Primary Care Monitoring of Long-Term Opioid Therapy among Veterans with Chronic Pain

Objective. To characterize long-term opioid prescribing and monitoring practices in primary care. Design. Retrospective medical record review. Setting. Primary care clinics associated with a large Veterans Affairs (VA) medical center. Patients. Adult patients who filled >= 6 prescriptions for opioid medications from the outpatient VA pharmacy between May 1, 2006 and April 30, 2007. Outcome Measures. Indicators of potential opioid misuse, documentation of guideline-recommended opioid-monitoring processes. Results. Ninety-six patients (57%) received a long-acting opioid, 122 (72%) received a short-acting opioid, and 50 (30%) received two different opioids. Indicators of some form of potential opioid misuse were present in the medical records of 55 (33%) patients. Of the seven guideline-recommended opioid-monitoring practices we examined, the mean number documented within 6 months was 1.7 (standard deviation [SD] 1.5). Pain reassessment was the most frequently documented process (N = 105, 52%), and use of an opioid treatment agreement was the least frequent (N = 19, 11%). Patients with indicators of potential opioid misuse had more documented opioid-monitoring processes than those without potential misuse indicators (2.4 vs 1.3, P < 0.001). After adjustment, potential opioid misuse was positively associated with the number of documented guideline-recommended processes (mean = 1.0 additional process, 95% confidence interval [CI] 0.4, 1.5). Conclusions. Guideline-recommended opioid management practices were infrequently documented overall but were documented more often for higher risk patients who had indicators of potential opioid misuse. The relationship between guideline-concordant opioid management and high-quality care has not been established, so our findings should not be interpreted as evidence of poor quality opioid management. Research is needed to determine optimal methods of monitoring opioid therapy in primary care.