Levels Study Background

Study Background and Justification

Study Background and Justification

Standard Glucose Testing

Metabolic health has historically been evaluated using fasted blood glucose (BG), oral glucose tolerance, and glycated hemoglobin (HbA1c). However, these tests do not capture real-world BG dynamics and their relation to health and behavior. Controlled testing conditions like fasting and oral glucose challenges allow for standardization in screening for existing glucose dysfunction, but do not capture the earliest deviations from healthy glycemic patterns that an individual may exhibit prior to overt manifestation of disease.

Fasting BG is used to diagnose normal, pre-diabetic, and diabetic status. This test leaves the vast majority of glycemia unevaluated as it is a single time point, repeated at annual or longer intervals in people without diabetes1. As fasting BG elevation may not occur until overt metabolic dysfunction is established, the earliest window for intervention may be missed.

The Oral Glucose Tolerance Test (OGTT) measures the metabolic response to a standardized oral intake of glucose (up to ~100g)2,3. While this test captures the glucose dynamics that inform status/onset of insulin resistance, beta-cell dysfunction, etc., it has several limitations. The OGTT assumes that all individuals are adapted to a high-carbohydrate diet2. Low-carbohydrate diets reduce metabolic reliance on glucose and increase fat oxidation efficiency4, and even the lack of a single high-carbohydrate meal prior to the OGTT can negatively impact test results5. Thus, impaired glucose tolerance on low-carbohydrate diets may not reflect poor metabolic health, but physiologic adaptation4. Additionally, like fasted BG, the OGTT does not capture the real-world behaviors and metabolic responses. Together, the OGTT provides a great tool for establishing metabolic response to a specific challenge, but provides a limited window into an individual’s response to glycemic challenges experienced in real life and may be confounded by recent and long-term dietary choices.

HbA1c evaluates the end product of hemoglobin glycation, which is associated with serum BG exposure over the lifespan of red blood cells, about 3-4 months6. Although HbA1c is a gold standard for assessing diabetes risk, and for diabetes management, it is not always elevated in concert with fasting BG or the OGTT7. Additionally, factors including iron and vitamin b-12 levels,  and genetic variants can confound the interpretation of an individual’s HbA1c8. This has led to the development of other tests for chronic glycemic dysregulation (e.g., fructosamine9), and demonstrates a need for more effective tools to evaluate glycemia at the level of the individual.

Continuous Glucose Monitors

Continuous glucose monitors (CGM) provide 5-minute resolution, 24-hour blood glucose using an enzymatic reaction that reflects glucose in the interstitial fluid (intermediate location) of reach peripheral cellular tissues (end product and uptake location). CGM also captures 24-hour/day glycemic dynamics, including circadian rhythms and postprandial responses10,11, that are more reflective of metabolic status than a single data capture time point (Fasting BG) or window (OGTT). This enhanced resolution has been the reason for recent shifts toward using CGM-based dynamics as more accurate measures of glycemic control in diabetic populations12.

Study Justification

Recent improvements in CGM device functionality, form factor, and cost, have greatly increased adoption among people with diabetes13, with associated improvements in subjective stress14, glycemia12, and management burden15. The potential benefits of metabolic awareness extend beyond people with diabetes: technology that can provide closed-loop feedback on the impact of lifestyle choices on glucose could support individuals across the spectrum of metabolic health. Researchers have begun to characterize continuous glucose dynamics in healthy participants to serve as a benchmark from which to understand progression to disease. However, few studies (<20) have evaluated continuous glucose in heterogeneous populations, and none have observed large populations (n>10,000) to generate reference glycemic values, patterns, usage analysis in people without diabetes. Most have had small sample sizes (tens, or, very occasionally, hundreds)16–18or have used early-generation CGMs17,19,20 with lower accuracy21. Many of these evaluations also involve controlled laboratory tests and interventions, which further influence participant glycemic control.

CGM data in controlled settings does not reflect the rapidly emerging use case of individuals wearing CGMs in everyday life, and in concert with insight-driven software for personal metabolic awareness and lifestyle improvement. Personally-relevant education is a key component of the movement toward preventative-CGM use, predicated on the idea that a lack of real-time input available to the individual about the relationship between one’s lifestyle and metabolic function contributes to the epidemic of metabolic disease. A personal feedback mechanism like informational CGM could greatly increase awareness of the impacts of daily choices on glycemia.

However, glycemic patterns under these circumstances have never been studied at scale and may:

1. Generate important discoveries around patterns of glycemia in the general population.
2. Evaluate the utility of the CGM & associated software as a tool for general metabolic education & behavior change
3. Evaluate associations among use of the CGM & software and health outcomes.

We propose that these tools can link understanding of the impact of one’s food choices, physical activity, sleep, and stress on glycemia and, with guidance, encourage healthier choices22. Though outcomes for this population have not yet been demonstrated in long-term studies, short term studies report automated screening for metabolic risk in agreement with traditional tests23, detection of prediabetic/diabetic glucose levels in people without diabetes24, clear impacts of meal composition19,25,26–28, high user acceptability17, and minimal adverse effects19,22–24.

Over long periods, real-time glycemic-response feedback available from a guided CGM experience could inform and motivate positive behavior change around nutrition, sleep, and exercise. Several studies have used CGM data to observe significant improvements in postprandial glucose levels18–20 and glycemic variability29 with relatively simple adjustments in macronutrient composition and order of consumption18,19,25. One study used CGM data to determine optimal timing for bouts of exercise to manage postprandial glucose response in people at risk for developing diabetes28, while another showed a positive correlation between light activity and reduced glycemic variability among lower-fitness subjects12. A number of recent and ongoing studies have demonstrated evidence that CGM data can be used to predict personalized meal responses and the interpersonal variability present in glycemic patterns with standardized meals and behaviors18–20,30,31. These findings are echoed in online reports from non-diabetic CGM users, who have discovered large meal-dependent variations in postprandial glucose along with negative effects from sleep-deprivation and other stressors32,33.

The epidemic of diabetes was estimated to cost the US over $327 billion in 201716, with both rates of disease and cost of treatment climbing year after year.16,34 The Diabetes Prevention Program Research Group called for a shift in response in order to reverse these trends, stating that: “methods of treating diabetes remain inadequate, prevention is preferable.35”. Ten years later, CGMs demonstrably improve diabetes treatment and reversal, and show early promise for prevention. Though untested at scale as a preventative measure, a CGM combined with personalized guidance is a low-risk technology, ideal for use in wellness education and research proposed here. If CGMs become a widespread wellness tool, the benefits of improved individual metabolic awareness and informed lifestyle decisions could have compounding effects at a larger societal scale. Data-driven consumers opting for healthier food and drink options could shift the demand balance, influencing suppliers to modify their offerings and driving down the cost of healthier food items. Wider adoption could also contribute to medical research, as large anonymized and aggregated CGM datasets would provide unprecedented insight into full-spectrum glycemia, generate a clear definition of “healthy” glucose regulation, and improve understanding of the onset, progression, and treatment targets for metabolic disorders.

The proposed study and the follow-on research it could inform will be groundbreaking, in that it will be large in scale and the first to assess some or all of the following:

• Reference glucose ranges in general population adults using unblinded CGM and associated software for wellness purposes. This represents real world data for a rapidly emerging CGM use case that has yet to be studied meaningfully.
  • This dataset will serve as the first and largest of its kind, contributing to an early understanding of free-living glucose metrics in individuals viewing and interacting with real-time glucose data as a wellness tool.
  • The dataset will be a resource for generating hypotheses for future studies that may focus on the efficacy and effectiveness of CGM in implementations beyond diabetes management.
• This data set will comprise the largest resource we know of for assessment of adverse metabolic events in the general population, many times larger than any known prior study employing CGMs in the general population. This will lay the foundation for future work related to safety and real-world evidence of CGM use for this population.
• Safety and effectiveness resource for a CGM system in the general population.
• Levels' remote monitoring software can also provide dynamic information to determine how demographic, anthropometric, and lifestyle data influence these values.

Bibliography

1. MD, R. H. S. Is blood sugar monitoring without diabetes worthwhile? Harvard Health https://www.health.harvard.edu/blog/is-blood-sugar-monitoring-without-diabetes-worthwhile-202106112473 (2021).
2. Eyth, E., Basit, H. & Smith, C. J. Glucose Tolerance Test. in StatPearls (StatPearls Publishing, 2022).
3. Jagannathan, R. et al. The Oral Glucose Tolerance Test: 100 Years Later. Diabetes Metab. Syndr. Obes. Targets Ther. 13, 3787–3805 (2020).
4. Kinzig, K. P., Honors, M. A. & Hargrave, S. L. Insulin sensitivity and glucose tolerance are altered by maintenance on a ketogenic diet. Endocrinology 151, 3105–3114 (2010).
5. Klein, K. R. et al. Carbohydrate Intake Prior to Oral Glucose Tolerance Testing. J. Endocr. Soc. 5, bvab049 (2021).
6. Weykamp, C. HbA1c: A Review of Analytical and Clinical Aspects. Ann. Lab. Med. 33, 393–400 (2013).
7. Lu, J. et al. Predictive Value of Fasting Glucose, Postload Glucose, and Hemoglobin A1c on Risk of Diabetes and Complications in Chinese Adults. Diabetes Care 42, 1539–1548 (2019).
8. Information, N. C. for B., Pike, U. S. N. L. of M. 8600 R., MD, B. & Usa, 20894. Some of the factors that influence HbA1c and its measurement. Use of Glycated Haemoglobin (HbA1c) in the Diagnosis of Diabetes Mellitus: Abbreviated Report of a WHO Consultation (World Health Organization, 2011).
9. Nansseu, J. R. N. et al. Fructosamine measurement for diabetes mellitus diagnosis and monitoring: a systematic review and meta-analysis protocol. BMJ Open 5, e007689 (2015).
10. Bandín, C. et al. Meal timing affects glucose tolerance, substrate oxidation and circadian-related variables: A randomized, crossover trial. Int. J. Obes. 2005 39, 828–833 (2015).
11. Mejean, L. et al. Circadian and Ultradian Rhythms in Blood Glucose and Plasma Insulin of Healthy Adults. Chronobiol. Int. 5, 227–236 (1988).
12. Rodriguez-Segade, S. et al. Continuous glucose monitoring is more sensitive than HbA1c and fasting glucose in detecting dysglycaemia in a Spanish population without diabetes. Diabetes Res. Clin. Pract. 142, 100–109 (2018).
13. DeSalvo, D. J. et al. Continuous glucose monitoring and glycemic control among youth with type 1 diabetes: International comparison from the T1D Exchange and DPV Initiative. Pediatr. Diabetes 19, 1271–1275 (2018).
14. Vesco, A. T., Jedraszko, A. M., Garza, K. P. & Weissberg-Benchell, J. Continuous Glucose Monitoring Associated With Less Diabetes-Specific Emotional Distress and Lower A1c Among Adolescents With Type 1 Diabetes. J. Diabetes Sci. Technol. 12, 792–799 (2018).
15. Lin, R., Brown, F., James, S., Jones, J. & Ekinci, E. Continuous glucose monitoring: A review of the evidence in type 1 and 2 diabetes mellitus. Diabet. Med. J. Br. Diabet. Assoc. 38, e14528 (2021).
16. Shah, V. N. et al. Continuous Glucose Monitoring Profiles in Healthy Nondiabetic Participants: A Multicenter Prospective Study. J. Clin. Endocrinol. Metab. 104, 4356–4364 (2019).
17. Liao, Y. & Schembre, S. Acceptability of Continuous Glucose Monitoring in Free-Living Healthy Individuals: Implications for the Use of Wearable Biosensors in Diet and Physical Activity Research. JMIR MHealth UHealth 6, e11181 (2018).
18. Zeevi, D. et al. Personalized Nutrition by Prediction of Glycemic Responses. Cell 163, 1079–1094 (2015).
19. Freckmann, G. et al. Continuous Glucose Profiles in Healthy Subjects under Everyday Life Conditions and after Different Meals. J. Diabetes Sci. Technol. Online 1, 695–703 (2007).
20. Derosa, G. et al. Continuous glucose monitoring system in free-living healthy subjects: results from a pilot study. Diabetes Technol. Ther. 11, 159–169 (2009).
21. Bailey, T. S. Clinical Implications of Accuracy Measurements of Continuous Glucose Sensors. Diabetes Technol. Ther. 19, S-51-S-54 (2017).
22. Ehrhardt, N. & Al Zaghal, E. Continuous Glucose Monitoring As a Behavior Modification Tool. Clin. Diabetes Publ. Am. Diabetes Assoc. 38, 126–131 (2020).
23. Chan, C. L. et al. Continuous Glucose Monitoring and its Relationship to Hemoglobin A1c and Oral Glucose Tolerance Testing in Obese and Prediabetic Youth. J. Clin. Endocrinol. Metab. 100, 902–910 (2015).
24. Borg, R. et al. Real-life glycaemic profiles in non-diabetic individuals with low fasting glucose and normal HbA1c: the A1C-Derived Average Glucose (ADAG) study. Diabetologia 53, 1608–1611 (2010).
25. González-Rodríguez, M. et al. Postprandial glycemic response in a non-diabetic adult population: the effect of nutrients is different between men and women. Nutr. Metab. 16, 46 (2019).
26. Higa, M. et al. Effect of High β-glucan Barley on Postprandial Blood Glucose Levels in Subjects with Normal Glucose Tolerance: Assessment by Meal Tolerance Test and Continuous Glucose Monitoring System. Clin. Nutr. Res. 8, 55–63 (2019).
27. Shukla, A. P. et al. The impact of food order on postprandial glycemic excursions in prediabetes. Diabetes Obes. Metab. 21, 377–381 (2019).
28. DiPietro, L., Gribok, A., Stevens, M. S., Hamm, L. F. & Rumpler, W. Three 15-min Bouts of Moderate Postmeal Walking Significantly Improves 24-h Glycemic Control in Older People at Risk for Impaired Glucose Tolerance. Diabetes Care 36, 3262–3268 (2013).
29. Avari, P., Moscardo, V., Jugnee, N., Oliver, N. & Reddy, M. Glycemic Variability and Hypoglycemic Excursions With Continuous Glucose Monitoring Compared to Intermittently Scanned Continuous Glucose Monitoring in Adults With Highest Risk Type 1 Diabetes. J. Diabetes Sci. Technol. 14, 567–574 (2020).
30. Hall, H. et al. Glucotypes reveal new patterns of glucose dysregulation. PLOS Biol. 16, e2005143 (2018).
31. Brynes, Audrey E et al. “The beneficial effect of a diet with low glycaemic index on 24 h glucose profiles in healthy young people as assessed by continuous glucose monitoring.” The British journal of nutrition vol. 93,2 (2005): 179-82.
32. [Using Freestyle Libre To See How Stress Effects My Blood Glucose Levels. Quantified Self (2018).](https://quantifiedself.com/blog/using-freestyle-libre-see-stress-effects-blood-glucose-levels/ (2018).)
33. QS Guide: Testing Food with Blood Glucose. Quantified Self (2019).
34. Kingsnorth, A. P., Whelan, M. E., Sanders, J. P., Sherar, L. B. & Esliger, D. W. Using Digital Health Technologies to Understand the Association Between Movement Behaviors and Interstitial Glucose: Exploratory Analysis. JMIR MHealth UHealth 6, e9471 (2018).
35. Guy’s and St Thomas’ NHS Foundation Trust. Predicting Inter-individual Differences in Biochemical and Behavioral Response to Meals With Different Nutritional Compositions Using Metabolomic and Microbiome Profiling (2021).