Higher Clinical Knowledge Scores Linked to Reduction in Opioid Prescriptions

While a shift in guidelines and evidence have led clinicians to be more concerned about opioid use and the impact of the opioid epidemic, data show prescriptions still remained inconsistent during the mid 2000s.

Investigators, led by Bradley M. Gray, PhD, American Board of Internal Medicine (ABIM), compared the association between clinical knowledge and opioid prescribing for Medicare patients’ new low back pain concern between 2009 - 2017.

They found physicians who performed well on ABIM internal medicine board examinations were less likely to prescribe opioids for back pain, compared to physicians who performed less well on the examination.

Study

Investigators performed a cross-sectional study that included 10,246 mid-career general internal medicine physicians in the United States who saw patients who were Medicare beneficiaries with Part D enrollment from 2009 - 2017.

The team compared clinical knowledge with opioid prescribing from 2009 - 2011 when prescribing peaked nationally to 2015 - 2017 after guidelines moved away from opioid prescribing.

Exclusions limited the sample to patients new eligible, new low back pain concern, diagnosis that could be confused as musculoskeletal LBP, and visits with surgical procedures within the prior 14 days and patients with a prior opioid prescription open at the data of visit.

A total of 55,387 visits were collected after exclusion.

Investigators included any opioid prescription and high dosage or long duration (HDLD) opioid prescription (<7 days or <50 daily morphine milligram equivalents), 7 days supplied, or a low-dosage or short-duration (LDSD) prescription if it was a less than or equal to 50 daily MME and less than or equal to a 7-day supply.

The measure of physician knowledge used equated scores from the internal MOC examination.

Further, ABIM examination questions replicated decision-making during real-world clinical scenarios faced by general internal medicine physicians in the outpatient setting.

The exam used a wide range of conditions treated by outpatient general internal medicine physicians. Pain management is not the specific focus of the exam, but it represents a general 2% of the examination.

Investigators used visit-level logit regression to estimate associations between opioid prescribing outcomes and knowledge quartile.

The key regression covariates included indicators for knowledge quartile and interactions with early (2009 - 2011), middle (2012 - 2014), and late (2015 - 2017) period dummies.

Results

Investigators included 55,387 LBP visits in the study, with 67.1% (n = 37185) female patients, 75.8% (n = 41978) White patients, with a mean age of 76.2 years.

Data show physicians in the lowest knowledge quartile answered a substantial number of fewer questions correctly than physicians in the highest quartile (mean examination percent correct, 63.5% versus 84.5%).

The rate of opioid prescribing was 21.6% (n = 11,978) for any opioid prescription and 17.6% (n = 9759) for HDLD prescriptions.

In addition, from 2009 - 2011, visits with physicians in the highest and lowest knowledge quartiles had similar adjusted opioid prescribing rates with 0.5 (95% CI, -1.9 - 3.0) percentage point difference.

Further, from 2015 - 2017, prescribing rates were 4.6 percentage points lower (95% CI, -7.5 to -1.8) for visits with physicians in the highest knowledge quartile prescribed compared to the lowest knowledge quartile (P = .002).

Investigators noted the late period knowledge had a stronger association with HDLD compared to LDSD prescriptions. Adjusted HDLD prescription rates were 4.8 (95% CI, -7.2 to - 2.1) percentage points lower in highest versus lowest knowledge quartiles (P < .001).

Data show it is not statistically significant for LDSD prescriptions (0.0; 95% CI, -1.1 to 1.1 percentage points, P = .98)

In addition, visits when HDLD opioids were prescribed showed a difference in late period when comparing physicians in highest and lowest knowledge quartiles with early period difference of -0.1 (95% CI, -2.4 - 2.2) and a late period difference of 4.8 (95% CI, -7.4 to -2.1 percentage points).

Conclusion

Investigators concluded the standard of care shifted away from routine prescribing of opioids during the late period of 2015 - 2017.

“Physicians who performed well on an ABIM examination assessing current knowledge were less likely to prescribe opioids for back pain than physicians who did not perform as well on the examination,” investigators wrote.

The study, “Clinical Knowledge and Trends in Physicians’ Prescribing of Opioids for New Onset Back Pain, 2009-2017,” was published online in JAMA Network Open.