New research reveals how dietary diversity affects stroke risks differently in men and women, challenging assumptions about universal nutrition advice.

Study: Healthy food diversity and the risk of major chronic diseases in the EPIC-Potsdam study. Image Credit: Dulce Rubia / Shutterstock

In a recent study published in the journal Scientific Reports, researchers in Germany investigated the association between the Healthy Food Diversity (HFD) Index and the risk of type 2 diabetes (T2D), myocardial infarction (MI), and stroke in a middle-aged German population.

Background

Major chronic diseases, including MI, stroke, and T2D, are leading global causes of mortality, with dietary behavior being a key modifiable risk factor. International dietary guidelines emphasize diverse diets to ensure nutrient adequacy and support health. However, evidence linking diet diversity with reduced chronic disease risk is limited and inconsistent. Most indices fail to differentiate between diversity within healthy and unhealthy foods, highlighting the need to integrate diet quality into measures of diversity. The study also underscores that combining diversity with diet quality can help clarify dietary impacts on chronic disease prevention, especially in diverse populations.

About the study

The European Prospective Investigation into Cancer and Nutrition (EPIC) study, initiated in 1992, investigates diet and chronic disease risk across Europe. The Potsdam cohort recruited 27,548 participants (aged 35–64 for women, 40–64 for men) between 1994 and 1998. Follow-up occurred every two years, achieving over 90% response rates, with additional passive follow-up through death certificates. Data up to the fifth follow-up in 2009 were included. Ethical approval was granted, and participants provided informed consent.

Participants with missing data, implausible energy intake (6000 kcal/day), or non-verified/prevalent cases of T2D, MI, or stroke at baseline were excluded. Final sample sizes were 25,063, 26,011, and 26,265 for T2D, MI, and stroke, respectively.

Dietary intake over 12 months was assessed using a validated 149-item food-frequency questionnaire (FFQ). The HFD Index was calculated to capture dietary diversity, proportionality, and quality based on German dietary guidelines. Incident cases of MI, T2D, and stroke were identified through self-reports, verified by medical records, and coded using the International Statistical Classification of Diseases and Related Health Problems (ICD) -10 classifications.

Cox proportional hazard models assessed associations between the HFD Index and disease risk, adjusting for confounders. Analyses explored sex and age modifications with sensitivity analyses. Statistical analyses were conducted using SAS 9.4.

Study results

The EPIC-Potsdam study analyzed 26,591 participants, with a mean age of 50.5 years (SD: 9.0), 60% of whom were women. Participants were predominantly well-educated and employed full-time, with 51% being current or former smokers and most consuming alcohol. Men with a higher HFD Index were more educated, less likely to exceed recommended alcohol limits, and smoked less, with similar but smaller differences observed in women. The mean body mass index (BMI) was 26.3 kg/m² (SD: 4.3), and 49% of participants had hypertension. Interestingly, participants with higher HFD Index scores consumed fewer calories, fat, saturated fatty acids, and cholesterol but had higher intakes of essential nutrients like vitamins C and E, magnesium, and fiber.

The mean HFD Index was 0.49 (SD: 0.10), slightly higher in women (0.51) than in men (0.46). Participants with higher HFD Index scores consumed fewer sweets, snacks, and high-fat animal foods while eating more vegetables, fruits, wholemeal products, and low-fat animal foods.

Over 10.5 years of follow-up, 1,537 participants developed T2D, 376 had MI, and 412 experienced strokes. No significant association was found between the HFD Index and T2D or MI in either sex. However, a key finding was that men with higher HFD Index scores had a lower risk of stroke [HR highest vs. lowest tertile: 0.52 (95% CI: 0.37, 0.74)], while women with higher scores showed an increased stroke risk [HR highest vs. lowest tertile: 1.44 (95% CI: 0.98, 2.12)]. The harmful association in women was more pronounced among those aged 51 or older.

Sensitivity analyses excluding early cases and investigating individual components of the HFD Index (Berry-Index and Health Value) confirmed accuracy. While the Health Value was strongly associated with outcomes, the Berry-Index showed weaker or no associations.

Conclusions

To summarize, a healthful and diverse diet may benefit chronic disease prevention, but evidence remains inconsistent. The HFD Index, combining dietary diversity and quality, addresses the limitations of traditional measures. In this longitudinal study, higher adherence to the HFD Index was linked to contrasting stroke risks by sex, suggesting that dietary guidelines should account for sex-specific and age-specific factors.