Predictors of opioid misuse in patients with chronic pain: a prospective cohort study
Creators: Ives, Timothy J, Chelminski, Paul R, Hammett-Stabler, Catherine A, Malone, Robert M, Perhac, J Stephen, Potisek, Nicholas M, Shilliday, Betsy, DeWalt, Darren A, Pignone, Michael P
- File Type: pdf | Filesize: 311.6 KB
- Date Deposited: 2012-09-05
- Date Created: 2006-04-04
Abstract Background Opioid misuse can complicate chronic pain management, and the non-medical use of opioids is a growing public health problem. The incidence and risk factors for opioid misuse in patients with chronic pain, however, have not been well characterized. We conducted a prospective cohort study to determine the one-year incidence and predictors of opioid misuse among patients enrolled in a chronic pain disease management program within an academic internal medicine practice. Methods One-hundred and ninety-six opioid-treated patients with chronic, non-cancer pain of at least three months duration were monitored for opioid misuse at pre-defined intervals. Opioid misuse was defined as: 1. Negative urine toxicological screen (UTS) for prescribed opioids; 2. UTS positive for opioids or controlled substances not prescribed by our practice; 3. Evidence of procurement of opioids from multiple providers; 4. Diversion of opioids; 5. Prescription forgery; or 6. Stimulants (cocaine or amphetamines) on UTS. Results The mean patient age was 52 years, 55% were male, and 75% were white. Sixty-two of 196 (32%) patients committed opioid misuse. Detection of cocaine or amphetamines on UTS was the most common form of misuse (40.3% of misusers). In bivariate analysis, misusers were more likely than non-misusers to be younger (48 years vs 54 years, p < 0.001), male (59.6% vs. 38%; p = 0.023), have past alcohol abuse (44% vs 23%; p = 0.004), past cocaine abuse (68% vs 21%; p < 0.001), or have a previous drug or DUI conviction (40% vs 11%; p < 0.001%). In multivariate analyses, age, past cocaine abuse (OR, 4.3), drug or DUI conviction (OR, 2.6), and a past alcohol abuse (OR, 2.6) persisted as predictors of misuse. Race, income, education, depression score, disability score, pain score, and literacy were not associated with misuse. No relationship between pain scores and misuse emerged. Conclusion Opioid misuse occurred frequently in chronic pain patients in a pain management program within an academic primary care practice. Patients with a history of alcohol or cocaine abuse and alcohol or drug related convictions should be carefully evaluated and followed for signs of misuse if opioids are prescribed. Structured monitoring for opioid misuse can potentially ensure the appropriate use of opioids in chronic pain management and mitigate adverse public health effects of diversion.