ANALISIS REGRESI HAZARD ADITIF DENGAN MODEL LIN DAN YING

• Main Authors: , RAHMASARI NUR AZIZAH, , Dr. Danardono, MPH.
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2013
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/123948/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=64066
• Institution: Universitas Gadjah Mada

Time to event data (survival data) is data of length of time until the event occurs. If the event time is affected by other independent variables, regression analysis can be used to analyze the effects of those independent variables. One of some kinds of regression analysis that can be used is additive hazard regression with Lin and Ying model. In Lin and Ying additive hazard model, the regression coefficients are constants, time-independent. The method that can be used to estimate regression coefficients in this model is similar with maximum partial likelihood method in Cox regression. The estimation of regression coefficients can be obtained from score equation which is obtained from mimicing the score equation from Cox model. Score equation of Cox model is the derrivative of the partial likelihood. In this paper, additive hazard regression analysis with Lin and Ying model is used to analyze some variables that affect the failure of medication of TBC patients in Puskesmas Mantang, Lombok Tengah. Risk Differences are also computed to explain each of the effects of independent variables. It is also presented here the alternative method, hazard regression analysis with Aalen model, which uses the graph of cumulative regression functions to interprete the effects of independent variables. It can be seen that additive hazard regression with Lin and Ying model has advantage in the interpretation of the effects of independent variables, compared to additive hazard regression with Aalen model