|
|
|
|
|
Orlando Gomez-Marin, PhD, MSc, Lora E. Fleming, MD PhD, William LeBlanc, PhD, David James Lee, PhD, Terry Pitman, BA, and Alberto Juan Caban, BSc. Epidemiology & Public Health and Pediatrics, University of Miami School of Medicine, Highland Professional Building, 1801 NW 9th Ave, Miami, FL 33136, (305) 243-6863, Ogomez@med.miami.edu
Objectives: In occupational epidemiologic research, data on current job are used as surrogate for longest held job and its exposures. Several measures of concordance between current and longest held job have been proposed, but an assessment of their performance has not been published. We evaluated three measures of concordance between current and longest held job in a large, population-based, representative sample of US workers.
Methods:We used combined data from the 1986 National Health Interview Survey (NHIS) Supplement on Longest Held Job and the 1988 NHIS Supplement on Occupational Health, and included over 49,000 workers, age 18 years and over, in occupational groups with an estimated employment of 100,000 persons or more. Three measures of concordance, Burnett's index, Dice’ s Coincidence Index and Cohen’s Kappa, were used and compared in terms of their ability to produce similar rankings for different sets of occupational categories, and different gender, race, and ethnic subgroups.
Results: Comparison of the rankings of the occupations showed highly significant values of Spearman Correlation Coefficients ranging from 0.73 for Dice's and Burnett's and also for Kappa and Burnett's indices (p = 0.006), to 0.96 for Kappa and Dice's index (p < 0.001).
Conclusions: High levels of agreement between current and longest held occupation were observed using three different indices. Although not directly comparable, the Kappa and Dice’s indices performance is almost identical and superior to that of Burnett’s index. Because of its known statistical properties and wide availability in most statistical packages, Kappa is the recommended index.
Learning Objectives:
Related Web page: www.rsmas.miami.edu/groups/niehs/niosh/
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.