My research has primarily focused on nutrition and cancer epidemiology. Main themes of my research have been determination of the role of plant-based diets, particularly fruit and vegetable intake, alcohol consumption, and more recently, vitamin D in the development of both common and less common cancers. Much of my research has been conducted in two international consortia that I have been leading/co-leading for over 16 years: the Pooling Project of Prospective Studies of Diet and Cancer (DCPP) and the Circulating Biomarkers and Breast and Colorectal Cancer Consortium (BBC3). Additionally, I have participated in several analyses of dietary factors including dietary patterns, anthropometric factors, and biomarkers within the Nurses' Health Study, Nurses' Health Study II, and Health Professionals Follow-Up Study, three large prospective cohort studies.
The Pooling Project of Prospective Studies of Diet and Cancer (DCPP) is an international consortium that was established in 1993 to examine associations between dietary and anthropometric factors and breast cancer risk. Since then, we have also evaluated associations with several other cancers including lung, colorectal, renal cell, ovarian, pancreatic, and prostate cancer, as well as amyotrophic lateral sclerosis and coronary heart disease. It provides a large collection of prospective data from which large-scale, comprehensive assessments of the role of nutrition (nutrients, foods, and anthropometry) in the risk of several cancers and other chronic diseases have been investigated. The consortium now includes over 35 cohort studies with over 2.5 million participants from 16 countries.
The Circulating Biomarkers and Breast and Colorectal Cancer Consortium (BBC3) was established in 2010 to examine associations between circulating 25(OH)D and risk of breast and colorectal cancers, overall and by tumor subtype and population subgroup. These initial analyses included 21 cohorts from North America, Europe, and Asia. The consortium is currently being expanded to examine additional biomarkers for these two cancers.
For both consortia, participant level data on the exposures of interest, confounding variables, and outcomes from each study have been centralized and harmonized and analyses have been conducted using standardized criteria across studies.