Scientific Updates /

Plant-based diet indices and lipoprotein particle subclass profiles: a cross-sectional analysis of middle- to older-aged adults

10 June 2024


Original research

2024 Alpro Foundation Award for best scientific publication

The Alpro Foundation Award winner of 2024, Patrick Elliott, considered the impact of varying plant-based diets on lipoprotein subclass distribution and the atherogenic outcomes related to these relationships in this novel study from University College Dublin.

Patrick Elliott
Patrick is a PhD student in University College, Dublin and his research is focused on developing new personalised nutrition approaches to help people eat sustainable and healthy diets.

He has previous research experience in plant-based diets and cardiometabolic health through his master’s degree at the same institution.

Patrick is also a science communicator and runs a blog on health- and nutrition-related topics.


This study aimed to address a gap in the literature regarding the relationship between plant-based diet indices (PDIs) and lipoprotein subclasses.

Lipoproteins = Vehicles by which cholesterol and other lipids are transported around the body.

Lipoproteins are either pro-atherogenic, containing an apolipoprotein B (apoB) molecule or non-atherogenic.

Having more of the pro-atherogenic lipoproteins and less high-density lipoproteins (HDL) is characterised as a less cardiometabolically healthy profile to have in the blood.

Therefore, this study aims to elucidate if there is a link between plant-based dietary patterns and the composition of the lipoprotein profile.


The researchers from University College Dublin utilised a study population from County Cork who were invited via letter from their GP to participate.

They received a substantial 67% response rate (n=1,986) of urban and rural individuals with a median age of 59.5 years.

These participants were invited to complete a series of documentation and procedures:

  • General Health Questionnaire

  • International Physical Activity Questionnaire

  • Standard Clinical Procedures

  • Nuclear Magnetic Resonance (NMR) Spectroscopy

  • Food Frequency Questionnaire (FFQ)

    • This was the principal method of dietary assessment utilising a modified version of the European Prospective Investigation into Cancer and Nutrition (EPIC FFQ) adapted for an Irish population consisting of 150 items.

Plant-based diet indices (PDIs)

Plant-based diet adherence was categorised into healthful (hPDI) and unhealthy (uPDI). This categorisation considered 18 food groups, subcategorised into quintiles of consumption leading to a score between 18-86 in this study.

Lipoprotein Particle Subclass Profiles

The concentrations and sizes of lipoproteins were calculated using NMR spectroscopy using the LP3 algorithm. Associations between the PDIs and lipid profiles were also considered. Total cholesterol (TC), LDL cholesterol (LDL-C), HDL cholesterol (HDL-C) and triglyceride concentrations were measured on Olympus 5400 biochemistry analysers.

Tests of normality and descriptive analyses were considered, and linear regressions were carried out on three models:

  1. Crude analysis.

  2. Age- and sex-adjusted.

  3. Adjusted for age, sex, BMI, energy intake, education, smoking, physical activity, alcohol intake, T2DM status and lipid lowering medication usage.

Key findings


  • 9% of the population live with T2DM

  • 15% were smokers

  • 16% were non-drinkers

  • 48% reported low physical activity

Most participants had at least secondary-level education, were moderate alcohol drinkers and did not use lipid lowering medications.

More favourable health, diet and lifestyle parameters were observed in females. Females also had greater total LDL and HDL cholesterol concentrations but lower TG concentrations. They also had more favourable lipoprotein subclass profiles, with les non-HDL particles.

Correlation and Regression Analyses:

UPDI and lipoprotein profile showed they were less favourable to cardiometabolic health. This was robust to correction for multiple testing. Higher scores were associated with greater TG concentrations (q=0.02), more large VLDL (q<0.001), and small LDL (q=0.03), greater LP-IR score (q=0.002) and smaller HDL particle size (q=0.01).

The relationship between PDI and lipoprotein profiles were mostly neutral.

hPDI and lipoprotein profiles found a relationship with more favourable cardiometabolic health, lower non-HDL cholesterol, less large VLDL amongst other measures, but these findings were not robust to correction for multiple testing.

Discussion and recommendations

The key result is that unhealthier PBDs, as determined by this study, are associated with less favourable, pro-atherogenic lipid and lipoprotein profiles in middle- to older-aged adults. The study drew upon a large sample size and utilised validated tools and statistical methods to reach this conclusion.

Despite a small magnitude of effect and not being able to draw causal inferences, this further contextualises previous findings on unhealthy PDIs leading to increased risk of CVD, T2DM and mortality.

“Not all plant-based diets are equal”

The implications of this are that minor changes to more unhealthy plant-based options are negative for metabolic health. Diets should be centred on healthy plant-based options like legumes, seeds, and nuts while reducing less healthy foods like sugar-sweetened beverages and refined carbohydrates. When advising on reducing meat or discussing a range of plant-based diets, quality as well as quantity of plant-based foods consumed should be at the fore.


  1. Elliott, P.S., Harrington, J.M., Millar, S.R., Otvos, J.D., Perry, I.J. and Phillips, C.M., 2023. Plant-based diet indices and lipoprotein particle subclass profiles: A cross-sectional analysis of middle-to older-aged adults. Atherosclerosis, 380, p.117190.

Save article as PDF

Original research

Plant-based diet indices and lipoprotein particle subclass profiles: A cross-sectional analysis of middle-to older-aged adults

Share this article on social media.

Original research

Plant-based diet indices and lipoprotein particle subclass profiles: A cross-sectional analysis of middle-to older-aged adults