Associations between dietary macronutrient quality and odds of hyperlipidemia: findings from the NEC-Biobank cohort

Abstract

Background: Associations between the overall quality of dietary macronutrients and the odds of hyperlipidemia remain unknown. This study aimed to first investigate the aforementioned associations, applying a novel multidimensional macronutrient quality index (MQI).

Methods: A large cross-sectional study (2018–2020) was carried out among 14,544 participants aged 18–79 years from the NEC-Biobank cohort (Northeast region, China). Dietary information was collected from a validated 110-item food frequency questionnaire. MQI was calculated based on fat quality index (FQI), protein quality index (PQI), and carbohydrate quality index (CQI). Binary logistic regression models were used to assess the odds ratio (OR) with 95% confidence intervals (CI).

Results: Higher MQI was associated with significantly reduced odds of hyperlipidemia (OR_{tertile 3 vs. tertile 1} = 0.82, 95% CI: 0.75–0.90; P _trend < 0.001). For MQI’s sub-indices, higher FQI (OR_{tertile 3 vs. tertile 1} = 0.84, 95% CI: 0.77–0.91; P _trend < 0.001) as well as higher PQI (OR_{tertile 3 vs. tertile 1} = 0.86, 95% CI: 0.79–0.94; P _trend < 0.001) were associated with lower odds of hyperlipidemia.

Conclusions: Our findings suggested that intake of high-quality macronutrients was associated with reduced odds of hyperlipidemia. These findings contribute novel knowledge to the field of nutrition and metabolic diseases, and lay the foundation for future studies.

Clinical trial number: Not applicable.

Supplementary Information: The online version contains supplementary material available at 10.1186/s12937-025-01174-w.

Keywords: Carbohydrate quality; Fat quality; Hyperlipidemia; Macronutrient quality; Protein quality.

Fig. 1

Flow chart of participants

Fig. 2

Stratification analysis of associations between dietary macronutrient quality and odds of hyperlipidemia. ^a The binary logistic regression model was used to calculate OR and 95% CI. The model was adjusted for age (continuous, year), sex (male or female), body mass index (continuous, kg/m²), total energy intake (continuous, kcal/day), physical activity (continuous, MET/hours/week), annual family income (continuous, RMB, ten thousand yuan), education (junior secondary or below, senior high/technical secondary school or above), smoking (yes or no), drinking (yes or no), hypertension (yes or no), diabetes (yes or no), heart disease (yes or no), and stroke (yes or no), unless a certain variable was the basis of the stratification. ^b Interaction was evaluated using the binary logistic regression model by including the product term of macronutrient quality index and its sub-indices (tertiles) and stratification variables (dichotomous). ^c Physical activity level was divided into “high” and “low” by median (96.93 MET/hours/week). ^d Income indicated annual family income (RMB, ten thousand yuan). ^e Education was divided into “low” (junior secondary or below) and “high” (senior high/technical secondary school or above). CI, confidence interval; CQI, carbohydrate quality index; FQI, fat quality index; MET, metabolic equivalent; MQI, macronutrient quality index; OR, odds ratio; PQI, protein quality index