Article Text

Download PDFPDF

The unknown citizen: epidemiological challenges in child mental health
  1. Amanda Kvalsvig1,
  2. Meredith O'Connor1,2,3,
  3. Gerry Redmond4,
  4. Sharon Goldfeld1,2,3
  1. 1Murdoch Childrens Research Institute, Melbourne, Australia
  2. 2Department of Paediatrics, University of Melbourne, Melbourne, Australia
  3. 3Centre for Community Child Health, Royal Children's Hospital, Melbourne, Australia
  4. 4School of Social and Policy Studies, Flinders University, Adelaide, South Australia, Australia
  1. Correspondence to Dr Sharon Goldfeld, Centre for Community Child Health, The Royal Children's Hospital, Melbourne, 2 East Clinical Offices, 50 Flemington Road, Parkville, VIC 3052, Australia; sharon.goldfeld{at}


Growing concern about the global burden of child mental health disorders has generated an increased interest in population-level efforts to improve child mental health. This in turn has led to a shift in emphasis away from treatment of established disorders and towards prevention and promotion. Prevention efforts are able to draw on a substantial epidemiological literature describing the prevalence and determinants of child mental health disorders. However, there is a striking lack of clearly conceptualised and measurable positive outcomes for child mental health, which may result in missed opportunities to identify optimal policy and intervention strategies. In this paper, we propose an epidemiological approach to child mental health which is in keeping with public health principles and with the WHO definition of health, and which is grounded in current thinking about child development. Constructs such as competence offer the opportunity to develop rigorous outcome measures for epidemiological research, while broader ideas about ‘the good life’ and ‘the good society’ derived from philosophical thinking can enable us to shape policy initiatives based on normative ideas of optimal child mental health that extend beyond individuals and undoubtedly beyond the traditional boundaries of the health sector.

  • Child Health
  • Epidemiology
  • Mental Health
  • Public Health Policy
  • Research Design in Epidemiology

Statistics from

Was he free? Was he happy? The question is absurd:

Had anything been wrong, we should certainly have heard.

WH Auden, The Unknown Citizeni

Over 70 years have elapsed since the publication of Auden's poem The Unknown Citizen, yet it remains current in the vivid depiction it gives us of the difficulties of capturing subjective individual experience using population-level statistics. This is nowhere more apparent than in the field of child mental health.

Child mental disorders represent a significant global burden; it is estimated that 20% of all children are affected.1 Some may be unwell only transiently but many continue to experience difficulties through adolescence and adulthood,1–3 with consequences that extend beyond affected individuals to their families and the societies in which they live.1

The burden of mental health disorders constitutes a strong argument for approaching child mental health as a public health concern. However, Auden's poem has an additional message for epidemiologists: our quote highlights the fact that mental health involves more than just the absence of mental disorder. Indeed as Lester Breslow argued over 40 years ago, the WHO definition of health as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity” is all too often disregarded,4 and this is arguably even more so where mental health is concerned.

There is a substantial epidemiological literature describing the prevalence of and risk factors for child mental health disorders.5–8 But do we have the evidence base we need to design population-level interventions and policies that promote the development of child mental health or wellness? We would argue that there is still much to be done. The focus on disorder currently seen in the epidemiological literature may limit the ability of public health planners to conceive of a society with high general levels of mental health, let alone to design effective interventions that promote child mental health, as distinct from preventing illness.

In this paper, we propose an epidemiological approach to child mental health which is in keeping with public health principles and with the WHO definition of health, and which is grounded in current thinking about child development. In considering concepts of positive child mental health we draw on philosophical ideas about the ‘good life’ and the ‘good society’, and on psychological investigations into the positive side of human functioning. Following Breslow,4 we argue that child mental health is to a large extent a relative concept and thus requires an understanding of structural-level determinants of the ‘good society’.

A population approach to child mental health

Growing concern has prompted national and international bodies to develop formal child mental health guidelines and recommendations.9–11 The reframing of child mental health as a public health concern has led to an increased emphasis on promotion and prevention, with calls to establish greater parity between preventive and treatment research.10 The literature on risk factors for child mental illness provides a useful basis for considering prevention strategies; the evidence on positive pathways, however, is less clear.

Reviews of child mental health promotion programmes agree that interventions focusing on early life influences such as infant attachment and warm parenting can lead to an array of desirable and long-lasting outcomes including reduced accident rates, teenage pregnancy, substance misuse, child abuse, conduct disorders and bullying.3 ,12 ,13

This is impressive, yet we are left with little indication of how these interventions might have affected children in more positive ways. As Auden might have asked: Were they free? Were they happy? The evidence appears absent.

Positive child mental health: the missing outcome

For thousands of years philosophers have debated the nature of ‘the good life’. Over 2000 years ago Plato advocated for ‘the good life’ as a happy or pleasurable life (hedonia), while Aristotle argued that ‘the good life’ was a moral life defined by virtues such as kindness, trust, loyalty and honesty (eudaimonia). Yet when the FrameWorks Institute undertook a survey of contemporary specialists in the field of early child development they found that “…when asked to talk about child mental health, the scientific experts we spoke to consistently defaulted to concepts and models of child mental illness2; the experts appeared to have no working concept of positive mental health.

One consequence of having a paradigm of disorder without a corresponding concept of health is the implicit assumption that if child mental disorders can be prevented, the resulting outcome is good mental health. This appears at odds with what parents want for their children (eg, to be happy) and with society's expectations for children, (eg, that they will grow up and be able to take on the role of adults within the community, contributing socially and economically). As Pittman et al14 point out: “Suppose we introduced an employer to a young person we worked with by saying, ‘Here's Johnny. He's not a drug user. He's not in a gang. He's not a dropout. He's not a teen father. Please hire him.’ The employer would probably respond, ‘That's great. But what does he know, what can he do?”

Another problematic consequence is the potential for confusing mental illness with low levels of well-being, when in fact people who experience long-term mental illness can have high levels of subjective well-being.15 In other words, it is misleading to conceptualise different states of mental illness and health as lying along a single continuum, where the absence of one implies the presence of the other.

Arising from these observations, the dual continuum model states that mental health and mental illness are “separate but correlated dimensions among the population”.16 Empirical research with both children and adults17 strongly supports modelling these as independent constructs.

If mental health cannot be inferred from the absence of mental disorder, it seems likely that an absence of predictors of disorders would not reliably predict health. Patel and Goodman18 have argued that protective and promotive factor epidemiology should not be seen as the converse of risk factor epidemiology, warning that “a failure to investigate protective and promotive factors rigorously may therefore hamper our ability to understand mental health problems and to promote good mental health.” Similarly, Miles et al11 observed that the terms prevention and promotion “are often used interchangeably, or sometimes prevention is used as a subset of promotion and other times the reverse is true”.

The dual continuum approach presents a route out of this confusion. It defines discrete goals for child mental health prevention and promotion, such as reducing mental illness or increasing mental health,16 and clarifies the likely population of interest in each case: for a prevention programme, this might be children who are at risk of mental illness; for promotion, on the other hand, this could include all children in a defined population. Given the conceptualisation of mental health and mental disorder as distinct yet correlated constructs, one would expect these pathways to overlap to a degree12 ,16; and empirical evidence in adults has shown that when the whole population is targeted, efforts to improve mental health for all may have the additional benefit of reducing mental health problems for some.11

Nevertheless, the advantage of maintaining a clear conceptual distinction between different pathways is illustrated by findings from a systematic review of mental health promotion in schools which concluded that long-term interventions promoting the positive mental health of all pupils in a school were likely to be more effective than short-term mental illness prevention programmes targeted at specific classes.19 We note, however, that the authors of the review found only four studies with positive mental health as an outcome, this being narrowly defined in all four as self-concept or self-esteem. As long as positive mental health remains a missing outcome, it will be difficult to make meaningful comparisons between mental health prevention and promotion strategies.

Ecological and societal perspectives on defining positive child mental health

Definitions of many common mental disorders in children are well described and widely accepted, but child mental health specialists have not produced parallel formal definitions of positive child mental health. Philosophers and psychologists, on the other hand, have made more progress towards a definition but have come at the issue from quite different starting points, with no clear consensus to date.

Traditionally, psychological research on child mental health has centred on observation and analysis of things that go wrong for individuals in the society in which they live. The emergence of positive psychology has attempted to remedy this with a focus on discovering people's inner strengths, their integrity and purpose. At minimum, it is about prevention of mental illness.20 At maximum, it is about enhancement of people's quality of life, given understanding of their social contexts.

What the health and psychology-oriented perspectives lack, and what the philosophical perspectives (particularly eudaimonic frameworks) offer, is the ambition to link mental health to more transformative goals, including the capacity to know oneself, to criticise, to step outside as well as to fit in and to be an ethical agent. In this setting, mental health may not necessarily derive from a sense of subjective well-being: individuals may choose to compromise their own well-being when taking an action that they feel to be ethically correct.21

Inherent in all these perspectives is the understanding that definitions of mental health are strongly contextual. Societies’ expectations about children's behaviour are informed by values and social norms, which together constitute a vision of ‘the good society’. This provides the context for ‘the good life’, which can inform, but also dissent from, these societal values and expectations.22

Philosophical approaches can be difficult to apply in practice, yet it is from these perspectives that we derive ideas about ‘the good life’ and the kind of society in which we might wish to live, and by extension, what positive mental health might look like at population level. Health and psychological approaches tend to lack a grander vision of ‘the good life’ or ‘the good society’, but it is undoubtedly from these perspectives that actual measurements of positive mental health are likely to come. The challenge, then, is to develop concepts of child mental health that are practical and measurable, yet not so narrow as to exclude these larger ideas.

Concepts and measures

The epidemiology of child mental health problems is now well established. Measures such as the Strengths and Difficulties Questionnaire,23 the ICD-10 diagnostic criteria24 and the Child Behaviour Checklist25 have been used to investigate the global and national-level prevalence of mental health disorders in children. It is notable, however, that these measures have been designed for the diagnosis of disorders, whereas we know very little about the prevalence of positive outcomes.

Measures of child mental disorder offer little by way of guidance for reporting positive mental health. We note that Huppert26 has operationalised positive mental health in adults by considering the Diagnostic and Statistical Manual and International Classification of Diseases criteria for anxiety and depression and then defining their opposites. However, this approach conflicts with the empirically supported dual continuum model which explicitly argues against defining health in terms of the lack of disorder. Measures of positive mental health need to be grounded in positive concepts.

An additional constraint is that constructs and measures of positive mental health developed for adults such as those reviewed by Barry12 and Patel and Goodman18 and Helliwell et al27 are not usually suitable for children, and become progressively less so in younger age groups. The general WHO definition of mental health, for example “a state of well-being in which every individual realizes his or her own potential, can cope with the normal stresses of life, can work productively and fruitfully, and is able to make a contribution to her or his community,”28 would be hard to operationalise in very young children. Measures of child mental health need to be developmentally contextualised.

No single perspective or measure has, to date, captured all of the requirements of a population measure of child mental health. Box 1 gives some examples of definitions that may be helpful to epidemiologists interested in reporting positive child mental health. While arguably none of these criteria speaks to bigger aspirations regarding ‘the good life’, they do speak about coping and competency in modern society. Definitions of competence from psychological work can present a starting point for developing measures of positive child mental health.

Box 1 Some example definitions of positive mental health relevant to children

“Children and adolescents with good mental health are able to achieve and maintain optimal psychological and social functioning and well-being. They have a sense of identity and self-worth, sound family and peer relationships, an ability to be productive and to learn, and a capacity to tackle developmental challenges and use cultural resources to maximize growth”.9

“Mental health in childhood and adolescence is defined by the achievement of expected developmental cognitive, social and emotional milestones and by secure attachments, satisfying social relationships and effective coping skills. Mentally healthy children and adolescents enjoy a positive quality of life; function well at home, in school, and in their communities; and are free of disabling symptoms of psychopathology”.29

“At its core, mental health or psychological wellbeing rests on the capacity of individuals to manage their thoughts, feelings and behaviour, as well as their interactions with others. It is essential that these core attributes of self-control, resilience and confidence be allowed to develop and solidify in the formative stages of life”.27

“Competence refers to adaptational success in the developmental tasks expected of individuals of a given age in a particular cultural and historical context. Competence by this definition is inherently multi-dimensional, because there are multiple developmental tasks salient in a given age period in a given place and time in society”.30

However, adapting this knowledge to epidemiological research is not straightforward. The authors of a recent systematic review of measures of social and emotional skills in childhood31 identified 189 measures in the published literature, but no single measure emerged as particularly well established. Similarly, although considerable progress has been made towards developing positively defined measures of subjective well-being in adults,27 there is a lack of comparable measures for children.32 ,33 In any case, as we have seen, positive mental health is not equivalent to subjective well-being, but also involves effective engagement with environmental demands. In addition, correlates of well-being in adults show considerable cultural variation.34 This is likely to be just as true of children, and may be particularly problematic in population studies where an outcome may be reported by a person (eg, a teacher) who does not share a cultural background with the child.

Competence (see box 1 for a definition) shows promise as a construct that aligns well with concepts of mental health and has a strong developmental foundation. It has been identified by the WHO as an important target for mental health promotion,28 and Miles et al11 advised exploring methods for measuring competence and its pathways as a next step in child mental health research. Measures and models of competence in individuals have been reported in the psychological literature30 ,35–36 and recently we have developed a measure of competence derived from national-level datasets of Australian children.37

Positive child mental health in context

The difference in requirements between psychometric testing and population-level investigation is not simply a difference of scale. An additional challenge of reporting child mental health in terms of developmental skills or well-being is to contextualise this information with data on the child's cultural and environmental circumstances.

In 1998, Secker38 argued for the development of definitions of positive mental health that were more consistent with health promotion principles, that is, incorporating holistic approaches to health; respect for diverse cultures and beliefs; promoting positive health as well as preventing ill health; working at structural not just individual levels; and using participatory methods.

Secker was critical of illness-centred definitions of mental health, but was also wary of the more positive definitions emerging in the psychological literature, commenting that conceptualising positive mental health solely in terms of its component parts was at odds with the principles of health promotion enumerated above. This focus on individual characteristics is likely to give an incomplete view of children's lives, in contrast to the accepted ecological model of child development outlined by Bronfenbrenner,39 where children are seen in the context of family, community and wider society.

Too narrow a focus on individuals may be ineffective if health promotion policy fails to acknowledge these powerful external influences on children's mental health. Instead, Barry12 and others40 advise that mental health promotion should not be confined to strengthening the capabilities of individuals but should also address structural determinants. As Miles et al explain: “Environments can be social, such as families, schools, communities and cultures, or physical, such as buildings, playgrounds, lakes and mountains”.11

In considering these environments, we move closer to examining ideas about ‘the good society’; and progress in this area may finally lead us to a better understanding of how societies shape children's mental health. Epidemiologists are experienced at modelling health in populations over multiple levels and are thus well placed to consider the effectiveness of strategies for mental health promotion that may require engaging with individuals, families, communities and the values of the broader society.


There is a need for the public health community to develop concepts and measures of positive child mental health that are developmentally appropriate and are not defined by disorder.

The current epidemiological research practice of reporting disorder to the exclusion of any other outcomes fails to document the full range of children's functioning. In particular, there is a lack of clearly conceptualised, measurable positive outcomes for child mental health promotion; this may result in missed opportunities to identify intervention strategies and evaluate policy decisions.

Constructs such as competence present an opportunity to develop rigorous outcome measures for epidemiological research, while broader ideas about ‘the good life’ and ‘the good society’ derived from philosophical thinking can enable us to shape policy initiatives based on normative ideas of optimal child mental health that extend beyond individuals and, undoubtedly, will take us to new frontiers beyond the traditional boundaries of the health sector.

What is already known on this subject?

  • Child mental health is an important determinant of outcomes in later life. There is increasing concern in the public health community about the global burden of child mental disorders.

  • Prevention programmes for children at risk of mental illness are able to draw on an extensive literature reporting the epidemiology of child mental disorders. However, there is no corresponding body of work describing the prevalence and predictors of positive child mental health in populations.

  • There is a lack of both concepts and measures of positive child mental health in epidemiological research.

What this study adds

  • A continuing focus on disorder prevention may limit the ability of public health planners to consider the full range of strategies available to them, particularly with regard to mental health promotion.

  • Epidemiologists should adopt the dual continuum model, considering mental health and mental illness as separate, though correlated, states. Positive mental health needs to be clearly conceptualised to inform the development of robust population-level measures.

  • Child mental health promotion should not be confined to strengthening individual competence (‘the good life’); to be effective it needs to be contextualised by society-level values (‘the good society’), taking into account the complex and reciprocal relationships between children and the environments in which they live.



  • Contributors The paper topic was proposed by AK. AK, MOC, GR and SG jointly contributed ideas and concepts to develop the argument of the paper. Similarly, all authors made a substantial contribution towards drafting of the manuscript and revising successive drafts before and after submission. All authors have given their final approval of the manuscript; AK is responsible for the overall content as guarantor.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • i The lines quoted are from The Unknown Citizen, in Another Time by WH Auden, published by Random House. Copyright Curtis Brown, Ltd.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.