Insulin therapy in type 2 diabetes roughly doubles all-cause mortality and raises cancer, renal and cardiovascular events, with a dose-response (each extra IU/kg/day +54% mortality), compared with metformin or sulfonylureas.
Original statement
ودراسة Currie الواسعة على 84,622 مريضًا وجدت أن الأنسولين منفردًا ضاعف الوفيات تقريبًا (HR 2.20) ورفع السرطان والفشل الكلوي مقارنةً بالميتفورمين. وأكدت دراسة Holden علاقة الجرعة بالوفاة: كل وحدة إضافية ترفع الوفيات 54%.
verbatim according to source. Translation by the project; the Arabic wording is authoritative.
Translation
Insulin therapy in type 2 diabetes roughly doubles all-cause mortality and raises cancer, renal and cardiovascular events, with a dose-response (each extra IU/kg/day +54% mortality), compared with metformin or sulfonylureas.
Origin
- Source status
- Archive editors
- Attribution
- Archive editors (not Elawady)
- Origin
- https://eltayebat.com/rule/insulin-injections/
- Attribution uncertainty
- Editorial narrative and study selection written by the archive operators.
- Claimed mechanism
Exogenous insulin (hyperinsulinaemia) directly causes mortality, cancer and vascular disease.
- Claimed health effect
Death, cancer, MI, stroke, renal failure
- Population concerned
Type 2 diabetes
- Geographic restriction by Elawady
none found
Scientific assessment
Mostly false – Essential parts of the statement are not supported by the evidence or are strongly exaggerated.
Certainty of evidence: low – Further research is likely to change the assessment considerably.
1. What was claimed?
Insulin itself doubles mortality and raises cancer and complications in a dose-dependent way.
2. What is demonstrably correct?
The figures are quoted correctly as associations: insulin monotherapy vs metformin aHR 2.197 for death, 1.437 for cancer (A-INS-08; this study did not analyse insulin dose); per 1-unit increase in IU/kg/day aHR 1.54 for death (A-INS-10); adding insulin vs sulfonylurea to metformin aHR 1.30 for the composite and 1.44 for death (A-INS-13); insulin use and cancer mortality HR 1.9 (A-INS-11) with a gradient by number of pharmacy dispensations (A-INS-12; cancer mortality only, called preliminary by its authors); insulin-based regimens vs combination oral agents HR 1.49 (A-INS-09).
3. What is wrong or exaggerated?
All of these are retrospective observational studies. In Currie et al. insulin users differed sharply at baseline (diabetes duration 5.2 vs 1.5 years; creatinine >130 umol/L in 21.1% vs 2.2%), follow-up was censored at regimen switch, and bias is judged critical; the study cannot support a causal 'doubling' or the dose-response element (A-INS-08). In Bowker et al. the 'dose' was the number of dispensations counted in the same window as death, which invites reverse causation (A-INS-12). Holden et al. state the association should be tested in interventional studies (A-INS-10); Bowker et al. state it is uncertain whether the risk reflects harm from insulin, protection by metformin, or unmeasured factors (A-INS-11); Roumie et al. call for further investigation (A-INS-13). Patients receive insulin, and higher doses, because their disease is more advanced and they are more insulin resistant, obese or ill (confounding by indication; reverse causation). In the randomised ORIGIN trial, glargine had neutral effects on cardiovascular outcomes (HR 1.02) and cancers (HR 1.00) over 6.2 years (A-INS-07, I-INS-05), and in UKPDS insulin-based intensive therapy reduced microvascular complications without excess macrovascular disease (A-INS-16) and was associated with fewer deaths at 10 years (I-INS-02).
4. What remains unclear?
RCT evidence in very insulin-resistant, high-dose, late-stage type 2 diabetes is limited; whether high insulin doses carry some independent risk cannot be excluded.
5. Which studies support it?
A-INS-08, A-INS-09, A-INS-10, A-INS-11, A-INS-12, A-INS-13 (associations only; the IU/kg/day dose-response comes from A-INS-10 alone).
6. Which studies contradict it?
A-INS-07, I-INS-05, A-INS-16, I-INS-02 (randomised evidence).
7. What applies to Germany and Europe?
European (EASD) and German (DDG) guidance follow the same evidence base as the ADA/EASD consensus: individualised HbA1c targets, insulin as mandatory therapy in type 1 diabetes, and SGLT2 inhibitors/GLP-1 receptor agonists for people with cardiovascular or kidney disease. No country-specific difference changes the assessment.
8. What is known about Egypt?
Egypt has one of the highest diabetes prevalences in the region; the same drug classes and insulins are used. Access barriers and cost may make advice to stop medication attractive, which increases the practical risk of this claim. No Egypt-specific evidence changes the assessment.
9. What practical conclusion is appropriate?
Observational signals are a reason for guidelines to prefer metformin and cardio-renally protective drugs and to avoid unnecessary insulin escalation in type 2 diabetes. They do not show that insulin kills, and they do not justify stopping insulin where it is needed.
Assessment details
- Rationale for certainty of evidence
The numbers are correctly reported from observational cohorts, but these are strongly confounded by indication (sicker patients with longer disease and failure of tablets receive insulin; doses rise with insulin resistance and illness); the authors themselves call for interventional studies. RCTs (ORIGIN, UKPDS) found no increase in mortality, cardiovascular events or cancer with insulin, so the causal claim is contradicted, as in C-INS-03 and C-INS-06. Certainty remains low because RCT populations (early dysglycaemia, newly diagnosed) differ from late-stage, high-dose insulin users.
- Source quality
Archive editorial quoting retrospective database cohorts.
- Regional transferability
UK, Canadian and US data; transferable as associations.
- Regional scope of the claim
- general (biological/medical)
- Last reviewed
- 25 October 2026 provisional
In plain language
Several large database studies found that people with type 2 diabetes who use insulin die earlier and get more cancer than people who use tablets. But insulin users are usually sicker and have had diabetes longer, so the comparison is unfair. When insulin was tested fairly in randomised trials, it did not increase deaths, heart disease or cancer. The study authors themselves said their results do not prove cause and effect.
Studies on this claim
Studies supporting the claim (6)
-
UK database: insulin users vs metformin users
Currie CJ et al.. J Clin Endocrinol Metab 2013 · PMID 23372169 · DOI 10.1210/jc.2012-3042Cohort studycited by the Tayebat archivefull text checked★ key studyPeople treated with insulin alone had about twice the death rate of metformin users, but they were also sicker.
Does the study support the specific claim? Partly · Methodological quality: very low · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Risk of adverse events with five glucose-lowering regimens
- How was the study done?
Retrospective cohort, UK General Practice Research Database
- How many people, animals or samples were studied?
84,622 patients; 105,123 exposure periods; Primary care type 2 diabetes
- What was actually found?
Insulin monotherapy associated with higher risks of death, MACE, cancer, complications.
- How large was the observed effect?
aHR 2.197 death
- What do the results mean in everyday life?
Shows insulin users fare worse, not that insulin causes it.
- What are the study's limitations?
Insulin users had longer diabetes (5.2 vs 1.5 years), more kidney and heart disease; ~3 years follow-up; censoring at treatment switch; authors note confounding by indication; authors have industry ties.
- What does it say about Elawady's specific claim?
Main observational support for C-INS-05.
- Do the results apply to Germany and Europe?
UK data.
- What can explicitly not be concluded from this study?
That insulin causes death, cancer or kidney failure; any insulin dose-response.
-
Long-term sugar level and survival in type 2 diabetes
Currie CJ et al.. Lancet 2010 · PMID 20110121 · DOI 10.1016/S0140-6736(09)61969-3Cohort studycited by the Tayebat archiveabstract only checkedBoth very low (6.4%) and very high (10.5%) HbA1c were linked to more deaths; lowest risk around 7.5%.
Does the study support the specific claim? Partly · Methodological quality: low · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Survival as a function of HbA1c
- How was the study done?
Retrospective cohort (GPRD)
- How many people, animals or samples were studied?
47,970 (27,965 + 20,005); Type 2 diabetes >=50 years intensified to combination orals or insulin
- What was actually found?
U-shaped mortality curve.
- How large was the observed effect?
HR 1.52 low; 1.79 high
- What do the results mean in everyday life?
Moderate control is associated with best survival; high sugar is also harmful.
- What are the study's limitations?
Observational; Confounding by indication; Industry funding
- What does it say about Elawady's specific claim?
Supports C-INS-04 (U-shape) and mixed for C-INS-05; also contradicts C-DIAB-01 (high HbA1c 10.5% HR 1.79).
- Do the results apply to Germany and Europe?
UK data.
- What can explicitly not be concluded from this study?
Causality.
-
Higher insulin doses and outcomes
Holden SE et al.. Diabetes Obes Metab 2015 · PMID 25399739 · DOI 10.1111/dom.12412Cohort studycited by the Tayebat archiveabstract only checked★ key studyPeople on higher insulin doses died more often, but higher doses are needed by sicker, more insulin-resistant people.
Does the study support the specific claim? Partly · Methodological quality: low · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Association of insulin dose with mortality, MACE, cancer
- How was the study done?
Retrospective cohort, UK CPRD, time-dependent dose exposure
- How many people, animals or samples were studied?
6,484; Type 2 diabetes progressing to insulin monotherapy from 2000
- What was actually found?
Dose-associated increases in death, MACE, cancer.
- How large was the observed effect?
aHR 1.54 per IU/kg/day
- What do the results mean in everyday life?
High doses mark severe disease; cause unclear.
- What are the study's limitations?
Dose reflects insulin resistance and illness; Reverse causation; Residual confounding
- What does it say about Elawady's specific claim?
- Do the results apply to Germany and Europe?
UK data.
- What can explicitly not be concluded from this study?
That lowering doses would save lives.
-
Diabetes drugs and cancer deaths in Canada
Bowker SL et al.. Diabetes Care 2006 · PMID 16443869 · DOI 10.2337/diacare.29.02.06.dc05-1558Cohort studycited by the Tayebat archiveabstract only checkedInsulin users had more cancer deaths than metformin users, possibly due to other differences.
Does the study support the specific claim? Partly · Methodological quality: low · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Antidiabetic therapy and cancer mortality
- How was the study done?
Population-based cohort (Saskatchewan Health)
- How many people, animals or samples were studied?
10,309; New users of metformin or sulfonylureas
- What was actually found?
aHR 1.9 for insulin.
- How large was the observed effect?
aHR 1.9 (1.5-2.4)
- What do the results mean in everyday life?
Association, cause unknown.
- What are the study's limitations?
No BMI/smoking; Confounding
- What does it say about Elawady's specific claim?
- Do the results apply to Germany and Europe?
Canadian data.
- What can explicitly not be concluded from this study?
Causation.
-
Amount of insulin and cancer deaths
Bowker SL et al.. Diabetologia 2010 · PMID 20407744 · DOI 10.1007/s00125-010-1750-8Cohort studycited by the Tayebat archivefull text checkedMore insulin prescriptions went with more cancer deaths, which may reflect that very ill people need more insulin.
Does the study support the specific claim? Partly · Methodological quality: low · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Glucose-lowering agents and cancer mortality
- How was the study done?
Population-based cohort with time-varying exposure
- How many people, animals or samples were studied?
10,309; New users of metformin/SU
- What was actually found?
More insulin dispensations went with more cancer deaths (HR up to 6.40); weaker (HR 2.47) when only new insulin users were compared.
- How large was the observed effect?
aHR 6.40 (>=12 dispensations)
- What do the results mean in everyday life?
Association only.
- What are the study's limitations?
No data on BMI, smoking, blood sugar control; insulin 'dose' is the number of pharmacy fills, which rises when people are seriously ill; few deaths.
- What does it say about Elawady's specific claim?
Association for C-INS-05.
- Do the results apply to Germany and Europe?
Canadian data.
- What can explicitly not be concluded from this study?
Causation.
-
Second drug after metformin: insulin or sulfonylurea
Roumie CL et al.. JAMA 2014 · PMID 24915260 · DOI 10.1001/jama.2014.4312Cohort studycited by the Tayebat archivefull text checked★ key studyVeterans who added insulin died more often than those who added a sulfonylurea, but heart attacks and strokes were not more frequent.
Does the study support the specific claim? Partly · Methodological quality: moderate · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Adding insulin vs sulfonylurea to metformin: AMI, stroke, death
- How was the study done?
Retrospective cohort with propensity matching and marginal structural models (VHA, Medicare, NDI)
- How many people, animals or samples were studied?
178,341 (matched 2,436 insulin vs 12,180 SU); Veterans initially treated with metformin
- What was actually found?
Composite aHR 1.30; death aHR 1.44; CV events not increased.
- How large was the observed effect?
aHR 1.30 (1.07-1.58)
- What do the results mean in everyday life?
Possibly reflects sicker patients getting insulin.
- What are the study's limitations?
Residual confounding (sicker patients get insulin); median follow-up about 1 year; mostly male veterans; extra deaths were mainly from cancer and other non-heart causes.
- What does it say about Elawady's specific claim?
Association for C-INS-05; mixed for C-INS-03 (no CV increase).
- Do the results apply to Germany and Europe?
US veterans.
- What can explicitly not be concluded from this study?
That insulin causes death.
No study in this group is marked as key.
Studies contradicting or limiting the claim (4)
-
Long-term insulin glargine in early diabetes (ORIGIN)
ORIGIN Trial Investigators; Gerstein HC et al.. N Engl J Med 2012 · PMID 22686416 · DOI 10.1056/NEJMoa1203858Randomised controlled trialcited by the Tayebat archivefull text checked★ key studyOver six years insulin did not change heart attacks, strokes or cancers, but caused more low-sugar episodes and 2 kg weight difference.
Does the study support the specific claim? No · Methodological quality: high · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Does basal insulin glargine targeting normal fasting glucose reduce CV events in dysglycaemia?
- How was the study done?
International multicentre RCT, 2x2 factorial (ORIGIN)
- How many people, animals or samples were studied?
12,537; CV risk factors plus IFG, IGT or type 2 diabetes; mean age 63.5
- What was actually found?
Neutral CV and cancer outcomes; more hypoglycaemia, slight weight gain.
- How large was the observed effect?
HR 1.02 CV; HR 1.00 cancer
- What do the results mean in everyday life?
Insulin itself did not cause heart disease or cancer in this large trial.
- What are the study's limitations?
Open-label; funded by the glargine manufacturer; early-diabetes population on moderate doses (~28 units/day for 70 kg).
- What does it say about Elawady's specific claim?
Key randomised evidence against C-INS-03, C-INS-05, C-INS-06; supports hypoglycaemia/weight side effects.
- Do the results apply to Germany and Europe?
Multinational/US/UK/European data; directly transferable to Germany/EU clinical practice.
- What can explicitly not be concluded from this study?
Safety in very high-dose, late-stage type 2 diabetes.
-
Glargine and cancer in a randomised trial
Bordeleau L et al.. Diabetes Care 2014 · PMID 24574355 · DOI 10.2337/dc13-1468Randomised controlled trialindependent cross-checkabstract only checked★ key studyCancer rates were identical with glargine and standard care (1.32 per 100 person-years each).
Does the study support the specific claim? No · Methodological quality: high · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Effect of glargine and n-3 fatty acids on incident cancers
- How was the study done?
Cancer substudy of the ORIGIN 2x2 factorial RCT
- How many people, animals or samples were studied?
12,537 (953 cancer events); Dysglycaemia with high CV risk, mean age 63.5
- What was actually found?
No difference in any cancer outcome.
- How large was the observed effect?
1.32 vs 1.32 per 100 PY
- What do the results mean in everyday life?
Glargine does not increase cancer within 6 years.
- What are the study's limitations?
Industry funding; 6 years may be short for some cancers
- What does it say about Elawady's specific claim?
- Do the results apply to Germany and Europe?
Multinational trial including European sites or directly applicable to EU practice.
- What can explicitly not be concluded from this study?
Effects after many decades.
-
UKPDS: tablets or insulin for new type 2 diabetes
UKPDS Group. Lancet 1998 · PMID 9742976Randomised controlled trialcited by the Tayebat archiveabstract only checked★ key studyBetter sugar control with tablets or insulin cut eye/kidney complications by a quarter; insulin caused more weight gain and low-sugar episodes.
Does the study support the specific claim? Partly · Methodological quality: high · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Intensive control with sulfonylurea or insulin vs diet on complications
- How was the study done?
Multicentre RCT (UKPDS 33)
- How many people, animals or samples were studied?
3,867; Newly diagnosed type 2 diabetes, median age 54
- What was actually found?
Microvascular -25%; more hypoglycaemia and weight gain with insulin.
- How large was the observed effect?
RR reduction 25% microvascular
- What do the results mean in everyday life?
Insulin helps prevent complications but has side effects.
- What are the study's limitations?
Open-label; Older drug regimens
- What does it say about Elawady's specific claim?
Supports weight/hypoglycaemia (C-INS-01); contradicts C-INS-03, C-INS-05, C-DIAB-01.
- Do the results apply to Germany and Europe?
UK; directly relevant.
- What can explicitly not be concluded from this study?
That insulin is harmful overall.
-
UKPDS legacy: 10 years after the trial
Holman RR et al.. N Engl J Med 2008 · PMID 18784090 · DOI 10.1056/NEJMoa0806470Randomised controlled trialindependent cross-checkabstract only checked★ key studyPeople treated intensively early with sulfonylureas or insulin had 15% fewer heart attacks and 13% fewer deaths years later.
Does the study support the specific claim? No · Methodological quality: high · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Do UKPDS benefits persist and do macrovascular effects emerge?
- How was the study done?
Post-trial monitoring of an RCT (UKPDS 80)
- How many people, animals or samples were studied?
4,209 randomised; 3,277 monitored; Newly diagnosed type 2 diabetes from UKPDS
- What was actually found?
Legacy reductions in MI and death.
- How large was the observed effect?
Death -13% (SU/insulin), -27% (metformin)
- What do the results mean in everyday life?
Early good control pays off for decades.
- What are the study's limitations?
Post-trial therapy not controlled; Questionnaire follow-up
- What does it say about Elawady's specific claim?
Contradicts C-INS-03, C-INS-05 and C-DIAB-01: early insulin/SU therapy reduced MI and death long-term.
- Do the results apply to Germany and Europe?
UK; directly relevant.
- What can explicitly not be concluded from this study?
Effects of extreme targets in late disease.
No study in this group is marked as key.
Studies with mixed or unclear results (1)
-
Adding metformin to insulin
Wulffelé MG et al.. Diabetes Care 2002 · PMID 12453950 · DOI 10.2337/diacare.25.12.2133Randomised controlled trialcited by the Tayebat archiveabstract only checkedMetformin improved sugar control and reduced insulin need and weight gain.
Does the study support the specific claim? Yes · Methodological quality: moderate · Certainty of evidence: see overall assessment of the claim
Original publication · Scientific details
Study explained simply
- What did the researchers want to find out?
Metformin added to intensive insulin in type 2 diabetes
- How was the study done?
Double-blind placebo-controlled RCT
- How many people, animals or samples were studied?
390; Insulin-treated type 2 diabetes
- What was actually found?
Better control, less insulin, less weight gain.
- How large was the observed effect?
GHb -0.7%
- What do the results mean in everyday life?
Combination lowers insulin needs.
- What are the study's limitations?
Short duration; Dropouts higher with metformin
- What does it say about Elawady's specific claim?
- Do the results apply to Germany and Europe?
Dutch data.
- What can explicitly not be concluded from this study?
That insulin can be stopped.
No study in this group is marked as key.
Regional context
Germany / EU
European (EASD) and German (DDG) guidance follow the same evidence base as the ADA/EASD consensus: individualised HbA1c targets, insulin as mandatory therapy in type 1 diabetes, and SGLT2 inhibitors/GLP-1 receptor agonists for people with cardiovascular or kidney disease. No country-specific difference changes the assessment.
Egypt
Egypt has one of the highest diabetes prevalences in the region; the same drug classes and insulins are used. Access barriers and cost may make advice to stop medication attractive, which increases the practical risk of this claim. No Egypt-specific evidence changes the assessment.
What applies in my country?
General scientific assessment
Mostly false Insulin users have worse outcomes in databases, but randomised trials do not show that insulin causes these deaths.
This statement concerns biological or medical relationships. The scientific assessment above applies regardless of the country; country-specific food data do not change it.
References
- Mortality and other important diabetes-related outcomes with insulin vs other antihyperglycemic therapies in type 2 diabetes.. Currie CJ et al.. J Clin Endocrinol Metab 2013 · PMID 23372169 · DOI 10.1210/jc.2012-3042
- Survival as a function of HbA(1c) in people with type 2 diabetes: a retrospective cohort study.. Currie CJ et al.. Lancet 2010 · PMID 20110121 · DOI 10.1016/S0140-6736(09)61969-3
- Glucose-lowering with exogenous insulin monotherapy in type 2 diabetes: dose association with all-cause mortality, cardiovascular events and cancer.. Holden SE et al.. Diabetes Obes Metab 2015 · PMID 25399739 · DOI 10.1111/dom.12412
- Increased cancer-related mortality for patients with type 2 diabetes who use sulfonylureas or insulin.. Bowker SL et al.. Diabetes Care 2006 · PMID 16443869 · DOI 10.2337/diacare.29.02.06.dc05-1558
- Glucose-lowering agents and cancer mortality rates in type 2 diabetes: assessing effects of time-varying exposure.. Bowker SL et al.. Diabetologia 2010 · PMID 20407744 · DOI 10.1007/s00125-010-1750-8
- Association between intensification of metformin treatment with insulin vs sulfonylureas and cardiovascular events and all-cause mortality among patients with diabetes.. Roumie CL et al.. JAMA 2014 · PMID 24915260 · DOI 10.1001/jama.2014.4312
- Basal insulin and cardiovascular and other outcomes in dysglycemia.. ORIGIN Trial Investigators; Gerstein HC et al.. N Engl J Med 2012 · PMID 22686416 · DOI 10.1056/NEJMoa1203858
- The association of basal insulin glargine and/or n-3 fatty acids with incident cancers in patients with dysglycemia.. Bordeleau L et al.. Diabetes Care 2014 · PMID 24574355 · DOI 10.2337/dc13-1468
- Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group.. UKPDS Group. Lancet 1998 · PMID 9742976
- 10-year follow-up of intensive glucose control in type 2 diabetes.. Holman RR et al.. N Engl J Med 2008 · PMID 18784090 · DOI 10.1056/NEJMoa0806470
- Combination of insulin and metformin in the treatment of type 2 diabetes.. Wulffelé MG et al.. Diabetes Care 2002 · PMID 12453950 · DOI 10.2337/diacare.25.12.2133