Failure of researchers, reviewers, editors, and the media to understand flaws in cancer screening studies: Application to an article in Cancer

Abstract

Observational studies present inferential challenges. These challenges are acute in cancer screening studies, in which lead‐time and length biases are ever present. These biases can make any study worthless. Moreover, a flawed study's impact on the public can be deleterious when its conclusions are publicized by a naïve media. Flawed studies can also make the public learn to be wary of any article or reports of articles claiming to be scientific. Here, the author addresses these and related issues in the context of a study published in Cancer. Cancer 2014;120:2784–2791. © 2014 American Cancer Society.

Socioeconomic inequalities in prostate cancer survival: A review of the evidence and explanatory factors

“…In this context, artificial effects as the lead time and length bias have to be considered while interpreting the results (Auvinen and Karjalainen, 1997;Berry, 2014;de Vries et al, 2010;Dickman and Adami, 2006;Kogevinas and Porta, 1997). Lead time bias increasingly occurred when new screening methods like PSA (prostate specific antigen) testing were introduced.…”

“…Length bias refers to slow-growing tumours that are more likely to be detected by screening programs than fast-growing ones. It suggests that these tumours have an improved survival time as they already had an inherently favourable prognosis quite apart from any advantage of early detection (Berry, 2014;de Vries et al, 2010). In this context, the stage migration bias, also known as the Will Rogers phenomenon, has to be regarded (Albertsen et al, 2005;Dickman and Adami, 2006;Feinstein et al, 1985).…”

Atmospheric fine particulate matter and breast cancer mortality: a population-based cohort study

“…Cancers diagnosed in the screening context are affected by length-time and lead-time bias and may also be less aggressive than those diagnosed outside of screening. 22 We next ran univariate and multivariate Cox proportional hazard models to estimate HRs with 95% CIs of BC death according to quartiles of PM 2.5 exposure. The multivariate model was stratified (separate baseline hazard functions for each variable category within the model) by age, grade, stage, diagnosis period and participation in screening to control for the possible confounding effects of these variables on mortality.…”

Mammography, Martin Yaffe, and Me: Response and Appreciation

“…Yaffe is on to something very important here-namely, the disruptive role of naïve and often downright ignorant media coverage (although, interestingly I should note, some screening critics see the opposite-a pro-screening bias in the popular media-as witness the posture of Dr. Donald Berry 3 ). But the point remains that much of popular media coverage of the mammography debate often borders on the appalling while also being steeped in unresolved potential biases.…”