JRHS 2014; 14(4): 258-263

JRHS
Journal of Research in Health Sciences

journal homepage: www.umsha.ac.ir/jrhs

Original Article

High Prevalence of Type 2 Diabetes Melitus and Its Risk Factors Among the
Rural Population of Pondicherry, South India
a* b
Saurabh RamBihariLal Shrivastava (MD) , Arun Gangadhar Ghorpade (MD)
a
Department of Community Medicine, Shri Sathya Sai Medical College & Research Institute, Kancheepuram, Tamil Nadu, India
b
Department of Community Medicine, Sri Manakula Vinayagar Medical College and Hospital, Pondicherry, India

ARTICLE INFORMATION ABSTRACT

Article history: Background: We aimed to assess the prevalence of type-2 diabetes in rural Pondicherry and to
Received: 15 August 2014 study the determinants of Diabetes Mellitus (DM) in the rural population of Pondicherry, south
Revised: 11 September 2014 India.
Accepted: 26 October 2014 Methods: It was a cross-sectional community-based study conducted from November 2010 to
Available online: 01 November 2014 January 2012 in two of the field practice villages affiliated to a Medical College in Pondicherry.
Sample size was calculated using open source software, Open Epi Version 2.3.10. The sampling
frame comprised individuals aged above 25 years and single stage cluster random sampling was
Keywords:
carried out. After obtaining the verbal informed consent each of the study participants were inter-
Diabetes Mellitus viewed face-to-face using a pre-tested structured questionnaire. Data were analyzed using the
Obesity SPSS version 16.
Tobacco Results: The age of the study participants ranged from 25 to 98 years with mean of 42.6 (±13.7)
India and majority of the study participants 339 (32.5%) from the age-group of 30-39 years. The
prevalence of diabetes was 19.8% (60-69 years), 17.1% (40-49 years), 16.8% (50-59 years),
and 13.6% (>69 years) among study subjects. In univariate analysis, higher age, being educat-
* Correspondence ed, unemployed and poor was associated with higher risk of diabetes mellitus (DM). Further-
Saurabh RamBihariLal Shrivastava (MD) more, a high triglyceride level was significantly associated with increase in the risk of DM (ad-
Tel: +91 98 84227224 justed odds ratio: 3.01; 95% CI: 1.86, 4.86).
E-mail: [email protected] Conclusions: Type-2 diabetes mellitus (T2DM) is an important public health problem in the
adults of rural Pondicherry. Among non-modifiable factors, higher age, better socio-educational
background and positive family history of diabetes was significantly associated with T2DM.

Citation: Shrivastava SR, Ghorpade AG. High Prevalence of Type 2 Diabetes Melitus and Its Risk Factors Among the Rural Population of Pondicherry, South
India. J Res Health Sci. 2014; 14(4): 258-263.

Introduction

D iabetes mellitus (DM) is a chronic metabolic disease

characterized by hyperglycemia either due to abso-
lute (Type 1 DM) or relative (Type 2 DM) insuffi-
ciency of the insulin hormone1. In contrast to the earlier as-
sumption that DM is a predominant disease diagnosed in de-
tioning of every system of the body, especially if blood sugar
remains deranged for a considerable period of time1,12.
Realizing the magnitude of the disease and dearth of
community-based studies especially in rural areas, the present
study has been conducted to assess the prevalence of type-2
veloped countries, recent findings have shown a considerable
diabetes in rural Pondicherry and to study the determinants of
rise in number of new cases of type 2 DM with an earlier
DM in the rural population of Pondicherry.
onset and associated complications, even in developing coun-
tries2-4. Furthermore, almost 347 million people worldwide
have diabetes and it is likely to be doubled by 2030 in ab- Methods
sence of any intervention5,6.
Study setting
In addition, more than 80% of diabetes deaths occur in
It was a cross-sectional descriptive study conducted from
low and middle-income countries and it will be ranked as the
November 2010 to January 2012 in two of the catchment
seventh leading cause of death in 2030 6,7. India has been des-
villages, namely Ramanathapuram and Pillaiyarkuppam un-
ignated as “diabetes capital” of the world, and diabetes now
der the JIPMER Rural Health Centre. Sample size was calcu-
affects a staggering 10-16% of urban population and 5-8% of
lated using a freely available open source software, Open Epi
rural population in India alone8,9. The causation of DM is
Version 2.3.10, with a 95% confidence interval (CI) and 80%
multi-factorial, and includes both non-modifiable (viz. age,
power, to detect a DM prevalence of 5.8% (±2.2), minimum
genetic defects, family history) and modifiable (viz. obesity,
of 434 subjects were needed11. Taking into account a design
sedentary lifestyle, diet, stress, alcohol, viral infections, soci-
effect of 2 for the cluster sampling and a non-response rate of
oeconomic status)2,10,11. Diabetes not only affects the quality
25%, sample size was calculated to be 1043. The sampling
of life of patients but also virtually hampers the routine func-
frame comprised individuals aged above 25 years (n=2608).
 Saurabh RamBihariLal Shrivastava et al 259

Single stage cluster random sampling was carried out. Using Statistical analysis
streets as the primary sampling unit, four streets in Rama-
Data were analyzed using the SPSS statistical package
nathapuram and six streets in Pillaiyarkuppam were chosen
version 16.0 for Windows (SPSS Inc., Chicago, United States
by lot method.
of America). The statistical significance was set at P<0.05.
Inclusion and exclusion criteria The associations were assessed using Chi-square test and
unpaired t- test the categorical (%) and continuous (mean
From the houses of the selected streets, all participants
±SD) variables respectively. Adjusted risk of diabetes was
aged more than 25 years were invited to take part in the
assessed with backward logistic regression model. Variables
study. Subjects not willing to participate (n=31) in the study
with P<0.2 in univariate analysis were forced in the regres-
were excluded. Data for pregnant women (2) and missing
sion model.
forms (7) were excluded from analysis.
Study tool Results
After obtaining the verbal informed consent each of the Table 1 describes the socio-demographic characteristics
study participants were interviewed face-to-face using a of study participants and their distribution with reference to
structured questionnaire. The questionnaire was pre-tested on the gender. The age of the study participants ranged from 25
a group of 30 individuals before its utilization in the current to 98 years with mean of 42.6 (±13.7). Majority of the study
study. participants 339 (32.5%) belonged to the age-group of 30-39
Study variables years. The higher proportion of women (44.6%) did not at-
tend the school as compared to men (15.3%). In females,
Socio-demographic parameters (viz. age, sex, education, nonworking status was more prevalent than males (47.2% vs
occupation, per capita income), family history, level of phys- 13.2%, P=0.004). Among the other socio-demographic and
ical activity, and addiction to tobacco / alcohol. In addition, lifestyle risk factors, men and women did not differ with re-
each of the study subjects was subjected to anthropometric spect to per capita income, physical activity and obesity.
measurements (viz. height, weight & waist circumference); However, central obesity as measured by waist circumference
assessment of pulse rate & blood pressure and laboratory was more in women than men (39.1% vs 29.5%, respective-
investigations (viz. lipid profile, estimation of fasting and ly).
postprandial blood glucose, and HbA1C levels, if diabetic).
Males consumed more amounts of calories and proteins
Operational definitions than females. Similarly, statistically significant difference
Education was classified using International Standard was observed between men and women pertaining to some of
Classification of Education as no formal schooling and at- the estimates of lipid profile (viz. total cholesterol, LDL and
tended school13. Census guidelines and B G Prasad modified HDL levels). In univariate analysis, higher age, being educat-
classification were utilized for classifying the work status and ed, unemployed and poor was associated with higher risk of
socioeconomic status respectively14,15. Physical activity was DM. Among the modifiable risk factors, physically inactive,
measured using the International Physical Activity Question- being obese, high blood pressure, and deranged lipid profile
naire15 (short version)16. Total metabolic equivalents/week were the significant risk factors of diabetes. Oil intake among
(MET/wk) were calculated and individuals grouped as physi- diabetics was higher than non-diabetics. People with diabetes
cally inactive (<600 MET/wk) and physically active (≥600 had higher resting pulse rate and pulse pressure than those
MET/wk)16. Smoking was defined as the current use of any without diabetes (Table 2).
tobacco product on a regular basis for ≥ six months 17. Alco- Table 3 reveals the diet pattern and its association by
hol use was defined as the consumption of any type of alco- gender and diabetes in study subjects. Among all the dietary
hol in the last one year17. Indian Council of Medical Research practices studied, consumption of fried food intake was sig-
guidelines were used for diagnosis and classification of dia- nificantly less among women than men (17.3% vs 12.4%,
betes and participants with fasting (>125 mg/dL) and/ or P=0.029). A considerable proportion of women (15%) were
postprandial (>200 mg/dL) blood glucose were diagnosed as consuming tobacco in chewable form as compared to men
diabetic18. Standard guidelines were followed to measure (4.1%). Among the modifiable risk factors, addiction to ciga-
height and weight. However, to negate observer’s bias, meas- rette smoking and alcohol usage were the significant predic-
urement was done twice one by the investigator and other by tor as they augmented the risk of DM by 2.37 and 1.62 times
a trained person and the average of two was taken. Body respectively.
mass index (BMI) was calculated and classified as per the
WHO classification (<23 kg/m2 as normal and ≥23 kg/m2 as Table 4 presents the association between risk factors and
overweight and obese)19. Blood pressure was checked with obesity using multivariate analysis. After adjusting for the
digital sphygmomanometer with study subjects sitting com- risk factors, increasing age, being educated, higher income
fortably. and positive family history were the non-modifiable signifi-
cant risk factors of diabetes. Although, subjects with high
Ethical considerations blood pressure have 1.63 times risk of diabetes but the asso-
The survey was conducted after taking approval from the ciation was not statistically significant (P=0.069). Among the
institutional Ethics Committee. Verbal informed consent was modifiable risk factors, physical activity was the strongest
obtained from all study participants before interviewing them predictor with being inactive raising the risk of DM by 5.34
and utmost care was taken to maintain privacy and confiden- times. In addition, having a waist circumference higher than
tiality. the normal recommended levels (viz. males - ≥90 cm, fe-
males - ≥80cm) increased the risk of DM by 1.86 times. Fur-
thermore, high triglyceride levels raises the risk of DM (ad-

JRHS 2014; 14(4): 258-263

 260 Diabetes in rural Pondicherry

justed odds ratio: 3.01; 95% CI: 1.86, 4.86). High pulse rate 10 led to 40% increased risk of diabetes.
was significantly associated with DM, as rise in pulse rate by
Table 1: The odds ratio (OR) estimates of diabetes by variables
Without diabetes With diabetes
Variables n=1043 n=127 OR (95% CI) P value
Age (yr)
25-29 85 3 1.00
30-39 339 18 1.53 (0.44, 5.33) 0.502
40-49 252 43 5.62 (1.70, 18.63) 0.005
50-59 190 32 5.54 (1.65, 18.62) 0.006
60-69 111 22 6.76 (1.95, 23.42) 0.003
>69 66 9 4.32 (1.12, 16.64) 0.034
Educational status
Illiterate 316 27 1.00
Attended school 727 100 1.70 (1.09, 2.67) 0.018
Occupational status
Worker 724 74 1.00
Non-workers 319 53 1.75 (1.20, 2.56) 0.004
Per capita income in Rs/month
≥3100 954 107 1.00
<3100 89 20 2.30 (1.34, 3.92) 0.002
Physical activity level (MET/wk)
Active (≥600) 968 100 1.00
Inactive (<600) 75 27 4.88 (2.92, 8.20) 0.001
Family history of T2 DM
Absent 776 74 1.00
Present 267 53 2.35 (1.60, 3.45) 0.001
Body mass index (kg/m2)
<25 758 73 1.00
≥25 285 54 2.19 (1.50, 3.21) 0.001
Waist circumference (cm)
Normal (<88) 684 52 1.00
High risk (≥88) 359 75 3.21 (2.20, 4.70) 0.001
Blood pressure (mmHg)
Normal (<120/80) 806 68 1.00
Above normal (≥120/80) 237 59 3.60 (2.45, 5.29) 0.001
Total cholesterol (mg/dl)
Normal (<200) 724 69 1.00
Above normal (≥200) 232 48 2.48 (1.66, 3.71) 0.001
Triglyceride (mg/dl)
Normal (<150) 758 66 1.00
Above normal (≥150) 198 51 3.64 (2.42, 5.46) 0.001
Low density lipoprotein (mg/dl)
Normal (<70) 753 80 1.00
Above normal (≥70) 203 37 1.88 (1.23, 2.87) 0.003
High density lipoprotein (mg/dl)
Normal (≥40l) 401 66 1.00
Above normal (<40) 117 51 0.51 (0.35, 0.76) 0.001

Table 2: Risk factors and its association by gender and diabetes in study subjects using independent t-test
Variables Women Men P value Normal Diabetics P value
Calories (kcal/day), mean (SD) 1674 (550) 2326 (856) 0.001 1987 (781) 2007 (815) 0.802
Proteins (gm/day), mean (SD) 42 (16) 57 (23) 0.001 49 (21) 49 (21) 0.872
Oil (ml/month), mean (SD) 760 (327) 745 (344) 0.457 743 (336) 814 (334) 0.026
Pulse rate, mean (SD) 79 (10) 78 (12) 0.064 77 (11) 84 (15) 0.001
Pulse pressure, mean (SD) 43 (12) 43 (11) 0.722 43 (11) 48 (15) 0.001

Discussion
The prevalence of diabetes among the study participants was 12.2%, which is much higher than a similar study done in a
rural community of Pondicherry where prevalence was 5.8% among study participants 9. Another study conducted among the
rural areas of Tamil Nadu revealed that prevalence of diabetes in studied population was 5.99% 20. Findings of a systematic
review and meta-analysis revealed that rural prevalence of diabetes in low-middle income countries has increased from 1.8%
(1985-89) to 7.5% (2005-11)21. However, a significantly higher prevalence was observed among migrants (14%) in a cross-
sectional study22. Most of the prevalence results have revealed that a definite hike in diabetes prevalence has been observed,
especially in rural settings of developing countries23. However, the heterogeneity in results could also be probably because of
socio-demographic variability and employment of differing guidelines (cut-off values) in reaching a diagnosis of diabetes.

JRHS 2014; 14(4): 258-263

 Saurabh RamBihariLal Shrivastava et al 261

Table 3: Diet pattern and its association by gender and diabetes in study subjects
Without diabetes With diabetes
Variablesa n=1041 n=125 OR (95% CI) P value
Diet preference
Vegetarian 61 11 1.00
Non-vegetarian 980 116 0.61 (0.31, 1.21) 0.151
Vegetable intake (g/day)
Low (<100) 669 84 1.00
High (≥100) 351 41 0.95 (0.64, 1.41) 0.798
Fruit intake
Absent 80 13 1.00
Present 961 114 0.69 (0.37, 1.30) 0.249
Salt intake (g/day)
Normal (<5) 403 52 1.00
High (≥5) 623 74 0.91 (0.62, 1.33) 0.625
Cooking oil
Mixed 28 6 1.00
Single type 991 117 0.49 (0.19, 1.24) 0.123
Fried food intake
Absent 155 13 1.00
Present 885 114 1.62 (0.89, 2.95) 0.115
Aerated drinks intake
Absent 498 65 1.00
Present 543 62 1.16 (0.80, 1.69) 0.421
Tobacco chewing
Absent 942 113 1.00
Present 101 14 1.18 (0.65, 2.15) 0.586
Cigarette smoking
Absent 988 114 1.00
Present 42 13 2.37 (1.24, 4.56) 0.008
Alcohol use
Absent 787 85 1.00
Present 256 42 1.62 (1.09, 2.42) 0.017
a
Dietary data missing for 2 subjects

Table 4: Multivariate analysis for association of risk factors and obesity

Without diabetes With diabetes
Variables n=1043 n=127 OR (95% CI) P value
Age (yr)
25-29 85 3 1.00 0.001
30-39 339 18 1.66 (0.43, 6.46) 0.464
40-49 252 43 6.98 (1.85, 26.42) 0.004
50-59 190 32 6.01 (1.56, 23.21) 0.009
60-69 111 22 8.48 (2.08, 34.48) 0.003
>69 66 9 3.82 (0.76, 19.16) 0.104
Educational status
Illiterate 316 27 1.00
Attended school 727 100 2.28 (1.24, 4.20) 0.008
Per capita income in Rs/month
≤3100 89 20 1.00
>3100 954 107 2.16 (1.07, 4.35) 0.031
Family history of Diabetes Mellitus
Absent 776 74 1.00
Present 267 53 1.96 (1.20, 3.23) 0.008
Physical activity level (MET/wk)
Inactive (<600) 75 27 1.00
Active (≥600) 968 100 5.34 (2.62, 10.88) 0.001
Waist circumference
Normal 684 52 1.00
High risk 359 75 1.86 (1.16, 2.98) 0.010
Blood pressure (mmHg)
Normal (<120/80) 806 68 1.00
Above normal (≥120/80) 237 59 1.63 (0.96, 2.75) 0.069
Triglyceride (mg/dl)
Normal (<150) 758 66 1.00
Above normal (≥150) 198 51 3.01 (1.86, 4.86) 0.001
Low density lipoprotein (mg/dl)
Normal (<70) 753 80 1.00
Above normal (≥70) 203 37 1.57 (0.93, 2.64) 0.089
Pulse pressure (mmHg)
Normal (<60) 966 43 1.00
Abnormal (≥60) 77 84 1.02 (1.00, 1.04) 0.055
Pulse rate (beats/minute)
Normal (60-100) 1000 79 1.00
Abnormal (<60 or >100) 43 48 1.04 (1.02, 1.06) 0.001

In the current study, a directly proportional relationship tained in studies done in other parts of the country24,25. The
was observed in the prevalence of diabetes and increasing age probable reason for rise in prevalence of diabetes with in-
among the study subjects. Similar sorts of results were ob- creasing age is because of the amplification of the physical

JRHS 2014; 14(4): 258-263

 262 Diabetes in rural Pondicherry

inactivity, obesity, adoption of harmful lifestyles over the pulse rate and adverse lipid profile should be monitored for
years. This clearly reflects the incompetency of the public development of diabetes.
health system and the health care professionals in preventing
people from adopting harmful lifestyles. Conflict of interest statement
The study subjects who were unemployed or were poor There was no conflict of interest to be stated.
had a higher prevalence of diabetes in the present study.
However, contrasting results were obtained in some of the
other studies done in different settings26,27. This relationship References
of diabetes and occupation is usually determined by the level 1. World Health Organization. Definition, diagnosis and
of physical activity and stress associated with job. In addi- classification of diabetes mellitus and its complications.
tion, our study reflected and even supported the findings of Geneva: WHO; 1999.
other studies that diabetes is no more a disease which is prev-
2. Kinra S, Bowen LJ, Lyngdoh T, Prabhakaran D, Reddy KS,
alent more in people from higher socio-economic class2,24. Ramakrishnan L, et al. Socio-demographic patterning of non-
The present study showed that subjects who were obese communicable disease risk factors in rural India: a cross
(those with BMI >25kg/m2) or were physically inactive had a sectional study. BMJ. 2010;341:c4974.
much higher prevalence of diabetes than non-obese individu- 3. Chuang LM, Tsai ST, Huang BY, Tai TY. The status of diabetes
als. This is not a new finding and many studies have eventu- control in Asia--a cross-sectional survey of 24317 patients with
ally showed quite identical results25,26. Thus, regular meas- diabetes mellitus in 1998. Diabet Med. 2002;19(12):978-985.
urement of BMI levels over a period of time can assist in 4. Narayanappa D, Rajani HS, Mahendrappa KB, Prabhakar AK.
early detection of the potential risk factor and implementation Prevalence of pre-diabetes in school-going children. Indian
of the desired preventive strategies in high-risk groups. Pediatr. 2011;48(4):295-299.
In this cross-sectional study, subjects, who were known to 5. Danaei G, Finucane MM, Lu Y, Singh GM, Cowan MJ,
have higher levels of blood pressure, also had higher preva- Paciorek CJ, et al. National, regional, and global trends in
lence of diabetes among them. This was even observed in fasting plasma glucose and diabetes prevalence since 1980:
another study conducted to estimate the risk factors of diabe- systematic analysis of health examination surveys and
tes, and is probably because of the same patho-physiologic epidemiological studies with 370 country-years and 2.7 million
participants. Lancet. 2011;378(9785):31-40.
mechanism involved in both the onset and the progression of
the disease28. 6. Mathers CD, Loncar D. Projections of global mortality and
burden of disease from 2002 to 2030. PLoS Med.
Furthermore, it was found that subjects who were addict- 2006;3(11):e442.
ed to tobacco (smokeless or smoking form) and/or alcohol
had a definite higher prevalence than their counterparts who 7. World Health Organization. Global status report on non-
communicable diseases-2010. Geneva: WHO; 2011.
were non-addicted. Consumption of tobacco or alcohol was a
significant determinant in another epidemiological study29. 8. Pradeepa R, Mohan V. The changing scenario of the diabetes
However, contrasting results have also been obtained in some epidemic: Implications for India. Indian J Med
other studies11,30. On performing the multivariate analysis, Res. 2002;116:121-132.
positive family history of diabetes aggravated the risk of dia- 9. Majgi SM, Soudarssanane BM, Roy G, Das AK. Prevalence of
betes significantly (OR - 1.96, CI - 1.20-3.23). Positive fami- diabetes mellitus and role of stress in diabetes in rural
ly history has been identified as one of the important deter- Pondicherry – An union territory of India. Global Journal of
minants predicting the onset of the diabetes in different set- Medicine and Public Health. 2012;1(5):40-45.
ting as well11. This again reiterates the importance of obtain- 10. Park K. Textbook of Preventive and Social Medicine. 20th ed.
ing the family history correctly, as it will help the health pro- Jabalpur: Banarsidas Bhanot; 2009.
fessionals to advice people to adopt to lifestyle modification
at an early age. 11. Majgi SM, Soudarssanane BM, Roy G, Das AK. Risk factors of
diabetes mellitus in rural Puducherry. Online J Health Allied
The strength of the present study is that it considered Scs. 2012;11(1):4.
most socio-demographic, lifestyle and anthropometric varia- 12. World Health Organization. Preventing chronic diseases: a vital
bles which provided adjusted association of risk factors with investment. Geneva: WHO press; 2005.
T2DM. In limitation, single contact data was collected for
dietary assessment by 24 hours recall method and family lev- 13. UNESCO. ISCED: International standard classification of
education; 1997 [cited 2014 June, 8]; Available from:
el aggregate information was obtained on vegetable, oil and
http://www.uis.unesco.org/Education/Pages/international-
salt intake. It may have added recall bias to the study results. standard-classification-of-education.aspx
14. Ministry of Home Affairs, India. Census of India 2001 [cited
Conclusions 2014 June, 22]; Available from:
The study has revealed that T2 DM is an important public http://www.censusindia.gov.in/2011-common/CensusData.html
health problem in the adults of rural Pondicherry. Among 15. Agarwal AK. Social classification: the need to update in the
non-modifiable factors, higher age, better socio-educational present scenario. Indian J Community Med. 2008;33(1):50-51.
background and positive family history of diabetes was relat- 16. International Physical Activity Questionnaire. IPAQ scoring
ed to T2DM. Physically inactivity, central obesity, high tri- protocol, 2005. [cited 2014 June, 22]; Available from:
glycerides levels, and raised pulse rate increased the risk of https://sites.google.com/site/theipaq/scoring-protocol
DM significantly. We recommend screening of those with
age above 30 years, positive family history of diabetes and
obesity for ruling out hyperglycemic risk. Those with raised

JRHS 2014; 14(4): 258-263

 Saurabh RamBihariLal Shrivastava et al 263

17. World Health Organization. STEPwise approach to surveillance 24. Kokiwar P, Gupta S, Durge P. Prevalence of diabetes in a rural
(STEPS) field manual appendices. Geneva: WHO Press. [cited area of central India. Int J Diab Developing Countries.
2014 June, 22]; Available from: 2007;27(1):8-10.
http://www.who.int/chp/steps/en/
25. Raghupathy P, Antonisamy B, Fall C, Geethanjali FS, Leary
18. Indian Council of Medical Research. Guidelines for SD, Saperia J, et al. High prevalence of glucose intolerance even
management of type 2 diabetes. New Delhi: ICMR press; 2005. among young adults in south India. Diabetes Res Clin Pract.
2007;77(2):269-279.
19. World Health Organization. BMI Classification. WHO Web
Site; 2014 [cited 2014 June, 22]; Available from: 26. Ramachandran A, Mary S, Yamuna A, Murugesan N,
http://apps.who.int/bmi/index.jsp?introPage=intro_3.html Snehalatha C. High prevalence of diabetes and cardiovascular
risk factors associated with urbanization in India. Diabetes
20. Gupta SK, Singh Z, Purty AJ, Kar M, Vedapriya D, Mahajan P, Care. 2008;31(5):893-898.
et al. Diabetes prevalence and its risk factors in rural area of
Tamil Nadu. Indian J Community Med. 2010;35(3):396-399. 27. Thomas C, Nightingale CM, Donin AS, Rudnicka AR, Owen
CG, Sattar N, et al. Socio-economic position and type 2 diabetes
21. Zabetian A, Sanchez IM, Narayan KM, Hwang CK, Ali MK. risk factors: patterns in UK children of South Asian, black
Global rural diabetes prevalence: A systematic review and meta- African-Caribbean and white European origin. PLoS One.
analysis covering 1990-2012. Diabetes Res Clin Pract. 2012;7(3):e32619.
2014;104(2):206-213.
28. Ferrannini E, Cushman WC. Diabetes and hypertension: the bad
22. Fosse-Edorh S, Fagot-Campagna A, Detournay B, Bihan H, companions. Lancet. 2012;380(9841):601-610.
Gautier A, Dalichampt M, et al. Type 2 diabetes prevalence,
health status and quality of care among the North African 29. Houston TK, Person SD, Pletcher MJ, Liu K, Iribarren C, Kiefe
immigrant population living in France. Diabetes Metab. CI. Active and passive smoking and development of glucose
2014;40(2):143-150. intolerance among young adults in a prospective cohort:
CARDIA study. BMJ. 2006;332(7549):1064-1069.
23. Anjana RM, Ali MK, Pradeepa R, Deepa M, Datta M,
Unnikrishnan R, et al. The need for obtaining accurate 30. Prabhakaran D, Chaturvedi V, Ramakrishnan L, Jeemon P, Shah
nationwide estimates of diabetes prevalence in India - rationale P, Snehi U, et al. Risk factors related to the development of dia-
for a national study on diabetes. Indian J Med Res. betes in men working in a north Indian industry. Natl Med J In-
2011;133(4):369-380. dia. 2007;20(1):4-10.

JRHS 2014; 14(4): 258-263