Growing up in Scotland: health inequalities in the early years

CHAPTER 1 INTRODUCTION

1.1 Context

At least four of the Scottish Government's 15 National Outcomes (Scottish Government, 2007) are directly relevant in the context of an analysis of health inequality, and the avoidance of negative outcomes, in the early years. These are:

Our children have the best start in life and are ready to succeed

We have improved the life chances for children, young people and families at risk

We have tackled the significant inequalities in Scottish society

We live longer healthier lives

This report uses data from the first four years of the Growing Up in Scotland study ( GUS) to explore health inequalities in the early years. The inequalities include a number of different indicators of physical and mental health, as well as indicators of exposure to known risk factors for poor health. There has been an increasing focus on health inequalities over the last decade or so in Scotland (Scottish Government, 2008), the rest of the UK (Marmot, 2010) and globally (Wilkinson and Marmot, 2003). Coupled with this, the emphasis on early years has also grown in prominence in this period. The two most significant recent policy developments in Scotland that bring these aspects together are Equally Well, the report of the 2007 Ministerial Taskforce on Health Inequalities (Scottish Government, 2008), and the Early Years Framework that resulted from a joint initiative between the Scottish Government and the Convention of Scottish Local Authorities (Scottish Government and COSLA, 2008).

GUS can help contribute to the development and implementation of early years policy in two ways. Firstly, it can help to provide an evidence base of both the overall extent of, and inequalities in, negative outcomes for young children. Information such as this can be used to identify important areas for action and to gauge, over time, whether the character or prevalence of these changes (for example, by comparing the two groups of children in the current study and by comparing them with future cohorts). Secondly, it can help to suggest factors that might potentially exacerbate or ameliorate poor outcomes which can then result in policies or initiatives being targeted in those areas. While official statistics exist for a number of health outcomes for children, largely based on routine surveillance data, GUS is unique in its ability to paint a very broad picture of children's life in the early years, and examine the factors that influence the quality of their experiences and outcomes.

GUS is an important longitudinal research project aimed at tracking the lives of two cohorts of Scottish children from the early years, through childhood and beyond. Its principal aim is to provide information to support policy-making, but it is also intended to be a broader resource that can be drawn on by academics, voluntary sector organisations and other interested parties. Focusing initially on a cohort of 5,217 children aged 0-1 years old (the birth cohort) and a cohort of 2,859 children aged 2-3 years old (the child cohort), the first wave of fieldwork began in April 2005 and annual data collection from both cohorts has been undertaken since that time. ¹ This report focuses on children in the birth cohort.

This report comprises two main sections plus a conclusion which address the following two questions:

• What is the extent and character of health inequalities in the early years?
• What factors, if any, correlate with the avoidance of negative early health outcomes, among families from disadvantaged backgrounds?

The first section (Chapters 1-3) outlines some of the important concepts that need to be taken into consideration when exploring health inequalities. This chapter provides an overview of health inequality as a concept, ways it is measured and its significance in early life. Chapter 2 includes a summary of the measures that are explored throughout the rest of the report. Chapter 3 addresses the first of the two questions above and maps out the extent and character of inequalities in the early years based on the measures outlined in Chapter 2. Chapter 3 also includes some discussion of the ways in which different socio-demographic factors influence early life outcomes and exposures.

The second section ( Chapters 4-5) addresses the second question above. It considers a number of factors that might help explain differences in outcomes for children and explores the extent to which they are relevant once disadvantage has been controlled for. This analysis specifically explores measures that are commonly framed in terms of resilience to negative outcomes and includes factors such as neighbourhood characteristics, parenting attributes as well as the extent of social support available to parents.

All of the statistics have been weighted by a specially constructed weight to adjust for non-response and sample selection. Both weighted and unweighted sample sizes are given in each table. All analyses have been weighted and the standard errors have been adjusted to take account of the clustered sampling.

1.2 What is health inequality?

Poor health in early life has been shown to have significant and long-term consequences that reach into adulthood. Some outcomes, including the incidence of certain diseases at very late stages in life, have been linked to pre-natal circumstances (Marmot, 2010; Claussen, Davey Smith, and Thelle, 2003; Wilkinson and Marmot, 2003; Ben-Shlomo and Yuh, 2002). The fact that poor health, as well as risk factors for poor health, is significantly socially patterned is also well established (Scottish Government, 2008; Marmot, 2010). Within developed countries such as Scotland and the UK, a child's risk of a wide range of negative health outcomes and circumstances such as mortality before the age of one, low birth weight, not being breastfed, exposure to maternal smoke and alcohol in the womb, smoke in the home and accidents have been shown to be greater for children from less advantaged backgrounds, measured in terms of their family's income, social class and experience of multiple deprivation (Wadsworth and Butterworth, 2006; Marmot, 2005; ONS, 2004; Roberts, 2000).

The report of the Scottish Government's Ministerial Task Force on Health Inequalities, Equally Well, highlighted the early years as a priority area of concern and recommended a number of actions be addressed at this crucial life stage (Scottish Government, 2008). The Scottish Government/ COSLA Early Years Framework is committed to levelling the outcomes and opportunities for all children and, of critical interest to this piece of work, to identifying those children at risk of poor outcomes (Scottish Government, 2009a).

The World Health Organisation's founding definition of health was "a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity". ² Equally Well adopted that definition in its work and defined health inequalities in the early years in two ways. Firstly, inequalities can relate to negative outcomes such low birth weight or other indicators of a failure to thrive. Secondly, it can mean inequalities in exposure to risk factors that increase the likelihood of, or perpetuate, poor health outcomes. These include poor diet, lack of physical exercise, parental drug or alcohol misuse, being in care, living in a poor physical environment and family poverty. The analysis in this report is based on these definitions of health and health inequalities.

1.3 Measures of health inequality

The academic literature on health inequality can be characterised as having three broad themes. Firstly, there is an obvious focus on which aspects of health are of most interest in terms of the extent of inequality in their distribution. The health outcomes for children most typically examined include low birth weight, all cause infant mortality (death in the first year of life), cause-specific infant mortality, as well as morbidity measures such as chronic illness, acute episodes and accident rates. More broadly, health inequalities are also considered in terms of exposure to social and behavioural factors likely to be hazardous to health, such as parental smoking (in the womb and in the home), poor diet (including low breast feeding rates), and limited opportunities for exercise and early cognitive development. No single standard indicator of health inequality exists, for good reason. In most cases the availability of data shapes what analysis can be performed. More importantly, multiple indicators of inequality are necessary if policy is going to be devised to help reduce inequalities because reducing the gap for one indicator could easily leave other gaps untouched. Also, because the extent of inequality varies depending on the measure explored, one measure with a very unequal distribution could potentially overstate the extent of inequalities, while a measure for which the gap between groups is much smaller could do the opposite.

The second focus of the literature on health inequality concerns the arguably more fundamental question: "inequality of what?" This issue exists on two levels. First there is the question of whether health inequality simply refers to any unequal distribution of a health outcome, or whether it is specific to differences in health across social groups that result in less affluent people having worse outcomes (Murray et al., 1999; Braveman et al., 2000; Murray et al., 2000). This report defines health inequality as the unequal socio-economic patterning of outcomes which disadvantages less affluent children. GUS is not well placed to explore inequalities in relation to race and ethnicity or parental sexual orientation, though it could be used to explore inequality related to gender and some religious groups. This leads us to the second aspect of the debates around the question of inequality of what - through which socio-economic groups should health be examined? A significant body of literature also exists in answer to this question, though no consensus exists on the ideal way of measuring health differences (for example, Galobardes et al., 2007, 2006a, 2006b; Kaplan and Lynch, 2000). This is not the place to explore fully the debates surrounding the numerous measures that exist. However, it is important to set out clearly why each measure being used has been chosen and what it contributes to our understanding of differences in health.

The third broad theme evident in the literature relates to the many different ways in which differences in health can be quantified. This is of critical importance because the conclusions drawn about the same set of data can be very different depending on what measure of inequality has been used (Wagstaff et al., 1991). Significant debate on this topic has taken place with the main issues being the role of absolute versus relative measures and simple versus more complex indices (Wagstaff et al., 1991; Mackenbach and Kunst, 1997; Regidor, 2004a, 2004b). This literature, and its approach, is evident in the choice of indicators being used in the Scottish Government's long-term monitoring of health inequalities project (Scottish Government, 2009b) and the Scottish Public Health Observatory's guide to measuring and monitoring health inequality (Munoz-Aroyo and Sutton, 2007). This report follows the recommendation in much of the literature that both absolute and relative measures should be utilised, and that complex analysis should be complemented by simpler descriptive measures (Mackenbach and Kunst, 1997). The sections below set out why health inequality matters and outline the specific measures that used in this report.

1.4 The significance of health inequality in the early years

Having considered some of the definitional issues and controversies surrounding the study of health inequality, an arguably more fundamental question arises of why health inequalities in childhood matter? Concern about socioe-conomic differences in health, among both adults and children, is clearly a major area of policy concern across the globe. In Scotland, the National Performance Framework, Early Years Framework, and Equally Well, collectively set the policy context and the intention to close the gap between children from the most and least advantaged groups in society. In addition, the 2010 Marmot Review of health inequalities in England made the early years its highest priority for action and recommended significant investments are made to reduce inequalities from the earliest stage in life. The kinds of policy interventions Marmot's review recommends echo much of what was set out in Equally Well and the Early Years Framework in terms of support for parents in the early years, the importance of high quality childcare, and targeted interventions in the pre- and post-natal windows.

In the health context, there has been a shift in recent years away from solely focusing on adult behavioural risk factors for conditions such as hypertension or cardiovascular disease, towards various pre and post-natal factors such as low birthweight (Power, et al., 2007; Ben-Shlomo and Kuh, 2002; Barker, 1997; Wadsworth et al., 1997). For example, the 2009 annual report from Scotland's Chief Medical Officer (Scottish Government, 2009c) sets out the evidence suggesting the importance of positive early life experiences for long-term health and wellbeing. For these reasons, socio-economic differences in health in the early years are not simply a matter of immediate concern, but they also have implications that stretch for decades to come. The most important point to note is that this emphasis on the early years is equally concerned with direct health outcomes at this stage in life and with the quality of the wider nurturing environment within the family and local community, and in particular the importance of consistent and supportive parenting. The second part of this report returns to this theme in its discussion of resilience and factors that protect against negative outcomes.

1.5 What this analysis adds

Official statistics are collected about health outcomes and risk factors in the early years which can demonstrate the extent of health inequalities for a number of measures, such as low birthweight, smoking in pregnancy, and hospitalisation and death rates from specific causes. These are usually disaggregated using area level deprivation, or, in the case of mortality figures, by a measure of the occupational status of the household. However, a data source such as GUS can be used to supplement these kinds of figures in two specific ways. Firstly, a much wider set of information about the child's health and development can be collected in GUS than could ever be collected at a national level about all children. Secondly, and perhaps more importantly, far more is known about the GUS children's family and home circumstances, for example their income, other aspects of the home environment and parental attitudes and behaviours. So, although area deprivation can be a useful proxy measure of individual level deprivation, the extent and range of measures explored in this report provide more insight than deprivation can alone.

As the discussion above notes, recognition of the importance of early childhood experiences has gained prominence in recent years and consequently there is increasing interest in the potential for screening children at an early age for factors such as their meeting of developmental milestones, readiness to learn, or for emotional, social or conduct disorders, all of which GUS routinely includes. GUS might therefore usefully contribute evidence to help inform policy development in this area.