InsideTracker’s recent peer-reviewed study in Public Library of Science (PLOS) analyzed longitudinal data from users on its health AI platform. The retrospective study, which involved over 20,000 individuals and at least two blood draws, found associations between platform use and measurable improvements in key health indicators across blood biomarkers and wearable data.
The study, “Improvements in Blood and Fitness Tracker Biomarkers in a Longitudinal Real-World Cohort of Digital Health Platform Users,” looked at real-world data over time and identified statistically significant improvements across 43 biomarkers between users’ baseline and follow-up blood draws. The biomarkers involve those associated with cardiometabolic health, inflammation and nutrient status, such as LDL cholesterol, HbA1c, glucose, vitamin D, hsCRP and testosterone.
“InsideTracker is sitting on something genuinely rare: a longitudinal cohort that combines blood biomarkers, wearable data, DNA and health surveys across tens of thousands of people over several years,” said Renée Deehan, PhD, SVP of Science and AI at InsideTracker and one of the study’s investigators. “This study is one piece of what we expect to be a much larger body of work from that dataset. The improvements we saw across cardiac, insulin sensitivity and inflammatory-related biomarkers were meaningful.”
Key takeaways
Deehan told NutraIngredients that at a population level, most people were able to improve their biomarkers, then sustain those improvements over time. Researchers also observed improvements in markers such as LDL cholesterol, APOB and iron levels by the second blood draw, with many users either maintaining or further improving those results over subsequent tests.
The findings also highlight the role genetics may play in personalized health outcomes. According to Deehan, “individuals with a higher inherited risk for certain traits (e.g., high LDL-c) tended to show smaller improvements,” suggesting genetic predisposition may influence how strongly people respond to lifestyle interventions. She added this could help explain why two people with high cholesterol may respond differently to the same lifestyle changes.
Additionally, improvements in metabolic health markers among users with elevated HbA1c levels were also observed. A subgroup of participants with baseline HbA1c levels in the diabetic range saw levels decline over five longitudinal tests, with the average reaching prediabetic levels by the fifth blood draw.
Lipid markers more resistant to change
Unoptimized vitamin biomarker measurements were some of the easiest to improve, with Deehan noting “Sixty-three-point-nine percent of individuals with low iron, 63.2% of individuals with low folate, 49.8% of users with low magnesium and 57.1% of individuals with low vitamin D were able to improve by their second test. Interestingly, vitamin B12 was a little more difficult, with only 36.1% of users improving by their second test.”
Researchers found lipid biomarkers were the most difficult to improve, with only 20.4% of users with elevated LDL cholesterol and 27.3% of those with low HDL cholesterol showing improvement by their second test.
“This does make intuitive sense, both because supplements may be easier to implement than diet or exercise, but also because vitamin levels can be more rapidly corrected than lipid levels,” she said. “This latter point may contribute to the reason why it was relatively harder for users to move their vitamin B12 levels, because it can take longer to see changes in B12 consumption. Absorption challenges may also contribute.”
Supplement recommendations tailored to user data
Dietary supplement recommendations are not one-size-fits-all wellness solutions, but data-driven health interventions integrated into a broader personalized health platform.
Deehan said the recommendations are driven primarily by blood biomarkers, though genetics, lifestyle preferences and health goals are also considered. For example, users with low vitamin B12 levels may be recommended a B12 supplement, while alpha lipoic acid (ALA) may be suggested for elevated triglycerides only if iron levels are not already low. The platform also filters recommendations based on dietary preferences, such as avoiding whey protein recommendations for vegan users.
According to InsideTracker, the platform’s most frequently selected supplement recommendations include vitamin D, vitamin B12, alpha lipoic acid (ALA), psyllium and probiotics, reflecting strong interest in nutrient optimization, metabolic health and digestive support.
Source: Public Library of Science (PLOS); doi: 10.1371/journal.pdig.0001271; “Improvements in blood and fitness tracker biomarkers in a longitudinal real-world cohort of digital health platform users.” Authors: N. Schneider, et al.