Artificial Sweeteners and Diabetes: The Ultimate Guide

What you’re about to get into?

3000 words – 30 min read.

Key Points

Artificial sweeteners do not increase blood glucose levels.
The safety of artificial sweeteners compounds in use is supported by numerous international agencies and scientists in the field.
Artificial sweeteners support fat loss and are a suitable replacement for calorically sweetened beverages/foods.

Artificial Sweeteners and Diabetes: The Ultimate Guide

If you spend any time on social media or reading about nutrition and health online, then you’ve likely seen the hysteria:

“artificial sweeteners will give you cancer!”
“aspartame will kill your brain cells!”
“artificial sweeteners trick your brain into releasing insulin and spiking your blood glucose!”

And these are just gen pop headlines, what about for someone like you – a diabetic?

If there is even a shred of truth in there, what could this mean for both your blood sugar management and long-term health?

Are artificial sweeteners just thrown into the food supply in some conspiracy between governments and the food industry?

Does anyone actually care whether they’re safe or not?

Fortunately, we do: and we’ve cut through the BS to see what the science really says, and to give you the right advice for using artificially sweetened products safely and effectively.

So, leave the “internet guru” nutrition nonsense you’ve heard at the door, and let’s get into this…

Role of Artificial Sweeteners and Global Use in the Food Supply

Non-nutritive sweeteners have become really popular within the food industry as a strategy to help reduce the energy content of popular foods and drinks, in particular from sugar-sweetened beverages, across the population (1).

Note the term ‘non-nutritive’ above: this term covers both synthetically derived artificial sweeteners (AS), like aspartame, and naturally derived non-caloric sweeteners, like stevia. This article will focus on AS, in particular aspartame, acesulfame-K, sucralose, and saccharin, as they are the predominant non-nutritive sweeteners in use (2).

AS have unique and distinct qualities due to their chemical differences, and each individual compound varies in potency of sweetness, sweetness duration, mouth feel, and aftertaste (3). This allows the food industry to be quite flexible with the type of taste it wants from a product, and it’s also likely why some people love Coke and some love Pepsi.

Each AS also possesses unique pharmacokinetic profiles – how it is absorbed, distributed throughout the body, metabolised, and ultimately excreted – that means when we assess the health effects of AS, we give individual consideration to each one (3). So, where appropriate throughout, this article will refer specifically to the particular AS in question, not simply the overall category.

Why use AS?

Simply, AS provide an alternative to sugars commonly used in the food supply – sucrose in Europe or high-fructose corn syrup in the US – allowing the food industry to maintain palatability in a given product, without compromising on the sweetness which gives a product a particular quality (4). They are characterised by contributing little to no calories, while having a sweetness potency of greater intensity than sugar (3).

Concerns over sugar consumption in the population have led to an expansion of the food produce containing AS. Historically, diet carbonated beverages have represented the product category with most common AS use, a fact which corresponds to this category being the most significant contribution of AS consumption at a population level (4).

However, specific product categories – in particular waters and yogurts – have a significant proportion of their market share in artificially sweetened versions (4). And AS use is now extending into low calorie alternatives to ice-cream, jams, syrups, grain products, and bars (4). The global market for AS products is projected to reach 2.2 billion by 2020, reflecting a strong consumer shift toward lower calorie or non-caloric alternatives to common food products (4).

However, despite significant research into these compounds, and ongoing safety monitoring programs by the European Food Standards Agency [EFSA; or the Food and Drug Administration in the US], there remain concerns voiced over the safety and efficacy of AS.

Of particular concern to diabetics is the suggestion that AS in fact stimulate blood glucose and insulin responses. Let’s look at that in more detail.

Effects of Artificial Sweeteners on Blood Glucose and Insulin

The concerns in relation to the potential impacts of AS on diabetes arise from some observational studies which found associations between high intake of AS and Type-2 diabetes (5;6).

However, an explanation for this observation is “reverse causality”. What this means is that people who are at high risk of diagnoses, are overweight and/or obese, and with a poor quality, high-calorie diet, are more likely to consume a lot of AS products (1). Basically, “I’ll have a Diet Coke” says the pre-diabetic who just ordered the KFC family bucket.

Thus, we need to look at controlled interventions assessing the potential impact of AS on relevant parameters of blood glucose regulation and insulin.

How might AS lead to increases in blood glucose and insulin? There are two suggestions offered.

The first is that the intense perception of sweetness alters insulin responses, in anticipation of a glycaemic response.

The second is that sweet-taste receptors in the intestinal lining are activated by AS, which increases glucose uptake from the digestive tract.

The AS sucralose has recently caused some eyebrow-raising in research. Prior to the EFSA, the Scientific Committee on Food [SCF] of the European Union undertook a review of sucralose. One trial included was a 6-month study in Type-2 diabetics which found that HbA1c was elevated above baseline in subjects administered 667mg sucralose daily (7).

However, there were no observed impacts on insulin or blood glucose levels, and thus the increase in HbA1c is difficult to reconcile as it occurred in the absence of impacts on other parameters of blood glucose regulation (7). In addition, the dose used in the trial significantly

exceeded even the highest levels of population consumption, and the SCF concluded that any potential effect would be minimal to the point of being clinically insignificant (7).

A recent trial in subjects with morbid obesity yet normal insulin sensitivity – which was assessed by HOMA-IR – found that pre-loading subjects with 48mg sucralose [a realistic dose] 10-minutes before an oral glucose tolerance test with 75g glucose led to a greater insulin response and peak plasma glucose levels (8).

However, both glucose and insulin responses were both within normal range for an OGTT, and the actual difference between the control group and sucralose group was nominal, again suggesting no material clinical significance to the finding (9).

The couple of trials suggesting some glycaemic interaction notwithstanding, multiple studies have assessed sucralose and glycaemic responses and this AS is generally considered to have no impact on carbohydrate metabolism (10). In particular, a trial that assessed the effects of 1,000mg sucralose – an enormous dose – on glucose homeostasis in both Type-1 and Type-2 diabetics found no effect on blood glucose levels in either group (11).

It’s important be aware of the fact that any of the positive findings in studies on different AS suggesting an impact on glucose metabolism may be a reflection of the trial design, not necessarily a clinically relevant finding. For example, the use of a placebo – like water – to compare to AS may lead to misleading findings when AS are designed to substitute for sugar, so a comparison with a caloric sweetener would arguably be a more appropriate design (1).

The reason for this is that, if we assume for one second that AS do in fact stimulate glucose uptake, this would occur in the presence of low concentrations of glucose in the gut (12). Conversely, a caloric sweetener would obviously result in greater glucose absorption from the digestive tract (12).

Given the specific purpose of AS is to substitute for the effects of sugar, not a benign placebo like water, this would arguably present a truer comparison of glycaemic effects. It could provide an explanation for the fact that in trials finding an impact of AS on glycaemic and insulin responses, the differences between intervention and placebo – often water – are in fact nominal, to the point of clinical insignificance (7;13).

In relation to sucralose specifically, it should be noted that the two trials cited above excepted, the majority of studies have found no effect on blood glucose levels (9).

A recent systematic review including 28 trials looking at the effects of different AS – including aspartame, saccharin, sucralose, and acesulfame-K – on glucose metabolism confirmed this overall position: the majority of trials have found no effect, while certain trials have concluded AS do impact on glucose metabolism (9).

The wide variability in characteristics of subjects, the particular AS used, the placebo issue, and the outcomes assessed, limits comparisons between studies (9). So, we need to take a closer look at the second issue identified above, that the underlying mechanism is the activation by AS of glucose transporters in the gut, in turn stimulating glucose uptake and insulin responses.

In particular, it has been suggested that AS activate glucagon-like peptide-1 [GLP-1], which itself stimulates insulin release and mediates glucose uptake from transporters in the gut (12;14). In a systematic review which included 11 studies that measured GLP-1, 1 study found a decrease in GLP-1 from aspartame while 2 found increases (9). What is interesting, however, is that neither measurable parameters of satiety nor subjective measures of appetite have been found to be influenced: this is interesting because increases in GLP-1 typically reduce appetite and delay gastric emptying (12;14).

This calls into question the clinical significance of the findings in a handful of studies of increased GLP-1 concentrations.

Trial design may again be an issue in the studies finding positive effects of AS on GLP-1. For example, a controlled trial in healthy subjects using a diet soda containing a combination of 46mg sucralose and 26mg acesulfame-K found an increase in GLP-1 in response to an oral glucose tolerance test after diet beverage administration, compared to water (15).

However, in the first instance the use of an oral glucose tolerance test is an issue, as glucose itself would activate GLP-1 and thus the increased GLP-1 levels in the diet beverage group may reflect the influence of other factors. In particular, the trial failed to control for other compounds in the diet beverage – citric acid, potassium citrate, phosphoric acid, potassium benzoate, and colourings – that could have influenced the results (15).

A subsequent trial addressed this limitation by administering only the AS themselves, and neither 52mg sucralose, 200mg acesulfame-K, or a combination of 46mg sucralose and 26mg acesulfame-K [the dose used in the aforementioned trial] were found to have any effect on insulin concentrations, blood glucose, or GLP-1, in healthy subjects (16).

In assessing the cumulative body of evidence assessing the impact of AS on GLP-1, 2 trials have found increased GLP-1 but it is important to note that these findings have occurred in the absence of any other impacts on insulin or blood glucose (17).

Also bear in mind that in one of the studies, the impact on GLP-1 from sucralose ingestion was only observed in healthy controls, not in subjects with T2DM (17).

Other interventions (16;18), and a systematic review including 11 studies that measured GLP-1 (12), taken together indicate that the majority of human trials have found no impact of AS on intestinal sweet-taste receptors and glucose uptake. As yet, no biological basis has been established to explain the findings of the limited number of studies showing such effect in humans (12).

Another factor that calls into question the studies finding positive associations is the relationship between GLP-1 and insulin, as increased GLP-1 stimulates insulin secretion and activation of sweet-taste receptors signals for insulin release (12;14). One of the studies which reported an increase in GLP-1 levels also found no effect on insulin (17).

Multiple studies in Type-2 diabetics have failed to find any impact on insulin (12;19;20;21). More particularly, trials that compare AS to a caloric sweetener, like sucrose [table sugar] have found an effect only from the sugar (12;17;22).

In a systematic review of studies assessing the impact of AS on glucose metabolism, of the included 14 studies which measured insulin, none found a negative impact of AS (9).

A significant criticism of this area of research has been that some authors have overreached on data from in vitro and animal model studies to make speculative assumptions on the effects of AS on sweet-taste receptors, insulin, and glucose homeostasis (12).

The majority of the research finding particular effects amounts to identification of potential mechanisms in vitro or in animal models, and there is limited human data supporting these findings. Indeed, the evidence for any direct effect of AS on glycaemic control is limited (1).

The weight of evidence suggests AS do not adversely impact on blood glucose regulation.

Artificial Sweeteners: Myths and Truths

And this is where we come to the real hysteria about AS, the kind you see on some naturopath’s blog, in particular that AS are carcinogenic, and neurotoxic.

Let’s deal with these briefly in turn.

Before we even get into those, I think it’s helpful to explain that these compounds are not simply unleashed on the population as some Orwellian experiment. There is ongoing toxicology monitoring for all of the AS in use, and a regulatory framework in which they are approved for use.

Because they have potential for high consumption in the population, AS are designated a high concern level: this means that there are specific directions for the design of toxicology studies to be carried out (3;8). Animal toxicology studies are designed to elicit a response and assess toxicity thresholds, but also for the potential for lower doses to impact on reproduction, development, cancer, genetic mutations, and the immune system (3).

Based on these toxicology studies, a threshold known as the ‘No Observable Adverse Effect Level’ [the NOAEL] is set, and this is the lowest dose at which any toxic effects were found in the animal studies (23). The Acceptable Daily Intake [ADI] is then set by dividing the NOAEL by an “uncertainty factor” of 100. This ultimately results in a highly conservative assessment of safety thresholds, factoring in anticipated concentrations of a given AS in food and beverages, compared to population levels of intake of those foods and drinks (23).

For some context, the NOAEL for the AS aspartame is set at 4g per kilogram of bodyweight per day, based on long-term toxicology studies (8;23). By dividing that by the uncertainty factor, the ADI was set at 40mg/kg, and actual average intake of aspartame is 4mg/kg (24;25).

To reach the ADI, a 60kg adult would have to consume 4L of beverages sweetened with the maximum amounts of aspartame permitted by regulation (23).

And in reality this cannot even happen, because recall that the permitted thresholds are set based on data on population consumption, thus no one food or drink contains the maximum level of any AS. Even the subsets of the population consuming the highest levels of aspartame consume 10mg/kg, i.e., the very highest levels of consumption observed in the population are still consuming only 30% of the ADI (24;25).

The European Union has ongoing toxicology monitoring programs, and in 2009 mandated that every AS in use be subject to full revaluation of toxicology and technical data (26). There is often a cynical view of AS expressed by some in relation to the efficacy of such programs, and the safety of AS.

I would highlight that to date, the EU’s toxicology monitoring has resulted in over 1,300 compounds being banned from use in personal care and cosmetic products, based on even preliminary evidence of toxicity (26).

An unbiased assessment of the regulatory framework in which AS are assessed and approved for use would see a well-regulated and ongoing consideration of safety, which acknowledges that there is always a degree of uncertainty where animal studies inform long-term human risk.

It’s important to understand the toxicology assessment framework in order to put to bed most of the myths generated about effects on the brain or cancer. All studies showing neurotoxic potential are in such animal studies.

As an example of where Internet-hysteria is generated, a recent study examined the neurotoxic potential of aspartame in rats using doses of 500mg/kg to 1,000mg/kg (27).

This experimental dose is 2,400% times greater than the ADI for aspartame, and thus of zero relevance to both typical population levels of consumption and established safety thresholds.

The same can be said for carcinogenicity. No significant relationship has been observed between aspartame at various experimental doses and tumour occurrence in meta-analysis of animal toxicology studies (28), which is consistent with observational evidence in humans (23).

Acesulfame-K was reviewed by the EU in 2009 and concluded to be safe, even for use in pregnancy and infancy, and in particular that concerns over carcinogenicity were unsubstantiated (29;30).

Sucralose is considered to have a strong safety profile, and a review of toxicology studies found no evidence of carcinogenicity either in long-term animal models or in studies in humans (31). Saccharin is also non-carcinogenic in human studies (32;33).

Ultimately, when it comes to the safety of AS across important health outcomes – from neurology to cancer – the safety of the various AS compounds in use is supported by numerous international agencies and scientists in the field (2).

Possible Considerations for People with Diabetics

The first and most relevant question is: are AS going to negatively impact on your blood glucose regulation?

And can AS beverages and foods be used in the context of blood glucose management safely in lieu of calorically sweetened products?

The answer to the former, based on the majority of human research, is no: use of AS beverages or foods will not negatively impact on insulin or blood glucose regulation.

The answer to the latter, is yes: substituting calorically sweetened beverages/foods for non-calorically AS sweetened can be a useful means of maintaining adherence and palatability in diabetics, a position supported by the National Academy of Nutrition and Dietetics (34).

The considerations, therefore, are more contextual. AS products should not displace overall diet quality:

they are simply “less unhealthy” versions of their calorically-sweetened cousins and cannot be said to be contributing to improved health as much as providing a sweet alternative.

However, that is the context in which you should feel comfortable in using them. Put this way: sugar-sweetened drinks are strongly associated with diabetes risk and profoundly impact on blood sugar levels. We are all human, and we want to be able to have some highly palatable, sweet foods. For you as a diabetic, the most important variable is staying on top of blood sugar management. So, if some AS products allow you to consume some sweet stuff, without negatively impacting on your blood glucose, then that is a positive – and the purpose for which they should be incorporated into your life.

Also, be mindful to avoid the pitfall of increasing calorie intake from other sources on the assumption that consuming AS products creates some leeway in this respect: if you’re looking to improve body composition, energy balance still matters!

And finally, consider the caffeine content of artificially sweetened drinks, particularly energy drinks, as caffeine can increase blood glucose levels and reduce insulin sensitivity (35;36).

Let’s put this all into a hierarchy of needs and quality, from drinks that can materially add to or improve your health, to one’s that won’t:

Water: obviously vital, and still the No.1 hydration agent.
Tea’s and coffee: careful not to exceed 400mg caffeine (or 200mg if pregnant). Tea’s and coffee were once believed to be dehydrating: that’s broscience. They can contribute to hydration.
AS beverages: be wary of caffeine content and keep to an absolute maximum of 1-2L per day (because higher levels inevitably mean displacing water and other health-promoting non-caloric products like green tea or coffee.
Sugar-sweetened beverages: ideally ZERO. They contribute nothing, and you can get your sweet hit from the AS version of your favourite product, i.e., the purpose for AS are used! So, use them.

So, there’s some truth on AS. Use them instead of sugar, not diet quality. And don’t believe the scaremongering: it doesn’t stand up to scrutiny.

References

Gardner, C. (2014). Non-nutritive sweeteners. Current Opinion in Lipidology, 25(1), pp.80-84.
Lohner, S., Toews, I. and Meerpohl, J. (2017). Health outcomes of non-nutritive sweeteners: analysis of the research landscape. Nutrition Journal, 16(1).
Roberts, A. (2016). The safety and regulatory process for low calorie sweeteners in the United States. Physiology & Behavior, 164, pp.439-444.
Sylvetsky, A. and Rother, K. (2016). Trends in the consumption of low-calorie sweeteners. Physiology & Behavior, 164, pp.446-450.
Nettleton, J., Lutsey, P., Wang, Y., Lima, J., Michos, E. and Jacobs, D. (2009). Diet Soda Intake and Risk of Incident Metabolic Syndrome and Type 2 Diabetes in the Multi-Ethnic Study of Atherosclerosis (MESA). Diabetes Care, 32(4), pp.688-694.
Hu, F. and Malik, V. (2010). Sugar-sweetened beverages and risk of obesity and type 2 diabetes: Epidemiologic evidence. Physiology & Behavior, 100(1), pp.47-54.
Scientific Committee on Food of the European Commission. Opinion of the Scientific Committee on Food on sucralose. SCF/CS/ADDS/EDUL/190 Final 12/9/2000.
Pepino, M., Tiemann, C., Patterson, B., Wice, B. and Klein, S. (2013). Sucralose Affects Glycemic and Hormonal Responses to an Oral Glucose Load. Diabetes Care, 36(9), pp.2530-2535.
Romo-Romo, A., Aguilar-Salinas, C., Brito-Córdova, G., Gómez Díaz, R., Valentín, D. and Almeda-Valdes, P. (2016). Effects of the Non-Nutritive Sweeteners on Glucose Metabolism and Appetite Regulating Hormones: Systematic Review of Observational Prospective Studies and Clinical Trials. PLoS One, 11(8), p.e0161264.
Brown, R. and Rother, K. (2012). Non-Nutritive Sweeteners and their Role in the Gastrointestinal Tract. The Journal of Clinical Endocrinology & Metabolism, 97(8), pp.2597-2605.
Mezitis, N., Maggio, C., Koch, P., Quddoos, A., Allison, D. and Pi-Sunyer, F. (1996). Glycemic Effect of a Single High Oral Dose of the Novel Sweetener Sucralose in Patients With Diabetes. Diabetes Care, 19(9), pp.1004-1005.
Renwick, A. and Molinary, S. (2010). Sweet-taste receptors, low-energy sweeteners, glucose absorption and insulin release. British Journal of Nutrition, 104(10), pp.1415-1420.
Grotz, V. and Jokinen, J. (2014). Comment on Pepino et al. Sucralose Affects Glycemic and Hormonal Responses to an Oral Glucose Load. Diabetes Care 2013;36:2530–2535. Diabetes Care, 37(6), pp.e148-e148.
Grieve, D., Cassidy, R. and Green, B. (2009). Emerging cardiovascular actions of the incretin hormone glucagon-like peptide-1: potential therapeutic benefits beyond glycaemic control?. British Journal of Pharmacology, 157(8), pp.1340-1351.
Brown, R., Walter, M. and Rother, K. (2009). Ingestion of Diet Soda Before a Glucose Load Augments Glucagon-Like Peptide-1 Secretion. Diabetes Care, 32(12), pp.2184-2186.
Steinert, R., Frey, F., Töpfer, A., Drewe, J. and Beglinger, C. (2011). Effects of carbohydrate sugars and artificial sweeteners on appetite and the secretion of gastrointestinal satiety peptides. British Journal of Nutrition, 105(09), pp.1320-1328.
Temizkan, S., Deyneli, O., Yasar, M., Arpa, M., Gunes, M., Yazici, D., Sirikci, O., Haklar, G., Imeryuz, N. and Yavuz, D. (2014). Sucralose enhances GLP-1 release and lowers blood glucose in the presence of carbohydrate in healthy subjects but not in patients with type 2 diabetes. European Journal of Clinical Nutrition, 69(2), pp.162-166.
Ma, J., Chang, J., Checklin, H., Young, R., Jones, K., Horowitz, M. and Rayner, C. (2010). Effect of the artificial sweetener, sucralose, on small intestinal glucose absorption in healthy human subjects. British Journal of Nutrition, 104(06), pp.803-806.
Grotz, V., Henry, R., McGill, J., Prince, M., Shamoon, H., Trout, J. and Pi-Sunyer, F. (2003). Lack of effect of sucralose on glucose homeostasis in subjects with type 2 diabetes. Journal of the American Dietetic Association, 103(12), pp.1607-1612.
Nauck, M., Niedereichholz, U., Ettler, R., Holst, J., Ørskov, C., Ritzel, R. and Schmiegel, W. (1997). Glucagon-like peptide 1 inhibition of gastric emptying outweighs its insulinotropic effects in healthy humans. American Journal of Physiology-Endocrinology and Metabolism, 273(5), pp.E981-E988.
Horwitz, D., McLane, M. and Kobe, P. (1988). Response to Single Dose of Aspartame or Saccharin by NIDDM Patients. Diabetes Care, 11(3), pp.230-234.
Abdallah, L., Chabert, M. and Louis-Sylvestre, J. (1997). Cephalic phase responses to sweet taste. The American Journal of Clinical Nutrition, 65(3), pp.737-743.
Marinovich, M., Galli, C., Bosetti, C., Gallus, S. and La Vecchia, C. (2013). Aspartame, low-calorie sweeteners and disease: Regulatory safety and epidemiological issues. Food and Chemical Toxicology, 60, pp.109-115.
Mortensen, A. (2006). Sweeteners permitted in the European Union: safety aspects. Food & Nutrition Research, 50(3).
Magnuson, B., Burdock, G., Doull, J., Kroes, R., Marsh, G., Pariza, M., Spencer, P., Waddell, W., Walker, R. and Williams, G. (2007). Aspartame: A Safety Evaluation Based on Current Use Levels, Regulations, and Toxicological and Epidemiological Studies. Critical Reviews in Toxicology, 37(8), pp.629-727.
REGULATION (EC) No 1333/2008 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 16 December 2008 on food additives.
Abhilash, M., Sauganth Paul, M., Varghese, M. and Nair, R. (2012). Long-term consumption of aspartame and brain antioxidant defense status. Drug and Chemical Toxicology, 36(2), pp.135-140.
Mallikarjun, S. and Sieburth, R. (2013). Aspartame and Risk of Cancer: A Meta-analytic Review. Archives of Environmental & Occupational Health, 70(3), pp.133-141.
Regulation (EC) No 1223/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products.
Statement of EFSA on the scientific evaluation of two studies related to the safety of artificial sweeteners. (2011). EFSA Journal, 9(2), p.2089.
Berry, C., Brusick, D., Cohen, S., Hardisty, J., Grotz, V. and Williams, G. (2016). Sucralose Non-Carcinogenicity: A Review of the Scientific and Regulatory Rationale. Nutrition and Cancer, 68(8), pp.1247-1261.
Shankar, P., Ahuja, S. and Sriram, K. (2013). Non-nutritive sweeteners: Review and update. Nutrition, 29(11-12), pp.1293-1299.
Gallus, S., Scotti, L., Negri, E., Talamini, R., Franceschi, S., Montella, M., Giacosa, A., Dal Maso, L. and La Vecchia, C. (2006). Artificial sweeteners and cancer risk in a network of case-control studies. Annals of Oncology, 18(1), pp.40-44.
Fitch, C. and Keim, K. (2012). Position of the Academy of Nutrition and Dietetics: Use of Nutritive and Nonnutritive Sweeteners. Journal of the Academy of Nutrition and Dietetics, 112(5), pp.739-758.
Greer, F., Hudson, R., Ross, R. and Graham, T. (2001). Caffeine Ingestion Decreases Glucose Disposal During a Hyperinsulinemic-Euglycemic Clamp in Sedentary Humans. Diabetes, 50(10), pp.2349-2354.
Keijzers, G., De Galan, B., Tack, C. and Smits, P. (2002). Caffeine Can Decrease Insulin Sensitivity in Humans. Diabetes Care, 25(2), pp.364-369.