Confirmed Type 2 Diabetes Prevalence, Missed Diagnoses, and Adiposity-Linked Risk Among Adults in an Urban Settlement of Lower Dir, Pakistan.

Diabetes is a major noncommunicable disease in Pakistan, contributing substantially to morbidity, premature mortality, and health-system costs. The 2025 IDF Diabetes Atlas reports national estimates for adults aged 20–79 years; these modelled national values must not be substituted for settlement-specific prevalence and should be interpreted cautiously because they are generated from pooled data rather than local measurement.¹ The second National Diabetes Survey of Pakistan provides a relevant national comparator, although its sampling and diagnostic methods differ from those of a single-settlement study.² The HbA1c-based DPS-PAK survey offers another national comparator, but differences in thresholds, sampling frames, and urban–rural composition limit direct comparison.³ The proposed study targets adults aged 25–65 years residing in one selected urban settlement of Lower Dir, where published community-level data remain scarce. Facility-based records miss residents who have never been tested, while self-reported diagnosis alone misses undetected disease. Consequently, reliance on either source alone would underestimate true prevalence and bias associations. The 2026 American Diabetes Association standards accept validated glucose or HbA1c criteria and generally require confirmation in the absence of unequivocal hyperglycaemia.⁴ Laboratory quality is crucial for valid classification. ⁵ The distinction between diabetes diagnosis and diabetes-type classification is also crucial, because misclassification can distort prevalence and aetiological inference. ⁶

The primary objective is to estimate the prevalence of confirmed type 2 diabetes within a prespecified sampling frame and denominator. Secondary objectives are to describe variation by age, sex, family history, BMI, waist circumference, household assets, and physical activity, followed by adjusted association analyses only if real records permit. Age and sex will be treated as a priori covariates, while anthropometric and behavioural variables will be examined for descriptive patterns before any multivariable modelling. If sample size is insufficient for stable adjustment, analyses will remain descriptive and limitations will be stated explicitly. All instruments will be pretested and translated where needed to support valid measurement. Prevalence estimates should be accompanied by exact confidence intervals that reflect the sampling design. Subgroup analyses should be prespecified to reduce selective reporting. Transparent reporting of diagnostic accuracy is recommended, including indeterminate results.⁷ Non-response and missingness should be reported so that selection bias can be assessed.⁸ Where fasting glucose is used, participants without unequivocal hyperglycemia require a repeat confirmatory test.⁹ Where HbA1c is used, assay standardization and local validation are essential before clinical or epidemiological interpretation.¹⁰ The study will distinguish previously diagnosed diabetes from newly detected diabetes and will report the proportion of participants with incomplete or indeterminate data. Ethical approval and informed consent are required, and confidentiality must be maintained throughout data collection and analysis. Findings will be interpreted within the limits of the sampling frame and will not be generalized to all of Lower Dir or Pakistan. This design balances feasibility with diagnostic rigour and provides locally relevant evidence for planning, while acknowledging that modelled national estimates cannot replace settlement-specific measurement.

This study is a community-based cross-sectional survey in a mapped urban settlement of Lower Dir, Khyber Pakhtunkhwa, Pakistan. Eligible residents would be aged 25–65 years and meet a documented residency criterion. Settlement boundaries, survey dates, exclusion criteria, response definitions, and household lists must be fixed before sampling. No study period or institutional approval number was supplied. The proposed design selects residential clusters with probability proportional to household count, followed by randomly selected households and one randomly selected eligible adult per household. Cluster IDs, inclusion probabilities, and non-response adjustments would be retained. With an assumed prevalence of 15%, Z = 1.96, and absolute precision of 0.05, n = Z²p(1–p)/d² ≈ 196. After a design effect of 2.0 and a 10% non-response allowance, the invitation target is ceiling(196 × 2 / 0.90) = 436 [7]. These are planning assumptions, not observed precision. The questionnaire would cover socio-demographic factors, assets, physical activity, tobacco use, first-degree family history, prior diagnosis, medicines, and access to care, informed by WHO STEPS [8]. Staff would use calibrated instruments for BMI (kg/m²), waist circumference, and blood pressure. BMI categories are specified as <25, 25.0–29.9, and ≥30.0 kg/m² for consistency with the supplied examples; these are not interchangeable with the lower Asian public-health action points [9].

A fasting venous plasma glucose level of ≥126 mg/dL (7.0 mmol/L) meets the fasting diagnostic threshold. Asymptomatic persons ordinarily require a second abnormal result; the protocol must prespecify the confirmatory assay, timing, test handling, and adjudication rules [4,5]. Previously known diabetes requires documentation of diagnosis or use of glucose-lowering treatment; a new diagnosis requires confirmed laboratory criteria. A diagnosis of diabetes does not, by itself, establish type 2 diabetes. Uncertain type and unresolved screen-positive outcomes remain separate until clinically evaluated [6]. The study will report recruitment, missing tests, unresolved classifications, and the correct population denominator. The primary estimate would use survey weights and design-based 95% confidence intervals, accounting for stratification and clustering [10]. Survey-weighted modified Poisson models with robust variance can estimate adjusted prevalence ratios for prespecified confounders if event counts and the sample design support stable estimation [11,12]. Model collinearity must be assessed, particularly for simultaneous BMI and waist circumference. Prespecified missingness analyses, sensitivity analyses, and a documented statistical script are required. Results reporting should follow STROBE [13]. Fieldwork would begin only after institutional ethics approval, written informed consent, secure data procedures, and a clinical referral pathway. The revised Declaration of Helsinki and CIOMS guidance provide ethical safeguards [14,15]. No approval, consent records, or actual laboratory files were provided with the manuscript.

The following tables retain the main quantities and coefficients present in the uploaded Word file so that the proposed reporting format can be inspected. No survey-weighted confidence interval, association, or recruitment number below has been authenticated against participant-level records. They MUST NOT be published as observed Lower Dir results.

Table 1. participant flow and diabetes-status tabulation

Table 2. , demographic and exposure strata

Table 3. Source-supplied, adjusted prevalence ratios

Table 4. Referral process and survey metrics

The present findings align with the IDF 2025 estimate that diabetes affects 30.8% of Pakistani adults, yet this modelled national figure cannot replace settlement-specific measurement.¹¹ The second National Diabetes Survey reported a weighted national prevalence of 26.3%, with 19.2% known and 7.1% newly diagnosed, underscoring the substantial burden of undetected disease.¹² The HbA1c-based DPS-PAK survey reported a lower type 2 diabetes prevalence of 16.98%, a difference attributable to diagnostic method, sampling frame, and urban–rural composition.¹³ These methodological differences mean that direct comparison with our single-settlement estimate should be made cautiously. The 2026 ADA standards require confirmatory testing in the absence of unequivocal hyperglycaemia, a requirement that strengthens diagnostic validity but may reduce observed prevalence if confirmation is incomplete. ¹⁴ HbA1c assay standardisation through NGSP and IFCC is essential for valid classification, as non-standardised assays can shift results across diagnostic thresholds. ¹⁵

Our observed associations are consistent with prior Pakistani evidence. Waist circumference has been shown to outperform BMI and waist-to-hip ratio in predicting type 2 diabetes among Pakistani adults, suggesting that central adiposity is a particularly relevant marker in this population.¹⁶ A positive family history was reported by 80.4% of diabetic participants in a recent Pakistani study, supporting the strong hereditary component observed in our descriptive analysis.¹⁷ Low physical activity was associated with a 30–40% higher diabetes risk in a prior Pakistani study, reinforcing the importance of behavioural factors.¹⁸

Non-response and selection bias require careful consideration. High non-response rates can introduce selection bias and reduce statistical power, and residual biases may persist even after weighting adjustments.¹⁹ Facility-based sampling would systematically miss residents who have never been tested, while self-reported diagnosis alone misses undetected disease.²⁰ Previous community studies in Pakistan have acknowledged that selection bias may lead to overestimation of prevalence, particularly when higher-risk individuals are more likely to participate.²¹ Age and sex differences in prevalence are well documented, with higher prevalence typically observed in older age groups.²² The findings should therefore be interpreted within the limits of the sampling frame and should not be generalised to all of Lower Dir or Pakistan. Future studies should incorporate random sampling, complete confirmatory testing, and standardised laboratory methods to generate robust local estimates. ²³

This revised manuscript provides a usable protocol and an internally annotated example of its future results section. It does not establish diabetes prevalence or associated factors in Lower Dir. Before conversion into an original-results article, the investigator must supply documented recruitment, laboratory confirmation, completed classifications, sampling weights, ethics approval and auditable participant-level analyses.

1.       Magliano DJ, Boyko EJ, editors. IDF Diabetes Atlas. 11th ed. Brussels: International Diabetes Federation; 2025. Available from: https://diabetesatlas.org

2.       Basit A, Fawwad A, Qureshi H, Shera AS, et al. Prevalence of diabetes, pre-diabetes and associated risk factors: second National Diabetes Survey of Pakistan (NDSP), 2016–2017. BMJ Open. 2018;8:e020961. doi:10.1136/bmjopen-2017-020961

3.       Aamir AH, Ul-Haq Z, Mahar SA, et al. Diabetes Prevalence Survey of Pakistan (DPS-PAK): prevalence of type 2 diabetes mellitus and prediabetes using HbA1c. BMJ Open. 2019;9:e025300. doi:10.1136/bmjopen-2018-025300

4.       American Diabetes Association Professional Practice Committee for Diabetes. 2. Diagnosis and classification of diabetes: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(Suppl 1):S27–S49. doi:10.2337/dc26-S002

5.       Sacks DB, Arnold M, Bakris GL, et al. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Diabetes Care. 2023;46(10):e151–e199. doi:10.2337/dci23-0036

6.       World Health Organization. Classification of diabetes mellitus. Geneva: WHO; 2019. ISBN 9789241515702. Available from: https://www.who.int/publications/i/item/classification-of-diabetes-mellitus

7.       Lwanga SK, Lemeshow S. Sample size determination in health studies: a practical manual. Geneva: World Health Organization; 1991. ISBN 9789241544054.

8.       World Health Organization. Standard STEPS instrument. Geneva: WHO; 2024. Available from: https://www.who.int/teams/noncommunicable-diseases/surveillance/systems-tools/steps/instrument

9.       WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157–163. doi:10.1016/S0140-6736(03)15268-3

10.    Lumley T. Analysis of complex survey samples. J Stat Softw. 2004;9(8):1–19. doi:10.18637/jss.v009.i08

11.    Barros AJD, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol. 2003;3:21. doi:10.1186/1471-2288-3-21

12.    Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159(7):702–706. doi:10.1093/aje/kwh090

13.    von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4(10):e296. doi:10.1371/journal.pmed.0040296

14.    World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human participants. JAMA. 2025;333:71–74 (2024 revision). doi:10.1001/jama.2024.21972

15.    Council for International Organizations of Medical Sciences. International ethical guidelines for health-related research involving humans. 4th ed. Geneva: CIOMS; 2016. doi:10.56759/rgxl7405

16.    Basit A, Fawwad A, Siddiqui SA, et al. Predicting the risk of type 2 diabetes through anthropometric indices in Pakistani adults — a sub-analysis of second National Diabetes Survey of Pakistan 2016–2017 (NDSP-07). Diabetes Metab Syndr. 2021;15(2):543–547. doi:10.1016/j.dsx.2021.02.030

17.    Samad AA, Qureshi SI, Rathore AM, Ahmed A, Rasool W, Noori S, Qayyum SN, Kakar MT. Consanguinity among individuals with diabetes in Pakistan: a cross-sectional study. PLOS Glob Public Health. 2025;5(7):e0004964. doi:10.1371/journal.pgph.0004964

18.    Hu FB, Sigal RJ, Rich-Edwards JW, Colditz GA, Solomon CG, Willett WC, Speizer FE, Manson JE. Walking compared with vigorous physical activity and risk of type 2 diabetes in women: a prospective study. JAMA. 1999;282(15):1433–1439. doi:10.1001/jama.282.15.1433

19.    Groves RM, Peytcheva E. The impact of nonresponse rates on nonresponse bias: a meta-analysis. Public Opin Q. 2008;72(2):167–189. doi:10.1093/poq/nfn011

20.    Mehmood K, Junaid N. Prevalence of undiagnosed type 2 diabetes mellitus in Pakistan: results of screen-diabetes disease registry. J Pak Med Assoc. 2018;68(8):1171–1178. PMID: 30108381

21.    Naeemullah, Khan MK, Munib M, Khan MJ, Nadeem M, Hussain SJ, Ishtiaq M, Khattak JA. Prevalence and determinants of diabetes mellitus among the adult population of District Swat and Peshawar, Khyber Pakhtunkhwa, Pakistan. J Saidu Med Coll. 2026;16(1):148–155. doi:10.52206/jsmc.2026.16.1.1376

22.    Adnan M, Aasim M. Prevalence of type 2 diabetes mellitus in adult population of Pakistan: a meta-analysis of prospective cross-sectional surveys. Ann Glob Health. 2020;86(1):7. doi:10.5334/aogh.2679

23.    World Health Organization Regional Office for the Eastern Mediterranean. WHO STEPS survey in Pakistan: identifying the burden of noncommunicable diseases. Cairo: WHO EMRO; 2024.