|Year : 2011 | Volume
| Issue : 1 | Page : 38-42
Demographic, breast-feeding, and nutritional trends among children with type 1 diabetes mellitus
Manash P Baruah1, Ariachery C Ammini1, Madan L Khurana2
1 Department of Endocrinology, Excel center, Ulubari, Guwahati, Assam-781 007, India
2 Department of Endocrinology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi-110 011, India
|Date of Web Publication||12-Mar-2011|
Manash P Baruah
Consultant Endocrinologist Excel Center, Maya Ville, Barthakur Mill Road, Ulubari, Guwahati - 781 007
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: The pathogenesis of type 1 diabetes mellitus (T1DM) requires a genetic predisposition to particular environmental triggers that may activate mechanisms leading to progressive loss of pancreatic beta cells. Aims: We tried to compare the impact of some demographic and environmental factors and breast-feeding on children (aged < 18 years) with recent onset diabetes mellitus (≤1 year) with that on age, sex, and socioeconomic status-matched controls. Material and Methods: A total of 43 consecutive patients (male, 24, mean age ± SD = 12.58 ± 9.6 years) and equal number of controls without diabetes mellitus or dysglycemia were included in this hospital-based case-control study. Results and Conclusions: A distinct peak in the incidence noted in the early adolescence with segregation in the winter months. Our patients did not differ significantly from the controls with regard to birth order, mode of delivery, parental age, parental education, dietary practices, breast-feeding, and migration in the family. Growth characteristics and nutritional status were also similar. A population study with more power will be better equipped to answer such queries.
Keywords: Type 1 diabetes mellitus demography, type 1 diabetes mellitus etiology, breast-feeding, diabetes
|How to cite this article:|
Baruah MP, Ammini AC, Khurana ML. Demographic, breast-feeding, and nutritional trends among children with type 1 diabetes mellitus. Indian J Endocr Metab 2011;15:38-42
|How to cite this URL:|
Baruah MP, Ammini AC, Khurana ML. Demographic, breast-feeding, and nutritional trends among children with type 1 diabetes mellitus. Indian J Endocr Metab [serial online] 2011 [cited 2019 Jan 22];15:38-42. Available from: http://www.ijem.in/text.asp?2011/15/1/38/77583
| Introduction|| |
The pathogenesis of type 1 diabetes mellitus (T1DM) requires a genetic predisposition to particular environmental triggers that may activate mechanisms leading to progressive loss of pancreatic beta cells. The polygenic nature T1DM and the influence of nongenetically determined factors in its causation are supported by data from studies using animal models and also from human studies. 
Among the various socioeconomic denominators, a tendency toward shorter education and more practical type of employment was seen among parents of diabetic children from Europe; however, a South Indian study found that higher rather than lower socioeconomic status had influenced the absolute incidence and early onset of T1DM. ,, Strong clustering of T1DM in young children is thought to be an indication of in utero exposure to infection or other environmental agents.  The precise impacts of migration on the incidence of T1DM still not clear. Some early reports showed a different childhood T1DM incidence in France (4.6/100,000/year) and Canada (8.2/100,000/year), suggesting that those moving from low-risk areas to a high-risk area tend to adopt the incidence rate of that community.  Similarly, the incidence among Asian communities living in Bradford, Yorkshire, UK, during 1978-1981 was 4 times lower than non-Asians. However, it rose steadily during the 1981-1990 period finally leading to similar incidence among the children of non-Asian parents and the first generation of Asian children born in the UK.  Studies conducted in various European countries have found a strong association of advanced maternal or paternal age at birth to the risk of T1DM in the later life of the offspring. ,, A South Indian study, however, found evidence to the contrary.  Etiologic importance of breast-feeding and pattern of weaning in T1DM has been studied extensively. The protective effect of breast-feeding in T1DM has been reported in number of studies. Diabetogenic effect of cow's milk in genetically susceptible animals, such as biobreeding rats and nonobese diabetic mouse have been observed by different workers.  A molecular mimicry between bovine -casein and glucose transport (GLUT2) peptide has been suggested. It is believed that early exposure to an ABBOS peptide present in bovine serum was responsible for autoimmune response directed against islet cells due to sequence homology to the islet antigen (ICA69) peptide.  Studies conducted among English and Scottish children showed that higher number of diabetic children discontinued breast-feeding earlier compared with normal children  ; however, an epidemiologic study from Sweden found that the population of children who were breast-fed was similar in diabetic and nondiabetic referral group.  The German BABYDIAB  and diabetes autoimmunity study in the young (DAISY),  both well designed for risk evaluation, did not find any association.
Seasonal onset of T1DM has also been highlighted in many studies indicating a link of etiopathogenesis of the disease with some environmental or even viral trigger. However, the findings are inconsistent regarding the dominance of a particular season over another. ,,,, There are insufficient data as regards the preponderance of a particular sex among T1DM patients. ,,,
Thus a review of literature available worldwide reveals conflicting data with regard to the influence of the different demographic, socioeconomic, and environmental issues on the etiopathogenesis of T1DM. Although there are only few studies from South India, there is no data on north Indian population in the available literature. Hence we planned to carry out this preliminary study among hospitalized T1DM patients at least to gather the basic demographic and socioeconomic data and analyze whether such factors actually influence the etiopathogenesis of T1DM.
| Material and Methods|| |
This was a tertiary care hospital-based, noninterventional, case-control study, spreading over a period of 20 months starting from May 2000. Consecutive diabetic patients were included who fulfilled the following criteria: (i) aged 18 years or less, (ii) diagnosed with diabetes mellitus within preceding 12 months, (iii) had unequivocal history of diabetic keto-acidosis or documentation of spontaneous ketonuria, and (iv) were treated in Endocrinology ward of the institute. Control group comprised children who were not found to have diabetes mellitus or any other metabolic, endocrine, or autoimmune disease. They were recruited in equal proportion to cases (1:1) from (i) different wards of the institute and (ii) among family members of medical and paramedical professionals. The subjects in both the groups were matched for age, sex, and socioeconomic class.
The reference subject (patient or control) and family member(s) inclusive of at least one parent were interviewed extensively. Special emphasis was put on information regarding antenatal and perinatal events, birth order, parental age, breast-feeding, weaning, animal milk exposure, feeding habits of the family (milk, sugar, fats, processed food, and others), family history of diabetes and other autoimmune disease, history of migration, including previous three generations, parents' educational, professional, and economic backgrounds and categorization using standard criteria.  Specific questions were asked with regard to age and season of onset of diabetes, type of presentation, episodes of keto-acidosis, and presence/absence of any precipitating factors, type of treatment received, additional problems, such as exocrine pancreatic defects and others. Baseline laboratory investigations, such as fasting blood glucose, glycated hemoglobin, triglycerides, and cholesterol were performed. Pancreatic imaging included a plain skiagram and ultrasound to rule out calcifications. The study was approved by the institutional ethics committee.
Descriptive statistics, that is, mean, standard deviation, and frequency distribution were calculated for each and every variable wherever applicable. To see the significant difference between the two groups for the continuous variable, we had applied Student's t test (unpaired). To see the association between the two groups for the categorical variables we had applied Chi-square test. P value of 0.05 or less was considered as statistically significant.
| Results|| |
Forty-three subjects were enrolled in each group. The study group consisted of 24 males and 19 females and the control group had 25 males and 18 females. Baseline characteristics, such as age, body mass index, and socioeconomic status were similar in both the groups [Table 1].
|Table 1: Baseline demographic and socioeconomic characteristics of patients and control subjects|
Click here to view
No significant difference existed between the patients and control (P = 0.326) with regard to the history of migration in the family [Table 1]. Similarly, place of residence also did not differ (P = 0.329) significantly between the two groups [Table 1]. There appeared to be a slight positive association of urban dwelling and previous migration (mostly rural to urban), which however was not significant due to small sample sizes.
Subjects were divided into 5 groups depending on the level of education received by at least one parent, namely, illiterate, primary, high school, college, and university. No significant difference was observed in the distribution of patient or control subjects at any level (P = 0.579) [Table 1]. Children born to illiterate parents were at an equal risk of developing diabetes mellitus compared with those who were born to highly educated parents.
Age of either parent at of the time of birth of the studied patients did not differ significantly from that of the control subjects [Table 1]. Mothers in either group were mostly between 21 and 25 years of age at the birth of the reference subjects. Children born to mothers above 35 years of age were slightly more at risk of developing diabetes in the later age.
Majority of children in both the groups belonged to either first or second order of birth. However, there was slight preponderance of early birth order in the control group, which was of borderline significance only (P = 0.06). Children belonging to the first birth order were relatively at a lower risk of developing diabetes compared with those belonging to the fifth birth order. There was insignificant (P = 0.825) difference between the 2 groups with regard to the mode of delivery, that is, institutional vs noninstitutional.
Duration (mean ± SD) of exclusive breast-feeding was comparable between the patients (5 ± 4 months) and control (6.6 ± 4 months). However, the duration (mean ± SD) of total breast-feeding (breast milk plus top feed) of the control group (18.6 ± 10.2 months) was significantly more (P < 0.05) than that of the patients (13 ± 7.5 months) [Table 2]. There was no significant difference between the 2 groups with regard to the age of initiation and duration of weaning [Table 2]. Both the groups underwent a rather prolonged course of weaning, 9 ± 4.7 months in patient group and 8.3 ± 3.2 months in the control group. The age at which the subject was fed with animal milk for the first time did not differ significantly between the 2 groups. Mean age of such exposure in the patient group was 6 ± 5 months and that for controls was 6.7 ± 6 months. Those children who were exposed to any type of animal milk prior to the age of 6 months were 1.6 times more at risk of developing diabetes compared with those who were not. The consumption of sugar, fats, and processed food in the families of patients did not significantly differ from that of the families of control subjects [Table 2].
|Table 2: Details of breast-feeding and consumption of processed food among the patients and control|
Click here to view
The prevalence across different age groups showed a distinct peak at early adolescence (around 12 years of age), irrespective of sex [Figure 1]. Reliable data were obtained regarding the exact month of onset of symptoms in 33 patients. Frequency distribution analysis did not show a significant increase of incidence in any particular month or season of the year; however, most of the cases occurred during autumn and winter [Figure 2].
|Figure 1: Agewise distribution of patients showing a peak incidence around 11– 12 years (the trend is similar in both sexes)|
Click here to view
|Figure 2: Distribution of patients according to the month of onset showing slight preponderance in winter months|
Click here to view
| Discussion|| |
The educational background of parents has been shown to be an important determinant in the etiology of diabetes in young in some studies. Bloom et al observed a tendency toward shorter, more practical type of education among parents of diabetic children compared with control subjects. In the present study, parents' educational background did not vary significantly between patients and control subjects. Most parents in either group were educated up to high school level, which might have reflected the prevailing situation in the population. Children of both illiterate and highly educated parents were equally at risk of having diabetes. Unlike others, our findings did not support the notion that higher socioeconomic status positively influences diabetes incidence in children. , Although a higher number of study children came from urban background, it could be a referral bias as comparable number of control were also from a similar background. Migration has been shown to be influential in the etiologic and clinical course of both type 1 and type 2 diabetes. , No significant difference in history and type of migration between the study and control groups was observed. All previous studies included large populations. Hence it is difficult to draw a conclusion from a hospital-based study, such as the current one.
Advanced maternal age has been shown to be associated with diabetes in young. ,, Such a correlation was not observed in the current study. Mean age of either parent at birth of the subject from both the group did not differ significantly. Data with regard to the mode of delivery, birth order, and total number of children in the referent family were comparable between 2 groups. Others have found increased risk of diabetes among lower birth order children. ,
Mean period of total breast-feeding in the current study was significantly longer in controls (P < 0.05). However, the duration of total breast-feeding among the patients ranged from birth till 2.5 years of age (mean ± SD = 13 ± 7.5 months). As such no conclusion can be drawn regarding the protective nature of breast-feeding on diabetes. Different researchers  earlier reported the protective nature of breast-feeding, whereas others found evidence to the contrary. ,, The diabetogenic effect of animal milk is also a matter of debate.  The consumption of processed food, sugar, milk, and fat did not differ between families of patients and controls.
A slight male preponderance (56%) found among diabetic children is similar to that of a large epidemiologic study among Swedish  and Swiss  children. One German study  did not find significant difference among either sex, whereas the Hawaiian IDDM registry  showed a higher incidence among girls. Influence of sex in etiology of diabetes is not clear. Peak incidence of diabetes was seen at 11-12 years of age, that is, early adolescence in this study, a finding which is similar to that of the EURODIAB-ACE (10-14 years) and a South Indian hospital-based study (11 years). ,
Various studies have indicated a high incidence of diabetes during winter and early spring  A Japanese study  reported highest occurrence (40%) during January-March. A higher incidence was observed during autumn and winter months in the current study also. Such variations at timing of onset suggest possible environmental triggers, such as viral infection, eating pattern, and so on.
In conclusion, children and adolescents with diabetes mellitus do not differ significantly from their normal healthy peers in nutritional status, growth characteristics, and socioeconomic, educational, and demographic characteristics of the family. The possible role of environmental socioeconomic, and environmental factors in triggering the onset and progression of T1DM has always generated great interest among scientists across the globe, leading to significant number of publications on this subject. Ideally, epidemiologic studies with large cohorts are required to address such issues. Unfortunately, such studies have rarely been performed in our country. Our study, despite being hospital based, and including a relatively smaller cohort of patients and controls is a humble initiation to generating some indigenous data on some hitherto unresolved issues. In many instances, data accumulation is completely dependent on the recalling power of parents or other relatives of the subject that is not entirely without bias. A long-term prospective study can overcome these limitations.
| References|| |
|1.||Kelly MA, Mijovic H, Banett AH. Genetics of type I diabetes: In Best practice and Research Clinical Endocrinology and Metabolism. Harcourt 2002;15:279-91. |
|2.||Blom L, Dalqust G, Nystrom L, Sandström A, Wall S. The Swedish childhood diabetes study: Social and prenatal determinants in childhood. Diabetologia 1989;32:7-13. |
|3.||Metcalfe MA, Baum JD. Family characteristics and insulin dependent diabetes. Arch Dis Child 1992;67:731-6. |
|4.||Ramachandran A, Snehlata C, Joseph TA, Vijay V, Vishthanathan M. Delayed onset of diabetes in children of low economic stratum- A study from South India. Diabetes Res Clin Pract 1994;22:171-734 . |
|5.||Law G, McKinney PA, Staines A, Willium R, Kelly M, Alexander F, et al. Clustering of childhood IDDM-Links with age and place of residence. Diabetes Care 1997;20:753-6. |
|6.||Rios MS, Goday A, Larrid TM. Migrant population and incidence of type I diabetes mellitus, An overview of the literature with a focus in the Spanish-Heritage countries in Latin America. Diabetes Metab Res Rev 1999;15:113-32. |
|7.||Bodansky HJ, Staines A, Stephenson P, Haigh D, Cartwright R. Evidence for environmental effect in etiology of insulin dependent diabetes in a transmigratory population. BMJ 1992;304:1020-2 . |
|8.||Bingley PJ, Douek IF, Christine AR, Gale EA. Rising incidence of insulin dependent diabetes in children aged under 5 years in the Oxford region: Time trend analysis. BMJ 1997;315: 713-7. |
|9.||Wagener F La, Porte RE, Orchard TJ, Cavender D, Kullar LH, Drash AL. The Pittsburg diabetes mellitus study 3: An increased prevalence with older maternal age. Diabetologia 1983;25:82-5. |
|10.||Ramachandran A, Snehlata C, Premila L, Mohan V, Viswanathan M. Maternal age and birth order of young IDDM patients study from South India. Diabetes Care 1993;16:636-7. |
|11.||Pociot F, Mandrup-Poulson T. Aetiogenesis and pathogenesis of insulin dependent diabetes mellitus -a general overview. In Leslie RD, editor. Molecular pathogenesis of diabetes mellitus. Front Hormone Res 1997;22:1-22. |
|12.||Cavallo MG, Fava D, Monetini L, Barone F, Pozzili P. Cell mediated immune response to b casein in recent onset diabetes: Implication for disease pathogenesis. Lancet 1996;348:926-8. |
|13.||Hummel M, Schenker M, Fuchtenbusch Mand Ziegler AG. No major association of breast feeding vaccination and childhood viral diseases with early Islet Autoimmunity in German BABYDIAB study. Diabetes Care 2000;23:609-14 . |
|14.||Norria JM, Beaty B, Kilnge-Smith G, Hoffman M, Chase P, Erlich HA, et al. Lack of association between early exposure to cow's milk protein and b- cell autoimmunity: Diabetes autoimmune study in the young (DAISY). J Am Med Assoc 1996;276:606-14. |
|15.||Levy-Marchal C, Patterson C, Green A. Variation by age group and seasonality at diagnosis of childhood IDDM in Europe. Diabetologia 1995;38:823-30 . |
|16.||Patrick SL, Kadohiro JK, Waseman SH, Curb JD, Orchard TJ, Dorman JS, et al. IDDM incidence in a multiracial population, The Hawaii IDDM Registry,1980-1990, Diabetes Care 1997;20:983-7. |
|17.||Higa S, Taira T, Nakaguchi H, Nagayoshi H, Ishikawa K, Murakami K, et al. Seasonal variation in the onset of IDDM in Miruma G, Qian RL, Murakami K, editors. Current status of diabetes mellitus in east Asia ; Elsevier Sc: 1994. p. 67-9. |
|18.||Schonle EJ, Lang-Muritano M, Gschwend S, Lainbacha J, Mullis PE, Torresanj T, et al. Epidemiology of type I diabetes mellitus in Switzerland: Steep rise in incidence under 5 years old children in past decades. Diabetologia 2001;44:286-9. |
|19.||Kadiki OA, Reddy MR, Marzouk AA. Incidence of insulin dependent diabetes (IDDM) and non insulin dependent diabetes (NIDDM) (0-34years at)onset in Benghazi, Libia. Diabetes Res Clin Pract 1996;32:165-73. |
|20.||Kuppuswami BR. Socio-economic status scale (Urban). Delhi: Mansayan; 1962. |
|21.||Nek A, Hub R, Kehren M, Ranke MB. Incidence of IDDM in German children aged 0-14 years-a 6 year population study. Diabetes Care 1997;20:530-3 . |
|22.||Ramaiya KL, Kodali VR, Alberti KG. Epidemiology of diabetes in Asian of Indian Subcontinent. Diabetes Metab Rev 1990;6:25-46. |
[Figure 1], [Figure 2]
[Table 1], [Table 2]