Citation:

Hilborn, E., S. Deflorio-Barker, G. Smith, M. Murphy, J. Stout, A. Ghio, E. Hudgens, K. Messier, J. Maillard, AND A. Egorov. Geospatial analysis of environmental risks for isolation prevalence of pulmonary nontuberculous mycobacteria in a population cohort. American Geophysical Union, San Francisco, California, December 09 - 13, 2019.

Impact/Purpose:

Nontuberculous mycobacteria (NTM) are a broad group of bacteria that include pathogens of importance to human health. They are transmitted from the environment, and are not transmitted person to person. NTM are difficult to study as infections may be extremely indolent, and the epidemiology and pathogenesis of infection is poorly described. Previous epidemiologic studies implicate drinking water and soil as important exposures associated with disease. We investigate and report associations between human NTM pulmonary isolation and environmental risk factors, including drinking water source. NTM are included on EPA's drinking water CCL.

Description:

The broad group of nontuberculous mycobacteria (NTM) includes pathogens that are transmitted from contaminated water and soil. Pulmonary NTM isolation represents evidence of exposure to environmental contamination. Previous observational studies of pulmonary NTM have identified acidic/hydric (peat) soil and distributed tap water from public water supply systems as potential environmental risk factors. We investigated environmental risk factors for pulmonary isolation using geospatial information among three counties in central North Carolina. We used reports of pulmonary NTM isolation (cases) from clinical laboratories from samples collected during 2006 - 2010. We used land cover and demographic data from: the 2011 National Land Cover Database; the Soil Survey Geographic Database; the National Wetlands Inventory, and the 2010 US Census. We also built a spatial information layer of type of drinking water (distributed public vs. private well) for each residence in the study area. The outcome variable was a count of NTM cases in a census block. We fitted generalized additive models with a two-dimensional smoothing function of geographic coordinates, a population offset term, and sociodemographic covariates to test associations of pulmonary NTM case counts with block level soil and water exposure variables. Of 22,104 census blocks in the study area, 13,495 blocks with both residential population and single-family homes were used in this analysis; they contained 507 geocoded NTM cases with a mean number of 0.04 cases per block. We found a 27% increase (95% Confidence Interval (CI): 1.8%, 58%), in mean number of NTM cases in blocks with at least 20% hydric soils and a 25% (-2.4%, 60%) increase in blocks with at least 50% acidic soils. Blocks with a majority area wetland had a -1.9% (-3.7%; 0.0%) smaller adjusted mean number of cases. We found no association between NTM pulmonary isolation and type of drinking water (distributed public vs. private well). Previous epidemiologic studies have implicated acidic/hydric (peat) soils, treated and distributed drinking water as sources of human NTM exposure. Our work supports the finding of hydric soil exposure risk for NTM isolation, but does not support previously observed associations with a distributed public drinking water. This abstract does not reflect EPA policy.

Record Details: