Healthful Plant-Based Diet Index and Physical Activity Are Associated with Adiposity and the Triglyceride-HDL Axis but Not Apolipoprotein B: NHANES 2007–2016

Abstract

Background: Apolipoprotein B (apoB) is the most direct indicator of the circulating atherogenic particle burden. Randomized trials indicate that vegetarian and vegan diets lower apoB. However, the association between population-level variation in plant-based diet quality, alone or with physical activity, and apoB has not been examined.

Objective: To evaluate associations between the healthful plant-based diet index (hPDI) and physical activity and apoB and a panel of atherogenic and metabolic risk markers in a nationally representative sample of US adults.

Methods: Data from 10,545 non-pregnant adults aged 20 years and older who underwent fasting blood collection in the National Health and Nutrition Examination Survey (2007–2016) were analyzed. Plant-based diet indices were derived from 24-hour dietary recalls linked to the United States Department of Agriculture (USDA) Food Patterns Equivalents Database. Physical activity was quantified as metabolic equivalent minutes per week using the Global Physical Activity Questionnaire. Survey-weighted linear regression with Taylor series linearization was applied throughout. Analyses were exploratory; apoB and seven additional atherogenic and metabolic outcomes were examined and adjusted for multiplicity using the Benjamini-Hochberg procedure. A saturated-fat-aware variant of the index, scoring total solid fat in place of animal fat, was examined as a sensitivity analysis.

Results: Higher hPDI was not significantly associated with apoB (-0.26 mg/dL per 1 standard deviation [SD], 95% CI -0.89 to 0.38, P = .42), low-density lipoprotein cholesterol (LDL-C), non-high-density lipoprotein cholesterol (non-HDL-C), or total cholesterol. In the same models, higher hPDI was associated with lower body mass index (-0.71 kg/m2, 95% CI -0.90 to -0.52), lower waist circumference (-1.67 cm, 95% CI -2.13 to -1.21), and higher HDL cholesterol (0.85 mg/dL, 95% CI 0.47 to 1.23), all q < .001. In a model omitting body mass index, which estimates the total rather than the adiposity-independent association, the apoB estimate was -0.70 mg/dL per 1 SD (95% CI -1.34 to -0.06). No significant interaction between diet and physical activity was detected. Equivalence testing excluded an apoB difference greater than 5 mg/dL across the 10th to 90th percentile of hPDI, against a bound benchmarked to trial effect sizes rather than an established clinical threshold. Adults in the highest hPDI quintile obtained 9.9% of energy from saturated fat, and saturated fat intake was itself associated with LDL-C. Across quintiles, nuts, whole grains, and whole fruit increased several-fold and sugar-sweetened beverages decreased by 87%, whereas dairy intake was essentially unchanged and total solid fat decreased by 40%. A saturated-fat-aware index gave the same null result for apoB (-0.24 mg/dL per 1 SD, 95% CI -0.88 to 0.40).

Conclusions: Higher hPDI scores and physical activity were associated with more favorable adiposity and triglyceride-HDL measures, but not with apoB or other atherogenic lipid markers after adjustment for body mass index. The hPDI gradient did not capture differences in saturated fat intake, or in the foods that supply it, large enough to distinguish dietary patterns associated with lower apoB, and an index amended to score total solid fat behaved no differently. These findings characterize the index as an exposure measure and do not address the effects of plant-based dietary interventions, which have lowered apoB in randomized trials. Improving plant-based diet quality may confer metabolic benefit, while additional targeted reduction of saturated fat may be needed to lower atherogenic lipoprotein burden.