American Journal of Epidemiology Vol. 126, No. 6: 1220
Copyright © 1987 by The Johns Hopkins University School of Hygiene and Public Health
correction |
ERRATA
They say the error does not affect in any way the main findings of the study but wished to alert readers to this minor problem. As stated on page 1002 of the report, the authors believed they were basing nutrient index calculations on a "usual serving size," which was the average serving size reported by the entire group for each food. Due to a computer programming error, 100 gm was instead the serving size assigned to each food. For many foods, 100 gm was similar to the usual serving size, but for other foods (e.g., butter and margarine), this was a substantial overestimate.The authors have re-computed all data based on average serving size rather than 100 gm units and find all of the relations presented in the paper to be essentially unchanged. They say it may be important for some readers to note, however, that the mean monthly intakes as presented in table 4, being based on 100 gm serving size units, are inflated estimates of actual intake, particularly for fat. The relations between the original, recalled. and current reports of intake are nonetheless still the same. Most importantly, the correlations (as presented in table 5) asd the patterns of categorical cross-classification (as presented in figure 1) show the same relations. The main finding of the study, that a retrospective dietary history may be a better indicator of true past diet than is the report of current diet, is therefore not affected by this minor error.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Van Looveren, G. Daube, L. De Zutter, J.-M. Dumont, C. Lammens, M. Wijdooghe, P. Vandamme, M. Jouret, M. Cornelis, and H. Goossens Antimicrobial susceptibilities of Campylobacter strains isolated from food animals in Belgium J. Antimicrob. Chemother., August 1, 2001; 48(2): 235 - 240. [Abstract] [Full Text] [PDF]
|
|