Hypoglycaemia: Independent Evidence on Causes, Glucagon & Driving

Table of contents

• Evidence summary
• What hypoglycaemia is
• Symptoms, diagnosis and red flags
• Causes and risk factors
• Treatments that work
• Medicines: glucose, glucagon and beyond
• Supplements and lifestyle claims, graded
• Risks of common treatments
• Driving rules: DVLA and US
• Who to see
• Follow the money
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

Independence tiers: 1 = regulator or government body, or a trial funded only by government or a charity with no product to sell; 2 = professional society or academic team with indirect industry ties; 3 = independent authors working with industry money; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D.

Our sources cluster in the UK and US. That matters most for driving, where rules are national, and for glucagon, where products and prices differ by country. The strongest signal is that UK government sources (NICE, NHS) and the industry-linked ADA give the same core advice. Their interests differ, so that agreement means more than either source alone.

What hypoglycaemia is

Hypoglycaemia (US spelling: hypoglycemia), often called a "hypo", means the level of glucose in the blood has fallen too low for the body and brain to work normally. The brain depends almost entirely on glucose for fuel, which is why confusion, odd behaviour and seizures can follow.

In people treated for diabetes, the ADA and the European Association for the Study of Diabetes use three levels. The thresholds came from the International Hypoglycaemia Study Group (IHSG 2017):

• Level 1: glucose below 3.9 mmol/L (70 mg/dL) and at or above 3.0 mmol/L (54 mg/dL). This is the "alert value" for treating and adjusting.
• Level 2: glucose below 3.0 mmol/L (54 mg/dL). This is clinically significant hypoglycaemia.
• Level 3: a severe event with altered mental or physical state in which the person needs someone else's help, whatever the glucose reading (ADA 2026).

The UK often uses a simpler rule: below 4 mmol/L is low and should be treated (NHS). The difference between 3.9 and 4.0 is rounding, not disagreement.

In people who do not have diabetes, the threshold is different. The Endocrine Society says hypoglycaemia should only be investigated when all three parts of Whipple's triad are documented: symptoms or signs, a low plasma glucose measured reliably at the time, and relief when glucose is raised (Cryer et al., Endocrine Society 2009).

How common is it?

In a UK study across six hospital diabetes clinics (383 people), people with type 1 diabetes for more than 15 years had about 3.2 severe hypos per person per year. People with type 2 diabetes on sulfonylurea tablets, or recently started on insulin, had 0.1 to 0.2 (UK Hypoglycaemia Study Group 2007; funding not verified). A meta-analysis of 46 population studies (532,542 people with type 2 diabetes) found that 21% of insulin users had had a severe hypo, compared with 5% of people on sulfonylurea regimens (Edridge 2015; no funding; some authors advise drug companies).

Symptoms, diagnosis and red flags

Early symptoms include hunger, dizziness, anxiety or irritability, sweating, shaking, tingling lips, a fast or pounding heartbeat, tiredness, blurred vision and confusion. Untreated, a hypo can cause seizures or loss of consciousness (NHS).

Hypos during sleep are easy to miss. Signs include crying out or nightmares, sweating enough to dampen the sheets, and waking tired, irritable or confused (NIDDK).

Impaired awareness of hypoglycaemia

After repeated lows, some people lose the early warning symptoms and only notice when glucose is already very low (NIDDK). Impaired awareness raises the risk of severe hypoglycaemia three to six times in type 1 diabetes (van Beers 2016). NICE asks clinicians to check awareness at every annual review using the Gold or Clarke questionnaire (NICE NG17). The ADA advises screening at least once a year (ADA 2026).

Diagnosis outside diabetes

For someone without diabetes, the first step is to look for obvious causes: medicines, serious illness, hormone deficiencies and some rare tumours. If none is found, doctors take blood during an episode. Tests include glucose, insulin, C-peptide, proinsulin, beta-hydroxybutyrate and a screen for diabetes tablets, plus insulin antibodies. This separates the body over-producing insulin from accidental or deliberate use of diabetes medicines (Endocrine Society). A home glucose meter reading alone is not enough to make the diagnosis.

Causes and risk factors

In people with diabetes

Insulin and drugs that make the pancreas release insulin (sulfonylureas such as gliclazide, glimepiride and glibenclamide, and meglitinides) cause almost all hypoglycaemia in diabetes (Endocrine Society). The ADA ranks the risk as highest with intensive insulin (multiple daily injections, pumps and automated systems), then basal insulin, then sulfonylureas. Using insulin and a sulfonylurea together raises it further (ADA 2026).

The ADA's main risk factors are:

• a level 2 or 3 hypo in the past 3 to 6 months
• impaired awareness
• kidney failure
• cognitive impairment or dementia
• previous weight-loss (metabolic) surgery
• age 75 or over
• social factors such as food insecurity, low income, unstable housing and fasting for religious reasons (ADA 2026)

NICE also lists injection-site problems, poor injection technique, exercise, alcohol, gastroparesis and changes in insulin sensitivity, including from kidney failure (NICE NG17).

Some antibiotics, including fluoroquinolones and clarithromycin, can greatly increase the effect of sulfonylureas. The ADA advises considering a temporary dose cut or pause while they are taken together (ADA 2026). Sulfonylureas are not all alike: a network meta-analysis of 27 trials (16,260 people) found gliclazide carried the lowest hypo risk when added to metformin (Andersen 2016; no external funding).

Alcohol

Alcohol interferes with the liver's release of glucose. In a small controlled study, six men with type 1 diabetes drank dry white wine in the evening. They had normal glucose overnight but lower glucose the next morning, and five needed treatment for hypoglycaemia after breakfast. None did after drinking water (Turner 2001). The NIH advises eating when drinking alcohol (NIDDK).

In people without diabetes

The NHS lists malnutrition, Addison's disease (adrenal failure) and pancreatic tumours among rare causes (NHS). The Endocrine Society also names drugs, critical illness, hormone deficiencies and non-islet-cell tumours. It notes that accidental, hidden or even malicious use of insulin or tablets must be ruled out (Endocrine Society). Three causes come up most often in searches.

• Insulinoma: an insulin-making tumour of the pancreas, usually benign. In a 60-year Mayo Clinic study, its incidence in one Minnesota county levelled off at about 4 cases per million person-years. About 6% were malignant and about 8% occurred with the inherited condition MEN1 (Service 1991; NIH-funded). Typical attacks happen when fasting.
• Post-bariatric hypoglycaemia: lows 1 to 3 hours after eating, usually starting more than a year after gastric bypass. Fewer than 1% need hospital care for it and fewer than 10% have a clinically recognised problem, although up to 38% report possible symptoms (Salehi 2018).
• "Reactive" (postprandial) hypoglycaemia: a popular label for shakiness, anxiety or tiredness after eating. In a classic study, 23% of referred patients and 25% of healthy controls had low readings after a sugary drink test. After ordinary mixed meals, 78% of patients still had symptoms but none had low glucose. The authors called this the "idiopathic postprandial syndrome" (Charles 1981). This is why a sugary drink test alone is a poor way to diagnose it.

Treatments that work

Treating a hypo: the "rule of 15"

For a conscious person with glucose below 3.9 mmol/L (70 mg/dL), the ADA advises:

• Glucose first: pure glucose is the preferred treatment, though any carbohydrate that contains glucose will work. Most people need 15 g.
• Recheck: test again after 15 minutes and repeat the treatment if glucose is still low.
• Avoid fat and protein for the first treatment, because they slow the response.
• Exceptions: people on automated insulin delivery usually need only 5 to 10 g. People taking acarbose must use pure glucose, because acarbose slows the breakdown of table sugar (ADA 2026).

The NIH says 15 to 20 g (NIDDK). The NHS examples are a small glass of fruit juice or sugary (not diet) drink, 5 glucose or dextrose tablets, 4 large jelly babies or 2 tubes of glucose gel. Check again after 10 to 15 minutes and repeat if still under 4 mmol/L. Once glucose is back up, eat a snack or meal with slower carbohydrate (NHS).

Severe hypos: glucagon

NICE advises that adults with type 1 diabetes who cannot safely swallow are given glucagon by a trained family member or friend. If they have not improved after 10 minutes, they need intravenous glucose from a professional. Once they can swallow safely they should eat, and someone should stay with them because the hypo can come back (NICE NG17). The ADA says glucagon should be prescribed for everyone taking insulin or at high risk of hypoglycaemia, and prefers products that do not need mixing (ADA 2026).

Preventing repeat episodes

After one or more level 2 or 3 hypos, the ADA advises reviewing the whole treatment plan. That can include reducing insulin or sulfonylureas, or switching to a medicine class with lower hypo risk (ADA 2026). For impaired awareness, NICE recommends a stepped approach:

1. Structured education in flexible insulin dosing.
2. Extra hypo-focused education if awareness is still impaired.
3. An insulin pump or real-time CGM.
4. Referral to a specialist centre if severe hypos continue (NICE NG17).

Notably, NICE advises against loosening individual glucose targets to deal with impaired awareness.

Medicines: glucose, glucagon and beyond

Glucose tablets and gels

These are cheap, widely sold and work if swallowed. The ADA recommends that first-aid kits contain oral glucose (ADA 2026). A Cochrane review for first-aid bodies found that glucose held in the cheek gave lower blood glucose at 20 minutes than glucose swallowed. A gel taken partly by mouth and partly in the cheek showed no clear difference from tablets or solution. All of this evidence was very low certainty (De Buck 2019).

Glucagon products

Glucagon is a hormone that makes the liver release stored glucose. The traditional kit is a powder that has to be mixed with liquid before injection, which is error-prone in an emergency. The newer products avoid this:

• Nasal glucagon (Baqsimi, 3 mg): approved by the FDA in July 2019 for severe hypoglycaemia in people aged 4 and over (FDA label). The current label, revised in March 2025, extends this to children aged 1 year and over (DailyMed, current Baqsimi label). In the pivotal US trial of 75 adults, it raised glucose successfully at 74 of 75 visits, against 75 of 75 for injected glucagon. It took 16 minutes on average, compared with 13 minutes for the injection (Rickels 2016).
• Ready-to-use liquid glucagon (Gvoke): approved by the FDA in September 2019 as a prefilled syringe or auto-injector (PharmTech).
• Dasiglucagon (Zegalogue): a modified, stable form of glucagon approved by the FDA on 22 March 2021 for people aged 6 and over (the auto-injector and prefilled syringe were reported discontinued in the US in 2026; see Follow the money) (Zealand Pharma). In its phase 3 trial, median recovery took 10 minutes, against 40 for placebo and 12 for standard glucagon (Pieber 2021).

All of these trials induced a hypo with insulin under hospital supervision, then compared products. None tested real-world emergencies, where speed depends mostly on whether a helper can find and use the device. That is why the ADA stresses training family, carers and school staff (ADA 2026).

Medicines for hypoglycaemia outside diabetes

For post-bariatric hypoglycaemia, a review lists acarbose, diazoxide and octreotide, with surgery reserved for severe cases. A tube into the bypassed stomach has helped some people for years. Partial removal of the pancreas is no longer recommended because of high complication rates and frequent relapse (Salehi 2018). The first placebo-controlled trial of acarbose, in 11 people, did not reduce time spent below 3.9 mmol/L. Some symptoms did improve (HypoBar trial 2026). Insulinoma is treated by surgical removal, and long-term survival after removing a benign tumour is normal (Service 1991).

Supplements and lifestyle claims, graded

We found no independent trial showing that any dietary supplement prevents or treats hypoglycaemia. Supplements sold to "balance blood sugar" are aimed at lowering it. For someone on insulin or a sulfonylurea, that is the wrong direction.

Risks of common treatments

• Glucagon side effects: in adult trials of nasal glucagon, 26% had nausea, 18% headache, 15% vomiting and 12% nose or throat irritation (FDA label). Injected glucagon causes similar nausea (35% vs 38% of visits in the head-to-head trial) (Rickels 2016). Someone who is drowsy should be kept in the recovery position (NHS).
• Glucagon interactions: people on beta-blockers may have a short rise in pulse and blood pressure. Indomethacin may stop glucagon working or even cause a low. Glucagon may increase the effect of warfarin (FDA label).
• Tight glucose targets: pushing glucose down too far increases hypos. In the ADVANCE trial, severe hypos occurred in 2.7% of the intensive-control group and 1.5% of the standard group over about five years. Severe hypos were followed by more heart events and deaths, though the authors say hypos may mark frailty as much as cause harm (Zoungas 2010; funded by Servier and Australia's NHMRC).
• Islet transplantation: in the NIH trial, 5 of 56 procedures through the skin into the portal vein caused bleeding that needed a transfusion or surgery. Kidney function declined on the immunosuppressants, and immunosuppression caused 43% of study-related serious adverse events (Hering 2016). The FDA-approved islet product Lantidra carries similar procedure and immunosuppression risks (FDA 2023).
• Drugs for post-bypass lows: diazoxide often causes fluid retention and swelling (Salehi 2018). In the HypoBar trial, acarbose caused moderate, common gut side effects, and one person stopped because of them (HypoBar trial).

Driving rules: DVLA and US

Hypoglycaemia is the main reason diabetes affects driving.

Great Britain (DVLA), cars and motorcycles: people treated with insulin can hold a licence if they have not had two or more severe hypos while awake in the past 12 months, and the most recent was more than 3 months ago. They must check glucose before the first journey and at least every 2 hours while driving. Anyone with impaired awareness must not drive until a GP or specialist confirms awareness has returned. Approved CGM systems may be used, but drivers must carry a finger-prick meter as backup (DVLA).

Great Britain, lorries and buses: the rules are stricter. There must be no severe hypo in the past 12 months, full awareness, and glucose checks at least twice daily even on days without driving. Since a law change reflected in November 2025, these drivers may also use approved CGM systems while driving (DVLA).

After a hypo: the NHS advises not driving for at least 45 minutes after glucose is back to normal (NHS).

United States: there is no single DMV rule. Licensing laws differ from state to state. The ADA's position is that a diabetes diagnosis alone should not decide fitness to drive. Instead, it calls for an individual assessment focused on whether the person can detect and treat hypoglycaemia (ADA, Diabetes and Driving). For interstate commercial drivers, a 2018 federal rule lets people on insulin qualify with an assessment form from their treating clinician. After any severe hypo they must stop driving until a clinician has evaluated them (FMCSA, Federal Register 2018).

Who to see

• Emergency services (999, 911 or local number): for anyone unconscious, fitting, unable to swallow, not recovering, or where alcohol is involved.
• Your diabetes team (GP, practice nurse, diabetes specialist nurse or diabetologist): after any level 2 or 3 hypo, frequent level 1 lows, night-time symptoms or loss of warning signs. The treatment plan may need changing (ADA). Ask about structured education and whether you qualify for CGM.
• A specialist hypoglycaemia or transplant centre: for recurrent severe hypos despite education and technology (NICE NG17).
• An endocrinologist: for anyone without diabetes who has a documented low glucose with symptoms. This is especially important for lows when fasting, which can point to insulinoma (Endocrine Society).
• Your bariatric team and dietitian: for symptoms 1 to 3 hours after meals following weight-loss surgery (Salehi 2018).
• A pharmacist: before adding antibiotics, supplements or new medicines if you take insulin or a sulfonylurea.

Follow the money: who funds the research, guidelines and treatments

Who sells hypoglycaemia treatments

• Baqsimi (nasal glucagon): the product was developed by Locemia Solutions of Montreal, whose staff co-authored the pivotal trial (Rickels 2016). It was launched by Eli Lilly. Lilly sold it to Amphastar Pharmaceuticals of Rancho Cucamonga, California, on 30 June 2023. Amphastar paid $500 million at closing and agreed to pay a further $125 million on the first anniversary, with up to $450 million more depending on milestones (Amphastar).
• Zegalogue (dasiglucagon): developed by Zealand Pharma of Copenhagen. In September 2022, Zealand licensed global rights to Novo Nordisk, which also makes insulin. The deal included DKK 25 million up front, up to DKK 265 million in milestones and royalties on sales (Zealand filing). In February 2026 a US Medicaid pharmacy notice said Novo Nordisk had announced the discontinuation of the Zegalogue auto-injector and prefilled syringe, and pointed prescribers to Baqsimi or Gvoke instead (MedImpact notice for Kentucky Medicaid). We could not find Novo Nordisk's own announcement.
• Gvoke (liquid glucagon): made by Xeris Pharmaceuticals (PharmTech).
• Lantidra (islet cells): made by CellTrans Inc. (FDA).

The companies that sell rescue treatments for hypoglycaemia include insulin makers, whose products cause most of it. That is not wrongdoing, but readers should know it.

Who funded the trials

Both pivotal glucagon trials we reviewed had maker involvement. The nasal glucagon trial was co-funded by Locemia, alongside the Helmsley Charitable Trust and the NIH (Rickels 2016). The dasiglucagon trial was sponsored by Zealand Pharma, which designed it and analysed the data (Pieber 2021).

The prevention evidence is cleaner. HypoCOMPaSS, which found that education did as well as pumps and sensors, was funded by Diabetes UK (Newcastle University). Its follow-up also lists the Wellcome Trust and Medical Research Council (Cambridge repository). The islet trial was paid for by NIH institutes (Hering 2016). The Dutch CGM trial was funded by Eli Lilly and Sanofi (van Beers 2016). The ADVANCE analysis linking severe hypos to death was co-funded by the drug company Servier and Australia's NHMRC (Zoungas 2010).

Who funds the guideline-makers

• NICE is a public body sponsored by the UK Department of Health and Social Care (GOV.UK).
• The ADA says its Standards of Care are funded from general revenue, with no industry money for the guideline, and that committee members are unpaid (ADA 2026 introduction). Its own disclosure for 2023 lists $24.8 million from drug and device makers. That includes $7.27 million from Eli Lilly, $6.40 million from Abbott, $2.74 million from Novo Nordisk, $1.85 million from Xeris, $0.73 million from Dexcom, $4,050 from Amphastar and $1,600 from Zealand (ADA disclosure). Earlier reporting found Lilly and Sanofi gave the ADA at least $3.9 million in 2016 (Nonprofit Quarterly).
• The International Hypoglycaemia Study Group, whose thresholds define the three levels, is supported by an unrestricted grant from Novo Nordisk (IHSG 2017).

Documented integrity events

We found no regulatory warning letters, recalls or legal actions about the glucagon products named here in the sources we reviewed. The main documented conflict is structural. The bodies setting thresholds and treatment advice receive support from companies that make insulin, glucagon and glucose monitors. This is disclosed, and NICE's government-funded advice broadly agrees with the ADA's.

Related research

For wider, evidence-graded context, see these Pure City Research guides:

• Type 2 diabetes: prevention and management evidence
• Diabetes supplements: benefits, limits and safety
• Berberine: glucose evidence, formulations and safety
• Diabetes and heart disease
• Drowsy driving: medicine effects and licensing context
• Vasovagal syncope: reflex fainting
• Panic disorder: assessment and treatment evidence
• Alcohol-related liver disease
• Chronic kidney disease and the heart

Frequently asked questions

Sources and funding notes

What we could not verify: We could not access the funding statements of the UK Hypoglycaemia Study Group's 2007 paper, the Endocrine Society's 2009 guideline or Charles et al. (1981). We could not find a published audited breakdown of ADA industry income beyond the 2023 draft figures. Xeris's company headquarters was not stated in the source we opened. We could not find Novo Nordisk's own announcement of the 2026 Zegalogue discontinuation, so its scope outside the US and its exact date are unconfirmed. We did not find independent, real-world head-to-head trials comparing glucagon products in actual emergencies. We did not review every US state's driving rules. NICE NG17 covers adults with type 1 diabetes; we did not review NICE guidance for children or type 2 diabetes for this article.

1. ADA Professional Practice Committee. 6. Glycemic goals, hypoglycemia, and hyperglycemic crises: Standards of Care in Diabetes—2026. Diabetes Care 2026 — guideline; ADA general revenue.
2. ADA. Introduction and Methodology: Standards of Care in Diabetes—2026 — funding statement.
3. ADA. Revenue received from pharmaceutical/device makers, 2023 (unaudited draft) — self-disclosure.
4. NICE NG17. Type 1 diabetes in adults: diagnosis and management — UK government-funded guideline.
5. GOV.UK: National Institute for Health and Care Excellence — status and sponsor.
6. NHS. Low blood sugar (hypoglycaemia) — UK public health service.
7. NIDDK. Low blood glucose (hypoglycemia) — US NIH.
8. DVLA. Diabetes mellitus: assessing fitness to drive; DVLA organisation page — UK government.
9. FMCSA. Qualifications of Drivers; Diabetes Standard. Federal Register 2018 — US government.
10. American Diabetes Association. Diabetes and driving. Diabetes Care 2013 — position statement; ADA.
11. International Hypoglycaemia Study Group. Glucose concentrations of less than 3.0 mmol/l should be reported in clinical trials. Diabetologia 2017 — Novo Nordisk unrestricted grant.
12. Cryer PE et al. Evaluation and management of adult hypoglycemic disorders: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2009 — Endocrine Society; funding statement not accessed.
13. FDA. Baqsimi (glucagon) nasal powder prescribing information, 2019 — regulator-approved label.
14. DailyMed. Baqsimi (glucagon) nasal powder, current prescribing information (Amphastar, revised 2025) — regulator-approved label.
15. MedImpact Healthcare Systems. Zegalogue discontinuation notice to Kentucky Medicaid providers (26 February 2026) — US pharmacy benefit manager notice.
16. Rickels MR et al. Intranasal glucagon for treatment of insulin-induced hypoglycemia in adults with type 1 diabetes. Diabetes Care 2016 — Helmsley Trust, NIH, Locemia Solutions.
17. Pieber TR et al. Dasiglucagon phase 3 randomized double-blind trial. Diabetes Care 2021 — sponsored by Zealand Pharma.
18. De Buck E et al. First aid glucose administration routes for symptomatic hypoglycaemia. Cochrane 2019 (PubMed) — Belgian Red Cross authors; commissioned via AHA for ILCOR.
19. Little SA et al. Recovery of hypoglycemia awareness in long-standing type 1 diabetes (HypoCOMPaSS). Diabetes Care 2014 — Diabetes UK (Newcastle University press release; follow-up funding record).
20. van Beers CA et al. CGM for type 1 diabetes and impaired awareness (IN CONTROL). Lancet Diabetes Endocrinol 2016 — Eli Lilly and Sanofi.
21. Hering BJ et al. Phase 3 trial of transplantation of human islets in type 1 diabetes complicated by severe hypoglycemia. Diabetes Care 2016 — NIAID and NIDDK (NIH).
22. FDA. FDA approves first cellular therapy to treat patients with type 1 diabetes (2023) — regulator.
23. Zoungas S et al. Severe hypoglycemia and risks of vascular events and death. N Engl J Med 2010 — Servier and NHMRC.
24. UK Hypoglycaemia Study Group. Risk of hypoglycaemia in types 1 and 2 diabetes. Diabetologia 2007 — funding not verified.
25. Edridge CL et al. Prevalence and incidence of hypoglycaemia in 532,542 people with type 2 diabetes. PLoS One 2015 — no funding; authors disclose industry consultancy.
26. Andersen SE, Christensen M. Hypoglycaemia when adding sulphonylurea to metformin. Br J Clin Pharmacol 2016 — no external funding.
27. Turner BC et al. Evening alcohol and next-morning glucose control in type 1 diabetes. Diabetes Care 2001 — PubMed lists an NIDDK grant.
28. Charles MA et al. Oral glucose tolerance tests and mixed meals in apparent idiopathic postabsorptive hypoglycemia. Diabetes 1981 — academic; funding not shown.
29. Service FJ et al. Functioning insulinoma: incidence, recurrence and long-term survival. Mayo Clin Proc 1991 — NIH (NIAMS) grant.
30. Salehi M, Vella A, McLaughlin T, Patti ME. Hypoglycemia after gastric bypass surgery. J Clin Endocrinol Metab 2018 — NIH grants; authors disclose ties to Eiger, Novo Nordisk, Novartis and Xeris.
31. Lobato CB et al. Acarbose or canagliflozin vs placebo for post-bariatric hypoglycaemia (HypoBar trial). Diabetes Obes Metab 2026 — Novo Nordisk Foundation, la Caixa Foundation, Danish hospital funds.
32. NCCIH. Diabetes and dietary supplements — US NIH.
33. Amphastar Pharmaceuticals completes acquisition of BAQSIMI from Lilly (2023) — company press release.
34. Zealand Pharma. Global licence agreement with Novo Nordisk for Zegalogue (2022); Zegalogue US availability (2021) — company filings.
35. PharmTech. Xeris Pharmaceuticals gets FDA nod for liquid glucagon (2019) — trade press.
36. Nonprofit Quarterly (2017) — reporting on insulin-maker donations to patient groups.

Educational research reviewed 7 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.