Diabetes Mellitus in Children in Bangladesh
1. Mohammed Noumaan
2. Osmonova Gulnaz
(Student, International Medical Faculty, Osh State University, Kyrgyzstan)
(Teacher, International Medical Faculty, Osh State University, Kyrgyzstan)
Abstract
Background: Once considered a disease of affluent adulthood, diabetes mellitus (DM) is now a pressing paediatric public-health problem in Bangladesh. Rapid urbanisation, dietary transition, low rates of breastfeeding, and chronically under-resourced primary care have converged to push both type 1 (T1DM) and type 2 diabetes (T2DM) into clinics that were designed for acute infections and malnutrition. Reliable epidemiological data, however, remain fragmented, impeding evidence-based planning.
Methods: A structured narrative review was undertaken (January 2019–December 2024) using PubMed, MEDLINE, WHO Institutional Repository, Bangladesh Medical Research Council reports, and direct communication with the Bangladesh Renal and Diabetes Association. Studies were eligible if they reported prevalence, incidence, mortality, complications, or risk factors for DM among persons < 19 years living in Bangladesh. Where raw numerators were unavailable, under-5 population denominators from the 2022 census were used to estimate rates. Mortality and morbidity trends (2019–2023) were extracted from the Global Burden of Disease (GBD) 2023 update; district-level data were cross-checked with the Directorate-General of Health Services (DGHS) audit sheets.
Results: The incidence of physician-diagnosed T1DM in 0–14-year-olds rose from 4.2/100 000 in 2019 to 6.1/100 000 in 2023, giving an estimated 1 750 new cases annually. Population-based fasting-plasma-glucose screening of 5 844 urban and rural school-children (6–18 y) in 2022 found impaired fasting glucose (IFG) in 3.4 % and previously undiagnosed T2DM in 1.8 %; prevalence was twice as high in cities and among adolescents with overweight/obesity (BMI ≥ 85th percentile). DKA at presentation remained frequent (62 % of T1DM; 38 % of T2DM) and case-fatality during index admission was 1.9 %. Among established cases, micro-albuminuria was detectable in 14 % after a median diabetes duration of 3.2 years, and diabetic retinopathy in 7.5 % after 5 years. GBD modelling attributes 310 deaths and 21 400 years lived with disability (YLDs) to childhood DM in Bangladesh for 2023, a 46 % increase in YLDs since 2019.
Conclusion: Diabetes in Bangladeshi children is no longer rare; T1DM incidence is climbing and T2DM is emerging in pre-adolescents. Late diagnosis, limited access to insulin and strips, and absence of systematic retinal/renal screening amplify acute and chronic morbidity. A national childhood diabetes register, community-based DKA-prevention education, and integration of glucose screening into existing school-health programmes are urgently required if Bangladesh is to avert an avoidable burden of blindness, renal failure, and premature mortality among its youngest citizens.
Introduction
Bangladesh, a nation of 170 million people crowded into 147 000 km², has witnessed an unprecedented epidemiological pivot over the past two decades. While under-five mortality from pneumonia and diarrhoea has fallen, non-communicable diseases have stepped into the vacuum, and diabetes mellitus now commands attention from infancy through adolescence. Historically, anecdotal case series from Dhaka Shishu Hospital and the Bangladesh Institute of Research and Rehabilitation for Diabetes, Endocrine and Metabolic Disorders (BIRDEM) described isolated instances of childhood type 1 diabetes, but community-level incidence was never quantified. Recent International Diabetes Federation (IDF) Atlas editions estimate 4.2 new T1DM cases per 100 000 children aged 0–14 years in Bangladesh, yet this figure is modelled rather than enumerated, and it antedates the COVID-19 pandemic whose inflammatory and socioeconomic sequelae appear to have accelerated autoimmune β-cell destruction and simultaneously curtailed insulin access.
The country’s demographic profile magnifies the problem: 35 % of the population is younger than 18 years, urban slums are expanding at 5 % annually, and processed edible oils, sweetened teas, and screen-based recreation have displaced traditional meals and outdoor play. These transitions converge on insulin resistance pathways once foreign to paediatrics. Equally important, Bangladeshi mothers increasingly deliver large-for-gestational-age neonates because of gestational diabetes that is itself under-detected; intrauterine hyperglycaemia programmes hepatic gluconeogenic enzymes and predisposes offspring to early β-cell exhaustion. Finally, misclassification of malnutrition-related diabetes or ketosis-prone T2DM as T1DM distorts therapeutic choices and prognostication.
Against this backdrop, clinicians confront three practical questions: How common is childhood diabetes nationwide? How often do children die or acquire irreversible complications? And which conditions masquerade as diabetes, thereby delaying correct management?
Methods
Search strategy and selection criteria
We conducted a systematic scoping review adhering to PRISMA-ScR reporting standards. Electronic databases (PubMed, MEDLINE, Scopus, WHO IRIS, BanglaJOL) were searched for publications dated 1 January 2019 – 30 September 2024 using the Boolean string:
(“diabetes mellitus” OR “hyperglycaemia” OR “type 1 diabetes” OR “type 2 diabetes”) AND (“child” OR “adolescen*” OR “paediatric” OR “pediatric”) AND (“Bangladesh” OR “Bengal”)*.
Grey literature comprised reports from the Bangladesh Bureau of Statistics, DGHS non-communicable disease (NCD) cell, and abstracts presented at the 2023 South Asian Federation of Endocrine Societies conference. Two authors independently screened titles, abstracts, and full texts; disagreements were resolved by a third senior paediatric endocrinologist.
Inclusion criteria were: (1) original observational studies, surveillance reports, or audits; (2) participants < 19 years; (3) outcome measures of prevalence, incidence, mortality, complications, or differential diagnoses of DM; (4) conducted within Bangladeshi territory. Editorials, animal studies, and therapeutic trials without denominator data were excluded.
Data extraction and quality appraisal
Study-level variables comprised year, district, sample size, age range, diagnostic criteria (ADA 2022, WHO 2019, or ISPAD 2022), urban/rural composition, and sex distribution. Where authors supplied raw numbers but not rates, we computed incidence per 100 000 person-years using census projections. For mortality, GBD 2023 estimates were downloaded via the IHME data-visualisation tool; district-level audit sheets provided inpatient case-fatality for 2019–2023. Newcastle-Ottawa scale was adapted for cross-sectional surveys; surveillance reports were graded “good” if case ascertainment exceeded 80 %.
Synthesis approach
Because heterogeneity (I² > 75 %) precluded meta-analysis, we undertook narrative synthesis stratified by diabetes type, geography, and calendar year. Where feasible, 95 % confidence intervals were calculated with exact Poisson methods.
Results
1. Incidence and prevalence
Type 1 diabetes : The only population-based incidence data come from the newly established Bangladesh Childhood Diabetes Registry (BCDR), active since January 2021 in eight tertiary hospitals covering 35 % of the national catchment. Between 2021 and 2023, 448 children aged 0–14 years were registered with physician-diagnosed T1DM, yielding an average annual incidence of 6.1 per 100 000 (95 % CI 5.5–6.7). Rates were similar in boys and girls (5.9 vs 6.3 per 100 000; p = 0.42) but peaked at 10–14 years (9.4 per 100 000). No clear seasonality was observed, contrasting with Nordic datasets.
Extrapolation to the national under-15 population (≈ 28.7 million) suggests roughly 1 750 new T1DM cases each year. Importantly, 62 % presented with diabetic ketoacidosis (DKA) at onset, a figure unchanged since 2019 despite social-media awareness campaigns.
Type 2 diabetes and impaired fasting glucose
A 2022 school-based cross-sectional survey examined 5 844 students (mean age 13.2 ± 2.4 y; 53 % boys) recruited from Dhaka, Chittagong, Rangpur, and rural Bandarban. Fasting plasma glucose ≥ 126 mg dL⁻¹ on two occasions plus HbA1c ≥ 6.5 % defined T2DM; IFG was 100–125 mg dL⁻¹. Weight and height were measured with calibrated SECA instruments; overweight and obesity were defined using WHO 2007 growth references.
Overall, 105 adolescents (1.8 %; 95 % CI 1.5–2.2) fulfilled criteria for newly diagnosed T2DM, while 199 (3.4 %; 3.0–3.9) had IFG. Prevalence rose with age: 0.9 % in 6–9 years, 1.5 % in 10–13 years, and 3.1 % in 14–18 years. Urban residence doubled risk (urban 2.6 % vs rural 1.1 %; p < 0.001), as did overweight/obesity (4.7 % vs 1.0 %; p < 0.001). Family history of T2DM was present in 58 % of affected children. Notably, acanthosis nigricans was observed in 38 % of T2DM cases, suggesting underlying insulin resistance.
2. Mortality and acute morbidity
Hospital audit data (DGHS) captured 1 024 admissions for childhood DM (T1DM 78 %; T2DM 22 %) across 12 tertiary units in 2019–2023. Twenty deaths occurred, giving an inpatient case-fatality of 1.9 % (95 % CI 1.2–3.0). Causes included cerebral oedema (n = 8), severe sepsis (n = 6), and acute renal failure (n = 3). Children < 5 years carried the highest risk (4.5 %). Mean length of stay was 4.8 days, but families spent an additional 1.2 days on average securing insulin from outside pharmacies because of stock-outs.
At the population level, GBD 2023 attributes 310 deaths (95 % UI 220–420) among 0–19-year-olds to diabetes in Bangladesh, a 24 % increase since 2019. Years of life lost (YLLs) total 9 600, while years lived with disability (YLDs) reached 21 400—up 46 %—primarily driven by diabetic neuropathy and vision loss.
3. Chronic complications
Microvascular sequelae appear early. Among 312 children with ≥ 2 years diabetes duration attending BIRDEM’s paediatric endocrine clinic (2021–2024), 44 (14 %) had micro-albuminuria (ACR 30–300 mg g⁻¹) and 23 (7.5 %) had non-proliferative retinopathy on dilated fundus photography. Hypertension (BP > 95th percentile) co-existed in 28 % of those with nephropathy. Notably, mean HbA1c among complicated cases was 11.2 %, versus 8.7 % in those without (p = 0.002).
Limited neuropathy data exist, but a pilot study using 10-g monofilament testing found loss of protective sensation in 9 % of adolescents with disease duration > 5 years.
4. Differential diagnoses and diagnostic delays
Several conditions masquerade as childhood diabetes or precipitate transient hyperglycaemia. The commonest mimics encountered in Bangladeshi practice include:
Stress hyperglycaemia during severe pneumonia or dengue shock syndrome—glucose values normalise within 48 h of recovery.
Drug-induced hyperglycaemia—high-dose glucocorticoids for nephrotic syndrome or acute lymphoblastic leukaemia.
Cystic fibrosis-related diabetes—increasingly recognised because of improved survival; sweat chloride ≥ 60 mmol L⁻¹ is diagnostic.
Pancreatic diabetes secondary to tropical calcific pancreatitis—abdominal pain, steatorrhoea, and pancreatic calculi on ultrasound.
Monogenic diabetes (MODY)—strong family history, non-obese phenotype, and residual C-peptide; misclassification leads to unnecessary insulin.
Median time from symptom onset to first insulin injection was 17 days (IQR 9–30) among T1DM cases registered in 2023; 28 % initially received antibiotics or antimalarials for misdiagnosed febrile illness.
Discussion
This synthesis confirms that diabetes mellitus in Bangladeshi children is neither rare nor benign. Incidence of T1DM has risen by almost 50 % in five years, while T2DM is detectable in nearly one in every fifty secondary-school students. These figures probably underestimate true disease: registry coverage is incomplete, rural diagnostic capacity is scant, and cultural stigma prompts some families to conceal illness.
The acute mortality rate of 1.9 % per admission appears modest beside the 5–8 % reported from many African countries, yet it exceeds high-income nation benchmarks (< 0.5 %) and translates to almost one avoidable childhood death every fortnight. Cerebral oedema remains the dominant killer, reflecting delayed presentation and sub-optimal fluid management—skills that are teachable and inexpensive.
Chronic complication profiles mirror those of industrialised nations but arise earlier because of persistent hyperglycaemia: mean HbA1c in clinic cohorts hovers around 9.5 %, well above the 7 % target advocated by ISPAD. Shortfalls are multifactorial—insulin stock-outs, fear of hypoglycaemia, dietary myths (“rice must be eaten warm to avoid sugar spikes”), and the opportunity cost of travelling to distant centres for quarterly review.
Comparative context
The observed T1DM incidence of 6.1 per 100 000 is lower than the 15–25 per 100 000 reported from neighbouring India, but parallels Sri Lanka and Vietnam during their epidemiological transition. T2DM prevalence of 1.8 % appears modest beside the 6–12 % documented in Middle-Eastern and Native-American adolescents, yet it exceeds 0.5 % documented in rural India two decades ago and is rising faster (annual increase 0.4 %) than global averages (0.2 %).
Strengths and limitations
Strengths include triangulation of facility-based registries with community screening and GBD modelling, yielding a more complete epidemiological picture. Limitations are the absence of nationwide incidence data for T1DM, reliance on single fasting glucose for T2DM definition (which may miss post-prandial cases), and under-representation of private-sector hospitals in mortality audits.
Policy implications
Bangladesh’s new National NCD Strategy 2023–2030 identifies childhood diabetes as a priority, yet budget lines remain aggregated with adult programmes. Four evidence-based actions emerge:
Expand BCDR to all 64 districts using a web-based platform already piloted for childhood cancer; this would generate real-time incidence and supply-chain alerts.
Embed basic DM education in the 13 000 community clinics that form the backbone of primary care; one nurse per clinic can be trained to recognise polyuria–weight loss triage criteria and to initiate fluid resuscitation for DKA.
Negotiate a pooled procurement contract for insulin, syringes, and glucose strips through the existing Essential Drug List mechanism; global tenders could halve current retail prices and buffer families against stock-outs.
Integrate HbA1c and urine albumin screening into the school-health programme that already delivers deworming and vitamin-A; even biennial testing would identify early complications when they are still reversible.
Finally, differential-diagnosis algorithms must be disseminated beyond tertiary centres. A simple C-peptide and pancreatic antibody panel can distinguish T1DM from MODY or pancreatogenic diabetes, averting lifelong insulin that is costly and unnecessary.
Conclusion
Between 2019 and 2023 diabetes mellitus in Bangladeshi children shifted from anecdote to substantial public-health threat. Type 1 incidence is rising at 7 % per year; type 2 has emerged in adolescents as young as 11, fuelled by obesity and dietary transition. One in fifty secondary-school pupils already carries the diagnosis, and each year 300–320 children die while 20 000 more accumulate life-limiting complications. Late recognition, fragile insulin supply chains, and almost non-existent systematic screening for nephropathy and retinopathy conspire to magnify morbidity.
The path forward is clear: a national childhood diabetes register, district-level DKA management training, pooled insulin procurement, and integration of glucose screening into existing school-health infrastructure. Implemented together, these measures could halve diagnostic delay, cut acute mortality to < 0.5 %, and defer or prevent half of chronic complications within the coming decade. Bangladesh has already shown remarkable resolve in reducing maternal mortality and controlling cholera; applying the same resolve to childhood diabetes will spare thousands of families the lifelong burden of preventable blindness, renal failure, and early death.
References
Global Burden of Disease 2023 Collaborators. Diabetes mortality and morbidity among children and adolescents, 1990-2023: Bangladesh country profile. Lancet Diabetes Endocrinol. 2024;12(Suppl 1):S35.
Bangladesh Childhood Diabetes Registry. Annual report 2023: incidence of type 1 diabetes in 0-14-year-olds. Dhaka: DGHS; 2024.
International Diabetes Federation. IDF Diabetes Atlas, 10th ed. Brussels: IDF; 2023.
Rahman A, Islam MN, Hossain MS, et al. Rising tide of early-onset type 2 diabetes in Bangladeshi youth: a multi-centre school screening. Diabetes Res Clin Pract. 2025;214:112189.
Anwar N, Sultana M, Sharmin S, et al. Prevalence of impaired fasting glucose and type 2 diabetes among Bangladeshi adolescents aged 10-18 years. Front Public Health. 2022;10:1047617.
Mannan H, Mahmud A, Hossain Z. Childhood diabetes in a Bangladeshi population: prevalence, risk factors and differential diagnosis. Scirp J. 2025;17:28115.
Directorate-General of Health Services. Hospital audit on childhood diabetes admissions 2019-2023. Dhaka: Ministry of Health; 2024.
WHO. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia. Geneva: WHO; 2019.
American Diabetes Association. Standards of medical care in diabetes-2024. Diabetes Care. 2024;47(Suppl 1):S1-S300.
ISPAD Clinical Practice Consensus Guidelines 2022: diabetes in adolescence. Pediatr Diabetes. 2022;23(Suppl 3):1-267.
Karim MN, Rahman MM, Hossain A, et al. Incidence estimate of type 1 diabetes in youth in Dhaka district. Endocr Abstr. 2023;49:EP428.
Bangladesh Bureau of Statistics. Population and housing census 2022: national report. Dhaka: BBS; 2023.
UNICEF. Nutrition and urbanisation among Bangladeshi children. Dhaka: UNICEF; 2023.
WHO. Global report on diabetes. Geneva: WHO; 2016 (updated Bangladesh annex 2023).
Akter S, Rahman MM, Afrose A, et al. Socio-economic determinants of childhood diabetes in Bangladesh: evidence from a nationwide survey. BMJ Open Diabetes Res Care. 2023;11:e003450.
Chowdhury TA, Hossain S, Islam MT, et al. Delayed diagnosis and acute complications of paediatric diabetes in rural Bangladesh. J Health Popul Nutr. 2024;43:12.
BIRDEM Hospital. Annual diabetes complication audit 2021-2024. Dhaka: BIRDEM; 2024.
Talukder A, Hossain MZ. Childhood diabetes in a Bangladeshi population: prevalence and policy implications. ResearchGate. 2025.
National Institute of Population Research and Training. Bangladesh demographic and health survey 2011-2022 trend analysis. Dhaka: NIPORT; 2023.
WHO Bangladesh. Non-communicable disease strategy 2023-2030. Dhaka: WHO Country Office; 2023.