• Jonathan D Ketcham and Andrew J Epstein.
• School of Health Management and Policy, W.P. Carey School of Business, Arizona State University, Tempe, AZ 85287-4506, USA. ketcham@asu.edu
• Pharmacoeconomics. 2006 Jan 1; 24 Suppl 3: 27-40.

BackgroundTo limit quickly rising prescription drug expenditures, many state Medicaid programmes have implemented preferred drug lists (PDLs) and prior authorisation (PA). Lessons from Medicaid efforts may be informative for Medicare, which started covering outpatient prescription drugs recently.ObjectivesTo examine how the cost of compliance with Medicaid PDLs for antihypertensives and statins varied across physicians, and to assess the effects of PDLs on physician prescribing patterns and access for Medicaid patients.Data And MethodsAn anonymous survey of primary care physicians and cardiologists in nine states with PDLs was conducted in December 2005 and January 2006. Survey responses were augmented with physician prescribing data, practice location characteristics, and publicly available information about state PDLs. We analysed six physician-level outcome measures: annual PDL-related costs; the proportion of Medicaid prescriptions covered by the PDL; the proportion of Medicaid prescriptions written for an alternative to a physician's most preferred drug because of PDLs; the proportion of times no drug was prescribed to a Medicaid patient because of PDLs; whether they restricted their new Medicaid patient caseload because of PDLs; and whether they decreased the proportion of prescriptions not covered by the Medicaid PDL for non-Medicaid patients. We assessed how these outcomes varied with Medicaid caseload, physician practice size, and socioeconomic characteristics of the practice's ZIP Code.ResultsCosts from complying with Medicaid PDLs for statins and antihypertensives were greatest for physicians in solo practices with high Medicaid caseloads located in poor areas. Although all physicians' prescribing patterns were influenced to some extent by PDLs, those with high volumes of Medicaid prescriptions were affected more. They more frequently prescribed Medicaid patients a less-preferred medication or nothing at all, and were more likely to alter their prescribing to non-Medicaid patients. Physicians with low Medicaid prescription volumes in areas with fewer minorities were more likely to limit their willingness to treat new Medicaid patients.ConclusionThe burden of Medicaid PDLs fell greatest on physicians in disadvantaged areas and their patients.

24 keywords...

hide…