Advertisement
Abstract| Volume 14, ISSUE 4, SUPPLEMENT , S24, April 2013

Development and psychometric evaluation of an instrument to assess prescribers’ perceptions of opioids - the Clinicians’ Attitudes about Opioids Scale (CAOS)

      Attitudes and policies surrounding the long-term use of opioids for chronic non-cancer pain have vacillated over time. Following recent increases in opioid prescribing, there has been a concomitant escalation in rates of misuse and abuse, as well as concerns regarding the long-term efficacy and safety of opioids, all of which have contributed to many political, regulatory, and clinical debates. The present study was designed to develop a reliable and valid psychometric measure – the Clinicians’ Attitudes about Opioids Scale (CAOS), which may be used to assess physicians’ current and evolving beliefs regarding opioid use in patients with chronic noncancer pain in the rapidly changing opioid climate. We developed the questionnaire in 4 sequential phases with a nationally representative sample of physicians in 10 medical specialties: (1) literature review, (2) focus groups conducted with physicians (N = 14) to refine the content of the questionnaire (3) pilot testing of the candidate items (N = 251 physicians) and then subsequently revision, and (4) the formal survey was administered (N = 1,535), and the stability (test-retest reliability) of the assessment instrument was assessed (N = 250). Finally, a principle component analysis of the resulting 38-item CAOS was performed. Five domains were identified: (1) Impediments and Concerns, (2) Perceived Effectiveness, (3) Schedule II vs. III Opioids, (4) Medical Education, and (5) Tamper Resistant Formulations and Dosing, accounting for 17.74%, 11.56%, 5.61%, 4.70%, and 3.56% of the variance, respectively, with the total variance accounted for being 43.17%. The CAOS was found to be internally consistent (Cronbach’s alpha = 0.87) and demonstrated reasonable stability over a 2-week period (reliability coefficients 0.62-0.79). The strong psychometric properties of this instrument should enable the monitoring of changes in opioid perceptions and practices as policy changes, formulations, and patterns of opioid use evolve, and comparisons across groups of prescribers.