Abstract
Objective
Early-onset type 2 diabetes (T2D) is an emerging public health concern. Despite its rising burden, national data on the prevalence and risk factors in Bangladesh remain scarce. This study aimed to estimate the prevalence of early-onset T2D and identify risk factors among young people in Bangladesh.
Methods
This nationwide, population-based survey encompassed 2,300 young people—adolescents (10-18 years) and young adults (19-34 years)—from rural and urban sites across 8 divisions of Bangladesh using multistage random sampling in 2024. Glycemic status (by oral glucose tolerance test except in pre-existing diabetes mellitus), fasting lipid profile, sociodemographics, lifestyle, anthropometry, and blood pressure were assessed. Glucose was measured by the glucose-oxidase method, and lipids by the enzymatic method.
Results
Prevalence of early-onset T2D and prediabetes was 4.5% (104/2,300) and 18.4% (423/2,300), respectively. Among those with diabetes, 66.3% (69/104) were newly diagnosed. Diabetes prevalence was higher in young adults (89/1,182; 7.5%) than in adolescents (15/1,118; 1.3%), and in urban residents (65/1,079; 6.0%) than in rural residents (39/1,221; 3.2%) (P < .05 for both). Higher odds for diabetes were linked with suboptimal physical activity (odds ratio [OR] 2.0, 95% CI: 1.1-4.0), smokeless tobacco use (OR 2.6, 95% CI: 1.4-4.9), central obesity (OR 2.0, 95% CI: 1.3-3.3), hypertension (OR 3.3, 95% CI: 1.6-6.7), and hypertriglyceridemia (OR 2.9, 95% CI: 1.8-4.5).
Conclusion
Early-onset T2D affects 4.5% of young people in Bangladesh, whereas 18.4% have prediabetes. Prevalence is higher in young adults and urban residents. Suboptimal physical activity, smokeless tobacco use, central obesity, hypertension, and hypertriglyceridemia are key risk factors.
Key words: early-onset type 2 diabetes, prediabetes, prevalence, risk factor, Bangladesh
Introduction
Diabetes mellitus (DM) is one of the most significant global noncommunicable diseases that impose a major public health challenge due to its chronic nature, rising prevalence, associated complications, and the need for long-term care. The impact of DM on young individuals is particularly significant because of their productive phase of life and potential role in shaping the nation’s future. The increasing worldwide prevalence of youth-onset DM, surpassing previous estimates, is therefore a matter of considerable concern.1,2 Moreover, early-onset type 2 diabetes (T2D) is characterized by an aggressive disease course, resulting from a rapid decline of beta cell secretory capacity and an increased risk of vascular complications.3,4 Given the rising incidence of early-onset T2D and its clinical phenotype, this condition is likely to impose a substantial socioeconomic burden in comparison to adult-onset DM.
Highlights
-
•
Early-onset type 2 diabetes and prediabetes are remarkably high in young Bangladeshis
-
•
7.5% of young adults have diabetes versus 1.3% of adolescents
-
•
Urban youth have higher diabetes prevalence (6.0% vs 3.2% rural)
-
•
Low physical activity, tobacco, and obesity are linked to higher diabetes odds
-
•
Hypertension and high triglycerides are key diabetes risk factors
Clinical Relevance
Early-onset type 2 diabetes affects 4.5% of the youth population in Bangladesh with a high rate of undiagnosed cases. This study identifies central obesity, hypertension, hypertriglyceridemia, and smokeless tobacco use as key independent risk factors. For the clinician, these results emphasize the necessity of early metabolic screening and aggressive lifestyle counseling for adolescents and young adults to mitigate the long-term socioeconomic and vascular complications associated with youth-onset diabetes.
It is now well recognized that South Asians are more likely to develop T2D at a younger age compared with Caucasian populations.5 In this population, diabetes often presents with distinct characteristics, including a lower body mass index (BMI), a more rapid decline in beta cell function leading to reduced insulin reserve, lower muscle mass, and increased ectopic fat deposition, particularly in the liver.6 Rapid unplanned urbanization and the adaptation of Western culture are shifting the lifestyle of the young population in South Asia. This lifestyle shift, coupled with changes in dietary habits and physical activity levels, may contribute to a higher risk of developing diabetes at an earlier age in this geographic area.7
Established research initiatives and registries have provided valuable longitudinal data on the prevalence, incidence, and complications of diabetes in young individuals.8,9 However, Bangladesh currently lacks a national registry dedicated to early-onset diabetes, and a nationally representative epidemiologic survey of its prevalence and associated risk factors has not yet been done. Prior research efforts within Bangladesh were often hospital based, limited to specific rural or urban regions, or constrained by inadequate sample sizes.10 Furthermore, many existing studies have not specifically targeted young populations with diabetes; instead, information on this age group was often derived from subset analyses, which might have diluted the statistical power necessary for robust conclusions. A comprehensive understanding of the prevalence of early-onset diabetes, prediabetes, and associated risk factors across the nation could significantly enhance awareness about the disease. This knowledge can also guide policymakers in developing targeted prevention and management strategies.11 Hence, the current study focused on obtaining a representative assessment of the prevalence of early-onset T2D in young Bangladeshi people and identifying the risk factors at individual, domestic, and community levels associated with it.
Methods
Study Design and Participants
This cross-sectional, nationwide, population-based study was conducted from March to September 2024 across the rural and urban sites of 8 administrative divisions of Bangladesh. It included young males and females aged 10 to 34 years (adolescents: 10-18 years and young adults: 19-34 years), who had been residents of the selected study sites for at least 6 months and provided written assent or consent. Participants who were pregnant, suffering from acute illness, or had severe comorbidities that could affect glucose metabolism were excluded from the study. Participants who reported current use of medications known to influence glucose metabolism were also excluded. These included systemic corticosteroids, thiazide diuretics, beta-blockers, certain antipsychotics (eg, olanzapine, clozapine), antiretroviral therapy, and immunosuppressants (eg, cyclosporine, tacrolimus). T2D was ascertained by excluding other diabetes types by clinical features, and if needed, by measuring C-peptide and autoantibodies.
Sample Size
The sample size was calculated considering an absolute precision of 1%, a two-sided confidence level of 95%, a design effect of 1.5, and a prevalence of 5%.12 Based on these assumptions, the minimum required sample size was 2,210 participants. Allowing for an anticipated nonresponse rate of approximately 10%, we targeted 2,400 participants in total, distributed equally between urban and rural sites across the 8 administrative divisions of Bangladesh (Appendix 1).
Sampling Technique
A multistage sampling design was employed, with simple random sampling at each stage (Fig. 1). At the final stage, a sampling frame was prepared by trained field workers who visited the households in the selected area, from which the required number of participants was randomly assigned (Appendix 2). If a selected individual did not respond, the next eligible individual within the same household or, if unavailable, the nearest household in the cluster was approached as a replacement to preserve representativeness.
Fig. 1.
Stages of multistage random sampling of the study.
Data Collection Procedure
Selected participants were instructed to report to a designated site on a specified day after an overnight fast. An oral glucose tolerance test (OGTT) was performed in participants without a self-reported history of diabetes, using 1.75 g/kg body weight (maximum 75 g) of anhydrous glucose for adolescents and a standard 75 g load for adults, dissolved in 250 to 300 mL of water. The World Health Organization criteria for nonpregnant adults were applied to determine glycemic status.13 In participants known to have DM, fasting and 2-hour postprandial plasma glucose were measured. Data on sociodemographic, lifestyle, and clinical variables were collected, followed by anthropometric measurements and relevant clinical examination.
Standing height was measured using an appropriate scale. A standard weighing scale was placed on a hard, flat surface and calibrated to zero before use. Participants were positioned at the center of the platform, wearing light clothing and no shoes. The grades of obesity in adolescents were defined based on BMI percentiles: underweight (<5th percentile), healthy weight (5th-84th percentile), overweight (85th-94th percentile), and obesity (≥95th percentile).14 In young adults, BMI was categorized as underweight: <18.5 kg/m2, healthy weight: 18.5 to 22.9 kg/m2, overweight: 23 to 24.9 kg/m2, and obesity: ≥25 kg/m2.15
Waist circumference (WC) was measured in the horizontal plane, midway between the lower border of the rib cage and the iliac crest, at the end of a normal expiration. WC ≥90 cm in males or ≥80 cm in females, or ≥70th percentile for age and sex in adolescents, was considered central obesity.16,17 Hip circumference was measured around the widest portion of the buttocks. The waist-to-hip ratio (WHR) was calculated as the ratio of WC to hip circumference. A WHR ≥0.9 in males or ≥0.85 in females was considered central obesity.18 The same cut-offs for WHR were used in adolescents owing to the absence of age-specific cut-offs. Waist-height ratio (WHtR) was calculated as the ratio of WC to height. WHtR ≥0.50 was considered abnormal.19
Blood pressure (BP) was measured after the participant rested in a chair for more than 5 minutes, with the arm supported on a table and the bladder empty. A well-calibrated and validated aneroid sphygmomanometer, equipped with an appropriately sized cuff, was used. Elevated systolic and diastolic BP were defined as ≥120 and ≥80 mm Hg, respectively.20 For adolescents, BP percentile was calculated using an online calculator based on age, sex, and height, with values ≥90th percentile considered as elevated BP.21,22
Physical activity was assessed using the Global Physical Activity Questionnaire version 2 that is validated in Bengali.23 The metabolic equivalent task (MET) in minutes per week was calculated, and participants were categorized into low (<600 MET min/wk), moderate (600-3,000 MET min/wk), and high (>3,000 MET min/wk) physical activity groups.24
Dietary practices, sedentary periods, screen time, sleep habits, and the use of smoked or nonsmoked tobacco were assessed using a questionnaire based on the World Health Organization stepwise approach to noncommunicable disease risk factor surveillance (STEPS) survey.12
Data Quality
A team of investigators visited each study site to train a group of field workers. On the day of the OGTT and data collection, the research team was present at the study site. The data were collected by the team of postgraduate endocrine residents, fellows, and faculty members. Senior investigators supervised each step of the fieldwork and reviewed the survey forms for any necessary corrections.
Biochemical Assay
Plasma glucose was assayed by a semiautomatic biochemical analyzer (BAS-150 TS Plus, Labomed, Inc) using the glucose-oxidase method on the same day of sample collection. For the measurement of lipid profile, blood samples were processed to separate serum at the site of sampling and transported in an ice box to be stored at –80 °C until assay. Total cholesterol (TC), triglyceride (TG), and high-density lipoprotein-cholesterol (HDL-C) were measured by the automated analyzer (Architect Plus ci8200) using the enzymatic method. Low-density lipoprotein-cholesterol (LDL-C) was calculated with the use of the Friedewald formula.25 Dyslipidemia was defined using the National Cholesterol Education Program Adult Treatment Panel III criteria, with the following cut-offs: elevated TC (adults ≥200 mg/dL, adolescents ≥170 mg/dL), elevated LDL-C (adults ≥130 mg/dL, adolescents ≥110 mg/dL), low HDL (both adults and adolescents ≤40 mg/dL), and elevated TG (≥150 mg/dL for both). Participants using lipid-lowering medication in the past 2 weeks or having high TC, TG, LDL-C, or low HDL-C were considered to have dyslipidemia.26,27
Statistical Analysis
All data were processed using IBM SPSS Statistics for Windows, Version 26 (SPSS Inc). A wealth index was calculated by applying principal component analysis on household asset data to assess socioeconomic status, which was categorized as lower, middle, and upper.28 Quantitative variables were tested for normality by the Shapiro-Wilk test and expressed as mean (±SD) if normally distributed and as median (with IQR) if skewed. They were compared between groups by unpaired Student’s t test for normally distributed data or by the Mann-Whitney U test for skewed data. Categorical variables were expressed as frequencies and percentages and were compared by the Χ2 test or Fisher’s exact test. Bivariate and multivariate analyses were employed to examine the relationship between the variables and diabetes. All explanatory variables with P < .10 in the bivariate analysis were inserted in the multivariate binary logistic regression model (except those with collinearity) to see the independent effect of each variable on the occurrence of DM. For all estimates, 95% CIs were reported. A P value of <.05 was considered statistically significant.
Ethical Consideration
Ethical approval for this study was obtained from the Institutional Review Board. Informed written consent/assent was obtained from each participant, with consent from a guardian when applicable.
Results
Characteristics of the Participants
A total of 2,340 participants participated in the study. However, 40 participants were excluded due to being in a nonfasting state (n = 5), pregnancy (n = 16), and exceeding the age limit (n = 19). Ultimately, 2,300 participants were included in the analysis.
The median age was 19 years (IQR 15-26), with nearly equal distribution between adolescents (48.6%) and young adults (51.4%), as well as between males (49.3%) and females (50.7%), and between rural (53.1%) and urban (46.9%) dwellers (Table 1). The majority had 0 to 5 years of schooling (40.9%) and were unmarried (68.1%). Most participants were students (63.2%). The distribution of education, marital status, and socioeconomic status significantly varied between male and female participants (Appendix 2, Supplementary Table 1). Regional variation was also observed in the proportion of age groups, gender, education, marital status, occupation, socioeconomic status, and religion (Appendix 2, Supplementary Table 2). The highest proportion of adolescent participants was observed in Sylhet (74.8%), whereas Barishal had the lowest (32.5%). Dhaka division had the highest proportion of affluent individuals (56.5%), whereas Barishal had the lowest (18.4%). The lifestyle and clinical characteristics of the participants are shown in Appendix 2, Supplementary Tables 3 and 4.
Table 1.
Sociodemographic Characteristics of Study Participants (n = 2,300)
| Characteristics | Value |
|---|---|
| Age [median (IQR); y] | 19 (15-26) |
| Age groups | |
| Adolescent | 1,118 (48.6) |
| Young adult | 1,182 (51.4) |
| Gender | |
| Male | 1,133 (49.3) |
| Female | 1,167 (50.7) |
| Years of schooling | |
| 0-5 | 941 (40.9) |
| 6-10 | 699 (30.4) |
| 11-15 | 660 (28.7) |
| Marital status | |
| Single | 1,572 (68.3) |
| Married | 728 (31.7) |
| Occupation | |
| Student | 1,453 (63.2) |
| Service | 178 (7.7) |
| Business | 98 (4.3) |
| Homemaker | 427 (18.6) |
| Others | 144 (6.3) |
| Religion | |
| Islam | 2,046 (89.0) |
| Other | 252 (11.0) |
| Type of community | |
| Rural | 1,221 (53.1) |
| Urban | 1,079 (46.9) |
| Division | |
| Dhaka | 184 (8.0) |
| Chattogram | 233 (10.1) |
| Rajshahi | 306 (13.3) |
| Khulna | 250 (10.9) |
| Barishal | 305 (13.3) |
| Sylhet | 337 (14.7) |
| Rangpur | 361 (15.7) |
| Mymensingh | 324 (14.1) |
Within parentheses are percentages over column total.
Prevalence of Early-Onset Type 2 Diabetes
The prevalence of early-onset T2D in these young participants was 4.5%, whereas the prevalence of prediabetes was 18.4%. Among the subtypes of prediabetes, impaired fasting glycemia was observed in 6.7%, impaired glucose tolerance in 10.0%, whereas 1.7% had both impaired fasting glycemia and impaired glucose tolerance (Fig. 2). Of the individuals with DM, only 35 of 104 (33.7%) had a prior diagnosis of DM, whereas the remaining were newly diagnosed.
Fig. 2.
Glycemic status of the study participants (n = 2,300). Percentages are over grand total. DM = diabetes mellitus; IFG = impaired fasting glycemia; IGT = impaired glucose tolerance; NGT = normal glucose tolerance.
There was a regional variation in DM prevalence, with Dhaka division having the highest prevalence (11.4%), followed by Khulna (6.8%) and Barishal (6.2%). The lowest prevalence was found in Rangpur (1.9%), Mymensingh (2.2%), and Sylhet (3.0%) (Fig. 3). The prevalence of prediabetes also followed the distribution of DM, except for a markedly high prevalence in Sylhet (33.8%) (Appendix 2, Supplementary Table 5). There was a higher prevalence of DM in young adults than adolescents (7.5% vs 1.3%, P < .001), in females than males (5.2% vs 3.8%, P < .001), and in urban residents than rural residents (6.0% vs 3.2%, P = .001) (Appendix 2, Supplementary Table 6). However, these differences were not consistent across all divisions (Appendix 2, Supplementary Table 7). DM prevalence shows a progressive increase with age, particularly after 24 years, at which point the rise becomes more pronounced. In contrast, the prevalence of prediabetes initially decreases with age but begins to increase again after 24 years (Appendix 2, Supplementary Fig. 1).
Fig. 3.
Prevalence of early-onset type 2 diabetes in different divisions of Bangladesh.
Sociodemographic Characteristics and Lifestyle Factors in DM
Young adults had significantly higher odds of DM compared with adolescents (odds ratio [OR] 6.0, 95% CI: 3.4-10.4, P < .001). Upper socioeconomic status (OR 1.5, 95% CI: 1.0-2.2, P = .047) and residing in urban areas (OR 1.9, 95% CI: 1.3-2.9, P = .001) were both associated with DM. Among lifestyle factors, low/middle physical activity (OR 2.4, 95% CI: 1.3-4.6, P = .006), the use of smokeless tobacco (OR 5.3, 95% CI: 3.0-9.4, P < .001), and having a late bedtime (after 11 pm) (OR 1.6, 95% CI: 1.1-2.5, P = .009) were significantly associated with DM. Conversely, no significant associations with DM were observed for gender, fruit and vegetable intake, the frequency of eating or snacking out, sedentary time, screen time, sleep duration, wake-up time, or smoking status (Table 2).
Table 2.
Association of Sociodemographic Characteristics and Lifestyle Factors With DM (n = 2,300)
| Characteristics | n | DM (n = 104) | Non-DM (n = 2,196) | OR(95% CI) | P valuea | |
|---|---|---|---|---|---|---|
| Age groups | Adolescent | 1,118 | 15 (1.3) | 1,103 (98.7) | Ref | |
| Young adult | 1,182 | 89 (7.5) | 1,093 (92.5) | 6.0 (3.4-10.4) | <.001 | |
| Gender | Male | 1,133 | 43 (3.8) | 1,090 (96.2) | Ref | |
| Female | 1,167 | 61 (5.2) | 1,106 (94.8) | 1.4 (0.9-2.1) | .098 | |
| Wealth index | Poor to middle | 1,533 | 60 (3.9) | 1,473 (96.1) | Ref | |
| Rich | 767 | 44 (5.7) | 723 (94.3) | 1.5 (1.0-2.2) | .047 | |
| Residence | Rural | 1,221 | 39 (3.2) | 1,182 (96.8) | Ref | |
| Urban | 1,079 | 65 (6.0) | 1,014 (94.0) | 1.9 (1.3-2.9) | .001 | |
| Fruits intake | <1 serving/d | 2,003 | 87 (4.3) | 1,916 (95.7) | Ref | |
| ≥1serving/d | 297 | 17 (5.7) | 280 (94.3) | 1.3 (0.8-2.3) | .285 | |
| Vegetables intake | <2 serving/d | 2,161 | 94 (4.3) | 2,067 (95.7) | Ref | |
| ≥2 serving/d | 139 | 10 (7.2) | 129 (92.8) | 1.7 (0.9-3.4) | .118 | |
| Meals outside home | <1/wk | 1,712 | 77 (4.5) | 1,635 (95.5) | Ref | |
| ≥1/wk | 588 | 27 (4.6) | 561 (95.4) | 1.0 (0.7-1.6) | .924 | |
| Snacks outside home | <3/wk | 1,596 | 77 (4.8) | 1,519 (95.2) | Ref | |
| ≥3/wk | 704 | 27 (3.8) | 677 (96.2) | 0.8 (0.5-1.2) | .293 | |
| Physical activity | High | 502 | 11 (2.2) | 491 (97.8) | Ref | |
| Low to middle | 1,798 | 93 (5.2) | 1,705 (94.8) | 2.4 (1.3-4.6) | .006 | |
| Sedentary period | ≤3 h/d | 1,238 | 52 (4.2) | 1,186 (95.8) | Ref | |
| >3 h/d | 1,062 | 52 (4.9) | 1,010 (95.1) | 1.2 (0.8-1.7) | .423 | |
| Screen time | ≤1 h/d | 1,187 | 58 (4.9) | 1,129 (95.1) | Ref | |
| >1 h/d | 1,113 | 46 (4.1) | 1,067 (95.9) | 0.8 (0.6-1.2) | .385 | |
| Sleep duration | ≥8 h/d | 1,638 | 67 (4.1) | 1,571 (95.9) | Ref | |
| <8 h/d | 662 | 37 (5.6) | 625 (94.4) | 1.4 (0.9-2.1) | .117 | |
| Bedtime | at/before 11 pm | 1,513 | 56 (3.7) | 1,457 (96.3) | Ref | |
| After 11 pm | 787 | 48 (6.1) | 739 (93.9) | 1.6 (1.1-2.5) | .009 | |
| Wake-up time | At/before 7 am | 1,553 | 65 (4.2) | 1,488 (95.8) | Ref | |
| After 7 am | 747 | 39 (5.2) | 708 (94.8) | 0.8-1.9 | .263 | |
| Smoking status | Nonsmokers | 2,031 | 87 (4.3) | 1,944 (95.7) | Ref | |
| Smokers | 269 | 17 (6.3) | 252 (93.7) | 1.5 (0.9-2.6) | .131 | |
| Smokeless tobacco | Nonusers | 2,205 | 87 (3.9) | 2,118 (96.1) | Ref | |
| Users | 95 | 17 (17.9) | 78 (82.1) | 5.3 (3.0-9.4) | <.001 | |
Abbreviations: DM = diabetes mellitus; OR = odds ratio.
Within parentheses percentages are over column total. Values in bold indicate statistical significance (P < .05).
By Χ2 test or by Fisher’s exact test, as applicable.
Symptoms, Medical History, and Family History in DM
The presence of classic diabetes-related symptoms, including polyuria (OR 2.5, 95% CI: 1.6-4.1, P < .001), polydipsia (OR 2.4, 95% CI: 1.5-3.8, P < .001), and unexplained weight loss (OR 2.2, 95% CI: 1.3-3.7, P = .001), was associated with DM. A prior history of hypertension (OR 8.4, 95% CI: 4.4-16.0, P < .001), dyslipidemia (OR 7.9, 95% CI: 2.8-22.3, P < .001), and thyroid disease (OR 7.9, 95% CI: 3.4-18.1, P < .001) also demonstrated markedly higher odds of DM. Although based on a small number of participants (n = 9), a history of gestational DM showed a trend toward higher DM odds (OR 5.3, 95% CI: 1.1-26.2, P = .076). A positive family history was significantly associated with DM (OR 2.1, 95% CI: 1.4-3.2, P < .001). Specifically, a maternal history of DM (OR 2.4, 95% CI: 1.5-3.9, P < .001) and a history of DM in siblings (OR 5.2, 95% CI: 2.5-11.1, P < .001) were significantly associated DM, whereas a paternal history of DM was not (OR 1.1, 95% CI: 0.6-2.2, P = .705) (Table 3).
Table 3.
Association of Symptoms, Medical History, and Family History With DM (n = 2,300)
| Characteristics | n | DM (n = 104) | Non-DM (n = 2,196) | OR (95% CI) | P valuea | |
|---|---|---|---|---|---|---|
| Polyuria | No | 2,043 | 80 (3.9) | 1,963 (96.1) | Ref | |
| Yes | 257 | 24 (9.3) | 233 (90.7) | 2.5 (1.6-4.1) | <.001 | |
| Polydipsia | No | 1,996 | 77 (3.9) | 1,919 (18.1) | Ref | |
| Yes | 304 | 27 (8.9) | 277 (91.1) | 2.4 (1.5-3.8) | <.001 | |
| Weight loss | No | 2,069 | 84 (4.1) | 1,985 (95.9) | Ref | |
| Yes | 231 | 20 (8.7) | 211 (91.3) | 2.2 (1.3-3.7) | .001 | |
| H/O HTN | No | 2,246 | 90 (4.0) | 2,156 (96.0) | Ref | |
| Yes | 54 | 14 (25.9) | 40 (74.1) | 8.4 (4.4-16.0) | <.001 | |
| H/O dyslipidemia | No | 2,281 | 99 (4.3) | 2,182 (95.7) | Ref | |
| Yes | 19 | 5 (26.3) | 14 (73.7) | 7.9 (2.8-22.3) | <.001 | |
| H/O thyroid disease | No | 2,269 | 96 (4.2) | 2,173 (95.8) | Ref | |
| Yes | 31 | 8 (25.8) | 23 (74.2) | 7.9 (3.4-18.1) | <.001 | |
| H/O GDM | No | 1,158 | 59 (5.1) | 1,099 (94.9) | Ref | |
| Yes | 9 | 2 (22.2) | 7 (77.8) | 5.3 (1.1-26.2) | .076 | |
| F/H of DM | No | 1,758 | 64 (3.6) | 1,694 (96.4) | Ref | |
| Yes | 542 | 40 (7.4) | 502 (92.6) | 2.1 (1.4-3.2) | <.001 | |
| F/H of DM in father | No | 2,053 | 94 (4.6) | 1,959 (95.4) | Ref | |
| Yes | 247 | 10 (4.0) | 237 (96.0) | 1.1 (0.6-2.2) | .705 | |
| F/H of DM in mother | No | 2,056 | 82 (4.0) | 1,974 (96.0) | Ref | |
| Yes | 244 | 22 (9.0) | 222 (91.0) | 2.4 (1.5-3.9) | <.001 | |
| F/H of DM in sibling | No | 2,252 | 95 (4.2) | 2,157 (95.8) | Ref | |
| Yes | 48 | 9 (18.8) | 39 (81.3) | 5.2 (2.5-11.1) | <.001 | |
Abbreviations: DM = diabetes mellitus; F/H = family history; GDM = gestational diabetes mellitus; H/O = history of; HTN = hypertension; OR = odds ratio.
Within parentheses percentages are over column total. Values in bold indicate statistical significance (P < .05).
By Χ2 test or by Fisher’s exact test, as applicable.
Obesity, Acanthosis Nigricans, Goiter, and BP in DM
Participants with overweight or obesity had significantly higher odds of DM compared with those with underweight or normal BMI (OR 3.3, 95% CI: 2.2-4.9, P < .001). Similarly, central obesity, defined by WC (OR 4.7, 95% CI: 3.2-7.0, P < .001), WHR (OR 4.5, 95% CI: 3.0-6.8, P < .001), and WHtR (OR 5.1, 95% CI: 3.4-7.7, P < .001), was significantly associated with DM. Elevated systolic BP (OR 1.8, 95% CI: 1.2-2.8, P = .006) and diastolic BP (OR 1.5, 95% CI: 1.0-2.3, P = .037) were both associated with DM and so was the acanthosis nigricans (OR 2.5, 95% CI: 1.5-4.3, P < .001) (Table 4).
Table 4.
Association of Obesity, Acanthosis, Goiter, Blood Pressure, and Dyslipidemia With DM (n = 2,300)
| Characteristics | n | DM (n = 104) | Non-DM (n = 2,196) | OR (95% CI) | P valuea | |
|---|---|---|---|---|---|---|
| BMI | ||||||
| Underweight/normal | 1,630 | 46 (2.8) | 1,584 (97.2) | Ref | ||
| Overweight/obese | 670 | 58 (8.7) | 612 (91.3) | 3.3 (2.2-4.9) | <.001 | |
| WC | ||||||
| Nonobese | 1,877 | 53 (2.8) | 1,824 (97.2) | Ref | ||
| Obese | 423 | 51 (12.1) | 372 (87.9) | 4.7 (3.2-7.0) | <.001 | |
| WHR | ||||||
| Nonobese | 1,581 | 36 (2.3) | 1,545 (97.7) | Ref | ||
| Obese | 719 | 68 (9.5) | 651 (90.5) | 4.5 (3.0-6.8) | <.001 | |
| WHtR | ||||||
| Nonobese | 1,659 | 37 (2.2) | 1,622 (97.8) | Ref | ||
| Obese | 641 | 67 (10.5) | 574 (89.5) | 5.1 (3.4-7.7) | <.001 | |
| Elevated SBP | ||||||
| No | 1,816 | 71 (3.9) | 1,745 (96.1) | Ref | ||
| Yes | 484 | 33 (6.8) | 451 (93.2) | 1.8 (1.2-2.8) | .006 | |
| Elevated DBP | ||||||
| No | 1,665 | 66 (4.0) | 1,599 (96.0) | Ref | ||
| Yes | 635 | 38 (6.0) | 597 (94.0) | 1.5 (1.0-2.3) | .037 | |
| Acanthosis | ||||||
| No | 2,103 | 85 (4.0) | 2,018 (96.0) | Ref | ||
| Yes | 197 | 19 (9.6) | 178 (90.4) | 2.5 (1.5-4.3) | <0.001 | |
| Dyslipidemiab | ||||||
| No | 544 | 18 (3.3) | 526 (96.7) | Ref | ||
| Yes | 1,756 | 86 (4.9) | 1,670 (95.1) | 1.5 (0.9-2.5) | 0.424 | |
| Elevated TC | ||||||
| No | 2,139 | 84 (3.9) | 2,055 (96.1) | Ref | ||
| Yes | 153 | 19 (12.4) | 134 (87.6) | 3.5 (2.0-5.9) | <0.001 | |
| Elevated LDL-C | ||||||
| No | 2,173 | 85 (3.9) | 2,088 (96.1) | Ref | ||
| Yes | 90 | 9 (10.0) | 81 (90.0) | 2.7 (1.3-5.6) | 0.011 | |
| Low HDL-C | ||||||
| No | 800 | 39 (4.9) | 761 (95.1) | Ref | ||
| Yes | 1,492 | 64 (4.3) | 1,428 (95.7) | 0.9 (0.6-1.3) | .416 | |
| Elevated TG | ||||||
| No | 1,893 | 53 (2.8) | 1,840 (97.2) | Ref | ||
| Yes | 399 | 50 (12.5) | 349 (87.5) | 5.0 (3.3-7.4) | <.001 | |
Abbreviations: BMI = body mass index; DBP = diastolic blood pressure; DM = diabetes mellitus; HDL-C = high-density lipoprotein-cholesterol; LDL-C = low-density lipoprotein-cholesterol; OR = odds ratio; SBP = systolic blood pressure; TC = total cholesterol; TG = triglyceride; WC = waist circumference; WHR = waist-to-hip ratio; WHtR = waist-height ratio.
Within parentheses percentages are over column total. Values in bold indicate statistical significance (P < .05).
By Χ2 test or by Fisher’s exact test, as applicable.
Lipid profile was not assessed in 8 participants, LDL-C could not be calculated in 29 participants.
Dyslipidemia and DM
Although dyslipidemia had no significant association with DM (OR 1.5, 95% CI: 0.9-2.5, P = .424), specific lipid abnormalities showed association. Elevated TC (OR 3.5, 95% CI: 2.0-5.9, P < .001) and LDL-C (OR 2.7, 95% CI: 1.3-5.6, P = .011) were both significantly associated with increased odds of DM. Furthermore, elevated TG showed a strong positive association with DM (OR 5.0, 95% CI: 3.3-7.4, P < .001). Conversely, low HDL-C levels were not significantly associated with DM (OR 0.9, 95% CI: 0.6-1.3, P = .416) (Table 4).
Multivariable Analyses
In a multivariate binary logistic regression, age range as young adults (OR 2.8, 95% CI: 1.5-5.2, P = .001), low to medium levels of physical activity (OR 2.0, 95% CI: 1.1-4.0, P = .036), smokeless tobacco use (OR 2.6, 95% CI: 1.4-4.9, P = .004), central obesity as defined by WHtR (OR 2.0, 95% CI: 1.3-3.3, P = .004), hypertension (OR 3.3, 95% CI: 1.6-6.7, P < .001), and hypertriglyceridemia (OR 2.9, 95% CI: 1.8-4.5, P < .001) were observed to be independent predictor of DM in young. However, other variables in the model (gender, residence, wealth index, bedtime, smoking, and family history of DM) were not significant (Fig. 4) (Appendix 2, Supplementary Table S8).
Fig. 4.
Predictors of early-onset type 2 diabetes on multivariate model (n = 2,300). Nagelkerke R square = 0.199; Hosmer and Lemeshow test P = .329. DM, diabetes mellitus; OR, odds ratio; WHtR = weight-height ratio.
Sex Stratified Results
In females, no additional variables were found to be associated with DM. However, in males, longer sedentary periods, late wake-up times, and tobacco smoking appeared to be associated with DM in addition to other factors (Appendix 2, Supplementary Tables S9-S14).
Community Stratified Results
In rural areas, females had higher odds of DM. On the other hand, in urban areas, dyslipidemia was identified as an additional factor associated with DM (Appendix 2, Supplementary Tables S15-S20).
Age Group Stratified Results
In adolescents, no variable was significantly associated with DM except for a history of dyslipidemia. On the contrary, in young adults, most of the identified factors were significantly associated, except for socioeconomic status, bedtime, and BP (Appendix 2, Supplementary Tables S21-S26).
Discussion
The current nationwide population-based survey in Bangladesh aimed to estimate the prevalence of early-onset T2D among young people and to identify its associated risk factors. Overall prevalence of T2D in young was found to be 4.5%, and a substantial proportion (18.4%) exhibited prediabetes. Alarmingly, nearly two-thirds of individuals with DM were unaware of their glycemic state. The young adults had a significantly higher prevalence of DM compared with adolescents, with an increased likelihood of DM among urban dwellers and those in the upper tertile of the wealth index. Specific lifestyle factors such as the use of smokeless tobacco and low physical activity were found to be significant predictors of DM in this cohort. In addition, obesity, hypertension, and hypertriglyceridemia were independently associated with DM. However, in subgroup analysis, a difference in risk profiles was observed, suggesting that these exposures may operate differently depending on age, sex, or other contextual factors.
Early-onset T2D is commonly defined as onset before the age of 40 years. However, in populations with high-risk ethnic backgrounds, such as South Asians, lower age cut-offs may be more appropriate.29 Accordingly, our study adopted a lower threshold to better reflect the risk profile of the Bangladeshi population. The observed DM prevalence in young Bangladeshi individuals indicates a lower prevalence in comparison to other national estimates of this country, such as 8.1% and 9.4% prevalence in adults aged 18 to 34 years, males and females, respectively, reported by the Bangladesh Demographic and Health Survey (BDHS), 2022.30 As BDHS survey enrolled the adults only with a lower cut-off for age at 18 years, the disparity with the current study is not unexpected. In a recent meta-analysis, the pooled prevalence of diabetes and prediabetes in the general population of Bangladesh was 7.8% and 10.1%, respectively.31 Here, the prevalence of diabetes in the age groups 20 to 30 and 31 to 40 was reported to be 2.8% and 6.5%, respectively, which aligns closely with our findings.
Our study also highlights regional disparities within Bangladesh, with higher prevalence observed in all urban areas compared with the rural parts of the divisions. The prevalence was highest in the Dhaka division, where the country’s capital city is located. In other divisions, there was a higher prevalence in the southwestern part than in the northeastern part of Bangladesh. A similar pattern was also observed in BDHS, 2022.30 These geographic differences likely reflect variations in urbanization, socioeconomic status, health care access, dietary patterns, and lifestyle factors across the country. These findings underscore the need for region-specific, tailored public health strategies. Disproportionately higher prevalence of prediabetes in the Sylhet division may be attributable to the inclusion of a high proportion of adolescents in the pubertal period at that particular site.
There is a remarkable proportion of undiagnosed diabetes in low- and middle-income countries worldwide.32 We also observed a higher proportion of undiagnosed DM in young people in the current study. There was also a relatively high prevalence of prediabetes in all divisions. In Bangladesh, limited access to health services, low awareness about diabetes, and cultural barriers to regular health check-ups likely contribute to these high rates. This underlines the importance of screening, early detection, and prevention of complications as well as the potential for intervention before the conversion to DM. However, universal screening is not advisable and may not be feasible. Therefore, it is necessary to conduct a risk assessment for this purpose.
Several factors have been identified that may be associated with early-onset T2D. Increasing age within the range of the current study was identified as a significant predictor, highlighting the cumulative effect of exposure to risk factors throughout an individual's development from adolescence to young adulthood. Lower physical activity was associated with higher odds of DM, reinforcing the importance of regular exercise for younger individuals. This emphasizes the creation of spaces like vast areas, playgrounds, swimming pools, parks, and well-designed educational campuses, as well as raising awareness among parents and guardians about the importance of adequate physical activity for young people. Sleep practice, like late bedtime, was also found to be a significant factor for DM in bivariate analysis. Late bedtimes have been linked in children to components of metabolic syndrome, like higher systolic BP.33 Furthermore, obesity emerged as an independent predictor, underscoring the importance of central adiposity in the pathogenesis of DM even in relatively younger individuals. Dietary practices are important determinants of obesity.34 Nevertheless, the present study found no relationship between DM and food habits, such as fruit or vegetable consumption, or taking snacks or meals outside the home, which may reflect measurement limitations, as these variables were based on self-reported, limited items. Future studies should adopt a more detailed and objective dietary assessment to explore the relationship.
Elevated TG levels also independently predicted DM, suggesting a key role of dyslipidemia in early-onset disease, which is typical of the South Asian phenotype.5,15 There was a high prevalence of low HDL-C in the participants with or without diabetes. Although an association with DM could not be appreciated, low HDL-C seems to be an important component of dyslipidemia in young Bangladeshi. The association of obesity and dyslipidemia with DM is also consistent with global findings across different age groups.35,36
The strong independent association of smokeless tobacco use is a finding that warrants further investigation and may be specific to the context of Bangladesh, where smokeless tobacco use is prevalent, particularly among certain demographic groups. Studies have reported an association between tobacco use and diabetes, but the specific impact of smokeless tobacco in young populations needs more focused research.37,38 Use of smoked tobacco was also important, especially in male participants. It highlights a potentially modifiable risk factor that has not been well investigated in the context of youth-onset diabetes.
Results of this study are consistent with findings from other regions in South Asia where the prevalence of early-onset T2D is rising at an alarming rate.39 Studies from India, the most neighboring country of Bangladesh, have similarly reported high rates of prediabetes and diabetes in youth, with socioeconomic and lifestyle factors playing a key role in their development.9 These risk factors align with global trends, where unhealthy lifestyles and changing dietary habits have been linked to the rise in youth-onset diabetes. The similarities between our findings and those of studies in neighboring countries might have been attributable to common socioeconomic and lifestyle factors in South Asia.39
This study has several strengths, including its nationwide, population-based design, which provides a representative estimate of early-onset T2D prevalence and associated risk factors in young Bangladeshis. The use of a multistage, random sampling method increases the reliability of the data. The inclusion of a variety of sociodemographic and lifestyle factors provides a comprehensive view of the determinants of DM in youth. The study employed an OGTT by venous plasma glucose measurement to determine the glycemic status. To mitigate issues related to poor reproducibility of OGTT, all personnel were rigorously trained, and strict protocols were enforced regarding patient preparation, timing of sampling, and sample handling. The portable semiautomatic biochemical analyzer was periodically cross-checked at the institute's central laboratory. However, certain limitations should be stated here. The cross-sectional design precluded the establishment of temporal relationships between the identified risk factors and the development of DM. Although the study included a large sample size, certain subgroups, such as those from remote rural areas with limited access to health care, may have been underrepresented. The study did not measure A1C and depended solely on OGTT to determine the glycemic status, which may lead to underestimation of dysglycemia. However, the use of OGTT was particularly advantageous in this context because A1C results in the target population may be unreliable due to a high prevalence of nutritional anemias and hemoglobin variants, conditions known to interfere with A1C measurement. The reliance on self-reported data for lifestyle factors, such as smoking, physical activity, and diet, could lead to reporting and recall bias. The relatively modest Nagelkerke R-squared value (0.199) in the multivariate model suggests that other unmeasured factors may also contribute to the development of DM in this population. Exclusion of maturity-onset diabetes of young by genetic testing was not possible.
Given the rising prevalence of early-onset T2D and prediabetes, there is an urgent need for national strategies to improve diabetes awareness, particularly targeting the younger population. Public health campaigns should focus on promoting healthy lifestyle changes, including reducing tobacco use, encouraging physical activity, and improving sleep hygiene. Furthermore, given the high proportion of undiagnosed DM, routine screening for prediabetes and DM should be incorporated into adolescent and young adult health check-ups, particularly for at-risk youth. Future research should explore the long-term outcomes of early-onset T2D in Bangladesh, including the progression of complications and the impact of early interventions on disease outcomes. Longitudinal follow-up studies are necessary to better understand the causal relationships between lifestyle factors and DM onset in youth, and to evaluate the effectiveness of preventative strategies over time.
Conclusion
This study provides baseline data on the prevalence of early-onset T2D and its associated risk factors in Bangladesh. A concerning prevalence of diabetes and prediabetes among young Bangladeshi individuals was observed, with a substantial undiagnosed proportion. There was a higher prevalence in young adults and urban populations. Low physical activity, tobacco use, central obesity, and elevated triglycerides were important associated factors. The findings highlight the significant burden of diabetes in the younger population and the need for targeted public health interventions promoting a healthy lifestyle and improving early detection. To mitigate the long-term health and socioeconomic consequences of this growing epidemic, there should be a concerted effort from policymakers, health care providers, and the community diabetes prevention and management among youth in Bangladesh.
Disclosure
The authors have no conflicts of interest to disclose.
Acknowledgment
The study was funded by the Non-communicable Disease Control (NCDC) Program, Directorate General of Health Services (DGHS), Mohakhali, Dhaka.
Author Contributions
M.A.H., M.H.: conceptualization. R.H.R., M.H.: data curation. M.H., M.S.M.: formal analysis. M.A.H., S.M.A., I.P.: funding acquisition. T.T., M.A.R., T.F., I.R., M.F.A., Y.A., K.K.S., R.H.R., S.A.S., K.A., A.N.B., P.C., M.H.I., S.B.I., K.A., H.Y., M.A.S., F.R., N.A., S.U.M.: investigation. M.A.H., M.H., K.K.S.: methodology. M.A.H., M.H., N.S., H.B.: project administration. M.A.H., M.H., N.A., B.H., M.S.M., N.S., H.B., T.T., M.A.R., T.F., I.R., M.F.A., Y.A.: supervision. S.M.A., I.P.: validation. M.H.: visualization. M.H.: Writing - Original draft. M.A.H., S.M.A., I.P., M.S.M., T.T., H.B., N.S., M.A.R.: Writing - Review and editing.
Ethical consideration
Ethical approval for this study was obtained from the Institutional Review Board of BMU, Dhaka (No. BSMMU/2024/3962). Informed written consent/assent was obtained from each participant, with consent from a guardian when applicable.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work the authors used ChatGPT (OpenAI) in order to improve the language, grammar, and readability of the manuscript. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Supplementary Material
References
- 1.Zhou Y., Chen Y., Tang Y., Zhang S., Zhuang Z., Ni Q. Rising tide: the growing global burden and inequalities of early-onset type 2 diabetes among youths aged 15-34 years (1990-2021) Diabetol Metab Syndr. 2025;17(1):103. doi: 10.1186/s13098-025-01673-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Xie J., Wang M., Long Z., et al. Global burden of type 2 diabetes in adolescents and young adults, 1990-2019: systematic analysis of the Global Burden of Disease Study 2019. BMJ. 2022;379 doi: 10.1136/bmj-2022-072385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.TODAY Study Group, Bjornstad P., Drews K.L., et al. Long-term complications in youth-onset type 2 diabetes. N Engl J Med. 2021;385(5):416–426. doi: 10.1056/NEJMoa2100165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lascar N., Brown J., Pattison H., Barnett A.H., Bailey C.J., Bellary S. Type 2 diabetes in adolescents and young adults. Lancet Diabetes Endocrinol. 2018;6(1):69–80. doi: 10.1016/S2213-8587(17)30186-9. [DOI] [PubMed] [Google Scholar]
- 5.Mohan V. Lessons learned from epidemiology of type 2 diabetes in South Asians: Kelly West Award Lecture 2024. Diabetes Care. 2025;48(2):153–163. doi: 10.2337/dci24-0046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shil K.K., Hasan M., Sultana N., Salam S.B., Mostafa S.N., Hasanat M.A. High prevalence of fatty liver in Bangladeshi adolescents and young adults with type 2 diabetes: key predictors and screening recommendations. Diabetes Metab Syndr Obes. 2025;18:2019–2027. doi: 10.2147/DMSO.S520806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Feldman A.L., Long G.H., Johansson I., et al. Change in lifestyle behaviors and diabetes risk: evidence from a population-based cohort study with 10 year follow-up. Int J Behav Nutr Phys Act. 2017;14(1):39. doi: 10.1186/s12966-017-0489-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lawrence J.M., Divers J., Isom S., et al. Trends in prevalence of type 1 and type 2 diabetes in children and adolescents in the US, 2001-2017. JAMA. 2021;326(8):717–727. doi: 10.1001/jama.2021.11165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Praveen P.A., Madhu S.V., Mohan V., et al. Registry of youth onset diabetes in India (YDR): rationale, recruitment, and current status. J Diabetes Sci Technol. 2016;10(5):1034–1041. doi: 10.1177/1932296816645121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rahman M.N., Alam S.S., Zobayed A., et al. Prevalence of diabetes mellitus and associated risk factors among adult individuals in selected areas of Bangladesh. Am J Public Health Res. 2020;8(6):209–213. [Google Scholar]
- 11.Hill-Briggs F., Adler N.E., Berkowitz S.A., et al. Social determinants of health and diabetes: a scientific review. Diabetes Care. 2020;44(1):258–279. doi: 10.2337/dci20-0053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.World Health Organization 2018 National STEPS Survey for Non-communicable Diseases Risk Factors in Bangladesh. https://www.who.int/bangladesh/about-us/publications/m/item/2018-national-steps-survey-for-non-communicable-diseases-risk-factors-in-bangladesh
- 13.World Health Organization . 2006. Definition and Diagnosis of Diabetes Mellitus and Intermediate Hyperglycaemia: Report of a WHO/IDF Consultation.https://www.who.int/publications/i/item/definition-and-diagnosis-of-diabetes-mellitus-and-intermediate-hyperglycaemia [Google Scholar]
- 14.Child and teen BMI categories. Centers for Disease Control and Prevention. https://www.cdc.gov/bmi/child-teen-calculator/bmi-categories.html
- 15.Misra A., Shrivastava U. Obesity and dyslipidemia in South Asians. Nutrients. 2013;5(7):2708–2733. doi: 10.3390/nu5072708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Siddiquee T., Bhowmik B., Munir S.B., et al. Optimal cut-off points for waist circumference in the definition of metabolic syndrome: a cross-sectional study in rural Bangladesh. BMJ Open. 2025;15(3) doi: 10.1136/bmjopen-2024-093159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Khadilkar A., Ekbote V., Chiplonkar S., et al. Waist circumference percentiles in 2-18 year old Indian children. J Pediatr. 2014;164(6):1358–1362.e2. doi: 10.1016/j.jpeds.2014.02.018. [DOI] [PubMed] [Google Scholar]
- 18.Sruthi K.G., John S.M., David S.M. Assessment of obesity in the Indian setting: a clinical review. Clin Epidemiol Glob Health. 2023;23 [Google Scholar]
- 19.Yoo E.G. Waist-to-height ratio as a screening tool for obesity and cardiometabolic risk. Korean J Pediatr. 2016;59(11):425–431. doi: 10.3345/kjp.2016.59.11.425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.McEvoy J.W., McCarthy C.P., Bruno R.M., et al. 2024 ESC Guidelines for the management of elevated blood pressure and hypertension. Eur Heart J. 2024;45(38):3912–4018. doi: 10.1093/eurheartj/ehae178. [DOI] [PubMed] [Google Scholar]
- 21.Flynn J.T., Kaelber D.C., Baker-Smith C.M., et al. Clinical practice guideline for screening and management of high blood pressure in children and adolescents. Pediatrics. 2017;140(3) doi: 10.1542/peds.2017-1904. [DOI] [PubMed] [Google Scholar]
- 22.Clinical calculators MSD Manuals. Professional Version. https://www.msdmanuals.com/professional/pages-with-widgets/clinical-calculators?mode=list
- 23.Bull F.C., Maslin T.S., Armstrong T. Global physical activity questionnaire (GPAQ): nine country reliability and validity study. J Phys Act Health. 2009;6(6):790–804. doi: 10.1123/jpah.6.6.790. [DOI] [PubMed] [Google Scholar]
- 24.Moniruzzaman M., Ahmed M.S., Zaman M.M. Physical activity levels and associated socio-demographic factors in Bangladeshi adults: a cross-sectional study. BMC Public Health. 2017;17(1):59. doi: 10.1186/s12889-016-4003-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Warnick G.R., Knopp R.H., Fitzpatrick V., Branson L. Estimating low-density lipoprotein cholesterol by the Friedewald equation is adequate for classifying patients on the basis of nationally recommended cutpoints. Clin Chem. 1990;36(1):15–19. [PubMed] [Google Scholar]
- 26.National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143–3421. [PubMed] [Google Scholar]
- 27.Thom T., Haase N., Rosamond W., et al. Heart disease and stroke statistics--2006 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2006;113(6):e85–e151. doi: 10.1161/CIRCULATIONAHA.105.171600. [DOI] [PubMed] [Google Scholar]
- 28.Rutstein S.O., Johnson K. The DHS wealth index. DHS Comparative Reports No. 6. Calverton, MD: ORC Macro. 2004 [Google Scholar]
- 29.Luk A., Wild S.H., Jones S., et al. Early-onset type 2 diabetes: the next major diabetes transition. Lancet. 2025;405(10497):2313–2326. doi: 10.1016/S0140-6736(25)00830-X. [DOI] [PubMed] [Google Scholar]
- 30.National Institute of Population Research and Training (NIPORT) and ICF. Bangladesh Demographic and Health Survey 2022: Final report. Dhaka, Bangladesh, and Rockville, Maryland, USA: NIPORT and ICF.
- 31.Akhtar S., Nasir J.A., Sarwar A., et al. Prevalence of diabetes and pre-diabetes in Bangladesh: a systematic review and meta-analysis. BMJ Open. 2020;10(9) doi: 10.1136/bmjopen-2019-036086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.International Diabetes Federation . 11th edn. 2025. IDF Diabetes Atlas.https://diabetesatlas.org [Google Scholar]
- 33.Mi S.J., Kelly N.R., Brychta R.J., et al. Associations of sleep patterns with metabolic syndrome indices, body composition, and energy intake in children and adolescents. Pediatr Obes. 2019;14(6) doi: 10.1111/ijpo.12507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hatami M., Taib M.N., Jamaluddin R., et al. Dietary factors as the major determinants of overweight and obesity among Iranian adolescents. A cross-sectional study. Appetite. 2014;82:194–201. doi: 10.1016/j.appet.2014.07.026. [DOI] [PubMed] [Google Scholar]
- 35.Oranika U.S., Adeola O.L., Egbuchua T.O., et al. The role of childhood obesity in early-onset type 2 diabetes mellitus: a scoping review. Cureus. 2023;15(10) doi: 10.7759/cureus.48037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Sunil B., Ashraf A.P. Dyslipidemia in pediatric type 2 diabetes mellitus. Curr Diab Rep. 2020;20(10):53. doi: 10.1007/s11892-020-01336-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Sawane H., Rajpurohit L., Sonawane S., Kharat P., Mathur A. Smokeless tobacco use and its association with type 2 diabetes: a case control study. Asian Pac J Cancer Prev. 2023;24(12):4209–4217. doi: 10.31557/APJCP.2023.24.12.4209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Manson J.E., Ajani U.A., Liu S., Nathan D.M., Hennekens C.H. A prospective study of cigarette smoking and the incidence of diabetes mellitus among US male physicians. Am J Med. 2000;109(7):538–542. doi: 10.1016/s0002-9343(00)00568-4. [DOI] [PubMed] [Google Scholar]
- 39.Nanditha A., Susairaj P., Satheesh K., Raghavan A., Snehalatha C., Ramachandran A. The rising prevalence of type 2 diabetes among the youth in southern India-an ancillary analysis of the Secular TRends in DiabEtes in India (STRiDE-I) study. J Diabetes. 2024;16(7) doi: 10.1111/1753-0407.13576. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.