Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Use of odds ratio questioned.

Use of odds ratio questioned. The authors are with the Los Angeles County Department of Health Services. Requests for reprints should be sent to Donnell Ewert, MPH, Los Angeles County Department of Health Services, Acute Communicable Disease Control, 313 N Figueroa Street, Room 231, Los Angeles, CA 90012. References 1. Davis RM. Current trends in cigarette advertising and marketing. N Engl J Med. 1987;316:725-732. 2. Brown A. Alcoholic beverages: industry outlooks 1990. Marketing and Media Decisions. 1990;25:34. Use of Odds Ratio Questioned Jaakkola et al.1 report their findings on associations between pollution levels and upper respiratory infections (URIs) in Finnish children. The data are intriguing, but the interpretation suggested by the authors is misleading. In Table 4 on page 1063, the authors report an adjusted odds ratio (OR) of 2.0 among 14- to 18-month-old children in the more polluted area of the polluted city (relative to those in the city's less polluted area) for risk of havmg experienced an upper respiratory infection during the past 12 months. On page 1062, the authors interpret this OR of 2.0 as showing "a 100% increase in the firquency of children with infection." If one looks at the crude data for this same comparison, given in table 2 on page 1062, one draws a different conclusion. The numbers of children with 0 and with 1 + infections are, in the less polluted area, 25 and 151 and, in the more polluted area, 15 and 173. The crude odds ratio from these data is 1.9, shown in Table 4. Actual prevalences of infection in the low and high pollution areas are 151/176 = 0.86, and 173/188 = 0.92. The prevalence ratio is thus 0.92V0.86 = 1.07, which is a 7% increase rather than the 100%o mentioned by Jaakkola et al. Although use of the odds ratio to estimate the risk ratio for cross-sectional data is often appropriate,2 this example illustrates why such an interpretation is generally not appropriate when prevalence of illness is high. Background prevalences for many ORs reported in this paper exceed 85%. In such circumstances, the odds ratio must stand on its own as a measure of effect. El MarkJ. Mendel4 MPH The author is with the Department of Biomedical and Environmental Health Sciences, University of California, Berkeley, and the Office of Environmental and Occupational Epidemiology, California Department of Health Services. Requests for reprints should be sent to Mark J. Mendell, MPH, Office of Environmental and Occupational Epidemiology, California Department of Health Services, 5900 Hollis Street, Suite E, Emeyville, CA 94608. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Public Health American Public Health Association

Use of odds ratio questioned.

M J Mendell
American Journal of Public Health , Volume 82 (6) – Jun 1, 1992
Download PDF
1 page

Loading next page...
 
/lp/american-public-health-association/use-of-odds-ratio-questioned-0lGZvaR28S

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
American Public Health Association
Copyright
Copyright © by the American Public Health Association
ISSN
0090-0036
eISSN
1541-0048
Publisher site
See Article on Publisher Site

Abstract

The authors are with the Los Angeles County Department of Health Services. Requests for reprints should be sent to Donnell Ewert, MPH, Los Angeles County Department of Health Services, Acute Communicable Disease Control, 313 N Figueroa Street, Room 231, Los Angeles, CA 90012. References 1. Davis RM. Current trends in cigarette advertising and marketing. N Engl J Med. 1987;316:725-732. 2. Brown A. Alcoholic beverages: industry outlooks 1990. Marketing and Media Decisions. 1990;25:34. Use of Odds Ratio Questioned Jaakkola et al.1 report their findings on associations between pollution levels and upper respiratory infections (URIs) in Finnish children. The data are intriguing, but the interpretation suggested by the authors is misleading. In Table 4 on page 1063, the authors report an adjusted odds ratio (OR) of 2.0 among 14- to 18-month-old children in the more polluted area of the polluted city (relative to those in the city's less polluted area) for risk of havmg experienced an upper respiratory infection during the past 12 months. On page 1062, the authors interpret this OR of 2.0 as showing "a 100% increase in the firquency of children with infection." If one looks at the crude data for this same comparison, given in table 2 on page 1062, one draws a different conclusion. The numbers of children with 0 and with 1 + infections are, in the less polluted area, 25 and 151 and, in the more polluted area, 15 and 173. The crude odds ratio from these data is 1.9, shown in Table 4. Actual prevalences of infection in the low and high pollution areas are 151/176 = 0.86, and 173/188 = 0.92. The prevalence ratio is thus 0.92V0.86 = 1.07, which is a 7% increase rather than the 100%o mentioned by Jaakkola et al. Although use of the odds ratio to estimate the risk ratio for cross-sectional data is often appropriate,2 this example illustrates why such an interpretation is generally not appropriate when prevalence of illness is high. Background prevalences for many ORs reported in this paper exceed 85%. In such circumstances, the odds ratio must stand on its own as a measure of effect. El MarkJ. Mendel4 MPH The author is with the Department of Biomedical and Environmental Health Sciences, University of California, Berkeley, and the Office of Environmental and Occupational Epidemiology, California Department of Health Services. Requests for reprints should be sent to Mark J. Mendell, MPH, Office of Environmental and Occupational Epidemiology, California Department of Health Services, 5900 Hollis Street, Suite E, Emeyville, CA 94608.

Journal

American Journal of Public Health – American Public Health Association

Published: Jun 1, 1992

There are no references for this article.