Maraviroc and raltegravir were not included in this study as FDA approval for these agents occurred near the end of our evaluation period. For persons starting more than one of the target medications over the study period, the first VHA out-patient prescription

for each target medication was counted. To reduce the number of prescriptions for patients whose care was transferred to the VHA and who were already on a medication of interest, only veterans with out-patient prescription records for at least 1 quarter (90 days) prior to their first prescription for a target medication were included. For each quarter post-approval, we measured the uptake of each medication, defined as the number of new patients with prescriptions for the medication. The start dates for the first quarter for each agent were: atazanavir, selleck June 2003; darunavir, June 2006; tipranavir, June 2005; and lopinavir/ritonavir, September 2000. For each quarter, we determined the number of providers who first wrote a new prescription for one of the four medications and the HM781-36B purchase number of providers who prescribed any antiretroviral. Based on provider type, providers were categorized as physician, physician trainee (student/resident/fellow) or physician extender (nurse/physician assistant/clinical pharmacist).

Clinics where prescriptions were initiated were categorized as infectious disease (ID), primary care or other. VAV2 For each quarter we determined

the cumulative number of facilities that had prescribed each of the target medications. Based on the facility location of the qualifying new out-patient prescription, the prescription was assigned to a Centers for Disease Control and Prevention (CDC) geographical region: Northcentral, Northeast, South or West. To provide a benchmark for the regional uptake of new antiretrovirals, we determined the total number of antiretroviral prescription fills during the period from March 2003 (3 months prior to the earliest approval date for a target medication) to December 2007. For comparison, if the uptake of a new medication matched the prescribing of other antiretrovirals, the percentage of the new prescriptions occurring in a specific region should match the percentage of all antiretroviral fills for that region. For example, if 20% of all antiretroviral fills occurred in the West then one would expect the West to account for 20% of the new prescriptions for a target medication; if the West accounted for >20% of the new prescriptions for a medication, that would indicate that the West had greater uptake of the medication than expected. Medication uptake by region was determined for three time periods: quarters 1 and 2 post FDA approval (period 1), quarters 3–6 post-approval (period 2), and quarters 7+ post-approval until 31 December 2007 (period 3).