A Semiparametric Mixture Cure Rate Model for Tuberculosis Data

Introduction: One of the essential questions in health research is apatient’s survival. The two primary families of the cure models are themixture cure model and the promotion time cure model. A mixturecure model is a type of survival theory that considers that there areboth susceptible and non-susceptible people in the population understudy, but they will never be exposed to the relevant event.Methods: A total of 1236 patients with pulmonary tuberculosis wereincluded in this work, and time to sputum culture conversion wasthe event of interest. A major comparison was done between thefailure time distribution and cure probability models for treatment,gender, weight, drug susceptibility test, and age between adjusted andunadjusted versions. The R studio version 1.1.447 statistical softwarewas used for all kinds of statistical analyses.Results: Models with two different aspects were compared. Theoutcomes of Mixture Cure Models (MCM) and Promotion Time CureModels (PTCM) were significant, as expected. Estimates for the cureprobability model perform better under MCM than those for the failuretime distribution model. The MCM’s performance has a significantimpact on how well the cure fraction is estimated. It additionallysupports determining the variables that influence the time to sputumconversion.Conclusions: Only the MCM model can distinguish between a treatment’simpact on an event’s timing and occurrence. It was seen that firstly, theresults of the cure probability model appeared differently in relationto those of the pure MCM’s failure time distribution model; secondly,the results of the comparison between MCMs and PTCMs are morefavourable. How to cite this article:Vijai B, Jayashree PR, Ponnuraja C. ASemiparametric Mixture Cure Rate Model forTuberculosis Data. Chettinad Health City Med J.2023;12(4):3-11. DOI: https://doi.org/10.24321/2278.2044.202364