The epidemiological burden of obesity in childhood: a worldwide epidemic requiring urgent action

Excess weight during childhood and adolescence remains one of the most important issues in global health, despite emerging as a concern several decades ago [1, 2]. Recent estimates suggest that 40 million children under the age of 5 years and more than 330 million children and adolescents aged 5–19 years were overweight or obese in 2016 [3]. Given the global emergency posed by excess weight in children, member states of the World Health Organization (WHO) endorsed “no increase in childhood overweight by 2025” as one of the six global nutrition targets in the ‘Comprehensive Implementation Plan for Maternal, Infant and Young Child Nutrition’ [4]. This is consistent with the same target for obesity and diabetes between 2010 and 2025 in the ‘WHO Global Action Plan for the Prevention and Control of Non-communicable Diseases 2013–2020’ [5, 6].

Overweight or obesity during childhood has important short-term and long-term consequences. In the short term, children who are overweight or obese are more likely to suffer from psychological comorbidities such as depression, anxiety, low self-esteem, a series of emotional and behavioural disorders [7, 8], asthma [9], low-grade systemic inflammation [10, 11], liver complications [12, 13], and musculoskeletal problems, especially in the lower extremities [14]. Children who are overweight and obese also have more metabolic and cardiovascular risk factors [15, 16], such as high blood pressure [17], dyslipidaemia [18], type 2 diabetes [19] and other abnormalities of the cardiovascular system [20]. In the long term, overweight or obesity during childhood increases the risk of developing cardiovascular diseases, diabetes, some cancers, and musculoskeletal disorders in adulthood, which can lead to disability [21] and premature death [22‐24]. In addition, the treatment of obesity in adulthood is difficult [25], with evidence suggesting that around three-quarters of children who are overweight or obese carry this status into adulthood [26]. Strong persistence of overweight status and low efficacy of available treatments highlight the need to prevent overweight and obesity at the earliest possible stage of life.

It is recognised that weight gain is partly caused by elevated energy intake, which often includes a disproportionate amount of refined carbohydrates and/or processed foods (increasing insulin release and fat storage), and decreased physical activity [27]. Weight gain is also promoted by environmental, behavioural, biological, and genetic factors, whose interactions have driven the current levels of worldwide obesity. Maternal health status during pregnancy, an obese intrauterine environment [28] and rapid changes in weight status during infancy [29] are other factors contributing to obesity in children. Moreover, the expanding ‘obesogenic’ environment increases the propensity of children to consume foods and beverages that are high in calories, energy-dense, or low in nutrients, as well as promoting sedentary lifestyles through reductions in opportunities for active mobility in daily lives [30]. Key drivers of the rapidly increasing worldwide occurrence of obesity and diabetes across populations are the globalised market and commercial interests that favour the production and distribution of inexpensive, energy-dense foods and beverages and limited political will to address the economic causes of the obesity epidemic [3], which include a strong association with socioeconomic inequalities [31, 32]. In high-income settings, higher prevalence of obesity is observed in disadvantaged and marginalised communities than in groups with higher socioeconomic status [33‐35]. In contrast, higher prevalence of obesity is seen in groups with higher socioeconomic status in some, but not all, low and middle-income settings [31].

Over the past decade, genome-wide association studies have been used to identify genetic markers that increase predisposition to weight gain, with the goal of explaining the biological mechanisms leading to obesity. For example, the FTO gene is recognised as being key to the regulation of energy intake, with variants predisposing individuals to greater caloric intake and reduced feelings of satiety [36]. Genetic and epigenetic factors also produce heterogeneity in obesity phenotypes across populations, including characteristic metabolic profiles and greater central body adiposity in south Asians [37]. However, groups with almost identical genotypes can have very different obesity phenotypes, as shown by the large differences in prevalence between Samoa and American Samoa [38]. In addition, obesity-associated genes cannot explain the rapid onset and scale of the current obesity epidemic, even if genetic predisposition makes some individuals more susceptible to the obesogenic environment [39].

Finally, obesity in childhood has important economic and social costs, with increased burdens on health systems as well as later reduced economic productivity [40‐43]. For example, in the USA, the estimated direct medical cost over the lifetime of a 10-year-old child with obesity, compared with a similar child with normal weight and allowing for weight gain in adulthood, is between US$12,660 and US$19,630 [44].

Over the past decade, there has been a global effort to provide reliable and detailed estimates of the worldwide epidemic of excess weight in children and adolescents. Here, we aim to provide a comprehensive description of this work, presenting global, regional and national trends based on the most up-to-date information available. To do so, we use data from the United Nations Children’s Fund (UNICEF)/WHO/World Bank Joint Child Malnutrition Estimates [45], the Institute for Health Metrics and Evaluation (IHME) Global Burden of Disease Study [46] and the Non-Communicable Disease (NCD) Risk Factor Collaboration (NCD-RisC) [47] (see Table 1). It should be noted that there are two definitions of obesity in childhood: the International Obesity Taskforce (IOTF) definition [48] and one based on the WHO growth reference curve [49]. They have different age-specific cut-offs and can therefore give different obesity estimates for a given set of data. In the following, estimates published by IHME use the IOTF definition, while estimates published by UNICEF/WHO/World Bank and NCD-RisC use the WHO growth reference. Readers interested in the differences in statistical models and regional definitions in these studies are referred to the original papers.

Over the past four decades, obesity in children of all ages has increased worldwide, as it has for adults [47]. However, obesity appears to have increased more rapidly in 5–19 year olds than in younger children, with an eight-fold increase between 1975 and 2016. This contrasts with an approximate doubling in obesity rates in children aged 2–4 years between 1980 and 2015, albeit using metrics that are not directly comparable. There is heterogeneity in the levels and trends in obesity prevalence between regions and countries, depending on the stage of the global obesity epidemic they are experiencing. In particular, there has been some flattening of trends, especially among those with high socioeconomic status in high-income countries [51].

The need for high-quality and comparable data is recognised as a key component for monitoring malnutrition [3]. Data from pooled analyses allow examination of change over time and the use of standardised, comparable metrics allows trends to be benchmarked across countries. Here, we have used data from three different sources, covering different ages and countries. This limits their comparability [45‐47] and, in particular, there is less standardised and comparable country-level information for children under the age of 5 years [45]. Equally, although national trends are of great interest, it is known that they mask subnational heterogeneity. Collection of disaggregated data at the subnational level and across specific groups of the population is therefore essential to identify groups that are at risk of malnutrition and to ensure progress in meeting global targets [3].

Despite overall increases in the prevalence of obesity in childhood, different forms of malnutrition coexist at global, national and subnational levels. Increases in obesity are linked to a reduction in the prevalence of children of normal weight, without there necessarily being decreases in the prevalence of children who are underweight. At a global level, the prevalence of underweight among children aged 5–19 years has remained unchanged over the past four decades [47]. Similar observations have been made in individual countries. For example, in Seychelles, only children in the upper percentiles of body mass index (BMI) have increased in weight, with little or no increase observed among those with median and low BMI [52]. More studies are needed to describe the shift in distribution of BMI over time in populations, e.g. estimates of the whole distribution to examine whether increases in BMI have occurred in all children or only in subgroups.

It is recognised that the main drivers of the current obesity epidemic are related to changing food systems and reduced physical activity [53‐55], with two key features. Firstly, there is increased availability of generally inexpensive, energy-dense and ultra-processed foods and beverages. The globalisation of food supply means that it is often economically more profitable to produce and market processed, energy-dense foods than fresh ones. Recent results from the Global Burden of Disease study show that consumption of healthy foods is suboptimal, whereas that of unhealthy options exceeds recommended levels [46]. Secondly, there have been increases in the number of people with sedentary lifestyles, with high levels of physical inactivity among children [56]. As children transition through childhood and adolescence, susceptibility to the food and physical environments increases. Increasingly, children can choose the foods they eat and how much exercise they do and this has a strong impact on current and future behaviour [57‐60]. This may, in part, explain the rapid increase in the prevalence of obesity in this group. Further investigation is required to explain the more rapid increase in obesity in boys, including studies of whether they are more susceptible to obesogenic pressures.

The need to improve the food environment requires governments, international organisations and other key stakeholders, including civil society and the private sector at local and global levels, to address global and local commercial determinants of obesity, including production and marketing of unhealthy, energy-dense foods and to improve availability and affordability of unprocessed healthy foods. Equally, healthy diets must be integrated with food systems in a sustainable way, such that long-term health benefits are possible [61]. A constructive dialogue with the food industry and effective regulations are needed to improve the availability of healthy foods and reduce that of unhealthy options, including prevention of unethical marketing of unhealthy foods aimed at low-income countries and other vulnerable members of the global population. The paradigm of energy imbalance (increased energy intake not balanced by energy consumption) is often used by the food industry to weaken policies aimed at tackling the use of energy-dense foods; i.e., it is argued that adequate levels of physical activity can compensate for this imbalance. Given the scale of the obesity epidemic, this argument must be viewed with scepticism. In the same vein, city and urban planners must rethink their role in society, given that current physical environments substantially restrict mobility patterns. As sedentarism becomes more common and future jobs require less activity, our children will amass a substantial cumulative burden of inactivity that will become difficult to reverse.

The effectiveness of interventions aimed at children who are overweight and obese has been widely studied [62‐65]. Most interventions have targeted behavioural changes, mainly in terms of nutrition and physical activity. There is evidence that some of these interventions have been effective in schools [66, 67]. In particular, promotion of physical activity in school-based settings may be beneficial given adequate resources [68], as well as being an important component of effective overweight prevention strategies in children [66]. Efforts to promote active mobility, such as cycling lanes, are being implemented in many cities in high-income countries – and, increasingly in low or middle-income countries, including the cities adhering to the Agita Mundo Network in Latin America [69]. Recently, declines in the level of obesity among pre-school-age children have been observed in New Zealand [70], Leeds (UK) [71, 72] and Amsterdam (the Netherlands) [73]. These declines are associated with interventions aimed at supporting families and communities by creating a healthier food environment and supporting families to enforce healthy habits in children with an approach of shared responsibilities among multiple actors. This may have important implications for future trends of obesity in childhood. However, caution in causal interpretation is necessary and more evidence is needed to establish that the implemented interventions are, in fact, responsible for the observed declines in obesity in childhood [70, 71]. Equally, data and evidence on the effectiveness of community-based approaches are limited compared to school-based programmes [74, 75]. Discussions about the role of community knowledge, attitudes and awareness towards obesity in accepting policy-level solutions continue, but data supporting such evidence are lacking or show unsuccessful examples [76, 77].

Despite this work, the effects of traditional behavioural change interventions will be too small to relieve the global burden of obesity in childhood – at least in the short to medium-term [53]. Community-wide approaches matched with changes in government policies related to food reformulation, advertising and affordability are therefore also required. Policies that can achieve such changes include, among others, excise tax on beverages containing sugar, subsidies or alleviation of trade tax for producing and distributing fruit and vegetables, regulations on food labelling, restrictions on advertising of unhealthy foods and beverages, and incentives or regulations to catalyse reformulation of processed foods toward healthier composition. Other policies include giving vouchers to mothers in the USA with low incomes to purchase fruits and vegetables, low-fat or skimmed milk and whole-grain instead of refined-grain products, among other changes. This has been shown to reduce obesity rates among children aged 2–4 years [78]. However, high levels of heterogeneity in policy are observed across countries, with low and middle-income countries relying more on such approaches (and implementing them earlier) than high-income countries. For example, Mexico was one of the first countries to implement a sugar-sweetened beverage tax; 2 years after implementation, consumption had decreased by 8.2% [79]. In 2014, Chile started to implement a series of policies aimed at reducing obesity. Tax on beverages with high sugar content was increased from 13 to 18%, while tax on drinks with low or no sugar content was reduced from 13 to 10%. In 2016, a labelling system using black octagons on packaging was introduced for food and beverages that are high in sugar, calories, sodium and saturated fats. In addition, foods and beverages with such labels have been banned from schools, while marketing of these products to children under the age of 14 years is no longer allowed [80, 81]. Initial results suggest a positive impact on knowledge and awareness, reductions in consumption of unhealthy foods and a positive response from the food industry. In turn, the food industry is decreasing the amount of sugar and sodium in some food categories.

The heterogeneity in levels of obesity across the world also has important implications for global targets and goals. It is necessary to aim only for “no increase in obesity by 2025” in those regions and countries in which a clear upward trend in obesity is observed. However, much stronger political action is needed in those regions and countries where the prevalence of obesity has plateaued at high levels, to raise the priority of multi-sectoral interventions to address obesity and other chronic conditions. In general, there is a need to examine how different policy agendas [5, 6, 82‐84] can be integrated and strengthened to promote healthy nutrition and regular physical activity, including preventing overweight among children, while also continuing to implement interventions against undernutrition. This will require additional efforts that should not overlook low and middle-income countries simply because some have moderate levels of obesity and high levels of undernutrition.

Tackling the obesity epidemic in children will require integrated efforts across multiple sectors to provide equitable access to economic resources, education, healthy food and urban environments and to universal health coverage. Most importantly, bolder political will and accountability is needed from actors including government, civil society, academia, the private sector and other key stakeholders, to spearhead efforts to promote production of and access to a healthier environment for all.