Introduction:

the design of the ASI-IMC methodology allows a correct application of the implicative statistical analysis in the studies of causality in health. Then the need arose to validate it.

Objective:

to validate the designed ASI-IMC methodology.

Methods:

a prospective analytical observational study of the case and control type nested in a cohort was carried out.The universe of study was made up of all women over 18 years of age, with a clinical and histological diagnosis of breast cancer, from the province of Santiago de Cuba, treated at the "Conrado Benítez" Oncological Hospital, between 2014 and 2019. Twenty-five assumed prognostic factors were used as covariates. Binary logistic regression was applied after verification of compliance with the required assumptions on a sample of 280 patients at the rate of one control per case, which constituted the same data set to which the statistical analysis was applied, in order to compare the results of both techniques. Regression was considered as the gold standard, for which 14 indicators were estimated: sensitivity, specificity, predictive values, likelihood ratios, diagnostic odds ratio, among others.

Results:

both statistical techniques identified biomarkers as good prognosis factors for breast cancer mortality in the study population, and advanced stage, metastasis, and chemotherapy as poor prognostic factors. The efficacy indicators showed values ​​in favor of the evaluated technique.

Conclusions:

the designed methodology was satisfactorily validated, proving to be effective for the identification of prognostic factors.

        · resumen en Español     · texto en Español     · Español ( pdf )