Assessing Opioid Abuse in Chronic Pain Patients: Further Validation of the Prescription Opioid Misuse Index (POMI) Using Item Response Theory

Abstract

AbstractDue the limitations of the previous validations, the purpose of this study was to further validate the Prescription Opioid Misuse Index (POMI) in a larger sample of chronic non-cancer pain (CNCP) patients and to examine differential item functioning (DIF) across sex. Participants (n=225 CNCP patients under long-term opioid therapy) completed patient characteristics, self-reported POMI and DSM-5 prescription opioid use disorder measurements. Reliability and factor structure were assessed using both item response theory and classical test theory. ROC curve analysis was used to establish the optimum cut-off score for detecting the presence of DSM-5 prescription opioid-use disorder. Concurrent validity was also tested. The POMI showed a unidimensional factor structure and acceptable internal consistency (ωcat =0.62). DIF analysis showed that males and females respond similarly to each item of the POMI, supporting unbiased measurement of the latent trait across both groups. A cut-off point of 2 is suggested in order to maximize the accuracy of the instrument as a first-screening tool for opioid misuse (AUC=0.78; p<0.001; CI 95%: 0.72–0.85). Concurrent validity of the POMI was high with DSM-5 moderate to severe opioid-use disorder criteria (OR=7.824, p<0.001). These results indicate that the POMI is a valid and clinically feasible screening instrument for detecting CNCP patients who misuse opioid medications. The short length of the scale could meet the needs of clinical practice as it allows clinicians to precisely identify and monitor prescription opioid misuse in both male and female patients.