The Learning About My Pain study protocol: Reducing disparities with literacy-adapted psychosocial treatments for chronic pain,a comparative behavioral trial

Abstract

Chronic pain is a critical public health problem that affects over 100 million Americans. Medical pain treatments carry undesirable side effects, whereas low-risk psychosocial treatments offer notable benefits, in combination or in isolation. This report presents the protocol for the Learning About My Pain study, one of the first comparative-effectiveness trials funded by the Patient-Centered Outcomes Research Institute. Adhering to published standards for clinical trials (e.g. Standard Protocol Items: Recommendations for Intervention Trials), it provides an overview of the trial (n = 294), comparing cognitive-behavioral and education pain interventions to usual care, and a detailed description of how its methodology reduces the risks from bias.

Keywords

chronic pain cognitive-behavioral therapy literacy pain education patient-centered research randomized controlled trial socioeconomic status study protocol

Chronic pain is a critical public health problem that affects over 100 million Americans (Institute of Medicine (IOM), 2011). Chronic pain conditions are a heterogeneous mix of diagnoses from back pain to diabetes-related neuropathy. Pain is labeled as chronic when it persists longer than the expected healing time, often operationalized as 3 months (Merskey and Bogduk, 1994). Effective, safe, accessible treatment is elusive despite that it is a major public health priority for the health academy, patients, and the general public (Esposito, 2014; IOM, 2011; Painter, 2014). Evidence suggests that psychological disorders, particularly depression, co-occur often with chronic pain and worsen pain outcomes (Hall et al., 2011). Biomedical treatments such as surgery and medication are costly, sometimes ineffective, and known to carry serious side effects. Psychosocial treatments for chronic pain, particularly cognitive-behavioral therapy (CBT), have been proven efficacious, cost-effective, and low in risk for patients (Eccleston et al., 2009; Thorn et al., 2011; Turk, 2002). Moreover, they provide valuable opportunities for combination therapy when they do not directly compete with medical interventions. However, the effectiveness of standard psychosocial treatments may be seriously jeopardized by individual characteristics such as literacy level, cognitive ability, social support, and sufficient financial resources to gain access to care.

The Learning About My Pain (LAMP) study is a multi-site, behavioral randomized controlled comparative-effectiveness trial of two health literacy–adapted psychosocial treatments for chronic pain, one providing CBT for pain and the other intensive pain education (EDU), in comparison to a usual-care control (full study details are available at http://pmt.ua.edu). The design is a parallel three-group superiority study, testing the benefits of two interventions over medical treatment as usual (TAU) for the management of pain. Using the Pragmascope tool based on the Pragmatic-Explanatory Continuum Indicator Summary (PRECIS) structure (Thorpe et al., 2009; Tosh et al., 2011), the LAMP study is a hybrid pragmatic–explanatory trial (score = 34) that features high controls on practitioner expertise and adherence but low controls on participant entry and retention (see supplementary materials). The study is patient-centered, community-based, and participatory in that it arises from a collaboration between the University of Alabama and Whatley Health Services, Inc. (WHS), a private, 501(c)3 non-profit corporation administering a network of federally qualified health clinics (FQHCs), who partnered to create and conduct a study with direct implications for patient care at the clinic and, by extension, similar settings. Dissemination goals were informed by the Reach Effectiveness Adoption Implementation Maintenance (RE-AIM) Framework (Dzewaltowski et al., 2004). The study was designed to answer the following research questions:

In individuals with chronic pain and socioeconomic disadvantage who are receiving care at an FQHC in the Southern United States, does participating in a health literacy–adapted psychosocial treatment group improve their self-reported pain intensity and interference in physical functioning by the end of treatment better than a group receiving standard medical care, and are these effects maintained 6 months later?

In these same individuals, does participation in the CBT pain management group improve depressive symptoms better than a pain education group by the end of treatment, and are these effects maintained 6 months later?

The interventions arise from 30+ years of research by the Principal Investigator (PI; B.E.T.) investigating the treatment of chronic pain (Thorn et al., 2011; Thorn, 2004). Previous research tailored two existing, effective psychosocial treatments—Group CBT and Group Pain Education (EDU)—for individuals with significant economic challenges and deficits in educational attainment (Kuhajda et al., 2011). Both are hypothesized to lead to improved knowledge of pain and pain self-management, a competency we term Pain Literacy. Furthermore, we hypothesize that CBT may show greater benefit for associated problems such as depression (a common comorbidity, Hall et al., 2011). The project is one of the first selected by the US Patient-Centered Outcomes Research Institute (PCORI; Contract No. 941) and is funded from 16 May 2013 through 15 May 2016.

Increasing evidence for the presence of study publication bias and outcome reporting bias has prompted professional organizations, researchers, journals, and funding bodies to call for publication of research protocols (Chan et al., 2004; Dwan et al., 2008). PCORI is a non-profit, non-governmental funding agency in the United States that operates with a number of research priorities. Among these are efforts to reduce disparities in health care and to bolster the use of more rigorous research methodologies, including greater transparency and consistency in study designs and reporting (PCORI Methodology Committee, 2012). Consistent with this is a requirement that all funded projects register with clinicaltrials.gov (NCT No. 01967342) and publish a study protocol adhering to guidelines published by the Standard Protocol Items: Recommendations for Intervention Trials (SPIRIT) Initiative (Chan et al., 2013) and consistent with the Consolidated Standards of Reporting Trials (CONSORT) guidelines (Boutron et al., 2008). This study and protocol were crafted to meet these aims and approach the highest level of methodological rigor for a behavioral clinical trial (Higgins et al., 2011; Spring et al., 2007).

Method

Participants

Setting

A total of 294 participants with chronic pain will be recruited from the patient population of WHS. With 13 clinics in 7 counties, WHS serves approximately 30,000 patients annually in the region of West Alabama, predominantly ethnic minorities with incomes below the poverty line. Its primary facility is located in Tuscaloosa, Alabama, which will also serve as the main site for the LAMP study. Additional sites will include other clinics in the WHS network. Participants are anticipated to be predominantly women and ~70 percent Black.

Inclusion criteria

Potential participants will have (a) an age of 19 years or greater (the age of consent in Alabama), (b) at least one diagnosis consistent with chronic pain, (c) pain most days of the month for 3 months (but not malignant pain as from cancer or HIV), (d) the ability to speak and understand English, and (e) a telephone or other avenue of communication for contact regarding the study.

Exclusion criteria

Potential participants must not have (a) significant cognitive impairment, as measured by the Short Portable Mental Status Questionnaire (Pfeiffer, 1975), (b) current uncontrolled serious psychological disturbance or active substance abuse, (c) less than minimal literacy skills (operationally defined as the inability to sign their name), (d) major changes in their current pain and psychotropic medication regimen in the last 4 weeks, or (e) current external psychosocial treatments for any pain condition (though they may be receiving psychotherapy for non-pain difficulties). In addition, participants cannot have previously participated in a psychosocial pain treatment study.

Recruitment

Given the patient population’s severe socioeconomic challenges, pronounced disparities, and a regional history promoting mistrust of research participation (Gamble, 1997), the PI committed early to seeking out skilled Participant Coordinators to facilitate communication between the study and this difficult-to-access participant population. These members of the community work at the FQHC’s main location and serve as the primary participant recruiters, communicators, and coordinators, as well as occasional assessors. To enroll the projected sample size of 294 participants, we anticipate approaching 758 persons with chronic pain to recruit 386 eligible persons.

Sample size

The PI’s study in a similar population produced pre-to-post treatment effects in the medium range that are the best available estimates for this study, with the proviso that effect sizes generated from pilot studies tend to be unreliable (Kraemer et al., 2006). A power analysis was completed using G*Power 3 (Faul et al., 2007) for the smallest anticipated between-group difference in treatment effects: the difference between treatment groups for improvements in depressive symptoms. Inputting values of (1 − β) = .80 and α = .05 and a medium interaction effect size of f = .25 (observed in Thorn et al., 2011), we tested a between-group analysis of covariance (ANCOVA) design for the therapeutic effects for the CBT versus EDU groups, including a subgroup analysis of race. The resulting total necessary sample was estimated at 158 subjects, 79 per treatment condition (237 including the TAU control). Allotting 15 percent for attrition (observed in Thorn et al., 2011), the total sample would need to be 279 participants. Thus, with 3 conditions × 14 cohorts × 7 members per group, a sample size of 294 participants (98 per condition) was set and is expected to produce adequate power to analyze subgroup differences in treatment effects for treatment outcomes and to promote the success of randomization for creating equivalent groups, reducing the potential impact of confounding variables.

Interventions

Overview of interventions

All participants will receive standard medical TAU for their pain conditions as provided at WHS (and external providers). The control group will be one-third of participants in a TAU condition receiving usual care, which may include specialty care such as chiropractic or physical therapy. TAU also features parallel study contact during recruitment (e.g. communication with Participant Coordinators), assessments, and intermittent phone contact to conduct administrative tasks such as scheduling and to facilitate participant retention. The other two-third of participants will receive either a cognitive-behavioral group intervention (CBT) or a pain education group intervention (EDU). Both interventions will be administered weekly for 90 minutes over 10 weeks in a closed group format and are the product of an extensive adaptation and refinement process begun in previous research (Kuhajda et al., 2011; Thorn et al., 2011). Participants will receive a CD at each session with a recap of the material covered in their respective condition and, for CBT, audio of a weekly relaxation exercise as well. (See supplemental materials for details about treatment adaptations and intervention content.)

CBT: “Learning About Managing Pain” Group

The CBT intervention is built off the PI’s published and empirically validated group CBT treatment for chronic pain (Thorn, 2004). It provides adapted cognitive-behavioral techniques and motivational reinforcement alongside education about chronic pain, its causes and effects, and pain management skills such as relaxation exercises.

EDU: “Learning About My Pain” Group

The construction of the EDU intervention was informed by existing Pain Education interventions designed for non-disadvantaged populations and the PI’s CBT manual (Ehde et al., 2005; Thorn, 2004). Its format was matched to standard group CBT treatment for pain on session length and frequency, therapist engagement, and pain relevance. Although they share many attributes, EDU is distinguished from CBT by its relative greater depth of information in some areas (e.g. comorbid conditions, health provider communication) and the intentional choice that it not incorporate patient skill-building activities or at-home practice, as does CBT. These characteristics may decrease its general efficacy, but they are also likely to increase its acceptability in our target population and promote its feasibility for a broad range of settings.

Adherence

Treatment receipt will be informally evaluated using verbal discussion about the material from the previous week at the beginning of each session, including participants’ reports about listening to the CDs. Participants will complete weekly post-session measures identifying how easy/difficult they think each session was and how likely they are to return the next week (both on numeric rating scales). For each, they provide a free-response explanation why they chose the score they did. These items will be reviewed and discussed immediately as well as preserved for later analysis.

Interventionists

All intervention providers will be doctoral-level or pre-doctoral-level professionals in the field of clinical psychology. Training will be provided by the PI and another study investigator to assure competency at providing the study interventions and assessments, and additional training will be conducted by the PI and other study personnel to promote cultural competency for all study personnel. All therapists (n = 8) will deliver both interventions and all sessions in a particular cohort.

Intervention fidelity

In order to ensure that all interventions are offered competently and consistent with research guidelines, all sessions will be digitally recorded and 30 percent will be randomly selected by a study investigator for evaluation. Fidelity will be determined with an adapted version of the Cognitive Adherence and Competence Scale (Barber et al., 2003), a measure of adherence, therapist competence, and the quality of the treatment provided. The sessions are rated weekly and any necessary feedback regarding adherence to the treatment manual will be provided in a timely manner. All sessions by a new therapist are reviewed for therapist feedback.

Outcomes

Primary and secondary outcomes

The outcomes selected for this study are important to patients, providers, and researchers. They are consistent with the published Initiative on Methods, Measurements, and Pain Assessment in Clinical Trials (IMMPACT) guidelines and previous studies by the PI (Dworkin et al., 2005, 2008; Thorn et al., 2011; Turk et al., 2003). The primary assessment period of interest is from pre-intervention to post-intervention, and clinically important change will be identified using IMMPACT guidelines (Dworkin et al., 2008).

All three conditions will be compared on the primary outcome, Pain Intensity, assessed using the mean change from baseline in scores on the Brief Pain Inventory (BPI) subscale for pain intensity (Keller et al., 2004). On the BPI, a minimal clinically important change is represented by a 15–20 percent decrease (~1 point), a meaningful change by a 30–36 percent decrease (~2 points), and a substantial change by 50 percent decrease (~4 points). In addition, the distribution of patients in these categories will be reported.

Secondary outcomes include Pain Interference, Depression, and Perceived Change. Pain Interference will be assessed using mean change from baseline in mean scores on the pain interference subscale of the BPI (Keller et al., 2004). A minimal clinically important change was set at 1 point. Depressive symptoms will be assessed using mean change from baseline in total scores on the Patient Health Questionnaire-9 (PHQ-9), a widely used self-report screener for depressive symptoms in primary care (Kroenke et al., 2001). A minimal clinically important change was set at one-half of the standard deviation of all PHQ-9 scores at the pretreatment assessment (Dworkin et al., 2008). Perceived Change will be assessed using the Patient Global Impression of Change (PGIC) tied to several domains (Dworkin et al., 2005). The PGIC is a self-reported point-estimate of perceived change since baseline, providing a patient assessment of improvement or worsening since the study started. The PGIC items assess five domains of interest: (a) Pain Intensity, (b) Pain Interference, (c) Pain Acceptance, (d) Pain Coping, and (e) Negative Thoughts about Pain. For each condition, we will publish the percentages of participants endorsing each of the possible outcomes for worsening, no change, or improvement. Additional exploratory measures may be reviewed by visiting http://pmt.ua.edu. These will allow investigation of differential response to treatment (heterogeneity in treatment effects) on demographic variables, adherence, quality of life, cognitive variables associated with pain, and affective variables.

Procedure

Participant entry

Participants enter the study through contact with the Participant Coordinators after (a) a provider referral from FQHC staff, (b) a targeted contact based on recorded medical diagnoses, or (c) a contact received from posted study information (e.g., flyers). These coordinators will contact and screen prospective participants and schedule those who are eligible for their initial assessment.

Participant timeline

The duration of participant involvement will be 9 months and feature 4 or 14 visits. Participants will be assessed at 4 time points: before the intervention period (pre-intervention assessment (PRE)), at 5 weeks or the midpoint of the intervention period (MID), at 10–14 weeks or the end of the intervention period (post-intervention assessment (POST)), and at the 6-month follow-up (6MO). Between the PRE and POST assessments, participants will engage in one of the two treatment conditions for 10 weeks or continue usual care alone in the TAU comparison condition.

Participant reimbursement

At each assessment, participants will be reimbursed US$45 (US$20 to defray the cost of transportation and US$25 for time and effort expended). Those in the treatment conditions will also receive US$20 per session to defray travel expenses resulting from participation in treatment groups. Thus, total possible reimbursement is US$380 for participants in the intervention groups and US$180 for TAU participants.

Allocation and implementation

Following the PRE assessment, participants will be cluster randomized to one of the three study conditions (TAU, CBT, or EDU) in sequential groups of 7. Randomization will be completed prior to starting the study by the statistical consultant using a computer-generated random-number table. Groups will be randomly assigned to condition with equal allocation (1:1:1) and stratified by site. Target enrollment is 14 groups in each condition (98 participants) for a total of 42 groups or 294 participants. Due to logistical constraints, the main site will host 24 groups, with another 9 groups each held at two other sites.

After 7 participants are assessed, the PI will inform the statistical consultant (who has no direct involvement in study operations) who will provide the group assignment. The PI will inform the group leaders, who will telephone their group members to inform them of assignment and group start-up.

Discontinuation

Attrition is estimated to occur at ~15 percent given the financial and health difficulties faced by our participants. Participants missing the first three groups will be considered non-initiators. Participants may withdraw at any time for any reason from the group treatments or the assessments. Participants who drop out of treatment groups will be asked to complete all major assessments as planned, and medical record data will be collected unless approval is explicitly rescinded. Participants will be monitored for emergency room visits and exacerbations of their pain. Any study-related negative health effects will prompt immediate discussion between the PI and the LAMP Safety Officer to determine implications for the participant or study. Because no negative health effects are anticipated, no stopping criteria are pre-established. However, a safety mechanism is in place to connect participants experiencing suicidal ideation (as identified using the P4 Suicidality Screener) with more intensive psychological care (Dube et al., 2010).

Bias reduction

Techniques for bias reduction in clinical trials are well established and feature prominently in grading systems for clinical trials designs (Higgins and Green, 2008; Jüni et al., 2001). Behavioral clinical trials often demonstrate significant challenges reducing bias from known causes due to the high risk of unmasking around interventions. However, where full implementation is impossible, partially implemented bias reduction techniques can result in rigorous behavioral trial designs that reduce the risks from bias even if they are unable to eliminate them. This trial implemented an ambitious bias reduction strategy that included protections against the largest sources of bias risk identified by the Cochrane Collaboration (Higgins and Green). The strategies used for each bias source are reported in Table 1. Patients are recruited and enrolled using uniform criteria across conditions and are assigned to condition using an a priori allocation table. In addition, group leaders are assigned to both treatment conditions to reduce bias due to group leader experience, and wherever possible, individuals will be masked to study hypotheses.

Table 1.

Strategies for reducing risk of bias in the LAMP study, based on the Cochrane Collaboration’s Tool for Assessing Risk of Bias (Higgins et al., 2011).

Bias domain and definition	Bias source Direction Judgment	Description of LAMP strategy
Selection bias: “Systematic differences between baseline characteristics of the groups that are compared.”	Random sequence generation.	An allocation table based on a random-number sequence was generated using statistical software that stratified assignment by study site (n = 3) and balanced by condition (n = 3). Stratification was intentionally weighted toward the main site (8:3:3).
	Direction uncertain.
	Risk: Low
	Allocation concealment.	The allocation table is held by the statistical consultant, who only unmasks allocation to the PI after all pre-assessments for one cohort are completed.
	Bias toward result.
	Risk: Low
Performance bias: “Systematic differences between groups in the care that is provided, or in exposure to factors other than the interventions of interest.”	Masking of participants.	Participants are well masked to treatment condition (CBT vs. EDU) but cannot be masked to control (TAU). Although this introduces the possibility of bias, this control condition was selected because it provides a meaningful comparator for our clinical partner and funding body (PCORI).
	Bias toward result.
	Risk:
	Low (between treatments).
	High (treatments vs control).
	Masking of personnel.	A culture of masking is maintained. Rigorous masking of treatment-group conditions will be preserved and open discussion of groups or group assignments are interdicted. All therapists, assigned by site, are masked to allocation condition except for their own groups.
	Bias toward result.
	Risk: Low
Detection bias: “Systematic differences between groups in how outcomes are determined.”	Masking of outcome assessment.	All assessors are masked to treatment condition, even for TAU. Masking is particularly important given the subjective outcomes that are standard for measuring pain. Masked status is recorded for each assessment. Post-study analyses will review unmasked assessments for evidence of bias, although none is expected. Masking is necessarily broken by qualitative interviews given at the end of the Post and 6-Month assessments. Multiple assessors are used at each site, and whenever possible, assessors are rotated following unmasking.
	Bias toward result.
	Risk:
	Low (PRE, POST).
	Unclear (6MO).
Attrition bias: “Systematic differences between groups in withdrawals from a study.”	Incomplete outcome data.	All participants will be tracked from first contact through study exit for later reporting. Deliberate techniques have been established to reduce attrition. These include treatment adaptations that increase their acceptability and strategies to reduce barriers to participation (e.g., payments for transportation). All participants wishing to attrite are asked to complete later assessments. Reasons for attrition are collected and reviewed for potential biasing of results.
	Bias toward result.
	Risk: Low
Reporting bias: “Systematic differences between reported and unreported findings.”	Selective reporting.	The LAMP Study hypotheses, primary outcomes, method, and analytic plan have been furnished to PCORI, registered with clinicaltrials.gov, and submitted for publication via this article. Researchers committed a priori to an analytic strategy.
	Bias toward result.
	Risk: Low

LAMP: Learning About My Pain; CBT: cognitive-behavioral group therapy for pain intervention; EDU: pain education group intervention; TAU: treatment as usual control comparison; PCORI: Patient-Centered Outcomes Research Institute; PRE: pre-intervention assessment; POST: post-intervention assessment; 6MO: 6-month-post-intervention assessment.

All definitions taken from Higgins and Green (2008) and Higgins et al. (2011).

Data and analyses

Data collection

Data will be collected at the four time points described above (PRE, MID, POST, and 6MO). All data will be collected by trained assessors masked to treatment condition. All assessment materials will be presented verbally to participants and entered directly into a digital assessment provider using portable tablet computers or recorded on a paper backup.

Measures

The four major assessments include a combination of outcome, process, and exploratory measures. Initial data are collected on literacy (Wide Range Achievement Test-4: Word Reading subtest), health literacy (Short-Test of Functional Health Literacy in Adults), and cognitive function (Digit Span) in addition to repeated measures (Baker et al., 1999; Lezak, 1995; Wilkinson and Robertson, 2006). Outcomes include the BPI: Intensity and BPI: Interference subscales (Keller et al., 2004), the PHQ-9 (Kroenke et al., 2001), and the PGIC (Dworkin et al., 2005). A few measures are provided at weekly group treatment sessions to measure incremental change in pain and pain-related cognitions and to assess session difficulty. Process measures attempt to quantify important common factors such as satisfaction, expectancy, group alliance, and the therapist–patient alliance. Exploratory measures include areas such as health-related quality of life, physical function, physical activity, cognitive function, social support, mood, and pain literacy. Full measure information is available from http://pmt.ua.edu.

Missing data

Methods to reduce lost or missing data include (a) a digital assessment with built-in data validation, range checks, and missing data prompts; (b) Participant Coordinators to facilitate assessment scheduling and participant follow-through; (c) expectation that those who attrite from groups will continue with assessments; (d) timely data verification and follow-up to complete portions of incomplete protocols; and (e) random review of entered data. Participants are provided with study contact information and receive multiple contacts about appointments. When attrition from assessments occurs, information will continue to be collected about pain diagnoses, clinic visit frequency, and medication usage. Finally, the inclusion of a midpoint assessment will allow a more robust multiple-imputation approach to replacement, likely a Markov chain Monte Carlo (MCMC) method (S = 10) accompanied by sensitivity measures, as recommended by the National Research Council (Panel on Handling Missing Data in Clinical Trials and National Research Council, 2010).

Retention and attrition

To improve group retention, extensive efforts were invested in reducing the barriers to engagement with the aim of maximizing intervention accessibility and participant retention. Missed sessions will prompt a follow-up phone call from a group leader to discover the cause of the absence and to provide a brief summary of the session, if desired. Although some degree of participant attrition is unavoidable, steps will be taken to reduce the negative impact of participant discontinuation. Participants who drop out of treatment groups will be contacted to verify their desire to discontinue treatment, and information will be collected to understand (a) what their reason is for discontinuation, (b) who decided that they would discontinue, and (c) whether discontinuation involves some or all types of participation.

Data and safety monitoring

All policies and procedures were reviewed and approved by the Institutional Review Board of the University of Alabama. Any changes to the study protocol will be posted to http://pmt.ua.edu. Participation in the LAMP Study is considered low risk for harms, with the primary risks involving disclosure of participation, exacerbation of existing pain or mood condition, and problematic interpersonal interactions. Where possible, procedures have been instituted to reduce these risks. These include the use of study ID numbers and rigorous prevention of disclosure, regular checks on hospital visits and symptom intensification, and participant exhortation to respect the other members and their privacy. The LAMP Safety Officer will review all procedures to avoid and report harms and all reported safety events. The safety officer will also work with the statistical consultant to identify study-related worsening on pain intensity, depression, or perceived change. If they detect significant change for the worse, they will decide on study stopping and inform the PI.

All data will be coded using unique, individualized study IDs for each participant. Participant names will be stored separately in a secured location. Most data will be entered directly into the Assessment Center platform by trained assessors at the time of the assessment (http://www.assessmentcenter.net). Assessment Center is a highly secure, National Institutes of Health (NIH)–funded, Health Insurance Portability and Accountability Act (HIPAA)–compliant computerized assessment tool. Paper protocols and digital data will be stored in secured university locations, behind at least two keyed locks or a digital firewall, respectively. All communications that include identifying patient information will be digitally encrypted.

Analyses and statistical methods

All analytic procedures were determined with guidance from PCORI Methodology and general best practices, including recommendations made by the IMMPACT group (Dworkin et al., 2008; PCORI Methodology Committee, 2013). Reports will include descriptive analyses that describe characteristics of the sample and individual conditions, such as demographic and pain-related variables, and analyses of treatment acceptability that will include attrition, attendance, and common factors. The core analysis will review treatment effectiveness (Research Question 1) by submitting primary and secondary outcome measures to a three-condition ANCOVA on POST scores, entering baseline PRE scores as covariates. Outcomes are Pain Intensity (BPI; primary), Pain Interference (BPI; secondary), and depressive symptoms (PHQ-9; secondary). Perceived Change (PGIC; secondary), a patient-reported estimate of change, will be analyzed using an analysis of variance (ANOVA) of scores at POST for each of five change domains. For Pain Intensity and Pain Interference, it is anticipated that both treatments will show significant improvement compared to TAU, and they will not significantly differ from each other. For depression scores (Research Question 2), it is anticipated that the CBT pain management treatment will show significant improvement over both other conditions, which may not differ from each other. Standardized mean differences will be reported for all between-treatment comparisons. Follow-up analyses will identify the clinical significance of observed treatment effects, both by using IMMPACT protocols and by comparing scores of much improved or very much improved on the PGIC to scores of no change. A similar analysis will identify symptom worsening. As recommended, final exploratory analyses will attempt to describe potential heterogeneity in treatment effects due to subgroups within the sample and potential moderators or mediators of treatment effects (Kraemer et al., 2002; PCORI Methodology Committee, 2013).

Discussion

This protocol provides a basic overview of the LAMP Study. The study is funded by PCORI due in part to its deep community integration and strong focus on producing results that can be used directly by patients. Building on the PI’s dedication to developing community partners, it integrates community administrators, providers, and members into the study team, which facilitates attention and responsivity to the community. The LAMP Research Advisory Board (established for this study) includes community administrators, providers, and patient representatives to review the study activity and progress periodically. Board recommendations serve to enrich the engagement of the community in the research and the study and provide an extra level of participant protection.

In part because of this strong collaboration, the results of the LAMP study will produce important data to inform patients, providers, administrators, and policymakers about the relative effectiveness, feasibility, and acceptability of these literacy-adapted psychosocial interventions to improve management of chronic pain. A comprehensive bias reduction strategy improves the reliability and future interpretability of the study results and inculcates a rigorous level of methodology and researcher accountability that harkens to calls from leaders in the healthcare field. Given the national crisis that is emerging in pain care, flexible and safe adjunctive treatment options should be an increasing priority, particularly in settings with few available treatment options. The LAMP Study may help meet this need by producing, evaluating, validating, and ultimately, if found effective, making available two low-cost treatment options for individuals and organizations with few financial resources and considerable unmet need.

Footnotes

Acknowledgements

We would like to thank our community partner for the LAMP trial, Whatley Health Services, Inc., particularly Ms Deborah Tucker and Dr Toya Burton; our external collaborators, Dr John Burns and Dr Lisa Campbell; and the members of the UA Behavioral Pain Management Team. Most of all, we wish to thank our community and patient partners, who have been willing to participate in this research, despite their notable personal challenges, in the hopes of helping their peers to have better, healthier, happier lives, less impacted by pain.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: All statements in this report, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute (PCORI), its Board of Governors, or Methodology Committee.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported (in part) through a Patient-Centered Outcomes Research Institute (PCORI) Award (No. 941) ().

References

Baker

Williams

Parker

. (1999) Development of a brief test to measure functional health literacy. Patient Education and Counseling 38(1): 33–42.

Barber

Liese

Abrams

(2003) Development of the cognitive therapy adherence and competence scale. Psychotherapy Research 13(2): 205–221.

Boutron

Moher

Altman

. ; CONSORT Group (2008) Extending the CONSORT statement to randomized trials of nonpharmacologic treatment: Explanation and elaboration. Annals of Internal Medicine 148(4): 295–309.

Chan

Hróbjartsson

Haahr

. (2004) Empirical evidence for selective reporting of outcomes in randomized trials: Comparison of protocols to published articles. JAMA: The Journal of the American Medical Association 291(20): 2457–2465.

Chan

Tetzlaff

Altman

. (2013) SPIRIT 2013 statement: Defining standard protocol items for clinical trials. Annals of Internal Medicine 158(3): 200–207.

Dube

Kurt

Bair

. (2010) The P4 screener: Evaluation of a brief measure for assessing potential suicide risk in 2 randomized effectiveness trials of primary care and oncology patients. Primary Care Companion to the Journal of Clinical Psychiatry 12(6): PCC.10m00978.

Dwan

Altman

Arnaiz

. (2008) Systematic review of the empirical evidence of study publication bias and outcome reporting bias. PLoS One 3(8): e3081.

Dworkin

Turk

Farrar

. (2005) Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain 113(1–2): 9–19.

Dworkin

Turk

Wyrwich

. (2008) Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. The Journal of Pain 9(2): 105–121.

10.

Dzewaltowski

Glasgow

Klesges

. (2004) RE-AIM: Evidence-based standards and a web resource to improve translation of research into practice. Annals of Behavioral Medicine 28(2): 75–80.

11.

Eccleston

Williams

Morley

(2009) Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2: CD007407.

12.

Ehde

Jensen

Engel

. (2005) Education Intervention Therapist Manual: Project II: Role of Catastrophizing in Chronic Pain. Seattle, WA: University of Washington.

13.

Esposito

(2014) Taking your life back from chronic pain. U.S. News & World Report, 15 July. Available at: http://health.usnews.com/health-news/patient-advice/articles/2014/07/15/taking-your-life-back-from-chronic-pain

14.

Faul

Erdfelder

Lang

. (2007) G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods 39(2): 175–191.

15.

Gamble

(1997) Under the shadow of Tuskegee: African Americans and health care. American Journal of Public Health 87(11): 1773–1778.

16.

Hall

Kamper

Maher

. (2011) Symptoms of depression and stress mediate the effect of pain on disability. Pain 152(5): 1044–1051.

17.

Higgins

Green

(2008) Cochrane Handbook for Systematic Reviews of Interventions. Wiley (Online).

18.

Higgins

JPT

Altman

Gøtzsche

. (2011) The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 343: d5928.

19.

Institute of Medicine (IOM) (2011) Relieving Pain in America: A Blueprint for Transforming Prevention, Care, Education, and Research. Washington, DC: The National Academies Press.

20.

Jüni

Altman

Egger

(2001) Assessing the quality of controlled clinical trials. BMJ 323(7303): 42–46.

21.

Keller

Bann

Dodd

. (2004) Validity of the Brief Pain Inventory for use in documenting the outcomes of patients with noncancer pain. The Clinical Journal of Pain 20(5): 309–318.

22.

Kraemer

Mintz

Noda

. (2006) Caution regarding the use of pilot studies to guide power calculations for study proposals. Archives of General Psychiatry 63(5): 484–489.

23.

Kraemer

Wilson

Fairburn

. (2002) Mediators and moderators of treatment effects in randomized clinical trials. Archives of General Psychiatry 59(10): 877–883.

24.

Kroenke

Spitzer

Williams

(2001) The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine 16(9): 606–613.

25.

Kuhajda

Thorn

Gaskins

. (2011) Literacy and cultural adaptations for cognitive behavioral therapy in a rural pain population. Translational Behavioral Medicine 1(2): 216–223.

26.

Lezak

(1995) Neuropsychological Assessment. New York: Oxford University Press.

27.

Merskey

Bogduk

(1994) Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms (2nd edn). Seattle, WA: IASP Press.

28.

Patient-Centered Outcomes Research Institute (PCORI) Methodology Committee (2012) Methodological standards and patient-centeredness in comparative effectiveness research: The PCORI perspective. JAMA: The Journal of the American Medical Association 307(15): 1636–1640.

29.

Painter

(2014) Painkiller prescription rates vary widely among states. USA Today, 1 July.

30.

Panel on Handling Missing Data in Clinical Trials and National Research Council (2010) The Prevention and Treatment of Missing Data in Clinical Trials. Washington, DC: The National Academies Press.

31.

Patient-Centered Outcomes Research Institute (PCORI) Methodology Committee (2013) The PCORI methodology report. Available at: www.pcori.org/research-we-support/research-methodology-standards

32.

Pfeiffer

(1975) A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. Journal of the American Geriatrics Society 23(10): 433–441.

33.

Spring

Pagoto

Knatterud

. (2007) Examination of the analytic quality of behavioral health randomized clinical trials. Journal of Clinical Psychology 63(1): 53–71.

34.

Thorn

(2004) Cognitive Therapy for Chronic Pain: A Step-by-Step Approach. New York: Guilford Publications.

35.

Thorn

Day

Burns

. (2011) Randomized trial of group cognitive-behavioral therapy compared with a pain education control for low-literacy rural people with chronic pain. Pain 152(12): 2710–2720.

36.

Thorpe

Zwarenstein

Oxman

. (2009) A pragmatic-explanatory continuum indicator summary (PRECIS): A tool to help trial designers. Journal of Clinical Epidemiology 62(5): 464–475.

37.

Tosh

Soares-Weiser

Adams

(2011) Pragmatic vs explanatory trials: The Pragmascope tool to help measure differences in protocols of mental health randomized controlled trials. Dialogues in Clinical Neuroscience 13(2): 209–215.

38.

Turk

(2002) Clinical effectiveness and cost-effectiveness of treatments for patients with chronic pain. The Clinical Journal of Pain 18(6): 355–365.

39.

Turk

Dworkin

Allen

. (2003) Core outcome domains for chronic pain clinical trials: IMMPACT recommendations. Pain 106(3): 337–345.

40.

Wilkinson

Robertson

(2006) Wide Range Achievement Test 4 (WRAT4). Lutz, FL: Psychological Assessment and Resources, Inc.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.24 MB