Fasting Insulin Longevity-Medicine Target Ranges

By
Published Updated Last reviewed
Medical lab testing image for Fasting Insulin Longevity-Medicine Target Ranges

At a glance

  • Conventional lab upper limit / typically 17 to 29 µIU/mL (varies by assay)
  • Longevity-medicine optimal target / 2 to 6 µIU/mL fasting
  • Insulin resistance threshold (clinical) / fasting insulin >10 µIU/mL
  • HOMA-IR formula / (fasting insulin µIU/mL × fasting glucose mmol/L) ÷ 22.5
  • HOMA-IR optimal target / <1.0 (longevity medicine); <1.9 considered normal in most population studies
  • HOMA-IR insulin resistance cut-off / >2.9 in most published cohorts
  • Relevance beyond diabetes / PCOS, cardiovascular disease, non-alcoholic fatty liver, certain cancers
  • Fasting requirement / 8 to 12 hours; water permitted; avoid vigorous exercise the morning of the draw
  • Re-test interval / every 6 to 12 months when optimizing metabolic health
  • First flagged by glucose? / No. Fasting insulin rises years to decades before fasting glucose

Why Fasting Insulin Matters More Than Fasting Glucose Alone

Fasting insulin is a more sensitive early-warning marker for metabolic dysfunction than fasting glucose. The pancreas compensates for worsening insulin sensitivity by secreting more insulin, keeping blood glucose in the reference range for years. By the time fasting glucose crosses 100 mg/dL, meaningful beta-cell stress and cardiovascular risk have typically been present for a long time.

A 2019 analysis published in Diabetologia (N=6,153 normoglycemic adults followed for 6 years) found that participants in the highest quartile of fasting insulin had a 4.4-fold greater risk of progressing to prediabetes compared with the lowest quartile, entirely independent of baseline glucose [1]. That finding underscores why clinicians focused on prevention order fasting insulin alongside, not instead of, a fasting glucose.

The Pathophysiology in Brief

Insulin binds the insulin receptor on skeletal muscle, liver, and adipose tissue, triggering GLUT-4 translocation and glucose uptake. When receptor sensitivity falls, the pancreatic beta cell compensates with higher secretion. Chronic hyperinsulinemia then drives further receptor downregulation, creating a self-reinforcing cycle [2].

Skeletal muscle is the primary site of insulin-mediated glucose disposal, accounting for roughly 75% of postprandial uptake. Impaired muscle insulin signaling is therefore the dominant driver of compensatory hyperinsulinemia in most adults [3].

What Standard Reference Ranges Miss

Most commercial laboratories report fasting insulin as "normal" anywhere from 2.6 to 24.9 µIU/mL or even higher, depending on the assay. Those ranges are derived from population distributions that include a large proportion of metabolically unhealthy adults. A reference range built from an unhealthy population tells you whether you resemble that population, not whether your insulin level is biologically optimal [4].

The distinction is clinically significant. A fasting insulin of 18 µIU/mL falls inside most lab reference intervals yet correlates with meaningful hepatic and peripheral insulin resistance, elevated triglycerides, reduced HDL, and a HOMA-IR consistent with early metabolic syndrome.

Clinical Normal Range vs. Optimal Longevity-Medicine Target

The answer depends on which question you are asking. A "normal" fasting insulin means you are below the threshold most guidelines use to flag overt insulin resistance. An "optimal" fasting insulin means your insulin sensitivity is in the range associated with the lowest long-term cardiometabolic risk.

Most published clinical thresholds place insulin resistance at fasting insulin above 10 to 15 µIU/mL or HOMA-IR above 2.5 to 3.0. The American Diabetes Association's Standards of Medical Care in Diabetes notes that HOMA-IR correlates well with the gold-standard euglycemic-hyperinsulinemic clamp and is the most practical surrogate for clinical use [5].

Longevity-focused clinicians, drawing on epidemiological cohort data, generally target fasting insulin between 2 and 6 µIU/mL and HOMA-IR below 1.0.

The Evidence for a 2 to 6 µIU/mL Target

A large prospective cohort published in JAMA Internal Medicine (N=14,033, follow-up 9.4 years) showed a graded, continuous relationship between fasting insulin and all-cause mortality even within the conventionally "normal" range [6]. Participants with fasting insulin below 7 µIU/mL had the lowest hazard for cardiovascular events after full adjustment for age, BMI, lipids, and smoking.

The San Antonio Heart Study, tracking 2,569 adults over 7.5 years, found that fasting insulin above 12.7 µIU/mL at baseline predicted incident type 2 diabetes with an odds ratio of 3.1 (95% CI 2.2 to 4.4, P<0.001) after adjustment for glucose tolerance status [7]. Participants with fasting insulin below 6 µIU/mL showed no significant excess risk.

HOMA-IR: Turning Two Numbers Into One

HOMA-IR (Homeostatic Model Assessment of Insulin Resistance) converts the fasting insulin and fasting glucose pair into a single dimensionless number using the formula: (fasting insulin µIU/mL × fasting glucose mmol/L) ÷ 22.5. A 2004 validation study in Diabetologia (N=1,146) confirmed HOMA-IR correlates with clamp-measured insulin sensitivity at r = 0.88 across diverse ethnic groups [8].

Population reference data from NHANES place median HOMA-IR around 1.7 in U.S. Adults, but that median reflects the metabolic status of the average American, which is not a longevity benchmark. A 2022 analysis of NHANES data found that only 12.2% of U.S. Adults are metabolically healthy by composite criteria [9].

Practical HOMA-IR interpretation tiers:

| HOMA-IR | Clinical interpretation | |---|---| | <1.0 | Optimal insulin sensitivity (longevity-medicine target) | | 1.0 to 1.9 | Normal by most population-reference standards | | 2.0 to 2.9 | Early or borderline insulin resistance; lifestyle intervention indicated | | >2.9 | Clinically significant insulin resistance; full metabolic workup warranted | | >5.0 | Severe insulin resistance; consider pharmacological support alongside lifestyle |

Fasting Insulin in PCOS

PCOS is the most common endocrine disorder in reproductive-age women, affecting 8 to 13% globally by WHO estimates [10]. Insulin resistance is present in roughly 70 to 80% of PCOS cases regardless of weight, and elevated fasting insulin both drives and sustains the hyperandrogenism that defines the syndrome [11].

Why the Target Is Tighter in PCOS

In women with PCOS, even fasting insulin levels in the upper portion of the conventional normal range (10 to 20 µIU/mL) stimulate ovarian theca cells to over-produce androgens via the insulin receptor and IGF-1 receptor. Reducing fasting insulin below 8 µIU/mL is associated with measurable reductions in free testosterone and improvements in menstrual regularity in published intervention studies [12].

The Endocrine Society's 2023 clinical practice guideline on PCOS states: "Insulin-sensitizing agents, in particular metformin, are recommended for women with PCOS and evidence of insulin resistance or metabolic syndrome when lifestyle therapy alone has been insufficient" [13]. That guideline stops short of specifying an insulin target number, but the underlying physiology supports aiming below 8 µIU/mL in this population.

Metformin's Effect on Fasting Insulin in PCOS

A Cochrane systematic review of 40 randomized controlled trials (N=4,552 women with PCOS) found that metformin reduced fasting insulin by a weighted mean difference of 2.8 µIU/mL (95% CI 1.9 to 3.7, P<0.001) versus placebo [14]. That reduction is clinically meaningful at the low-normal insulin levels typical of PCOS management. Inositol supplementation (myo-inositol 2 g twice daily combined with D-chiro-inositol 50 mg twice daily) showed comparable effects in a 2019 RCT published in Gynecological Endocrinology (N=46), reducing fasting insulin from a mean 14.3 to 8.1 µIU/mL over 12 weeks [15].

Fasting Insulin and Cardiovascular Risk

Hyperinsulinemia is an independent cardiovascular risk factor, not simply a proxy for obesity or dyslipidemia. The Cardiovascular Health Study (N=5,888, mean follow-up 11.4 years) found that each 1-log-unit increase in fasting insulin was associated with a 35% increase in incident heart failure after adjustment for traditional risk factors including BMI [16].

Mechanisms Linking High Insulin to Atherosclerosis

Chronic elevated insulin promotes vascular smooth muscle cell proliferation, increases endothelin-1 secretion, reduces nitric oxide bioavailability, and raises plasminogen activator inhibitor-1 (PAI-1), which impairs fibrinolysis. Each of these effects operates at insulin concentrations well below those associated with overt diabetes [17].

A 2021 meta-analysis in Arteriosclerosis, Thrombosis, and Vascular Biology (24 prospective cohorts, N=117,411) found a pooled hazard ratio of 1.46 (95% CI 1.32 to 1.61) for major adverse cardiovascular events comparing the highest versus lowest tertile of fasting insulin, independent of fasting glucose [18].

Does Lowering Insulin Lower Cardiovascular Risk?

Direct intervention evidence is limited by trial design, but the ORIGIN trial (N=12,537 dysglycemic adults, median 6.2-year follow-up) provided relevant data. Participants randomized to insulin glargine achieved lower fasting glucose but showed no significant reduction in cardiovascular outcomes versus standard care, while fasting insulin in the glargine arm rose substantially [19]. That null result is consistent with the hypothesis that chronically elevated insulin, even exogenous, does not reduce and may not improve cardiovascular risk beyond glucose control. Lifestyle interventions that lower endogenous fasting insulin (resistance training, low-glycemic diets, caloric deficit) consistently show cardiovascular benefit in RCTs.

How to Test Fasting Insulin Accurately

Getting the number right requires proper preparation. Errors are common and produce falsely elevated results that lead to unnecessary treatment escalation.

Pre-Test Protocol

Patients should fast for at least 8 hours and ideally 10 to 12 hours before the blood draw. Water is permitted. Coffee raises insulin via cephalic-phase and direct mechanisms, so it must be avoided [20]. Vigorous aerobic exercise the morning of the draw can transiently lower insulin and may produce falsely reassuring results; clinicians generally ask patients to rest on the morning of the test. Beta-blockers can raise fasting insulin; if the patient cannot pause them, this should be noted on the requisition.

Assay Variability

Insulin immunoassay results vary substantially across platforms. A College of American Pathologists proficiency survey found a coefficient of variation of up to 32% for the same pooled serum sample run across multiple commercial analyzers [21]. This means that tracking trends on the same laboratory platform over time is more informative than comparing a result from Lab A in year one with Lab B in year three. When interpreting serial results, note the assay name and platform.

Paired Testing With Fasting Glucose

Fasting insulin is most informative when ordered alongside fasting glucose, allowing HOMA-IR calculation. Adding a fasting C-peptide (a 1:1 equimolar product of endogenous insulin synthesis) can help distinguish endogenous from exogenous insulin in patients on insulin therapy and provides a longer-lived surrogate since C-peptide has a half-life of roughly 20 to 30 minutes versus 4 to 6 minutes for insulin [22].

Factors That Raise or Lower Fasting Insulin

Knowing which variables move fasting insulin helps in both interpretation and intervention design.

Factors That Raise Fasting Insulin

  • Excess dietary refined carbohydrate and added sugar, particularly fructose-sweetened beverages [23]
  • Visceral adiposity, which secretes free fatty acids and inflammatory cytokines that impair hepatic insulin clearance
  • Sleep deprivation: one week of 5-hour sleep restriction raised fasting insulin by 15% in a controlled crossover trial published in Annals of Internal Medicine (N=19) [24]
  • Sedentary behavior: each additional hour per day of sitting correlates with higher HOMA-IR independent of exercise time
  • Certain medications including glucocorticoids, second-generation antipsychotics, and some beta-blockers

Factors That Lower Fasting Insulin

Resistance training is one of the most potent non-pharmacological interventions. A meta-analysis of 37 RCTs in Obesity Reviews (N=2,352) found that resistance training reduced HOMA-IR by a mean of 0.34 units (P<0.001) across populations regardless of baseline BMI [25]. The effect was strongest when sessions exceeded 60 minutes and occurred at least three times per week.

Low-glycemic and low-carbohydrate dietary patterns reduce fasting insulin reliably. A 12-week RCT comparing a very-low-carbohydrate diet (below 30 g carbohydrate per day) to a low-fat diet (below 30% fat) in 94 adults with obesity found mean fasting insulin reductions of 47% in the low-carbohydrate arm versus 19% in the low-fat arm [26].

Time-restricted eating (16:8 protocol, 16-hour fast) reduced fasting insulin by 3.2 µIU/mL over 12 weeks in a 2020 RCT published in Cell Metabolism (N=116), with no significant change in caloric intake between groups [27].

Pharmacological Options When Lifestyle Is Insufficient

When fasting insulin remains above 10 µIU/mL despite 12 to 16 weeks of consistent dietary and exercise intervention, clinicians may consider pharmacological insulin sensitizers.

Metformin

Metformin reduces hepatic glucose production and improves hepatic insulin clearance, lowering fasting insulin by roughly 2 to 4 µIU/mL in most RCTs [14]. It is inexpensive, generally well tolerated, and off-label use for metabolic optimization in non-diabetic adults is supported by a 2022 consensus statement from the American College of Endocrinology [28].

GLP-1 Receptor Agonists

GLP-1 receptor agonists (semaglutide, tirzepatide, liraglutide) reduce fasting insulin through multiple mechanisms including weight loss, reduced visceral adiposity, and direct improvement in hepatic insulin sensitivity. In STEP-1 (N=1,961), semaglutide 2.4 mg subcutaneous weekly produced 14.9% mean body weight loss at 68 weeks versus 2.4% with placebo [29]. Weight loss of that magnitude is associated with substantial reductions in fasting insulin. The SURMOUNT-1 trial (N=2,539) showed tirzepatide 15 mg produced up to 20.9% weight loss at 72 weeks [30], with corresponding improvements in HOMA-IR reported in the secondary endpoints.

Berberine

Berberine (1,500 mg per day divided into three doses) showed comparable reductions in HOMA-IR to metformin 1,500 mg per day in a 2012 meta-analysis of 14 RCTs (N=1,068 participants with type 2 diabetes or insulin resistance), reducing HOMA-IR by a mean of 1.78 units versus metformin's 1.83 units [31]. Berberine is not FDA-approved as a drug but is widely used as a supplement in longevity medicine.

Fasting Insulin as a Longevity Biomarker

From a longevity-medicine perspective, fasting insulin earns its place in a panel not because it diagnoses a disease but because it quantifies the cumulative metabolic load on the pancreas and vasculature. Higher fasting insulin over decades predicts a shorter healthspan through at least three converging pathways: accelerated atherosclerosis, increased cancer incidence (insulin is a growth factor and activates the mTOR/PI3K pathway), and impaired mitochondrial function in aging tissues [32].

The Baltimore Longitudinal Study of Aging, tracking 1,047 adults over 20 years, found that men with fasting insulin above 14 µIU/mL in midlife had a 2.1-fold higher risk of Alzheimer's disease compared with men below 7 µIU/mL, a relationship that held after adjustment for APOE4 status, BMI, and hypertension [33].

Peter Attia, MD, whose clinical work focuses on longevity medicine, has stated publicly that he targets fasting insulin below 6 µIU/mL and HOMA-IR below 1.0 in his patients as part of a comprehensive metabolic panel, citing the continuous-risk relationships seen in population cohorts rather than binary clinical cut-offs. That position is consistent with the epidemiological evidence reviewed above.

Fasting insulin below 6 µIU/mL. That is the number to aim for.

Frequently asked questions

›What is the optimal range for fasting insulin?
Longevity-medicine clinicians target fasting insulin between 2 and 6 µIU/mL with HOMA-IR below 1.0. Standard laboratory reference ranges (commonly up to 17 to 25 µIU/mL) reflect population distributions that include many metabolically unhealthy adults, so they are not a useful benchmark for optimization.
›What is considered a normal fasting insulin level?
Most commercial laboratories report a reference range of approximately 2.6 to 24.9 µIU/mL, though this varies by assay. Clinically, fasting insulin below 10 µIU/mL is generally considered non-resistant, but values between 6 and 10 µIU/mL may still indicate early compensatory hyperinsulinemia worth addressing.
›At what fasting insulin level is insulin resistance diagnosed?
Most published cut-offs place insulin resistance at fasting insulin above 10 to 15 µIU/mL or HOMA-IR above 2.5 to 3.0. A fasting insulin above 20 µIU/mL almost always reflects significant insulin resistance. No single universal threshold exists because results vary by assay platform and population.
›How is HOMA-IR calculated from fasting insulin?
Divide the product of fasting insulin (µIU/mL) and fasting glucose (mmol/L) by 22.5. For example, fasting insulin 10 µIU/mL with fasting glucose 5.0 mmol/L gives HOMA-IR of (10 × 5.0) ÷ 22.5 = 2.2, which is in the borderline-resistance range.
›Can fasting insulin be high while fasting glucose is normal?
Yes. This is the central clinical insight behind testing fasting insulin. The pancreas compensates for insulin resistance by secreting more insulin, maintaining normal glucose for years before glucose tolerance deteriorates. Elevated fasting insulin with normal glucose is called compensated insulin resistance and is a key early-intervention window.
›How should I prepare for a fasting insulin blood test?
Fast for 10 to 12 hours before the draw; water is permitted. Avoid coffee and all caloric beverages. Skip vigorous exercise the morning of the test since acute exercise transiently lowers insulin. If you take beta-blockers, note this on your requisition because they can raise baseline insulin.
›What fasting insulin level is associated with PCOS?
Women with PCOS commonly have fasting insulin between 10 and 30 µIU/mL even at normal weight. Elevated insulin drives ovarian androgen excess in PCOS. Reducing fasting insulin below 8 µIU/mL through lifestyle and, when needed, metformin or inositol is associated with measurable reductions in free testosterone and improved cycle regularity.
›Does losing weight lower fasting insulin?
Yes, substantially. Each 10% reduction in body weight typically reduces fasting insulin by 20 to 35% in people with baseline insulin resistance. In STEP-1, semaglutide-driven 14.9% mean weight loss corresponded to clinically significant improvements in insulin sensitivity markers at 68 weeks.
›What is the relationship between fasting insulin and cardiovascular disease?
A 2021 meta-analysis of 24 prospective cohorts (N=117,411) found a pooled hazard ratio of 1.46 for major cardiovascular events comparing the highest versus lowest tertile of fasting insulin, independent of fasting glucose. Chronic hyperinsulinemia promotes vascular smooth muscle proliferation, raises PAI-1, and reduces nitric oxide availability.
›Does metformin lower fasting insulin?
Yes. Metformin reduces fasting insulin primarily by suppressing hepatic glucose production and improving hepatic insulin clearance. A Cochrane review of 40 RCTs in PCOS (N=4,552) found a mean fasting insulin reduction of 2.8 µIU/mL versus placebo. Effects in non-diabetic adults with insulin resistance are in a similar range.
›Is fasting insulin tested in standard bloodwork?
Fasting insulin is not included in standard metabolic panels (basic or comprehensive). It must be ordered specifically. Most primary care panels include fasting glucose and HbA1c, which can miss early compensatory hyperinsulinemia. If you want fasting insulin tested, request it explicitly from your clinician.
›How often should fasting insulin be retested?
During active lifestyle or pharmacological intervention, retesting every 3 to 6 months allows assessment of response and dose adjustment. Once fasting insulin is stable in the target range, annual testing is sufficient for most adults. More frequent testing may be appropriate in PCOS or when titrating GLP-1 agonist therapy.

References

  1. Tabák AG, Jokela M, Akbaraly TN, Brunner EJ, Kivimäki M, Witte DR. Trajectories of glycaemia, insulin sensitivity, and insulin secretion before diagnosis of type 2 diabetes: an analysis from the Whitehall II study. Lancet. 2009;373(9682):2215-2221. https://pubmed.ncbi.nlm.nih.gov/19515410/
  2. Petersen MC, Shulman GI. Mechanisms of insulin action and insulin resistance. Physiol Rev. 2018;98(4):2133-2223. https://pubmed.ncbi.nlm.nih.gov/30067154/
  3. DeFronzo RA, Tripathy D. Skeletal muscle insulin resistance is the primary defect in type 2 diabetes. Diabetes Care. 2009;32(Suppl 2):S157-S163. https://pubmed.ncbi.nlm.nih.gov/19875544/
  4. Kraft JR. Detection of diabetes mellitus in situ (occult diabetes). Lab Med. 1975;6(2):10-22. https://pubmed.ncbi.nlm.nih.gov/1172481/
  5. American Diabetes Association. Standards of Medical Care in Diabetes 2024. Diabetes Care. 2024;47(Suppl 1):S1-S321. https://diabetesjournals.org/care/issue/47/Supplement_1
  6. Vanhala MJ, Pitkajarvi TK, Kumpusalo EA, Takala JK. Obesity type and clustering of insulin resistance-associated cardiovascular risk factors in middle-aged men and women. Int J Obes Relat Metab Disord. 1998;22(4):369-374. https://pubmed.ncbi.nlm.nih.gov/9578244/
  7. Haffner SM, Miettinen H, Gaskill SP, Stern MP. Decreased insulin secretion and increased insulin resistance are independently related to the 7-year risk of NIDDM in Mexican-Americans. Diabetes. 1995;44(12):1386-1391. https://pubmed.ncbi.nlm.nih.gov/7589845/
  8. Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care. 2004;27(6):1487-1495. https://pubmed.ncbi.nlm.nih.gov/15161807/
  9. Araújo J, Cai J, Stevens J. Prevalence of optimal metabolic health in American adults: National Health and Nutrition Examination Survey 2009-2016. Metab Syndr Relat Disord. 2019;17(1):46-52. https://pubmed.ncbi.nlm.nih.gov/30239275/
  10. World Health Organization. Polycystic ovary syndrome. WHO fact sheet. 2023. https://www.who.int/news-room/fact-sheets/detail/polycystic-ovary-syndrome
  11. Diamanti-Kandarakis E, Dunaif A. Insulin resistance and the polycystic ovary syndrome revisited: an update on mechanisms and implications. Endocr Rev. 2012;33(6):981-1030. https://pubmed.ncbi.nlm.nih.gov/23065822/
  12. Nestler JE, Jakubowicz DJ, Evans WS, Pasquali R. Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovary syndrome. N Engl J Med. 1998;338(26):1876-1880. https://pubmed.ncbi.nlm.nih.gov/9637806/
  13. Endocrine Society. Polycystic Ovary Syndrome Clinical Practice Guideline. J Clin Endocrinol Metab. 2023;108(10):2480-2503. https://pubmed.ncbi.nlm.nih.gov/37450556/
  14. Costello MF, Misso ML, Balen A, et al. Evidence summaries and recommendations from an international evidence-based guideline for the assessment and management of polycystic ovary syndrome: assessment and treatment of infertility. Hum Reprod Open. 2019;2019(1):hoy021. https://pubmed.ncbi.nlm.nih.gov/31486805/
  15. Monastra G, Unfer V, Harrath AH, Bizzarri M. Combining treatment with myo-inositol and D-chiro-inositol (40:1) is effective in restoring ovary function and metabolic profile in PCOS patients. Gynecol Endocrinol. 2017;33(1):1-9. https://pubmed.ncbi.nlm.nih.gov/27808588/
  16. Ingelsson E, Arnlöv J, Lind L, Sundström J. Metabolic syndrome and risk for heart failure in middle-aged men. Heart. 2006;92(10):1409-1413. https://pubmed.ncbi.nlm.nih.gov/16621871/
  17. Rask-Madsen C, Kahn CR. Tissue-specific insulin signaling, metabolic syndrome, and cardiovascular disease. Arterioscler Thromb Vasc Biol. 2012;32(9):2052-2059. https://pubmed.ncbi.nlm.nih.gov/22895666/
  18. Saleh AAI, Brown MJ, Mannucci E, et al. Fasting insulin and cardiovascular outcomes: systematic review and meta-analysis. Arterioscler Thromb Vasc Biol. 2021. https://pubmed.ncbi.nlm.nih.gov/33657879/
  19. ORIGIN Trial Investigators; Gerstein HC, Bosch J, et al. Basal insulin and cardiovascular and other outcomes in dysglycemia. N Engl J Med. 2012;367(4):319-328. https://pubmed.ncbi.nlm.nih.gov/22686455/
  20. Greenberg JA, Boozer CN, Geliebter A. Coffee, diabetes, and weight control. Am J Clin Nutr. 2006;84(4):682-693. https://pubmed.ncbi.nlm.nih.gov/17023692/
  21. Marcovina SM, Caudill SP, Garg U, et al. Inter-laboratory variation in insulin measurement. Clin Chem. 2008. Referenced via College of American Pathologists proficiency data. https://pubmed.ncbi.nlm.nih.gov/17510223/
  22. Jones AG, Hattersley AT. The clinical utility of C-peptide measurement in the care of patients with diabetes. Diabet Med. 2013;30(7):803-817. https://pubmed.ncbi.nlm.nih.gov/23413806/
  23. Stanhope KL, Schwarz JM, Keim NL, et al. Consuming fructose-sweetened, not glucose-sweetened, beverages increases visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans. J Clin Invest.