Reacción a "The Basque Country’s colorectal cancer screening programme was associated with a 50 per cent reduction in mortality from this cause among adults aged 50 to 69"

Assessment of the press release

The press release accurately summarises the main findings of the study, but there are some substantive issues that should be corrected before it is published, as they could lead to a distorted interpretation of the study:

1. The authors themselves explicitly state that the observational and retrospective design prevents any causal inference from being drawn. Throughout the manuscript, the term ‘association’ is used consistently, never ‘direct causality’. The press release, however, states in the headline and in the body: ‘the programme… has reduced mortality… by 30 per cent’ and ‘has prevented more than 6,000 deaths’. However, its observational-retrospective design precludes causal inference. I recommend including ‘associated with a reduction’ rather than ‘has reduced’ or ‘has prevented’.
2. The reference in the first paragraph to savings on surgery, hospital admissions and cancer treatments is not supported by the data presented in this study, which is purely epidemiological (mortality) and contains no cost component. Attributing this to the JAMA Network Open study would be inaccurate and could prompt an awkward question from a journalist asking for the specific figure; therefore, I would avoid this comment in the press release.
3. The figure of “more than 6,000 deaths prevented” does not appear anywhere in the manuscript or in the tables/figures, and should therefore not be presented as a result of the study.

Assessment of the study

This is a large-scale population-based study with a very long follow-up period: it includes virtually the entire population of the Basque Country over 21 consecutive years, allowing the phenomenon to be observed at different stages of the screening programme and even during the disruption to healthcare caused by COVID-19. The data are drawn from high-quality population registers (the Basque Institute of Statistics and the screening programme’s own register), which reduces the risk of selection bias or loss to follow-up.

From a methodological perspective, the study employs a robust analytical design, using age-adjusted rates, segmented regression to objectively identify the point at which the trend changed, and counterfactual projections that estimate what would have happened had the screening programme not existed. Furthermore, it incorporates an unscreened reference group (those under 50 years of age) as an internal control and confirms its results through sensitivity analyses (excluding deaths from COVID-19). It is this combination of a large sample size, long-term follow-up and the triangulation of several independent analytical approaches that lends greater weight to the observed association, although, as in any observational study, it cannot establish direct causality with the same certainty as a randomised trial.

As for the most relevant findings, the study shows that the reduction in colorectal cancer mortality was particularly marked in the target population (aged 50–69), with a 50 per cent decrease compared with pre-screening rates, well above that observed in other population-based European programmes, which typically report reductions of 9–10 per cent in the first decade of follow-up. This highly favourable result is largely explained by the high and sustained participation in the Basque programme, which remained at around 70 per cent throughout the period, together with an adherence rate to diagnostic colonoscopy following a positive test of over 90 per cent.