Errors In Methods

Abstract

In ReplyThe starting point for our study was the promotion of cancer screening as lifesaving-a powerful claim. However, our study findings showed that this statement is not backed by trial evidence. As we stated in the conclusion of our metaanalysis, we do not advocate for abandoning all screening; rather, we find it important to call for moderation and factbased information instead of promising a longer life with cancer screening. 1 We do not believe it is informative to speculate whether lifetime gains would become evident with longer follow-up periods, as Brenner and Hoffmeister suggest. We analyzed data from all relevant randomized trials, which constitute the best available evidence for current screening recommendations. 1 The findings and conclusions of the included trials have been used to promote longer life with screening. We used intentionto-treat data because these are most reliable. We agree that intention-to-treat analyses may underestimate effect sizes, as we highlighted in the Discussion section of our article.Brenner and Hoffmeister as well as Watt et al are correct in stating that none of the included screening trials were powered for all-cause mortality. As we have explained in the article, the included trials still provide the most trustworthy precision estimates for lifetime gains with screening. 1,2 Regardless of hypothetical power estimates, with 95% probability, the real effects on prolonged lifetime by screening are between the upper and lower bounds of the stated confidence intervals. 1 It is up to societies and individuals to decide if these gains are large enough to prioritize screening.Kupiers and Spaander claim that population life expectancy is a crude measure to determine the pros and cons of individual health care measures. We agree but remind readers that this is what is used to promote screening among individuals. We believe that individuals who are considering whether to undergo cancer screening should know whether they can reasonably expect their lifetime to be extended by doing so. Furthermore, screening is not only an individual health care decision. It is a public health measure in most Western countries, and the effect of public health interventions on longevity is of relevance also to decision-makers.Lastly, in reply to Watt et al, we applied for registration with PROSPERO in October 2021 but received a notice stating that PROSPERO was prioritizing submissions related to COVID-19 and that our meta-analysis did not meet the requirements. We did not pursue the matter any further.