Diet-quality and its association with cardiovascular diseases and cancer incidence and all-cause mortality: a prospective cohort study from UK Biobank

Abstract

Introduction

Newly available data from big scale studies conducted in the UK, such as the UK Biobank, offers the possibility to further explore the prospective association between a diet-quality score and health outcomes after accounting for the effect of important confounding factors. The aim of this work, therefore, was to investigate the association between a diet-quality score, with the incidence of cardiovascular diseases (CVDs), cancer and all-cause mortality.

Material and methods

This study includes 345,343 participants (age range: 39–73, 55.1% women) from the UK Biobank, a prospective population-based study. Using 21 standardised variables of diet (alcohol, bread, bread type, cereal, dried fruit, water, coffee, tea, cheese, oily fish, non-oily fish, salt added to food, spread type, fresh fruit, cooked vegetable, raw vegetables, milk type, poultry, beef, lamb, and pork) we created a diet-quality score (very healthy, healthy, unhealthy and very unhealthy) using principal-component factor analysis. Associations between the dietary-quality score (very unhealthy individuals were the reference group) and health outcomes (all-cause mortality, CVD and cancer incidence) were investigated using Cox-proportional hazard models. All analyses were performed using STATA 14 statistical software.

Results

In comparison to individuals with a very unhealthy diet, those with a better diet-quality had a lower risk of all-cause mortality and cancer as well as incidence of CVD and cancer. For example, individuals classified in the very healthy group had a 12% lower risk of all-cause mortality (HR: 0.88 [95% CI: 0.82 to 0.95]), 12% lower risk of CVD incidence (HR: 0.88 [95% CI: 0.80 to 0.98]), 17% of all-cancer mortality (HR: 0.83 [95% CI: 0.75 to 0.93]), and 10% lower risk all-cancer incidence (HR: 0.90 [95% CI: 0.85 to 0.94]). Those in the healthy group had a 12% lower risk of all-cause (HR: 0.88 [95% CI: 0.83 to 0.93]) and 15% lower risk of all-cancer mortality (HR: 0.85 [95% CI: 0.78 to 0.93]). There was no significant association between CVD mortality and any diet-quality group. These findings were independent of major confounding factors including socio-demographic covariates, prevalent of diseases and lifestyle factors.

Discussion

Our findings indicate that individuals with a healthy diet in the UK biobank cohort are associated with a lower risk of premature mortality, and incidence of CVDs and cancer independently of major confounding factors.