To the Editor: Dr. Robertson raises important issues that have been addressed consistently by the American Board of Family Practice (ABFP) for more than 20 years. Traditionally, the ABFP has used a “Content Blueprint” that described the substance of the discipline of family practice and defined both the subject areas and the proportion of questions in the ABFP certification and recertification examinations. The ABFP conducted 3 content validity studies during this time to guarantee that the examination was assessing the knowledge necessary to practice the full scope of family medicine.
The first study was conducted for the Board in 1982 by the University of Massachusetts. This was a task analysis that identified the knowledge, skills, and abilities of practicing family physicians. In 1993, a validity study was conducted by researchers at Jefferson Medical College that included surveys of patient mix and clinical experiences, as well as a review of data from the National Ambulatory Medical Care Survey. These results were consistent with the previous data collected on the practices of family physicians, but two major differences were apparent. First, the severity of illness reported by the physicians underscored the breadth and knowledge demanded of family physicians as they managed the advance stages of many diseases. Second, the data documented the role that many family physicians play in the treatment of emergent conditions. Unfortunately, the return rates of these surveys were low, placing the results in question. Therefore, in 1998, the Board asked the same researchers to replicate the study to verify whether the previous results were valid. In 1999, another patient mix study was conducted that produced sound data consistent with the findings from the work performed previously.
Based upon the recommendations from the 1999 study, the Board recently convened a blue ribbon task force of experts representing each of the constituencies of the “family” of family medicine. The American Board of Family Practice’s Examination Committee served as the core for the Blueprint Task Force and was joined by 3 experts from each of the following organizations: the American Academy of Family Physicians, the Association of Departments of Family Medicine, the Association of Family Practice Residency Directors, and the Society of Teachers of Family Medicine. The Blueprint Task Force met in late February 2003 and early June 2003 to work on the project. Before, between, and after these meetings, additional work was done via e-mail.
After much discussion, coupled with study of the design and methodology of recently created “blueprints” from other American Board of Medical Specialties’ Boards, the Task Force reached the conclusion that the structure of the current ABFP Content Blueprint was inadequate to meet the challenges of representing the current content of family practice and prevented a much larger, and better categorized, bank of test items from being developed. Therefore, they created a new blueprint that used a robust, multidimensional database that more than adequately characterized the breadth and depth of our specialty. A sample of recently certified diplomates will soon be used to weight the multiple cells within the blueprint, and it will be presented to the ABFP Board of Directors for approval in October. Once approved, the new blueprint will be used by our item writers to guide their development of test items from many of the evidence-based sources that Dr. Robertson has mentioned in his letter. It will also be posted on our web site so that it can be used by physicians to prepare for our certification and recertification examinations.
I would like to thank Dr. Robertson for raising some very critical questions about the design and content of the ABFP’s examinations. They have provided me with the opportunity to bring the readership up to date with initiatives that the Board has undertaken to guarantee the content validity of its examinations.
Notes
The following letter is a reply to the letter above.