Updated Estimates of Pharmaceutical Company Payments to Physicians in Vermont

Methods

Our methods have been described previously.¹ Vermont law requires disclosure of payments of $25 or more.² We categorized each payment by recipient and purpose; recipient names were typically not disclosed as part of the settlement. We conducted a descriptive analysis, summarizing all payments over the study period stratified by whether or not they were initially designated trade secret. We focused on payments of $100 or more because these exceed guidelines by the American Medical Association and Pharmaceutical Research and Manufacturers of America for gifts to physicians.^3,4 We used χ² and Brown-Mood 1-way analyses of variance with medians to compare trade-secret– and non–trade-secret–designated payments. Analyses were performed using SAS 9.1 (SAS Institute Inc, Cary, North Carolina). All statistical tests were 2-tailed, using a type I error rate of .002 to account for multiple comparisons.

Results

From July 1, 2002, to June 30, 2004, there were 21 409 payments of any value to all health care professionals and organizations, totaling $4.90 million (median, $52; range, $0.22-$63 458). This sum is $0.69 million less than the $5.59 million calculated by the Vermont attorney general,⁵ a discrepancy accounted for by our exclusion of discount and rebate payments (not required to be disclosed by law) reported by a single company. Of these 21 409 payments, 9182 (42.9%) totaling $2.72 million (55.5%) were designated trade secret. There was a nonsignificant difference between the median trade-secret– and non–trade-secret–designated payments ($53 vs $51; P = .06), although a greater proportion of trade-secret–designated payments were for $100 or more (29.6% vs 25.0%; P < .001).

There were 5539 payments that exceeded $100, totaling $4.20 million (median, $250; range, $100-$63 458) (Table). Of these, 2713 (49.0%) totaling $2.44 million (58.1%) were designated trade secret; the median trade-secret–designated payment was significantly greater than the median non–trade-secret–designated payment ($500 vs $184; P < .001).

Approximately 86% of trade-secret and non–trade-secret–designated payments exceeding $100 were to physicians, comprising 4743 payments totaling $3.20 million (median, $250; range, $100-$20 000). Among these, the median trade-secret–designated payment was significantly greater than the median non–trade-secret–designated payment ($500 vs $177; P < .001) (Table). Trade-secret–designated payments to physicians were more likely to be for education or speakers, and less likely to be for detailing or unspecified purposes, when compared with non–trade-secret–designated payments (P values < .001) (Table), and were less likely to be in the form of food (25.3% vs 67.5%; P < .001).

Among all payments of any value, 30 companies did not designate any trade secrets, 7 designated all trade secrets, and 14 designated some but not all trade secrets. Among payments to physicians that exceeded $100, 46 companies provided 4743, totaling $3.20 million (median number per company, 21; range, 1-973; median total amount, $13 549; range, $116-$994 567).

Comment

Data about nearly half of payments exceeding $100 to physicians from pharmaceutical companies in Vermont, accounting for two-thirds the total value, were obtained through litigation; they were available only to the plaintiff, not the general public. Trade-secret–designated payments were of greater value and many were for food. Currently proposed federal legislation (S.2029 and HR.5605) does not permit trade-secret designation of payments. However, proposed state bills, such as in Washington,⁶ do include trade-secret provisions, which prevent disclosure of substantial numbers of payments. Variation in designation among companies and among payment purposes raises concerns about the appropriateness of trade-secret designation and its practical usage.