Ultra-processed food intake, genetic polymorphisms and the risk of dyslipidaemia in the adult Korean population.

Objective: This study aimed to examine the association between ultra-processed food intake and dyslipidaemia risk and whether this association varied by the polygenic score for dyslipidaemia in the adult Korean population.

Design: Prospective cohort study.

Setting: Ultra-processed foods were identified under the NOVA classification. Participants were categorised into < 5, 5 to < 10, 10 to < 15, 15 to < 20 and ≥ 20 %E/d of ultra-processed food intake. The polygenic scores for dyslipidaemia were calculated from 53 950 SNPs. ORs and 95 % CIs were estimated using multivariate logistic regression models.

Participants: 20 044 Korean adults aged ≥ 40 years in the Health Examinees (HEXA) study, the Cardiovascular Disease Association Study (CAVAS) and the Korea Association Resource (KARE) study.

Results: During median follow-ups of 4·09, 8·67 and 15·67 years in the HEXA, CAVAS and KARE studies, respectively, there were a total of 7331, 786 and 1732 incident dyslipidaemia events. Ultra-processed food intake was not significantly associated with dyslipidaemia risk. Compared with < 5 %E/d, the pooled OR (95 % CI) of ≥ 20 %E/d of ultra-processed food intake for dyslipidaemia incidence was 1·01 (0·90, 1·13; P for trend = 0·83). There was no interaction by dyslipidaemia-related genetic variations; ORs (95 % CIs) were 1·04 (0·89, 1·22; P for trend = 0·91) and 0·98 (0·84, 1·15; P for trend = 0·72) for individuals with high- and low-polygenic scores, respectively (P for interaction = 0·90).

Conclusions: No significant association was observed between ultra-processed food intake and the overall risk of dyslipidaemia, nor in subgroups of polygenic scores for dyslipidaemia among Korean adults with low ultra-processed food intake.