Shortening the Current Opioid Misuse Measure (COMM) via computer-based testing: a retrospective proof-of-concept study

An Interview with Stephen F. Butler, PhD

Stephen Butler, PhD, is Senior Vice-President and Chief Science Officer of Inflexxion, Inc. He was trained as a clinical and research psychologist with a background in addiction and scale development. He has been the Principal Investigator for a number of NIDA-funded research grants. Some of this research formed the foundation of NAVIPPRO, a pharmaceutical risk management program that monitors abuse of prescription drugs reported by individuals entering substance abuse treatment at hundreds of treatment centers around the country. As part of other NIDA-funded work, he led the effort, along with Dr. Bob Jamison of Brigham and Women’s Hospital Pain Management Center, to develop, validate, and cross validate the Screener and Opioid Assessment for Patients with Pain (SOAPP® and SOAPP®-R) and the Current Opioid Misuse Measure (COMM)®. The SOAPP and COMM are recommended by treatment guidelines published by the American Pain Society, American Academy of Pain Management, and Canadian national guidelines for the treatment of chronic non-cancer pain.

Elsbeth McSorley: 
Can you explain what the Current Opioid Misuse Measure (COMM) is in more detail for those who are not familiar with it?

Stephen F. Butler, PhD: The COMM helps clinicians identify whether a patient, currently on long-term opioid therapy, may be exhibiting aberrant medication-related behaviors associated with misuse or abuse of opioid medications. Since the COMM examines concurrent (that is, past 30-day) misuse, it is ideal for helping clinicians monitor patients’ aberrant medication-related behaviors over the course of treatment.


EM: Previously, COMM was a paper and pencil questionnaire. Is it now available electronically?

SFB: Yes that is correct. We are very excited to launch PainCAS®, a computerized, systematic assessment of chronic pain patients that will also offer decision support resources to healthcare providers. PainCAS will streamline and standardize the pain assessment process in the clinical setting, taking it off the clipboard and putting it into an electronic format. It will offer clinical resources to healthcare providers, including the validated risk assessment tools SOAPP and COMM, along with risk-level based monitoring recommendations. Patients will receive opioid medication safety information upon completion of the assessment.


EM: That’s very exciting news! So, back to this recent study, can you give some background?

SFB: In 2007, we introduced the COMM to assist clinicians in recognizing aberrant drug-related behavior among respondents who have been prescribed opioids. The COMM is a 17-item self-report questionnaire. Most respondents finish the full-length, pencil and paper version of the COMM in a reasonably short amount of time. We are aware, however, that many settings are pressed for time, so that time spent administering and scoring a screener can be tight. 2013 guidelines for primary care identify 63 different conditions that these providers should screen for. Any given screener should be as time efficient as possible, particularly a screener like COMM, which is intended for ongoing monitoring and repeated administration. Electronic administration using PainCAS may help streamline administration, scoring and generating an easy-to-use printout.

Computer-based administration also gives us other, more powerful options to reduce patient burden. Dr. Matthew Finkelman from Tufts University School of Dental Medicine, proposed using statistical methods called curtailment and stochastic curtailment so that the fewest number of questions are asked to determine whether a patient’s COMM answers reach the cutoff for a positive COMM score—or not. Briefly, while the patient is taking a computerized version of the COMM, the computer continually calculates the score and terminates as soon as the patient is classified as positive or negative (curtailment), or, in the case of stochastic curtailment, is likely to be positive or negative based on some acceptable threshold.


EM: How did you conduct the study?

SFB: This retrospective study used data from a large, two-part study with different samples. A total of 415 patients were divided into a “training” sample of 214 patients that were used to develop the statistical algorithms, and a “test” sample of 201 patients we used to test the sensitivity and specificity of the algorithms. All patients had chronic non-cancer pain, had taken the full-length COMM and were classified as either exhibiting aberrant medication-related behaviors or not using the Aberrant Drug Behavior Index (ADBI).


EM: What were the results?

SFB: Results suggest that both curtailment and stochastic curtailment reduce the respondent burden of the COMM by an average of 22% for curtailment and up to 59% for stochastic curtailment without compromising sensitivity and specificity. Such reductions in screener length are made possible by electronic administration and could be real time-savers in the clinical setting.


EM: Was this study published?

SFB: Yes it was published in BMC Medical Research Methodology on October 20, 2013.


EM: Where can readers go to view the full details of the study?

SFB: The study is published on BMC Medical Research Methodology’s website.