Multiple Imputation to Account for Measurement Error in Marginal Structural Models.

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Marginal structural models are an important tool for observational studies. These models typically assume that variables are measured without error. We describe a method to account for differential and nondifferential measurement error in a marginal structural model.

Related collections

Most cited references 35

Record: found
Abstract: not found
Article: not found

Marginal Structural Models and Causal Inference in Epidemiology

James M. Robins, Babette A. Brumback, Miguel Ángel Hernán (2000)

0 comments Cited 565 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: not found
Article: not found

Marginal Structural Models to Estimate the Joint Causal Effect of Nonrandomized Treatments

Babette A. Brumback, James M. Robins, Miguel A. Hernán (2001)

0 comments Cited 83 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

When to initiate combined antiretroviral therapy to reduce mortality and AIDS-defining illness in HIV-infected persons in developed countries: an observational study.

Lauren E. Cain, Roger Logan, James M. Robins … (2011)

Most clinical guidelines recommend that AIDS-free, HIV-infected persons with CD4 cell counts below 0.350 × 10(9) cells/L initiate combined antiretroviral therapy (cART), but the optimal CD4 cell count at which cART should be initiated remains a matter of debate. To identify the optimal CD4 cell count at which cART should be initiated. Prospective observational data from the HIV-CAUSAL Collaboration and dynamic marginal structural models were used to compare cART initiation strategies for CD4 thresholds between 0.200 and 0.500 × 10(9) cells/L. HIV clinics in Europe and the Veterans Health Administration system in the United States. 20, 971 HIV-infected, therapy-naive persons with baseline CD4 cell counts at or above 0.500 × 10(9) cells/L and no previous AIDS-defining illnesses, of whom 8392 had a CD4 cell count that decreased into the range of 0.200 to 0.499 × 10(9) cells/L and were included in the analysis. Hazard ratios and survival proportions for all-cause mortality and a combined end point of AIDS-defining illness or death. Compared with initiating cART at the CD4 cell count threshold of 0.500 × 10(9) cells/L, the mortality hazard ratio was 1.01 (95% CI, 0.84 to 1.22) for the 0.350 threshold and 1.20 (CI, 0.97 to 1.48) for the 0.200 threshold. The corresponding hazard ratios were 1.38 (CI, 1.23 to 1.56) and 1.90 (CI, 1.67 to 2.15), respectively, for the combined end point of AIDS-defining illness or death. CD4 cell count at cART initiation was not randomized. Residual confounding may exist. Initiation of cART at a threshold CD4 count of 0.500 × 10(9) cells/L increases AIDS-free survival. However, mortality did not vary substantially with the use of CD4 thresholds between 0.300 and 0.500 × 10(9) cells/L.