A revised measure of symptom-specific health care use.

The objective was to determine physicians' ratings of the clinical importance of common adult symptoms and to use these ratings to develop a new measure of symptom-specific use of health care services. We developed a list of common symptoms using the National Ambulatory Medical Care survey. We surveyed a random sample of internists, family physicians, general practitioners and emergency medicine physicians from the American Medical Association's Physician Masterfile and asked them to rate symptoms' seriousness, impact on quality of life and urgency of need for medical attention. The symptoms' prevalences were determined in a general population survey (National Access to Care Survey). Eleven items were classified as "serious" (median seriousness ratings of 7 or higher on scale from 1 to 10); most of these also were judged to have high impact on quality of life. Another 12 items that were not judged "serious" had median quality of life impact scores of 7 or higher and were classified as "morbid". Sixteen items did not meet criteria for either "serious" or "morbid" symptoms and were classified as "intermediate". Six other items had median seriousness and quality of life impact scores of 3 or less and were classified as "benign". A total of 24 of these items (7 "serious", 8 "morbid", 8 "intermediate" and 1 "benign") were selected to form the symptom-response measure. In the national survey, 26.3% reported one or more serious symptom, 30.6% reported one or more morbid symptom and 29.6% reported one or more intermediate symptom. This new instrument expands on earlier "symptom-response" measures by including a larger number and a broader spectrum of symptoms. This measure should be useful to assess differences in patterns of health care use for particularly serious or morbid conditions; such variations may indicate problems with access to medical care.