Use of laboratory test data to describe histoplasmosis epidemiology in the United States

Histoplasmosis can be a devastating invasive fungal disease caused by Histoplasma capsulatum. Twenty-six states are considered endemic for H. capsulatum, as defined using skin testing data collected from US military recruits between 1958-1969. Today, histoplasmosis surveillance is a notifiable disease in only 11 states, limiting our knowledge of histoplasmosis epidemiology. We explored the use of Histoplasma antigen laboratory test (HALT) data to describe the epidemiology of histoplasmosis. HALT data were extracted from the laboratory information system of ARUP Laboratories for the years 2007, 2008, and 2009. Data were cleaned and analyzed using version 9.2 of the SAS System (SAS Institute Inc, Cary, NC, USA). Positivity proportions and confidence intervals for HALT were calculated using exact binomial methods. We assessed the precision of histoplasmosis proportion estimates for each state using standard error (SE). HALT data were available for 45 states (n=30623) with TX, CA, OH, AR, and KY, constituting >15,000 tests. For 2007-2009, 14, 18, and 20 states respectively provided high resolution (SE<1.0%) estimates of histoplasmosis prevalence. The states were ranked in order of prevalence. Most cases of histoplasmosis consistently occurred in the age group 30-39 years. This study was delimited to HALT data from a single reference laboratory. With complete laboratory data from all labs in the US, HALT data may be a useful source for histoplasmosis surveillance.