Type 1 diabetes mellitus

• Type 1 diabetes mellitus is an autoimmune disease in which the body's own immune system destroys the insulin-producing beta cells of the pancreas. The result is an absolute lack of insulin, without which the body cannot use glucose for fuel and begins to break down fat, risking diabetic ketoacidosis.
• It is the most common form of diabetes in children and young adults, but it can begin at any age, including well into adult life. About 5 to 10% of all diabetes is type 1.^[1]
• Most people still have no family history: roughly 90% of those who develop type 1 diabetes have no affected first-degree relative.^[1]
• The disease announces itself years before symptoms through islet autoantibodies. Its natural history is staged from 1 (silent autoimmunity, normal glucose) through 2 (silent autoimmunity, abnormal glucose) to 3 (clinical diabetes).^[1]
• Treatment rests on lifelong exogenous insulin, usually as a basal-bolus regimen, together with carbohydrate counting, self-monitoring or continuous glucose monitoring, and rapid response to low blood glucose. One disease-modifying treatment, teplizumab, can delay the onset of clinical disease when it is given during stage 2.^[6]

By the end of this page you should be able to:

1. Explain the autoimmune destruction of beta cells, including the islet autoantibodies and the HLA risk alleles, and why other autoimmune diseases cluster with type 1 diabetes.
2. Define and apply the three stages of type 1 diabetes and their risk of progression.
3. List the diagnostic criteria for diabetes and know when a random glucose with symptoms is enough.
4. Plan insulin replacement (basal-bolus, multiple daily injections versus pump), carbohydrate counting, glucose monitoring and hypoglycaemia treatment.
5. State the targets for A1C and time in range and interpret a glucose profile.
6. Plan screening for associated autoimmune disease in a person with type 1 diabetes, and explain management in a low-resource setting.

Type 1 diabetes mellitus (T1DM) is a metabolic disorder in which autoimmune destruction of the pancreatic beta cells leads, eventually, to an absolute deficiency of insulin. Insulin is the hormone that enables glucose to enter most cells; without it the body cannot use dietary glucose, the liver continues to release glucose, and blood glucose rises. Once hyperglycaemia is established, the cells starve despite high blood sugar, and the body breaks down fat, releasing ketone bodies that can produce diabetic ketoacidosis (DKA).

The classification matters, because treatment follows from the underlying mechanism, not from the age of the patient. Type 2 diabetes arises from progressive beta-cell failure, frequently on a background of insulin resistance, and may initially be treated without insulin. Type 1 diabetes carries an absolute insulin deficiency and cannot be managed without exogenous insulin for long. Other forms of diabetes, such as monogenic diabetes or diabetes after pancreatitis, must also be considered when the clinical picture is atypical.^[1]

A useful clinical mnemonic is the AABBCC approach, where Age (under 35 years favours type 1), Autoimmunity (personal or family autoimmune disease), Body habitus (body mass index under 25 kg/m2), Background (family history of type 1 diabetes), Control (inability to achieve goals on non-insulin therapy) and Comorbidities (such as checkpoint-inhibitor therapy) are weighed together.^[1] Other features classically attributed to type 1 diabetes, such as ketosis without acidosis, osmotic symptoms, family history alone or a history of autoimmune disease are weak discriminators on their own, and classification is misdiagnosed in about 40% of adults with new type 1 diabetes.^[1]

Autoimmune type 1 diabetes accounts for 5 to 10% of all diabetes.^[1] It is the most common form of diabetes in children and adolescents, but there are more adults living with and diagnosed with type 1 diabetes than children, because adults live with it for decades.^[5]

The presentation is typically acute. Children often present with the classic triad of polyuria, polydipsia and weight loss, and approximately half present in diabetic ketoacidosis.^[1] Adults may present more slowly, without the classic symptoms, and may retain enough beta-cell function to delay insulin dependence, sometimes for years; remission with decreased insulin needs for months or years is possible.^[1] Among adults, type 1 diabetes may be diagnosed late and mislabelled as type 2 diabetes.

Epidemiological markers from current practice include:

• Newly diagnosed type 1 diabetes presents with life-threatening DKA in 25 to 50% of cases.^[1]
• A first-degree relative with type 1 diabetes increases risk, but roughly 90% of those who develop it have no affected relative.^[1]
• Among children who developed more than two autoantibodies, almost 70% developed clinical diabetes within 10 years and 84% within 15 years.^[1]
• The incidence and prevalence of type 1 diabetes are increasing, with several countries introducing population screening programmes.^[1]

Type 1 diabetes is, at its core, an autoimmune attack on the beta cells. The immune system loses tolerance to beta-cell antigens and, over months to years, T-lymphocytes infiltrate the islets (insulitis) and destroy the insulin-secreting cells. The rate of destruction varies: it can be rapid, particularly in infants and children, or slow, mainly in adults. Once hyperglycaemia develops, all forms of diabetes share the same chronic complication risk.^[1]

Autoimmune (type 1) diabetes includes latent autoimmune diabetes in adults (LADA), every form mediated by autoimmune beta-cell destruction regardless of age of onset. The term LADA is common and clinically useful because it flags adults in whom autoimmune destruction progresses slowly, accelerating insulin initiation before glycaemic deterioration or ketoacidosis.^[1]

The markers of the autoimmune process are islet autoantibodies. Current guidance uses autoantibodies to:

• glutamic acid decarboxylase (GAD), often as GAD65;
• insulin (more useful before insulin has been started);
• islet antigen 2 (IA-2) and IA-2b;
• zinc transporter 8 (ZnT8).^[1]

Having two or more confirmed islet autoantibodies is a near-certain predictor of clinical diabetes.^[1] The autoantibody data come from large birth cohorts and trials, and have been standardised through the Islet Autoantibody Standardization Program. For general population screening, the advice is that the first positive autoantibody test be confirmed on a second sample within 3 months. Up to half of children who have only a single autoantibody revert to being antibody-negative during follow-up.^[1]

The disease has strong HLA associations, particularly with the DQB1 and DRB1 haplotypes. The predisposing alleles are DRB10301-DQB10201 (written DR3-DQ2) and DRB10401-DQB10302 (DR4-DQ8). Protective alleles include DRB11501 and DQA10102-DQB1*0602.^[1] Genetic risk scores built from these and other loci can, in simulation, identify the majority of future cases within a small fraction of the general population, but many studies have been performed in populations of European ancestry, and discriminatory ability may differ in other populations.^[1]

Genetic and environmental factors combine, and the environmental triggers are still poorly defined. Viruses, particularly enteroviruses such as Coxsackievirus B, have been associated with type 1 diabetes, and the COVID-19 pandemic was associated with an increase in new diagnoses and in DKA. Checkpoint inhibitors used for cancer can precipitate a fulminant autoimmune diabetes, often with low or undetectable C-peptide and with fewer than half of patients carrying classic autoantibodies. Immune checkpoint inhibitor-associated diabetes occurs in just under 1% of treated individuals.^[1]

• Younger age at onset and lower body mass index are associated with type 1 rather than type 2 diabetes; older adults may present more slowly.^[1]
• Obesity is increasingly common in type 1 diabetes and should not preclude testing; many people have features of both types.
• Family history matters, but most cases are "sporadic".
• Other autoimmune diseases: Hashimoto thyroiditis, Graves disease, coeliac disease, Addison disease, vitiligo, autoimmune hepatitis, myasthenia gravis, pernicious anaemia, and polyglandular autoimmune syndromes including immune dysregulation with polyendocrinopathy, enteropathy and X-linked (IPEX) syndrome.

The sequence is the lesson. It explains the staging that follows.

1. Genetic susceptibility (HLA and other loci) plus an environmental trigger begins autoimmunity.
2. Seroconversion: one, then two or more islet autoantibodies appear. Seroconversion rarely occurs before 6 months of age, with a peak between 9 and 24 months in those followed from birth.^[1]
3. Beta-cell mass declines gradually while the person remains outwardly well. Glucose and A1C may start to rise months before clinical diagnosis.
4. Dysglycaemia (impaired fasting glucose, impaired glucose tolerance or A1C in the pre-diabetes range) marks stage 2.
5. Clinical diabetes (stage 3), with hyperglycaemia, osmotic symptoms and, in some, ketoacidosis, appears once insulin secretory capacity is inadequate.
6. Remission (honeymoon) may follow initiation of insulin, with residual beta-cell function lowering insulin needs for months or years.
7. Eventually near-absent C-peptide, and lifelong insulin dependence, with ongoing risk of DKA.

C-peptide is the connecting peptide cleaved from proinsulin. It is secreted in equimolar amounts with insulin and is a practical marker of how much insulin the beta cells still make. Low or undetectable C-peptide signals severe insulin deficiency. For classification, a C-peptide level of at least 600 pmol/L (at least 1.8 ng/mL) regardless of circumstances, a random sample within 5 hours of eating, and repetition if the value is between 200 and 600 pmol/L when the glucose or fasting state is uncertain are used; very low levels under 80 pmol/L need not be repeated. C-peptide should not be tested within 2 weeks of a hyperglycaemic emergency.^[1]

The symptoms of hyperglycaemia are the same whatever the type of diabetes, and they come from osmotic diuresis.

• Polyuria (passing urine frequently, including nocturia) and polydipsia (thirst) from glucose exceeding the renal threshold.
• Weight loss despite normal eating, because glucose cannot enter cells and fat and protein are consumed.
• Fatigue, blurred vision and recurrent infections (thrush, skin, urinary) from hyperglycaemia.
• Abdominal pain, nausea, vomiting and breathlessness suggest DKA and are an emergency.
• Some adults present with milder symptoms over months, or with complications already present.

The adult presentation is variable. Adults may not have the classic symptoms seen in children and may progress to insulin replacement more slowly. Misdiagnosis is common and can occur in roughly 40% of adults with new type 1 diabetes, for example adults with type 1 diabetes misdiagnosed as having type 2.^[1] A useful safeguard: if an adult with "type 2 diabetes" has low C-peptide, is lean, or does not respond to oral agents, reconsider the diagnosis.

There is no single sign. Look for:

• dehydration, tachycardia and hypotension if DKA is present;
• fruity (acetone) breath in DKA;
• signs of weight loss and muscle wasting;
• retinopathy or other complications in long-standing disease;
• evidence of associated autoimmune disease (goitre, vitiligo), which should be looked for in every patient.

After diagnosis and the start of insulin, many people enter a remission phase in which residual beta-cell function improves and insulin needs fall for months or years. This is a reprieve, not a cure: eventually beta-cell function declines and insulin requirements rise again, and the person remains at risk of DKA.^[1] Stringent A1C goals may be possible during this phase.^[5]

The main contrast is with type 2 diabetes, because treatment and prognosis differ. Use a table rather than memory.

Diabetes is diagnosed on A1C or plasma glucose. The criteria are:

• A1C at least 6.5% (at least 48 mmol/mol); or
• Fasting plasma glucose at least 126 mg/dL (at least 7.0 mmol/L, after at least 8 hours without calories); or
• 2-hour plasma glucose at least 200 mg/dL (at least 11.1 mmol/L) during a 75 g oral glucose tolerance test; or
• Random plasma glucose at least 200 mg/dL (at least 11.1 mmol/L) in an individual with classic hyperglycaemic symptoms or a hyperglycaemic crisis.^[1]

In the absence of unequivocal hyperglycaemia or a crisis, diagnosis requires two abnormal results, either from two different tests taken together, or the same test repeated at another time.^[1] Plasma glucose is an acute measure of the last few hours; A1C reflects average glycaemia over roughly 2 to 3 months and is less disturbed by recent meals, but it is limited by conditions that alter red cell turnover (anaemia, haemolysis, pregnancy, recent transfusion, iron deficiency). If the result is borderline, repeat testing in 3 to 6 months and educate the patient about symptoms.^[1]

For assessing the type of diabetes, the most useful features in adults are younger age at diagnosis, body mass index, unintentional weight loss, ketoacidosis at presentation and a marked plasma glucose elevation above 360 mg/dL (above 20 mmol/L).^[1] Autoantibody testing and C-peptide are used when the type is unclear, and referral to a specialised centre is appropriate.^[1]

• Fasting and 2-hour glucose measure acute glucose handling and are inexpensive, but they vary day to day and require proper sample handling.
• A1C is the primary monitoring tool; 6.5% or above is also a diagnostic threshold. It cannot measure glycaemic variability or hypoglycaemia.
• Glucose monitoring (capillary or continuous) complements A1C, particularly in people using insulin.
• Autoantibodies tell you the diabetes is autoimmune; their number, titre and specificity change the stage and the risk of progression.
• C-peptide estimates residual beta-cell function and helps distinguish type 1 from type 2 and monogenic diabetes; it is only interpretable when interpreted with the concurrent glucose and clinical context.

The three-stage framework for type 1 diabetes is now used for research, regulatory decisions and the new disease-modifying treatments. Stage 1 and stage 2 are usually found through screening (for example, of relatives of a person with type 1 diabetes, or in population programmes), while stage 3 is how most patients first present.

People with stage 1 or 2 disease have presymptomatic type 1 diabetes and will almost certainly develop stage 3 unless the autoimmune process is interrupted or modified.^[6] Screening for presymptomatic type 1 diabetes may be done by detection of autoantibodies to insulin, GAD, IA-2 or ZnT8, and autoantibody-based screening should be offered to those with a family history or known elevated genetic risk.^[1] When multiple islet autoantibodies are identified, referral to a specialised centre for further evaluation or consideration of a prevention trial or an approved therapy to delay stage 3 is recommended.^[1]

All people with type 1 diabetes require exogenous insulin. The aim is to reproduce, as far as possible, normal pancreatic secretion: a low background (basal) level at all times, with surges (bolus) for meals. Basal-bolus therapy is therefore central. Patient education in insulin self-titration and the safe use of technology is essential.^[3]

Multiple daily injections (MDI). A typical regimen uses one or two long-acting basal injections plus rapid-acting bolus insulin with each meal and for corrections. MDI suits most adults and children and does not require a pump.

Continuous subcutaneous insulin infusion (CSII, insulin pump). A pump delivers a continuous basal rate and user-triggered boluses. Pumps (including smart pumps with bolus calculators) and automated insulin delivery (AID) systems that link a pump to a continuous glucose monitor through an algorithm are now offered in youth and adults who are capable of using them safely.^[3][5] Early initiation, including at diagnosis, is recommended for CGM, CSII and AID, based on individual needs and preferences.^[3]

Group to remember. Both approaches are "intensive insulin therapy". Both carry the highest hypoglycaemia rates of any treatment. Neither is better if the patient cannot use it safely.

CGM should be offered early in type 1 diabetes, even at diagnosis, including in children and adults on any insulin therapy.^[3] Initiation early after diagnosis has been shown to decrease A1C in young people.^[3] Real-time and intermittently scanned systems both have roles; real-time systems with alerts are generally preferred where hypoglycaemia is a concern.^[3] CGM data, including time in range and the ambulatory glucose profile, make the day-to-day management of insulin visible.

In practice:

• Initiate CGM at diagnosis wherever possible.
• Use rtCGM or isCGM in people with type 1 diabetes on insulin.
• If a device is not available, capillary blood glucose monitoring remains effective: in one large database study of almost 27,000 children and adolescents with type 1 diabetes, each additional daily check was associated with an A1C about 0.2 percentage points lower and fewer acute complications.^[3] In people on intensive insulin therapy, 6 to 10 checks a day may be needed.^[3]
• Restarting or interrupting CGM worsens outcomes, so continuity of access matters.^[3]

Carbohydrate content is the main determinant of the meal-time insulin dose, and counting (or experience-based estimation) is a key component of glycaemic management.^[5] Education covers:

• the insulin-to-carbohydrate ratio and correction factor for high glucose;
• the effect of high-fat and high-protein meals, which cause early hypoglycaemia and delayed post-prandial rises, and may need split or increased doses;^[5]
• the timing of rapid-acting insulin 15 minutes before the meal (or immediately for ultra-rapid preparations) rather than after.
• the impact of exercise, illness, stress and alcohol on glucose, and how to adjust.

Medical nutrition therapy should be individualised by an experienced dietitian, particularly in children and when coeliac disease coexists.^[5]

Hypoglycaemia is the major limiting factor in type 1 diabetes management. Recognise it early and treat it fast.

Levels and definitions:

• Level 1: glucose under 70 mg/dL (under 3.9 mmol/L) and at least 54 mg/dL (at least 3.0 mmol/L).
• Level 2: glucose under 54 mg/dL (under 3.0 mmol/L), where neuroglycopenic symptoms begin.
• Level 3: a severe event with altered mental or physical function requiring assistance from another person, irrespective of the measured glucose.^[2]

Treatment of the conscious person:

• the preferred treatment is glucose, a serving of fast-acting carbohydrate; any glucose-containing carbohydrate works, but avoid foods high in fat or protein for the initial treatment. Repeat after 15 minutes if the glucose is still under 70 mg/dL (under 3.9 mmol/L).^[2]

Treatment of the unconscious or unable-to-swallow person:

• glucagon. Glucagon should be prescribed for everyone taking insulin or at high risk for hypoglycaemia, and family, carers and school staff should know where it is kept and how to give it. Preparations that do not require reconstitution are preferred.^[2]
• give intravenous glucose where available, then oral carbohydrate once recovered.
• call emergency services. A severe or repeated Level 2 or Level 3 episode should prompt a prompt review and adjustment of the treatment plan.^[2]

Hypoglycaemic awareness may be reduced after repeated episodes, and fear of hypoglycaemia is common. Screening for impaired awareness, structured education and CGM reduce risk.^[2]

Illness raises glucose and ketones even when someone cannot eat. Practical points, adapted to local guidance:

• Never stop basal insulin completely; insulin needs often rise during illness.
• Check glucose more frequently; check blood or urine ketones when glucose is high or the person is unwell.
• If ketones are present, use extra (bolus or correction) insulin according to the agreed sick-day plan, take sugar-free fluids to avoid dehydration, and take carbohydrate-containing fluids if eating is reduced.
• Seek urgent care for vomiting, dehydration, persistent high glucose, moderate or large ketones, breathlessness, confusion or drowsiness.

The choice of insulin determines how closely one can reproduce normal secretion. The table summarises the principal classes. Concentrations, brands and availability vary by country; check the local formulary.

Two mnemonics of practice:

• Basal: glargine, detemir, degludec or NPH. Given once (or twice for NPH and detemir) to hold fasting glucose and suppress ketogenesis.
• Bolus: rapid- or short-acting with each meal.
• Biphasic premixes are convenient where intensive regimens are not feasible; they trade flexibility and can increase hypoglycaemia.

Teplizumab is a humanised anti-CD3 monoclonal antibody and the first disease-modifying treatment approved for type 1 diabetes. It is immunomodulatory rather than immunosuppressive. It is given as a single 14-day intravenous course, with the dose escalating over the first 5 days from 65 micrograms/m2 on day 1 to 1030 micrograms/m2 on days 5 to 14.^[6]

In the TrialNet teplizumab prevention study, a single 14-day course in people with stage 2 type 1 diabetes delayed the median time to stage 3 by about 24 months; longer follow-up showed a delay of about 32 months, with about half of treated participants still free of clinical diabetes after 5 years against about 22% of controls.^[6] FDA approval is for people aged 8 years or older with stage 2 disease (a 2026 approval extended this to children over 1 year).^[6] Teplizumab is not approved for stage 3 disease; trials in newly diagnosed patients show beta-cell preservation but it is not yet licensed for that use.^[6]

Practical constraints matter: live vaccines must be avoided for 8 weeks before and 12 months after treatment, inactivated vaccines for 2 weeks before and 6 weeks after; viral screening (cytomegalovirus, Epstein-Barr virus) and tests of blood count, liver function, tuberculosis, HIV and hepatitis are required before infusion. Lymphopenia is expected in around 80% of patients and cytokine release syndrome in around 5.8%. Teplizumab is not used in pregnancy or breastfeeding. Access currently depends on national and local availability, and the cost and infrastructure to deliver a 14-day intravenous course are real barriers in many settings.^[6]

Targets individualized to the patient, their phase of life, risk of hypoglycaemia and comorbidities. Shared decision-making is central.

• A1C under 7.0% (under 53 mmol/mol) is appropriate for many non-pregnant adults and for many children and adolescents; during the honeymoon phase, even lower targets may be possible without significant hypoglycaemia.^[2][5]
• A1C higher (up to about 8%) may be appropriate where life expectancy is limited, where hypoglycaemia is severe, or where harms of treatment outweigh benefits.^[2]
• Time in range (CGM, 70 to 180 mg/dL or 3.9 to 10.0 mmol/L) above 70% aligns with an A1C of roughly 7%.^[2]
• Time below range under 4% below 70 mg/dL and under 1% below 54 mg/dL; coefficient of variation under 36%.^[2]
• Targets by glucose: before meals 80 to 130 mg/dL (4.4 to 7.2 mmol/L); peak after meals under 180 mg/dL (under 10.0 mmol/L); overnight glucose within range.^[2]
• Pregnancy has its own stringent targets and is an indication for urgent specialist team care; pregnancy-specific time-in-range targets are tighter.

Two cautions from the DCCT and its follow-up:

• better glycaemic control in type 1 diabetes reduced microvascular complications by 50 to 76%, with benefits persisting for decades even after glycaemic separation disappeared.^[2]
• hypoglycaemia is a potent marker of cardiovascular and mortality risk, so targets must be safety-aware.^[2]

For people at risk of hypoglycaemia, CGM is beneficial and recommended, and glucagon must be available.^[2]

Type 1 diabetes clusters with other autoimmune diseases. Screening should be planned at diagnosis and thereafter.

• Autoimmune thyroid disease is the most common: it occurs in 17 to 30% of people with type 1 diabetes, and about 25% of children have thyroid autoantibodies at the time of diagnosis. Hyperthyroidism occurs in about 0.5%.^[4][5] Screen soon after diagnosis by measuring thyroid antibodies and thyroid-stimulating hormone; if normal, recheck every 1 to 2 years, or sooner with symptoms, goitre, growth problems or glycaemic variability. In adults, screening soon after diagnosis and thereafter if clinically indicated is recommended.^[4][5]
• Coeliac disease occurs in 1.6 to 16.4% of people with type 1 diabetes against 0.3 to 1% in the general population.^[5] Screen youth soon after diagnosis with IgA tissue transglutaminase antibodies (and total IgA), and repeat within 2 years and again after 5 years, or sooner with symptoms. In adults, screen in the presence of gastrointestinal symptoms, signs or laboratory suspicion (diarrhoea, malabsorption, abdominal pain, osteoporosis, iron or vitamin deficiencies).^[4][5] A confirmed diagnosis needs a gluten-free diet and dietetic support.
• Pernicious anaemia (vitamin B12 deficiency) is common enough to matter; check B12 in people with type 1 diabetes and peripheral neuropathy or unexplained anaemia.^[4]
• Other autoimmune diseases, including autoimmune liver disease, primary adrenal insufficiency (Addison disease), vitiligo, myasthenia gravis and polyglandular syndromes, should be assessed and monitored when clinically suggested.^[4][5]

Remember that family members of a person with type 1 diabetes are also at high risk; relatives should be offered islet autoantibody testing through appropriate programmes.^[5]

Type 1 diabetes is the commonest form of diabetes in youth, although there are more adults living with it.^[5] Special considerations include growth and puberty, childcare and school, neurological vulnerability to hypo- and hyperglycaemia, and the cognitive impact of DKA.^[5]

• Glycaemic targets: A1C under 7% for many; under 7.5% if the child cannot articulate symptoms or lacks reliable monitoring; under 8% with severe hypoglycaemia or limited life expectancy; under 6.5% can be reasonable where it is achievable without significant hypoglycaemia.^[5]
• Monitor glucose frequently, and offer CGM and AID or pumping early.^[5]
• Medical nutrition therapy includes carbohydrate counting (or experience-based estimation) and education on high-fat and high-protein meals.^[5]
• Psychosocial screening, school support, exercise plans and management of cardiovascular risk factors are all part of routine paediatric care.
• Screening for thyroid and coeliac disease follows the timelines above.

As in adults, residual beta-cell function may give a phase of reduced insulin requirement. Set goals accordingly, and beware that the honeymoon ends and insulin requirements return.

Pregnancy is a specialist area. Type 1 diabetes in pregnancy needs a multi-disciplinary team, tighter glycaemic targets, preconception counselling and contraception advice from puberty. Inactivated and live vaccines need individual planning. Refer early; this page does not attempt to cover it fully.

Both thyroid dysfunction and coeliac disease can worsen glycaemic control or increase hypoglycaemia. Sudden, unexplained changes in insulin requirements, poor growth or frequent hypoglycaemia are reasons to look again.

In many settings, continuous glucose monitors, pumps and HbA1c testing are not available. The principles of good care remain the same; the tools change.

• Insulin. Human insulin (regular and NPH) is widely available and inexpensive and remains effective. A single morning and evening mixed insulin regimen, twice daily, is workable where basal-bolus and pumps are impossible. Store insulin cool but avoid freezing; once in use, follow the expiry stored at room temperature.
• Monitoring. Capillary blood glucose monitoring, even a few checks a day, transforms safety. Urine ketone testing is cheap and valuable for sick days. HbA1c, when available, tracks long-term control.
• Education. Structured education on insulin injection, site rotation, sharps disposal, carbohydrate estimation from local foods, sick-day rules and recognition of DKA often delivers most of the benefit.
• Access and advocacy. Insulin, syringes and strips should be free or heavily subsidised. Interruptions in supply are dangerous in type 1 diabetes; communities and clinics should keep a small safety stock and plan for illness.
• Shared care. Where a specialist is not available, a trained generalist with clear protocols, remote specialist support and simple registers can deliver safe care.
• DKA triage. Fluids, electrolytes and intravenous insulin should be available even in district hospitals; mortality from DKA falls sharply when protocolised care is used.

Even without advanced technology, a well-educated patient on human insulin with regular glucose checks can achieve good outcomes.

Disease classification and autoimmune disease: type 2 diabetes mellitus, coeliac disease, hypothyroidism, Addison disease.

Insulin and therapy: insulin preparations, insulin pump therapy, carbohydrate counting, continuous glucose monitoring.

Complications: diabetic ketoacidosis, hypoglycaemia, chronic kidney disease in diabetes, diabetic retinopathy.

Paediatrics and pregnancy: type 1 diabetes in children and adolescents, diabetes and pregnancy.