A daily portion of nuts (28 g) reduces the risk of cardiovascular disease by about 21% and premature death by about 22%, supported by both meta-analyses and a large randomised trial. A proven effect on body weight or diabetes has not yet been established.
A handful of nuts per day (roughly 28 grams) is associated with a clearly lower risk of cardiovascular disease. Large umbrella analyses of existing meta-analyses show that this risk drops by about 21% compared with eating no nuts at all. The likelihood of dying prematurely from all causes combined was also approximately 22% lower among regular nut eaters in those analyses (PMID 36041171, 29800597).
The strongest direct evidence comes from the randomised PREDIMED trial (7,447 participants, mean follow-up 4.8 years). People who followed a Mediterranean diet supplemented with nuts had a 30% lower chance of a serious cardiovascular event such as a heart attack, stroke, or death from cardiovascular disease than the control group that received low-fat dietary advice (hazard ratio 0.70; PMID 24829485, 25940230). The same study showed that the nut group also experienced a far greater reduction in metabolic syndrome: a 13.7% reduction, compared with 6.7% and 2.0% in the other groups.
The main mechanism appears to be an improvement in the blood lipid profile: nuts lower LDL cholesterol and improve the ratio of apolipoproteins (the proteins that transport fats). In addition, they reduce oxidative stress and inflammation and improve blood vessel function (PMID 29800597, 36041171). The protective effect on stroke is present, but weaker and less well supported than the effect on coronary heart disease such as heart attack (PMID 29800597).
For body weight, the picture is less clear. Nuts are calorie-dense, but the available studies give no reason to think that one portion per day causes weight gain. However, a proven independent slimming effect cannot be substantiated on the basis of these sources either (PMID 36041171, 25447615). In studies, nuts are almost always examined as part of a healthy dietary pattern, not as a stand-alone weight-loss agent.
Whether nuts reduce the risk of type 2 diabetes remains unclear for the time being. In PREDIMED, the effect of the nut group on diabetes was not statistically reliable (hazard ratio 0.82; 95% confidence interval 0.61-1.10), and broader review analyses give mixed signals. The evidence for cardiovascular disease is much stronger than for diabetes (PMID 24829485, 36041171). Finally, approximately 1 to 2 percent of adults have a nut allergy or another adverse reaction. For them, this advice obviously does not apply (PMID 36041171).
Based on two umbrella reviews/large meta-analyses (PMID 36041171, 29800597) and the randomised PREDIMED trial (PMID 24829485, 25940230, 25447615). PREDIMED was later partly re-analysed due to a randomisation error at one sub-centre, but the revised analysis confirmed the main findings.