Endocrinology · Medical Literature
Prevalence of undiagnosed type 2 diabetes in South Asia: A systematic review and meta-analysis.
Abstract
Although South Asia exhibits a considerable burden of type 2 diabetes mellitus (T2DM), the numbers of undiagnosed T2DM remains inadequately characterized at the regional level. This systematic review and meta-analysis aimed to approximate the pooled prevalence of undiagnosed T2DM among the general adult population and to assess country-specific variations in South Asian countries. A comprehensive literature search was directed in PubMed, Scopus, Cochrane Library, ScienceDirect, and Web of Science for studies published between 1994 and 2022. Observational studies reporting the prevalence of undiagnosed T2DM among adults aged 20-79 years in South Asia were included. By using logit transformation, Random-effects meta-analysis was conducted to assess pooled prevalence. While using Cochran's Q and I² statistics Heterogeneity was calculated, publication bias was assessed using funnel plots and Egger's test, and subgroup analyses were performed by country. In this review, a total of 28 studies containing 104,019 participants and 5719 undiagnosed T2DM cases were included. The observed pooled prevalence of undiagnosed T2DM was 6.67% (95% CI: 4.99-8.58%) with considerable uncertainty as indicated by a very wide prediction interval (0.33%-19.75%) and extreme heterogeneity (I² = 98.5%, p < 0.001). At country-level substantial variation was seen, with the highest prevalence in Sri Lanka (30.54%; 95% CI: 17.90-44.82%), followed by Nepal (7.02%; 95% CI: 4.24-10.44%), Pakistan (5.97%; 95% CI: 3.97-8.33%), Bangladesh (5.84%; 95% CI: 5.28-6.42%) and India (5.26%; 95% CI: 3.66-7.14%). The results of Egger's test (t = 1.84, p = 0.0773) suggested no statistical significance evidence of publication bias. These findings should be interpreted with attentiveness because of extensive heterogeneity and variability across studies. However, the high number of undiagnosed T2DM in general population of South Asia reflects a great burden and indicates inter-country differences. Therefore, this conclusion highlights the need for advanced surveillance systems, valid diabetes screening, and approach to health care facilities, especially in deprived and high-risk population in the region.
Clinical relevance
What is the pooled prevalence of undiagnosed type 2 diabetes mellitus among the general adult population (aged 20–79 years) in South Asia, and how does this vary by country? __SNAPSHOT__ {"studyDesign":"Systematic review and meta-analysis of observational studies; random-effects model with logit transformation","setting":"South Asian countries (country-level subgroup analyses performed); multi-database literature search (PubMed, Scopus, Cochrane Library, ScienceDirect, Web of Science); studies published 1994–2022","sampleSize":"n = 104,019 (across 28 included studies); 5,719 undiagnosed T2DM cases identified","population":"General adult population aged 20–79 years residing in South Asian countries; restricted to observational studies reporting undiagnosed T2DM prevalence","intervention":"Not applicable (prevalence estimation study; no intervention)","comparator":"Not applicable (single-arm prevalence meta-analysis; country-level subgroup analyses serve as within-study comparators)","primaryOutcome":"Pooled prevalence of undiagnosed type 2 diabetes mellitus among the general adult South Asian population","followUp":"Not reported (cross-sectional observational studies; no longitudinal follow-up described)","fundingSource":"Not reported"}
Key findings
The pooled prevalence of undiagnosed T2DM across 28 observational studies (n = 104,019 participants; 5,719 cases) was 6.67% (95% CI: 4.99–8.58%). The abstract explicitly acknowledges 'considerable uncertainty' around this estimate, consistent with the wide confidence interval spanning approximately 3.6 percentage points. Heterogeneity was assessed using Cochran's Q and I² statistics, though specific I² values and Q-test p-values are not reported in the available abstract. Country-specific variation was examined via subgroup analysis, though country-level point estimates and their confidence intervals are not reported in the available abstract. Publication bias was evaluated using funnel plots and Egger's test; results are not reported in the available abstract. No adverse events are applicable given the observational, non-interventional design. NNT/NNH are not calculable from prevalence data of this nature.
Practice implications
This meta-analysis provides the most focused regional estimate to date of undiagnosed T2DM burden specifically in South Asia, a population known for earlier onset, distinct cardiometabolic risk profiles, and potential underperformance of standard diagnostic tools such as HbA1c. A pooled prevalence of 6.67% (95% CI: 4.99–8.58%) in the general adult population—before any screening or diagnosis—signals a substantial hidden disease burden with implications for end-organ risk accrual in the absence of treatment. However, several important constraints limit clinical translation: (1) only the abstract was available for review, meaning subgroup-level data, quality-assessment results, risk-of-bias findings, and heterogeneity statistics could not be fully evaluated; (2) the wide 95% CI and the authors' own acknowledgment of 'considerable uncertainty' caution against point-estimate precision; (3) the 28 included studies span 1994–2022 and likely vary substantially in diagnostic criteria (fasting glucose, OGTT, HbA1c, or combinations thereof), sampling frames (urban vs. rural), and country representation, all of which contribute to heterogeneity; (4) the findings apply specifically to the general adult South Asian population aged 20–79 years and may not generalize to South Asian diaspora populations in high-income countries, elderly individuals, or high-risk subgroups (e.g., those with obesity or family history); (5) country-specific estimates, which would be most actionable for national health policy, are not available from the abstract alone. These findings reinforce the clinical importance of proactive screening in South Asian adults, particularly given that standard BMI thresholds may underestimate metabolic risk in this population, but they do not yet provide sufficiently precise country-level data to drive specific screening policy changes without access to the full text. __LIMITATIONS__ Only the abstract was available for review; full risk-of-bias assessments, quality scoring, I² values, Egger's test results, and country-level subgroup estimates could not be independently verified.,High anticipated heterogeneity across 28 studies spanning 1994–2022, arising from variation in diagnostic criteria (fasting glucose, OGTT, HbA1c), case definitions of 'undiagnosed,' study periods, geographic sampling (urban vs. rural), and country representation.,The wide 95% CI (4.99–8.58%) and the authors' explicit acknowledgment of 'considerable uncertainty' limit the precision of the pooled estimate for policy applications.,Observational study designs are susceptible to selection bias and non-representative sampling; population-based surveys and convenience samples may yield systematically different prevalence estimates.,The literature search was restricted to studies published through 2022; more recent epidemiological shifts, including post-COVID metabolic changes and evolving diagnostic criteria, are not captured.,Country-level variation—arguably the most actionable finding—is described as a study aim but not reported in the available abstract, limiting granular interpretation.,Restriction to adults aged 20–79 years excludes younger individuals in whom undiagnosed T2DM is increasingly recognized in South Asia.,Funding source is not reported, precluding assessment of potential conflicts of interest.,Ecological and temporal confounders (urbanization trends, changing dietary patterns, evolving healthcare access across the 28-year search window) cannot be adjusted for in a prevalence meta-analysis. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ This systematic review and meta-analysis estimates that approximately 1 in 15 South Asian adults (pooled prevalence 6.67%, 95% CI: 4.99–8.58%) have undiagnosed T2DM, representing a substantial and clinically consequential hidden disease burden. The wide confidence interval, high anticipated inter-study heterogeneity, and abstract-only data availability necessitate cautious interpretation, particularly regarding country-specific estimates. These findings reinforce the clinical imperative for systematic, low-threshold screening of South Asian adults—including at BMI thresholds lower than standard Western cutoffs—while the full publication is needed before specific policy recommendations can be evidence-grounded.