Statistical modeling is commonly used in both predictive and explanatory studies in health research. Its use in Cuba continues to grow, although it is sometimes employed inappropriately, which can lead to errors that imperil validity. This article attempts to shed light on faulty practices in statistical modeling by examining and discussing the main differences between explanatory and predictive models, with reference to the following: study objectives, theoretical considerations in model-building, aspects requiring assessment, variable and algorithm selection, analysis of confounders, treatment of multicollinearity, and reporting results.

KEYWORDS Prognosis, risk factor, protective factor, causality, statistical models, linear models, predictive models, explanatory models, logistic regression, Cuba