Smoking and lung cancer: recent evidence and a discussion of some questions.

By J. Cornfield, W. Haenszel, Hammond Ec, A. Lilienfeld, M. Shimkin, Wynder El. 1959

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Causal Inference through a Witness Protection Program (2014, methodology): “The linear programming approach of Dawid (2003), inspired by Balke and Pearl (1997) and further refined by Ramsahai (2012), is then modified to generate bounds on the ACE by introducing constraints on some causal paths, motivated as relaxations of Entner et al.”
Identifying Causal Effects With Proxy Variables of an Unmeasured Confounder (2016, background): “The approach is along with the spirit of sensitivity analysis initiated by Cornfield et al. (1959) that relates different causative hypotheses to sensitivity parameters, and assesses their plausibility by checking whether extreme values of the sensitivity parameters are plausible.”
Effects attributable to treatment: Inference in experiments and observational studies with a discrete pivot (2001, background): “An earlier sensitivity analysis by Cornfield et al. (1959) had an important influence on the debate about the effects of smoking on the risk of lung cancer.”
A confounding bridge approach for double negative control inference on causal effects (2018, unknown): “Sensitivity analysis methods (Cornfield et al., 1959; Rosenbaum & Rubin, 1983a) are widely used to evaluate the impact of unmeasured confounding and to assess the robustness of causal inferences, but in general they cannot completely correct for confounding bias.”
War and Wages (2008, methodology): “For alternative models and methods of sensitivity analysis, see, e.g., Copas and Eguchi 2001; Cornfield et al. 1959; Gastwirth 1992; Imbens 2003; Lin, Psaty, and Kronmal 1998; Robins, Rotnitzky, and Scharfstein 1999; Wang and Krieger 2006.)”
Three principles of data science: predictability, computability, and stability (PCS) (2018, background): “Sensitivity analyses test the stability of a causal effect relative to unmeasured confounding (57, 58).”
Good practices for quantitative bias analysis. (2014, background): “As with Cornfield et al.,3 simple bias-sensitivity analyses might suffice to demonstrate robustness or sensitivity of inferences.”
Effect Modification and Design Sensitivity in Observational Studies (2013, background): “The first sensitivity analysis in an observational study was conducted by Cornfield et al. (1959) in their discussion of heavy smoking as a cause of lung cancer, concluding that only very large biases could explain away the observed association as something other than an effect caused by smoking.”
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration (2007, unknown): “famously showed that a relative risk of 9 for cigarette smoking and lung cancer was extremely unlikely to be due to any conceivable confounder, since the confounder would need to be at least nine times as prevalent in smokers as in non-smokers [127].”
A Distributional Approach for Causal Inference Using Propensity Scores (2006, background): “Following Cornfield et al. (1959), a common approach to sensitivity analysis is to investigate the consequences of leaving out a relevant but unmeasured covariate.”
Split Samples and Design Sensitivity in Observational Studies (2009, background): “For other sensitivity analyses, see Cornfield et al. (1959), Breslow and Day (1980, section 2.7), Rosenbaum and Rubin (1983), Gastwirth (1992), Marcus (1997), Lin, Psaty, and Kronmal (1998), Robins, Rotnitzky, and Scharfstein (1999), Copas and Eguchi (2001), and Imbens (2003).”