Chronic musculoskeletal pain conditions are among the most common problems seen in primary care. As the importance of these conditions for the health of individual patients and society has been increasingly recognized, use of long-term opioid therapy for chronic musculoskeletal pain has grown exponentially. Research has not kept pace with this change in prescribing practice. Although evidence supports the ability of opioid analgesics to produce short-term reductions in pain intensity, long-term trials evaluating opioid effectiveness are not available. Evidence for effects of opioids on function and quality of life are limited. Furthermore, the long-term safety of opioids is poorly described in the literature.
The main objective of the Strategies for Prescribing Analgesics Comparative Effectiveness (SPACE) Trial was to compare opioid therapy to non-opioid medication therapy for chronic back pain or hip or knee osteoarthritis pain over 12 months. SPACE aimed to compare 1) effects of opioid therapy versus non-opioid medication therapy on pain-related function and pain intensity and 2) adverse effects of opioid therapy versus non-opioid medication therapy. Secondarily, we examined effects of opioid therapy versus non-opioid medication therapy on health-related quality of life, pain sensitivity, and adherence-related behaviors. We also conducted a secondary qualitative analysis to better understand patients' perceptions of their response to the intervention and the value of its components.
This was a pragmatic randomized comparative effectiveness trial with masked outcome assessment comparing opioid therapy versus non-opioid medication therapy over 12 months for patients with chronic back pain or hip or knee osteoarthritis pain. Eligible veterans were seen in primary care for chronic back or hip/knee osteoarthritis pain and had moderate-severe pain at baseline despite analgesic use. Those currently receiving long-term opioid therapy were excluded. After providing informed consent, participants were randomized to opioid therapy or non-opioid medication therapy. Medications in each arm were adjusted to target improvement in pain, while considering individual patient preferences and responses. Interventions were delivered in a care management model using the assigned prescribing strategies, automated symptom monitoring, and a structured decision-making approach to guide medication adjustment. Outcome assessors masked to treatment assignment conducted interviews to assess patient-reported outcomes at 0, 3, 6, 9, and 12 months and assess pain sensitivity, physical performance, and cognitive function at 0, 6, and 12 months. The primary outcome (Aim 1) was the Brief Pain Inventory (BPI) interference scale. The primary adverse effects outcome (Aim 2) was medication-related adverse symptoms, assessed with a patient-reported checklist. The target sample size was 115 completers in each arm, for 80% power to detect a 1 point difference in BPI interference between groups, assuming 2-sided alpha=0.05. Analysis used an intent-to-treat approach, including all participants in the arm to which they were originally assigned.
We enrolled 265 patients and randomized 240. Randomized patients were 86.7% male, 87% white, and had an average age of 58.3 years (SD 13.7), with an age range of 21-80 years. The primary pain diagnosis was back pain for 156 (65%) and hip/knee osteoarthritis for 84 (35%). At baseline, BPI interference was 5.5 (SD 1.9) and BPI severity was 5.4 (SD 1.3). Mental health comorbidity was common; at baseline 53 (22%) screened positive for depression (PHQ-9 score 10), 22 (9%) screened positive for anxiety (GAD-7 score 10), and 50 (21%) screened positive for post-traumatic stress disorder (PC-PTSD score 3).
Two patients withdrew after randomization and primary patient-reported outcomes were collected from 234 (98%) participants. Mean BPI interference, representing pain interference with function, improved over time in both groups and did not differ between groups (p=0.58). BPI interference at 12 months was 3.4 (SD 2.5) in the opioid group versus 3.3 (SD 2.6) in the non-opioid group. BPI severity improved more in the non-opioid group (p=0.03). Mean BPI severity at 12 months was 4.0 (SD 2.0) in the opioid group versus 3.5 (SD 1.9) in the non-opioid group. Patients in the opioid group had significantly more medication-related adverse symptoms than those in the non-opioid group; at 12 months, the mean between-group difference was 0.9 symptoms (95% confidence interval 0.3, 1.5; p=0.03).
This is the first completed randomized controlled trial of long-term opioid therapy for chronic pain. SPACE was designed as a pragmatic comparative effectiveness trial to achieve maximum relevance to clinical decision-making in VA primary care. Results show no long-term advantage of opioid therapy compared with non-opioid medication therapy for patients with back or hip/knee osteoarthritis pain who did not respond to previous analgesics. Opioids did not differ from non-opioids on the primary outcome of pain-related function and were significantly worse than non-opioids for pain intensity and tolerability. Our findings support the CDC Guideline recommendation that non-opioid therapies are preferred for chronic pain.
External Links for this Project
Grant Number: I01HX000671-01A1
- Matthias MS, Donaldson MT, Jensen AC, Krebs EE. "I Was a Little Surprised": Qualitative Insights From Patients Enrolled in a 12-Month Trial Comparing Opioids With Nonopioid Medications for Chronic Musculoskeletal Pain. The journal of pain : official journal of the American Pain Society. 2018 Sep 1; 19(9):1082-1090. [view]
- Chen CX, Kroenke K, Stump T, Kean J, Krebs EE, Bair MJ, Damush T, Monahan PO. Comparative Responsiveness of the PROMIS Pain Interference Short Forms With Legacy Pain Measures: Results From Three Randomized Clinical Trials. The journal of pain : official journal of the American Pain Society. 2019 Jun 1; 20(6):664-675. [view]
- Krebs EE, Jensen AC, Nugent S, DeRonne B, Rutks I, Leverty D, Gravely A, Noorbaloochi S, Bair MJ, Kroenke K. Design, recruitment outcomes, and sample characteristics of the Strategies for Prescribing Analgesics Comparative Effectiveness (SPACE) trial. Contemporary clinical trials. 2017 Nov 1; 62:130-139. [view]
- Kroenke K, Stump TE, Chen CX, Kean J, Bair MJ, Damush TM, Krebs EE, Monahan PO. Minimally important differences and severity thresholds are estimated for the PROMIS depression scales from three randomized clinical trials. Journal of affective disorders. 2020 Apr 1; 266:100-108. [view]
- Koffel E, Kats AM, Kroenke K, Bair MJ, Gravely A, DeRonne B, Donaldson MT, Goldsmith ES, Noorbaloochi S, Krebs EE. Sleep Disturbance Predicts Less Improvement in Pain Outcomes: Secondary Analysis of the SPACE Randomized Clinical Trial. Pain medicine (Malden, Mass.). 2020 Jun 1; 21(6):1162-1167. [view]
- Krebs EE, Gravely A, Noorbaloochi S. Opioids vs Nonopioids for Chronic Back, Hip, or Knee Pain-Reply. [Letter to the Editor]. JAMA. 2018 Aug 7; 320(5):508-509. [view]
- Nugent SM, Krebs EE, Jensen AC, DeRonne BM, Rutks IR, Leverty DM. Automating Symptom Management Monitoring in an RCT: Strategies for Prescribing Analgesics Comparative Effectiveness (SPACE) Trial. Poster session presented at: VA HSR&D / QUERI National Meeting; 2015 Jul 10; Philadelphia, PA. [view]
- Krebs EE, Gravely AA, Jensen AC, Bair M, Kroenke K. Expectations and analgesic preferences among participants in a 12-month trial comparing opioid versus non-opioid analgesics for chronic pain. Poster session presented at: American Academy of Pain Medicine Annual Meeting; 2016 Feb 20; Palm Springs, CA. [view]
Treatment - Efficacy/Effectiveness Clinical Trial, Treatment - Comparative Effectiveness