American Journal of Epidemiology Advance Access published online on August 5, 2008
American Journal of Epidemiology, doi:10.1093/aje/kwn164
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Practice of Epidemiology |
Constructing Inverse Probability Weights for Marginal Structural Models
1 Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
2 Department of Epidemiology, Harvard School of Public Health, Boston, MA
3 Harvard–MIT Division of Health Sciences and Technology, Cambridge, MA
Correspondence to Dr. Stephen R. Cole, Department of Epidemiology, School of Public Health, University of North Carolina at Chapel Hill, McGavran-Greenberg Hall, Campus Box 7435, Chapel Hill, NC 27599-7435 (e-mail: cole{at}unc.edu) (present address).
Received for publication January 22, 2008. Accepted for publication May 12, 2008.
The method of inverse probability weighting (henceforth, weighting) can be used to adjust for measured confounding and selection bias under the four assumptions of consistency, exchangeability, positivity, and no misspecification of the model used to estimate weights. In recent years, several published estimates of the effect of time-varying exposures have been based on weighted estimation of the parameters of marginal structural models because, unlike standard statistical methods, weighting can appropriately adjust for measured time-varying confounders affected by prior exposure. As an example, the authors describe the last three assumptions using the change in viral load due to initiation of antiretroviral therapy among 918 human immunodeficiency virus-infected US men and women followed for a median of 5.8 years between 1996 and 2005. The authors describe possible tradeoffs that an epidemiologist may encounter when attempting to make inferences. For instance, a tradeoff between bias and precision is illustrated as a function of the extent to which confounding is controlled. Weight truncation is presented as an informal and easily implemented method to deal with these tradeoffs. Inverse probability weighting provides a powerful methodological tool that may uncover causal effects of exposures that are otherwise obscured. However, as with all methods, diagnostics and sensitivity analyses are essential for proper use.
bias (epidemiology); causality; confounding factors (epidemiology); probability weighting; regression model
Abbreviations: AIDS, acquired immunodeficiency syndrome; HAART, highly active antiretroviral therapy; HIV, human immunodeficiency virus; HIV-1, human immunodeficiency virus type 1
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. R. Cole, L. P. Jacobson, P. C. Tien, L. Kingsley, J. S. Chmiel, and K. Anastos Using Marginal Structural Measurement-Error Models to Estimate the Long-term Effect of Antiretroviral Therapy on Incident AIDS or Death Am. J. Epidemiol., November 24, 2009; (2009) kwp329v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. McWilliams, E. Meara, A. M. Zaslavsky, and J. Z. Ayanian Medicare Spending for Previously Uninsured Adults Ann Intern Med, October 5, 2009; (2009) 0000605-200912010-00149v1. [Abstract] [Full Text]
|
|
|
|
|
|
R. W. Platt, E. F. Schisterman, and S. R. Cole Time-modified Confounding Am. J. Epidemiol., September 15, 2009; 170(6): 687 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
What Happens to Your Manuscript: Characteristics of Papers Published in Volumes 167 and 168 Am. J. Epidemiol., May 15, 2009; 169(10): 1275 - 1276. [Full Text] [PDF]
|
|