4. The relationship between childhood obesity and educational outcomes

Marion Devaux
Sabine Vuik

This chapter investigates the relationship between childhood obesity and a variety of educational outcomes. A data analysis covering 32 countries explores the association between performance at school and obesity in children aged 11 to 15, and assesses the degree of inequality across countries. A longitudinal analysis investigates a potential causal relationship between childhood obesity and academic performance as well as education attainment in five countries. Finally, the chapter discusses the broader consequences of the relationship between obesity and educational outcomes, for individuals and the economy.

    

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

Key findings
  • Obesity and educational outcomes are interrelated, and mediated by biological (e.g. diseases), behavioural (e.g. lack of physical activity), emotional and mental health factors (e.g. low self-esteem, poor social connection).

  • An analysis of Health-Behaviours in School-based Children (HBSC) data shows that children with obesity have lower life satisfaction, and are more prone to being bullied by schoolmates. This can lead to lower class participation and reduced educational performance.

  • The relationship between obesity and bullying is more pronounced in girls than in boys. In OECD countries, girls with obesity are 3 times more likely to be bullied than are healthy-weight girls, while this ratio is 1.8 times in boys.

  • The association between obesity and poor academic performance is well established in the 32 studied countries, after controlling for confounders. Healthy-weight children are 13% more likely to report good performance at school than children with obesity.

  • Analysis of longitudinal data from national surveys suggests that this relationship may be causal, as the presence of obesity at a young age appears to affect academic performance and education attainment later on in life.

  • Childhood obesity has long-lasting effects, including health-threatening consequences over the life course. In addition, the relationship between childhood obesity and educational outcomes can constrain the formation of human capital and future socio-economic status. As such, it affects the individual as well as society and the economy.

  • Policy makers should invest in a wide range of policy interventions aimed to tackle childhood obesity, reduce the obesity stigma, reduce bullying, and improve the wellbeing and mental health of overweight children.

4.1. The relationship between childhood obesity and educational outcomes is mediated by different factors

4.1.1. The evidence points to a significant relationship between childhood obesity and educational outcomes

Obesity1 rates among children have substantially increased over the past decades, similar to the rising obesity rates in adults (see Chapter 2). Childhood obesity rates will continue to rise if nothing is done to tackle the obesogenic factors (OECD, 2010[1]; OECD, 2017[2]). The consequences of overweight in childhood are important and widespread: they affect individuals in their early life and may have an impact over their entire life course; including deteriorating health and well-being, emotional and mental health problems, bullying, and poor educational outcomes. This chapter seeks to shed light on the relationship between childhood obesity and educational outcomes.

Overall, a great number of studies published in the literature converge to show a significant association between childhood obesity and poor academic performance. These studies provide evidence for a variety of educational outcome measures (such as school grades, absences, engagement, repeating a grade), and for different age groups. For instance, high body weight is associated with lower academic performance, as measured by grades, in United States university students (Anderson and Good, 2017[3]). Obesity in the Unites States is also related to more school problems and less school engagement in adolescents (Carey et al., 2015[4]), and to more absence from school in children aged 6 to 11 (Li et al., 2012[5]), as well as in adolescents aged 12 to 17 (Pan et al., 2013[6]; Carey et al., 2015[4]).

However, some of these studies reach different conclusions once they controlled for confounders2 and mediating factors. For instance, while obesity in US schoolchildren appeared to be related to the likelihood of repeating a grade, the relationship disappeared when controlling for confounders (Carey et al., 2015[4]). Similarly, in Spain, the relationship between obesity and test scores in children aged 9 to 11 vanished once controlled for confounders (Torrijos-Niño et al., 2014[7]). This exposes the complexity of the relationship between obesity and educational outcomes, which is influenced by and related to many other factors.

4.1.2. The relationship between obesity and educational outcomes is mediated by biological, behavioural and emotional factors

There are different pathways through which obesity is linked to educational outcomes. These pathways involve various mediating factors as described in Figure 4.1: biological factors (e.g. diseases), behavioural factors (poor nutrition and lack of physical activity), and emotional and mental health factors (e.g. low self-esteem, poor social connection).

Figure 4.1. Relationship between obesity and educational performance
Figure 4.1. Relationship between obesity and educational performance

Source: OECD analysis.

Biological factors may adversely affect both body weight and cognitive functions

Obesity and its related diseases, such as metabolic syndrome, may have a direct effect on cognitive functions and concentration at school. For instance, metabolic syndrome was found to have an impact on cognitive functions and brain structure through physiological impairments (Yates et al., 2012[8]). Another study found a direct link between childhood obesity and lower cognitive performance, independently of physical activity, sleep, and diet (Hjorth et al., 2016[9]).

Behaviour may be associated with childhood obesity and lower concentration

There exists an interrelationship between unhealthy behaviours, obesity and educational outcomes. Behavioural risk factors, such as poor nutrition and lack of physical activity, may directly play a role on both obesity and low concentration at school, resulting in poor performance at school. For example, insufficient levels of physical activity can be both a cause and a consequence of obesity, and can lead to lower concentration (Bustillo et al., 2016[10]).

A clear relationship between physical activity and cognition has been documented. A recent systematic review of 64 studies found that physical activity has a positive influence on cognitive functions as well as brain structure and function (Donnelly et al., 2016[11]). Similarly, a meta-analysis of 44 studies showed that physical activity has a positive association with cognition in children (Sibley and Etnier, 2003[12]).

Regarding the effect of physical activity on school achievement, a systematic review of 14 longitudinal studies found evidence for a significant longitudinal positive relationship between physical activity and academic performance, although the dose-response relationship needs further investigation (Singh et al., 2012[13]). In a Cochrane review, physical activity interventions produced small yet significant improvements in mathematics achievements (mean difference (MD) of 3 points on a scale of 0 to 100, P-value=0.008), executive function (MD 3, P-value=0.04), and working memory (MD 3, P-value=0.02) (Martin et al., 2014[14]). However, there was no evidence to suggest an effect on reading, vocabulary and language achievements, attention, inhibitory control and simultaneous processing. Hence, while the association between physical activity, cognitive function and school achievement is evident, the best way to incorporate physical activity within schools to improve academic achievement (e.g. activity breaks versus active lessons) is less clear (Donnelly et al., 2016[11]).

Emotional and mental health problems, related to bullying and stigma, can lead to poor academic performance

Emotional and mental health problems can also mediate the relationship between obesity and education performance. Overweight children and adolescents may often be excluded from friendships and bullied by other children. Because of this, overweight children may feel isolated, lonely or socially disconnected, they may have lower self-esteem, poor well-being and suffer from emotional and mental health problems (Russell-Mayhew et al., 2012[15]; Strauss, 2000[16]).

These problems may have deleterious effects on educational outcomes. Children who are bullied and socially excluded by others engage less in class: they step aside and refuse to speak up for fear of being bullied (Ladd, Ettekal and Kochenderfer-Ladd, 2017[17]). Moreover, behavioural problems in schools, such as disobedience and violence, may emerge.

OECD analysis of the Health-Behaviours in School-based Children (HBSC) survey 2013-14 points to a significant association between obesity and emotional and mental health problems (see data and methods in Box 4.1). In the HBSC survey, children aged 11-15 were invited to rate their life satisfaction from 0 (lowest) to 10 (highest). The average life satisfaction was 7.4 in boys with obesity compared to 7.8 in boys with a healthy-weight, and 6.8 in girls with obesity compared to 7.4 in girls with a healthy-weight, all other things being equal (Figure 4.2).

Figure 4.2. Life satisfaction by BMI category, children aged 11-15, OECD countries, 2013-14
Predicted score of life satisfaction, with 95% confidence interval
Figure 4.2. Life satisfaction by BMI category, children aged 11-15, OECD countries, 2013-14

Note: Analysis adjusted for age, family affluence, smoking and drunkenness in lifetime. OECD countries only.

Source: OECD analysis based on HBSC 2013-14.

 StatLink https://doi.org/10.1787/888934007354

Bullying and obesity are also strongly associated. OECD analysis shows that 16% of the 11-15 year-old boys who had obesity were bullied by schoolmates, compared to 9% in healthy-weight boys. This gap is even larger in girls (19% versus 8%) (Figure 4.3). This relationship was adjusted for family affluence and life satisfaction, meaning that a high BMI has a direct relationship with bullying, not mediated by socio-economic status or emotional health. A country-specific analysis confirms the obesity-related differences in the probability of being bullied in each country studied (Annex Figure 4.A.1).

Italy, Czech Republic and Germany have the largest obesity-related inequalities in bullying among girls, while Germany, Malta, Austria and Norway show the largest inequalities among boys (Figure 4.4). The relationship between obesity and bullying is much more pronounced in girls than in boys in all countries. On average in OECD countries, girls with obesity are 3.1 times more likely to be bullied than their healthy weight counterparts, compared to 1.8 times in boys. This difference between sexes is especially marked in Iceland, the Czech Republic, Israel and the Russian Federation, where the effect of obesity on bullying is almost two times larger in girls than in boys.

Differences in bullying with regard to obesity have remained stable over the last decade in girls, while they have increased in boys, on average in OECD countries (Annex Figure 4.A.2). A time-trend analysis showed that the probability of being bullied in boys has significantly decreased in boys with a healthy-weight and pre-obesity compared to boys with obesity, widening inequalities. By contrast, bullying has not significantly changed in girls, on average in OECD countries.

Emotional and mental health consequences in children and adolescents are comparable to the well-known effects of the obesity stigma in adults. Weight bias and obesity stigma are associated with poor body image, low self-esteem, loneliness, suicidal thoughts and acts, depression, and anxiety (WHO Europe, 2017[18]). Obesity stigma leads to exclusion and marginalisation of persons with obesity, and to inequities. For example, people with obesity may not receive adequate health care or may be discriminated against in the workplace or in educational settings (WHO Europe, 2017[18]).

Figure 4.3. Probability of being bullied by BMI category, children aged 11-15, OECD countries, 2013-14
Predicted probability of being bullied, with 95% confidence interval
Figure 4.3. Probability of being bullied by BMI category, children aged 11-15, OECD countries, 2013-14

Note: Analysis adjusted for age, family affluence, smoking and drunkenness in lifetime. OECD countries only.

Source: OECD analysis based on HBSC 2013-14.

 StatLink https://doi.org/10.1787/888934007373

Figure 4.4. Relative index of inequality for being bullied by BMI category, children aged 11-15, 2013-14, by sex and by country
Figure 4.4. Relative index of inequality for being bullied by BMI category, children aged 11-15, 2013-14, by sex and by country

Note: The RII is calculated as the ratio between the probability of children with obesity being bullied divided by the probability for healthy-weight children, taking into account the average BMI in each group. An RII greater than one means that the likelihood of being bullied increases with BMI. The higher the RII, the larger the inequality across BMI categories. For example, in Italy, the most obese girls are 3.8 times more likely to be bullied than healthy-weight girls. Data on bullying for Switzerland is missing. The OECD average is derived from a pooled countries analysis weighted by the national sample size.

Source: OECD analysis based on HBSC 2013-14.

 StatLink https://doi.org/10.1787/888934007392

Discrimination against people with obesity has also been observed for young people. A study in the United States provides evidence that interviewers for graduate programmes favoured thinner applicants, which could be due to a (conscious or unconscious) bias against applicants with obesity. However, it is also possible that stereotype threat or social identity threat led the applicant to under-perform (Burmeister et al., 2013[19]).

4.2. There is a clear association between childhood obesity and school performance in OECD countries

An OECD analysis based on HBSC data (Box 4.1) shows that children with obesity have significantly lower performance at school than their healthy weight counterparts, and teenagers with obesity are more often absent from school. These relationships remain significant after controlling for mediating and confounding factors such as family affluence, life satisfaction, and bullying.

Box 4.1. Data and methods for the analysis on obesity and educational outcomes

Analyses are based on the 2013-14 Health-Behaviours in School-based Children survey (HBSC) which collects information on school students aged 11, 13 and 15 years old every four years. School and health information are self-reported by children (e.g. school performance compared to classmates, bullying, life satisfaction, height and weight, smoking and alcohol drunkenness). Body mass index (BMI), calculated from height and weight, was categorised into healthy-weight, pre-obese and obese, using the WHO age- and sex-specific BMI cut-off points for children (de Onis et al., 2007[20]). Analyses presented cover 32 HBSC countries: 26 European Union countries plus Canada, Iceland, Israel, Norway, Switzerland, and the Russian Federation.

A pooled-country multivariate logistic regression analysis was performed to assess the probability of being bullied and a linear model was used for the assessment of life satisfaction. Country-specific analyses of the probability of being bullied and having good performance at school were performed using a mixed logit model with random effects on the intercepts and the BMI coefficient at the country level, while controlling for individual characteristics. Predicted probabilities of being bullied and of self-perceived good performance at school were estimated for each BMI category. The relative index of inequality, which is a summary measure of inequality, was used to gauge obesity-related inequalities in bullying and school performance across countries.

Trends analysis

Four waves of the HBSC survey (2001-02, 2005-06, 2009-10, 2013-14) were combined for the trends analysis. A pooled-country logistic model of the probability of good performance at school was used to assess the effect of BMI categories, survey year (continuous), and their interaction term, while controlling for age, smoking status, drinking status and socio-economic background. Separate models were run for boys and girls. The overall statistical difference in the trends coefficient for the three BMI categories was tested. A similar approach was used for the probability of being bullied.

School absence analysis

The United States’ NHANES data was used to study school absence. Four waves of NHANES, from 2001-02 to 2007-08, collect information on missed days of school. The analysis focused separately on children aged 6-11 and teenagers aged 12-19. BMI categories were defined using the WHO age- and sex-specific BMI cut-off points for children (de Onis et al., 2007[20]).

Multivariate logistic regression analysis was performed to assess the probability of missing school days according to BMI category, while adjusting for age, sex, ethnicity and socio-economic status. A negative binomial model was used for the number of days of absence.

Figure 4.5. Probabilities of good performance at school by BMI level, children aged 11-15, 2013-14, by sex and by country
Predicted probability of good performance at school, with 95% confidence intervals
Figure 4.5. Probabilities of good performance at school by BMI level, children aged 11-15, 2013-14, by sex and by country

Note: Mixed model with random slope. Adjusted for age, family affluence, smoking and drunkenness in lifetime. Covariates are set at fixed values (Age 13, Middle family affluence, Never drunk, Never smoke).

Source: OECD analysis based on HBSC 2013-14.

 StatLink https://doi.org/10.1787/888934007411

4.2.1. Children with obesity have lower performance at school than children with a healthy-weight

The analysis of HBSC data shows that high BMI levels in 11-15 year-olds are associated with poorer self-perceived school performance in the 32 countries studied (Figure 4.5). The strength of the relationship varied across countries. For instance in Italy, 57% of healthy-weight boys are likely to rate their performance above the class average compared to 51% of boys with obesity. The gradient is steeper in France, where 58% of healthy-weight boys perform above average compared to 49% of boys with obesity. Girls have better school performance than boys in all countries studied but Portugal.

The gradient of inequalities – as measured by the relative index of inequality – confirms that the higher the BMI in childhood, the lower the performance at school, in all countries (Figure 4.6). On average across all 32 countries in the analysis, boys and girls with a healthy weight are 13% more likely to report good school performance, compared to their peers with obesity. When the OECD countries are pooled, boys with a healthy weight are 12% more likely to report good school performance and girls 11%.

Obesity-related inequalities in school performance are greater for boys in 18 countries and for girls in 14 countries. France and Belgium have the largest inequalities among girls. In both cases, girls with a healthy weight are about 27% more likely to report good school performance compared to girls with obesity. Germany and Latvia display the largest inequalities among boys, with an increased probability for boys with a healthy weight to report good school performances of 24% and 23%, respectively.

Figure 4.6. Relative index of inequality for good performance at school by BMI category, children aged 11-15, 2013-14, by sex and by country
Figure 4.6. Relative index of inequality for good performance at school by BMI category, children aged 11-15, 2013-14, by sex and by country

Note: The RII is calculated as the ratio between the probability of good school performance for children with obesity divided by the probability for healthy-weight children, taking into account the average BMI of each group. An RII greater than one means that the likelihood of good performance at school decreases with BMI. The greater the RII, the larger the inequality across BMI categories. For example, in France, healthy-weight girls have 27% more chance of performing well at school than the most obese girls. The OECD average is derived from a pooled countries analysis weighted by the national sample size.

Source: OECD analysis based on HBSC 2013-14.

 StatLink https://doi.org/10.1787/888934007430

Differences in school performance have slightly increased in boys and slightly narrowed in girls over the last decade. A time-trend analysis showed that from 2002 to 2014 there was a significant increase in school performance in healthy-weight boys over time, while the trends have remained stable in those who were pre-obese and obese, on average across OECD countries. In girls, there was a significant decrease in school performance in healthy-weight girls over time, while the trends have remained stable in those who were pre-obese and obese, narrowing inequalities (Annex Figure 4.A.3).

4.2.2. Teenagers with obesity are more likely to miss school

Analysis based on the NHANES data shows that teenagers with obesity, aged 12-19, are more likely to miss school (Figure 4.7). In the previous 12 months, 69% of adolescents with obesity had missed school days compared to 66% of healthy-weight adolescents. The difference is not significant in children aged 6-11. The pattern is similar for boys and girls, and results are therefore presented together.

Figure 4.7. School absence and BMI categories, children aged 12-19, United States, 2001-08
Predicted probability of absence, with 95% confidence interval
Figure 4.7. School absence and BMI categories, children aged 12-19, United States, 2001-08

Note: Analysis adjusted for age, sex, ethnicity and socio-economic status.

Source: OECD analysis based on US NHANES 2001-2008.

 StatLink https://doi.org/10.1787/888934007449

When absent from school, adolescents with obesity had significantly more days of absence than healthy-weight adolescents aged 12 and above (Figure 4.8). Among those aged 12-14, adolescents with obesity reported 5.9 days of absence per year compared to 5.0 days in healthy-weight adolescents. Adolescents aged 15-19 with obesity or a healthy weight reported 7.6 days versus 5.4 days of absence, respectively.

Figure 4.8. Average number of days of absence, by age group and BMI category, United States, 2001-08
Number of days of absence per year
Figure 4.8. Average number of days of absence, by age group and BMI category, United States, 2001-08

Note: Analysis adjusted for age, sex, ethnicity and socio-economic status.

Source: OECD analysis based on US NHANES 2001-2008.

 StatLink https://doi.org/10.1787/888934007468

4.3. A potential causal effect of childhood obesity on educational outcomes is found in five countries

From a policy perspective, it is important to understand the nature of the association between obesity and educational outcomes. If the relationship is causal, public health policy could help improve not only child health, but also educational outcomes such as bullying at school, absence, academic performance, and education attainment.

4.3.1. The literature presents mixed evidence for the causal effect of obesity on education

A number of studies have looked at the causal effect of obesity on educational outcomes, but with mixed results emerging. While some studies found a temporal association between obesity and educational outcomes, studies looking at younger children found no relation between obesity and educational outcomes, and yet others found different results for boys and girls.

A retrospective cohort study found a causal relationship between overweight and future academic performance among female high school students in Saudi Arabia. Girls who were overweight in 10th grade had 3.73 higher odds of experiencing a decline in grades between the 10th and 12th grade over healthy-weight students (Adaili, Mohamed and Alkhashan, 2017[21]). Similar evidence from Sweden suggests that male obesity may lead to lower educational attainment. Swedish men who had obesity at age 18 had 70% less chance of completing at least 15 years of education than healthy-weight subjects, even when adjusting for intelligence and socio-economic factors (Karnehed et al., 2006[22]).

Among younger children, evidence for a causal relationship is lacking. At least two studies found no significant relationship between child obesity and educational performance in primary school. A Dutch birth cohort study found no statistical association between the number of years that children were overweight and final year test scores or teacher assessment (Ruijsbroek et al., 2015[23]). Likewise, a Taiwanese study following elementary school children over six years found that BMI changes were not significantly associated with changes in academic performance (Chen et al., 2012[24]). Regarding education attainment, an analysis of the 1970 British Cohort Study found no significant relationship between obesity at 10 years old and educational attainment at 30 years old, after adjusting for confounders such as social class, parental BMI and education (Viner and Cole, 2005[25]).

Two studies found evidence for a causal association in girls, but not in boys. A UK study based on the ALSPAC cohort showed an impact of overweight on future test scores in girls. Obesity at age 11 decreased the English marks of girls at age 13 and 16, compared to healthy-weight girls, even after controlling for confounders. However, the effect was not significant for boys (Booth et al., 2014[26]). Similarly, a US study highlighted that becoming overweight between kindergarten entry and third grade was associated with reductions in test scores and teacher ratings of social-behavioural outcomes and approaches to learning for girls, but no such effects were found for boys (Datar and Sturm, 2006[27]).

4.3.2. An OECD analysis points to a negative causal relationship between obesity and educational outcomes in five countries

An OECD analysis based on longitudinal data available in five countries suggests that the relationship between obesity and educational outcomes may be causal. The data and methods are described in Box 4.2.

Box 4.2. Description of the longitudinal analyses

The objective of this analysis was to identify a potential causal relationship between obesity and educational outcomes. To investigate whether the relationship is causal, longitudinal datasets were used. By measuring obesity in one wave, and the educational outcomes in a later wave, temporal precedence can be established – one of the requirements for causality (Oppewal, 2010[28]). Lagged regression models were used to test this relationship.

The results in this section are based on data from longitudinal cohort studies in the United Kingdom (the 1970 British Cohort Study), the United States (the National Longitudinal Study of Adolescent to Adult Health, or Add Health), the Russian Federation (Russia Longitudinal Monitoring Survey, or RLMS), Germany (The German Health Interview and Examination Survey for Children and Adolescents, or KiGGS) and the Netherlands (The Prevention and Incidence of Asthma and Mite Allergy, or PIAMA). These longitudinal cohorts were selected as they included school-aged children and collected data on obesity and educational performance or attainment.

Educational outcomes were measured as educational performance and educational attainment. Educational performance is the performance of a student during his or her time in school. This included, for example, grades obtained in school subjects, teacher’s assessment of performance relative to other students, or tests scores. Educational attainment is the level of education ultimately achieved. This was measured as the number of years spent in full-time education, or whether the student completed any degree-level higher education.

An effort was made to standardise the analyses across the different country datasets. However, due to differences in the collected and reported data, different variables and concepts were used per country. Obesity was deduced from body mass index (BMI) using the child cut-off values defined by WHO (de Onis et al., 2007[20]). To correct for confounders, the models were adjusted for age, ethnicity or minority status, social class and/or income, and alcohol consumption - depending on the availability of data.

For more details on the datasets and the methods, please refer to OECD Health Working Paper No. 109 (Vuik, Devaux and Cecchini, 2019[29]).

Childhood obesity is associated with lower educational performance later on

In the United States, Russia and the Netherlands, a lagged relationship between obesity and educational performance was found (see Table 4.1). In the US study, there was a statistically significant relationship between BMI or obesity and grade point average (GPA) a year later, even when correcting for confounders such as age, sex, family income and ethnicity (note that all results presented here are adjusted for confounders). The GPA of girls with obesity was 0.26 points lower than that of girls with healthy weight. This equates to a student with a median GPA of 2.75 dropping to the 45th percentile. For boys, obesity was associated with a 0.11 point lower GPA a year later – a smaller but still significant effect. Similarly, a point increase in BMI was associated with 0.009 points (95% CI: 0.0001 to 0.0178) decrease in GPA for boys; and 0.021 points (95% CI: 0.015 to 0.028) for girls.

Table 4.1. Lagged relationship between obesity and educational performance

Country

Outcome

Method

Exposure

Boys

Girls

United States

GPA (1 to 4)

Lagged linear regression

Obesity (vs healthy weight)

Coefficient: -0.11*

Coefficient: -0.26***

 

 

 

Pre-obesity (vs healthy weight)

Coefficient: -0.00

Coefficient: -0.15***

 

 

 

BMI

Coefficient: -0.009**

Coefficient: -0.021***

 

 

 

BMI (quadratic)

Coefficient: -0.0002**

Coefficient: -0.0004***

Russian Federation

Average grade (1 to 5)

Lagged linear regression

Obesity (vs healthy weight)

Coefficient: 0.03

Coefficient: -0.11***

 

 

 

Pre-obesity (vs healthy weight)

Coefficient: 0.03

Coefficient: -0.02

 

 

 

BMI

Coefficient: 0.007**

Coefficient: -0.007***

 

 

 

BMI (quadratic)

Coefficient: 0.0001*

Coefficient: -0.0001**

Germany

Average grade (6 to 1; NOTE: lower grade is better performance)

Lagged linear regression

Obesity (vs healthy weight)

Coefficient: 0.08

Coefficient: 0.05

 

 

 

Pre-obesity (vs healthy weight)

Coefficient: 0.06

Coefficient: -0.02

 

 

 

BMI

Coefficient: 0.005

Coefficient: -0.010

 

 

 

BMI (quadratic)

Coefficient: 0.0162

Coefficient: -0.0002

Netherlands

High level of high school

Lagged logistic regression

Overweight (vs healthy weight)

Risk ratio: 0.80***

Risk ratio: 0.89

 

 

 

BMI

Odds ratio: 0.91***

Odds ratio: 0.94*

 

 

 

BMI (quadratic)

Odds ratio: 0.997***

Odds ratio: 0.998*

 

High school level below teacher assessment level

Lagged logistic regression

Overweight (vs healthy weight)

Risk ratio: 0.79

Risk ratio: 0.82

 

 

 

BMI

Odds ratio: 0.98

Odds ratio: 1.003

 

 

 

BMI (quadratic)

Odds ratio: 0.999

Odds ratio: 1.00002

Note: *: significant at 0.1 level; **: significant at 0.05 level; ***: significant at 0.01 level

Results shown are adjusted for age, ethnicity or minority status, social class and/or income, and alcohol consumption (United States only). The coefficients of linear regression models can be interpreted as the increase in outcome for each unit increase in exposure; relative risks show how much more or less likely one group is to experience the outcome, with a value greater than one signifying a higher likelihood; odds ratios are similar to risk ratios but are based on odds rather than risk.

Source: OECD analysis of the US National Longitudinal Study of Adolescent to Adult Health, the Russia Longitudinal Monitoring Survey, the German Health Interview and Examination Survey for Children and Adolescents, and the Dutch Prevention and Incidence of Asthma and Mite Allergy data.

 StatLink https://doi.org/10.1787/888934007544

In the Russian data sample, BMI was also associated with lower grades for girls, by 0.007 (0.002-0.012) points per point BMI. There also was a significant effect of BMI on grades for boys, but the effect was in the opposite direction: a one-point increase in BMI was linked to a 0.007 (0.001-0.012) point increase in average grade a year later.

In the Netherlands, performance was measured based on the different levels of high school that exist. The analysis showed that boys who were overweight at age 11 were 20% less likely to attend a higher level of high school at the age of 17 (RR: 0.80, 0.65 to 0.95), while the effect for girls was not significant. However, when looking at performance relative to the teacher’s assessment at age 11, there was no significant effect of overweight or BMI in either sex.

In Germany, no significant relation was found between obesity and educational performance six years later.

Childhood obesity during school years is associated with lower educational attainment

OECD analysis based on longitudinal data available in three countries points to a negative impact of obesity on educational attainment in the United States and the United Kingdom, while no significant relationship was found in Russia (see Table 4.2).

Table 4.2. Results of obesity and educational attainment analyses

Country

Outcome

Method

Exposure

Boys

Girls

United States

Any higher education

Lagged log-binomial regression

Obesity (vs healthy weight)

Risk ratio: 0.88

Risk ratio: 0.72***

 

 

 

Pre-obesity (vs healthy weight)

Risk ratio: 0.97

Risk ratio: 0.79***

 

 

 

BMI

Risk ratio: 1.00

Risk ratio: 0.98***

 

 

 

BMI (quadratic)

Risk ratio: 0.9999

Risk ratio: 0.9995***

United Kingdom

Any higher education

Lagged log-binomial regression (or logistic regression for odds ratio)

Obesity (vs healthy weight)

Risk ratio: 0.42**

Risk ratio: 1.02

 

 

 

Pre-obesity (vs healthy weight)

Risk ratio: 0.83

Risk ratio: 0.86

 

 

 

Overweight (vs healthy weight)

Risk ratio: 0.76**

Risk ratio: 0.88

 

 

 

BMI

Odds ratio: 0.95**

Risk ratio: 0.98*

 

 

 

BMI (quadratic)

Odds ratio: 0.9987**

Risk ratio: 0.9996**

Age left full-time education

Lagged linear regression

Obesity (vs healthy weight)

Coefficient: -0.91*

Coefficient: -0.35

 

 

 

Pre-obesity (vs healthy weight)

Coefficient: -0.19

Coefficient: -0.35*

 

 

 

Overweight (vs healthy weight)

Coefficient: -0.33

Coefficient: -0.35*

 

 

 

BMI

Coefficient: -0.045*

Coefficient: -0.044**

 

 

 

BMI (quadratic)

Coefficient: -0.0008*

Coefficient: -0.0010**

Russian Federation

Any higher education

Lagged log-binomial regression

Overweight (vs healthy weight)

Risk ratio: 0.73

Risk ratio: 0.94

BMI

Risk ratio: 1.04

Risk ratio: 0.98

BMI (quadratic)

Risk ratio: 1.0009

Risk ratio: 0.9995

Note: *: significant at 0.1 level; **: significant at 0.05 level; ***: significant at 0.01 level

Results shown are adjusted for age, ethnicity or minority status, social class and/or income, smoking status (except for the United States), and alcohol consumption. The coefficients of linear regression models can be interpreted as the increase in outcome for each unit increase in exposure; relative risks show how much more or less likely one group is to experience the outcome, with a value greater than one signifying a higher likelihood; odds ratios are similar to risk ratios but are based on odds rather than risk.

Source: OECD analysis of the US National Longitudinal Study of Adolescent to Adult Health, the 1970 British Cohort Study and the Russia Longitudinal Monitoring Survey data.

 StatLink https://doi.org/10.1787/888934007563

While neither obesity nor BMI in Wave 1 predicted whether a boy would complete higher education in the United States, both variables were significant predictors for girls. A girl who had obesity in Wave 1 was 38% less likely to complete higher education 14 years later than someone of healthy weight (RR: 0.72, 0.59 to 0.88).

Conversely, in the United Kingdom, the relationship between obesity and educational attainment was significant only for boys, who were 58% less likely to have completed higher education at the age of 29 if they had obesity at age 16 (RR: 0.42, 0.16 to 0.95). However, a higher BMI did have a significant effect in girls: for each point increase in BMI at age 16, girls spent 0.044 (0.004 to 0.085) years less in higher education – approximately half a month.

In Russia, no significant relationship was found between obesity and educational attainment using lagged regression models.

Comparing effect sizes across countries is complicated by differences in survey data. The diverging results that were found across countries may be the result of the national context; but they may also be caused by differences in the data. All cohorts collected data at different ages and at different intervals. While efforts were made to standardise the variables used for analysis, the data collected in each cohort was not always fully comparable. In particular, the type of variables available on socio-economic status and ethnicity/minority status varied across datasets, and may have caused differences in the results. Moreover, the known difference between self-reported and measured BMI may have also contributed to differences between the countries (Devaux et al., 2011[30]).

4.4. The short-term and long-term consequences of childhood obesity are of concern for individuals and societies

The consequences of overweight in children are important because they affect the entire life span. Childhood obesity may have short-term direct effects on physical and mental health, and educational outcomes such as bullying, school absence, low concentration, and poor grades, as discussed in the previous sections. In addition to this, long-term consequences may emerge, affecting both health and economic outcomes. Figure 4.9 depicts the potential consequences of childhood obesity on adult economic outcomes. In particular, it highlights the double burden: (a) through obesity in adulthood, which in turn affects labour market outcomes, and (b) through potentially lower educational outcomes, which in turn affect adult socio-economic status.

Obesity in childhood has long-lasting health consequences. Children with obesity are more prone to be obese as adults, and they have a higher risk of developing non-communicable diseases such as diabetes, cardiovascular diseases and musculoskeletal disorders (Kelsey et al., 2014[31]). As the age of onset and the duration of obesity matter for NCDs, young children with obesity are more at risk of premature death and disability in adulthood (Abdullah et al., 2011[32]).

In addition to the health implications, obesity also carries adverse labour market outcomes. Childhood obesity is a leading determinant for adult obesity and related diseases, which in turn, have consequences on employment and work productivity. People who suffer from obesity and its related chronic diseases have a lower chance of being employed, are more absent from work, and retire earlier (Devaux and Sassi, 2015[33]). Further evidence is provided in Chapter 3.

As shown in the analyses contained in this chapter, overweight in childhood is clearly linked to educational outcomes. In addition to disadvantaging the individual, this effect also multiplies the negative impact of obesity on society and the economy at large.

Figure 4.9. The double burden of childhood obesity on poor economic outcomes
Figure 4.9. The double burden of childhood obesity on poor economic outcomes

Note: Grey bubbles refer to childhood and adolescence, while blue ones refer to adulthood.

Source: OED analysis.

First, school performance and educational outcomes are key determinants for the formation of human capital. As such, reduced school performance may affect a country’s economic growth. An OECD report using Programme for International Student Assessment (PISA) data shows that improving the cognitive skills of the population can lead to significant economic gains and that relatively small improvements to labour force skills can largely impact the future well-being of a nation (OECD, 2010[34]). A modest goal of all OECD countries boosting their average PISA scores by 5% (25 points) over the next 20 years would increase OECD gross domestic product (GDP) by USD 115 trillion over the lifetime of the generation born in 2010 (OECD, 2010[34]).

Second, differences in health and health behaviours can reinforce existing social inequalities, which in turn affect a country’s social welfare. Differences in health at a young age perpetuate into adulthood, generating further inequalities in health status in adulthood but also social inequalities such as job prospects and income gaps (Michael Marmot, Peter Goldblatt, Jessica Allen, 2010[35]). Inequalities and social injustice can jeopardise a nation’s social welfare. This is a main concern for European countries, which have recently agreed on the European Pillar of Social Rights that seeks to guarantee new and more effective rights to citizens. These range from equal opportunities and access of the labour market to fairer working conditions and social protection and inclusion (Tajani and Juncker, 2017[36]).

4.5. Reducing childhood obesity will help build better future lives and stronger societies

A significant association exists between childhood obesity and educational outcomes. This relationship is driven through mediating factors such as disease, poor diet, physical inactivity, and emotional and mental health problems. An OECD analysis based on longitudinal data from five countries suggests that the presence of obesity at a young age can, in some cases, affect school grades and educational attainment later on in life. As education is a determinant of the formation of human capital, future individual socio-economic status and GDP, this effect can exacerbate the negative impact of obesity on society and the economy.

This chapter describes the implications of childhood obesity on well-being and mental health, educational outcomes, and furthermore, on the economy and the welfare of societies. Policy makers should invest in a wide range of policy interventions aimed at tackling childhood obesity, reducing the obesity stigma, tackling bullying, and improving the well-being and mental health of overweight children. Such interventions have the potential to improve the lives of children by improving educational performance and attainment, future labour market prospects and overall health and wellbeing.

Data acknowledgements

The authors would like to thank the following organisations and institutions for providing data that was used in the analyses presented in this chapter. These organisations and institutions do not bear any responsibility for the analysis or interpretation of the data.

  • The Health Behaviours in School-based Children survey (2001-02, 2005-06, 2009-10, and 2013-14) produced by the HBSC international network coordinated by the HBSC International Coordinating Centre based at the Child and Adolescent Health Research Unit in the University of St Andrews, Scotland. The HBSC data are managed by the HBSC Data Management Centre based at the Department of Health Promotion and Development in the University of Bergen, Norway.

  • The National Health and Nutrition Examination Survey (NHANES) (2001-02, 2003-04, 2005-06 and 2007-08) produced by the National Center for Health Statistics (NCHS), Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention.

  • The 1970 British Cohort Study, managed by the Centre for Longitudinal Studies at University College London, funded by the Economic and Social Research Council, and accessed through the UK Data Service.

  • The National Longitudinal Study of Adolescent to Adult Health (Add Health), managed by the Carolina Population Center, University of North Carolina at Chapel Hill, and accessed through the CPC Dataverse.

  • The Russia Longitudinal Monitoring Survey (RLMS), conducted by the Higher School of Economics and ZAO “Demoscope” together with the Carolina Population Center, University of North Carolina at Chapel Hill and the Institute of Sociology RAS, accessed through the CPC Dataverse.

  • The German Health Interview and Examination Survey for Children and Adolescents (KiGGS), run by The Robert Koch Institute, made available for this study upon request.

  • The Prevention and Incidence of Asthma and Mite Allergy (PIAMA) study, run by the National Institute for Public Health and the Environment (RIVM), the University of Utrecht, the University Medical Centre (UMC) Utrecht, and the UMC Groningen, made available for this study upon request.

References

[32] Abdullah, A. et al. (2011), “The number of years lived with obesity and the risk of all-cause and cause-specific mortality”, International Journal of Epidemiology, Vol. 40/4, pp. 985-996, http://dx.doi.org/10.1093/ije/dyr018.

[21] Adaili, M., A. Mohamed and H. Alkhashan (2017), “Association of overweight and obesity with decline in academic performance among female high-school students, Riyadh, Saudi Arabia.”, Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit, Vol. 22/12, pp. 887-893, http://www.ncbi.nlm.nih.gov/pubmed/28181664 (accessed on 22 June 2017).

[3] Anderson, A. and D. Good (2017), “Increased body weight affects academic performance in university students”, Preventive Medicine Reports, Vol. 5, pp. 220-223, http://dx.doi.org/10.1016/j.pmedr.2016.12.020.

[26] Booth, J. et al. (2014), “Obesity impairs academic attainment in adolescence: findings from ALSPAC, a UK cohort”, International Journal of Obesity, Vol. 38/10, pp. 1335-134240, http://dx.doi.org/10.1038/ijo.2014.40.

[19] Burmeister, J. et al. (2013), “Weight bias in graduate school admissions”, Obesity, Vol. 21/5, pp. 918-920, http://dx.doi.org/10.1002/oby.20171.

[10] Bustillo, A. et al. (2016), “Relationship between Low School Performance and Obesity in Adolescents: An Article Review”, World Journal of Nutrition and Health, Vol. 4, 2016, Pages 10-15, Vol. 4/1, pp. 10-15, http://dx.doi.org/10.12691/JNH-4-1-3.

[4] Carey, F. et al. (2015), “Educational outcomes associated with childhood obesity in the United States: cross-sectional results from the 2011-2012 National Survey of Children’s Health.”, The international journal of behavioral nutrition and physical activity, Vol. 12 Suppl 1/Suppl 1, p. S3, http://dx.doi.org/10.1186/1479-5868-12-S1-S3.

[24] Chen, L. et al. (2012), “A Longitudinal Study of Childhood Obesity, Weight Status Change, and Subsequent Academic Performance in Taiwanese Children”, Journal of School Health, Vol. 82/9, pp. 424-431, http://dx.doi.org/10.1111/j.1746-1561.2012.00718.x.

[27] Datar, A. and R. Sturm (2006), “Childhood overweight and elementary school outcomes”, International Journal of Obesity, Vol. 30/9, pp. 1449-1460, http://dx.doi.org/10.1038/sj.ijo.0803311.

[20] de Onis, M. et al. (2007), “Development of a WHO growth reference for school-aged children and adolescents”, Bulletin of the World Health Organization, Vol. 85/9, http://dx.doi.org/10.2471/BLT.07.043497.

[33] Devaux, M. and F. Sassi (2015), “The Labour Market Impacts of Obesity, Smoking, Alcohol Use and Related Chronic Diseases”, OECD Health Working Papers, No. 86, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jrqcn5fpv0v-en.

[30] Devaux, M. et al. (2011), “Exploring the Relationship Between Education and Obesity”, OECD Journal: Economic Studies, Vol. 1, http://dx.doi.org/10.1787/19952856.

[11] Donnelly, J. et al. (2016), “Physical Activity, Fitness, Cognitive Function, and Academic Achievement in Children: A Systematic Review”, http://dx.doi.org/10.1249/MSS.0000000000000901.

[9] Hjorth, M. et al. (2016), “Normal weight children have higher cognitive performance – Independent of physical activity, sleep, and diet”, Physiology & Behavior, Vol. 165, pp. 398-404, http://dx.doi.org/10.1016/j.physbeh.2016.08.021.

[22] Karnehed, N. et al. (2006), “Obesity and Attained Education: Cohort Study of More Than 700,000 Swedish Men*”, Obesity, Vol. 14/8, pp. 1421-1428, http://dx.doi.org/10.1038/oby.2006.161.

[31] Kelsey, M. et al. (2014), “Age-Related Consequences of Childhood Obesity”, Gerontology, Vol. 60/3, pp. 222-228, http://dx.doi.org/10.1159/000356023.

[17] Ladd, G., I. Ettekal and B. Kochenderfer-Ladd (2017), “Peer Victimization Trajectories From Kindergarten Through High School: Differential Pathways for Children’s School Engagement and Achievement?”, http://dx.doi.org/10.1037/edu0000177.

[5] Li, Y. et al. (2012), “Association between increased BMI and severe school absenteeism among US children and adolescents: findings from a national survey, 2005–2008”, International Journal of Obesity, Vol. 36, pp. 517-523, http://dx.doi.org/10.1038/ijo.2012.15.

[14] Martin, A. et al. (2014), “Lifestyle intervention for improving school achievement in overweight or obese children and adolescents”, Cochrane Database of Systematic Reviews, http://dx.doi.org/10.1002/14651858.CD009728.pub2.

[35] Michael Marmot, Peter Goldblatt, Jessica Allen, E. (2010), Fair Society, Healthy Lives (The Marmot Review), http://www.instituteofhealthequity.org/resources-reports/fair-society-healthy-lives-the-marmot-review/fair-society-healthy-lives-full-report-pdf.pdf (accessed on 3 September 2018).

[2] OECD (2017), Obesity Update 2017, OECD, Paris, http://www.oecd.org/health/obesity-update.htm.

[1] OECD (2010), Obesity and the Economics of Prevention: Fit not Fat, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264084865-en.

[34] OECD (2010), The highcost of low educational performance, the long-run economic impact of improving PISA outcomes, OECD Publishing Paris.

[28] Oppewal, H. (2010), “Concept of Causality and Conditions for Causality”, in Wiley International Encyclopedia of Marketing, John Wiley & Sons, Ltd, http://dx.doi.org/10.1002/9781444316568.wiem02059.

[6] Pan, L. et al. (2013), “The Association of Obesity and School Absenteeism Attributed to Illness or Injury Among Adolescents in the United States, 2009”, JAH, Vol. 52, pp. 64-69, http://dx.doi.org/10.1016/j.jadohealth.2012.04.003.

[23] Ruijsbroek, A. et al. (2015), “School Performance: A Matter of Health or Socio-Economic Background? Findings from the PIAMA Birth Cohort Study”, PLOS ONE, Vol. 10/8, p. e0134780, http://dx.doi.org/10.1371/journal.pone.0134780.

[15] Russell-Mayhew, S. et al. (2012), “Mental health, wellness, and childhood overweight/obesity.”, Journal of obesity, Vol. 2012, p. 281801, http://dx.doi.org/10.1155/2012/281801.

[12] Sibley, B. and J. Etnier (2003), “The Relationship Between Physical Activity and Cognition in Children: A Meta-Analysis”, Pediatric Exercise Science, Vol. 15, pp. 243-256, https://peandhealth.wikispaces.com/file/view/Sibley+and+Etnier+2003.pdf (accessed on 4 January 2018).

[13] Singh, A. et al. (2012), “Physical Activity and Performance at School”, Archives of Pediatrics & Adolescent Medicine, Vol. 166/1, p. 49, http://dx.doi.org/10.1001/archpediatrics.2011.716.

[16] Strauss, R. (2000), “Childhood obesity and self-esteem.”, Pediatrics, Vol. 105/1, p. e15, http://dx.doi.org/10.1542/PEDS.105.1.E15.

[36] Tajani, A. and J. Juncker (2017), “European Pillar of Social Rights EUROPEAN PILLAR OF SOCIAL RIGHTS European Pillar of Social Rights”, https://ec.europa.eu/commission/sites/beta-political/files/social-summit-european-pillar-social-rights-booklet_en.pdf (accessed on 13 December 2017).

[7] Torrijos-Niño, C. et al. (2014), “Physical Fitness, Obesity, and Academic Achievement in Schoolchildren”, The Journal of Pediatrics, Vol. 165/1, pp. 104-109, http://dx.doi.org/10.1016/j.jpeds.2014.02.041.

[25] Viner, R. and T. Cole (2005), “Adult socioeconomic, educational, social, and psychological outcomes of childhood obesity: a national birth cohort study”, BMJ, Vol. 330/7504, p. 1354, http://dx.doi.org/10.1136/bmj.38453.422049.E0.

[29] Vuik, S., M. Devaux and M. Cecchini (2019), “Exploring the causal relation between obesity and alcohol use, and educational outcomes”, OECD Health Working Papers, No. 109, OECD Publishing, Paris, https://dx.doi.org/10.1787/7bcd4669-en.

[18] WHO Europe (2017), Weight bias and obesity stigma: considerations for the WHO European Region, WHO Regional Office for Europe, Copenhagen, http://www.euro.who.int/__data/assets/pdf_file/0017/351026/WeightBias.pdf?ua=1 (accessed on 2 January 2019).

[8] Yates, K. et al. (2012), “Impact of metabolic syndrome on cognition and brain: a selected review of the literature.”, Arteriosclerosis, thrombosis, and vascular biology, Vol. 32/9, pp. 2060-7, http://dx.doi.org/10.1161/ATVBAHA.112.252759.

Annex 4.A. Further analyses on inequality
Annex Figure 4.A.1. Probability of being bullied, by BMI level, children aged 11-15, 2013-14, by sex and by country
Predicted probability of being bullied, with 95% confidence intervals
Annex Figure 4.A.1. Probability of being bullied, by BMI level, children aged 11-15, 2013-14, by sex and by country

Note: Mixed model with random slope. Adjusted for age, family affluence, smoking and drunkenness in lifetime. Covariates are set at fixed values (Age 13, Middle family affluence, Never drunk, Never smoke).

Source: OECD analysis based on HBSC 2013-14.

 StatLink https://doi.org/10.1787/888934007487

Annex Figure 4.A.2. Time evolution of the probability of being bullied, by BMI category and by sex, children aged 11-15, OECD countries, 2013-14
Predicted probability of being bullied, with 95% confidence intervals
Annex Figure 4.A.2. Time evolution of the probability of being bullied, by BMI category and by sex, children aged 11-15, OECD countries, 2013-14

Note: Predicted probabilities obtained from logistic model, and adjusted for BMI categories, survey year, and their interaction term, as well as age, smoking status, drinking status and family socio-economic background.

Source: OECD analysis based on four waves of HBSC survey.

 StatLink https://doi.org/10.1787/888934007506

Annex Figure 4.A.3. Time evolution of the probability of good school performance, by BMI category and by sex, OECD countries, 2013-14
Predicted probability of good performance at school, with 95% confidence intervals
Annex Figure 4.A.3. Time evolution of the probability of good school performance, by BMI category and by sex, OECD countries, 2013-14

Note: Predicted probabilities obtained from logistic model, and adjusted for BMI categories, survey year, and their interaction term, as well as age, smoking status, drinking status and family socio-economic background.

Source: OECD analysis based on four waves of HBSC survey.

 StatLink https://doi.org/10.1787/888934007525

Notes

← 1. Throughout this chapter, the nutritional status of individuals is defined according to WHO guidelines and thresholds and uses body-mass index (BMI). Overweight is defined as a BMI higher than 25 kg/m2; pre-obesity is defined as a BMI of 25-30 kg/m2; and obesity is defined as a BMI higher than 30 kg/m2. Obesity can be further divided into class I, class II and class III obesity. Class I obesity is the milder form of obesity and is defined as a BMI of 30-35 kg/m2; class II obesity is defined as a BMI of 35-40 kg/m2; while class III obesity is defined as a BMI over 40 kg/m2. Morbid obesity includes class II and class III obesity and is defined as a BMI higher than 35 kg/m2. Further information can be found in Chapter 2 - Box 2.1. Using body mass index (BMI) to define levels of adiposity.

← 2. A confounder is a variable that influences both the dependent variable (educational outcomes) and independent variable (obesity), causing a spurious (false, or fake) association.

4. The relationship between childhood obesity and educational outcomes