American Journal of Epidemiology Vol. 139, No. 9: 963-964
Copyright © 1994 by The Johns Hopkins University School of Hygiene and Public Health
correction |
ERRATUM
ABSTRACTThe Journal has been notified by Dr. Rebecca Lipton of errors that appeared in tables 6 and 7 of a recently published article (1). Odds ratios were incorrectly calculated from regression coefficients for several models of incident non-insulin-dependent diabetes mellitus. The correct data are presented below.
In table 6, the correct odds ratio for race (black vs. white) in the overall model is 5.57, while the odds ratio for race among women is 4.81. The correct odds ratio for a 2-mmHg change in systolic blood pressure among women is 1.02. Among men, the correct odds ratio for baseline age (per 10 years) is 1.37. In table 7, the correct odds ratio for subscapulantriceps skinfold ratio is 1.59 among white men, and the odds ratio for baseline age (per year) is 1.01 among white women. The regression coefficients are correct as shown in both of the tables.
The authors regret any inconvenience this may have caused.