Prevalence of overweight and obesity in Irish school children, using four different definitions

Abstract

Obesity is a major public health problem in Ireland, with 39 and 18% of Irish adults reported to be overweight and obese, respectively (McCarthy et al., 2001). Since 1990, there has been no national data on the weight status of children, however, a recent study of inner city Dublin schoolchildren indicated that the levels of obesity among 10–12 years old has risen dramatically (Griffin et al., 2004). Ireland is not alone, the World Health Organization (WHO, 1998) have reported that the prevalence of both adult and childhood obesity has reached epidemic proportions worldwide. Obesity is defined as excess body fat and can be measured using various techniques such as dual-energy X-ray absorptiometry, underwater weighing and magnetic resonance imaging (WHO, 1998). However, these methods require expensive equipment and are therefore limited to a clinical research setting. Alternatively, body mass index (BMI) (weight/height2) is a valid and reproducible method that is cheap and convenient for the assessment of weight status (where weight becomes a surrogate for body fatness) in epidemiological studies. BMI is commonly used in adults and more recently recommended for use in children and adolescents (Power et al., 1997; Bellizzi and Dietz, 1999; Dietz and Bellizzi, 1999; Bini et al., 2000; Reilly et al., 2000; Widhalm et al., 2001). However, the assessment of weight status in young people is more complex, when using BMI to assess adiposity in children and adolescents it is essential that age- and sex-specific BMI cutoffs be used to define overweight and obesity (Widhalm et al., 2001; Field et al., 2003). Given the economic, environmental, genetic and nutrition influences on body size, having growth standards that are nationally representative is an important consideration (Livingstone, 2000). However, as was highlighted in the recent study of Dublin schoolchildren, there is no standard method for weight status assessment among children in Ireland (Griffin et al., 2004). Recently, the use of BMI standard deviation (s.d.) scores has also been recommended to monitor trends in childhood obesity (Rudolf et al., 2006). Childhood obesity is associated with a number of health problems, including type II diabetes, increased cardiovascular disease risk factors, respiratory problems, sleep apnoea, gall bladder disease, orthopaedic and psychosocial problems (Dietz, 1998; WHO, 1998; Must and Strauss, 1999; NTO, 2005). Furthermore, the persistence of childhood obesity into adulthood (Guo et al., 2000) and the prospect of the associated increases in mortality and morbidity in the long term have prioritized childhood obesity as a major public health crisis (WHO, 1998; Ebbeling et al., 2002; NTO, 2005). Data on weight status that is both timely and representative is a fundamental requirement of public health campaigns to prevent and treat childhood obesity. The National Children's Food Survey (NCFS) has recently collected the most detailed anthropometric, nutritional, physical activity and attitudinal data ever available on a nationally representative sample of Irish children. Such data is vital for formulating nutrition policy and for the development of effective health promotion strategies for Irish children. Therefore, the aim of this study was to use the weight and height data collected and determine the prevalence of overweight and obesity using four different weight-for-height methods to define the overweight and obese child, secondly to indicate secular trends from pre-existing data and finally to consider the issue of which of these methods may be the most appropriate for determining the current prevalence of overweight and obesity in Irish children. As the same assessment methods were used, archived weight (kg) and height (cm) data collected in children aged 8–12 years during the Irish National Nutrition Survey (INNS), a cross-sectional and nationally representative survey conducted in 1990, were analysed and compared with the current data (Lee and Cunningham, 1990). In the INNS, 14 primary schools were selected with probability proportionate to size and were stratified by county (Lee and Cunningham, 1990). Comparisons were also made with the 1948 National Nutrition Survey, using mean height and weight data that was recorded on 14 835 primary school children and then classified by area of residence, father's occupation, age and sex (DOH, 1952). In order to establish crude trends in mean weight (kg), height (m) and BMI (kg/m2) over the last 57 years, data for 8–12 year olds from the 1948 NNS, 1990 INNS and 2005 NCFS were plotted as a percentage increase since 1948. Age-and-sex-specific BMI charts were also used to establish the prevalence of overweight and obesity in this sample. Age-and-sex-specific BMI charts compare a child's BMI to the BMI distribution of a reference sample of children of the same age (Flegal et al., 2002). Centile cutoffs are then used to define the weight status of the child. Owing to the absence of age-and-sex-specific BMI charts for an Irish reference population, the Centres for Disease Control and Prevention (CDC) BMI charts, developed to measure the nutritional status of US boys and girls aged 2–20 years, were used (Kuczmarski et al., 2000). The cutoffs for use with these charts to define overweight status included describing all children with a BMI between the 85th and 95th percentile as 'at risk of overweight' and those with a BMI on or above the 95th percentile as 'overweight' (Kuczmarski et al., 2000). The second age-and sex-specific BMI charts used in this study were the UK 1990 BMI reference curves for boys and girls (UK90). The criteria used in these charts to define 'overweight' were having a BMI between the 91st and 98th percentile, whereas the criteria for 'obesity' was a BMI on or above the 98th percentile (Cole et al., 1995). Finally, the recently published International Obesity Task Force (IOTF) age-and-sex-specific BMI cutoffs for...