OUTCOMES OF AN INSURANCE COMPANY-SPONSORED MULTICHANNEL CHEMISTRY SCREENING INITIATIVE

BACKGROUND. The use of serum chemistry panels as screening tests has b een studied in a variety of clinical and nonclinical settings. None of the studies, however, has attempted to carefully examine any potentia l harm done to participants, and none has measured the impact on healt h-related quality of life. METHODS. Consenting participants in an insu rance company-sponsored screening initiative completed a questionnaire before and 6 months after having blood drawn for a 25-item chemistry panel and a lipid profile; for men older than 50, a prostate-specific antigen (PSA) test was also done. The prescreening questionnaire inclu ded demographic and health information. The postscreening questionnair e included questions about specific outcomes. Included in both questio nnaires were single-item measures of self-rated health and self-rated worry about health, and the 17-item Duke Health Profile (DUKE), a meas ure of health-related quality of life. Various outcomes were examined, including the numbers of new diagnoses, numbers and types of new trea tment recommendations, change in self-reported health and worry, and c hange in DUKE subscale scores. Participants who were potentially helpe d and those who were potentially harmed by the serum chemistry panels screening program were identified and further characterized. RESULTS. Of the 2249 subjects who satisfactorily completed both questionnaires, 2012 (89%) had at least one abnormal test result, but only 985 of the se (49%) remembered having discussed their lest results with a physici an. A total of 342 individuals received new treatment advice. However, 29 (10%) of them indicated that they would be ''somewhat unlikely'' t o ''very unlikely'' to follow it. Following the intervention questionn aire, there were statistically significant average decrements in the G eneral Health, Physical Health, and Pain subscales of the DUKE for par ticipants with abnormal results. Self-rated health status did not chan ge, but level of worry about health increased significantly. At least 250 (11%) subjects were potentially helped by the screening initiative , but at least 574 (26%) were potentially harmed by it. CONCLUSIONS. T he use of serum chemistry panels as screening tests in nonclinical set tings should probably be discouraged, since health-related quality of life is not improved and the intervention may harm more individuals th an it benefits.