Development and Preliminary Evaluation of an Integrated Cognitive-Behavior Treatment for Chronic Pain and Substance Use Disorder in Patients with the Hepatitis C Virus

Abstract

Objective. Individuals with the hepatitis C virus (HCV) have high rates of both chronic pain and substance use disorder (SUD). Despite high comorbidity, there are limited data available on effective methods of treatment for co-occurring chronic pain and SUD. In this study, we sought to develop and conduct preliminary testing of an integrated cognitive-behavior therapy (CBT) for chronic pain and SUD in patients with HCV.

Design. Descriptive, including pretreatment, posttreatment, and follow-up testing.

Setting and Patients. Outpatient clinic as part of one VA Medical Center.

Participants. Veterans with chronic pain, SUD, and HCV.

Intervention. Eight-session integrated group CBT for chronic pain and SUD in patients with HCV.

Methods. Participants completed standardized measures of pain, function, depression severity, and alcohol and substance use at baseline, post-treatment, and 3-month follow-up.

Results. Generalized estimating equations identified improvements in pain interference, reducing cravings for alcohol and other substances, and decreasing past-month alcohol and substance use. The proportion of participants who met diagnostic criteria for current SUD demonstrated a four-fold decrease over the course of the study from 24% at baseline to 15% at post-treatment and 6% at 3-month follow-up. On response to a global impression of change, 94% of participants noted improvement from baseline.

Conclusions. Results from this pilot study suggest that a customized CBT for patients with both chronic pain and SUD (CBT-cp.sud) may be beneficial in improving important pain and addiction-related outcomes in patients with HCV. Larger scale investigations of this intervention appear warranted.

Introduction

Chronic pain co-occurs with substance use disorders (SUDs) at high rates [1,2], thereby impeding potential recovery. Although a sizeable body of literature is available on treatments for pain disorders and addictive disorders, a paucity of data are available regarding integrated treatment. Research in pain management has frequently excluded patients with SUD [1]. Recent data suggest that patients with chronic pain and comorbid SUD need more intensive interventions than usual care in order to attain clinically significant improvements in pain-related function [3].

Clinical treatment approaches for chronic pain and SUDs have often been isolated, with chronic pain programs mandating a period of abstinence or that patients be free of current SUDs prior to admission [4]. Data are mixed on SUD outcomes for patients with co-occurring chronic pain [5–8]. Unfortunately, abstinence from substances may be difficult for patients with chronic pain, particularly when pain is a primary trigger for alcohol or substance use [9].

We have identified only two published studies that have examined the potential impact of integrating chronic pain treatment with substance abuse treatment. Currie and colleagues [10] found that a 10-week outpatient group based within a substance abuse treatment program was effective in improving pain-related outcomes in 44 patients; results were maintained for 12 months. Similarly, Ilgen and colleagues developed a 12-week group cognitive-behavior therapy (CBT) for chronic pain, which was examined in 17 patients recruited from a substance abuse treatment program, who were predominately in the early stages of recovery [11]. They found that the integrated intervention resulted in significant improvements in pain-related outcomes, mental health, and substance use severity. Another recent pilot study found that interventions for chronic pain are well-tolerated by patients with pain who are enrolled in methadone maintenance programs [12].

Although these prior studies are promising and suggest that an integrated approach for chronic pain and SUD may be feasible, the studies have limitations. They required participants to be abstinent from alcohol and illicit substances during treatment [11] or have current enrollment in SUD specialty care [12]. Additional research is needed about the efficacy of an integrated intervention for chronic pain and substance abuse among patients who have SUD, but not necessarily enrolled in specialty SUD treatment settings.

The purpose of this research was to develop and conduct a preliminary investigation of an integrated cognitive-behavior treatment for chronic pain and SUD (CBT-cp.sud) among patients with the hepatitis C virus (HCV) (not restricted to patients previously identified as having an SUD and who were receiving treatment within an SUD program). We specifically targeted patients with HCV because previous research has demonstrated that this clinical population has both high rates of SUD [13–15] and chronic pain [16,17]. Furthermore, HCV is the most common blood-borne infection in the United States, affecting nearly 2% of the general population [18]. Many commonly used medications for chronic pain may not be appropriate for patients with HCV. For example, clinicians may not prescribe opioid medications to patients with HCV due to histories of SUD [19,17] and medications containing acetaminophen may not be appropriate for patients with cirrhosis due to concerns about liver toxicity [20]. Thus, patients with HCV may be especially appropriate for non-pharmacological interventions for chronic pain and SUD.

The present report includes a description of our process for developing an integrated treatment for chronic pain and SUD among patients with HCV, as well as providing an overview of the treatment itself. We report preliminary results from a pilot trial in which the integrated intervention was tested. The pilot study was conducted to examine the acceptability and preliminary efficacy of an integrated CBT for chronic pain-related and substance abuse-related clinical outcomes among patients with HCV.

Methods

Participants

All participants were recruited from the VA Portland Health Care System via posted advertisements in the Medical Center and announcements made in mental health classes. Potential participants were informed that the study would test the effectiveness and acceptability of a new integrated treatment designed to improve pain-related function and reduce symptoms of SUD in patients with HCV. In order to be included in the study, participants must have had a current diagnosis of HCV (defined as a positive HCV antibody result confirmed with detectable HCV RNA level on polymerase chain reaction test), self-reported current pain ≥ 4 on a scale of 0–10 (where 0 = no pain and 10 = worst pain imaginable), self-report of experiencing pain for at least 6 months, history of SUD assessed with the Structured Clinical Interview for DSM-IV, over age 18, and were able to speak English. Participants were excluded for age greater than 70 years, history of antiviral therapy or plan to begin antiviral therapy during the study period (because antiviral therapy can lead to somatic symptoms that could confound outcomes of the intervention), current unstable psychiatric disorder (defined as current clinically significant suicidal ideation or untreated schizophrenia or bipolar disorder), or pending litigation or disability compensation for pain. Whether patients were taking medication(s) for chronic pain, or if they had previously participated in a substance abuse treatment program, did not impact study eligibility.

A total of 87 patients were screened for study eligibility. Of these, 38 were ineligible (16 did not have HCV, nine had recently undergone antiviral therapy, five were over age 70, and eight were excluded for other reasons). Eighteen patients were eligible, but declined to participate in the study (eight declined participation and 10 patients had a standing conflict at the scheduled group meeting time). An additional 10 eligible patients signed consent to participate but did not attend any group treatment sessions. A total of 21 patients enrolled in the study, participated in at least one treatment session, and completed a follow-up assessment.

This study was approved by the Institutional Review Board of the VA Portland Health Care System. All participants provided written informed consent to participate.

CBT-Cp.Sud Treatment Development

The biopsychosocial model is a common conceptualization for both chronic pain and SUD and serves as the basis for CBT [21–23]. We conducted a multi-step process to develop an integrated intervention (CBT-cp.sud) designed to improve pain-related function and substance abuse outcomes among patients with HCV. In Step 1, we reviewed relevant treatment manuals to identify shared components of treatment. CBT has been shown to be effective in improving pain-related outcomes [24] and treatment for SUDs [25]. Given the high degree of similarity between CBT for chronic pain and for SUD, we initially examined commonly used treatment manuals for each condition [26,27] to identify potential areas of overlap. We also consulted with authors and examined treatment manuals of recent studies that have integrated pain and addiction treatment [10–11]. Based on this review and discussions, we developed a preliminary outline of an integrated 10-session group SUD and chronic pain CBT-cp.sud.

In Step 2, we conducted a cross-sectional assessment of biopsychosocial factors associated with chronic pain in 119 patients with HCV [16]. The sample was recruited from a single VA Medical Center and included adult participants who reported moderate to severe pain and had a current pain-related diagnosis documented in the medical record. Results indicated that biopsychosocial constructs and processes accounted for 37% of the variance in pain severity and 56% of the variance in pain interference in this sample of patients with HCV. The most significant factors associated with pain severity were pain catastrophizing [28] and availability and perception of social support. Modifiable variables significantly associated with pain interference were prescription opioid use, which was negatively associated, and chronic pain self-efficacy [29]. A secondary analysis with this dataset examined associations between depressive symptoms and pain interference. We found that clinically significant comorbid depressive symptoms and chronic pain affect 26% of patients with HCV, and pain-related anxiety may mediate the relationship between depressive symptoms and pain interference [30]. Based on these results, we determined the intervention should have a cognitive-behavior treatment orientation (i.e., to address pain catastrophizing and self-efficacy for managing pain) and decided to include two group treatment sessions focused on identifying maladaptive beliefs and cognitive restructuring (rather than one). We also planned to increase our focus on the impact of co-occurring mental health issues in Sessions 1 and 2.

In Step 3 of treatment development, we conducted two focus groups (total n = 7) with patients with HCV. The purpose was to discuss treatment approaches for pain and addiction, obtain an understanding of participants’ experiences with different forms of treatment, and to discuss our intended treatment model. Based on focus group findings, we made four substantive changes to our research methods and treatment model, and received general support of the overall treatment focus.

1. All participants universally recommended that we not limit treatment to patients with current alcohol or SUD, and that we should include participants with a remote history of SUD. Participants stated that substance abuse remains an ongoing concern, even if they had been abstinent for many years, and that urges to use substances are heightened during times when pain management is poor. As a result, we modified the study inclusion criteria to include patients with any history of SUD and altered our treatment plan to discuss SUD-specific issues during each treatment session.

2. Based on participant perceptions and feedback, we replaced a session on general assertiveness training to focus explicitly on strategies for talking with one’s primary care physician (PCP) about pain management strategies and functional goals.

3. We originally planned to devote a full session on healthy living with HCV [31]. However, participants stated that this information was thoroughly discussed following their initial diagnosis and is frequently revisited during annual primary care follow-ups. As a result, we replaced this topic with a more comprehensive session on sleep hygiene.

4. Some focus group participants expressed reservation about our intended treatment length of 10 weeks. They stated that they would be more inclined to participate in an intervention of shorter duration. Based on this feedback, we modified the intervention to consist of eight weekly sessions.

In addition to the above, focus group participants expressed strong support for the general CBT-cp.sud treatment model. They indicated that they had not previously had discussions about the inter-relationship between pain and addiction and welcomed the opportunity. Participants provided support for many of the specific concepts that would be discussed in treatment (e.g., pacing for pain, identifying maladaptive beliefs, relapse prevention). They liked the group therapy format, rather than individual therapy, because they could connect with other participants, receive support, and share coping strategies. They agreed with the intended focus on functional outcomes, rather than a focus on pain intensity. Finally, participants expressed interest in the opportunity to discuss and share alternative behavioral strategies for chronic pain management, and preferred to not focus exclusively on pain medications.

In Step 4 of the treatment development process, we pilot-tested the intervention with three patients in individual therapy. Although the treatment was designed for group therapy, we chose to pilot in individual therapy to develop a better understanding of the critical ingredients of the treatment without the potential confounding factors associated with group therapy dynamics. Additionally, these individual treatment sessions provided each of the planned group therapy facilitators (BJM and DWG, both licensed psychologists) an opportunity to increase their comfort with the materials. Following each treatment session, the two treatment facilitators met to review experiences and problem-solve strategies for the group intervention. This process led to a change in the order of session topics and development of strategies for explicitly discussing addiction-related issues in the context of chronic pain management. See Table 1 for a comparison of the current integrated treatment with traditional CBT approaches for chronic pain and for SUD.

Table 1.

Outline of an integrated cognitive-behavior therapy (CBT) for chronic pain and substance use disorder (SUD) in patients with the Hepatitis C virus

Session number	Traditional CBT for chronic Pain	Traditional CBT for SUD	Integrated CBT for chronic pain and SUD*
1	Education about chronic pain	Assessing high risk situations	Education about chronic pain and addiction
2	Relaxation training	Functional analysis	Inter-relationship between chronic pain and substance use
3	Cognitive distortions	Increasing pleasant activities	Identifying automatic thoughts
4	Cognitive restructuring	Seemingly irrelevant decisions	Cognitive restructuring
5	Pacing	Urge surfing	Increasing pleasant activities and pacing for chronic pain
6	Sleep hygiene	Assertiveness training	Pain communication and pain medications
7	Anger management	Cognitive restructuring	Sleep hygiene
8	Planning for flare-ups	Relapse prevention	Relapse prevention

Note: * = The inter-relationship between chronic pain and SUD was discussed during each of the eight sessions.

Overview of CBT-Cp.Sud Treatment

The overarching treatment model for CBT-cp.sud was CBT, with focus on improving pain-related function and reducing harms from SUD. Although not necessarily explicit in the title of each session, there was focus on the interaction of pain and SUD during each treatment session. Every session included the introduction of a relaxation exercise, which participants were encouraged to practice daily for 10 or more minutes. Participants enrolled in group treatment sessions as a cohort. All group treatment sessions were facilitated by the same two psychologists who closely followed the intervention manual and met following completion of each session to discuss session content, group process, and treatment fidelity.

Session 1 focused on education. Information was provided about the etiology of chronic pain and addiction, with a focus on the biopsychosocial model and a discussion of the differences between acute and chronic pain. In Session 2, the principal focus was on the inter-relationship between pain and addiction. Participants were asked about triggers for chronic pain (what causes flare-ups) and triggers for substance use (when participants have increased urges to use), and encouraged to explore areas of overlap. We conducted a functional analysis to examine a recent experience in which participants used (or had increased urge to use) alcohol or illicit substances as mechanisms for coping with pain; this included a discussion of the anticipated and actual consequences associated with this episode of substance use. A central emphasis of CBT is the role of cognitive distortions in pain and SUD, thus Session 3 focused on identifying automatic thoughts and their impact on both pain severity and urge to abuse alcohol or illicit substances, while in Session 4 participants were taught strategies for challenging automatic thoughts, cognitive distortions, and maladaptive beliefs. A great deal of attention in treatment for chronic pain is on pacing of activities, but less attention has focused on the inclusion of positive, discretionary versus required activities (e.g., household chores). Session 5 was an integration of traditional pacing for pain with increasing pleasant activities (originally developed for relapse prevention). In chronic pain treatment, pacing involves identifying a reasonable amount of time one can engage in given activity without an increase in pain or fatigue, then setting predetermined quotas for the amount of time to engage in the activity. Participants were encouraged to consider social or leisure activities that could be enhanced, while also setting limits about the extent of participation. A discussion about pain medications and strategies for communicating with PCPs was the focus of Session 6. We considered different methods for discussing chronic pain, how participants’ history of addiction could impact their conversations with medical staff, and encouraged participants to focus on meeting functional goals rather than changes in pain intensity. As problems with sleep are frequently identified as a problem for individuals with both chronic pain and SUD, Session 7 addressed specific behavioral strategies for sleep hygiene. For lifestyle change program of any type, relapse following initial behavioral change is common. Relapse prevention was addressed in Session 8. This included addressing inevitable pain flare-ups and increased urges to use alcohol or illicit substances. Treatment also focused on increasing the regular use of pain and SUD coping strategies as treatment prevention, which involved ongoing practice of treatment skills.

In summary, this eight-session group CBT-cp.sud in patients with HCV integrates what we believe are the primary ingredients of essential treatment for pain and addiction with an emphasis on self-management and problem-solving. The intervention includes regular and ongoing discussion of the inter-relationship between exacerbations in pain with increased urges to abuse alcohol or illicit substances. Patients with chronic pain and SUD may be at increased risk of relapse when pain-related symptoms are heightened. This treatment focuses on pain-related function, reducing risk for relapse, and ceasing the use of alcohol or illicit substances as mechanisms to address increases in pain intensity.

Measures

Participants completed assessments at baseline, post-treatment, and 3 months post-treatment. Demographic characteristics were collected by self-report. These included age, gender, race, marital status, and yearly income.

Pain and Mental Health Outcomes

Consistent with recommendations from the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT), we assessed pain severity and interference, depressive symptomatology, and global impression of change [32,33]. Pain severity and pain interference were assessed with the Multidimensional Pain Inventory (MPI) severity and interference scales, respectively [34]. The MPI is a frequently used and well-validated 52-item self-report measure that assesses pain and its physical and emotional impact in the past week. We used the pain severity (three items) and interference (nine items) subscales. Scale scores on the MPI range from zero to six, with higher scores indicating more pain or interference.

The Beck Depression Inventory – Second Edition (BDI-II) is a widely used 21-item self-report measure that assesses the severity of depressive symptoms in the past two weeks [35]. Scores are summed and higher scores indicate more severe symptoms of depression.

A one-item Global Impression of Change [36], administered at post-intervention and 3-month follow-up visits, assessed perceived change in function since the start of treatment. The measure is scored on a seven-point Likert scale from very much worse to very much better. Pain severity, pain interference, and depression were measured at all study visits.

Substance Use Outcomes

Substance use outcomes were assessed at all study visits. The TimeLine Follow Back (TLFB) is a frequently used method that uses calendar prompts to elicit frequency and intensity of substance use within a specific timeframe [37]. This study assessed the number of days in the past month in which alcohol, marijuana, methamphetamine, cocaine, heroin, and prescription opioid medications were used. All participants were queried about the past month use of all illicit substances regardless of their substance use history.

The Penn Alcohol Craving Scale (PACS) is a five-item self-report measure of alcohol and other drug craving in the past week [38]. Scores range from 0—30 with higher scores indicating greater craving for alcohol.

The Structured Clinical Interview for DSM-IV (SCID) was used to assess current and past SUDs [39]. For the purpose of this study, a participant was coded as having an SUD only if s/he endorsed symptoms in the past month. Participants who were concurrently enrolled in an opioid treatment program (e.g., methadone maintenance or buprenorphine therapy) were not diagnosed with SUD if they did not endorse current symptoms, as DSM-IV diagnostic criteria that were used at the time of this study categorize opioid use disorders treated with opioid agonist therapy as “in remission.” SCID interviews were conducted by masters-level research clinicians or students enrolled in graduate-level clinical psychology or social work programs. A licensed psychologist provided initial and ongoing training of SCID interviewers.

All participants completed urine drug tests (UDTs) at each study assessment. Substances tested included ethanol, amphetamines, benzodiazepines, barbiturates, cannabis, cocaine, opiates, and phencyclidine.

Analyses

Our primary objective of this pilot trial was to assess feasibility of our protocol and establish variance estimates for treatment outcome variables that can be used for sample size calculations for a future large-scale trial [40]. To examine the preliminary efficacy of group CBT-cp.sud to improve pain and SUD outcomes, we employed generalized estimating equations to account for the non-independence of observations within patients due to longitudinal assessments. For all models, we specified model-based variance estimators and unstructured correlation matrices to examine the effect of time on pain, depression severity, and substance use outcomes. Models of continuous variables—pain severity, pain interference, depression, and alcohol/drug craving—specified a normal distribution and identity link function. For the outcome variable number of days in the past month on which substances were used (not including prescription opioid medications), we specified a negative binomial distribution with a log link function because data screening procedures indicated data violated assumptions of a Poisson distribution due to the variance of this count variable exceeding its mean. For the binary outcome variables of current SUD diagnosis obtained from structured clinical interviews and urine drug test positive for a non-prescribed substance, a binomial distribution with a logit link function was specified. Sensitivity analyses were conducted for all study outcomes controlling for time-varying covariates of receipt of prescription opioid therapy and opioid agonist therapy; the results were unchanged and are available upon request from the corresponding author. Descriptive statistics characterized participants’ global impressions of change at post-intervention and 3-month follow-up. All inferential tests were performed for participants who attended the first group therapy session, and models utilized all available data. Effect sizes are presented as Cohen’s d for continuous outcome variables and odds ratios for count and binary outcome variables. Although we report P values for inferential tests, we also describe trends as we are underpowered to detect statistically significant effects.

Results

Group treatment sessions were completed with four cohorts of participants (total n = 21). The average age of participants was 57.6 (SD = 6.3) years, 88% were male, 54% reported white race, and 59% were separated or divorced. Participants had mixed pain-related diagnoses: 38% had back pain as primary pain-related diagnosis, 28% had joint pain (including arthritis), and 33% had another pain-related diagnosis (bone spur, testicular pain, carpal tunnel syndrome, etc.). The average self-reported duration of pain was greater than 13 years. Eleven participants (52%) were prescribed an opioid at entry into the study; seven of these patients received these medications as part of enrollment in an opioid treatment program. The rates of opioid use did not change over the course of the 8-week treatment.

At baseline, five participants (24%) met diagnostic criteria for a current SUD, these included alcohol, heroin, cocaine, and other stimulant use disorder. All participants had an extensive past history of SUD: 81% had history of alcohol use disorder, 69% cocaine use disorder, 44% cannabis use disorder, 44% stimulant use disorder, 38% heroin use disorder, 38% sedative use disorder, 31% hallucinogen use disorder, 21% polysubstance use disorder, and 19% prescription opioid use disorder. Seven participants (33%) were enrolled in an opioid treatment program at the time of study participation, and three participants (14%) completed substance abuse treatment within the year prior to enrolling in this study. Eleven participants (52%) had no past-year participation in substance abuse treatment.

Participants attended an average of 5.9 of the eight group treatment sessions. Seventy percent of participants attended five or more sessions. Number of treatment sessions attended was not significantly associated to study outcomes at post-treatment or follow-up. Rates of assessment completion at post-intervention were high with 20 participants (95%) completing the post-intervention assessment. Retention declined by 3 months follow-up; 16 participants (76%) completed this final assessment.

Pain and Mental Health Outcomes

Both pain severity and pain interference significantly declined from baseline to post-intervention. Improvements were sustained for pain interference through 3-month follow-up, while pain severity rebounded slightly. No changes were observed in depression symptom severity from baseline to post-intervention or from post-intervention through follow-up. See Table 2 for mean values of all pain, depression, and substance use outcome variables across the three assessments, effect sizes, test statistics, and P values.

Table 2.

Changes in pain, depression, and substance use disorder  (SUD) outcomes

Outcome	Baseline mean  (SE) (n = 21)	Post-intervention mean  (SE) (n = 20)	Follow-up mean (SE)  (n = 17)	Effect size comparing post-inter vention to baseline	Effect size comparing follow-up to post-intervention	P value for overall effect of time
MPI pain severity	3.86  (0.85)a	3.27  (1.08)b	3.54  (1.18)a,b	0.69	0.21	Wald χ2(2) = 9.86, P = 0.007
MPI Pain Interference	4.61  (1.42)a	3.98  (1.23)b	3.82  (1.16)b	0.44	0.13	Wald χ2(2) = 9.73, P = 0.008
BDI-II	19.00  (12.56)a	19.25   (14.43)a	17.13  (11.16)a	0.02	0.19	Wald χ2(2) = 0.94, P = 0.633
TLFB Substance Use*	10.76  (15.42)a	8.19  (13.13)b	6.75  (12.00)b	OR = 0.76 95% CI = 0.58–1.00	OR = 0.79 95% CI = 0.60-1.03	Wald χ2(2) = 7.43, P = 0.024
PACS Alcohol and Drug Cravings	7.42  (5.88)a	7.25  (7.29)a,b	6.33  (7.26)b	0.03	0.13	Wald χ2(2) = 79.79, P < 0.001
% who tested positive for a non-prescribed substance**	14%a	14%a	10%a	OR = 1.00 95% CI = 0.29–3.19	OR = 0.79 95% CI = 0.26–2.42	Wald χ2(2) = 0.17, P = 0.920
% with SUD diagnosis based on SCID***	24%a	15%a	6%a	OR = 0.54 95% CI = 0.17-1.70	OR = 0.64 95% CI = 0.19–2.17	Wald χ2(2) = 2.63, P = 0.269

Note: Superscript letters characterize statistical tests comparing  (1) post-intervention to baseline, and  (2) follow-up to post-intervention. Mean values with different symbols significantly differ at the P < 0.05 level.

^*Effect size reported as odds ratio with 95% confidence interval.

^**Proportion of sample testing positive for a non-prescribed substance. Effect size reported as odds ratio with 95% confidence interval.

^***Proportion of sample meeting criteria for current SUD. Effect size reported as odds ratio with 95% confidence interval.

Responses to the Global Impression of Change suggest general improvement. At the 3-month follow-up, 16 respondents noted moderate to very much improvement while only one participant reported no change or worsening of symptoms.

Substance Use Outcomes

Alcohol and drug cravings as assessed by the PACS decreased from baseline to follow-up; however, neither change from baseline to post-intervention nor post-intervention to follow-up reached statistical significance. Participants reported a reduction from baseline to post-intervention in the number of days on which they consumed alcohol or used other illicit substances, with slight but non-significant reductions from post-intervention to follow-up. Positive urine screens for non-prescribed substances remained consistently low across the study, with 14%, 14%, and 10% of participants testing positive at baseline, post-intervention and follow-up, respectively. Non-significant declines in rates of current SUD occurred over the course of the study. At baseline, 24% of participants met criteria for a current SUD based on diagnostic interview, while 15% and 6% met criteria at the post-intervention and follow-up assessments, respectively (see Table 2).

Discussion

This study describes a process of developing, modifying, and testing the feasibility and acceptability of a unique integrated treatment, CBT-cp.sud for chronic pain and SUD in patients with HCV. We used an iterative process to adapt and combine commonly used treatments for chronic pain and SUD [26,27], The hope is that the integrated intervention provides robust treatment for improving pain-related function, reducing symptoms of SUD, and enhancing quality of life in patients with HCV.

In early focus groups with patients with HCV, we received feedback that the treatment was acceptable and that patients thought the focus was appropriate. Pain specialty treatment programs often require a period of abstinence from alcohol or illicit substances prior to initiating treatment. However, requiring abstinence from illicit substances may not be feasible for patients who use substances as a mechanism for coping with pain [9,41]. Feedback from study participants suggested that a treatment that integrated pain and substance abuse would be preferable to a linear approach.

Though we were able to successfully recruit cohorts of patients to enroll in this pilot trial, treatment engagement may be more difficult than in traditional studies of chronic pain management. In the present study, 57% of the patients who screened eligible for the study ultimately did not enroll. Of note, of the 18 patients who were eligible but declined participation, 10 patients (55.6%) reported a conflict with the scheduled group time, suggesting a significant portion of patients had logistical barriers (e.g., work schedules, childcare needs, transportation difficulty) impeding participation. This rate of non-participation may be somewhat higher than usual pain treatment and could reflect a patient population with greater ambivalence toward treatment [42], increased psychosocial stressors [43], or other factors. Future treatment and research focused on integrating pain and SUD may include case management, motivational interviewing, and/or other approaches designed to optimize patient engagement. These pilot data suggest that, once enrolled in treatment, most people attend the majority of group treatment sessions.

Results from this study suggest that integrating pain and substance abuse treatments is suitable to patients and may be clinically appropriate. Furthermore, results from the 3-month follow-up assessment are important as they suggest that the integrated CBT-cp.sud may be helpful in improving pain-related function while also reducing cravings for alcohol and other substances, and decreasing past-month alcohol and substance use. These findings are consistent with other recent cohort studies examining psychosocial pain treatment interventions among patients currently enrolled in substance abuse treatment programs [10,11] and indicate that patients with SUD may benefit from integrated pain and SUD treatment regardless of substance abuse treatment status. Future research, which would include an active control group, is needed to examine the potential utility of integrated treatment on pain-related function and SUD-related variables. The present study was conducted in an outpatient setting with a convenience sample of participants; future research should also evaluate the optimal clinical setting for this intervention (e.g., primary care, SUD specialty care, pain clinic).

There exist potential concerns associated with integrating pain and SUD specialty treatment. Like treatment for chronic pain [44], alcohol and SUD treatment may be best completed in a multidisciplinary setting [45] and patients often benefit from medical consultation and pharmacotherapy [46]. A limitation of an integrated treatment for pain and SUD may involve patient perception that this brief group treatment would replace comprehensive SUD or chronic pain treatment. The current intervention may likely be used as an adjunct to more intensive interventions rather than stand-alone treatment.

Although preliminary findings suggest potential for the benefit of an integrated treatment, there exist weaknesses of this study. The primary limitations include the small sample size and lack of control group. Further testing of the intervention in a larger trial, including patients recruited from diverse settings and backgrounds, is needed. Future research should also include longer-term follow-up. The intervention in this preliminary trial was delivered by licensed psychologists, which may not be feasible in other settings; future research may employ a variety of clinicians representative of the health care system. Our study inclusion and exclusion criteria were intentionally broad, particularly related to pain-related and SUD diagnoses; interventions that include more homogenous patient populations (e.g., restricted to patients with only one type of SUD) might alter the sessions to address substance-specific issues. Not all participants met criteria for current SUD at study entry and SUD outcomes for this study are more difficult to evaluate as a result. This intervention included treatment components specifically related to HCV and future studies may extend this to other pain and SUD populations that do not necessarily include patients with HCV. Finally, our retention for the 3-month follow-up was relatively low.

In summary, we detail a process for adapting traditional CBT to integrate treatment for chronic pain with substance abuse treatment, with a focus on treatment in patients with HCV. The CBT-cp.sud was developed through an iterative process, which included peer consultation, integrating contemporary data to inform treatment development, focus group testing with potential consumers, and pilot-testing in individual therapy. Preliminary evaluation of the intervention suggested improvements on important pain-related and SUD-related outcomes. Further evaluation of integrated treatment approaches for pain and SUD, in larger trials and with an active control group, and longer-term follow-up, are indicated.