Prevalences of overweight and obesity among children in remote Aboriginal communities in central Australia

Introduction: The chronic diseases associated with overweight and obesity are major contributors to the excess disease burden of Aboriginal Australians. Surveillance of overweight and obesity is required to monitor these conditions, and to develop and evaluate interventions to improve health and wellbeing. Remote Aboriginal communities in Australia’s Northern Territory (NT) are where approximately two-thirds of the NT Aboriginal people live, a proportion which has been stable over many years. However the remote communities suffer significant socioeconomic disadvantage including limited education and employment opportunities, and poor quality and overcrowded housing. Approximately one-third of Aboriginal people in NT live in central Australia, which consists of the Alice Springs and Barkly districts. The Healthy School-Aged Kids Program includes health promotion and child health screening, and is run in remote Aboriginal communities of NT. This report provides estimates of prevalences of overweight and obesity among children in central Australia who participated in health checks as part of Healthy School-Aged Kids Program in 2010. Methods: All children in remote central Australian Aboriginal communities were invited to participate in health checks. Children who attended were weighed and measured. Date of birth, sex, height and weight for each child were used to determine prevalence of overweight (≥+1 standard deviation [SD] BMI-for-age) and obesity (≥+2 SD BMI-for-age) according to WHO Growth Standards. Differences in proportions of overweight and obesity by age group and sex, and their statistical significance were calculated.


Introduction:
The chronic diseases associated with overweight and obesity are major contributors to the excess disease burden of Aboriginal Australians.Surveillance of overweight and obesity is required to monitor these conditions, and to develop and evaluate interventions to improve health and wellbeing.Remote Aboriginal communities in Australia's Northern Territory (NT) are where approximately two-thirds of the NT Aboriginal people live, a proportion which has been stable over many years.However the remote communities suffer significant socioeconomic disadvantage including limited education and employment opportunities, and poor quality and overcrowded housing.Approximately one-third of Aboriginal people in NT live in central Australia, which consists of the Alice Springs and Barkly districts.The Healthy School-Aged Kids Program includes health promotion and child health screening, and is run in remote Aboriginal communities of NT.This report provides estimates of prevalences of overweight and obesity among children in central Australia who participated in health checks as part of Healthy School-Aged Kids Program in 2010.Methods: All children in remote central Australian Aboriginal communities were invited to participate in health checks.Children who attended were weighed and measured.Date of birth, sex, height and weight for each child were used to determine prevalence of overweight (≥+1 standard deviation [SD] BMI-for-age) and obesity (≥+2 SD BMI-for-age) according to WHO Growth Standards.Differences in proportions of overweight and obesity by age group and sex, and their statistical significance were calculated.

Introduction
High prevalences of chronic diseases at young ages in remote Northern Territory (NT) Aboriginal populations have been documented.For example, by 15 years of age 8.6% have renal disease, 0.4% have type 2 diabetes and 0.3% have ischaemic heart disease 1 .Compared with all Australians the overall burden of disease suffered by the Aboriginal people of NT is close to four times higher; much of this excess is due to ischaemic heart disease, renal disease and diabetes, for which overweight and obesity are important risk factors 2 .This report contributes to the limited published data on population prevalence of overweight and obesity in Aboriginal children, with community based data on children in remote communities in central Australia.
Remote Aboriginal communities in NT have populations of from less than 50 to more than 1000 people, and over 90% of people in these communities are Aboriginal 3 .These communities are classified by Australian Remoteness Index for Areas (ARIA+) as Very Remote, indicating the highest levels of remoteness from service centres [3][4][5] .The populations of these communities are growing, leading to increasing levels of overcrowding and limited educational and employment opportunities 3,6 .However approximately twothirds of Aboriginal people in NT live in these communities rather than move to urban centres, and this proportion is not declining 7 .
In this report, 'central Australia' consists of Alice Springs and Barkly districts of NT, where 20 000 or 31.3% of the NT Aboriginal people live 3 .
The Healthy School-Aged Kids Program is a joint initiative of the NT Departments of Health and Education, run throughout remote Aboriginal communities in NT.It is a population health program involving health promotion, service integration and health checks.An evaluation was performed in 1998, leading to a revision of the manual published in 2007 8,9 .The manual provides detailed guidelines for child health and remote area nurses and Aboriginal health workers, including health checks to identify and manage common health problems in children aged 5

Methods
The Healthy School-Aged Kids Program entails engagement of health staff and community Elders, teachers and parents of children in remote communities.Health checks are offered to every child aged 5-15 years.Parents or guardians are asked to give written consent for their children to be checked.Children not at school are encouraged to participate 9 .
Participating children are measured by nurses and Aboriginal health workers using stadiometers and digital scales.The guidelines include measuring and weighing children at 5,10 and 15 years 9 .Weight is measured to 0.1 kg, and height to the nearest centimetre.Equipment is calibrated monthly 11 .For this report, the children's date of birth, sex, height and weight were entered into WHO AnthroPlus application (http://www.who.int/growthref/tools/en/) to calculate BMI and BMI-for-age z-scores based on WHO 2007 growth standards.The significance of the difference between proportions was calculated using two-tailed z-tests for independent proportions, utilising the VassarStats (http:// faculty.vassar.edu/lowry/VassarStats.html) statistical calculator.

Results
The program ran between March and October 2010.Participating children's ages ranged from 3 to 17 years, although guidelines recommend measuring weight and height only at 5, 10 and 15 years 9 .
The BMI-for-age data were available for 996 children in 26 communities, with between 4 and 109 children measured in each community.This included 883 in the target population of 1764 children aged 5 to 15 years, giving 50% coverage among these children.It was found that 21.4% of children were overweight (≥+1.0 standard deviation [SD] BMI-forage; 95% CI 18.6-24.5%)and 5.4% were obese (≥+2.0SD BMI-for-age; 95% CI 4.2-7.0%).No community had BMIfor-age distribution significantly different from the overall distribution.
The BMI-for-age according to age groups and sex is shown (Table 1).Overweight increased with age (z=3.28,p=0.0011), while obesity did not (z=0.68;p=0.50).There was no difference in the prevalence of overweight or obesity between boys and girls (overweight z=0.52; p=0.60; for obesity z=0.50; p=0.62).

Discussion
Health checks performed on children in remote Aboriginal communities in 2010 showed that 21.4% of children were overweight and 5.4% obese.Other cardiovascular risk factors were not assessed in the Healthy School Aged Kids Program.
Comparative data from other recent studies of prevalence of overweight and obesity among Australian Aboriginal children shows a diversity of results (Table 2).Data from the NTER defined 'overweight' as BMI-for-age >95th centile, which is consistent with other definitions of obesity.The prevalence of overweight reported in the NTER was significantly higher than the prevalence of obesity in this study 12 .The lower prevalence of overweight and obesity among the remotedwelling children in the Aboriginal Birth Cohort Study is also notable because there are lower prevalences of chronic diseases among Aboriginal people in the 'Top End' (the northern region of NT) compared with central Australian Aboriginal people of NT, although different standards for overweight and obesity were used 1,[13][14][15] .Alternatively a lower prevalence reported in the Aboriginal Birth Cohort Study approximately 10 years before the present study may reflect an increased prevalence of overweight and obesity over the past 10 years.The higher prevalence of obesity among the children at Inala in Brisbane may reflect differences between urban and rural populations noted in the Aboriginal Birth Cohort Study 13,16 .
The 2007-2008 National Health Survey identified that 25% of children aged 5 to 17 years were overweight, including 8% obese.Data on Aboriginal children were not reported separately 18 .

Conclusion
The high burden of chronic diseases attributable to overweight among Aboriginal people in central Australia is a devastating problem.Rates of childhood overweight and obesity described here do not reflect the four-times greater burden of disease in this population compared with all Australians 2 .Possible reasons for the discrepancy include BMI-for-age not adequately predicting future cardiovascular risk in this population; selective non-participation of overweight children; and increasing overweight after school age.The NHMRC Clinical practice guidelines for management of overweight and obesity in children and adolescents recommended further research to determine whether children's BMI-for-age correlates with medical morbidity, and that Aboriginal children may need separate consideration 20 .These Healthy School Aged Kids Program health check data results support this recommendation, and the proposition that waist circumference may be a better predictor of cardiovascular risk than BMI-for-age in this population 21,22 .
Attention to the underlying causes of the disease burden of Aboriginal people in remote central Australia includes measures to promote healthy eating and physical activity.A review of current Healthy School-Aged Kids Program guidelines and implementation, particularly the monitoring of waist circumference, should be considered.In the absence of standardised survey methods, conclusions about relative rates of overweight and obesity in different populations of Aboriginal and non-Aboriginal children cannot be drawn.Aboriginal children in particular are diverse populations.
-15 years.The recommended checks include height and weight of children aged 5, 10 and 15 years 9 .The poor health and socioeconomic status of NT Aboriginal children is of national concern, and led to the NT Emergency Response (NTER).The 2007-2008 phase of the NTER involved visiting teams offering health checks to all children aged up to 16 years 10 .Implementation of the Healthy School-Aged Kids Program was variable during the NTER but returned to the standard program in 2010.This report describes the prevalence of overweight and obesity among children in central Australia who underwent health checks in Healthy School-Aged Kids Program in 2010.

Table 1 :
Prevalence of overweight and obesity among children in remote central Australia, according to age group interval; SD, standard deviation: ≥ +1 SD BMI-for-age is threshold for childhood overweight, ≥ +2 SD BMI-for-age is threshold for childhood obesity.**p< 0.01, for 5-9 year-olds compared with 10-15 year-olds.No other differences statistically significant.

Table 2 :
Reports of prevalence of overweight and obesity among Australian Aboriginal children 12,13,15-17 Not available; NT, Northern Territory.SD, standard deviation.†Overweight based on >85th centile and ¶obesity based on >95th centile of the Australian Health and Fitness Survey (1985) [15].The sample was limited to children who attended the health checks, and included children whether or not they attended school.The sample of 996 included 883 in the age range 5-15 years, of 1764 children in remote central Australia.Aboriginal status was not recorded in the health checks which are targeted at the needs of the Aboriginal children.Australian Bureau of Statistics reports that 92.4% of 5-15 year-olds in remote central Australia are Aboriginal 19 .