Expert Ratings of NHANES Occupational Categories

Research Trainee: Jennifer Chang D’Souza, PhD Student, Dept. of Environmental Health Sciences at the University of Michigan

Faculty Researchers: Alfred Franzblau, MD, Professor of Environmental Health Sciences at the University of Michigan and W. Monroe Keyserling, PhD, Professor of Industrial and Operations Engineering at the University of Michigan

Certain occupational activities such as stair climbing, heavy lifting and squatting have been frequently associated with knee osteoarthritis (OA), and are considered established risk factors for knee OA. However, the exposure data in these epidemiologic studies are frequently dichotomized, self-reported or based on job titles (Cooper, 1994; Maetzel et al, 1997; Schouten et al, 2002). These exposure assessment methods often do not take into account previous work history, do not represent a range of exposures and thus do not adequately characterize occupational exposures. These inadequacies represent a challenge in further describing the relationship between distinct occupational exposures and the later development of knee osteoarthritis (i.e., dose-response relationships, threshold effects), as well as developing effective interventions. An additional challenge in studying knee OA is the expense and difficulty in obtaining knee radiographs to validly assess the outcome. Anderson et al (1986) analyzed the association between knee radiographs and current occupation in The First National Health and Nutrition Examination Survey (NHANES I), which was a cross-sectional survey of the nation’s health. Job categories were characterized by averaging the ratings performed by the Department of Labor (DOL) and published in the Dictionary of Occupational Titles (DOT). However the knee bending requirement variable was dichotomous and did not differentiate between various knee bending activities (e.g. stooping, kneeling, crouching or crawling). The strength variable was rated on a 5-point scale, but was dichotomized. The NHANES III is the most recent survey and contains important work history variables not included in NHANES I and bilateral knee radiographs. However, examining the relationship between the 40 occupation categories in the NHANES III and knee OA is limited by the lack of sufficient exposure data for the job categories. In order to further describe these occupation categories beyond dichotomous variables used by Anderson et al (1986), the proposed project aims to develop job ratings for the occupation categories based on the consensus of ergonomic experts (aka Delphi survey). Ergonomic experts would individually rate the job categories on various physical variables associated with knee OA, and then come to a consensus rating through discussion with other experts. The hypotheses of the study address the following areas: 1) between- and within-expert rating variability of occupational categories 2) comparisons with the occupational ratings performed by the Department of Labor (DOL) and published in the “O*NET Analyst” database.


Publications resulting from this project:
D’Souza JC, Keyserling WM, Werner RA, Gillespie B, Franzblau A. Expert consensus ratings of job categories from the Third National Health and Nutrition Examination Survey (NHANES III). Am J Ind Med. 2007;50(8):608-616. doi:10.1002/ajim.20487.

D’Souza JC, Werner RA, Keyserling WM, et al. Analysis of the Third National Health and Nutrition Examination Survey (NHANES III) using expert ratings of job categories. Am J Ind Med. 2008;51(1):37-46. doi:10.1002/ajim.20512.


Research trainee’s current position:
Jennifer D’Souza completed her PhD in 2006 and is currently a Research Analyst in the Dept. of Epidemiology at the University of Michigan.