Incremental Effect of the Addition of Prescriber Restrictions on a State Medicaid's Pharmacy-Only Patient Review and Restriction Program.

Abstract

Patient review and restriction programs (PRRPs), used by state Medicaid programs to limit potential abuse and misuse of opioids and related controlled medications, often restrict members to a single pharmacy for controlled medications. While most states use a restricted pharmacy access model, not all states include restricted prescriber access. Oklahoma Medicaid (MOK) added a restricted prescriber access feature to its PRRP in July 2014. To evaluate the incremental effect that the addition of a prescriber restriction to MOK's pharmacy-only PRRP had on the pharmacy and resource utilization of the enrolled members. MOK members with at least 6 months of enrollment in the pharmacy-only PRRP were restricted to a maximum of 3 prescribers for controlled substances in July 2014 and were identified as "cases." Using a propensity score method, cases were matched to controls from the MOK non-PRRP enrolled population based on demographics and baseline health care utilization. Data from January 1, 2014, through December 31, 2014, were evaluated. Each member's monthly health care resource utilization, defined in terms of medical and pharmacy costs, prescription counts, and opioid use per member per month (PMPM), was analyzed. A difference-indifferences (DID) regression estimated the change in resource utilization following the July 2014 policy change. This study included 378 controls and 126 cases after propensity matching. No differences were noted for daily morphine equivalents, benzodiazepine prescriptions, or maintenance prescriptions. There were decreases in mean PMPM use for both groups for short-acting opioid (SAO) claims (P < 0.001), overall opioid claims (P = 0.007 for controls and P < 0.001 for cases), prescribers (P = 0.01 for controls and P < 0.001 for cases), and number of pharmacies for cases (P < 0.001). DID analyses indicated that cases had a larger decrease in mean SAO claims (difference: -0.15, 95% CI: -0.25 to -0.04, P = 0.008); prescribers (difference: -0.25, 95% CI: -0.36 to -0.15, P < 0.001); and pharmacies (difference: -0.20, 95% CI: -0.28 to -0.13, P < 0.001) relative to controls. The difference for overall opioid claims was greater for cases than controls but did not reach statistical significance (difference: -0.12, 95% CI: -0.25 to 0.00, P = 0.050). Although there was no evidence that overall opioid claims were affected, the addition of prescriber restrictions may have resulted in an incremental change to SAO, prescriber, and pharmacy use in the PRPP population. Use of PRRPs may be an effective tool in reducing inappropriate use of prescription opioids within payer systems. The question remains whether these changes result in long-term changes to behavior outside the payer system. Future research into the effects of PRRPs on patient behavior beyond the payer system is needed. No outside funding supported this research. All authors disclose either employment by the Oklahoma Health Care Authority or contractual work for this employer. In addition, Keast discloses unrelated funding through unrestricted research grants from Gilead Sciences and Purdue Pharma. Study concept and design were contributed by Keast and Pham, along with Teel and Nesser. Keast and Pham collected the data, along with Teel, and data interpretation was provided by Keast and Pham, with assistance from Teel and Nesser. The manuscript was written primarily by Keast, along with Pham and Teel, and revised by all the authors.