Publication:
The occurrence of Simpson's paradox if site-level effect was ignored in the TREAT Asia HIV Observational Database

Issued Date

2016-08-01

Resource Type

Article

ISSN

18785921
08954356

DOI

10.1016/j.jclinepi.2016.01.030

Other identifier(s)

2-s2.0-84959216697

Rights

Mahidol University

Rights Holder(s)

SCOPUS

Bibliographic Citation

Journal of Clinical Epidemiology. Vol.76, (2016), 183-192

Suggested Citation

Awachana Jiamsakul, Stephen J. Kerr, Ezhilarasi Chandrasekaran, Aizobelle Huelgas, Sineenart Taecharoenkul, Sirinya Teeraananchai, Gang Wan, Penh Sun Ly, Sasisopin Kiertiburanakul, Matthew Law, P. S. Ly, V. Khol, F. J. Zhang, H. X. Zhao, N. Han, M. P. Lee, P. C.K. Li, W. Lam, Y. T. Chan, N. Kumarasamy, S. Saghayam, C. Ezhilarasi, S. Pujari, K. Joshi, S. Gaikwad, A. Chitalikar, T. P. Merati, D. N. Wirawan, F. Yuliana, E. Yunihastuti, D. Imran, A. Widhani, S. Oka, J. Tanuma, T. Nishijima, J. Y. Choi, S. Na, J. M. Kim, B. L.H. Sim, Y. M. Gani, R. David, A. Kamarulzaman, S. F. Syed Omar, S. Ponnampalavanar, I. Azwa, M. Mustafa, N. Nordin, R. Ditangco, E. Uy, R. Bantique, W. W. Wong, W. W. Ku, P. C. Wu, O. T. Ng, P. L. Lim, L. S. Lee, R. Martinez-Vega, P. Phanuphak, K. Ruxrungtham, A. Avihingsanon, P. Chusut, S. Sungkanuparph, L. Chumla, N. Sanmeema, R. Chaiwarith, T. Sirisanthana, W. Kotarathititum, J. Praparattanapan, P. Kantipong, P. Kambua, W. Ratanasuwan, R. Sriondee, K. V. Nguyen, V. H. Bui, D. T.H. Nguyen, D. T. Nguyen, T. T. Pham, D. D. Cuong, H. L. Ha, B. Petersen, D. A. Cooper, M. G. Law, D. C. Boettiger The occurrence of Simpson's paradox if site-level effect was ignored in the TREAT Asia HIV Observational Database. Journal of Clinical Epidemiology. Vol.76, (2016), 183-192. doi:10.1016/j.jclinepi.2016.01.030 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/41271

Research Projects

Organizational Units

Authors

Journal Issue

Thesis

Title

The occurrence of Simpson's paradox if site-level effect was ignored in the TREAT Asia HIV Observational Database

Author(s)

Awachana Jiamsakul
 Stephen J. Kerr
 Ezhilarasi Chandrasekaran
 Aizobelle Huelgas
 Sineenart Taecharoenkul
 Sirinya Teeraananchai
 Gang Wan
 Penh Sun Ly
 Sasisopin Kiertiburanakul
 Matthew Law
 P. S. Ly
 V. Khol
 F. J. Zhang
 H. X. Zhao
 N. Han
 M. P. Lee
 P. C.K. Li
 W. Lam
 Y. T. Chan
 N. Kumarasamy
 S. Saghayam
 C. Ezhilarasi
 S. Pujari
 K. Joshi
 S. Gaikwad
 A. Chitalikar
 T. P. Merati
 D. N. Wirawan
 F. Yuliana
 E. Yunihastuti
 D. Imran
 A. Widhani
 S. Oka
 J. Tanuma
 T. Nishijima
 J. Y. Choi
 S. Na
 J. M. Kim
 B. L.H. Sim
 Y. M. Gani
 R. David
 A. Kamarulzaman
 S. F. Syed Omar
 S. Ponnampalavanar
 I. Azwa
 M. Mustafa
 N. Nordin
 R. Ditangco
 E. Uy
 R. Bantique
 W. W. Wong
 W. W. Ku
 P. C. Wu
 O. T. Ng
 P. L. Lim
 L. S. Lee
 R. Martinez-Vega
 P. Phanuphak
 K. Ruxrungtham
 A. Avihingsanon
 P. Chusut
 S. Sungkanuparph
 L. Chumla
 N. Sanmeema
 R. Chaiwarith
 T. Sirisanthana
 W. Kotarathititum
 J. Praparattanapan
 P. Kantipong
 P. Kambua
 W. Ratanasuwan
 R. Sriondee
 K. V. Nguyen
 V. H. Bui
 D. T.H. Nguyen
 D. T. Nguyen
 T. T. Pham
 D. D. Cuong
 H. L. Ha
 B. Petersen
 D. A. Cooper
 M. G. Law
 D. C. Boettiger

Other Contributor(s)

University of New South Wales (UNSW) Australia
 The HIV Netherlands Australia Thailand Research Collaboration
 Academic Medical Centre, University of Amsterdam
 VHS Medical Centre India
 Gokila
 Research Institute for Health Sciences
 Beijing Ditan Hospital
 National Center for HIV/AIDS
 Mahidol University

Abstract

© 2016 Elsevier Inc. 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.

Keyword(s)

Medicine

Availability

URI

https://repository.li.mahidol.ac.th/handle/20.500.14594/41271

Collections

Scopus 2016-2017