Abstract
Background
Prediabetes Risk Test (PRT) has been found valid in the Western population for screening prediabetes. However, ethnicity, race, geographical and other biological characteristics have been linked to the development of prediabetes. There is a dearth of literature on the external validity of PRT in the Indian population. So, the objective of this study was to assess the external validity of the PRT in the Indian population for screening prediabetes.
Methods
The study contained 522 participants aged between 18 and 60 years. The medical history, physical activity level and anthropometric measures were assessed. Prediabetes was diagnosed using fasting blood sugar and HbA1C levels. External validation of PRT was performed using specificity, sensitivity, positive predictive value, negative predictive value, positive likelihood ratio and negative likelihood ratio. The receiver operating curve was plotted to identify the optimum cut-off value for Indians.
Results
The study found that the sensitivity of PRT was 48.1%, specificity 95.5%, positive predictive value 66.1% and negative predictive value 90.9% for screening prediabetes in the Indian population. Receiver operating curve analysis revealed that the optimum cut-off of PRT was around 3 for Indians.
Conclusion
The results showed that PRT might not be useful in the Indian population to identify the true positives of prediabetes as it has a sensitivity of 48.1%. However, it can be helpful to identify the true negatives as the specificity is 95.5%. Further study is required to modify PRT for the Indian context to make it more appropriate for the Indian population.
Keywords: Diabetes risk, Prediction model, Centre for Disease control and prevention, External validation, Prediabetes
Introduction
Prediabetes is a significant risk factor for type 2 diabetes mellitus (T2DM), comprising 10.6% of the global population.1 Furthermore, the largest percentage of undiagnosed diabetes has been found in low and middle income nations. The prevalence of undiagnosed diabetes is 53% in India.2 This statistic emphasises the critical need to enhance the ability to diagnose persons with prediabetes and give appropriate treatment to lower the risk of progressing to T2DM. Early intervention in the prediabetes state helps to cut down the future burden of T2DM and its complications to a certain degree.3 Thereby, early detection of prediabetes is essential to begin early intervention.
Biochemical testing is the only way to confirm a diagnosis of prediabetes. World Health Organization and American Diabetes Association (ADA) are the most often used diagnostic criteria.4,5 Biochemical testing could be financially taxing in case of large sample size and low-cost setup. In such cases, screening tools can be used to determine the potential high-risk cohort that can further evaluate for biochemical parameters. Screening tools have gained popularity in epidemiological studies as they are cost-effective. In a community, prediabetes risk is measured using a risk questionnaire, anthropometric and body composition measurements.6,7 Among all, using a prediction model is the most accessible form of a screening tool, as it does not require any modalities or trained instructors. ‘Indian diabetes risk score’, ‘Finnish diabetes risk score’ and ‘Diabetes risk questionnaire by ADA’ are a few risk prediction models used to screen the risk of diabetes.7, 8, 9 Most of the time, prediction models are rarely used because of their unknown validity.10 Lack of external validation was one of the key reasons for poor screening for the risk of diabetes.11 Literature has suggested that the external validation and recalibration of the existing predictor tool are always the better option than developing a new model.11
The Centre for Disease Control and Prevention (CDC) and the ADA administer the Prediabetes Risk Test (PRT) to assess prediabetes risk (https://www.cdc.gov/prediabetes/pdf/Prediabetes-Risk-Test-Final.pdf).12 It was developed based on a questionnaire published in 1995 and validated in 2008 on the American population.13,14 PRT is a simple and cost-effective tool to screen for prediabetes. However, it has developed and validated on the American population. Ethnicity, race, geographical and other biological characteristics have been linked to the development of prediabetes and T2DM.11,15 There is a dearth of literature on the external validity of PRT in the Indian population. So, the main objective of this study was to assess the external validity of the PRT in the Indian population for screening prediabetes. Furthermore, we intended to find out the optimum cut-off level of the tool for screening prediabetes in the Indian scenario.
Materials and methods
Study design
This study was conducted and reported as per the guidelines of ‘Standard of Reporting Diagnostic accuracy studies’. The study was conducted after getting permission from Institutional Research Committee and Institutional Ethics Committee. A written informed consent was taken from all patients/participants for inclusion in the study. Before directing this study, an email confirmation was received from CDC that it does not have a reservation about doing the validation study of PRT. All guidelines as per the Declaration of Helsinki and good clinical practice guidelines were followed.
Participants
The sample size was calculated using the following formula:
| nSe = Z21 − α/2 Se (1−Se)/d2 × Prevalence |
where anticipated sensitivity (Se) = 70%, prevalence = 14%, alpha = 5% (Z table value = 1.96), precision (d) = 12%. The minimum required sample size was ∼396. Six hundred twenty voluntary participants of either gender aged 18–60 years were screened to test out the study's inclusion criteria. The convenience sampling method was used to recruit participants. Known cases of T2DM were excluded after screening. Also, the individuals on any form of oral hypoglycaemics were excluded. Pregnant women and individuals with chronic neurological, cardiovascular or musculoskeletal injuries that impact regular physical activity were also excluded. After obtaining written consent for data following Helsinki guidelines, 522 individuals were recruited.
Study setting
The data were collected from the community screening and the medicine Out Patient Department (OPD) of tertiary care hospitals in South India from November 2019 to May 2021.
Test method
Demographic assessment
After reporting the age and gender, the participant's occupation was recorded. Physical activity was recorded using the Global Physical Activity Questionnaire. Participants were considered physically active if they were completing 600 Met. Min. of activity per week; else, they were considered inactive. The history of hypertension (HTN) and the family history of T2DM were also noted down. Female participants recorded a history of gestational diabetes mellitus (GDM).
Anthropometric assessment
It included the assessment of height, weight, body mass index (BMI) and waist circumference. Height was recorded using a wall-mounted stadiometer or measuring tape. Weight was measured using a standard weighing machine (Essae DS-415). Waist circumference was recorded at an umbilical level using a nonstretchable measuring tape.
Biochemical assessment
Participants were asked to perform biochemical tests to confirm their glycaemic status; fasting blood sugar (FBS) and glycated haemoglobin (HbA1C). FBS was conducted in the biochemistry laboratory using venous blood after 8–10 h of fasting. Valid HbA1C tests from the previous 3 months were used to verify the results.16 The ADA criteria have been used to confirm the prediabetes diagnosis. When the HbA1C was between 5.7% and 6.4% and the FBS was between 100 and 125 mg/dL, participants were deemed to be in the prediabetes state.4
Assessment of Prediabetes Risk Test
The Prediabetes Risk Test was updated in 2018 for prediabetes screening.17 The PRT contains seven questions. Each question carries a score ranging from zero to three. Ten points are the highest possible score. A score of 5 points or more is considered an increased risk of prediabetes.
The questions are based on age, gender, weight category, known HTN condition, history of GDM and family history of T2DM. Participants under the age of 40 years were given a score of zero. Between the ages of 40 and 49 years, one point was given; between the ages of 50 and 59 years, two points were given. Participants above the age of 60 years were given three points. Male participants received one point, whereas females received zero. Each of the following criteria carried one point: a history of GDM, a family history of T2DM, known HTN and physical inactivity (PA > 600 Met. Min. per week). Participants in the weight category were given a score ranging from 0 to 3 points based on their weight and height12 (Supplementary file).
Statistical analysis
Statistical analysis was performed using EZR software. Demographic characteristics of participants were provided using descriptive statistics, mean and standard deviation. The PRT was validated using specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV) and positive and negative likelihood ratios. Furthermore, we analysed its performance at a different cut-off level using specificity, sensitivity, PPV and NPV. The receiver operating curve (ROC) and area under the curve (AUC) were used to determine the PRT's ideal cut-off value for the Indian population.
Results
The participant demographic information is presented in Table 1. A total of 522 participants in total, 213 men and 309 women, made up this study. Fig. 1 represents the flow of participants with prediabetes and PRT-positive results. Fifty-nine participants received a PRT score of five or higher. The exact number of participants diagnosed with prediabetes was 81.
Table 1.
Demographic details of participants.
| N = 522 M = 213 (40.8%) F = 309 (59.2%) |
Mean ± SD |
|---|---|
| Age (years) | 39.3 ± 10.3 |
| Weight (kg) | 63.2 ± 12.4 |
| BMI (kg/m2) | 24.6 ± 4.2 |
| FBS (mg/dL) | 93.8 ± 15.5 |
| HbA1C (%) | 5.3 ± 0.4 |
| Waist circumference (cm) | 86.7 ± 9.3 |
BMI, body mass index; FBS, fasting blood sugar; HbA1C, glycated haemoglobin; SD, standard deviation.
Fig. 1.
STARD diagram to present the flow of participants in the study.
Table 2 indicates the prevalence of different risk factors in percentage among participants. Based on a Global Physical Activity Questionnaire score of <600 Met. Min per week, the study found that the prevalence of physical inactivity was highest, followed by a family history of T2DM, HTN and GDM.
Table 2.
Prevalence of prediabetes risk factors in participants.
| Risk factor | N = 522 M = 213 F = 309 |
Prevalence |
|---|---|---|
| Family history | M | 26.2% |
| F | 33.6% | |
| GDM | M | – |
| F | 2.9% | |
| HTN | M | 16.4% |
| F | 10.4% | |
| Physical inactivity | M | 40.4% |
| F | 41.7% |
GDM, gestational diabetes mellitus; HTN, hypertension.
Table 3 indicates the performance of the PRT in the Indian population for screening prediabetes. It showed a sensitivity of 48.1%, specificity 95.5%, PPV 66.1% and NPV 90.9%. The positive likelihood ratio was 10.6%, and the negative likelihood ratio was 0.5%.
Table 3.
Performance of prediabetes risk test in screening prediabetes Indians population.
| Validation parameters | Estimation | 95% CIs |
|---|---|---|
| True prevalence | 0.155 | 0.125, 0.189 |
| Sensitivity | 0.481 | 0.369, 0.595 |
| Specificity | 0.955 | 0.931, 0.972 |
| Positive predictive value | 0.661 | 0.526, 0.779 |
| Negative predictive value | 0.909 | 0.879, 0.934 |
| Likelihood ratio of a positive test | 10.617 | 6.542, 17.229 |
| Likelihood ratio of a negative test | 0.543 | 0.44, 0.671 |
CI, confidence interval.
ROC analysis reveals that the optimum cut-off of PRT is around 3, with AUC 0.815 (95% confidence interval: 0.764–0.867) for Indians (Fig. 2).
Fig. 2.
Optimum cut-off level of Prediabetes Risk Test for Indians.
Table 4 indicates the performance of PRT at different cut-off levels.
Table 4.
Specificity and sensitivity for different cut-off levels of PRT.
| Cut-off points | ≥2 |
≥3 |
≥4 |
≥5 |
||||
|---|---|---|---|---|---|---|---|---|
| Estimation | 95% CI | Estimation | 95% CI | Estimation | 95% CI | Estimation | 95% CI | |
| True prevalence | 0.16 | 0.13, 0.19 | 0.16 | 0.13, 0.19 | 0.16 | 0.13, 0.19 | 0.16 | 0.13, 0.19 |
| Sensitivity | 0.95 | 0.88, 0.99 | 0.82 | 0.71, 0.89 | 0.62 | 0.50, 0.72 | 0.48 | 0.37, 0.60 |
| Specificity | 0.37 | 0.33, 0.42 | 0.63 | 0.59, 0.68 | 0.83 | 0.79, 0.86 | 0.96 | 0.93, 0.97 |
| PPV | 0.22 | 0.18, 0.26 | 0.29 | 0.23, 0.35 | 0.39 | 0.31, 0.48 | 0.66 | 0.53, 0.78 |
| NPV | 0.98 | 0.94, 0.99 | 0.95 | 0.92, 0.97 | 0.92 | 0.89, 0.95 | 0.91 | 0.88, 0.93 |
| LR +ve | 1.51 | 1.39, 1.65 | 2.22 | 1.89, 2.60 | 3.54 | 2.71, 4.61 | 10.62 | 6.54, 17.23 |
| LR –ve | 0.13 | 0.05, 0.35 | 0.29 | 0.18, 0.47 | 0.46 | 0.35, 0.61 | 0.54 | 0.44, 0.67 |
CI, confidence interval; LR +ve, Likelihood Ratio of Positive Test; LR -ve, Likelihood Ratio of Negative Test; NPV, negative predictive value; PPV, positive predictive value.
Discussion
This cross-sectional study intended to externally validate the PRT in the Indian population for screening prediabetes. Five hundred twenty-two participants aged 18–60 years were included. For the Indian population, PRT showed a sensitivity of 48.1%, specificity of 95.5%, PPV of 66.1% and NPV of 90.9% for identifying prediabetes. According to ROC analysis, the ideal PRT cut-off for Indians was around 3.
The CDC used the ADA questionnaire to construct PRT. The ADA questionnaire has suggested compatibility for the diagnosis of both prediabetes and T2DM. It was found valid for screening prediabetes with 73% sensitivity and 57% specificity.14 For a good test, sensitivity and specificity are equally important factors. However, tests can be more sensitive and less specific or more specific and less sensitive. PRT is a screening tool and does not diagnose prediabetes. Literature indicates that it may not be feasible to use the screening test with less specificity as many people without a disease will screen positive and receive unnecessary diagnostic procedures for further confirmation.18
The previous study on Caucasians stated a tool's sensitivity was 73%.14 Different ethnicity, cultural variation, geographical and other biological characteristics have been linked to the development of prediabetes.11,15 Factors impacting prediabetes might affect differently in Indians vis-à-vis Caucasians. Indians are highly susceptible to developing prediabetes and diabetes because of biological and cultural variation.19,20 Because of low birth weight, Indians have high insulin levels at the time of birth itself compared with Caucasians.21
In addition, Indians have higher adiposity and lower lean mass than Caucasians, which increases insulin resistance. Hence, Indians are more prone to developing diabetes/prediabetes at a young age than Caucasians.22 Diabetes at a young age is defined as diabetes at age <35 years.23,24 The onset of diabetes/prediabetes at a young age is associated with lower beta cell function and lower C-peptide levels. Also, family history is a significant factor.25 Strong genetic factor influences the trait related to diabetic phenotype. Impaired glucose tolerance was higher among the Indian offspring of both diabetic parents than among Caucasians.26 In PRT, the prediabetes risk is calculated for a minimum age of 40 years, ignoring the above aspect. Indians are prone to developing diabetes/prediabetes at a lower BMI than Caucasians.22 In PRT, the prediabetes risk is identified at a minimum BMI of 25 kg/m2, whereas the literature indicated the 23 kg/m2 as a cut-off for risk of diabetes/prediabetes in Indians.27 More than total body weight, abdominal fat plays a significant role in diabetes risk. PRT does not capture the abdominal fat. Indians have more visceral fat levels than Caucasians,28 increasing the risk of prediabetes.
The above aspects might have resulted in lower sensitivity of PRT in Indians. Probability of prediabetes increases when the risk score is 5 or more on PRT. Cut-offs specified in PRT might be more appropriate for Caucasians and need modifications for Indians. In the present study, ROC analysis reveals that the optimum cut-off of PRT is around 3 for Indians. Test sensitivity increased to 82% when the cut-off value was considered at 3.
Specificity indicates the ability of a screening test to detect true negative results. NPV indicates that people with negative screening test results do not have the condition.29 In our study, specificity and NPV are comparatively higher than sensitivity and PPV for a given cut-off value. PRT might not be useful to identify true positive prediabetes, as it has 48.1% sensitivity. However, it can be helpful to identify the true negatives, as the specificity is 95.5%. It can identify the group who are unlikely to develop prediabetes. Hence, the larger population group can be excluded from further investigations to reduce the burden of biochemical testing to confirm the diagnosis of prediabetes.
The prevalence of different risk factors indicated the highest prevalence of physical inactivity, whereas GDM was the least. In India, the rapid epidemiological transition has accelerated urbanisation. Socioeconomic development has influenced the lifestyle that increased sedentary duration and physical inactivity. The previous study to find out India's physical activity and inactivity pattern stated that the prevalence of physical inactivity was 54.4%.30 Physical inactivity has a significant implication for increasing the prevalence of noncommunicable diseases and is a fourth leading cause of mortality worldwide. It is a warning sign that community members need to be more conscious of the importance of promoting physical activity.
Strength and limitations
This was the first study conducted to assess the external validity of the PRT on the Indian population. This study identified the optimum cut-off value of a tool for an Indian setting as 3. According to ADA recommendations, the diagnosis of prediabetes is made using the FBS, HbA1C and oral glucose tolerance values. In this study, we did not consider the value of oral glucose tolerance for diagnosing prediabetes. In the future, the study can be replicated on a robust sample size using all criteria of prediabetes diagnosis to confirm the results.
Conclusion
The study found that the sensitivity of PRT was 48.1%, specificity 95.5%, PPV 66.1% and NPV 90.9% for screening prediabetes in the Indian population. PRT might not be useful in the Indian population to identify the true positives of prediabetes. However, it can be helpful to identify the true negatives. To make PRT more suitable for Indians, we need further research to modify the tool in the Indian scenario.
Patients/ Guardians/ Participants consent
Patients informed consent was obtained.
Ethical clearance
Institute/hospital ethical clearance certificate was obtained.
Source of support
Nil.
Disclosure of competing interest
The authors have none to declare.
Acknowledgements
The authors would like to acknowledge ‘Center for Diabetic Foot Care and Research’ for providing support to this study. The authors are also thankful to Professor S Sudhindra, Academic President, MAHE, Dubai for his advise on statistical interpretations.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.mjafi.2022.11.001.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Sun H, Saeedi P, Karuranga S, et al. IDF diabetes Atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. Published online 2021:109119. 10.1016/j.diabres.2021.109119. [DOI] [PMC free article] [PubMed]
- 2.International Diabetes Federation IDF Diabetes Atlas 10th ed. https://diabetesatlas.org/idfawp/resource-files/2021/07/IDF_Atlas_10th_Edition_2021.pdf Published online 2021.
- 3.Lindström J., Peltonen M., Eriksson J.G., et al. Improved lifestyle and decreased diabetes risk over 13 years: long-term follow-up of the randomised Finnish Diabetes Prevention Study (DPS) Diabetologia. 2013;56(2):284–293. doi: 10.1007/s00125-012-2752-5. [DOI] [PubMed] [Google Scholar]
- 4.ADA - American Diabetes Association. American Diabetes Association: diagnosis and classification of diabetes mellitus. Diabetes Care. Published online 2014 37: S81-S90. [DOI] [PubMed]
- 5.WHO (World Health Organization) 2005. Definition and Diagnosis of Diabetes Mellitus and Intermediate Hyperglycemia.http://whqlibdoc.who.int/publications/2006/9241594934_eng.pdf [Google Scholar]
- 6.Vineetha K., Nayak Ramdas, Nayak K.R., Vidyasagar S., Kamath A. Body composition analysis, anthropometric indices and lipid profile markers as predictors for prediabetes. PLoS One. 2018;13(8):1–14. doi: 10.1371/journal.pone.0200775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mohan V., Deepa R., Deepa M., Somannavar S., Datta M. A simplified Indian Diabetes Risk Score for screening for undiagnosed diabetic subjects. J Assoc Phys India. 2005;53(Sept):759–763. [PubMed] [Google Scholar]
- 8.Diabetes Association American Standards of medical care in. J Clin Appl Res Educ. 2018;33(supplment_1):1–24. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2797382&tool=pmcentrez&rendertype=abstract [Google Scholar]
- 9.Lindström J., Tuomilehto J. The diabetes risk score. Diabetes Care. 2003;26(3):725–731. doi: 10.2337/diacare.26.3.725. [DOI] [PubMed] [Google Scholar]
- 10.Noble D., Mathur R., Dent T., Meads C., Greenhalgh T. Risk models and scores for type 2 diabetes: systematic review. BMJ. 2011;343(7836):1243. doi: 10.1136/bmj.d7163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Asgari S., Khalili D., Hosseinpanah F., Hadaegh F. Prediction models for type 2 diabetes risk in the general population: a systematic review of observational studies. Int J Endocrinol Metabol. 2021;19(3) doi: 10.5812/IJEM.109206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Center for Disease Control and Prevention, American Diabetes Association Prediabetes Risk Test. https://www.cdc.gov/diabetes/prevention/pdf/Prediabetes-Risk-Test-Final.pdf Published 2018.
- 13.Herman W.H., Smith P.J., Thompson T.J., Engelgau M.M., Aubert R.E. A new and simple questionnaire to identify people at increased risk for undiagnosed diabetes. Diabetes Care. 1995;18(3):382–387. doi: 10.2337/diacare.18.3.382. [DOI] [PubMed] [Google Scholar]
- 14.Bang H. Development and validation of a patient self-assessment score for diabetes risk. Ann Intern Med. 2009;151(11):775. doi: 10.7326/0003-4819-151-11-200912010-00005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rosella L.C., Mustard C.A., Stukel T.A., et al. The role of ethnicity in predicting diabetes risk at the population level. Ethn Health. 2012;17(4):419–437. doi: 10.1080/13557858.2012.654765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sherwani S.I., Khan H.A., Ekhzaimy A., Masood A., Sakharkar M.K. Significance of HbA1c test in diagnosis and prognosis of diabetic patients. Biomark Insights. 2016;11:95–104. doi: 10.4137/Bmi.s38440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Centers for Disease Control and Prevention Prediabetes-Risk-Tests. https://nationaldppcsc.cdc.gov/s/article/Prediabetes-Risk-Tests-1525314769238 Published 2021.
- 18.Maxim L.D., Niebo R., Utell M.J. Screening tests: a review with examples. Inhal Toxicol. 2014;26(13):811–828. doi: 10.3109/08958378.2014.955932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mohan V. Why are Indians more prone to diabetes? J Assoc Phys India. 2004;52:468–474. Published online. [PubMed] [Google Scholar]
- 20.Wells JCK, Pomeroy E, Walimbe SR, Popkin BM, Yajnik CS. The elevated susceptibility to diabetes in India: an evolutionary perspective. Front Public Health. Published online 2016. doi:10.3389/fpubh.2016.00145 [DOI] [PMC free article] [PubMed]
- 21.Yajnik C.S., Lubree H.G., Rege S.S., et al. Adiposity and hyperinsulinemia in Indians are present at birth. J Clin Endocrinol Metab. 2002;87(12):5575–5580. doi: 10.1210/jc.2002-020434. [DOI] [PubMed] [Google Scholar]
- 22.Yajnik CS, Ganpule-Rao A V. The obesity-diabetes association: what is different in Indians? Int J Low Extrem Wounds. Published online 2010. doi:10.1177/1534734610380028 [DOI] [PubMed]
- 23.Sahoo S., Zaidi G., Vipin V., et al. Heterogeneity in the aetiology of diabetes mellitus in young adults: a prospective study from north India. Indian J Med Res. 2019;149(4):479–488. doi: 10.4103/ijmr.IJMR_1004_17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Thanabalasingham G., Pal A., Selwood M.P., et al. Systematic assessment of etiology in adults with a clinical diagnosis of young-onset type 2 diabetes is a successful strategy for identifying maturity-onset diabetes of the young. Diabetes Care. 2012;35(6):1206–1212. doi: 10.2337/dc11-1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Siddiqui M.K., Anjana R.M., Dawed A.Y., et al. Young-onset diabetes in Asian Indians is associated with lower measured and genetically determined beta cell function. Diabetologia. 2022;65(6):973–983. doi: 10.1007/s00125-022-05671-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Viswanathan M., Mohan V., Snehalatha C., Ramachandran A. High prevalence of Type 2 (non-insulin-dependent) diabetes among the offspring of conjugal Type 2 diabetic parents in India. Diabetologia. 1985;28(12):907–910. doi: 10.1007/BF00703134. [DOI] [PubMed] [Google Scholar]
- 27.Snehalatha C., Viswanathan V., Ramachandran A. Cutoff values for normal anthropometric variables in Asian Indian adults. Diabetes Care. 2003;26(6):1380–1384. doi: 10.2337/diacare.26.5.1380. [DOI] [PubMed] [Google Scholar]
- 28.Williams R., Periasamy M. Genetic and environmental factors contributing to visceral adiposity in asian populations. Endocrinol Metab. 2021;35(4):681–695. doi: 10.3803/ENM.2020.772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Swift A., Heale R., Twycross A. What are sensitivity and specificity? Evid Base Nurs. 2020;23(1) doi: 10.1136/ebnurs-2019-103225. [DOI] [PubMed] [Google Scholar]
- 30.Anjana R.M., Pradeepa R., Das A.K., et al. Physical activity and inactivity patterns in India - results from the ICMR-INDIAB study (Phase-1) [ICMR-INDIAB-5] Int J Behav Nutr Phys Activ. 2014;11(1):1–11. doi: 10.1186/1479-5868-11-26. [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.