SURVEY OF PHYSICIAN PRACTICE BEHAVIORS RELATED TO DIABETES-MELLITUS IN THE UNITED-STATES 1. DESIGN AND METHODS

OBJECTIVE - To conduct a survey among a representative sample of prima ry care physicians in the U.S. to assess practice behaviors, treatment goals, and beliefs related to management of diabetes mellitus and pre vention of its complications. RESEARCH DESIGN AND METHODS - A mail sur vey with telephone follow-up was conducted among 3481 primary care phy sicians in active practice in the continental U.S. A stratified probab ility sample was selected using the files of the American Medical Asso ciation and American Osteopathic Association. Four specialties were se lected to be included in the study: family physician, general practiti oner, internist, and pediatrician. Two versions of a questionnaire wer e constructed: one for pediatricians containing questions about IDDM o nly and one for the other three specialties containing questions about both IDDM and NIDDM. Physicians who were not actively engaged in prac tice or did not see patients with diabetes were excluded. RESULTS - Co mpleted questionnaires were received from 1502 of 3481 sampled physici ans. Based on various assumptions of eligibility among nonresponders, an overall response rate to the survey was estimated to be between 65. 7 and 86.5%. Discrepancies between specialty identifications as noted on the American Medical Association/American Osteopathic Association f iles and as self-designated were noted. CONCLUSIONS - This report desc ribes the methodology used in the design and conduct of the survey, an d data are provided to document the technical success of survey execut ion. This report provides the methodological basis for a series of sep arate reports on demographic characteristics of the physicians, their practices and their patients, and on specific attitudes, beliefs, and practice behaviors of primary care physicians in the U.S. with regard to diabetes mellitus.