Article Text
PDF
Abstract
Objective Short stature is associated with increased risk of ill health and mortality and can negatively impact on an individual's economic opportunity and psychological well-being. The aim of this study was to investigate the association between height and area-level deprivation by ethnic group in children in England.
Design Cross-sectional analysis of data gathered from the National Child Measurement Programme 2008/2009 to 2012/2013.
Participants/methods Children (n=1 213 230) aged 4–5 and 10–11 years attending state-maintained primary schools in England. Mean height SD score (SDS) (based on the British 1990 growth reference) was calculated for children by Income Deprivation Affecting Children Index as a measure of area-level deprivation. Analyses were performed by sex and age group for white British, Asian and black ethnicities.
Results For white British children mean height decreased 0.2 SDS between the least and the most deprived quintile. For Asian children the relationship was weaker and varied between 0.08 and 0.18 SDS. For white British boys the magnitude of association was similar across age groups; for Asian boys the magnitude was higher in the age group of 10–11 years and in white British girls aged 10–11 years the association decreased. Height SDS was similar across all levels of deprivation for black children.
Conclusions Social inequalities were shown in the height of children from white British and Asian ethnic groups. Further evaluation of height in black children is warranted. Action is needed to reduce inequalities in height by addressing the modifiable negative environmental factors that prevent healthy growth and development of children.
- Epidemiology
- Statistics
Statistics from Altmetric.com
What is already known on this topic
Short stature is linked with increased risk of ill health and mortality and reduced economic opportunity and can also impact on psychological well-being.
Environmental factors that negatively influence height in childhood (exposure to cigarette smoke, poor nutrition, infections) are modifiable through public health policy and interventions.
Evidence from relatively small British studies suggests inequalities in height are diminishing.
What this study adds
Inequalities in the height of children from white British and Asian but not black ethnic groups were found.
Highlights the need for routine monitoring of growth in childhood and targeting of interventions particularly in more economically deprived areas.
Introduction
Many factors, both genetic and environmental, influence how children develop and grow from conception through to adulthood. Around 80% (with some estimates up to 90%) of variance in height is attributed to genetic factors, leaving 20% explained by environmental influences such as exposure to cigarette smoke, poor nutrition, infectious diseases, adverse living conditions and psychological stress.1–4 These factors are more prevalent in people living in the most deprived areas in England.
Previous studies of British children have reported a social gradient in height whereby children from lower socioeconomic groups have a lower average height than their more affluent peers.5–9 The difference in height between children in the lowest and highest socioeconomic groups varies between 4 cm for 9-year-olds born in 1947 and 198710 and 1.4 cm (boys) and 1.7 cm (girls) for 10-year-olds born in 1991–1992.11 Furthermore, more recent research asserts that inequalities in height are diminishing.12–14 However, these studies have been unable to examine the relationship between height and socioeconomic status of children by ethnic group due to relatively small or unrepresentative study samples.5–9 Additionally, many of them do not draw from up-to-date data sets; therefore, their findings may not be generalisable to the current child population in England.
Short stature increases the risk of ill health and mortality, reduces economic opportunity and negatively impacts on psychological well-being though poorer educational and employment prospects and low social mobility.5 ,6 ,15–20 Therefore, it is important that inequalities in child growth are identified, particularly if the determining factors are modifiable through policy change or public health interventions.
Our aim was to investigate the association between height and socioeconomic status by ethnic group in boys and girls aged 4–5 years and 10–11 years using cross-sectional data collected by the National Child Measurement Programme (NCMP).
Methods
Sample
Data were gathered from the NCMP, an annual surveillance programme that measures the height and weight of around 1 million children aged 4–5 and 10–11 years in primary schools in England. In 2012/2013, the seventh year of data collection, 93% (1 076 824) of all eligible children attending state schools participated.
The sample comprises data for white British children gathered during 2012/2013. For Asian and black children the sample comprises data for 5 years (2008/2009 to 2012/2013) to create a sample of sufficient size to explore patterns by sex and deprivation. Data from independent and special schools are excluded from this analysis in line with guidance from the Health and Social Care Information Centre.21
Data measurement
A full description of the methods for measuring height for the NCMP has been previously published.22
Ethnicity is based on parental report and is categorised according to National Health Service definitions.23
Deprivation was classified using Income Deprivation Affecting Children Index (IDACI) 2010 which as a measure of child poverty is the best indicator of socioeconomic status in children. This was done ecologically using the 2001 Lower Super Output Area (LSOA) they are resident in. A deprivation score was assigned to each LSOA derived from the IDACI data set which represents the proportion of children aged 0–15 years in the LSOA that are living in ‘income-deprived households’ whereby higher scores indicate more deprived areas.
Statistical analysis
The 5-year trend in height for individual ethnic groups within the Asian and black ethnic groups by deprivation quintile was found to be relatively consistent so we combined Indian, Pakistani and Bangladeshi ethnicities to create one Asian ethnic group, and black African and black Caribbean ethnicities to form one black ethnic group to achieve a large enough sample for each to explore height by sex and IDACI quintile.
To account for variation in children’s ages within a particular school year, the height of individual children was adjusted for age and sex using the UK90 growth reference and is presented as a height SD score (SDS).24–26 Categorical variables were summarised using frequencies and percentages. IDACI scores were summarised in quintiles.
The descriptive analysis was performed in Microsoft Access 2007. Simple linear regression was used to determine the relationship between height and socioeconomic status in each ethnic group by age and sex using IBM SPSS Statistics V.19.
Results
The sample included 1 217 099 children of which 3869 were excluded due to invalid or missing postcode data (table 1). Therefore, these analyses included 646 097 white British children from 2012/2013, and 374 668 Asian and 192 465 black children from 2008/2009 to 2012/2013 (n=1 213 230).
Demographic information of all children measured by NCMP 2008/2009–2012/2013 (total); and for children in the sample for analysis
In each ethnic group, boys aged 4–5 years are taller than girls aged 4–5 years (figure 1A), whereas girls are taller than boys in the older age group (figure 1B). Black Caribbean and black African children are the tallest, whereas Bangladeshi boys and girls are the shortest.
Average height (centimetres) by ethnic group of boys and girls aged (A) 4–5 and (B) 10–11 years in England.
For boys and girls of white British ethnicity aged 4–5 years height SDS decreased significantly for each increase in quintile of deprivation (table 2) with a difference of 0.2 SDS between the least and the most deprived quintile (figure 2). Additionally, girls resident in the most deprived 40% of areas in England and boys resident in the most deprived 20% of areas were shorter compared with children in the 1990 baseline. The association between height and level of deprivation was similar for white British boys aged 10–11 years and weaker for girls aged 10–11 years.
Relation between height SDS and IDACI score (quintile) by ethnic group, age and sex
Mean height SD score (SDS) by Income Deprivation Affecting Children Index (IDACI) quintile, by age and sex for white British, Asian and black ethnic groups.
The association between height and deprivation for Asian children aged 4–5 years and 10–11 years was similar, albeit weaker, compared with white British children (table 2). The difference between the least and the most deprived quintile varied between 0.08 and 0.18 SDS (figure 2).
The pattern for children of black ethnicity was different from that of white British and Asian children in both age groups. Black girls aged 4–5 years resident in the most deprived areas have a higher mean height SDS compared with black girls aged 4–5 years in the least deprived areas (figure 2). For black children aged 10–11 years there was little variation in mean height SDS and deprivation (table 2).
Children of white British ethnicity are distributed relatively evenly across the deprivation scale, the proportion of children in each quintile ranges from 18% to 21%. Since whites represent the vast majority of the population and the quintiles are chosen to divide the population into fifths, this is to be expected. However, over 50% of children from Asian ethnic groups live in the most deprived 20% areas of England. The inequality is greater among children of black ethnicity, where over 65% of children live in the most deprived 20% of areas (figure 3).
Proportion of children in each Income Deprivation Affecting Children Index (IDACI) quintile by ethnic group.
Discussion
These data show a relationship between height and area-level deprivation which differed by ethnic group and in some instances by age group and sex.
White British children, living in the least deprived areas of England on average are the tallest and white children living in the most deprived areas are the shortest. This is similar to findings from the Avon Longitudinal Study of Parents and Children (ALSPAC) study for white British children aged 10 years using the educational attainment of the mother as a measure of socioeconomic status.11 ,12 The difference in height between the lowest and highest deprivation quintiles among children aged 10–11 years in our sample is 1.6 cm for boys and 1.2 cm for girls. These differences are similar to those reported in the ALSPAC study of data gathered on children born in 1991–1992 in England.11 The inequality in height by deprivation level among Asian children corresponds to about 0.4 cm in children aged 4–5 years, and 0.9 and 1.3 cm for girls and boys aged 10–11 years, respectively. The magnitude of the physiological, health and social effects of these differences is unclear. One striking finding was the difference in mean height SDS (around 0.2) for white children aged 10–11 years in relation to the British 1990 reference population, perhaps reflecting a tempo effect due to the earlier age of puberty now than in 1990.27
Interestingly, children of black ethnicity were the tallest across all age, sex and ethnic groups with little variation across levels of deprivation. A possible explanation for this finding is that black children are mostly resident in the more deprived areas of England, with <10% resident in 50% of the least deprived areas. We assigned deprivation level ecologically using the IDACI rather than an individual-level measure; therefore, at a population level, children from black ethnic groups living in more deprived areas may be actually from higher income households and of higher socioeconomic status. This may reflect less social mobility in black communities. In an equitable society it would be expected that around 20% of the population would be resident in each quintile of deprivation, assuming some degree of social gradient.
Our findings provide some support for prenatal to adult growth in boys being more sensitive to environmental factors compared with girls as shown in previous studies.5 ,18 ,28 In the older age group white British and, to a lesser extent, Asian boys showed a greater increase from the 1990 baseline than girls in the least deprived quintile, whereas in the most deprived quintile height of boys has increased less than that of girls since 1990. However, this pattern was not evident among children aged 4–5 years.
Strengths and limitations
The NCMP provides valuable data due to the large sample size greatly enhancing our understanding of the epidemiology of child growth. It can be used to examine patterns by socioeconomic and ethnic group with national coverage unlike smaller data sets.
The measure of socioeconomic status was ecologically assigned, rather than individually assigned, by area of residence for each child so may misclassify some children. However, this study was interested in the effects of the neighbourhood environment and living conditions on height of children; therefore, IDACI was an appropriate measure of social and economic deprivation.
However, there are limitations due to the surveillance approach of the NCMP including lack of data for potentially important determinants of growth such as parental height, and restriction to two age groups, therefore not covering all stages of child growth and development. Lastly, it is possible that using broad categories for ethnicity may hide differing patterns within particular ethnic groups.
Conclusions
These findings indicate that social inequalities in height persist. Since short stature at a population level may indicate poor nutrition and poor living conditions, relevant stakeholders working with local communities should ensure that growth is monitored during childhood, particularly among children in more deprived areas.
Public health practitioners have been working to address the environmental factors that can negatively influence height and growth in childhood; however, continued education of practitioners and the general public of these influences on child growth may lead to future improvements in child growth and development across the population.
Acknowledgments
We thank the Health and Social Care Information Centre and Public Health England (PHE) for enabling the use of the NCMP data.
References
Footnotes
Contributors This work was undertaken by CH for her master’s dissertation and supervised by LS and SB. CH contributed to the conception and design of the study, carried out data analyses and interpretation of findings and wrote the first draft of the manuscript. SB contributed to the conception and design of the study, reviewed and revised the manuscript and approved the final manuscript as submitted. LS contributed to the design of the study and interpretation of findings, reviewed and revised the manuscript and approved the final manuscript as submitted.
Funding Core funding for the project was provided by the NHS Strategic Health Authority bursary.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.