We provide biostatistics expertise for a number of NIEHS and NTP researchers, as well as researchers at universities and other government agencies who are conducting research related to the mission of the NIEHS. In these collaborations, we provide advice about study design, analyze data, interpret statistical results, and participate in writing papers for publication. Over the past year, we have been involved in the five research areas described below. We have an ongoing collaboration with an intramural laboratory that is studying genetic changes and pathways related to fibroids, a common uterine tumor in women. We assisted with the design, statistical analysis and interpretation of studies that investigated gene expression, particularly in response to estrogen, in human uterine tissue cultures. Thus far, our findings suggest that estrogen interacts with the receptor tyrosine kinase pathway to regulate cell proliferation. As part of an NTP interagency agreement, we collaborated with researchers at NIOSH and the University of Cincinnati on a case-control study of the effects of genes and exposure to diisocyanates on the occurrence of occupational asthma. Based on our statistical analyses, we identified several cytokine and antioxidant defense gene polymorphisms that are associated with occupational asthma. The polymorphism having the strongest association is within the superoxide dismutase (SOD2) gene. We collaborated with several NIEHS researchers who focus on mechanisms of DNA repair, mutation, tumorigenesis and receptor activation. By testing for differences between genotype groups, calculating mutation probabilities and/or generating computer simulations of mutations, we found genes and sequences of events involved in DNA repair and activation of the glucocorticoid receptor, and we identified mutation hotspots. We collaborated with NIEHS and extramural researchers who are mining survey data to discover associations between environmental exposures and disease. In particular, we examined the association between tubal ligation and breast cancer risk and found no increased risk. In the statistical analysis of another study, we found that coronary stents coated with a drug were more effective in preventing thrombosis than bare metal stents. Finally, we also evaluated data from two assays. In the first, we examined the stability of bisphenol A concentrations measured in urine specimens across a variety of experimental conditions. The results of these analyses will help determine the procedures that should and should not be used when this assay is used in an epidemiological study. In the second study, the NTP was interested in possibly adopting a new assay to measure mitochondrial respiration, an indicator of cellular health. We analyzed data from this assay to compare mitochondrial toxicities of approximately 15 perfluorinated alkyl compounds and to examine the reliability of results from replicate runs of the assay.
Showing the most recent 10 out of 44 publications
