[Skip to Navigation]
Clinical Observation
August 12/26, 2002

Should We Screen for Hemochromatosis?An Examination of Evidence of Downstream Effects on Morbidity and Mortality

Author Affiliations

From the Department of Family Medicine, Medical University of South Carolina, Charleston (Dr Mainous and Mr Pearson), and the Department of Family and Community Medicine, Christiana Care Health System, Wilmington, Del (Dr Gill).

Arch Intern Med. 2002;162(15):1769-1774. doi:10.1001/archinte.162.15.1769

Background  Population-based hemochromatosis screening has been suggested with the rationale that identification and treatment of subclinical disease would decrease morbidity and mortality due to hemochromatosis.

Objective  To examine the prevalence of elevated serum transferrin saturation levels and the burden of illness of hemochromatosis in terms of ambulatory visits, hospitalizations, and death in the United States.

Participants and Methods  Four nationally representative data sets were used for the analysis of the prevalence of hemochromatosis as well as ambulatory care, hospitalizations, and deaths related to hemochromatosis. Participants included men and nonpregnant women aged 18 years and older in the Third National Health and Nutrition Examination Survey (1988-1994) and the 1996, 1997, and 1998 National Ambulatory Care Survey, National Hospital Discharge Survey, and Underlying Cause-of-Death Mortality Files. The data sets were based on single measurements of serum transferrin saturation levels, serum ferritin levels, and healthcare provider–recorded diagnoses according to the International Classification of Diseases, Ninth Revision, Clinical Modification, code for hemochromatosis.

Results  The prevalence of elevated serum transferrin saturation levels ranged from 1% to 6%. When an elevated serum transferrin saturation level of 55% is combined with an elevated serum ferritin level, the prevalence decreases from 1.9% to 0.65%. The proportion of diagnosed hemochromatosis utilization out of total ambulatory visits, hospitalizations, and deaths is stable across the measures and the 3 years of data ranging from 0.01% to 0.03%. When white men were examined separately, the relationships remained the same as those among the general population of adults.

Conclusions  Although a substantial proportion of adults whose condition is not currently diagnosed would be identified in a population-based screening program for subclinical hemochromatosis, diagnosed morbidity or mortality owing to hemochromatosis is considerably lower than would be expected. Recommendations for screening programs may need to be revisited.

Add or change institution