Executive Summary  Updated June 26, 2008

Why we need a scorecard

The AMSA PharmFree Scorecard 2008 evaluates conflict-of-interest policies at the 151 medical colleges and colleges of osteopathic medicine in the United States. Using letter grades to assess schools’ performance in eleven potential areas of conflict, the Scorecard offers a comprehensive look at the landscape of conflict-of-interest policies across American medical education, as well as more in-depth assessment of individual policies that govern industry interaction with medical school faculty and trainees.

The Scorecard assesses policies related to potential conflicts of interest created by industry marketing at the level of the individual physician and trainee. The Association of American Medical Colleges, which recently proposed strong guidelines for many of these domains, has also addressed institutional and research conflicts of interest in earlier reports.

Methods

AMSA worked with the Prescription Project to develop the PharmFree Scorecard 2008.  Assessed domains were broadly consistent with those identified in recent literature – primarily Brennan et al. Health Industry Practices that Create Conflicts of Interest:  A Policy Proposal for Academic Medical Centers.  JAMA 2006; 295(4): 429-433. A qualitative decision-tree instrument was created to capture key criteria within each domain.

The Scorecard assesses policies related to: acceptance of gifts and meals from industry; consulting relationships; speaking relationships; disclosure of financial conflicts; pharmaceutical samples; individuals with financial conflicts participating in university purchasing decisions; financial support for educational events (on- and off-campus); industry support for scholarships and trainee funds; access of industry sales personnel to the medical school or hospital; and inclusion of education about conflict of interest within the academic curriculum. Additionally, the presence of oversight and sanctions was examined, but not included in grade calculation.

Policies were requested from every U.S. medical school, and follow-up continued until every institution had either responded, formally declined to respond, or failed to respond after at least four contact attempts. The scoring itself was completed by two formally trained independent assessors, blinded to the institution of origin.

The letter grade is calculated based on cumulative domain scores, according to a formula (see methodology). Any institution submitting a formal notification that its policies were currently under review received a grade of In process, or I, which may stand for up to one year. Non-reporting institutions receive a grade of F. An institution may request an explanation or re-assessment of its score if policies change or if it feels it has been scored inappropriately.
 

Summary of results

Of the 151 US medical schools, 8 receive As (5%), 14 Bs (9%), 5 Cs (3%), and 24 Ds (16%). These totals include 11 institutions assessed since the original Scorecard release on May 30, 2008.

52 schools (34%) receive a grade of F. This includes 16 that either submitted policies graded as F or indicated they had no relevant policies, as well as 13 schools that declined to submit policies and 23 that did not respond to repeated attempts at follow-up.

Assigning a score of F to a school that has not submitted policies may overestimate the true prevalence of inadequate policies. Nevertheless, it seems likely that most of the schools that declined to participate are not schools with strong policies.

As of June 26, 2008, 115 of 151 medical institutions had participated in the Scorecard, a 76% participation rate. (Non- participants include 13 schools that declined to submit policies and 23 that have not yet responded.)

36 institutions (24%) received a grade of In Process because they indicated that they are currently revising or creating new policies. This is an important finding: roughly 1 in 4 US medical schools is in the process of developing new, and presumably more stringent, policies.

An additional 12 schools (8%) have been given an In Process designation for pending assessments. This includes 5 new submissions received after the launch date of May 30, 2 late submissions, and 2 requests for reassessment. (Also included are 3 additional I scores, which reflect policies in a missing courier delivery. AMSA is working with the schools on resubmission.)

Trends by domain

The areas that garnered the greatest number of perfect scores were those addressing industry support of scholarships, off-campus education, purchasing and gifts. The areas with fewest perfect scores were site access and disclosure (for which only one AMC received a perfect score).

Scholarships – 35
Off campus education – 28
Purchasing – 26
Gifts – 20
Consulting – 16
Samples – 14
Curriculum – 13
Speaking – 6
On campus education – 5
Site Access – 2
Disclosure – 1

Trends and analysis

Approximately 13 percent of medical students in the U.S. are studying at an A or B school. 

The state of California, a stronghold of clinical care and medical education, fared extremely well on the Scorecard; 3 medical schools in the University of California system received As, reflecting the University's system-wide adoption of strong policies in certain domains (modeled after those first introduced at UC Davis). One UC school and Stanford School of Medicine received Bs. Other results in California included one C, and four Is.

Massachusetts, with four medical schools, presents an uneven picture. Two schools, the University of Massachusetts Medical School (A) and Boston University (B), have strong policies incorporating model language. However, two other medical schools, Tufts and Harvard, do not yet have policies in place. 

New York State, another hub of medical care and education, has 5 non-responders, but a cluster of strong grades, including Mount Sinai School of Medicine (A), Columbia College of Physicians and Surgeons (B), and State University of New York – Upstate (B). Other results included two Ds, one scored F and three Is. The University of Rochester, an I, has submitted new policies fore review. Assessment is pending.

The cluster of Illinois medical schools, some of national renown, fared poorly, receiving 4 Ds, 2 non-responders and five In Process (Northwestern’s Feinberg School of Medicine, University of Chicago Pritzker School of Medicine, University of Illinois College of Medicine at Chicago, University of Illinois College of Medicine at Urbana-Champaign, and University of Illinois College of Medicine at Rockford.
AMSA has recently been made aware that the UIC campus schools are able to set their own policy. Chicago campus has indicated they are In Process, and AMSA is initiating outreach to the Rockford and Urbana-Champaign campuses).

Medical schools in the Potomac region, including Washington, DC, Virginia and Maryland, also fared poorly, with 3 Fs, 3 In Processes and 4 Ds. The regional bright spot is the Uniformed Services University of the Health Sciences in Bethesda, MD, which received an A.

Next steps/assistance

The Scorecard serves not only to measure the strength of policies, but also to provide a valuable resource for institutions to develop and refine new policies. The inclusion of full text policies on the website (where permission has been given) will facilitate this goal.

To help medical students promote policy change, AMSA provides toolkits, talks, and training institutes.

The Prescription Project works with leaders within the profession, as well as policymakers, and offers toolkits that include practical considerations and best-practice policies from leading academic medical centers.

Another resource is the Association of American Medical Colleges (AAMC) task force report on Industry Funding of Medical Education.