Reporting dichotomous data using Logistic Regression in Medical Research: The scenario in developing countries
Logistic regression is one of the powerful analytical techniques for use when the outcome variable is dichotomous and the coefficients derived from logistic regression can be interpreted as odds ratio. The analysis under logistic regression gives the β coefficient along with the odds ratio and 95% confidence interval for odds ratio. Since the β coefficients are the natural logarithm of the risk, we have to exponentiate them to get the estimate of risk (odds ratio). Currently, the most widely used statistical methodology for reporting dichotomous data in developed countries is logistic regression analysis. But in developing countries, several research studies with dichotomous data are reported using percentages or cross tabulation method because of the ignorance of the application and knowledge of this methodology among the authors, readers, reviewers and editors. This trend in developing countries should be changed through the use of appropriate statistical methods with the help of professional data analysts, otherwise research cannot investigate vital findings.
Nepal Journal of Epidemiology 2011;1(4):111-113
- Upon acceptance Copyright on any research article is transferred in full to the Confederation of Epidemiological Associations (CEA) and International Nepal Epidemiological Association (INEA). The copyright transfer includes the right to reproduce and distribute the article in any form of reproduction (printing, electronic media or any other form).
- Articles in the Nepal Journal of Epidemiology are Open Access articles published under the Creative Commons CC BY License (https://creativecommons.org/licenses/by/4.0/)
- This license permits use, distribution and reproduction in any medium, provided the original work is properly cited.