Abstract

BIRDEM is the largest referral center of diabetes in the world. It registered more than 300,000 diabetic patients from 1956 to 2005. This retrospective study compared the biophysical characteristics of diabetic patients registered in 1995 to those registered in 2005. Information on social (income, education), clinical (height, weight, blood pressure) and oral glucose tolerance (OGTT) of patients registered in 1995 and 2005 were retrieved from the BIRDEM registry. The age group ³ 20y was considered eligible. Overall, there were 11489 patients for 1995 and 19580 for 2005. Compared with the registry of 1995, a significant increase of registry for female patients were observed (39.5 vs. 46.7%, p<0.001) and also the rural population (31.9 vs. 47.4%, p<0.001). Likewise, the number of poor social class was also found higher in 2005 (5.2 vs. 25.5%, p<0.001). Young aged (<40y) registry was also significantly higher in 2005 (34.4 vs. 37.1%, p<0.001). Compared with the registered patients of 1995, adjusted for sex and area, those of 2005 had a significantly higher BMI, higher FPG and higher 2hPG (for all, p<0.001). In contrast, a significantly lower age, lower height and lower blood pressure were observed in those of 2005. We conclude that the age at registration for diabetes has decreased significantly in 2005 compared to that in 1995 indicating an earlier onset of diabetes. Significantly higher obesity in the year 2005 than 1995 indicates that there has been an increase in obesity that might be an important contributing factor for earlier onset of diabetes.

Ibrahim Med. Coll. J. 2008; 2(1): 1-3

Indexing words: Diabetes, age of onset, contributing factors.

Address for Correspondence: Dr. Parvin Akter Khanam, Sr Research Officer, Dept. of Epidemiology & Biostatistics, Research Division, BIRDEM

Introduction

The prevalence of type 2 diabetes mellitus (T2DM) is on the increase worldwide. This increase has also been reported in Bangladesh1-3. Additionally, there are reports indicating an increased number of registrations at BIRDEM, a central referral center of Bangladesh situated in Dhaka. This is known to be the largest referral center of diabetes in the world. Since the establishment of BIRDEM in 1956 the cumulative frequency of registration was 1,185, 26,349 and 221,166 respectively in 1960, 1980 and 20004-6. The registration of diabetic subjects includes social, anthropometric, clinical and biochemical information. It may be noted that more than 1600 diabetic patients visit BIRDEM daily for follow up. This retrospective study compares the biophysical characteristics of diabetic patients registered in 1995 to those registered in 2005.

Materials and Methods

When a patient is suspected of having diabetes, s(he) is usually referred to BIRDEM. The referral system is maintained not only for Dhaka but for almost all of Bangladesh. During registration each patient is interviewed with regard to social and demographic information. The variables are age, sex, family income, education, occupation and area of residence, height and weight and calculated body mass index (BMI). Systolic and diastolic blood pressure is also taken. Two-sample oral glucose tolerance (OGTT, WHO criteria) of each patient is undertaken for confirmation of diabetes. All information of the registered patients is preserved in computer using the SPSS package. These records of the diabetic patients registered in 1995 and 2005 were retrieved for analytical purposes.

Statistical analyses – All the categorical data were expressed in percentages and all the quantitative data were shown in mean with standard deviation. The comparisons were made between rural diabetic men and women of 1995 with those of 2005. Urban men and women were compared with their counterparts. Student’s t-test was applied for quantitative and Chi-sq was used for categorical data. SPSS 11.5 was used for all analyses. All statistical analyses were considered significant at 5% level.

Results

A total of 30,588 diabetic subjects were investigated. 56% were males and 44% were females. Of them, 11671 (M= 7062, F= 4609) diabetic subjects were taken from 1995 and 18914 (M= 10077, F= 8837) were taken from 2005.

Biophysical characteristics between 1995 and 2005: Table 1 shows the comparison of rural men and women of 1995 with their counterparts of 2005. Similar comparisons were made for urban and women of 1995 with those of 2005 (Table 2). Thus, adjusted for sex and area, compared with the registered patients of 1995, those of 2005 had a significantly higher BMI, higher FPG and higher 2hPG (for all, p<0.001). In contrast, significantly lower age, lower height and lower blood pressure were observed in those of 2005.

Table-1: Comparison of characteristics between 1995 and 2005 for rural men (n: 1995/2005= 2093/4344) and women (n: 1995/2005=1232/3590)

Table-2: Comparison of characteristics between 1995 and 2005 for urban men (n: 1995/2005= 4320/4730) and women (n: 1995/2005= 2895/4236)

Table-3: Comparison of variables between subjects registered in 1995 and 2005.

The registration of women increased from 39.5% in 1995 to 46.7% in 2005 (p<0.001) (Table 3). Likewise, the registration from rural diabetics had also increased (31.9 vs. 47.4%, p<0.001). Again, compared with 1995, the registration from poor class increased manifolds in 2005 (5.2 vs. 25.5%, p< 0.001). More interesting finding was that the diabetes registration from lower age (<40y) group had increased in 2005 than in 1995 (37.1 vs. 34.4 %, p<0.001). The prevalence of obesity (BMI>22.0) and insulin treatment at the time of registration was found significantly higher in 2005 than that of 1995.

Discussion

In this analysis it was observed that the total number of registration had increased 160% from 1995 to 2005. This figure is consistent with previous studies, both population based1-3 and registration based4,5. In the past, on or before 1995, more than two-thirds of the diabetic subjects were urban4,5 indicating a higher prevalence in the urban than in the rural population. In contrast, the registration in 2005 showed that the previous proportion was no longer maintained. More and more diabetics were registered from the rural community and it had increased from 31.9% in 1995 to 47.4% in 2005. It appears that there was either an increased awareness or increased prevalence in the rural communities that lead to an increased referral to BIRDEM. The alarming finding was that there had been an increasing number of registrations from the younger group. This indicates that the onset of diabetes probably starts earlier in recent years than in the past. Obesity has also increased, which may be a contributory factor for early development of diabetes in Bangladesh. This finding is consistent with other studies9-11. These studies observed that obesity starts in childhood and adolescence that prompts development of diabetes at an earlier age. Possibly, Bangladesh is experiencing the same transition. The obesity starts early and the abnormal fat distribution in younger age contribute to develop diabetes or glucose intolerance at a much younger age. However, this is a speculation. A population based study in the younger age group may help explain these findings.

The registration from rural areas has also increased in 2005. It was reported that severe glycemia and proteinuria were registered from the rural population6. As already discussed, very few diabetic patients were registered at BIRDEM from rural community although 70% of Bangladeshis live in the rural area7. However, the policy makers should keep in mind that there is a possibility of undetected diabetes cases in the rural community and their number is not negligible.

Conclusions

In conclusion, compared with 1995 the number of registration has increased in 2005. The registration from female and rural population and from poor class has increased in 2005 than seen in 1995. More interesting finding is that the diabetes registration from lower age group (age<40y) has increased in 2005. Significantly higher obesity in the year 2005 than 1995 indicates that there has been an increasing obesity that might be an important contributing factor for an earlier onset of diabetes.

References

1.  Sayeed MA, Banu A, Khanam PA, Mahtab H, Azad Khan AK. Prevalence of hypertension in Bangladesh: Effect of socioeconomic risk factors on difference between rural and urban community. Bang Med Res Counc Bull 2002; 28(1): 7-18.

2.  Sayeed MA, Mahtab H, Khanam PA, Latif ZA, Ali SMK, Banu A, Ahren B, Azad Khan AK. Diabetes and impaired fasting glycemia in a rural population of Bangladesh. Diabetes Care 2003; 26(4): 1035-1039.

3.  Sayeed MA, Mahtab H, Khanam PA, Ahsan KA, Banu A, Bazlur Rashid ANM and Azad Khan AK. Diabetes and impaired fasting glycemia in the tribes of the Khagrachari Hill Tracts of Bangladesh. Diabetes Care 2004; 27(5): 1054-1059.

4.  Azad Khan A K , Mahtab H, Latif Z A, Hussain M Z. Characteristics of diabetics registered at BIRDEM 1981. J Diab Assoc Bang 1992; 20(2): 36-48.

5.  Sayeed M A, Hussain M Z, Islam M A, Azad Khan A K. Characteristics of the diabetic subjects: BIRDEM diabetes registry, 1984. J Diab Assoc Bang 1994; 22(1): 8-20.

6.  Sayeed MA, Khanam PA, Hossain MZ, Mahtab H and Azad Khan A K. Diabetic subjects with severe hyperglycemia and proteinuria are more prevalent in rural than urban and in poor than rich people. Diab Endocr J 1999; 27(2): 35-40.

7.  Yusuf F H (Ed.). Statistical pocket book of Bangladesh 2004. Dhaka, Bureau of Statistics, Statistical Division, Ministry of Planning, Government of the People’s Republic of Bangladesh, 1984-5.

8.  Kriska AM, Lporte RE, Pettitt DJ, Charles MA, Nelson RG, Kuller LH, Bennett PH, Knowler WC. The association of physical activity with obesity, fat distribution and glucose intolerance in Pima Indians. Diabetologia 1993; 36: 863-869.

9.  Harris KM, Gordon-Larsen P, Chantala K, Udry JR. Longitudinal trends in race/ethnic disparities in leading health indicators from adolescence to young adulthood. Arch Pediatr Adolesc Med 2006; 160(1): 74-81.

10.Rappaport EB, Robbins JM. Overweight in Southeastern Pennsylvania children: 2002 household health survey data. Public Health Rep 2005; 120(5): 525-31.