On behalf of the American College of Obstetricians and Gynecologists (ACOG), the nation’s premier women’s health care membership organization with over 58,000 members, thank you for the opportunity to provide comments on the Centers for Medicare & Medicaid Services’ (CMS) proposed measure specification on Potential Opioid Overuse.

As discussed in ACOG Committee Opinion #711: Opioid Use and Opioid Use Disorder in Pregnancy, the use of opioids in pregnancy has escalated dramatically in recent years and parallels the epidemic seen in the general population throughout the country. Women who use opioids during pregnancy represent a diverse group and it is critically important to recognize and differentiate between opioid use in the context of medical care, opioid misuse, and untreated opioid use disorder. Referral for treatment of pregnant women with opioid use and opioid use disorder are proven to improve both maternal and infant outcomes. The importance of pregnant women receiving adequate treatment for opioid use and opioid use disorder cannot be overstated; inadequate maternal methadone dosage may result in mild to moderate opioid withdrawal signs and symptoms that may cause fetal stress and maternal drug cravings, which increase the likelihood of relapse and treatment discontinuation. If a woman has been treated with a stable methadone dose before pregnancy, pharmacokinetic and physiologic changes that occur during pregnancy may require dose adjustments, especially in the third trimester. Because of metabolic changes in pregnancy, a single daily dosage may not control withdrawal symptoms over a 24-hour period. Rapid metabolism often develops during pregnancy, especially in the third trimester and in these cases, split dosages may be optimal. Due to the unique timing and dosing issues associated with treatment and the dire consequences that stopping or limiting treatment has on a pregnant woman and her developing fetus, we are asking that you exclude pregnant women from the denominator of this measure. Further, we ask that when developing the exclusion criteria, you consider that the use of live birth data alone will not identify all pregnancies. Careful consideration should be made to ensure all pregnant women, regardless out outcome, are excluded.

Additionally, we would like to express concern regarding the ability to accurately parse out the data when attempting to identify buprenorphine use for pain management vs. for treatment, both in pregnant women and in the general population. Obtaining this level of data will be extremely difficult, if not impossible, to do accurately. Failing to recognize buprenorphine use for pain management versus treatment will have significant negative consequences on both the patients receiving this treatment and on their providers. Further, consistently identifying accurate dosage levels for all listed opioids will also be incredibly difficult given the limited availability of consistently accurate prescribing data. Failing to obtain accurate dosage data will affect patients receiving buprenorphine or methadone – both of which are given in doses that may be smaller than 90 mg for periods longer than 90 days. Currently there are data to support that doses between 4 mg to 24 mg a day may be appropriate for buprenorphine stabilization in the non-pregnant, general population. This circumstance is due in part to variability in sublingual absorption of buprenorphine, its subsequent metabolism, and patient response. As with methadone, the primary goal in choosing a stable dose of buprenorphine for a given patient should be to attain a level that suppresses opioid withdrawal effects, and hence, provides the best opportunity to retain the patient in treatment. Failing to accurately identify smaller dosages in patients who are receiving either treatment option for periods longer than 90 days will result in suboptimal performance results for providers who are providing treatment. These results may lead to patients being pulled off their treatment regimen, which will result in even greater consequences.

Lastly, we question the selection of both the 90-day duration and 90 mg dosage as the selected performance thresholds in this measure. Should this measure move forward we would like to see additional data to support these as the selected thresholds.