Dataset Information

Revisiting the g-null Paradox.

ABSTRACT: The (noniterative conditional expectation) parametric g-formula is an approach to estimating causal effects of sustained treatment strategies from observational data. An often-cited limitation of the parametric g-formula is the g-null paradox: a phenomenon in which model misspecification in the parametric g-formula is guaranteed in some settings consistent with the conditions that motivate its use (i.e., when identifiability conditions hold and measured time-varying confounders are affected by past treatment). Many users of the parametric g-formula acknowledge the g-null paradox as a limitation when reporting results but still require clarity on its meaning and implications. Here, we revisit the g-null paradox to clarify its role in causal inference studies. In doing so, we present analytic examples and a simulation-based illustration of the bias of parametric g-formula estimates under the conditions associated with this paradox. Our results highlight the importance of avoiding overly parsimonious models for the components of the g-formula when using this method.

SUBMITTER: McGrath S

PROVIDER: S-EPMC9831349 | biostudies-literature | 2022 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Revisiting the g-null Paradox.

McGrath Sean S Young Jessica G JG Hernán Miguel A MA

Epidemiology (Cambridge, Mass.) 20220101 1

The (noniterative conditional expectation) parametric g-formula is an approach to estimating causal effects of sustained treatment strategies from observational data. An often-cited limitation of the parametric g-formula is the g-null paradox: a phenomenon in which model misspecification in the parametric g-formula is guaranteed in some settings consistent with the conditions that motivate its use (i.e., when identifiability conditions hold and measured time-varying confounders are affected by p ...[more]

PMID: 34711734

Similar Datasets

Project description:A paradox is a seemingly absurd or impossible concept, proposition, or theory that is often difficult to understand or explain, sometimes apparently self-contradictory, and yet ultimately correct or true. How is it possible, for example, that oxygen "a toxic environmental poison" could be also indispensable for life (Beckman and Ames Physiol Rev 78(2):547-81, 1998; Stadtman and Berlett Chem Res Toxicol 10(5):485-94, 1997)?: the so-called Oxygen Paradox (Davies and Ursini 1995; Davies Biochem Soc Symp 61:1-31, 1995). How can French people apparently disregard the rule that high dietary intakes of cholesterol and saturated fats (e.g., cheese and paté) will result in an early death from cardiovascular diseases (Renaud and de Lorgeril Lancet 339(8808):1523-6, 1992; Catalgol et al. Front Pharmacol 3:141, 2012; Eisenberg et al. Nat Med 22(12):1428-1438, 2016)?: the so-called, French Paradox. Doubtless, the truth is not a duality and epistemological bias probably generates apparently self-contradictory conclusions. Perhaps nowhere in biology are there so many apparently contradictory views, and even experimental results, affecting human physiology and pathology as in the fields of free radicals and oxidative stress, antioxidants, foods and drinks, and dietary recommendations; this is particularly true when issues such as disease-susceptibility or avoidance, "healthspan," "lifespan," and ageing are involved. Consider, for example, the apparently paradoxical observation that treatment with low doses of a substance that is toxic at high concentrations may actually induce transient adaptations that protect against a subsequent exposure to the same (or similar) toxin. This particular paradox is now mechanistically explained as "Adaptive Homeostasis" (Davies Mol Asp Med 49:1-7, 2016; Pomatto et al. 2017a; Lomeli et al. Clin Sci (Lond) 131(21):2573-2599, 2017; Pomatto and Davies 2017); the non-damaging process by which an apparent toxicant can activate biological signal transduction pathways to increase expression of protective genes, by mechanisms that are completely different from those by which the same agent induces toxicity at high concentrations. In this review, we explore the influences and effects of paradoxes such as the Oxygen Paradox and the French Paradox on the etiology, progression, and outcomes of many of the major human age-related diseases, as well as the basic biological phenomenon of ageing itself.

Dataset Information

Revisiting the g-null Paradox.

Publications

Revisiting the g-null Paradox.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets