Lydia Lera; Bárbara Leyton & Pablo A Lizana

Logistic regression is a statistical technique widely used in biomedical research to model the relationship between independent variables and a categorical dependent variable. Unlike linear regression, this methodology is based on the logistic function, which allows predicting probabilities between 0 and 1 without requiring strict assumptions such as normality, homoscedasticity, or linearity. One of its main strengths is the interpretation of the coefficients in terms of odds ratio (OR), which allows quantification of the strength of association between exposure and outcome. Logistic regression is a robust and flexible tool in biomedical studies, provided its application is rigorous and accompanied by an adequate contextual interpretation of the results. The correct model specification and the researcher's experience are key elements to guarantee the validity of the inferences. This paper presents the concepts, applications, and key logistic regression examples, an important tool in biomedical research. The objective is to facilitate the understanding and correct application of this technique in various studies, thus generating solid and relevant scientific evidence for decision-making in the health and biological sciences.

KEY WORDS: Odds Ratio; Logistic models; Epidemiological methods; Confidence intervals.

LERA, L.; LEYTON, B. & LIZANA, P. A. Logistic regression and its application in biomedical research. Int. J. Morphol, 43(5):1545-1552, 2025.