Coffee drinking and cancer risk: an umbrella review of meta-analyses of observational studies
Zhao LG, Li ZY, Feng GS, Ji XW, Tan YT, Li HL, Gunter MJ, Xiang YB · BMC Cancer · 2020 February 05
DOI 10.1186/s12885-020-6561-9
Objective
To investigate the association between coffee intake and cancer risk.
Conclusions
The study concludes that there is highly suggestive evidence for an inverse association between coffee intake and the risk of liver and endometrial cancer.
Critical Appraisal
Published research can offer valuable insights, but a paper’s conclusions may go beyond what its results support. Critical appraisal examines the methods, findings and limitations to clarify what can reasonably be concluded—and what remains uncertain.
This article has not yet been critically appraised by the Research Hub.
Learn more about Critical Appraisal
Insights
Coffee consumption shows a strong association with decreased risk of liver and endometrial cancer.
The randomized umbrella review included a total of 28 individual meta-analyses and 36 summary associations for 26 different types of cancer. In order to collect this data, a variety of databases like PubMed, Embase, Web of Science and the Cochrane database were searched to find systematic reviews and meta-analyses of relationships between coffee consumption and cancer incidence. The summary effect size was estimated for each association using both the fixed and random effects model. A detailed evaluation of heterogeneity, small-study effects, and excess significance bias was also carried out.
The results showed 17 out of 28 meta-analyses to be significant. For the highest versus the lowest categories, 4 out of 26 associations showed a more stringent value. There were notable dose-response associations for five types of cancer. Majority of the studies (69%) displayed low heterogeneity. There was evidence of excessive significance bias and publication bias in three and six associations respectively. Data exhibited that coffee intake was inversely proportional to the risk of contracting liver and endometrial cancer. All this remained constant even when the analysis was restricted only to the meta-analysis of cohort studies.
