Diet Quality and Colorectal Outcomes
In this PLCO cohort analysis, higher HEI-2020 scores were associated with favorable colorectal cancer incidence-trend and mortality findings, but the observational design and a nonsignificant extreme-quartile incidence comparison require a bounded interpretation.
> Research explainer: This briefing examines verified primary research published 78 days before the briefing date. It is not a same-day research update and does not provide medical advice.
This Research explainer examines a source published on 2026-06-11. It reports a prospective analysis of diet quality and colorectal cancer (CRC) outcomes among U.S. adults enrolled in the PLCO cohort. [pmid:42370359]
Evidence
The analysis included 101,709 PLCO participants aged 55–74 years. During the study period, investigators identified 1,100 incident CRC cases and 314 deaths. Diet was measured with the Diet History Questionnaire, and researchers calculated Healthy Eating Index 2020 (HEI-2020) scores; higher scores represented closer alignment with the 2020–2025 Dietary Guidelines for Americans. [pmid:42370359]
For CRC incidence, the fully adjusted comparison of the highest HEI-2020 quartile with the lowest produced a hazard ratio of 0.86, with a 95% confidence interval of 0.73–1.02. Because that interval includes 1.0, this particular highest-versus-lowest comparison was not statistically significant. [pmid:42370359]
The study nevertheless reported a statistically significant decreasing trend in CRC incidence risk across HEI-2020 quartiles, with P for trend=0.021. This trend result and the nonsignificant highest-versus-lowest estimate describe different statistical summaries and should be read together. [pmid:42370359]
For CRC-specific mortality, the highest HEI-2020 quartile was associated with a lower risk estimate than the lowest quartile: hazard ratio 0.61, 95% confidence interval 0.45–0.85, with P for trend=0.001. The source states that this mortality association remained consistent in competing-risk analyses. [pmid:42370359]
The authors also reported site-specific patterns: lower risks for distal colon cancer mortality, a borderline lower point estimate for rectal cancer mortality, and a decreasing trend in proximal colon cancer mortality. The supplied source characterizes these as site-specific analyses rather than providing a single uniform result for every colorectal site. [pmid:42370359]
Analysis — Association pattern and interpretation
The central finding is an association pattern, not evidence that diet quality caused either lower CRC incidence or lower CRC-specific mortality. The incidence results are especially important to separate carefully: the quartile trend was statistically significant, whereas the direct comparison between the highest and lowest quartiles was not. Thus, the source supports saying that higher HEI-2020 scores tracked with a decreasing incidence trend in this cohort; it does not support treating the extreme-quartile incidence result as independently conclusive. [pmid:42370359]
The mortality result was more precise in the reported highest-versus-lowest comparison, with a confidence interval below 1.0, and was described as consistent in competing-risk analyses. That consistency can increase confidence that the reported association was not dependent on that particular analytic framing, but it does not change the observational nature of the evidence. [pmid:42370359]
HEI-2020 is a whole-diet quality measure rather than a measurement of one food, nutrient, or treatment. Accordingly, this study evaluates how a score derived from questionnaire-reported dietary information was associated with outcomes in the PLCO population. It should not be read as a test of a specific diet intervention or as proof that changing an individual’s diet will change cancer outcomes. [pmid:42370359]
Limitations
This was a prospective observational cohort analysis, so measured associations may reflect factors other than HEI-2020 adherence and cannot establish causation. [pmid:42370359]
Dietary information came from the Diet History Questionnaire, making the HEI-2020 assessment dependent on those reported data. [pmid:42370359]
Direct applicability is limited by the analyzed population: U.S. adults aged 55–74 enrolled in PLCO. Findings from this group cannot by themselves establish the same associations in other populations. [pmid:42370359]
Finally, the incidence evidence is mixed in the specific sense reported above: there was a significant trend across quartiles, but the highest-versus-lowest quartile estimate was not statistically significant. This briefing therefore describes the results as observational associations and offers no medical advice, prediction, or patient-specific conclusion. [pmid:42370359]