“Eat more vegetables”. You must have heard it, read it, or said it. Probably your mom said that when you left peas or carrots on your plate at dinner. Indeed, a massive body of research has shown an association between eating vegetables and lower risk of death and diseases, including obesity, cancers, and heart diseases. Therefore, it is universally recommended to include plenty of vegetables in our everyday diet. 

 

It has been noticed that people who eat more vegetables tend to have an overall healthier lifestyle. For example, they may also eat less processed meat and smoke less. All these factors are known to reduce disease risk. They are called ‘confounders’. Their presence makes it very difficult to draw a definite conclusion whether eating more vegetables could truly reduce disease risk, as the reduced risk may be caused by high vegetable intake, or it may result from an individual’s other healthy lifestyle factors. 

 

Therefore, it is important to control for confounders, when we aim to examine the pure effect of eating vegetables. What researchers usually do is collect data on confounders and control for them in data analysis. However, this method cannot completely remove the disturbing effects of confounders on the vegetable-diseases association. Researchers may miss some confounders in data collection and/or analysis. For example, data on some confounders may be collected inaccurately (which is very common), or omitted because researchers do not know they exist. This part of confounding, which has not been completely controlled for, is called ‘residual confounding’. It can still be troublesome. So in this study, we examined the effect of eating vegetables on cardiovascular risk, and we further examined the role of residual confounding on vegetable-disease association.    

 

We recruited 400,000 adults aged 40-69 from England, Scotland and Wales. We asked how many heaped tablespoons of cooked and raw vegetables they eat on an average day, respectively. One heaped tablespoon is roughly one serving. We recorded their age, sex, ethnicity, socioeconomic status, body mass index, medical history, dietary habits as well as other lifestyle factors. All were free of cardiovascular diseases at the start, and were followed for 12 years (through March 2021). We checked the National death registry and hospital admission databases to see who developed cardiovascular diseases and fatalities. 

 

As might be expected, our results support that eating raw vegetables reduced risks of cardiovascular diseases, after controlling for all the confounders we measured. Specifically, people eating more raw vegetables showed 10% lower risk of cardiovascular diseases and 15% lower cardiovascular mortality. This impact is similar to what previous research has reported. We observed the following about the adults who ate more raw vegetables: They were more likely to be women, higher educated, and in higher socioeconomic status. They also tended to exercise more, have a healthy weight, smoke less, eat less processed meat, and eat more fresh fruit. We further found that 80-87% of the associations could be explained by residual confounding of lifestyle factors and socioeconomic status. That means if all of these factors had been perfectly measured, 80-87% of the associations would be diluted. This suggests that very likely, much of the associations we have observed are due to residual confounding. Although we cannot exclude the possibility that there is a true effect, the effect should be small, if it does exist. Additionally, we found no association between eating cooked vegetables and cardiovascular risk. 

 

What can we learn from our findings? Our findings imply that solely eating more vegetables, if other lifestyle habits are poor, may have limited protection against cardiovascular diseases. It is relatively easier to change lifestyle habits than to change socioeconomic status. So, eating more vegetables and improving lifestyle would be more effective for disease prevention. For example, Tom eats 1 serving of vegetables and 2 servings of processed meat every day, and exercises rarely. If he eats 3 more servings of vegetables, this may not reduce his cardiovascular risk. Yet, if he increases vegetable intake by 3 servings and decreases processed meat intake by 1 serving, plus exercises more regularly, this will be very helpful. Therefore, the take-home message is to keep a balanced diet and a healthy lifestyle. NHS has kindly offered “Eat well” dietary guidance. 

 

Of note, we only looked at cardiovascular diseases. The effects of eating vegetables on other health outcomes, such as cancers and diabetes, are not discussed here. Also, the diet measured in this study may better represent typical dietary patterns in the UK, rather than other populations.