The European Journal of Public Health Advance Access originally published online on September 28, 2006
The European Journal of Public Health 2007 17(3):291-296; doi:10.1093/eurpub/ckl235
Infectious and allergic diseases, cancer, and diabetes |
Type 2 diabetes and impaired fasting blood glucose in rural Bangladesh: a population-based study
Akhtar Hussain1, Stein Vaaler2, M. A. Sayeed3, Hajera Mahtab3, S. M. Keramat Ali4 and A. K. Azad Khan3
1 Institute of General Practice and Community Medicine, Department of International Health, University of Oslo Norway
2 Centre for Clinical Epidemiology, National Hospital/University of Oslo Norway
3 Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM) Dhaka, Bangladesh
4 Institute of Nutrition and Food Sciences, University of Dhaka Bangladesh
Correspondence: Dr Akhtar Hussain, Department of International Health, Faculty of Medicine, University of Oslo, P.O. Box 1130 Blindern, 0317 Oslo, Norway, tel.: +47 22 850641, fax: +47 22 850501, e-mail: akhtar.hussain{at}medisin.uio.no
Received July 5, 2005, accepted August 28, 2006
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionsReferences |
|---|
Background: Diabetes is a fast expanding global health problem but more so in the developing countries. Therefore, it is of particular interest to study the epidemiological transition of the state and to identify the risk factors in order to recognize the extent of the problem.
Methods: A random sample of 5000 rural individuals (age 
20 years) were included in a cross-sectional study. Fasting capillary blood glucose levels were measured from 4757 individuals. Height, weight, waist, hips including blood pressure and demographic information was collected.
Results: The study population was lean [mean body mass index (BMI) 19.4] with a low prevalence of type 2 diabetes but relatively high impaired fasting glucose. No relationship between type 2 diabetes and BMI in men, but an overall relationship was observed for women (P = 0.04) (data not shown). Age, sex, and waist/hip ratio appeared to be important risk factors for the occurrence of type 2 diabetes in this population.
Conclusions: Low prevalence of type 2 diabetes and relative high impaired fasting blood glucose was observed. The factors associated with the occurrence of diabetes in this population appeared to differ than its known relations with BMI. This may indicate that the risk factors for type 2 diabetes are likely to differ in different population. Our results are likely to be in line with the Indian data suggesting that a revised guideline for anthropometric measures in the South Asian population is called for, in order to classify people at risk.
Keywords: Bangladesh, body mass index, obesity, type 2 diabetes, waist/hip ratio
The WHO report on diabetes prevalence alarmed that diabetes has posed a serious threat to developing countries with respect to their existing health care services.1 Further, the prevalence of diabetes is predicted to increase dramatically over the next 25 years, mostly as a result of type 2 diabetes.2
Diabetes, insulin resistance hyperinsulinaemia, and other coronary risk factors are more prevalent in Bangladeshis compared with other South Asian migrants (Indian, Pakistani) settled in United Kingdom.3,4 It has also been reported that Bangladeshis among all other South Asian immigrants had highest morbidity and mortality from CHD in the United Kingdom.5 Higher prevalence of glucose intolerance and hypertension were also shown in a number of small epidemiological studies in Bangladesh.6–10
Obesity is a known risk factor for the development of type 2 diabetes. However, obesity as measured by body mass index (BMI) and its association with type 2 diabetes varied in different ethnic groups, possibly as a consequence of different body stature.11 Epidemiological studies have suggested that genetic factors and central obesity as measured by waist/hip ratio (WHR) is a major contributing factor to insulin resistance and is associated with diabetes, hypertension, dislipidaemia [high triglycerides (TG) and low high-density lipoprotein-c (HDL-c)].12–16
The prevalence of type 2 diabetes has shown to vary in different population probably as a consequence of food habits and obesity. Fat deposition is shown to vary in different population as a consequence of different body stature and lifestyle. A more promising intermediate trait is abdominal adiposity and its association with metabolic disturbances including resistance to insulin. Large-scale population-based studies to identify the extent of the problem and its associated risk factors are scarce. The study was conducted to estimate the prevalence of type 2 diabetes and impaired fasting glucose (IFG) along with potential risk factors in a rural Bangladeshi population.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionsReferences |
|---|
Selection of study area
The study areas were selected from a rural community
35 miles north of Dhaka city between Dhaka and Tangail. The areas may still be characterized as rural but due to fast expanding urbanization the localities will transfer in to semi-urban areas in a short while. Rationality for choosing the areas is to observe the transition of the disease as a consequence of changed lifestyle. Rural areas were included as 80% of the total people in Bangladesh live in the countryside.17 Characteristics of rural life were defined to reflect the livelihood primarily related to agriculture or related to agrarian activities like ploughing, fishing, pottering, etc. The dwellers of the area did not have access to the municipal (urban) facilities like housing provided with water, gas, and sanitation. Accordingly, 5 areas with 10 villages were selected.
Selection of study subjects
There were 20 171 individuals and subjects 20 years of age, identified by a census conducted prior to the commencement of the survey. The census was conducted by the Bangladesh Bureau of statistics to update the demographic characteristics of the locality which is done every second year. All the individuals were given an identification number including a household number by the programme administration. Among those 5000 individuals, subjects were selected following simple random procedure with the presumption that the combined prevalence of diabetes and IFG will be close to 15%. Of the selected subjects 4757 individuals agreed to participate in the study.
Ethics
Verbal consent was received from each of the individuals prior to inclusion, as majority of the participants were illiterates. They were informed of their rights to withdraw from the study at any stage or to restrict their data from the analysis. Identified cases for (type 2 diabetes mellitus) T2DM were referred to the diabetic hospital (BIRDEM) for follow-up and treatment. The protocol was approved by the Norwegian and Bangladeshi ethical committee for medical research.
Survey procedures
The community leaders were invited for a meeting with the project team. They were oriented of the purpose of the study and requested for their opinion and/or comments. Their co-operation was sought in a participatory manner. Each of them were given specific tasks (organizing, collecting voter's list, co-ordination with the field team, and feedback to the programme supervisors) based on their background and interest.
Sixteen volunteers were recruited in four teams from the local community and trained by the programme managers. In addition, four physicians were employed to supervise and to measure blood pressure (BP). Each team investigated around 30 subjects a day. Residents were informed of the objectives of the study including their approval by the locally recruited volunteers. Further, every one was made aware of the fasting state of a minimum of 8 h prior to the test. The investigating team moved from village to village. The identified people were reminded of the importance of the fasting state prior to the day of investigation and verbal confirmation was made prior to blood test. Three days of training (both theoretical and field) for the project workers were conducted prior to the commencement of the programme. The census and the population were defined during the months of September–October 1999. Fasting blood glucose (FBG and biophysical examination was conducted for the selected 5000 people during the months of November 1999 to January 2000).
Anthropometry and measurement of BP. Measurements for height, weight, waist, and hip are taken with light clothes without shoes. The weighing tools (Salter 918 Electronic) were calibrated daily by known standard weight (10 kg). For height, the subject stood in erect posture vertically touching occiput, back, buttocks, and heels on the wall gazing horizontally in front keeping tragus and lateral orbital margin in the same horizontal plane. Waist girth was measured by placing a plastic tape horizontally midway between 12th rib and iliac crest on the mid-axillary line. Hip was measured horizontally on the greater trochanters.
Measurement of BP needs special precaution. Variation in BP was minimized by (i) ensuring 10 min rest before BP record, (ii) using standard cuffs for adults fitted with standard mercury sphygmomanometer, and (iii) placing the stethoscope bell lightly over the pulsatile brachial artery on the right hand. The physicians were particularly trained to record BP. All the four physicians had 10 years of experience in working with diabetic patients at the hospital.
Blood glucose estimation. FBG from capillary whole blood was performed from 4757 individuals following the newly proposed diagnostic criteria.18 FBG > 6.1 mml/l (>180 mg/100 ml) and were used to classify diabetic cases and >5.6–6.0 mml/l (>100 mg/ml) to identify IFG. The estimation was performed by the HEMOCUE glucose analyser in the field. The machine was calibred everyday with the calibration cuvette prior to estimations. The microcuvettes were stored in a refrigerator in the field and ice bags were used during transport of the cuvettes. Open packs were used within 3 weeks. The sensitivity and specificity of the HemoCue glucose analyser was reported in previous studies.19
Data analysis and statistical methods
The data was registered using Microsoft Access data entry. Control of data entry was secured through both programme appliance and manually. The prevalence rates of DM were determined by simple percentages. These rates were further standardized for the ‘New World Population’ following WHO suggestions in order to make a uniform prevalence data.20 The odds ratio (OR) with 95% confidence interval (CI) for risk factor(s) were calculated taking the least prevalence of complication or clinically relevant criteria as a reference value. All P-values presented are two-tailed. Multiple logistic regression was executed to adjust for potential confounding factors. All statistical analyses were performed using SPSS 9.0 software.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionsReferences |
|---|
Diabetes prevalence increased with increasing age both for males and females (table 1). The standardized rate was also provided in parenthesis for the total number and prevalence following the "New World Population" proposed by the WHO. Though non-significant, females had higher prevalence of diabetes in all age groups compared with males. The difference in prevalence by sex widened in the older age group (>50 years). The same was true for IFG but the difference by sex was less prominent. Only six people were reported to have known diabetes but none of them were under medication.
|
Characteristics of the participants were presented in (table 2) with 10 years age interval. Male subjects were older compared to the female participants but with almost no differences in BMI. However, the total population was lean with a mean BMI of 19.3 and 19.4 for the male and female population, respectively. This picture is also reflected in the assessment of BP.
|
Elevated levels of FBG were observed with growing age both for males and females (figure 1). The difference in glucose values was somewhat similar for both sexes until the age of 50 after which the glucose values increase notably for women. The differences were not statistically significant.
|
Age, sex, systolic BP, and WHR for men showed to be important risk factors for the occurrence of type 2 diabetes (table 3). The findings were also evident in the multivariate model adjusted for age, sex, BMI, diastolic, and systolic BP. The risk for diabetes was almost 2-fold higher in subjects aged >40 compared with the age group 20–30, systolic BP >140 mm Hg, and WHR for men. An over all significant association of BMI and type 2 diabetes (P = 0.04) was also observed for females (data not shown). BMI >30 showed to be exceedingly risky state for the occurrence of diabetes but with only 11 subjects in the group the apparent statistical significance was not observed in the multivariate model. WHR (>0.9) were found to be significantly associated with the diabetic state in men both in the univariate and multivariate model adjusted for age, systolic, and diastolic BP and BMI.
|
|
Discussions |
|---|
|
TopAbstractMethodsResultsDiscussionsReferences |
|---|
We have observed a low (2.8%) prevalence of type 2 diabetes with a relatively high IFG in the population aged 20 years and above. This may indicate a possibility for exponential increase of type 2 diabetes in this population unless necessary measures are taken for the specified group. Type 2 diabetes were found in 7% of the adult population (>25 years of age) and 10–50% in the minority communities including Asians in the United States.21 The rate was found to be 14.8 in a population aged >20 years in Kuwait.22 In a random survey of 1195 subjects aged >40 years in Shimla, India, the prevalence of type 2 diabetes were shown to be 4.9%.23 Among other rural population in India the prevalence of type 2 diabetes was found to be 2.8%.24 The prevalence of type 2 diabetes was found to be 3.8% in rural Bangladesh among subjects in the age category 30–64 years10 and the same in a recent study in Bangladesh among 30–70 years.25 Our results are in agreement with the Indian rural population but show somewhat lower prevalence compared to other Bangladeshi studies10 probably because of the selection of the study population in the other study were from an adjacent location of Dhaka city. Our population was also leaner. However, other published data from Bangladesh25 may also indicate varied prevalence of diabetes within the country. Besides, the results may also have been influenced by the diagnostic tool used for FBG estimation. The other studies in Bangladesh have employed autoanalyser where as our estimation was performed by HemoCue glucose analyser. The sensitivity and specificity of HemocCue glucose analyser was mentioned earlier.19 Further, the most recent HemoCue glucose meter measures B-glucose and convert the result to a plasma equivalent glucose concentration and was found to be satisfactory for diagnostic determination.26 The non-responders were only 4.9% (243) is not likely to explain the lower prevalence observed.
We have observed a higher, though non-significant, occurrence of the disease among females in all age categories compared with males especially in the highest age strata. We have observed higher mean FBG among women aged more than 50 years compared to men. Finding of the pre-dominancy of the disease by sex was not consistent in the previous studies. European investigations have shown a higher prevalence of type 2 diabetes in males compared to females.27 Non-significant higher prevalence of type 2 diabetes was also observed in women (2.2% compared with 1.6% in men) among Mexican Indians.28 The exact mechanism for this finding is not well understood but a selection process in the phases of female survival in a society where female mortality is higher may have influenced our findings.
Systolic hyper tension (sHTN > 140 mm Hg) was associated with the occurrence of diabetes in our population both in crude and adjusted analysis. Another study in Bangladesh found that the rate for sHTN and dHTN as 23.2 and 13.6%, respectively, among the newly diagnosed type 2 diabetic patients.8 Observed higher hypertension in the mentioned study compared with ours may have been due to selection of samples. The participants were selected among the newly diagnosed diabetic type 2 cases not a population-based study. Systolic hyper tension was found to be associated with the prevalence of diabetes in southern Taiwan,29 Nigeria,30 China,31 and Australia.32
Although generalized obesity appeared to ensue type 2 diabetes, we have not observed any association of overweight (as defined by the standard BMI criterions)33 and diabetes in our material. Our samples appeared to exemplify a lean population with only 3.6% people defined as overweight, while 25% of the men and 59% of the women had higher WHR. The association with BMI and type 2 diabetes appeared to differ in different ethnic groups. Epidemiological data from Asian Indians (AI), and Mexican Americans (MA), and non-Hispanic Whites (NHW) from San Antonio heart study showed that MA had the highest rate of obesity and highest prevalence of diabetes (men 19.6%; women 11.8%). NHW had also high rates of obesity but a low prevalence of diabetes (men 4.4%; women 5.7%). Although AI had lower BMI than MA, the risk conferred by BMI was similarly high in AI and MA than NHW.11
WHR ratio appeared to be significantly associated with the occurrence of diabetes in men but not in women. In a review of 59 references it was found that average waist/hip circumference ratios are higher in South Asians than in Europeans of similar BMI.34 Previous study in Bangladesh also showed that the prevalence of diabetes was related to WHR.35 We have observed central obesity even among people with normal BMI in our population (data not shown).
Measures applied to define obesity in relation to the occurrence of type 2 diabetes in different population appeared to provide inconsistent results. Individuals with a predominance of abdominal fat exhibit numerous metabolic disturbances, including insulin resistance compared with those having fat primarily distributed subcutaneously over the lower extremities.36 Our data showed low BMI and relatively high central obesity with low prevalence of type 2 diabetes but rather high fasting impaired blood glucose level may indicate genetically predetermined fat deposition. Further this may also indicate that so-called ‘lean diabetes’ may be explained by higher genetic risk factor and a low calorie intake in this population. This may call for a new guide line for anthropometric measures in South Asian population in line with the Indian data37 suggesting different cut-off values in order to classify people at risk. More studies are needed on metabolic, hormonal, and biophysical profiles in order to understand the transitional epidemiology of lean diabetes.
|
Acknowledgments |
|---|
We acknowledge the contribution of our field medical officer Dr Shamim Talukder, the village leaders, and the volunteers for their sincere and enduring contribution to the collection and quality of data. We express our gratitude to the Division of Medical Statistics including Mr Ishtiaq Khushi (Computer Manager), Centre for Clinical Epidemiology, National Hospital, Norway, for their input in the statistical analysis. We also thank all the participants in the study for their active co-operation. Finally, our gratitude to Prof. Emiratus Jak Jervell for his continued support.
| Key points
|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionsReferences |
|---|
1 King H, Rewers M. Global estimates for prevalence of diabetes mellitus and impaired glucose tolerance in adults. Diabetes Care (1993) 16:157–77.[Abstract]
2 King H, Aubert RE, Herman WH. Global burden of diabetes 1995–2025. Prevalence, numerical estimates and projections. Diabetes care (1998) 21:1414–31.[Abstract]
3 McKeigue PM, Miller GJ, Marmot MG. Coronary heart disease in South Asians overseas—a review. J Clin Epidemiol (1989) 42:597–609.[CrossRef][Web of Science][Medline]
4 McKeigue PM, Bela Shah, Marmot MG. Relation of central obesity and insulin resistance with high diabetes prevalence and cardiovascular risk in South Asians. Lancet (1991) 337:382–6.[CrossRef][Web of Science][Medline]
5 McKeigue PM, Pierpoint T, Ferrie JE, Marmot MG. Relationship of glucose intolerance and hyperinsulinemia to body fat pattern in South Asians and Europeans. Diabetologia (1992) 35:785–91.[Web of Science][Medline]
6 West KM, Kalbfleisch JM. Glucose tolerance, nutrition and diabetes in Uruguay, Venezuela, Malaya and East Pakistan. Diabetes (1966) 15:9–18.[Web of Science][Medline]
7 Mahtab H, Ibrahim M, Banik NG, et al. Diabetes detection survey in a rural and semiurban community in Bangladesh. Tohoku J Exp Med (1983) 141:211–7.[Web of Science][Medline]
8 Sayeed MA, Khan AR, Banu A, Hussain MZ. Prevalence of diabetes and hypertension in a rural population of Bangladesh. Diabetes Care (1995) 18:555–8.[Abstract]
9 Sayeed MA, Hussain MZ, Banu A, et al. Prevalence of diabetes in a suburban population of Bangladesh. Diabetes Res Clin Pract (1997) 34:149–55.[CrossRef][Web of Science][Medline]
10 Sayeed MA, Hussain MZ, Banu A, et al. Effect of socio-economic risk factor on difference between rural and urban in the prevalence of diabetes in Bangladesh. Diabetes Care (1997) 20:551–5.[Abstract]
11 Ramachandran A, Snehalatha C, Viswanathan V, et al. Risk of noninsulin dependent diabetes mellitus conferred by obesity and central adiposity in different ethnic groups: a comparative analysis between Asian Indians, Mexican Americans and Whites. Diabetes Res Clin Pract (1997) 36:121–5.[CrossRef][Web of Science][Medline]
12 Zavaroni I, Dall'Aglio E, Alpi O, et al. Evidence for an independent relationship between plasma insulin and concentration of high density lipoprotein cholesterol and triglyceride. Atherosclerosis (1985) 55:259–66.[CrossRef][Web of Science][Medline]
13 Orchard TJ, Becker DJ, Bates M, et al. Plasma insulin and lipoprotein concentrations: an atherogenic association? Am J Epidemiol (1983) 118:326–37.
14 Cambien F, Warnet M, Eschwège E, et al. Body mass, blood pressure, glucose and lipids. Does plasma insulin explain their relationship? Arteriosclerosis (1987) 7:197–202.
15 Modan M, Halkin H, Almog S, et al. Hyperinsulinemia: a link between hypertension, obesity and glucose intolerance. J Clin Invest (1985) 75:809–17.[Web of Science][Medline]
16 Reaven GM, Greenfield MS. Diabetes hyperglyceridemia. Evidence for three clinical syndromes. Diabetes (1981) 30:66–75.[Web of Science][Medline]
17 Islam W. Bangladesh Bureau of Statistics. In: Statistical Pocket Book of Bangladesh 1997 Ed. Bangladesh: Statistical Division, Ministry of planning, Government of the People's Republic of Bangladesh.
18 Alberti KGMM, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional report of a WHO consultation. Diabet Med (1998) 15:539–53.[CrossRef][Web of Science][Medline]
19 Hannestad U, Lunsblad A. Accurate and precise isotope dilution mass spectrometry method for determining glucose in whole blood. Clin Chem (1997) 43:794–800.
20 Ahmad OB, Boschi-Pinto C, Lopez AD, et al. Age standardization of rates: a new WHO standard. EPI/GPE/EBD. Series No.31. Geneva: WHO.
21 Florez H. Steps toward the primary prevention of type II diabetes mellitus. Various epidemiological considerations. Invest Clin (1997) 38:39–52.[Medline]
22 Abdella N, Al Arouj M, Al Nakhi A, et al. Non-insulin dependent diabetes in Kuwait: prevalence rates and associated risk factors. Diabetes Res Clin Pract (1998) 42:187–96.[CrossRef][Web of Science][Medline]
23 Dhadwal D, Ahluwalia SK, Das Gupta DJ, et al. Prevalence of NIDDM in the general population (<40 years) in Shimla. Indian J Med Sci (1997) 51:459–64.[Medline]
24 Sing RB, Bajaj S, Niaz MA, et al. Prevalence of type 2 diabetes mellitus and risk of hypertension and coronary artery disease in rural and urban population with low rates of obesity. Int J Cardiol (1998) 66:65–72.[CrossRef][Web of Science][Medline]
25 Sayeed MA, Mahtab H, Khanam PA, et al. Diabetes and impaired fasting glycemia in a rural population of Bangladesh. Diabetes care (2003) 26:1034–9.
26 Niels Fogh. Evaluation of HemoCue glucose meter (201+) converting B-glucose to P-glucose. Point of Care. (2004) 3:172–5.
27 Gatling W, Budd S, Walters D, et al. Evidence of an increasing prevalence of diagnosed diabetes mellitus in the Poole area from 1983 to 1996. Diabet Med (1998) 15:1015–21.[CrossRef][Web of Science][Medline]
28 Castro-Sanchez H, Escobedo-de la Pena J. Prevalence of non insulin dependent diabetes mellitus and associated risk factors in the Mazatec population of the State of Oaxaca, Mexico. Gac Med Mex (1997) 133:527–34.[Medline]
29 Lu FH, Yang YC, Wu JS, et al. A population-based study of the prevalence and associated factors of diabetes mellitus in southern Taiwan. Diabet Med (1998) 15:564–72.[CrossRef][Web of Science][Medline]
30 Olatunbosun ST, Ojo PO, Fineberg NS, Bella AF. Prevalence of diabetes mellitus and impaired glucose tolerance in a group of urban adults in Nigeria. J Natl Med Assoc (1998) 90:293–301.[Medline]
31 Pan XR, Yang WY, Li GW, Liu J. Prevalence of diabetes and its risk factors in China, 1994. National Diabetes Prevention and Control Cooperative Group. Diabetes Care (1997) 20:1664–9.[Abstract]
32 Donnelly R, Molyneaux L, McGill M, Yue DK. Detection and treatment of hyper tension with non-insulin-dependent diabetes mellitus: does the "rule of halves" apply to a diabetic population? Diabetes Res Clin Pract (1997) 37:35–40.[CrossRef][Web of Science][Medline]
33 Shils ME, Olson JA, Shike M, eds. Modern nutrition in health and disease, Vol 1, 8th edition. Philadelphia: Lea & Febiger, 1994.
34 McKeigue PM. Metabolic consequences of obesity and body fat pattern: lessons from migrant studies. Ciba Found Symp (1996) 201:54–64.[Medline]
35 Sayeed MA, Banu A, Malek MA, Khan AK. Blood pressure and coronary heart disease in NIDDM subjects at diagnosis: prevalence and risks in Bangladeshi population. Diabetes Res Clin Pract (1998) 39:147–55.[CrossRef][Web of Science][Medline]
36 Mohan V, Vijayaprabha R, Rema M, et al. Clinical profile of lean NIDDM of south India. Diabetes Res Clin Pract (1997) 38:149–50.
37 Snehalatha C, Viswanathan V, Ramachandran A. Cutoff values for normal anthropometric variables in Asian Indian adults. Diabetes Care (2003) 26:1380–84.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||