PMH57 - Inverse Correlation of Decreased Opioid Utilization and Increased Medical Assisted Therapy in U.S. Medicaid Managed Care

Abstract

To evaluate opioid and medication-assisted treatment (MAT) utilization in Medicaid managed care plans in six states and the District of Columbia. A post-hoc analysis evaluated opioid and MAT claim utilization (per thousand members [PK]) and expenditures (per member per month) from AmeriHealth Caritas’ claim database over a 24-month period (October 2015 to September 2017). Pearson’s correlation analyses were performed between MAT and opioid use with one-month lag periods. All analyses were performed with SAS EG 7.1 software. AmeriHealth Caritas paid out 2,695,303 distinct opioid claims (N=561,682 members) over the study period; 26,019 members with ≥24 claims and 389 prescribers accounted for 32.4% and 20.0% of all opioid claims, respectively. Members with multiple claims in high-risk strata included ≤18 year-olds; expectant moms; “Triple-Threat” users (opioid+benzodiazepine+carisoprodol); and moderate users. Opioid utilization and expenditures declined by 28.2% and 28.9%, respectively, while individual member claims declined by 18.3% (4.49% to 3.67%). For opioid-based pain management, decreases were observed in hydrocodone-acetaminophen (−19.4%), oxycodone-acetaminophen (−38.0%), tramadol (−19.5%), oxycodone (−30.3%), and acetaminophen-codeine #3 (−31.3%). For MAT-administered opioid addiction treatment, brand buprenorphine-naloxone decreased (−70.7%), while increases were observed in generic buprenorphine-naloxone (+903.0%) and buprenorphine (+68.8%), and the opioid receptor antagonists injectable naltrexone (+208.3%) and naloxone (+197.5% in 2017). Despite a 9.1% increase in MAT claims PK, MAT expenditures declined by −20.5% PK, correlating with decreased brand-to-generic buprenorphine-naloxone utilization (r=0.82, p<0.001). An inverse cross-correlation was observed between MAT and opioid use (r=−0.74, p<0.001). Inter-MAT and inter-opioid analysis show high correlations between current and prior-month’s use (r=0.56 and 0.77, respectively). Pain-related service utilization and opioid-related emergency department visits PK both decreased slightly over the study period. Fewer member claims for opioid pain management medications caused declines in opioid utilization, inversely correlated with increased MAT utilization for treating opioid addiction. Decreased brand-versus-generic buprenorphine-naloxone utilization was also correlated with lower opioid expenditures.