Estimating the causal effect of zidovudine on CD4 count with a marginal structural model for repeated measures

Abstract

Even in the absence of unmeasured confounding factors or model misspecification, standard methods for estimating the causal effect of a time-varying treatment on the mean of a repeated measures outcome (for example, GEE regression) may be biased when there are time-dependent variables that are simultaneously confounders of the effect of interest and are predicted by previous treatment. In contrast, the recently developed marginal structural models (MSMs) can provide consistent estimates of causal effects when unmeasured confounding and model misspecification are absent. We describe an MSM for repeated measures that parameterizes the marginal means of counterfactual outcomes corresponding to prespecified treatment regimes. The parameters of MSMs are estimated using a new class of estimators - inverse-probability of treatment weighted estimators. We used an MSM to estimate the effect of zidovudine therapy on mean CD4 count among HIV-infected men in the Multicenter AIDS Cohort Study. We estimated a potential expected increase of 5.4 (95 per cent confidence interval -1.8,12.7) CD4 lymphocytes/l per additional study visit while on zidovudine therapy. We also explain the theory and implementation of MSMs for repeated measures data and draw upon a simple example to illustrate the basic ideas.

Deferasirox therapy is associated with reduced mortality risk in a medicare population with myelodysplastic syndromes

“…By contrast, MSM can control for dynamic therapy and health status changes over the observation period [10][11][12]. Several studies have used MSM to estimate the effectiveness of therapies in various clinical settings [10,12,[17][18][33][34][35][36]. Observation of a survival benefit to ICT in a transfusion-requiring MDS cohort in this MSM analysis adds confidence to the conclusion from previous case-control literature that the observed impact of ICT is unlikely to be attributable solely to selection of healthier patients and persistence in treatment.…”

“…The analyses included univariate and bivariate descriptions of the characteristics of patients in the cohort, exposures to therapies and outcomes of interest. MSM were used to examine the contributions of DFX use to the morbidity outcomes of incident AML, CHF, diabetes, hypothyroidism, renal disease and mortality [11,[17][18]. Cases with prior experience of the outcome of interest were excluded from the analysis of that outcome, but not from the mortality analysis.…”

When to Start Treatment? A Systematic Approach to the Comparison of Dynamic Regimes Using Observational Data

“…Like in previous analyses of observational HIV data (Cole et al, 2003;Hernán et al, 2000;Hernán et al, 2002;Sterne et al, 2005) we assumed that treatment was never stopped once initiated. Therefore, for each individual, the factors in the denominator of the weights W t were set to 1 for times t subsequent to treatment initiation, and estimated from the data for all other times, i.e., times when 0.…”

Relationship between Epoetin Alfa Dose and Mortality

“…Such confounding occurs when prognostic factors are markers for therapy and affected by therapy (6,7). EPO dose is titrated in response to hemoglobin concentration, which reflects previous dose and is a prognostic factor (8,9).…”

“…This weighting balances confounding factors across treatment groups. Given specific assumptions, the treatment estimate from a MSM may have the same causal interpretation as an estimate from a randomized clinical trial (7,13).…”