AUTHOR=Wabo Therese M. C. , Wang Yifeng , Nyamao Rose M. , Wang Wenjie , Zhu Shankuan TITLE=Protein-to-carbohydrate ratio is informative of diet quality and associates with all-cause mortality: Findings from the National Health and Nutrition Examination Survey (2007–2014) JOURNAL=Frontiers in Public Health VOLUME=10 YEAR=2022 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2022.1043035 DOI=10.3389/fpubh.2022.1043035 ISSN=2296-2565 ABSTRACT=Background

Dietary protein and carbohydrate intake and health outcomes have received extensive attention in recent years. However, the nutritional context in which these associations occur is less studied.

Objectives

We aimed to examine the dietary context associating protein-to-carbohydrate ratio and all-cause mortality in US adults.

Methods

Data from 17,814 adults enrolled in the 2007–2014 NHANES was analyzed. Information on mortality was obtained from the US mortality registry updated in December 2015. Diet quality was assessed using the Healthy Eating Index (HEI) and Total Nutrients Index (TNI). ANCOVA was used to test the mean differences in HEI and TNI scores across %E P:C quintiles. Linear regression examined the association of HEI and TNI with %E P:C. Cox proportional hazards regression evaluated the association between %E P:C and all-cause mortality. A restricted cubic spline examined the non-linear relationship between %E P:C and death.

Results

Low %E P:C was associated with lower HEI and TNI scores while higher %E P:C was associated with healthier HEI and TNI scores. HEI and TNI were positively associated with %E P:C (β = 0.22, 95% CI: 0.19–0.25, and β = 0.16, 95% CI: 0.14–0.18), respectively. Low %E P:C was associated with an increased risk of death from all-cause. The higher HRs (95% CIs) of all-cause mortality were 1.97(1.46–2.65), and 7.35 (2.57–21.03) in the second quintile for the age-sex-ethnicity model, and the fully adjusted model, respectively. There was a significant reverse U-shape relationship between %E P:C and all-cause mortality with P, non-linearity < 0.001.

Conclusion

This study indicates that a low %E P:C that gives emphasis to unhealthy foods increases the risk of death. Hence, it would be useful to consider the complete diet associated with protein intake when making dietary recommendations for populations.