Testing a Program to Increase Patient Activation among Patients Prescribed Opioid Medicine for Chronic Pain

BACKGROUND: Prescription opioid use can present challenges for patients with chronic pain as well as for health systems. These patients typically have multiple chronic conditions and many points of contact with the health system. They can feel disempowered and report concerns about undertreatment of pain, difficulties in obtaining medication, and stigma. Use of prescription opioids is now a controversial approach to managing pain long term, with the prescribing environment becoming more restrictive. At the same time, patients may lack the skills and knowledge to talk to their physicians about their pain and opioid use, and to navigate the health care system. As patients reduce their opioid use, whether voluntarily or involuntarily, physicians must find innovative and effective ways to support these patients and improve their quality of life and function. OBJECTIVES: This study aimed to develop a patient-centered intervention curriculum and compare the effectiveness of the intervention on patient-centered outcomes among patients receiving long-term opioid therapy. The goal of the intervention was to empower patients and move them toward a greater partnership role in their health care team. Our hypotheses were that patients in the intervention would increase their activation; improve self-management of pain; improve pain, function, and quality of life; decrease prescription opioid use; decrease utilization of acute services; and report improved satisfaction and communication with physicians. METHODS: In partnership with patient, clinical, and health system stakeholders, we conducted a pragmatic, randomized trial to examine the effectiveness of a 4-session, 90-minute group-based patient activation intervention for patients receiving long-term opioid therapy in 2 large primary care clinics in an integrated health care system. We randomized 376 patients to either the intervention arm (n = 189) or usual care-only arm (n = 187). The intervention curriculum covered patient activation and empowerment; risks of prescription opioid use; how to talk to physicians about prescription opioid use; and self-management strategies for pain, including how to navigate the health care system and the online patient portal. We developed the study questions, intervention curriculum, patient-centered outcome measures, and dissemination plan with our patient partners and other stakeholders. We conducted interviews at baseline, 6 moths, and 12 months. The primary outcome was patient activation. We also analyzed many secondary outcomes, including overall health, quality of life, depression, pain severity, function, patient-provider communication, patient satisfaction, prescription opioid use, pain management strategies, and health care utilization. We analyzed questionnaire and electronic health record data using bivariate comparisons and repeated-measures, mixed-effects regression models. RESULTS: Among the 376 enrolled patients, 354 (94%) and 342 (91%) completed the 6-month and 12-month follow-up assessments, respectively. We found no effect of the intervention on patient activation scores (estimate of the interaction of intervention and time in the mixed-effects model = 0.17; 95% CI, −1.54 to 1.88; P = .85). For the secondary outcomes, bivariate results indicated that participants in the intervention arm, compared with usual care, reported significantly higher overall health scores at 6 and 12 months (P = .001 and .007, respectively), as well as function scores at 12 months (P = .02, everyday physical activity; P = .01, social activity and roles), and lower prevalence of moderate to severe depression at 12 months (P = .02). Participants in the intervention also demonstrated more of a decrease in the prevalence of depression over time compared with the usual care arm (estimate of the interaction of intervention and time in mixed-effects model = −0.56; 95% CI, −1.02 to −0.10; P = .02). Participants in the intervention arm were more likely to use pain management strategies such as mindfulness and meditation at 6 and 12 months (<0.001) and exercise at 12 months (<0.001). The intervention group demonstrated a greater increase in the use of exercise for pain management compared with the usual care arm over the 3 time points (estimate of the interaction of intervention and time in mixed-effects model = 0.44; 95% CI, 0.07-0.81; P = .02). Participants in the intervention arm were more likely to use the online portal's health and wellness resources at 6 and 12 months (P = .02 and < .001, respectively), and the trend was significantly higher for the intervention arm over time (estimate of the interaction of intervention and time in mixed-effects model = 0.41; 95% CI, 0.02-0.79; P = .04). We found no effect of the intervention on quality of life, pain severity, patient-provider communication, patient satisfaction, health care utilization, or prescription opioid use. CONCLUSIONS: The intervention did not have an observable effect on patient activation. However, we found modest effects on select secondary outcomes that reflect increased self-care and greater engagement with the health care system. These findings suggest that a limited intervention focused on patient activation and nonpharmacological pain management set in primary care may be an effective adjunct to treatment for patients with chronic pain. LIMITATIONS: Changes in the prescribing environment necessitated expanding study eligibility criteria from patients starting regular opioids to patients taking long-term opioids. Recruitment overall was lower than anticipated, resulting in a modest sample size and potentially limiting our ability to identify smaller effects. Multiple comparisons also require caution in interpreting findings.