component of this modular intervention. We made the group sessions optional based on our findings in developmental work that patients had mixed reactions to the idea of groups. Group support is often applied to PSM and other CBT-based interventions, based on the idea that supporting others can build self-efficacy, offer positive reinforcement, and other benefits. Studies have demonstrated the positive impact of group psychotherapy on depression for people listed for transplantation and the ability of such approaches to reduce anxiety and improve social support, despite patient reluctance to participate.^36–39 These findings were consistent with ours, in that only 11 of 21 participants opted to participate in the groups. Despite lower group participation, participants did well overall, indicating the need to disaggregate the impacts of individual and group sessions in future studies. Several anecdotes indicate unmeasured benefits of these groups: One included a younger participant with asymptomatic disease alongside a participant who had refractory ascites. The experience of seeing someone manage decompensated disease was beneficial to the participant with asymptomatic liver disease. For example, the coach noted that the participant without advanced disease started to take the work more seriously when they entered the group environment. Another group included multiple people with MASH cirrhosis, who spent much of the time doing support around weight loss, with good effects on pain, mood, and weight, according to people in the group.

While this was a unique pilot study for people with cirrhosis, there were several notable limitations. As a result of identified challenges, we learned several lessons that will inform future work. First, it was challenging to recruit and retain patients with cirrhosis, given the repeated hospitalizations, high mortality rates, and complex psychosocial situations (eg, unstable housing). COVID was challenging in terms of in-person recruitment, and the need to rely on clinician referral and online recruitment limited our ability to track the number of patients approached. Second, this trial was not designed to be an efficacy trial and was not powered to detect changes in clinical outcomes, and small pilot trials can preclude certain analyses. For example, due to the heterogeneity of analgesic types and doses used by patients in the trial, we could not interpret changes in medication use. While we recruited both men and women and participants, with a range of disease etiologies and severities, the patients who participated were predominantly white and from a single center. Women were likely overrepresented in this pilot. However, women are generally more likely than men to have chronic pain and are overrepresented in trials of chronic pain management approaches (ie, 70% of trial participants are women), so this was not unexpected. ⁴⁰ However, given this potential for selection bias, future trials should consider methods to recruit men or to block randomize by sex. Likewise, using only one health coach may limit the generalizability of the findings, but this will be addressed by a larger, multi-site trial. While our completion of at least 5 sessions did not reach our 80% goal, this is in part explained by the high level of illness and symptoms experienced by our complex population (eg, 2 participants died over follow-up). In addition, had the participant who attended 4 but not 5 sessions completed 1 more session, we would have met this measure. The non-response to the post-survey may have unintentionally biased our findings toward the positive (ie, people who completed the survey were more likely to benefit and stay engaged in the intervention). This concern is somewhat mitigated by the similarities between the groups that did and did not complete the intervention, and by the limited extent of noncompletion.

There were several key lessons that will inform subsequent trials. While the completion rate of 80% was not achieved, we used a strict a priori definition of completion and noted that the study was particularly challenging to conduct during COVID and with a very medically ill population. That said, the 76% completion rate exceeds that found in a meta-analysis of PSM studies and is considered to be acceptable in the field. ⁴¹ Future study will assume a higher (35%–40%) attrition rate in their design as a result of the pilot. We will also use a block-randomized design that will account for decompensation and allow us to assess heterogeneity of treatment effects. We would recommend that future studies consider assessing the minimal number of intervention sessions that are associated with clinically meaningful outcomes, rather than setting an a priori threshold for dose, following an intent-to-treat approach. Future work should take into account the high degree of illness experienced by patients and maximize flexibility to meet the needs of patients who are being hospitalized. Despite the limitations of this single-arm, single-site pilot trial, these promising preliminary data indicated that the LEAP intervention merits further study. Thus, this pilot provides critical information for a future large-scale effectiveness trial, where we now plan to offer a support intervention to the control group (rather than an inactive control).

In conclusion, this single-arm pilot trial of a unique behavioral intervention for pain for people with cirrhosis found changes in health behaviors and symptom improvements. While future trials can uncover mechanisms of change and test the intervention against attention control, PSM approaches should be considered for clinicians aiming to address this under-addressed and challenging problem of pain management in hepatology.

Supplementary Material

hc9-10-e0955-s001.docx ^{(35.3KB, docx)}