Premium
Dairy consumption and risk of esophageal squamous cell carcinoma: A meta‐analysis of observational studies
Author(s) -
Li Bailing,
Jiang Gengxi,
Xue Qing,
Zhang Hao,
Wang Chong,
Zhang Guanxin,
Xu Zhiyun
Publication year - 2016
Publication title -
asia‐pacific journal of clinical oncology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.73
H-Index - 29
eISSN - 1743-7563
pISSN - 1743-7555
DOI - 10.1111/ajco.12183
Subject(s) - medicine , observational study , confounding , meta analysis , confidence interval , relative risk , esophageal squamous cell carcinoma , food science , carcinoma , biology
Aim Inconsistent results regarding the relations between consumption of dairy products and the risk of esophageal squamous cell carcinoma ( ESCC ) have been reported. In this report, we summarized the evidence by a meta‐analysis of observational studies. Methods Eligible studies published up to J anuary 31, 2013 were retrieved via both computer searches and a manual review of references. Random‐effects models were used to calculate the summary relative risk ( SRR ) based on high versus low and dose–response analysis. Results A total of 19 studies with 4315 ESCC cases were included in this meta‐analysis. Overall, there were no significant associations between intakes of total dairy products, milk, cheese and butter and ESCC for the highest versus lowest intake categories (total dairy products: SRR 1.03, 95% confidence interval [ CI ]: 0.60–1.77; milk: SRR 0.93, 95% CI : 0.74–1.16; cheese: SRR 0.84, 95% CI : 0.61–1.15; butter: SRR 1.77, 95% CI 0.85–3.75). A significant inverse association was found for yogurt consumption ( SRR 0.73, 95% CI : 0.54–0.98). There was high heterogeneity among studies on total dairy products, milk and butter; however, little or no heterogeneity was observed among studies on cheese and yogurt. Conclusion N o associations between consumption of milk, dairy products, butter or cheese and risk of ESCC were found, while yogurt consumption may have a protective effect. However, these associations may be subject to high levels of heterogeneity or confounding, and further efforts should be made to confirm these findings.