The occurrence of Simpson's paradox if site-level effect was ignored in the TREAT Asia HIV Observational Database
| Autor: | P.L. Lim, C. Ezhilarasi, Aizobelle Huelgas, Thira Sirisanthana, N. Han, Tuti Parwati Merati, W.W. Wong, Wilai Kotarathititum, P. Kambua, F. Yuliana, Anchalee Avihingsanon, R. David, M.G. Law, W. Lam, Sirinya Teeraananchai, David C Boettiger, D.T.H. Nguyen, S Pujari, W.W. Ku, Mahiran Mustafa, Iskandar Azwa, R. Bantique, Evy Yunihastuti, Adeeba Kamarulzaman, R. Sriondee, Elenore Judy B. Uy, Shinichi Oka, P.C.K. Li, B.L.H. Sim, T. Nishijima, Y.M. Gani, Alvina Widhani, H.L. Ha, Vohith Khol, R. Martinez-Vega, Ezhilarasi Chandrasekaran, Nicolas Durier, P. Chusut, Winai Ratanasuwan, Sasisopin Kiertiburanakul, Suneeta Saghayam, N. Nordin, N. Sanmeema, Junko Tanuma, Dewa Nyoman Wirawan, Sineenart Taecharoenkul, H.X. Zhao, Somnuek Sungkanuparph, P.C. Wu, R. Ditangco, Sharifah Faridah Syed Omar, K.V. Nguyen, T.T. Pham, Matthew Law, S. Gaikwad, D.D. Cuong, M.P. Lee, O.T. Ng, B. Petersen, Jutarat Praparattanapan, N. Kumarasamy, Gang Wan, D. Imran, Awachana Jiamsakul, D.T. Nguyen, K Ruxrungtham, Kedar Joshi, J.M. Kim, Sasheela Ponnampalavanar, A.H. Sohn, F.J. Zhang, J.Y. Choi, Stephen J. Kerr, A. Chitalikar, Y.T. Chan, Praphan Phanuphak, Penh Sun Ly, L. Chumla, Pacharee Kantipong, S. Na, Romanee Chaiwarith, V.H. Bui, David A. Cooper, L.S. Lee, A. Jiamsakul, P.S. Ly |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Adult
Male 0301 basic medicine Cart Asia Biomedical Research Databases Factual Anti-HIV Agents Epidemiology HIV Infections Article Simpson's paradox Cohort Studies 03 medical and health sciences 0302 clinical medicine Statistics Humans Medicine 030212 general & internal medicine Mortality Survival analysis Aged Proportional Hazards Models Aged 80 and over Proportional hazards model business.industry Hazard ratio Fixed effects model Middle Aged Survival Analysis 030112 virology Research Design Data Interpretation Statistical Cohort Female business Forecasting Demography Cohort study |
| Zdroj: | Journal of Clinical Epidemiology. 76:183-192 |
| ISSN: | 0895-4356 |
| DOI: | 10.1016/j.jclinepi.2016.01.030 |
| Popis: | Objectives In multisite human immunodeficiency virus (HIV) observational cohorts, clustering of observations often occurs within sites. Ignoring clustering may lead to “Simpson's paradox” (SP) where the trend observed in the aggregated data is reversed when the groups are separated. This study aimed to investigate the SP in an Asian HIV cohort and the effects of site-level adjustment through various Cox regression models. Study Design and Setting Survival time from combination antiretroviral therapy (cART) initiation was analyzed using four Cox models: (1) no site adjustment; (2) site as a fixed effect; (3) stratification through site; and (4) shared frailty on site. Results A total of 6,454 patients were included from 23 sites in Asia. SP was evident in the year of cART initiation variable. Model (1) shows the hazard ratio (HR) for years 2010–2014 was higher than the HR for 2006–2009, compared to 2003–2005 (HR = 0.68 vs. 0.61). Models (2)–(4) consistently implied greater improvement in survival for those who initiated in 2010–2014 than 2006–2009 contrasting findings from model (1). The effects of other significant covariates on survival were similar across four models. Conclusions Ignoring site can lead to SP causing reversal of treatment effects. Greater emphasis should be made to include site in survival models when possible. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect