Jurnal Biometrika dan Kependudukan
ISSN 2302707X
Vol. 2 / No. 1 / Published : 2013-07
Order : 10, and page :75 - 81
Related with : Scholar Yahoo! Bing
Original Article :
Mixture modeling survival case study of hiv / aids at vct clinic / hospital dr cst. kariadi semarang
Author :
- Sigit Ari Saputro *1
- Soenarnatalina Melaniani*2
- Arief Wibowo*3
- Bambang Wijanarko*4
- Muchlis Achsan Sofro*5
- Departemen Biostatistika dan Kependudukan Fakultas Kesehatan Masyarakat, Universitas Airlangga, Surabaya
- Departemen Biostatistika dan Kependudukan Fakultas Kesehatan Masyarakat, Universitas Airlangga, Surabaya
- Departemen Biostatistika dan Kependudukan Fakultas Kesehatan Masyarakat, Universitas Airlangga, Surabaya
- Jurusan Statistika Fakultas Matematika dan Ilmu Pengetahuan Alam Institut Teknologi Sepuluh Nopember Surabaya
- RSUP Dr. Kariadi Semarang
Abstract :
A cohort study HIV/AIDS was conducted in Kariadi hospital and used simple random sampling method. Theobjective were to determine which factors are associated with survival time of HIV/AIDS. Factors that infl uencewith survival time of HIV/AIDS patient’s included age, gender, education level, working status, marital status,antiretroviral therapy, CD4 counts, opportunistic infections, functional status, stadium and adherence. Thisresearch employed Mixture Survival analysis with cox regression of proportional hazard. This model consist oftwo distribution of survival .They are higher risk sub population and lower risk sub population of HIV infection.The result of cox proportional hazard regression mixture analysis by its population at risk classifi cation showsthat the resulting model for each component risk is different based on percentage of survival time. Analysis ofmultivariate Cox proportional hazards models were constructed for each to evaluate trends in the RR of HIVrelated death. Multivariate cox regression in higher status risk group resulted that education level (HR = 1. 826 ,CI: 1.048–3.182 ), CD4 counts (HR = 0.995 , CI: 0.991–0.999), functional status (HR = 3.063, CI: 1.670–5.617)and adherence (HR = 0.235, CI: 0.127–0.436) have signifi cant with survival time. The other non risk grouprepresented age (HR = 0.903, CI: 0.825–0.988), marital status (HR = 0.031, CI: 0.002–0.575), CD4 counts(HR = 0.992, CI: 0.986–0.999), opportunistic infections (HR = 7.734, CI: 1.477–40.503) and adherence(HR = 0.247, CI: 0.098–0.625) have signifi cant with survival time. Estimation mixture Weibull parameter showsmodel contribution that 96,37% survival time from higher risk sub population. Further research is needed todetermine the other survival modeling why such disparities of subpopulation hazard proportion.
Keyword :
HIV/AIDS, mixture survival, cox proportional hazard,
References :
Collet,(1994) Modelling Survival Data in Medical Research London : Chapman and Hall
Grambsch, T.M.T.A.P.M.,(2007) Modelling Survival Data Extending Cox Model Minnesota USA : Springer
Kalbleisch,(2002) The Statistical Analysis of Failure Time Data London : John Willey Intercience
Kleinbaum, D.G.,(2012) Survival Analysis London : Springer
UNAIDS,(2012) UNAIDS Report on the Global AIDS Epidemic 2012 Geneva : WHO
Archive Article
Cover Media | Content |
---|---|
Volume : 2 / No. : 1 / Pub. : 2013-07 |
|