DIABETICMedicine DOI: 10.1111/dme.12591
Research: Epidemiology The influence of family history of diabetes on disease prevalence and associated metabolic risk factors among Sri Lankan adults P. Katulanda1,2, P. Ranasinghe3, R. Jayawardena1,4, R. Sheriff1 and D. R. Matthews2 1 Diabetes Research Unit, Department of Clinical Medicine, Faculty of Medicine, University of Colombo, Sri Lanka, 2Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, UK, 3Department of Pharmacology, Faculty of Medicine, University of Colombo, Sri Lanka and 4Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia
Accepted 19 September 2014
Abstract Aims To describe the influence of family history on diabetes prevalence and associated metabolic risk factors in a nationally representative sample from Sri Lanka. Methods A cross sectional national survey was conducted among 5000 adults in Sri Lanka. Family history was evaluated at three levels: (1) parents, (2) grandparents (paternal and maternal) and (3) siblings. A binary-logistic regression analysis controlling for confounders (age, gender, BMI and physical activity) was performed in all patients with ‘presence of diabetes’ as the dichotomous dependent variable and using family history in father, mother, maternal grandmother/grandfather, paternal grandmother/grandfather, siblings and children as binary independent variables.
The sample size was 4485, mean age was 46.1 15.1 years and 39.5% were males. In all adults, the prevalence of diabetes was significantly higher in patients with a family history (23.0%) than those without (8.2%) (P < 0.001). When family history was present in both parents, the prevalence of diabetes was 32.9%. Presence of a family history significantly increased the risk of diabetes [odds ratio (OR): 3.35, 95% confidence interval (CI): 2.78– 4.03], obesity (OR: 2.45, 95% CI: 1.99–2.99), hypertension (OR: 1.25, 95% CI: 1.08–1.45) and metabolic syndrome (OR: 2.28, 95% CI: 1.97–2.63). In all adults, the presence of a family history of diabetes in a father (OR: 1.29, 95% CI: 1.02–1.63), mother (OR: 1.23, 95% CI: 1.11–1.36), paternal grandfather (OR: 1.27, 95% CI: 1.14–1.41), siblings (OR: 4.18, 95% CI: 3.34–5.22) and children (OR: 5.47, 95% CI: 2.93–10.19) was associated with a significantly increased risk of developing diabetes. Results
Conclusions Family history and diabetes had a graded association in the Sri Lankan population, because the prevalence increased with the increasing number of generations affected. Family history of diabetes was also associated with the prevalence of obesity, metabolic syndrome and hypertension. Individuals with a family history of diabetes form an easily identifiable group who may benefit from targeted interventions.
Diabet. Med. 00, 000–000 (2014)
Introduction Diabetes has become an important health concern in South Asia. It is estimated that the prevalence of diabetes in the region will increase about 151% between 2000 and 2030 [1]. It is well known that South Asians are at an increased risk of developing diabetes in comparison with other ethnic groups [2]. There is strong evidence to suggest that South Asians are more insulin resistant than Caucasians, even at younger ages
Correspondence to: Prasad Katulanda. E-mail: [email protected]
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
and comparative low levels of BMI, possibly due to a greater accumulation of visceral fat [3]. South Asians also experience an early decline in b–cell function, compared with other ethnic groups [3,4]. This predisposition, coupled with recent urbanization leading to unhealthy lifestyle changes, such as physical inactivity, changes in dietary habits and stress, are known to be contributory factors driving the diabetes epidemic in South Asia [5]. Because a significant number of immigrants from the region are living in affluent Western nations, a disease such as Type 2 diabetes affecting ethnic South Asians will have potential implications on global health.
1
DIABETICMedicine
Family history of diabetes and disease prevalence P. Katulanda et al.
What’s new? • There are no studies on the association between family history of diabetes and the prevalence of the disease in a large cohort of ethnic South Asian adults. • Family history and diabetes showed a graded association in Sri Lankan adults, as the prevalence increased with the increasing number of generations affected. • Family history of diabetes was also associated with the prevalence of obesity, metabolic syndrome and hypertension. • Family history could be used as a tool to detect those at risk of diabetes and a constellation of other cardiovascular risk factors. Family history is a common non-modifiable risk factor for most of the chronic non-communicable diseases, including cardiovascular disease, Type 2 diabetes and hypertension. It is a collective reflection of the genetic susceptibility, shared environments and behaviours that, when present, influences the probability of developing a disease [6]. Information on family history may serve as a unique and useful tool for public health and preventive medicine [6]. Most health screening guidelines usually include family history as a marker to assess health risks and initiate interventions. The advantages of family history as a risk assessment tool is its lower cost, greater acceptability and that it is a reflection of the shared genetic and lifestyle factors. Type 2 diabetes is a common chronic disease that results from an interaction between genetic and environmental factors. Studies have shown that the risk of Type 2 diabetes increases 2–5 times when one or both parents are affected [7,8]. Furthermore, the presence of a family history of diabetes is known to be associated with higher fasting plasma glucose, lipids, systolic blood pressure and BMI [9,10]. The increased genetic predisposition among South Asians probably makes family history more important in risk assessment than in other ethnic groups. In addition, family history reflects both shared genetic and lifestyle factors, thus it may serve as a better predictor of diabetes risk than either factor taken alone. It can be used to identify individuals at risk and influence health-promoting behaviour; in addition, prevention efforts can also be extended to the family members at risk. To our knowledge, there is currently no detailed analysis of the association between family history of diabetes and the prevalence of the disease in a large cohort of ethnic South Asian adults. This study aims to describe the influence of family history on diabetes prevalence and associated metabolic risk factors in a large cohort of South Asian adults, from a nationally representative sample from Sri Lanka.
2
Methods Study population and sampling
This study was a cross sectional community-based national survey conducted between August 2005 and September 2006 in seven of the nine provinces in Sri Lanka. Detailed sampling has been described elsewhere [11]. In brief, 5000 adults (> 18 years) were invited to take part in the study using a multistage stratified cluster sampling technique. A total of 100 clusters of 50 adults each were divided among the seven provinces using a probability proportional to size (PPS) technique based on total provincial populations. In each province, the required number of clusters was selected from a list of all ‘Village Officers Units’ in that province by using a computer-generated random number list. The voter registration lists of the selected ‘Village Officers Units’ were used to randomly select the first household in each cluster and a uniform criterion was used to select the remaining 49 households. Ethical approval for the study was obtained from the Ethics Review Committee, Faculty of Medicine, University of Colombo, Sri Lanka.
Data collection
An interviewer-administered questionnaire was used to collect data, which included: age, gender, area of residence, ethnicity, level of education, household monthly income [Sri Lankan Rupees (LKR)], duration of diabetes, family history, height, weight, waist circumference and hip circumference. Each study participant was asked by direct questioning whether any of his/her family members (living or not) had diabetes, as diagnosed by a physician. The family history was evaluated at three levels: (1) parents, (2) grandparents (paternal and maternal) and (3) siblings. If a positive family history was present, the age of onset of diabetes at each of the above levels was evaluated. The options provided were ‘yes’, ‘no’ and ‘don’t know’. For the purpose of analysis, those who said that they were unaware (‘don’t know’) of the family history were excluded. In addition, family history was grouped into the following generations: first generation, siblings; second generation, parents; and third generation, grandparents. We also looked for the presence of diabetes among the children of the study participants. Data on self-reported physical activity were collected using the short version of the International Physical Activity Questionnaire (IPAQ). Height was measured using Harpenden stadiometers (Chasmors Ltd, London, UK) to the nearest 0.1 cm. Body weight was measured using a Salter 920 digital weighing scale (Salter Ltd, Tonbridge, UK) to the nearest 0.1 kg. BMI was calculated as weight in kilograms divided by height squared in metres (kg/m2). Waist circumference was measured midway between the iliac crest and the lower rib margin at the end of normal expiration and hip
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
circumference was measured at the widest level over the greater trochanters using a plastic flexible tape to the nearest 0.1 cm. Waist-to-hip ratio and waist-to-height ratio were calculated as waist circumference divided by hip circumference and height, respectively. Seated blood pressure was measured after at least a 10 min rest using Omron IA2 digital blood pressure monitors (Omron Healthcare, Singapore). Fasting venous blood samples were obtained for glucose and lipid estimation from all participants, details of analysis have been described previously [11].
DIABETICMedicine
procedure was used and a P–value of 0.10 was considered to be the cut-off for removal of variables. A similar binary logistic regression analysis with above dependant and independent variables was also performed separately for both males and females. We present the results of the logistic regression analysis as adjusted odds ratio controlling for confounders (age, gender, BMI and physical activity). In all statistical analyses P < 0.05 was considered significant.
Results
Definitions
Sample characteristics
Participants were considered to have ‘diagnosed diabetes’ if they had been previously diagnosed at a government hospital or by a registered medical practitioner. New cases (‘undiagnosed diabetes’) were diagnosed according to the American Diabetes Association (ADA) and World Health Organization (WHO) criteria [12,13]. Pre-diabetes was also defined according to the above WHO criteria. Young-onset diabetes was defined as those who develop diabetes at an age < 40 years. Hypertension was defined as systolic blood pressure > 130 mmHg and/or diastolic blood pressure > 85 mmHg and/or being on anti-hypertensive treatment. Central obesity was classified as a waist circumference > 90 cm for males and > 80 cm for females (Asian cut-offs) [14]. Obesity was defined as a BMI ≥ 25 kg/m2, based on criteria for Asians [14]. Metabolic syndrome was defined according to the International Diabetes Federation (IDF) criteria [14]. Urban and rural sectors were defined according to the classification of the Sri Lankan government. Physical activity was classified in to three categories (‘inactive’, ‘moderately active’ and ‘highly active’) based on the total metabolic equivalent of task (MET) minutes/week derived from the IPAQ short version [11]. Peripheral neuropathy was assessed using the validated Toronto Clinical Scoring System (TCSS) (score > 5) [15]. Ophthalmological evaluation for retinopathy was performed with dilated indirect ophthalmoscopy using slit lamp biomicroscopy.
Of the 5000 invited individuals, 4485 participated in the study (response rate 89.7%). Mean age ( SD) was 46.1 15.1 years and 39.5% (n = 1772) were males. Most were residing in rural areas (n = 3530, 78.7%) and Sinhalese in ethnicity (n = 3877, 86.4%). The crude prevalence of diabetes was 12.0% [n = 536, 95% confidence interval (CI): 11.0–12.9], of which 344 (64.2%) were patients with ‘diagnosed diabetes’. The prevalence of obesity (BMI ≥ 25 kg/m2) was 9.8% (n = 438, 95% CI: 8.9–10.6), and central obesity (based on waist circumference) was present in 30.3% (n = 1358, 95% CI: 28.9–31.6). Pre-diabetes was prevalent in 14.6% (n = 577, 95% CI: 13.5–15.7), and 27.2% (n = 1219, 95% CI: 26.1–28.7) had hypertension and 26.6% (n = 1193, 95% CI: 25.3–27.9) had metabolic syndrome.
Statistical analysis
Data were analysed using SPSS v14 (SPSS Inc., Chicago, IL, USA) statistical software package. The significance of the differences between proportions and means was tested using a z-test and Student’s t-test or analysis of variance (ANOVA), respectively. A binary logistic regression analysis was performed in all patients with ‘presence of diabetes’ as the dichotomous dependent variable (0 = diabetes absent; 1 = diabetes present) and using family history in father, mother, maternal grandmother/grandfather, paternal grandmother/grandfather, siblings and children as the binary independent variables (0 = No, 1 = Yes). The explanatory variables selected above were subsequently included in a binary logistic regression model, a backward elimination
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Family history, prevalence of diabetes and metabolic risk factors
The overall prevalence of a family history of diabetes in the population was 25.5% (n = 1141, 95% CI: 24.0–27.0), irrespective of disease status. A family history of diabetes (parents, grandparents or siblings) was present in 49.0% (n = 262, 95% CI: 45.0–53.0) of patients with diabetes and 22.3% (n = 879, 95% CI: 21.0–24.0) participants without diabetes. In all adults, the prevalence of diabetes was significantly higher in patients with a family history (23.0%, 95% CI: 20.5–25.4) than in those without a family history (8.2%, 95% CI: 7.2–9.1) (P < 0.001). A similar result was observed independently in both males (23.7% vs. 7.4%) and females (22.6% vs. 8.7%). Presence of a family history significantly increased the risk of diabetes [odds ratio (OR): 3.35; 95% CI 2.78–4.03]. The prevalence of diabetes was higher in those with a family history of diabetes at all levels (mother, father, maternal grandmother/grandfather, paternal grandmother/ grandfather, siblings and children) than those without a family history (Table 1) (An example for a contingency table for Table 1 is provided in the Supplementary file as Table S4.) However, this difference was significant only in those with a family history in mother, father, maternal grandfather, siblings and children. There was no significant difference between the prevalence of diabetes in those with a family
3
Family history of diabetes and disease prevalence P. Katulanda et al.
DIABETICMedicine
history in the father (20.8%, 95% CI: 16.3–25.3) or mother (22.3%, 95% CI: 18.6–26.0). When family history of diabetes was present in both parents, the prevalence of diabetes was 32.9% (95% CI: 21.8–43.9). Among the different ethnicities, the presence of a family history was highest in Muslims (47.0%), followed by Sinhalese (24.5%) and Tamils (16.2%). The prevalence of diabetes in those with a family history was higher in each generation than in those without a family history [first generation (siblings), 34.2% vs. 9.1%, P < 0.001; second generation (parent), 20.6% vs. 10.3%, P < 0.001; third generation (grandparents) 13.3% vs. 11.9%, P = 0.338]. In addition, the prevalence of diabetes also increased with the number of affected generations (one generation, 33.9%; two generations, 34.2%; and three generations, 37.1%; P = 0.608). In patients without diabetes, obesity (34.5% vs. 17.7%; P < 0.001), central obesity (46.7% vs. 24.6%; P < 0.001) and metabolic syndrome (30.6% vs. 20.2%; P < 0.001) were more prevalent in those with a family history of diabetes than in those without. In this group, presence of family history of diabetes increased the risk of obesity (OR: 1.87; 95% CI 1.62–2.17), central obesity (OR: 1.95; 95% CI 1.74–2.18) and metabolic syndrome (OR: 1.52; 95% CI 1.34–1.72). Similarly in patients with diabetes, the prevalence of obesity (61.4% vs. 46.7%; P < 0.01), central obesity (57.4% vs. 39.4%; P < 0.001) and metabolic syndrome (52.3% vs. 39.9%; P < 0.01) were all significantly higher in those with a family history of diabetes than in those without. In patients with diabetes, presence of family history of diabetes increased the risk of obesity (OR: 1.32; 95% CI 1.08– 1.60), central obesity (OR: 1.45; 95% CI 1.21–1.75) and metabolic syndrome (OR: 1.31; 95% CI 1.05–1.64).
Association of age of onset, clinical and biochemical parameters with family history
The age at diagnosis of diabetes was significantly lower in patients with a family history (48.9 11.6 years) than in
those without (52.6 12.3 years) (P < 0.01). Table 2 summarizes the association between age, clinical and biochemical parameters with family history in patients with and without diabetes. In both patients with diabetes and without diabetes, those with a family history were significantly younger and had a higher mean BMI, waist circumference, hip circumference and fasting blood glucose. However, in the participants without diabetes, in addition to the above-mentioned variables, those with a family history also had a higher total cholesterol, LDL cholesterol, triglycerides and lower HDL cholesterol (Table 2). In patients with diabetes the BMI, waist and hip circumferences increased significantly with the increasing number of generations affected by diabetes (Table 3). A similar trend was also observed among the participants without diabetes. Furthermore, in the patients with diabetes, an increase was also observed in systolic and diastolic blood pressures, fasting blood glucose, total cholesterol and triglycerides with the increasing number of generations affected by diabetes, although this difference was not significant (Table 3). By contrast, in patients without diabetes, there was a significant decrease in total cholesterol, LDL cholesterol and triglycerides with increasing number of generations (Table 3). In those with a family history of diabetes, the mean age of onset ( SD) of diabetes in the fathers, mothers, brothers, sisters, sons and daughters was 57.5 ( 10.6), 58.2 ( 11.8), 45.6 ( 9.2), 46.1 ( 9.3), 37.9 ( 7.8) and 41.5 ( 8.5) years, respectively. In patients with diabetes, a significant correlation was observed between the age of onset of diabetes and the age of onset in the mother (r = 0.25, P < 0.05), brother (r = 0.56, P < 0.001), sister (r = 0.60, P < 0.001), son (r = 0.79, P < 0.001) and daughter (r = 0.72, P < 0.01). In patients with young-onset diabetes (< 40 years), the age of onset of diabetes in brothers (38.3 6.6 vs. 47.6 9.3 years) and sisters (37.7 8.9 vs. 48.4 8.2 years) was significantly lower than in patients developing diabetes at > 40 years of age. The age of onset of diabetes in the patients decreased significantly with the
Table 1 Presence of family history in different generations in patients with and without diabetes
Mother Father Maternal grandmother Maternal grandfather Paternal grandmother Paternal grandfather Siblings Children
Patients with diabetes (n = 535)
Patients without diabetes (n = 3947)
Number with family history (prevalence of diabetes)
Number with family history (% without diabetes)
Present
Absent
Not known
P*
Present
Absent
Not known
P†
107 66 15 10 9 7 175 24
399 434 333 327 323 317 360 511
29 35 186 197 202 210 0 0
< 0.001 < 0.001 0.131 < 0.05 0.377 0.178 < 0.001 < 0.001
373 251 97 46 72 42 336 25
3525 3637 3096 3094 3044 3048 3610 3922
49 59 753 806 830 856 0 0
< 0.001 < 0.001 0.131 < 0.05 0.377 0.178 < 0.001 < 0.001
(22.3) (20.8) (13.4) (17.9) (11.1) (14.3) (34.2) (49.0)
(10.2) (10.7) (9.7) (9.6) (9.6) (9.4) (9.1) (11.1)
(37.2) (37.2) (19.8) (19.6) (19.6) (19.7)
(77.7) (79.2) (86.6) (82.1) (88.9) (85.7) (65.8) (51.0)
(89.8) (89.3) (90.3) (90.4) (90.4) (90.6) (90.9) (88.5)
(62.8) (62.8) (80.2) (80.4) (80.4) (80.3)
*Comparison between prevalence of diabetes in those with and without a family history. Comparison between the percentage without diabetes in those with and without a family history.
†
4
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Table 2 Association of age, clinical and biochemical parameters with family history
Age (years) BMI (kg/m2) Waist circumference (cm) Hip circumference (cm) Waist-to-hip ratio Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting blood glucose (mg/dl) Total cholesterol (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglycerides (mg/dl)
Patients with diabetes (n = 535)
Patients without diabetes (n = 3947)
Family history
Family history
Present Mean ( SD)
Absent Mean ( SD)
53.6 24.5 86.8 94.7 0.91 136.2 80.3 145.9 215.3 137.9 45.9 157.2
56.3 23.2 82.7 91.0 0.91 138.1 79.5 135.7 219.8 141.3 46.3 161.0
( ( ( ( ( ( ( ( ( ( ( (
11.5) 4.1) 10.4) 9.3) 0.07) 19.9) 11.3) 58.7) 40.6) 36.2) 9.2) 75.2)
( ( ( ( ( ( ( ( ( ( ( (
P*
13.1) 3.9) 10.9) 8.6) 0.07) 23.3) 11.9) 55.9) 48.4) 42.3) 10.3) 97.8)
< < <
70
Age in years
(b)
70
(b)
40
60
35
50
30
Prevalence
Prevalence of Diabetes
45
25 20 15
30 20
10
10
5 0 25
Body Mass Index (kgm2) Family History - Present
Family History - Absent
FIGURE 1 Prevalence of diabetes by (a) age category and (b) BMI in patients with and without a family history of diabetes.
result was observed in the participants without diabetes (3817.78 vs. 5248.01 min, P < 0.001). Patients with diabetes with a family history were more ‘inactive’ (24.8% vs. 13.9%, P < 0.01) and less ‘highly active’ (38.9% vs. 45.8%, P < 0.01) when self-reported physical activity was classified in to the three categories based on the total MET minutes/week. Even in patients without diabetes, those with a family history were more ‘inactive’ (13.5% vs. 8.9%, P < 0.001) and less ‘highly active’ (51.5% vs. 65.6%, P < 0.001). Family history and complications of diabetes
The prevalence of macrovascular complications of diabetes was higher among patients with diabetes with a family history than in those without; ischaemic heart disease (16.2% vs. 14.6%, P = 0.629), stroke/TIA (4.6% vs. 2.2%, P = 0.153) and peripheral vascular disease (5.4% vs. 4.5%, P = 0.803). However, of the microvascular complications, only nephropathy was more prevalent in the patients with a family history; nephropathy (8.1% vs. 3.2%, P < 0.05), retinopathy (26.8% vs. 28.1%, P = 0.893) and neuropathy (68.5% vs. 73.5%, P = 0.545). Erectile impotence in males with diabetes was more prevalent in patients with a family history (8.1%) than in those without (3.3%) (P < 0.05).
6
40
0 25
Body Mass Index (kgm-2) Hypertension Metabolic Syndrome
Family history - Present
Family history - Absent
Family history - Present
Family history - Absent
FIGURE 2 Prevalence of hypertension and metabolic syndrome by (a) age category and (b) BMI in patients with and without a family history of diabetes.
Logistic regression analysis
The results of the binary logistic regression analysis in all adults using the dichotomous variable ‘presence of diabetes’ as the dependent factor and other independent variables are shown in Table 4. The overall model was statistically significant and the Cox & Snell R-square and Nagelkerke R-square values were 0.064 and 0.123, respectively. The results indicate that in all adults, the presence of a family history of diabetes in father (OR: 1.29), mother (OR: 1.23), both parents (OR: 1.64), paternal grandfather (OR: 1.27), siblings (OR: 4.18) and children (OR: 5.47) were associated with significantly increased risk of developing diabetes (Table 4). Family history in siblings and children were also associated with developing diabetes in both males and females independently (Table 4). However, family history in father (OR: 1.58) was associated with increased risk of diabetes only in males, whereas in females, it was the family history in the mother (OR: 1.35) that significantly increased the risk.
Discussion To our knowledge, this is the first comprehensive survey evaluating the association between family history of diabetes,
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Table 4 Binary logistic regression analysis in all adults, males and females Adjusted odds ratio (95% CI)a Co-variants (Family history)
All adults
Father Mother Both parents Paternal grandfather Maternal grandfather Siblings Children
1.29 1.23 1.64 1.27 0.99 4.18 5.47
(1.02–1.63)‡ (1.11–1.36)† (1.10–2.10)‡ (1.14–1.41)* (0.77–1.27) (3.34–5.22)* (2.93–10.19)*
Male 1.58 1.42 0.73 2.19 0.78 5.73 5.82
Female (1.11–2.22)† (0.84–2.39) (0.22–2.46) (1.30–3.69)† (0.53–1.14) (3.95–8.31)* (1.94–17.44)†
1.18 1.35 1.31 1.08 1.26 3.39 4.66
(0.72–1.94) (1.18–1.54)* (0.56–3.05) (0.78–1.49) (1.09–1.45)† (2.56–4.51)* (2.13–10.21)*
*P < 0.001, †P < 0.01, ‡P < 0.05; a, adjusted odds ratio controlling for confounders (age, gender, BMI and physical activity).
disease prevalence and metabolic risk factors from a nationally representative, large cohort of ethnic South Asian adults from Sri Lanka. Our results clearly demonstrate that the family history of diabetes has a graded association with diabetes, because the prevalence of diabetes increased with the number of generations with a family history. In the current cohort, those with a family history of diabetes were 1.2–5 times more likely to develop diabetes than those without a family history. Previous studies from Germany, Norway, South Africa, Taiwan and the USA have shown similar increased risks, in which compared with people without a family history of diabetes, those who have a family history were 2–6 times more likely to develop Type 2 diabetes [16]. However, the observed risks were much higher in European populations than reported in our cohort [8]. The risk associated with family history in the current population appeared to be independent of other known risk factors, including age, anthropometric parameters (BMI) and lifestyle factors (physical activity). The InterAct study, a large prospective case–cohort study involving eight European countries, also corroborated these findings, where prominent lifestyle (diet and physical activity) and anthropometric (BMI, waist and hip circumference) risk factors explained only a marginal proportion of the excess risk associated with family history [8]. These findings further enumerate the fact that family history is a strong and independent risk factor for diabetes [17]. Furthermore, family history and diabetes showed a graded association, because the prevalence increased with the increasing number of generations affected. A similar study in the USA showed that the strength of the association was related to the type and number of relatives involved [18]. Participants with a family history of diabetes displayed a significantly higher BMI, waist and hip circumference than those without a family history, irrespective of diabetes status, a finding that is in keeping with other similar studies from Europe [19]. In another recent study from India, it was found that apparently healthy individuals with a family history of diabetes had higher anthropometric values than their counterparts [20]. Furthermore, the anthropometric parameters in the present cohort showed a graded increase with the ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
increasing number of generations affected by diabetes. We also observed that, irrespective of disease status, participants with a family history had a lower total weekly MET minutes and were more ‘inactive’ in self-reported physical activity than those without a family history. Objectively measured physical activity is known to be associated with insulin resistance and dyslipidaemia independent of sex, age and obesity in those with a family history of diabetes [21]. These findings further support the hypothesis that expression of complex traits like diabetes results from an interaction between shared genes, environments and behaviours [22]. Furthermore, individuals with a family history of diabetes form an easily identifiable group who may benefit from targeted intervention to prevent the development of diabetes through increased physical activity. Hence, family history of diabetes is included in most diabetes risk scores [23]. Patients with a family history were both younger in age and had a younger age of onset of diabetes than patients without a family history of diabetes, a finding which is consistent with previous reports among European adults [19]. In addition, the age of onset showed an inverse relationship with the increasing number of generations with a family history of diabetes. We also observed that the age of onset of diabetes in patients was closely correlated with the age of onset in siblings and children, rather than with the age of onset in parents. These findings suggest that the age of diabetes onset might be genetically determined. However, another simpler/alternative explanation could be that those with a family history of diabetes are more aware about the disease and have frequent contact with medical practitioners, leading to an earlier diagnosis. Our data showed that a history of diabetes in the sibling was associated with a higher risk of diabetes than parental history. Sibling pairs have higher shared environmental factors than parents and offspring, and some common environments in young childhood have been postulated to be associated with future adult diseases [24]. Furthermore, we did not observe a difference in the prevalence of diabetes in patients with a family history in either mother or father. However, previous studies from different parts of the world have shown that there is an excess maternal transmission of diabetes [19]. A finding
7
DIABETICMedicine
Family history of diabetes and disease prevalence P. Katulanda et al.
refuted in the Framingham population, where maternal and paternal diabetes conferred equal risk for overt diabetes among offspring, similar to the findings in the present study [25]. Family history of diabetes is associated with the prevalence of obesity, metabolic syndrome and hypertension. In a cohort of Asian Indians, a positive family history of diabetes was significantly associated with the prevalence of metabolic syndrome [26]. A study conducted in Sweden demonstrated that family history of diabetes was not only related to the presence of obesity on one occasion, but also to an examination 12 years later [27]. Previous studies have also demonstrated an association between family history of diabetes and the presence of hypertension and the clustering of these disorders in the offspring of affected patients [28]. However, a study in Taiwanese patients, found that a parental family history of diabetes conferred a lower risk of hypertension in Type 2 diabetes, suggesting the need for further study of blood pressure control in relation to family history [29]. In a study by Haffner et al., people with a positive parental history of diabetes had a more atherogenic pattern of cardiovascular risk factors compared with those without a parental history of diabetes [30]. A similar result was observed in the present cohort, where in patients with and without diabetes, the BMI, waist and hip circumferences increased significantly with the increasing number of generations affected by diabetes. Furthermore, although statistically not significant in the patients with diabetes, an increase was also observed in systolic and diastolic blood pressure, fasting blood glucose, total cholesterol and triglycerides with the increasing number of generations affected by diabetes. The genetic factors associated with the occurrence of familial diabetes may increase risk of cardiovascular disease beyond the risk among people without a family history [31]. Hence, family history could be used as a tool to detect not only those at risk of diabetes, but also a constellation of other cardiovascular risk factors, and they might form a group of individuals that benefit the most from targeted risk factor interventions. Family history is a common non-modifiable risk factor for most chronic non-communicable diseases, as it is a collective reflection of the genetic susceptibility, shared environments and behaviours [6]. Hence, information on family history may serve as a unique and useful tool for public health and preventive medicine [6]. The advantages of family history as a risk assessment tool are its lower cost, greater acceptability, and that it is a reflection of shared genetic and lifestyle factors. Although family history by itself is a non-modifiable risk factor, it is useful for raising awareness of risk, risk stratification, targeting interventions, and positively influencing health behaviours [16]. Risk perception is one factor that positively influences health-related behaviour change in relation to diabetes prevention [32]. There is enough evidence from randomized controlled trials and observational studies to show that Type 2 diabetes can be prevented or delayed by adopting simple, healthy lifestyle changes.
8
Accurate risk perception might positively influence behaviour change, and aid in prevention of diabetes [16]. Hence, overall there is promising potential for the use of family history as a public health tool aiding prevention of diabetes. The strengths of the present study include the large and nationally representative nature of the sample, random selection of participants from a well defined and homogenous target population, the high response rate and the detailed nature of the clinical and demographic assessment using well validated tools. The major limitation of the study was the, cross sectional nature. The usefulness of family history in risk prediction should be tested in large prospective studies. Furthermore, in developing countries like Sri Lanka, large sections of the community remain undiagnosed and therefore the accuracy of self-reported family history is a challenge. Other potential limitations include the recall bias for family history, however, previous studies have shown that this is minimal in diabetes [33]. In order to minimize recall bias, we have carried out a separate analysis on patients with newly diagnosed diabetes, which is presented in the Supporting Information (Tables S1–S4). The prevalence of diabetes in newly diagnosed patients was significantly higher in those with a family history in the mother or in siblings (Tables S1– S4). However, in the binary logistic analysis, only a family history in siblings was shown to be significantly associated with the presence of diabetes among newly diagnosed patients. Furthermore, newly diagnosed patients with a family history were younger in age, and had a higher BMI, waist and hip circumference. We were also unable to accurately assess the impact of family history on glycaemic control, because data on HbA1c values were not available for the majority of the population. In this study, we have made no distinction between Type 1 and Type 2 diabetes in the study population. However, previous large-scale studies in Sri Lanka have shown that nearly 95% of Sri Lankan population with diabetes are having Type 2 diabetes.
Funding sources
The National Science Foundation (NSF) of Sri Lanka.
Competing interests
None declared.
Acknowledgements
The National Science Foundation (NSF) of Sri Lanka was the main source of funding for the study. The additional support provided from the Oxford Centre for Diabetes Endocrinology and Metabolism UK, and the NIHR Biomedical Research Centre Programme is gratefully acknowledged. We thank the Diabetes Association of Sri Lanka and the World Health Organization Office in Colombo for the support for lipid assays.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
References 1 Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004; 27: 1047–1053. 2 Mather HM, Keen H. The Southall Diabetes Survey: prevalence of known diabetes in Asians and Europeans. BMJ 1985; 291: 1081– 1084. 3 Gujral UP, Pradeepa R, Weber MB, Narayan KM, Mohan V. Type 2 diabetes in South Asians: similarities and differences with white Caucasian and other populations. Ann NY Acad Sci 2013; 1281: 51–63. 4 Kooner JS, Saleheen D, Sim X, Sehmi J, Zhang W, Frossard P et al. Genome-wide association study in individuals of South Asian ancestry identifies six new type 2 diabetes susceptibility loci. Nat Genet 2011; 43: 984–989. 5 Jayawardena R, Ranasinghe P, Byrne NM, Soares MJ, Katulanda P, Hills AP. Prevalence and trends of the diabetes epidemic in South Asia: a systematic review and meta-analysis. BMC Public Health 2012; 12: 380. 6 Yoon PW, Scheuner MT, Peterson-Oehlke KL, Gwinn M, Faucett A, Khoury MJ. Can family history be used as a tool for public health and preventive medicine? Genet Med 2002; 4: 304–310. 7 Alcolado JC, Alcolado R. Importance of maternal history of noninsulin dependent diabetic patients. BMJ 1991; 302: 1178–1180. 8 InterAct C. The link between family history and risk of type 2 diabetes is not explained by anthropometric, lifestyle or genetic risk factors: the EPIC-InterAct study. Diabetologia 2013; 56: 60–69. 9 Mitchell BD, Valdez R, Hazuda HP, Haffner SM, Monterrosa A, Stern MP. Differences in the prevalence of diabetes and impaired glucose tolerance according to maternal or paternal history of diabetes. Diabetes Care 1993; 16: 1262–1267. 10 Viswanathan M, McCarthy MI, Snehalatha C, Hitman GA, Ramachandran A. Familial aggregation of type 2 (non-insulin-dependent) diabetes mellitus in south India; absence of excess maternal transmission. Diabetes Med 1996; 13: 232–237. 11 Katulanda P, Constantine GR, Mahesh JG, Sheriff R, Seneviratne RDA, Wijeratne S et al. Prevalence and projections of diabetes and pre-diabetes in adults in Sri Lanka—Sri Lanka Diabetes, Cardiovascular Study (SLDCS). Diabetes Medicine 2008; 25: 1062–1069. 12 American Diabetes Association. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997; 20: 1183–1197. 13 World Health Organization. Part 1: diagnosis and classification of diabetes mellitus. Definition, diagnosis and classification of diabetes mellitus and its complications. Geneva: WHO 1999. 14 Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009; 120: 1640–1645. 15 Perkins BA, Olaleye D, Zinman B, Bril V. Simple screening tests for peripheral neuropathy in the diabetes clinic. Diabetes Care 2001; 24: 250–256. 16 Harrison TA, Hindorff LA, Kim H, Wines RC, Bowen DJ, McGrath BB et al. Family history of diabetes as a potential public health tool. Am J Prevent Med 2003; 24: 152–159. 17 Valdez R, Yoon PW, Liu T, Khoury MJ. Family history and prevalence of diabetes in the U.S. population: the 6–year results from the National Health and Nutrition Examination Survey (1999–2004). Diabetes Care 2007; 30: 2517–2522.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
DIABETICMedicine
18 Annis AM, Caulder MS, Cook ML, Duquette D. Family history, diabetes, and other demographic and risk factors among participants of the National Health and Nutrition Examination Survey 1999–2002. Prev Chron Dis 2005; 2: A19. 19 Papazafiropoulou A, Sotiropoulos A, Skliros E, Kardara M, Kokolaki A, Apostolou O et al. Familial history of diabetes and clinical characteristics in Greek subjects with type 2 diabetes. BMC Endocrin Disord 2009; 9: 12. 20 Padaki S, Vijayakrishna K, Dambal A, Ankad R, Manjula R, Surekharani C et al. Anthropometry and physical fitness in individuals with family history of type-2 diabetes mellitus: a comparative study. Ind J Endocrinol Metab 2011; 15: 327– 330. 21 Ekelund U, Griffin SJ, Wareham NJ. Physical activity and metabolic risk in individuals with a family history of type 2 diabetes. Diabetes Care 2007; 30: 337–342. 22 Ghosh A, Liu T, Khoury MJ, Valdez R. Family history of diabetes and prevalence of the metabolic syndrome in U.S. adults without diabetes: 6–year results from the National Health and Nutrition Examination Survey (1999–2004). Public Health Genom 2010; 13: 353–359. 23 Noble D, Mathur R, Dent T, Meads C, Greenhalgh T. Risk models and scores for type 2 diabetes: systematic review. BMJ 2011; 343: d7163. 24 Smith GD, Hart C, Blane D, Hole D. Adverse socioeconomic conditions in childhood and cause specific adult mortality: prospective observational study. BMJ 1998; 316: 1631–1635. 25 Meigs JB, Cupples LA, Wilson PW. Parental transmission of type 2 diabetes: the Framingham Offspring Study. Diabetes 2000; 49: 2201–2207. 26 Mithun D, Susil P, Arnab G. Family history of type 2 diabetes and prevalence of metabolic syndrome in adult Asian Indians. J Cardiovasc Dis Res 2012; 3: 104–108. 27 Lapidus L, Bengtsson C, Lissner L, Smith U. Family history of diabetes in relation to different types of obesity and change of obesity during 12-yr period. Results from prospective population study of women in Goteborg, Sweden. Diabetes Care 1992; 15: 1455–1458. 28 Wada K, Tamakoshi K, Yatsuya H, Otsuka R, Murata C, Zhang H et al. Association between parental histories of hypertension, diabetes and dyslipidemia and the clustering of these disorders in offspring. Prevent Med 2006; 42: 358–363. 29 Tseng CH. Effect of parental hypertension and/or parental diabetes on hypertension in Taiwanese diabetic patients. Eur J Clin Invest 2007; 37: 870–877. 30 Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK, Ferrannini E. Parental history of diabetes is associated with increased cardiovascular risk factors. Arteriosclerosis 1989; 9: 928–933. 31 Park JW, Yun JE, Park T, Cho E, Jee SH, Jang Y et al. Family history of diabetes and risk of atherosclerotic cardiovascular disease in Korean men and women. Atherosclerosis 2008; 197: 224–231. 32 Harwell TS, Dettori N, Flook BN, Priest L, Williamson DF, Helgerson SD et al. Preventing type 2 diabetes: perceptions about risk and prevention in a population-based sample of adults > or = 45 years of age. Diabetes Care 2001; 24: 2007–2008. 33 Kahn LB, Marshall JA, Baxter J, Shetterly SM, Hamman RF. Accuracy of reported family history of diabetes mellitus. Results from San Luis Valley Diabetes Study. Diabetes Care 1990; 13: 796– 798. 34 Katulanda P, Shine B, Katulanda GW, Silva A, Asfir EL, Sheriff R et al. Diabetes mellitus among young adults in Sri Lanka—role of GAD antibodies in classification and treatment: the Sri Lanka Young Diabetes study. Diabetologia 2008; 51: 1368–1374.
9
DIABETICMedicine
Family history of diabetes and disease prevalence P. Katulanda et al.
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Presence of family history in different generations in newly diagnosed patients with diabetes.
10
Table S2. Association of age, clinical and biochemical parameters with family history in patients with newly diagnosed diabetes. Table S3. Binary logistic regression analysis in all adults, males and females. Table S4. An example of a Contingency Table for Table 1.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research: Epidemiology The influence of family history of diabetes on disease prevalence and associated metabolic risk factors among Sri Lankan adults P. Katulanda1,2, P. Ranasinghe3, R. Jayawardena1,4, R. Sheriff1 and D. R. Matthews2 1 Diabetes Research Unit, Department of Clinical Medicine, Faculty of Medicine, University of Colombo, Sri Lanka, 2Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, UK, 3Department of Pharmacology, Faculty of Medicine, University of Colombo, Sri Lanka and 4Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia
Accepted 19 September 2014
Abstract Aims To describe the influence of family history on diabetes prevalence and associated metabolic risk factors in a nationally representative sample from Sri Lanka. Methods A cross sectional national survey was conducted among 5000 adults in Sri Lanka. Family history was evaluated at three levels: (1) parents, (2) grandparents (paternal and maternal) and (3) siblings. A binary-logistic regression analysis controlling for confounders (age, gender, BMI and physical activity) was performed in all patients with ‘presence of diabetes’ as the dichotomous dependent variable and using family history in father, mother, maternal grandmother/grandfather, paternal grandmother/grandfather, siblings and children as binary independent variables.
The sample size was 4485, mean age was 46.1 15.1 years and 39.5% were males. In all adults, the prevalence of diabetes was significantly higher in patients with a family history (23.0%) than those without (8.2%) (P < 0.001). When family history was present in both parents, the prevalence of diabetes was 32.9%. Presence of a family history significantly increased the risk of diabetes [odds ratio (OR): 3.35, 95% confidence interval (CI): 2.78– 4.03], obesity (OR: 2.45, 95% CI: 1.99–2.99), hypertension (OR: 1.25, 95% CI: 1.08–1.45) and metabolic syndrome (OR: 2.28, 95% CI: 1.97–2.63). In all adults, the presence of a family history of diabetes in a father (OR: 1.29, 95% CI: 1.02–1.63), mother (OR: 1.23, 95% CI: 1.11–1.36), paternal grandfather (OR: 1.27, 95% CI: 1.14–1.41), siblings (OR: 4.18, 95% CI: 3.34–5.22) and children (OR: 5.47, 95% CI: 2.93–10.19) was associated with a significantly increased risk of developing diabetes. Results
Conclusions Family history and diabetes had a graded association in the Sri Lankan population, because the prevalence increased with the increasing number of generations affected. Family history of diabetes was also associated with the prevalence of obesity, metabolic syndrome and hypertension. Individuals with a family history of diabetes form an easily identifiable group who may benefit from targeted interventions.
Diabet. Med. 00, 000–000 (2014)
Introduction Diabetes has become an important health concern in South Asia. It is estimated that the prevalence of diabetes in the region will increase about 151% between 2000 and 2030 [1]. It is well known that South Asians are at an increased risk of developing diabetes in comparison with other ethnic groups [2]. There is strong evidence to suggest that South Asians are more insulin resistant than Caucasians, even at younger ages
Correspondence to: Prasad Katulanda. E-mail: [email protected]
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
and comparative low levels of BMI, possibly due to a greater accumulation of visceral fat [3]. South Asians also experience an early decline in b–cell function, compared with other ethnic groups [3,4]. This predisposition, coupled with recent urbanization leading to unhealthy lifestyle changes, such as physical inactivity, changes in dietary habits and stress, are known to be contributory factors driving the diabetes epidemic in South Asia [5]. Because a significant number of immigrants from the region are living in affluent Western nations, a disease such as Type 2 diabetes affecting ethnic South Asians will have potential implications on global health.
1
DIABETICMedicine
Family history of diabetes and disease prevalence P. Katulanda et al.
What’s new? • There are no studies on the association between family history of diabetes and the prevalence of the disease in a large cohort of ethnic South Asian adults. • Family history and diabetes showed a graded association in Sri Lankan adults, as the prevalence increased with the increasing number of generations affected. • Family history of diabetes was also associated with the prevalence of obesity, metabolic syndrome and hypertension. • Family history could be used as a tool to detect those at risk of diabetes and a constellation of other cardiovascular risk factors. Family history is a common non-modifiable risk factor for most of the chronic non-communicable diseases, including cardiovascular disease, Type 2 diabetes and hypertension. It is a collective reflection of the genetic susceptibility, shared environments and behaviours that, when present, influences the probability of developing a disease [6]. Information on family history may serve as a unique and useful tool for public health and preventive medicine [6]. Most health screening guidelines usually include family history as a marker to assess health risks and initiate interventions. The advantages of family history as a risk assessment tool is its lower cost, greater acceptability and that it is a reflection of the shared genetic and lifestyle factors. Type 2 diabetes is a common chronic disease that results from an interaction between genetic and environmental factors. Studies have shown that the risk of Type 2 diabetes increases 2–5 times when one or both parents are affected [7,8]. Furthermore, the presence of a family history of diabetes is known to be associated with higher fasting plasma glucose, lipids, systolic blood pressure and BMI [9,10]. The increased genetic predisposition among South Asians probably makes family history more important in risk assessment than in other ethnic groups. In addition, family history reflects both shared genetic and lifestyle factors, thus it may serve as a better predictor of diabetes risk than either factor taken alone. It can be used to identify individuals at risk and influence health-promoting behaviour; in addition, prevention efforts can also be extended to the family members at risk. To our knowledge, there is currently no detailed analysis of the association between family history of diabetes and the prevalence of the disease in a large cohort of ethnic South Asian adults. This study aims to describe the influence of family history on diabetes prevalence and associated metabolic risk factors in a large cohort of South Asian adults, from a nationally representative sample from Sri Lanka.
2
Methods Study population and sampling
This study was a cross sectional community-based national survey conducted between August 2005 and September 2006 in seven of the nine provinces in Sri Lanka. Detailed sampling has been described elsewhere [11]. In brief, 5000 adults (> 18 years) were invited to take part in the study using a multistage stratified cluster sampling technique. A total of 100 clusters of 50 adults each were divided among the seven provinces using a probability proportional to size (PPS) technique based on total provincial populations. In each province, the required number of clusters was selected from a list of all ‘Village Officers Units’ in that province by using a computer-generated random number list. The voter registration lists of the selected ‘Village Officers Units’ were used to randomly select the first household in each cluster and a uniform criterion was used to select the remaining 49 households. Ethical approval for the study was obtained from the Ethics Review Committee, Faculty of Medicine, University of Colombo, Sri Lanka.
Data collection
An interviewer-administered questionnaire was used to collect data, which included: age, gender, area of residence, ethnicity, level of education, household monthly income [Sri Lankan Rupees (LKR)], duration of diabetes, family history, height, weight, waist circumference and hip circumference. Each study participant was asked by direct questioning whether any of his/her family members (living or not) had diabetes, as diagnosed by a physician. The family history was evaluated at three levels: (1) parents, (2) grandparents (paternal and maternal) and (3) siblings. If a positive family history was present, the age of onset of diabetes at each of the above levels was evaluated. The options provided were ‘yes’, ‘no’ and ‘don’t know’. For the purpose of analysis, those who said that they were unaware (‘don’t know’) of the family history were excluded. In addition, family history was grouped into the following generations: first generation, siblings; second generation, parents; and third generation, grandparents. We also looked for the presence of diabetes among the children of the study participants. Data on self-reported physical activity were collected using the short version of the International Physical Activity Questionnaire (IPAQ). Height was measured using Harpenden stadiometers (Chasmors Ltd, London, UK) to the nearest 0.1 cm. Body weight was measured using a Salter 920 digital weighing scale (Salter Ltd, Tonbridge, UK) to the nearest 0.1 kg. BMI was calculated as weight in kilograms divided by height squared in metres (kg/m2). Waist circumference was measured midway between the iliac crest and the lower rib margin at the end of normal expiration and hip
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
circumference was measured at the widest level over the greater trochanters using a plastic flexible tape to the nearest 0.1 cm. Waist-to-hip ratio and waist-to-height ratio were calculated as waist circumference divided by hip circumference and height, respectively. Seated blood pressure was measured after at least a 10 min rest using Omron IA2 digital blood pressure monitors (Omron Healthcare, Singapore). Fasting venous blood samples were obtained for glucose and lipid estimation from all participants, details of analysis have been described previously [11].
DIABETICMedicine
procedure was used and a P–value of 0.10 was considered to be the cut-off for removal of variables. A similar binary logistic regression analysis with above dependant and independent variables was also performed separately for both males and females. We present the results of the logistic regression analysis as adjusted odds ratio controlling for confounders (age, gender, BMI and physical activity). In all statistical analyses P < 0.05 was considered significant.
Results
Definitions
Sample characteristics
Participants were considered to have ‘diagnosed diabetes’ if they had been previously diagnosed at a government hospital or by a registered medical practitioner. New cases (‘undiagnosed diabetes’) were diagnosed according to the American Diabetes Association (ADA) and World Health Organization (WHO) criteria [12,13]. Pre-diabetes was also defined according to the above WHO criteria. Young-onset diabetes was defined as those who develop diabetes at an age < 40 years. Hypertension was defined as systolic blood pressure > 130 mmHg and/or diastolic blood pressure > 85 mmHg and/or being on anti-hypertensive treatment. Central obesity was classified as a waist circumference > 90 cm for males and > 80 cm for females (Asian cut-offs) [14]. Obesity was defined as a BMI ≥ 25 kg/m2, based on criteria for Asians [14]. Metabolic syndrome was defined according to the International Diabetes Federation (IDF) criteria [14]. Urban and rural sectors were defined according to the classification of the Sri Lankan government. Physical activity was classified in to three categories (‘inactive’, ‘moderately active’ and ‘highly active’) based on the total metabolic equivalent of task (MET) minutes/week derived from the IPAQ short version [11]. Peripheral neuropathy was assessed using the validated Toronto Clinical Scoring System (TCSS) (score > 5) [15]. Ophthalmological evaluation for retinopathy was performed with dilated indirect ophthalmoscopy using slit lamp biomicroscopy.
Of the 5000 invited individuals, 4485 participated in the study (response rate 89.7%). Mean age ( SD) was 46.1 15.1 years and 39.5% (n = 1772) were males. Most were residing in rural areas (n = 3530, 78.7%) and Sinhalese in ethnicity (n = 3877, 86.4%). The crude prevalence of diabetes was 12.0% [n = 536, 95% confidence interval (CI): 11.0–12.9], of which 344 (64.2%) were patients with ‘diagnosed diabetes’. The prevalence of obesity (BMI ≥ 25 kg/m2) was 9.8% (n = 438, 95% CI: 8.9–10.6), and central obesity (based on waist circumference) was present in 30.3% (n = 1358, 95% CI: 28.9–31.6). Pre-diabetes was prevalent in 14.6% (n = 577, 95% CI: 13.5–15.7), and 27.2% (n = 1219, 95% CI: 26.1–28.7) had hypertension and 26.6% (n = 1193, 95% CI: 25.3–27.9) had metabolic syndrome.
Statistical analysis
Data were analysed using SPSS v14 (SPSS Inc., Chicago, IL, USA) statistical software package. The significance of the differences between proportions and means was tested using a z-test and Student’s t-test or analysis of variance (ANOVA), respectively. A binary logistic regression analysis was performed in all patients with ‘presence of diabetes’ as the dichotomous dependent variable (0 = diabetes absent; 1 = diabetes present) and using family history in father, mother, maternal grandmother/grandfather, paternal grandmother/grandfather, siblings and children as the binary independent variables (0 = No, 1 = Yes). The explanatory variables selected above were subsequently included in a binary logistic regression model, a backward elimination
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Family history, prevalence of diabetes and metabolic risk factors
The overall prevalence of a family history of diabetes in the population was 25.5% (n = 1141, 95% CI: 24.0–27.0), irrespective of disease status. A family history of diabetes (parents, grandparents or siblings) was present in 49.0% (n = 262, 95% CI: 45.0–53.0) of patients with diabetes and 22.3% (n = 879, 95% CI: 21.0–24.0) participants without diabetes. In all adults, the prevalence of diabetes was significantly higher in patients with a family history (23.0%, 95% CI: 20.5–25.4) than in those without a family history (8.2%, 95% CI: 7.2–9.1) (P < 0.001). A similar result was observed independently in both males (23.7% vs. 7.4%) and females (22.6% vs. 8.7%). Presence of a family history significantly increased the risk of diabetes [odds ratio (OR): 3.35; 95% CI 2.78–4.03]. The prevalence of diabetes was higher in those with a family history of diabetes at all levels (mother, father, maternal grandmother/grandfather, paternal grandmother/ grandfather, siblings and children) than those without a family history (Table 1) (An example for a contingency table for Table 1 is provided in the Supplementary file as Table S4.) However, this difference was significant only in those with a family history in mother, father, maternal grandfather, siblings and children. There was no significant difference between the prevalence of diabetes in those with a family
3
Family history of diabetes and disease prevalence P. Katulanda et al.
DIABETICMedicine
history in the father (20.8%, 95% CI: 16.3–25.3) or mother (22.3%, 95% CI: 18.6–26.0). When family history of diabetes was present in both parents, the prevalence of diabetes was 32.9% (95% CI: 21.8–43.9). Among the different ethnicities, the presence of a family history was highest in Muslims (47.0%), followed by Sinhalese (24.5%) and Tamils (16.2%). The prevalence of diabetes in those with a family history was higher in each generation than in those without a family history [first generation (siblings), 34.2% vs. 9.1%, P < 0.001; second generation (parent), 20.6% vs. 10.3%, P < 0.001; third generation (grandparents) 13.3% vs. 11.9%, P = 0.338]. In addition, the prevalence of diabetes also increased with the number of affected generations (one generation, 33.9%; two generations, 34.2%; and three generations, 37.1%; P = 0.608). In patients without diabetes, obesity (34.5% vs. 17.7%; P < 0.001), central obesity (46.7% vs. 24.6%; P < 0.001) and metabolic syndrome (30.6% vs. 20.2%; P < 0.001) were more prevalent in those with a family history of diabetes than in those without. In this group, presence of family history of diabetes increased the risk of obesity (OR: 1.87; 95% CI 1.62–2.17), central obesity (OR: 1.95; 95% CI 1.74–2.18) and metabolic syndrome (OR: 1.52; 95% CI 1.34–1.72). Similarly in patients with diabetes, the prevalence of obesity (61.4% vs. 46.7%; P < 0.01), central obesity (57.4% vs. 39.4%; P < 0.001) and metabolic syndrome (52.3% vs. 39.9%; P < 0.01) were all significantly higher in those with a family history of diabetes than in those without. In patients with diabetes, presence of family history of diabetes increased the risk of obesity (OR: 1.32; 95% CI 1.08– 1.60), central obesity (OR: 1.45; 95% CI 1.21–1.75) and metabolic syndrome (OR: 1.31; 95% CI 1.05–1.64).
Association of age of onset, clinical and biochemical parameters with family history
The age at diagnosis of diabetes was significantly lower in patients with a family history (48.9 11.6 years) than in
those without (52.6 12.3 years) (P < 0.01). Table 2 summarizes the association between age, clinical and biochemical parameters with family history in patients with and without diabetes. In both patients with diabetes and without diabetes, those with a family history were significantly younger and had a higher mean BMI, waist circumference, hip circumference and fasting blood glucose. However, in the participants without diabetes, in addition to the above-mentioned variables, those with a family history also had a higher total cholesterol, LDL cholesterol, triglycerides and lower HDL cholesterol (Table 2). In patients with diabetes the BMI, waist and hip circumferences increased significantly with the increasing number of generations affected by diabetes (Table 3). A similar trend was also observed among the participants without diabetes. Furthermore, in the patients with diabetes, an increase was also observed in systolic and diastolic blood pressures, fasting blood glucose, total cholesterol and triglycerides with the increasing number of generations affected by diabetes, although this difference was not significant (Table 3). By contrast, in patients without diabetes, there was a significant decrease in total cholesterol, LDL cholesterol and triglycerides with increasing number of generations (Table 3). In those with a family history of diabetes, the mean age of onset ( SD) of diabetes in the fathers, mothers, brothers, sisters, sons and daughters was 57.5 ( 10.6), 58.2 ( 11.8), 45.6 ( 9.2), 46.1 ( 9.3), 37.9 ( 7.8) and 41.5 ( 8.5) years, respectively. In patients with diabetes, a significant correlation was observed between the age of onset of diabetes and the age of onset in the mother (r = 0.25, P < 0.05), brother (r = 0.56, P < 0.001), sister (r = 0.60, P < 0.001), son (r = 0.79, P < 0.001) and daughter (r = 0.72, P < 0.01). In patients with young-onset diabetes (< 40 years), the age of onset of diabetes in brothers (38.3 6.6 vs. 47.6 9.3 years) and sisters (37.7 8.9 vs. 48.4 8.2 years) was significantly lower than in patients developing diabetes at > 40 years of age. The age of onset of diabetes in the patients decreased significantly with the
Table 1 Presence of family history in different generations in patients with and without diabetes
Mother Father Maternal grandmother Maternal grandfather Paternal grandmother Paternal grandfather Siblings Children
Patients with diabetes (n = 535)
Patients without diabetes (n = 3947)
Number with family history (prevalence of diabetes)
Number with family history (% without diabetes)
Present
Absent
Not known
P*
Present
Absent
Not known
P†
107 66 15 10 9 7 175 24
399 434 333 327 323 317 360 511
29 35 186 197 202 210 0 0
< 0.001 < 0.001 0.131 < 0.05 0.377 0.178 < 0.001 < 0.001
373 251 97 46 72 42 336 25
3525 3637 3096 3094 3044 3048 3610 3922
49 59 753 806 830 856 0 0
< 0.001 < 0.001 0.131 < 0.05 0.377 0.178 < 0.001 < 0.001
(22.3) (20.8) (13.4) (17.9) (11.1) (14.3) (34.2) (49.0)
(10.2) (10.7) (9.7) (9.6) (9.6) (9.4) (9.1) (11.1)
(37.2) (37.2) (19.8) (19.6) (19.6) (19.7)
(77.7) (79.2) (86.6) (82.1) (88.9) (85.7) (65.8) (51.0)
(89.8) (89.3) (90.3) (90.4) (90.4) (90.6) (90.9) (88.5)
(62.8) (62.8) (80.2) (80.4) (80.4) (80.3)
*Comparison between prevalence of diabetes in those with and without a family history. Comparison between the percentage without diabetes in those with and without a family history.
†
4
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Table 2 Association of age, clinical and biochemical parameters with family history
Age (years) BMI (kg/m2) Waist circumference (cm) Hip circumference (cm) Waist-to-hip ratio Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Fasting blood glucose (mg/dl) Total cholesterol (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglycerides (mg/dl)
Patients with diabetes (n = 535)
Patients without diabetes (n = 3947)
Family history
Family history
Present Mean ( SD)
Absent Mean ( SD)
53.6 24.5 86.8 94.7 0.91 136.2 80.3 145.9 215.3 137.9 45.9 157.2
56.3 23.2 82.7 91.0 0.91 138.1 79.5 135.7 219.8 141.3 46.3 161.0
( ( ( ( ( ( ( ( ( ( ( (
11.5) 4.1) 10.4) 9.3) 0.07) 19.9) 11.3) 58.7) 40.6) 36.2) 9.2) 75.2)
( ( ( ( ( ( ( ( ( ( ( (
P*
13.1) 3.9) 10.9) 8.6) 0.07) 23.3) 11.9) 55.9) 48.4) 42.3) 10.3) 97.8)
< < <
70
Age in years
(b)
70
(b)
40
60
35
50
30
Prevalence
Prevalence of Diabetes
45
25 20 15
30 20
10
10
5 0 25
Body Mass Index (kgm2) Family History - Present
Family History - Absent
FIGURE 1 Prevalence of diabetes by (a) age category and (b) BMI in patients with and without a family history of diabetes.
result was observed in the participants without diabetes (3817.78 vs. 5248.01 min, P < 0.001). Patients with diabetes with a family history were more ‘inactive’ (24.8% vs. 13.9%, P < 0.01) and less ‘highly active’ (38.9% vs. 45.8%, P < 0.01) when self-reported physical activity was classified in to the three categories based on the total MET minutes/week. Even in patients without diabetes, those with a family history were more ‘inactive’ (13.5% vs. 8.9%, P < 0.001) and less ‘highly active’ (51.5% vs. 65.6%, P < 0.001). Family history and complications of diabetes
The prevalence of macrovascular complications of diabetes was higher among patients with diabetes with a family history than in those without; ischaemic heart disease (16.2% vs. 14.6%, P = 0.629), stroke/TIA (4.6% vs. 2.2%, P = 0.153) and peripheral vascular disease (5.4% vs. 4.5%, P = 0.803). However, of the microvascular complications, only nephropathy was more prevalent in the patients with a family history; nephropathy (8.1% vs. 3.2%, P < 0.05), retinopathy (26.8% vs. 28.1%, P = 0.893) and neuropathy (68.5% vs. 73.5%, P = 0.545). Erectile impotence in males with diabetes was more prevalent in patients with a family history (8.1%) than in those without (3.3%) (P < 0.05).
6
40
0 25
Body Mass Index (kgm-2) Hypertension Metabolic Syndrome
Family history - Present
Family history - Absent
Family history - Present
Family history - Absent
FIGURE 2 Prevalence of hypertension and metabolic syndrome by (a) age category and (b) BMI in patients with and without a family history of diabetes.
Logistic regression analysis
The results of the binary logistic regression analysis in all adults using the dichotomous variable ‘presence of diabetes’ as the dependent factor and other independent variables are shown in Table 4. The overall model was statistically significant and the Cox & Snell R-square and Nagelkerke R-square values were 0.064 and 0.123, respectively. The results indicate that in all adults, the presence of a family history of diabetes in father (OR: 1.29), mother (OR: 1.23), both parents (OR: 1.64), paternal grandfather (OR: 1.27), siblings (OR: 4.18) and children (OR: 5.47) were associated with significantly increased risk of developing diabetes (Table 4). Family history in siblings and children were also associated with developing diabetes in both males and females independently (Table 4). However, family history in father (OR: 1.58) was associated with increased risk of diabetes only in males, whereas in females, it was the family history in the mother (OR: 1.35) that significantly increased the risk.
Discussion To our knowledge, this is the first comprehensive survey evaluating the association between family history of diabetes,
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Table 4 Binary logistic regression analysis in all adults, males and females Adjusted odds ratio (95% CI)a Co-variants (Family history)
All adults
Father Mother Both parents Paternal grandfather Maternal grandfather Siblings Children
1.29 1.23 1.64 1.27 0.99 4.18 5.47
(1.02–1.63)‡ (1.11–1.36)† (1.10–2.10)‡ (1.14–1.41)* (0.77–1.27) (3.34–5.22)* (2.93–10.19)*
Male 1.58 1.42 0.73 2.19 0.78 5.73 5.82
Female (1.11–2.22)† (0.84–2.39) (0.22–2.46) (1.30–3.69)† (0.53–1.14) (3.95–8.31)* (1.94–17.44)†
1.18 1.35 1.31 1.08 1.26 3.39 4.66
(0.72–1.94) (1.18–1.54)* (0.56–3.05) (0.78–1.49) (1.09–1.45)† (2.56–4.51)* (2.13–10.21)*
*P < 0.001, †P < 0.01, ‡P < 0.05; a, adjusted odds ratio controlling for confounders (age, gender, BMI and physical activity).
disease prevalence and metabolic risk factors from a nationally representative, large cohort of ethnic South Asian adults from Sri Lanka. Our results clearly demonstrate that the family history of diabetes has a graded association with diabetes, because the prevalence of diabetes increased with the number of generations with a family history. In the current cohort, those with a family history of diabetes were 1.2–5 times more likely to develop diabetes than those without a family history. Previous studies from Germany, Norway, South Africa, Taiwan and the USA have shown similar increased risks, in which compared with people without a family history of diabetes, those who have a family history were 2–6 times more likely to develop Type 2 diabetes [16]. However, the observed risks were much higher in European populations than reported in our cohort [8]. The risk associated with family history in the current population appeared to be independent of other known risk factors, including age, anthropometric parameters (BMI) and lifestyle factors (physical activity). The InterAct study, a large prospective case–cohort study involving eight European countries, also corroborated these findings, where prominent lifestyle (diet and physical activity) and anthropometric (BMI, waist and hip circumference) risk factors explained only a marginal proportion of the excess risk associated with family history [8]. These findings further enumerate the fact that family history is a strong and independent risk factor for diabetes [17]. Furthermore, family history and diabetes showed a graded association, because the prevalence increased with the increasing number of generations affected. A similar study in the USA showed that the strength of the association was related to the type and number of relatives involved [18]. Participants with a family history of diabetes displayed a significantly higher BMI, waist and hip circumference than those without a family history, irrespective of diabetes status, a finding that is in keeping with other similar studies from Europe [19]. In another recent study from India, it was found that apparently healthy individuals with a family history of diabetes had higher anthropometric values than their counterparts [20]. Furthermore, the anthropometric parameters in the present cohort showed a graded increase with the ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
increasing number of generations affected by diabetes. We also observed that, irrespective of disease status, participants with a family history had a lower total weekly MET minutes and were more ‘inactive’ in self-reported physical activity than those without a family history. Objectively measured physical activity is known to be associated with insulin resistance and dyslipidaemia independent of sex, age and obesity in those with a family history of diabetes [21]. These findings further support the hypothesis that expression of complex traits like diabetes results from an interaction between shared genes, environments and behaviours [22]. Furthermore, individuals with a family history of diabetes form an easily identifiable group who may benefit from targeted intervention to prevent the development of diabetes through increased physical activity. Hence, family history of diabetes is included in most diabetes risk scores [23]. Patients with a family history were both younger in age and had a younger age of onset of diabetes than patients without a family history of diabetes, a finding which is consistent with previous reports among European adults [19]. In addition, the age of onset showed an inverse relationship with the increasing number of generations with a family history of diabetes. We also observed that the age of onset of diabetes in patients was closely correlated with the age of onset in siblings and children, rather than with the age of onset in parents. These findings suggest that the age of diabetes onset might be genetically determined. However, another simpler/alternative explanation could be that those with a family history of diabetes are more aware about the disease and have frequent contact with medical practitioners, leading to an earlier diagnosis. Our data showed that a history of diabetes in the sibling was associated with a higher risk of diabetes than parental history. Sibling pairs have higher shared environmental factors than parents and offspring, and some common environments in young childhood have been postulated to be associated with future adult diseases [24]. Furthermore, we did not observe a difference in the prevalence of diabetes in patients with a family history in either mother or father. However, previous studies from different parts of the world have shown that there is an excess maternal transmission of diabetes [19]. A finding
7
DIABETICMedicine
Family history of diabetes and disease prevalence P. Katulanda et al.
refuted in the Framingham population, where maternal and paternal diabetes conferred equal risk for overt diabetes among offspring, similar to the findings in the present study [25]. Family history of diabetes is associated with the prevalence of obesity, metabolic syndrome and hypertension. In a cohort of Asian Indians, a positive family history of diabetes was significantly associated with the prevalence of metabolic syndrome [26]. A study conducted in Sweden demonstrated that family history of diabetes was not only related to the presence of obesity on one occasion, but also to an examination 12 years later [27]. Previous studies have also demonstrated an association between family history of diabetes and the presence of hypertension and the clustering of these disorders in the offspring of affected patients [28]. However, a study in Taiwanese patients, found that a parental family history of diabetes conferred a lower risk of hypertension in Type 2 diabetes, suggesting the need for further study of blood pressure control in relation to family history [29]. In a study by Haffner et al., people with a positive parental history of diabetes had a more atherogenic pattern of cardiovascular risk factors compared with those without a parental history of diabetes [30]. A similar result was observed in the present cohort, where in patients with and without diabetes, the BMI, waist and hip circumferences increased significantly with the increasing number of generations affected by diabetes. Furthermore, although statistically not significant in the patients with diabetes, an increase was also observed in systolic and diastolic blood pressure, fasting blood glucose, total cholesterol and triglycerides with the increasing number of generations affected by diabetes. The genetic factors associated with the occurrence of familial diabetes may increase risk of cardiovascular disease beyond the risk among people without a family history [31]. Hence, family history could be used as a tool to detect not only those at risk of diabetes, but also a constellation of other cardiovascular risk factors, and they might form a group of individuals that benefit the most from targeted risk factor interventions. Family history is a common non-modifiable risk factor for most chronic non-communicable diseases, as it is a collective reflection of the genetic susceptibility, shared environments and behaviours [6]. Hence, information on family history may serve as a unique and useful tool for public health and preventive medicine [6]. The advantages of family history as a risk assessment tool are its lower cost, greater acceptability, and that it is a reflection of shared genetic and lifestyle factors. Although family history by itself is a non-modifiable risk factor, it is useful for raising awareness of risk, risk stratification, targeting interventions, and positively influencing health behaviours [16]. Risk perception is one factor that positively influences health-related behaviour change in relation to diabetes prevention [32]. There is enough evidence from randomized controlled trials and observational studies to show that Type 2 diabetes can be prevented or delayed by adopting simple, healthy lifestyle changes.
8
Accurate risk perception might positively influence behaviour change, and aid in prevention of diabetes [16]. Hence, overall there is promising potential for the use of family history as a public health tool aiding prevention of diabetes. The strengths of the present study include the large and nationally representative nature of the sample, random selection of participants from a well defined and homogenous target population, the high response rate and the detailed nature of the clinical and demographic assessment using well validated tools. The major limitation of the study was the, cross sectional nature. The usefulness of family history in risk prediction should be tested in large prospective studies. Furthermore, in developing countries like Sri Lanka, large sections of the community remain undiagnosed and therefore the accuracy of self-reported family history is a challenge. Other potential limitations include the recall bias for family history, however, previous studies have shown that this is minimal in diabetes [33]. In order to minimize recall bias, we have carried out a separate analysis on patients with newly diagnosed diabetes, which is presented in the Supporting Information (Tables S1–S4). The prevalence of diabetes in newly diagnosed patients was significantly higher in those with a family history in the mother or in siblings (Tables S1– S4). However, in the binary logistic analysis, only a family history in siblings was shown to be significantly associated with the presence of diabetes among newly diagnosed patients. Furthermore, newly diagnosed patients with a family history were younger in age, and had a higher BMI, waist and hip circumference. We were also unable to accurately assess the impact of family history on glycaemic control, because data on HbA1c values were not available for the majority of the population. In this study, we have made no distinction between Type 1 and Type 2 diabetes in the study population. However, previous large-scale studies in Sri Lanka have shown that nearly 95% of Sri Lankan population with diabetes are having Type 2 diabetes.
Funding sources
The National Science Foundation (NSF) of Sri Lanka.
Competing interests
None declared.
Acknowledgements
The National Science Foundation (NSF) of Sri Lanka was the main source of funding for the study. The additional support provided from the Oxford Centre for Diabetes Endocrinology and Metabolism UK, and the NIHR Biomedical Research Centre Programme is gratefully acknowledged. We thank the Diabetes Association of Sri Lanka and the World Health Organization Office in Colombo for the support for lipid assays.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
References 1 Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004; 27: 1047–1053. 2 Mather HM, Keen H. The Southall Diabetes Survey: prevalence of known diabetes in Asians and Europeans. BMJ 1985; 291: 1081– 1084. 3 Gujral UP, Pradeepa R, Weber MB, Narayan KM, Mohan V. Type 2 diabetes in South Asians: similarities and differences with white Caucasian and other populations. Ann NY Acad Sci 2013; 1281: 51–63. 4 Kooner JS, Saleheen D, Sim X, Sehmi J, Zhang W, Frossard P et al. Genome-wide association study in individuals of South Asian ancestry identifies six new type 2 diabetes susceptibility loci. Nat Genet 2011; 43: 984–989. 5 Jayawardena R, Ranasinghe P, Byrne NM, Soares MJ, Katulanda P, Hills AP. Prevalence and trends of the diabetes epidemic in South Asia: a systematic review and meta-analysis. BMC Public Health 2012; 12: 380. 6 Yoon PW, Scheuner MT, Peterson-Oehlke KL, Gwinn M, Faucett A, Khoury MJ. Can family history be used as a tool for public health and preventive medicine? Genet Med 2002; 4: 304–310. 7 Alcolado JC, Alcolado R. Importance of maternal history of noninsulin dependent diabetic patients. BMJ 1991; 302: 1178–1180. 8 InterAct C. The link between family history and risk of type 2 diabetes is not explained by anthropometric, lifestyle or genetic risk factors: the EPIC-InterAct study. Diabetologia 2013; 56: 60–69. 9 Mitchell BD, Valdez R, Hazuda HP, Haffner SM, Monterrosa A, Stern MP. Differences in the prevalence of diabetes and impaired glucose tolerance according to maternal or paternal history of diabetes. Diabetes Care 1993; 16: 1262–1267. 10 Viswanathan M, McCarthy MI, Snehalatha C, Hitman GA, Ramachandran A. Familial aggregation of type 2 (non-insulin-dependent) diabetes mellitus in south India; absence of excess maternal transmission. Diabetes Med 1996; 13: 232–237. 11 Katulanda P, Constantine GR, Mahesh JG, Sheriff R, Seneviratne RDA, Wijeratne S et al. Prevalence and projections of diabetes and pre-diabetes in adults in Sri Lanka—Sri Lanka Diabetes, Cardiovascular Study (SLDCS). Diabetes Medicine 2008; 25: 1062–1069. 12 American Diabetes Association. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997; 20: 1183–1197. 13 World Health Organization. Part 1: diagnosis and classification of diabetes mellitus. Definition, diagnosis and classification of diabetes mellitus and its complications. Geneva: WHO 1999. 14 Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009; 120: 1640–1645. 15 Perkins BA, Olaleye D, Zinman B, Bril V. Simple screening tests for peripheral neuropathy in the diabetes clinic. Diabetes Care 2001; 24: 250–256. 16 Harrison TA, Hindorff LA, Kim H, Wines RC, Bowen DJ, McGrath BB et al. Family history of diabetes as a potential public health tool. Am J Prevent Med 2003; 24: 152–159. 17 Valdez R, Yoon PW, Liu T, Khoury MJ. Family history and prevalence of diabetes in the U.S. population: the 6–year results from the National Health and Nutrition Examination Survey (1999–2004). Diabetes Care 2007; 30: 2517–2522.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
DIABETICMedicine
18 Annis AM, Caulder MS, Cook ML, Duquette D. Family history, diabetes, and other demographic and risk factors among participants of the National Health and Nutrition Examination Survey 1999–2002. Prev Chron Dis 2005; 2: A19. 19 Papazafiropoulou A, Sotiropoulos A, Skliros E, Kardara M, Kokolaki A, Apostolou O et al. Familial history of diabetes and clinical characteristics in Greek subjects with type 2 diabetes. BMC Endocrin Disord 2009; 9: 12. 20 Padaki S, Vijayakrishna K, Dambal A, Ankad R, Manjula R, Surekharani C et al. Anthropometry and physical fitness in individuals with family history of type-2 diabetes mellitus: a comparative study. Ind J Endocrinol Metab 2011; 15: 327– 330. 21 Ekelund U, Griffin SJ, Wareham NJ. Physical activity and metabolic risk in individuals with a family history of type 2 diabetes. Diabetes Care 2007; 30: 337–342. 22 Ghosh A, Liu T, Khoury MJ, Valdez R. Family history of diabetes and prevalence of the metabolic syndrome in U.S. adults without diabetes: 6–year results from the National Health and Nutrition Examination Survey (1999–2004). Public Health Genom 2010; 13: 353–359. 23 Noble D, Mathur R, Dent T, Meads C, Greenhalgh T. Risk models and scores for type 2 diabetes: systematic review. BMJ 2011; 343: d7163. 24 Smith GD, Hart C, Blane D, Hole D. Adverse socioeconomic conditions in childhood and cause specific adult mortality: prospective observational study. BMJ 1998; 316: 1631–1635. 25 Meigs JB, Cupples LA, Wilson PW. Parental transmission of type 2 diabetes: the Framingham Offspring Study. Diabetes 2000; 49: 2201–2207. 26 Mithun D, Susil P, Arnab G. Family history of type 2 diabetes and prevalence of metabolic syndrome in adult Asian Indians. J Cardiovasc Dis Res 2012; 3: 104–108. 27 Lapidus L, Bengtsson C, Lissner L, Smith U. Family history of diabetes in relation to different types of obesity and change of obesity during 12-yr period. Results from prospective population study of women in Goteborg, Sweden. Diabetes Care 1992; 15: 1455–1458. 28 Wada K, Tamakoshi K, Yatsuya H, Otsuka R, Murata C, Zhang H et al. Association between parental histories of hypertension, diabetes and dyslipidemia and the clustering of these disorders in offspring. Prevent Med 2006; 42: 358–363. 29 Tseng CH. Effect of parental hypertension and/or parental diabetes on hypertension in Taiwanese diabetic patients. Eur J Clin Invest 2007; 37: 870–877. 30 Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK, Ferrannini E. Parental history of diabetes is associated with increased cardiovascular risk factors. Arteriosclerosis 1989; 9: 928–933. 31 Park JW, Yun JE, Park T, Cho E, Jee SH, Jang Y et al. Family history of diabetes and risk of atherosclerotic cardiovascular disease in Korean men and women. Atherosclerosis 2008; 197: 224–231. 32 Harwell TS, Dettori N, Flook BN, Priest L, Williamson DF, Helgerson SD et al. Preventing type 2 diabetes: perceptions about risk and prevention in a population-based sample of adults > or = 45 years of age. Diabetes Care 2001; 24: 2007–2008. 33 Kahn LB, Marshall JA, Baxter J, Shetterly SM, Hamman RF. Accuracy of reported family history of diabetes mellitus. Results from San Luis Valley Diabetes Study. Diabetes Care 1990; 13: 796– 798. 34 Katulanda P, Shine B, Katulanda GW, Silva A, Asfir EL, Sheriff R et al. Diabetes mellitus among young adults in Sri Lanka—role of GAD antibodies in classification and treatment: the Sri Lanka Young Diabetes study. Diabetologia 2008; 51: 1368–1374.
9
DIABETICMedicine
Family history of diabetes and disease prevalence P. Katulanda et al.
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Presence of family history in different generations in newly diagnosed patients with diabetes.
10
Table S2. Association of age, clinical and biochemical parameters with family history in patients with newly diagnosed diabetes. Table S3. Binary logistic regression analysis in all adults, males and females. Table S4. An example of a Contingency Table for Table 1.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
