Although previous investigations have proposed an association between Dietary Approaches to Stop Hypertension (DASH)-style diet and lower mortality from chronic diseases, the exposure-response relationship is not clear. The present systematic review and meta-analysis aimed to explore the linear and non-linear dose-response association between adherence to the DASH diet and all-cause and cause-specific mortality.

Database search was performed in PubMed, Scopus, and EMBASE for prospective cohort studies investigating the association between adherence to the Dietary Approaches to Stop Hypertension (DASH) diet and risk of mortality. Summary hazard ratios (HRs) and 95% confidence intervals (CI) were estimated with the use of a random-effects model for the linear and nonlinear relationships. The two-stage hierarchical regression model was applied to test the potential non-linear dose-response associations.

The inclusion criteria were met by 17 studies (13 publications). The scores reported for adherence to the DASH diet in different studies were converted to a conventional scoring method in which the adherence score might range between 8 to 40. The linear analysis revealed that summary HRs were 0.95 (95% CI: 0.94–0.96, I² = 91.6%, n = 14) for all-cause, 0.96 (95% CI: 0.95–0.98, I² = 82.4%, n = 12) for CVD, 0.97 (95% CI: 0.96–0.98, I² = 0.00%, n = 2) for stroke, and 0.97 (95% CI: 0.95–0.98, I² = 63.7%, n = 12) for cancer mortality per each 5-point increment of adherence to the DASH diet. There was also evidence of non-linear associations between the DASH diet and all-cause and cause-specific mortality as the associations became even more evident when the adherence scores were more than 20 points (P < 0.005).

Even the modest adherence to the DASH diet is associated with a lower risk of all-cause and cause-specific mortality. The higher adherence to the diet also strengthens the risk-reducing association.

This review was registered in the international prospective register of systematic reviews (PROSPERO) database (registration ID: CRD42018086500).