BACKGROUND: Primary care providers manage most patients with chronic pain. Pain is a complex problem, particularly in underserved populations. A technology-enabled, point-of-care decision support tool may improve pain management outcomes.
METHODS: We created an electronic health record (EHR)-based decision support tool, the Pain Management Support System-Primary Care (PMSS-PC), and studied the tool-plus-education in 6 Federally Qualified Health Center practices using a randomized, wait-list controlled design. The PMSS-PC generated "best practice alerts," gave clinicians access to a pain assessment template, psychological distress and substance use measures, guidelines for drug and non-drug therapies, and facilitated referrals. Practices were randomly assigned to early vs delayed (after 6 months) implementation of the intervention, including technical support and 6 webinars. The primary outcome was change in worst pain intensity scores after 6 months, assessed on the Brief Pain Inventory-Short Form. Changes in outcomes were compared between the practices using linear multilevel modeling. The EHR provided clinician data on PMSS-PC utilization.
RESULTS: The 256 patients in the early implementation practices had significantly improved worst pain (standardized effect size [ES] = -.32) compared with the 272 patients in the delayed implementation practices (ES = -.11). There was very low clinician uptake of the intervention in both conditions.
CONCLUSIONS: Early implementation of the PMSS-PC improved worst pain, but this effect cannot be attributed to clinician use of the tool. Further PMSS-PC development is not indicated, but practice-level interventions can improve pain, and studies are needed to identify the determinants of change.
The improvement in "worst pain" measures may be attributed to the Hawthorne effect on the part of the patient, as clinicians weren't actively adopting the tool.
The study design is reasonable, but there is a significant subjective component. Also the physician opportunity to bypass the the intervention in the EMR could create bias.
As an internist, managing pain properly is important. The study does show improvement with time, but not due to the intervention. The use of decision aids in EHR sounds promising, but I'm not sure why they did not work!