Navigating plant-based diets: increased cardiovascular risk in those with unhealthy plant-based dietary plans in people with pre-diabetes or diabetes Authors: Sonia Butalia, Kara Nerenberg Published: 2026 Commentary on: Zhuang P, Wang F, Yao J, et al. Unhealthy plant-based diet is associated with a higher cardiovascular disease risk in patients with prediabetes and diabetes: a large-scale population-based study. BMC Med 2024;22:485. Implications for practice and research Assessment of plant-based dietary plans’ content is important such that unhealthy food groups, associated with increased cardiovascular disease risk in people with prediabetes or diabetes, are avoided. Assessment of plant-based dietary plans’ content is important such that unhealthy food groups, associated with increased cardiovascular disease risk in people with prediabetes or diabetes, are avoided. Further work is needed to support population-level interventions to minimise unhealthy plant-based foods and to enhance equitable access to healthy foods. Further work is needed to support population-level interventions to minimise unhealthy plant-based foods and to enhance equitable access to healthy foods. Context Given the global epidemic of diabetes and pre-diabetes, there are continuing interests in population-level interventions to reduce the risk of cardiovascular disease (CVD) in this high-risk population. General healthy behaviours (nutrition and exercise) remain the foundation to management despite clinical research gaps on specific nutritional interventions for CVD prevention in those with diabetes or pre-diabetes. This study by Zhuang et al addresses an important gap by examining the association of different levels of adherence to plant-based dietary plans and the risk of CVD in people with pre-diabetes or diabetes.1 Methods This population-based cohort study used the UK Biobank nutritional dataset with longitudinal clinical outcomes.2 Individuals with pre-diabetes or diabetes and who completed greater than or equal to one 24-hour recall dietary assessment were included. The exposure, Plant-based Diet Indexes (PDIs), were categorised into an overall PDI, a healthy PDI (hPDI) and an unhealthy PDI (uPDI) based on consumption of food groups, including plant- and animal-based foods. Adherence to dietary categories was scored, with a higher score representing higher adherence and subsequently split into tertiles. The outcome of CVD was obtained from linked hospital administrative data. Analyses examined the risks of CVD associated with plant-based diet category (PDI, hPDI or uPDI) tertiles using Cox proportional hazard models (adjusted for known CVD risk factors) and the contributions of individual dietary components and blood biomarkers. Findings A total of 7798 individuals with diabetes and 17 926 with pre-diabetes were followed up for ~10 years. Importantly, higher adherence to an unhealthy plant-based diet (ie, higher tertiles of the uPDI compared with the lowest tertile) was associated with an increased risk of CVD in people with pre-diabetes and diabetes. This association was partly explained by higher intake of sugar-sweetened beverages and animal fats. Finally, an inverse association was found between higher adherence to a healthy plant-based diet and CVD risk in people with pre-diabetes, but not in those with diabetes. Commentary Alarmingly, CVD remains a leading cause of morbidity and mortality in those with pre-diabetes or diabetes, highlighting the need for prevention strategies.3 Zhuang et al demonstrated that an ‘unhealthy’ plant-based dietary pattern was associated with an increased risk of CVD among those with pre-diabetes or diabetes.1 This increased risk was largely driven by the consumption of sugar-sweetened drinks and animal fats. These findings are consistent across epidemiological studies in diverse populations and are supported by mechanistic studies supporting the role of inflammation underlying the association between unhealthy diets and CVD.4 5 However, the relationship between ‘healthy’ plant-based diets (represented by the hPDI) and CVD risk was more complex and less consistent. For example, while higher adherence to a healthy PDI pattern in individuals with pre-diabetes was protective, the same finding was not found in those with diabetes. Furthermore, there was no clear pattern of association of the hPDI across planned subgroup analyses for known risk factors for CVD. From a mechanistic perspective, the team’s exploratory analyses of biomarkers, diet and CVD yielded inconsistent associations which identify interesting hypotheses to be tested in future studies which may address this study’s methodological limitations associated with the temporal difference in collection of biomarkers and dietary assessments. The take-home messages for people living with diabetes or pre-diabetes centres on avoiding unhealthy plant-based dietary patterns and limiting consumption of sugar-sweetened beverages and animal fats. For interdisciplinary clinicians, the study identifies a need for regular assessments of the composition and adherence to plant-based diets along with individualised counselling. For public health and policy planners, this study supports ongoing population-level interventions to limit unhealthy plant-based diets and equitable access to healthy foods. Finally, future work is needed to examine specific preventive impacts of nutrition on CVD for those with pre-diabetes or diabetes.