Chocolate and risk of chronic disease: a systematic review and dose-response meta-analysis

PURPOSE: Evidence for the association between chocolate intake and risk of chronic diseases is inconclusive. Therefore, we aimed to synthesize and evaluate the credibility of evidence on the dose-response association between chocolate consumption with risk of all-cause mortality, coronary heart disease (CHD), stroke, heart failure (HF), type 2 diabetes (T2D), colorectal cancer (CRC), and hypertension. METHODS: Prospective studies were searched until July 2018 in PubMed, Embase, and Web of Science. Random-effects meta-analyses comparing highest versus lowest intake categories, linear, and non-linear dose-response analyses were conducted. The credibility of evidence was evaluated with the NutriGrade scoring-system. RESULTS: Overall, 27 investigations were identified (n = 2 for all-cause mortality, n = 9 for CHD, n = 8 for stroke, n = 6 for HF, n = 6 for T2D, n = 2 for hypertension and CRC, respectively). No associations with HF (RR 0.99, 95% CI 0.94, 1.04) and T2D (RR 0.94, 95% CI 0.88, 1.01) per each 10 g/day increase in chocolate intake were observed in the linear dose-response meta-analyses. However, a small inverse association for each 10 g/daily increase could be shown for the risk of CHD (RR 0.96, 95% CI 0.93, 0.99), and stroke (RR 0.90, 95% CI 0.82, 0.98). The credibility of evidence was rated either very low (all-cause mortality, HF, T2D, CRC or hypertension) or low (CHD, stroke). CONCLUSION: Chocolate consumption is not related to risk for several chronic diseases, but could have a small inverse association with CHD and stroke. Our findings are limited by very low or low credibility of evidence, highlighting important uncertainty for chocolate–disease associations. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s00394-019-01914-9) contains supplementary material, which is available to authorized users.