Development and Internal Validation of a Multivariable Prediction Model for the Sexual Function of Cervical Cancer Survivors

Background & Objective: Sexual dysfunction is the symptom with the greatest negative impact on the quality of life of cervical cancer survivors. However, the evaluation of sexual function is challenging, and there is ongoing debate concerning its determinants. We aimed to identify predictive variables for sexual function after cervical cancer treatment.Materials & Methods: One hundred and four cervical cancer survivors participated in a prospective cohort study that we carried out employing FACT–Cx v.4.0 and FSFI questionnaires.Results: We developed a beta generalized linear model with a predictive accuracy of 78% (C–index=0.78) and based on vaginal shortening (Cx4) (P=0.077), age (P=0.0002), and ovarian preservation (P=0.01) as risk factors, and functional well–being (P<0.0001) and follow–up duration (P=0.015) as protective factors.Conclusion: To the best of our knowledge, we may have created the first reliable and internally validated prediction model for cervical cancer survivors based on predictors like vaginal shortening, age, ovarian preservation, functional well–being, and follow–up duration that significantly affect female sexual function as targets for potential intervention.