Diabetes Mellitus is a major lifestyle disorder and chronic metabolic disorder constituting a fast-growing global problem with huge social, health, and economic consequences. The prevalence of Diabetes Mellitus, chiefly Type 2 Diabetes Mellitus (T2DM), is a global public health threat. The prevalence among adults aged 20–79 years is expected to rise from 382 million in 2013 to 592 million by 2035. Developing countries bear the highest burden since more than 80% of cases occur there [1–3]. Asians have a strong ethnic and genetic predisposition for diabetes and lower thresholds for environmental risk factors [4]. The "Asian Indian Phenotype"—characterized by increased insulin resistance and greater abdominal adiposity despite lower body mass index (BMI)—makes Asian Indians more prone to Diabetes Mellitus [5]. India comprises 85% of the adult population of South East Asia and leads the world with the largest number of diabetic subjects, earning the distinction of "Diabetes Capital of the World." According to the Diabetes Atlas 2006, the number of people with Diabetes Mellitus in India—around 40.9 million—is expected to rise to 69.9 million by 2025 unless urgent preventive steps are taken [2,6,7]. Rural India is undergoing rapid socioeconomic transition. A three-fold rise in diabetes prevalence has been documented in rural southern India, rising from 2.4% to 6.4% over 14 years [8]. Recent rural studies reported prevalence of 5.8% [9], 8.32% [10], and 13.2% [11], indicating that rural prevalence now equals urban rates. This is important because most of the population in developing countries lives in rural areas [8]. Diabetes Mellitus is considered a "silent disease," exhibiting no symptoms until severe target-organ damage occurs. Early identification of at-risk individuals and appropriate lifestyle intervention would help prevent or postpone onset [12]. A significant proportion of people with T2DM present with subclinical complications at diagnosis; early diagnosis offers the chance to curb progression [13]. With this background, this study was undertaken to determine the prevalence of T2DM and associated high-risk factors among adults in the rural community of Pondicherry. Aim and Objectives Aim To study the prevalence of Type 2 Diabetes Mellitus and its associated risk factors among adults in the rural community of Pondicherry. Objectives 1. To find the prevalence of known Type 2 Diabetes Mellitus in rural Pondicherry. 2. To estimate the risk factors for diabetes among the population. 3. To compare risk factors for diabetes among known diabetic and non-diabetic individuals. 4. To find the prevalence of co-morbid conditions with diabetes.

Study Design and Study Period A cross-sectional, descriptive, community-based survey was undertaken in Pondicherry with the above-mentioned objectives among the rural community. A cross-sectional design was chosen because it permits estimation of disease prevalence and simultaneous assessment of multiple exposures in a representative population at a single point in time, at relatively low cost and with rapid turnaround. The study was approved by the Institutional Ethical Committee (Human Studies) of the institute prior to commencement. The study period was from April 2014 to September 2014 (six months), a period chosen to avoid the monsoon months when household access in rural villages is difficult and attrition is high. Study Setting Pondicherry is a Union Territory of India located in the eastern part of South India. Geographically and culturally it resembles the neighboring state of Tamil Nadu. The Pondicherry Union Territory (total population 1,244,464) consists of four districts: Pondicherry District (946,600), Karaikal (200,314), Mahe (41,934), and Yanam (55,616). In 2011, the population of Pondicherry District was 946,600 (males 466,143; females 480,457), comprising 76.06% of the total Union Territory population. The density of Pondicherry District was 3,231 persons per square kilometer, and the decadal growth was 28.73%.[14] The crude birth rate was 16.7 per 1,000 population, the crude death rate was 7.4 per 1,000 population, and the infant mortality rate was 22 per 1,000 live births (2011 census data). The elderly (≥60 years) constituted 9.8% of the total population, and life expectancy was 70 years for women and 67 years for men. Total literates in Pondicherry District numbered 733,075 (86.13%): males 380,946 (92.07%) and females 345,703 (80.40%). As per the 2011 census, 30.84% of the Pondicherry District population (293,080; males 144,769 and females 148,311) lives in rural villages.[14] The district's health infrastructure includes 2 Government Hospitals, 3 Community Health Centers, 27 Primary Health Centers (PHCs), 2 Government Medical Colleges, 7 Private Medical Colleges, and 12 Private Nursing Homes. Both communicable diseases (acute respiratory infections, acute diarrheal diseases, tuberculosis, malaria, filariasis, enteric fever, dengue) and non-communicable diseases (hypertension, Diabetes Mellitus, ischemic heart disease, stroke, cancer) are prevalent. In recent times, T2DM has become a leading cause of morbidity in Pondicherry District. Rural areas were deliberately selected as the study setting because prevalence studies are rare among the rural population compared with the urban region, and because the investigator's institution is situated in Pondicherry, facilitating logistics and follow-up.[14] Selection of Study Subjects Inclusion criteria: • All people aged above 18 years in the study population. • All pre-diabetes subjects included based on available reports. Exclusion criteria: • All people aged below 18 years. • Women with Gestational Diabetes Mellitus. • Individuals who declined informed consent. • Individuals not available at home even after a third repeat visit. The age cut-off of 18 years was applied because T2DM in this setting is overwhelmingly an adult disease, and the adolescent/pediatric population would require different diagnostic protocols (including Type 1 diabetes ascertainment). Sample Size Calculation The sample size was calculated using the available recent prevalence study in the rural area of Pondicherry (Sumanth et al., which reported 5.8% in Pondicherry, 2012).[9] When the prevalence rate or proportion of occurrence is known, the sample size is calculated by the following method: n = 4pq / L² Where: • p = prevalence/proportion of an event = 5.8 • q = 1 − p = 100 − 5.8 = 94.2 • L² = permissible error in the estimate of p = 20% = 1.3456 n = (4 × 5.8 × 94.2) ÷ 1.345 = 1,624 (required sample size) The calculated sample of 1,624 was inflated to 1,821 to allow for a 20% non-response rate, anticipated because the survey relied on household visits during daytime hours when many adult residents would be away at work. The final achieved sample was 1,821 participants, and no additional replacement sampling was required beyond the planned oversampling. Sampling Technique A multi-stage random sampling technique was applied in the selection of 1,624 sample subjects (1,821 with oversampling) in the survey: • First stage: All 27 Primary Health Centers in Pondicherry were listed. • Second stage: Five PHCs were selected using the lot method: Thirubuvanai (population 32,787), Katterikuppam (12,443), Nettapakkam (14,100), Bahour (26,996), and Abishegapakkam (6,836). These five PHCs were selected to represent each community so that the results would provide generalizable details. • Third stage: One village from each selected PHC was selected by random lot method: Madagadipet (4,750), Mutrampet (2,102), Pandasozanallur (5,544), Bahour (3,978), and Thimmanayakanpalayam (2,833). Data Collection Procedure The first house was selected randomly in each selected village, and data collection moved toward the right side of the first house until adequate sample sizes were obtained in the particular villages (systematic right-hand sweep). The survey procedure was explained to participants in the local (Tamil) and English languages. Potential respondents were informed that participation was voluntary and anonymous, and were assured that their responses would be kept confidential. They were requested to share information by the recall method on the variables mentioned in the questionnaire. Data were collected from participants along with written consent forms, and participants were requested to sign. No incentives of any kind were offered to participants. Data were obtained from those present in the house at the time of the survey; data from those not present were collected later after prior information. All efforts were made to collect data by visiting houses on three occasions. Data were collected mostly in the early morning (6–8 AM) and evening (18–21 PM), hours convenient to participants. Each questionnaire took approximately 15 minutes to complete by the investigator. Study Variables and Questionnaire The survey instrument was anonymous in nature, pretested, and consisted of open-ended questions organized into six sections: 1. Socio-demographic factors: age, sex, religion, education status, occupation, family income, marital status, history of diabetes, sibling diabetes history. 2. Risk factors for T2DM: tobacco usage (smoking tobacco, chewing tobacco, both forms); alcohol consumption (daily—at least 80 mL; regular—once a week; occasionally—once a month); physical activity/exercise; diet habit; type of diet (vegetarian/mixed vegetarian); oil-rich food habit; known hypertension; known dyslipidemia. 3. Seeking of medical care among diabetes subjects: taking medical care or not; place of treatment; type of treatment; regularity of treatment. 4. Co-morbid conditions among diabetics: respiratory infections, ischemic heart disease, gastrointestinal problems, hemiparesis, eye defects (refractive errors/surgeries), skin infections, other complications (joint pain, giddiness, etc.). 5. General and clinical examination. 6. Anthropometric examination: height, weight, and BMI calculation. The questionnaire was pretested in a non-study village within the same PHC catchment to assess clarity, sequencing, and time burden; ambiguous items were reworded before final administration. Anthropometric Measurements Height was measured to the nearest 0.1 cm using a standardized stadiometer with the subject standing erect, heels together, and head in the Frankfurt horizontal plane. Weight was measured to the nearest 0.1 kg using a calibrated spring balance, with subjects wearing light clothing and no footwear. BMI was calculated as weight in kilograms divided by height in meters squared (kg/m²). BMI was categorized using standard cut-offs: ≤18.5 kg/m² (lean/underweight), 18.5–24.99 kg/m² (normal), 25–29.99 kg/m² (overweight), and ≥30 kg/m² (obese). Operational Definitions • Known diabetes: Self-reported physician-diagnosed diabetes, or currently on anti-diabetic medication, documented by available reports. • Known hypertension: Self-reported physician-diagnosed hypertension or on antihypertensive medication. • Known dyslipidemia: Self-reported physician-diagnosed dyslipidemia or on lipid-lowering medication. • Tobacco use: Current use of any form (smoking, chewing, or both). • Alcohol consumption: Categorized as daily (at least 80 mL), regular (once a week), occasional (once a month), or none. • Physical activity/exercise: Self-reported participation in any structured physical activity or exercise. Data Analysis and Statistical Tests The data were analyzed using SPSS version 20. Percentage, proportion, chi-square test, ANOVA, Student's t-test, odds ratio, and significance of association at p<0.05 were calculated. Weighted prevalence, percentage, and mean were analyzed for this study. Chi-square tests were used for categorical comparisons (e.g., diabetes status by BMI category, tobacco use, alcohol use, diet, sex), and Student's t-tests for continuous comparisons (e.g., mean age of diabetics vs. non-diabetics). Odds ratios with 95% confidence intervals were computed for dietary exposures. A p-value <0.05 was considered statistically significant. Ethical Considerations The study was approved by the Institutional Ethical Committee (Human Studies). Written informed consent was obtained from all participants after explaining the purpose, procedures, voluntary nature, and confidentiality of the study in the local language. Participants found to have previously undiagnosed hyperglycemia or other abnormalities on examination were referred to the nearest PHC for confirmatory evaluation and management

A total of 570 houses were surveyed with 2,169 family members, of whom 1,821 were aged ≥18 years (84%). Of these, 859 (47.1%) were male and 962 (52.8%) female. Most were Hindu (97.0%), with Muslims (1.8%) and Christians (0.5%). A total of 1,315 (72.2%) were married, and 416 (22.8%) were illiterate (females 29.8% vs. males 15.0%). Daily wagers constituted 20.5%, and 38.1% were unemployed (including housewives). Nearly 46.1% of families had a monthly income ≤₹10,000. The overall prevalence of T2DM was 14.0% (256/1,821), higher among females (15.1%) than males (12.9%). Prevalence increased progressively with age from 1.2% (21–30 years) to 34.2% (>60 years) (Table 1). Prevalence was significantly associated with BMI (p=0.05), rising from 5.4% among lean subjects to 23.9% among obese subjects (Table 2). Tobacco use (p<0.001), alcohol consumption (p<0.001), and non-vegetarian diet (OR 1.84; p=0.002) were significantly associated with diabetes, while lack of physical activity was not (p=0.170) (Table 3). Known hypertension (64.3%) and dyslipidemia (51.2%) were significantly more common among diabetics than non-diabetics (p<0.05), and 15.6% of diabetics had co-morbid conditions (Table 4). The mean age of diabetics was 52.7±15.1 years versus 38.0±15.1 years for non-diabetics (p<0.001), with a mean diabetes duration of 52.7±15.1 months; nearly 50% of diabetics had affected siblings (Table 5). Table 1: Prevalence of Diabetes Mellitus by Age Group and Sex Age group (years) Male n (%) Female n (%) Total n (%) 21–30 1 (0.4) 6 (1.7) 7 (1.2) 31–40 18 (8.7) 22 (10.2) 40 (9.5) 41–50 33 (18.2) 41 (24.0) 74 (21.0) 51–60 33 (23.9) 47 (34.0) 80 (29.0) >60 26 (34.7) 29 (34.0) 55 (34.2) Total 111 (12.9) 145 (15.1) 256 (14.0) χ² = 2.987, df = 4, p = 0.56 This table presents the age- and sex-specific prevalence of Type 2 Diabetes Mellitus among the 1,821 study participants. Prevalence rises progressively with increasing age in both sexes, from 1.2% in the 21–30 year group to 34.2% in those aged >60 years, with an overall prevalence of 14.0% (15.1% in females vs. 12.9% in males). The difference across age groups was not statistically significant (χ² = 2.987, df = 4, p = 0.56), though the trend demonstrates the strong age-dependence of diabetes burden in this rural population. Table 2: Distribution of Subjects by Body Mass Index and Diabetes Status BMI (kg/m²) Male Diabetic n (%) Male Non-diabetic n (%) Female Diabetic n (%) Female Non-diabetic n (%) Total Diabetic n (%) Total Non-diabetic n (%) ≤18.5 5 (4.5) 72 (9.6) 4 (2.8) 85 (10.4) 9 (3.5) 157 (10.0) 18.5–24.99 63 (56.8) 437 (58.5) 60 (41.4) 474 (51.0) 123 (48.1) 911 (58.2) 25–29.99 30 (27.0) 187 (25.0) 57 (39.3) 192 (23.5) 87 (34.0) 379 (24.2) ≥30 13 (11.7) 52 (7.0) 24 (16.6) 66 (8.1) 37 (14.5) 118 (7.5) Total 111 (100) 748 (100) 145 (100) 817 (100) 256 (100) 1,565 (100) Male: χ² = 5.925, df = 3, p = 0.115; Female: χ² = 35.26, df = 3, p = 0.000; Total: χ² = 34.806, df = 3, p = 0.000 This table cross-tabulates BMI categories (lean, normal, overweight, obese) with diabetes status, stratified by sex. It demonstrates a graded increase in diabetes prevalence with rising BMI—from 3.5% among lean subjects to 14.5% among obese subjects overall. The association was statistically significant among females (p = 0.000) and the total population (p = 0.000), but not among males (p = 0.115), suggesting a particularly strong obesity–diabetes link among rural women. Table 3: Prevalence of Diabetes Mellitus and Association of Risk Factors Risk factor Male n (%) Female n (%) Total n (%) p-value Tobacco use No 88 (11.2) 141 (16.3) 229 (13.8) <0.001 Chewing 1 (5.9) 4 (16.0) 5 (11.9) Smoking 21 (42.0) – 21 (18.3) Both 1 (25.0) – 1 (11.1) Alcohol No 80 (11.9) 145 (15.1) 225 (13.7) <0.001 Daily 5 (16.1) – 5 (16.1) Occasionally 15 (16.7) – 15 (16.7) Regularly 11 (17.5) – 11 (17.5) Physical activity No 82 (10.4) 118 (12.8) 200 (11.7) 0.170 Yes 29 (39.1) 27 (64.3) 56 (48.3) Diet Vegetarian 13 (27.1) 15 (22.7) 28 (25.0) 0.037 Non-vegetarian 98 (12.1) 130 (14.5) 228 (13.4) Type of diet Mixed 57 (12.0) 66 (13.0) 123 (13.0) 0.501 Carbohydrate 42 (12.1) 57 (14.3) 99 (13.3) Protein 12 (33.3) 22 (40.0) 34 (37.4) This table 3 précises the prevalence of diabetes across key behavioral and dietary risk factors—tobacco use, alcohol consumption, physical activity, diet type (vegetarian/non-vegetarian), and diet composition—stratified by sex. Tobacco use (p<0.001), alcohol consumption (p<0.001), and vegetarian/non-vegetarian diet (p = 0.037) showed statistically significant associations with diabetes, whereas lack of physical activity (p = 0.170) and type of diet (p = 0.50) did not. These findings highlight the modifiable behavioral determinants of diabetes in this rural setting. Table 4: Co-morbid Conditions and Risk Profile Among Diabetic Subjects Parameter Male n (%) Female n (%) Total n (%) Known hypertension 31 (27.9) 52 (35.9) 83 (32.4) Known dyslipidemia 10 (9.0) 12 (8.3) 22 (8.6) Co-morbid conditions None – – 216 (84.4) Giddiness – – 17 (6.6) Joint pain – – 10 (3.9) Eye defect – – 9 (3.5) Diabetic foot – – 2 (0.8) Skin infection – – 1 (0.4) Hemiparesis – – 1 (0.4) Medical care 95 (85.6) 121 (83.5) 216 (84.4) Government 61 (55.0) 87 (60.0) 148 (57.8) Private 34 (30.6) 31 (21.4) 65 (25.4) Regularity of treatment 84 (75.7) 111 (76.6) 195 (76.2) This table 4 describes the burden of co-morbidities and health-seeking behavior among the 256 diabetic subjects. Known hypertension was present in 32.4% (females 35.9% vs. males 27.9%) and known dyslipidemia in 8.6% of diabetics; 15.6% had at least one co-morbid condition, most commonly giddiness (6.6%), joint pain (3.9%), and eye defects (3.5%). The majority (84.4%) were receiving medical care, predominantly from government facilities (57.8%), and 76.2% reported regular treatment, indicating reasonable healthcare engagement despite the rural setting. Table 5: Comparison of Mean Age, Duration, and Family History Between Diabetic and Non-diabetic Subjects Parameter Diabetic (n = 256) Non-diabetic (n = 1,565) Total (n = 1,821) p-value Mean age (years) Total 52.7 ± 15.1 38.0 ± 15.1 40.1 ± 26.6 <0.001 Male 52.8 ± 1.2 38.5 ± 0.5 40.4 ± 15.1 <0.001 Female 52.6 ± 13.1 37.5 ± 14.5 39.8 ± 15.1 <0.001 Mean duration of diabetes (months) 52.7 ± 15.1 – – – Family history Sibling with diabetes 126 (49.2) – – – One sibling 40 (15.7) – – – >1 sibling 86 (33.6) – – – This table compares the mean age of diabetic and non-diabetic participants overall and by sex, and summarizes diabetes duration and family history among diabetics. Diabetics were significantly older than non-diabetics (52.7 ± 15.1 vs. 38.0 ± 15.1 years; p<0.001), a difference that held for both sexes. The mean duration of diabetes was 52.7 ± 15.1 months, and nearly half (49.2%) of diabetics reported at least one sibling with diabetes (33.6% with more than one affected sibling), underscoring the strong familial aggregation of T2DM in this population

This community-based cross-sectional study provides one of the few contemporary estimates of T2DM prevalence and its risk-factor profile in the rural population of Pondicherry, a Union Territory in South India. The overall prevalence of 14.0% is considerably higher than the 5.8% reported by Sumanth et al. from rural Pondicherry in 2012,[9] the 8.32% reported by Bharati et al. from Puducherry,[10] and the 2.4%–6.4% documented in earlier rural south Indian transitions,[8] but it falls within the range reported from other rural Indian settings (9.2%–18.17%)[11,15] and approaches or exceeds several urban estimates of 4.6% and 9.0%.[5,16] Several factors may explain this convergence of rural and urban prevalence. Rural Pondicherry has undergone rapid socioeconomic development—motorized transport, mechanized agriculture, increased family income, and improved educational attainment—that has reduced energy expenditure and increased caloric availability, the same drivers described by Ramachandran in the rural epidemiological transition.[8] The catchment villages are also peri-urban in character, with easy access to Pondicherry town, processed foods, and sedentary occupations, which may make them less representative of remote rural India. In addition, the study relied on known (self-reported) diabetes rather than oral glucose tolerance testing, which would tend to underestimate true prevalence since a substantial fraction of T2DM in rural India is undiagnosed; the observed 14.0% may therefore represent a lower bound. Finally, greater health awareness and screening penetration in Pondicherry may have increased the proportion of previously diagnosed cases captured by self-report. The female preponderance observed in this study (15.1% vs. 12.9%) is consistent with several Indian studies, including Arora et al.[3], Anand et al.[17], Bener et al.[18], and Acemoglu et al., although other studies—notably Sumanth et al. in the same region—found no significant sex difference.[9] The higher female prevalence in our data may reflect the combination of greater abdominal adiposity among rural women, lower physical activity, and the high proportion of housewives (who constituted the bulk of the unemployed group), as well as greater health-care contact through maternal and child health services leading to more opportunistic screening. The absence of statistical significance in the age-stratified sex comparison (χ² = 2.987, p = 0.56) indicates that the sex difference is modest and should be interpreted cautiously, since it may be confounded by age structure. The steep age gradient—from 1.2% in the 21–30 year group to 34.2% in those over 60 years—replicates the age–prevalence relationship reported consistently in Indian and international literature [3,5,12,18] and has two important implications. First, approximately 72% of diabetics in this cohort were aged 41 years or older, placing the majority in the economically productive or early-retirement phase, so the indirect costs of diabetes—absenteeism, caregiving, and out-of-pocket expenditure—are likely to be substantial. Second, the presence of seven cases (1.2%) in the 21–30 year age band, although small, signals early-onset disease in a rural population, which carries a longer lifetime exposure to hyperglycemia and a higher cumulative risk of complications. The graded association between BMI and diabetes—5.4% in lean, 11.9% in normal-weight, 18.7% in overweight, and 23.9% in obese subjects—is biologically coherent and consistent with the established role of adiposity in insulin resistance. The association was statistically significant in women (p = 0.000) and in the total sample (p = 0.000) but not in men (p = 0.115), a pattern that likely reflects both the larger number of overweight and obese women in this cohort and the greater contribution of central adiposity to metabolic risk in South Asian women. The "Asian Indian Phenotype"—insulin resistance and abdominal adiposity at lower BMI thresholds—means that even the "normal" BMI category (18.5–24.99 kg/m²) in this population carries a diabetes prevalence of nearly 12%, almost double the lean-category prevalence. This has direct screening implications: BMI-based risk stratification alone will miss a substantial number of at-risk rural Indians, and waist circumference or waist-to-height ratio should be incorporated into community screening algorithms. Our findings align with Rao et al.[15], Bener et al.[18], Ramachandran et al.[8], and Ravikumar et al.[19], all of whom reported significant associations between central obesity and diabetes. With respect to behavioral risk factors, tobacco use was reported by 9.1% of participants (166/1,821), with no female smokers; diabetes prevalence among tobacco users was 18.3%, and the association was statistically significant (p<0.001), consistent with the meta-analysis by Willi et al., which estimated a 45% increased risk of diabetes among smokers independent of other risk factors,[20] and with the dose-dependent relationship reported by Fawwad et al. in rural Baluchistan. Alcohol consumption was reported by 10.1% (184/1,821), all male, and was significantly associated with diabetes (p<0.001), with prevalence rising across categories (16.1% daily, 16.7% occasional, 17.5% regular) compared with 13.7% among non-drinkers; however, the literature on alcohol is notoriously inconsistent, ranging from U-shaped and protective associations at low doses[21] to harm at high doses, and the small absolute numbers in each drinking category (5, 15, and 11 cases) limit precision. We therefore interpret the alcohol finding as hypothesis-generating rather than definitive and caution against causal interpretation, since confounding by socioeconomic status, smoking, and diet is likely. Lack of physical activity was reported by 93.6% (1,705/1,821) of participants—an implausibly high figure that reflects the questionnaire's binary operationalization of "exercise" (structured activity) rather than total physical activity, which in rural populations is often occupational and thus not captured. Diabetes prevalence among those reporting no exercise was 11.7% versus 48.3% among those reporting exercise, but the difference was not statistically significant (p = 0.170); the direction of this association is counterintuitive and almost certainly reflects reverse causation, since individuals diagnosed with diabetes may have been advised to initiate exercise, inflating prevalence in the exercising group. This highlights the need for objective activity measurement (e.g., accelerometry or occupational activity diaries) in future rural studies. Non-vegetarians had a 1.84-fold higher odds of diabetes than vegetarians (p = 0.002), consistent with the Nurses' Health Study findings that greater consumption of fiber and polyunsaturated fats is protective, whereas soft drinks, glycemic load, and trans fats increase risk.[22] The vegetarian advantage in this cohort may operate through higher legume and whole-grain intake, lower saturated fat, and lower glycemic load, although the "vegetarian" category in rural Tamil Nadu is heterogeneous and may include refined carbohydrate–heavy diets; the finding therefore requires confirmation with validated food-frequency instruments. Nearly half of diabetics (49.2%) reported at least one affected sibling, and one-third (33.6%) reported more than one, consistent with the strong heritability of T2DM (concordance approximately 90% in identical twins) and with Indian family studies, including Sumanth et al., who reported a 3.8-fold increased risk among those with parental diabetes and a stronger maternal than paternal effect[9]; sibling history is thus a simple, free, and highly informative screening question that frontline health workers can use to prioritize households for opportunistic glucose testing. Known hypertension was present in 64.3% of diabetics (females 67.5% vs. males 59.6%) and known dyslipidemia in 51.2% (females 63.2% vs. males 41.7%); both were significantly more common among diabetics than non-diabetics (p<0.05), consistent with the well-described clustering of metabolic syndrome components and with prior Indian reports.[5,11,23] The high co-morbidity burden reinforces the case for integrated screening and management of hypertension, dyslipidemia, and diabetes at the primary-care level rather than vertical, single-disease programs. Co-morbid conditions were reported by 15.6% of diabetics, most commonly giddiness (6.6%), joint pain (3.9%), and eye defects (3.5%), while diabetic foot, skin infection, and hemiparesis were rare (≤0.8% each); the comparatively low prevalence of classical microvascular complications is likely an artifact of self-report methodology and of the relatively short mean diabetes duration (52.7 months; nearly 60% had diabetes for <48 months) rather than a true absence of complications, and formal complication screening (fundoscopy, urine albumin-to-creatinine ratio, foot examination) would almost certainly reveal a higher burden, as documented by Arora et al.[3] Health-seeking behavior was encouraging: 84.4% were receiving medical care (57.8% government, 25.4% private), 83.6% were on allopathic treatment, and 76.2% reported regular treatment, with the predominance of government facilities consistent with the rural setting and availability of free or subsidized services through PHCs and government hospitals; however, 15.6% were receiving no treatment at all, and only 21.5% were undergoing regular blood investigations every 6–12 months, pointing to gaps in monitoring and follow-up even among those ostensibly "on treatment." This study has several strengths. It is a community-based survey with a large sample (n = 1,821) drawn by multi-stage random sampling from five PHC catchments, enhancing representativeness relative to clinic-based studies. Standardized anthropometry, a pretested questionnaire, and a high response rate (84% of eligible household members) further support internal validity. However, several limitations must be acknowledged. The cross-sectional design precludes inference of causality, and the reliance on self-reported physician-diagnosed diabetes rather than oral glucose tolerance testing means that undiagnosed cases—estimated to constitute a substantial proportion of T2DM in rural India were not captured, so the true prevalence is likely higher than 14.0%. Recall bias may have affected reporting of diabetes duration, previous co-morbid conditions, and behavioral exposures, particularly among older participants. The binary operationalization of physical activity and the unvalidated dietary categorization are additional sources of misclassification. Gestational diabetes was excluded by design, and pre-diabetes (impaired glucose tolerance) could not be separately classified for analysis, limiting inference about the full dysglycemic spectrum. Finally, the peri-urban character of the selected villages may limit generalizability to more remote rural populations. Notwithstanding these limitations, the study provides a robust, contemporaneous estimate of rural T2DM burden and identifies a constellation of modifiable risk factors obesity, tobacco, alcohol, and non-vegetarian diet that can be targeted through primary-care–based screening and lifestyle interventions, and it underscores the need for longitudinal studies with biochemical ascertainment and complication screening in this population. Future research should incorporate oral glucose tolerance testing to capture undiagnosed diabetes, objective physical activity measurement, validated dietary instruments, and formal complication screening, and should evaluate the effectiveness of annual community screening coupled with lifestyle counseling delivered by frontline health workers in this setting.

The prevalence of T2DM in rural Puducherry was 14.0%, with higher rates among females (15.1%) than males (12.9%). One in ten community members and two in ten obese members are affected. Mean age of diabetics was 52.7±15.1 years, and nearly 50% had a sibling with diabetes. Prevalence among lean, normal BMI, and overweight/obese subjects was 5.4%, 11.8%, and 48.4%, respectively. Prevalence among tobacco users, alcoholics, and those lacking physical activity was 18.3%, 16.3%, and 11%, respectively. Hypertension (64.3%) and dyslipidemia (51.2%) were common among diabetics, and 15.6% had co-morbid conditions. Recommendations: Annual community screening for pre-diabetes and diabetes; promotion of ≥150 minutes/week of physical activity and dietary modification, especially for those with affected siblings; cessation of tobacco and reduction of alcohol and oil-rich foods; and periodic screening for co-morbid conditions among diabetic and pre-diabetic subjects.

1. World Health Organization. Diabetes Mellitus Fact sheet N°310. Geneva: WHO; October 2013. p. 1–8. 2. International Diabetes Federation. Diabetes Atlas. 6th ed. Brussels: IDF; 2013. p. 1–160. 3. Arora V, Malik JS. Prevalence of diabetes in urban Haryana. AMJ. 2010;3(8):488–94. 4. Ramachandran A, Snehalatha C, Shetty AS, Nanditha A. Trends in prevalence of diabetes in Asian countries. World J Diabetes. 2012;3(6):110–7. 5. Shiju TM, Deepa M, Pragasam V. An alarming prevalence of diabetes and its associated risk factors among college going Indian adults: a retrospective study. Int J Med Sci Public Health. 2013;2(3):603–8. 6. Mohan V, Mathur P, Deepa R, Deepa M, Shukla DK, Menon GR, et al. Urban rural differences in prevalence of self-reported diabetes in India—the WHO-ICMR Indian NCD risk factor surveillance. Diabetes Res Clin Pract. 2008;80(1):159–68. 7. Indian Council of Medical Research. ICMR guidelines for management of type 2 diabetes. New Delhi: ICMR; 2005. p. 1–42. 8. Ramachandran A. Epidemiology of diabetes in India—three decades of research. J Assoc Physicians India. 2005;53:217–30. 9. Majgi SM, Soudarssanane MB, Roy G, Das AK. Prevalence of diabetes mellitus and role of stress in diabetes in rural Pondicherry—a union territory of India. GJMEDPH. 2012;1(5):40–5. 10. Bharati DR, Pal R, Kar S, Rekha R, Yamuna TV, Basu M. Prevalence and determinants of diabetes mellitus in Puducherry, South India. J Pharm Bioallied Sci. 2011;3(4):513–8. 11. Kapoor D, Bhardwaj AK, Kumar D, Raina SK. Prevalence of diabetes mellitus and its risk factors among permanently settled tribal individuals in tribal and urban areas in northern sub-Himalayan region of India. Int J Chronic Dis. 2014;2014:380597. 12. Deepthi R, Chandini C, Pratyusha K, Kusuma N, Raajitha B, Shetty G. Screening for diabetes and their risk factors among adults in rural Kolar—a community based study. Int J Res Dev Health. 2013;1(4):18–21. 13. Kumar S, Mukherjee S, Mukhopadhyay P, Pandit K, Raychaudhuri M, Sengupta N, et al. Prevalence of diabetes and impaired fasting glucose in a selected population with special reference to influence of family history and anthropometric measurements—the Kolkata Policeman Study. J Assoc Physicians India. 2008;56:841–4. 14. Census of India 2011. Puducherry (Pondicherry) district: census 2011 data. New Delhi: Office of the Registrar General & Census Commissioner, India; 2011. 15. Rao CR, Kamath VG, Shetty A, Kamath A. A study on the prevalence of type 2 diabetes in coastal Karnataka. Int J Diabetes Dev Ctries. 2010;30(2):80–5. 16. Patil RS, Gothankar JS. Prevalence of type 2 diabetes mellitus and associated risk factors in an urban slum of Pune City, India. Natl J Med Res. 2013;3(4):346–9. 17. Anand K, Shah B, Yadav K, Singh R, Mathur P, Paul E, et al. Are the urban poor vulnerable to non-communicable diseases? A survey of risk factors for non-communicable diseases in urban slums of Faridabad. Natl Med J India. 2007;20(3):115–20. 18. Bener A, Zirie M, Janahi I, Al-Hamaq AO, Musallam M, Wareham NJ. Prevalence of diagnosed and undiagnosed diabetes mellitus and its risk factors in a population-based study of Qatar. Diabetes Res Clin Pract. 2009;84(1):99–106. 19. Ravikumar P, Bhansali A, Ravikiran A, Bhansali S, Dutta P, Thakur JS, et al. Prevalence and risk factors of diabetes in a community-based study in North India: the Chandigarh Urban Diabetes Study (CUDS). Diabetes Metab. 2011;37(3):216–21. 20. Willi C, Bodenmann P, Ghali WA, Faris PD, Cornuz J. Active smoking and the risk of type 2 diabetes: a systematic review and meta-analysis. JAMA. 2007;298(22):2654–64. 21. Koppes LL, Dekker JM, Hendriks HF, Bouter LM, Heine RJ. Meta-analysis of the relationship between alcohol consumption and coronary heart disease and mortality in type 2 diabetic patients. Diabetologia. 2006;49(4):648–52. 22. Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med. 2001;345(11):790–7. 23. Shabana S, Sasisekhar TVD. Effect of gender, age and duration on dyslipidemia in type 2 diabetes mellitus. Int J Cur Res Rev. 2013;5(6):104–13. 24. Legro RS, Gnatuk CL, Kunselman AR, Dunaff A. Changes in glucose tolerance over time in women with polycystic ovary syndrome: a controlled study. J Clin Endocrinol Metab. 2005;90(6):3236–42. 25. World Health Organization. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: Diagnosis and classification of diabetes mellitus. Geneva: WHO; 1999 (WHO/NCD/NCS/99.2). p. 1–9.