Dataset Information

Parametric g-formula implementations for causal survival analyses.

ABSTRACT: The g-formula can be used to estimate the survival curve under a sustained treatment strategy. Two available estimators of the g-formula are noniterative conditional expectation and iterative conditional expectation. We propose a version of the iterative conditional expectation estimator and describe its procedures for deterministic and random treatment strategies. Also, because little is known about the comparative performance of noniterative and iterative conditional expectation estimators, we explore their relative efficiency via simulation studies. Our simulations show that, in the absence of model misspecification and unmeasured confounding, our proposed iterative conditional expectation estimator and the noniterative conditional expectation estimator are similarly efficient, and that both are at least as efficient as the classical iterative conditional expectation estimator. We describe an application of both noniterative and iterative conditional expectation to answer "when to start" treatment questions using data from the HIV-CAUSAL Collaboration.

SUBMITTER: Wen L

PROVIDER: S-EPMC9044498 | biostudies-literature | 2021 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Parametric g-formula implementations for causal survival analyses.

Wen Lan L Young Jessica G JG Robins James M JM Hernán Miguel A MA

Biometrics 20200706 2

The g-formula can be used to estimate the survival curve under a sustained treatment strategy. Two available estimators of the g-formula are noniterative conditional expectation and iterative conditional expectation. We propose a version of the iterative conditional expectation estimator and describe its procedures for deterministic and random treatment strategies. Also, because little is known about the comparative performance of noniterative and iterative conditional expectation estimators, we ...[more]

PMID: 32588909

Similar Datasets

Project description:IntroductionOccupation can contribute to differences in risk and stage at diagnosis of breast cancer. This study aimed at determining whether occupation, along with skill level and the socio-professional category, affect the breast cancer survival (BCS) up to 10 years after diagnosis.Materials and methodsWe used cancer registry records to identify women diagnosed with primary invasive breast cancer in western Switzerland over the period 1990-2014 and matched them with the Swiss National Cohort. The effect of work-related variables on BCS was assessed using non-parametric and parametric net survival methods.ResultsStudy sample included 8,678 women. In the non-parametric analysis, we observed a statistically significant effect of all work-related variables on BCS. Women in elementary occupations, with low skill level, and in paid employment not classified elsewhere, had the lowest BCS, while professionals, those with the highest skill level and belonging to top management and independent profession category had the highest BCS. The parametric analysis confirmed this pattern. Considering elementary occupations as reference, all occupations but Craft and related trades had a hazard ratio (HR) below 1. Among professionals, technicians and associate professionals, and clerks, the protective effect of occupation was statistically significant and remained unchanged after adjustment for age, calendar period, registry, nationality, and histological type. After adjusting for tumor stage, the HRs increased only slightly, though turned non-significant. The same effect was observed in top management and independent professions and supervisors, low level management and skilled laborers, compared to unskilled employees.ConclusionThese results suggest that work-related factors may affect BCS. Yet, this study was conducted using a limited set of covariates and a relatively small study sample. Therefore, further larger studies are needed for more detailed analyses of at risk occupations and working conditions and assessing the potential interaction between work-related variables and tumor stage.

Dataset Information

Parametric g-formula implementations for causal survival analyses.

Publications

Parametric g-formula implementations for causal survival analyses.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets