Cardiovascular disease and diabetes, significantly influenced by diet and lifestyle, are leading global causes of death. While dietary and lifestyle interventions can reduce disease risk, diet-related diseases continue to rise, potentially due to poor adherence to general guidelines and significant inter-individual variability in responses to food. Adherence to dietary guidelines is low in both the US and UK, highlighting the need for more effective approaches. Personalized nutrition programs, tailoring advice to individual biological and lifestyle factors, offer a promising alternative. This study investigated whether a personalized dietary program (PDP) would be superior to standard dietary advice in improving cardiometabolic health outcomes.

Existing literature reveals substantial variation in individual responses to dietary interventions. Studies like the DIETFITS trial demonstrated different weight loss responses to low-fat and low-carbohydrate diets, and research shows low adherence to general dietary guidelines. Previous studies on personalized nutrition have shown mixed results; some report improvements in dietary behaviors but not in weight or other clinically relevant outcomes. The Food4Me study showed improved dietary behaviors with personalized advice but no significant difference in body weight compared with a non-personalized group, and other studies showed no significant differences in weight loss between personalized and standard diets. These inconsistencies highlight the need for more comprehensive and precisely tailored approaches to personalized nutrition.

This 18-week, parallel-design, randomized controlled trial (RCT) was conducted remotely in the US. 347 participants (aged 41-70 years) representing a general US population were randomized to either a personalized dietary program (PDP, n=177) or a control group receiving standard dietary advice based on the 2020-2025 USDA Dietary Guidelines (n=170). The PDP utilized a 2022 algorithm incorporating food characteristics, individual postprandial glucose and triglyceride responses, microbiome composition, cardiovascular risk factors, and health history to generate personalized food scores. The control group received advice via online resources, check-ins, video lessons, and a leaflet. Assessments included baseline and 18-week measurements of serum triglycerides (TGs) and low-density lipoprotein cholesterol (LDL-C) (primary outcomes), along with secondary outcomes such as body weight, waist circumference, HbA1c, diet quality (HEI score), and microbiome composition. Data collection included venous blood draws, anthropometric measurements, questionnaires, and stool samples. The PDP involved a 4-week period of generalized advice followed by personalized recommendations and an action plan delivered through a mobile app. Statistical analyses were conducted using intention-to-treat and per-protocol analyses.

Intention-to-treat analysis revealed a significant reduction in triglycerides in the PDP group compared to the control group (-0.13 mmol/l, P=0.016). Changes in LDL-C were not significant. Secondary outcome analysis showed significantly greater improvements in the PDP group for body weight (-2.46 kg, P<0.05), waist circumference (-2.35 cm, P<0.05), HbA1c (-0.05%, P<0.05), and diet quality (7.08 points increase in HEI score, P<0.05). The PDP group also showed a greater and more sustained impact on gut microbiome composition (beta-diversity), with more favorable changes in microbial species associated with cardiometabolic health. Subgroup analysis indicated that highly adherent PDP participants experienced greater improvements in weight loss, approaching clinically meaningful levels (4.7% weight loss). The study also reported improvements in participant self-reported energy levels, sleep quality, mood, and reduced hunger levels in the PDP group. Post-hoc per-protocol analysis further confirmed the significant reduction in triglycerides in the PDP group.

This study provides evidence supporting the use of a PDP, incorporating multiple biological and lifestyle factors, for improving cardiometabolic health compared to standard dietary advice. The significant reduction in triglycerides, coupled with improvements in other cardiometabolic risk factors and microbiome composition, highlights the potential benefits of this personalized approach. The superior outcomes in highly adherent PDP participants emphasizes the importance of engagement and adherence in achieving optimal results. While the weight loss observed was moderate, it aligns with previous research showing that even modest weight loss can improve health outcomes and may be sustained long-term. The findings of this study contrast some previous research, which reported no difference in weight between personalized and standard diets, and this may be due to the multilevel personalization used in this study, including aspects of postprandial responses and microbiome composition.

This RCT demonstrates the efficacy of a personalized nutrition program in improving cardiometabolic health, particularly triglyceride levels. The incorporation of multiple biological and lifestyle factors into the PDP design proves superior to standard dietary guidelines. Future research should focus on expanding the diversity of participants, refining the prediction algorithm, and incorporating additional lifestyle factors to further enhance the effectiveness of personalized nutrition interventions.

Limitations include the inability to accurately capture physical activity levels and potential differences in contact or intensity between the intervention groups, which could have influenced the results. The generalizability of the findings may be limited by the primarily female and White participant makeup and the exclusion of children and older adults. Future studies should address these limitations to strengthen the evidence base for PDPs.