Nut consumption on all-cause, cardiovascular, and cancer mortality risk: a systematic review and meta-analysis of epidemiologic studies.

BACKGROUND Recent pooled analyses supported a beneficial impact of nut consumption on health, but to our knowledge, whether nuts are associated with overall decreased mortality has not been previously reviewed. OBJECTIVES We aimed to systematically review prospective studies that explored the effects of nut consumption on all-cause, cardiovascular disease (CVD), and cancer mortality and quantify the size effect through a meta-analysis. We also reviewed confounding factors associated with nut consumption to assess potential clustering with other covariates. DESIGN We searched PubMed and EMBASE for studies published up to June 2014. Study characteristics, HRs, and 95% CIs were generated on the basis of quantitative analyses. A dose-response analysis was performed when data were available. RESULTS Seven studies for all-cause mortality, 6 studies for CVD mortality, and 2 studies for cancer mortality were included in the meta-analysis with a total of 354,933 participants, 44,636 cumulative incident deaths, and 3,746,534 cumulative person-years. Nut consumption was associated with some baseline characteristics such as lower body mass index and smoking status as well as increased intakes of fruit, vegetables, and alcohol. One-serving of nuts per week and per day resulted in 4% (RR: 0.96; 95% CI: 0.93, 0.98) and 27% (RR: 0.73; 95% CI: 0.60, 0.88) decreased risk of all-cause mortality, respectively, and decreased risk of CVD mortality [RR: 0.93 (95% CI: 0.88, 0.99) and 0.61 (95% CI: 0.42, 0.91), respectively]. Effects were primarily driven by decreased coronary artery disease deaths rather than stroke deaths. Nut consumption was also associated with decreased risk of cancer deaths when highest compared with lowest categories of intake were compared (RR: 0.86; 95% CI: 0.75, 0.98), but no dose-effect was shown. CONCLUSION Nut consumption is associated with lower risk of all-cause, CVD, and cancer mortality, but the presence of confounding factors should be taken into account when considering such findings.