Why screening and treating 3-year-olds for mental health problems is not such a good idea

The Australian Federal Government is embarking upon a campaign to screen 3-year-old children for early signs of mental illness, using a GP-administered ‘Healthy Kids Check’, followed by referral to psychologists or paediatricians for intervention where this is deemed to be needed.

What does the research on screening and early intervention with young children say about the effectiveness of such actions in reducing mental illness in later life? How accurate can we be in identifying children who need and would benefit from intervention? What proportion of false-positives and false-negatives coming from the screening process are likely? What are the dangers of prematurely labelling children as having a disorder on the basis of unreliable and/or insufficient evidence? Alternatively, how likely are we to miss those who have real and serious adjustment difficulties but do not meet criteria for problem status on the day, and are not sent on for assessment and treatment? Prediction of later mental health problems from childhood is a doubtful undertaking, and resources are at risk of being wasted and outcomes are uncertain on both an individual and a population basis.

Children at this stage of development are very labile: they can be happy one minute/hour/day/week and miserable the next. They are sensitive to changing family and social contexts and have not yet developed stable patterns of self-regulated emotions and behaviour, management of peer interactions, confidence, and competence in communication skills; they are in the early stages of learning empathy, social reciprocity, and impulse control. Their adjustment and well-being is very influenced by the family environment and especially the behaviour of their parents; hence, a medical model of judging children and their mood and behaviour is misguided. Development and change are ‘normal’, and labelling at this stage is highly problematic. Who wants to give a dog (puppy) a bad name?

It is dangerous to assume simplicity and specificity in screening and diagnostic categorisation. For children generally, developmental difficulties almost always come in packages, not as single problems. Developing communication, behavioural, learning, cognitive, emotional, and attachment factors interact and combine in a child’s presentation, rather than conforming to specific psychiatric disorder categories ready for tailored treatment packages. Australian researchers have shown that treatment for parents with troubled children can be the key target for focus rather than the child alone (Barrett et al., 1996; Sanders, 2008).

Post screening, where will these potentially psychologically troubled children go for help?

Along the proposed pathways for this initiative are a number of critical stumbling blocks, especially regarding the size and accessibility of the workforce needed for such an enterprise. The proposal in the first instance is to load upon family doctors yet another duty – to screen the population of children coming into their surgeries (i.e. in a medical setting, which is likely to promote the concept of illness, rather than a continuum of developmental variation, which is what one actually sees). Most GPs are not thoroughly trained in family or paediatric psychology unless they have a personal and particular interest in this domain and have sought professional education for themselves. Do we now require them to further educate themselves so as to feel confident and competent in these added duties? Can families have confidence that their GP is abreast not only of normal child psychosocial development but, importantly, of the complexities of family life and child-rearing practices which so powerfully influence children’s mental health? The proposed measure, the Child Health Checklist, is made up of open-ended questions to be answered by parents. The reliability of this instrument appears to be unknown, and parent reports are often variable and subject to bias. To what extent will the checklist deliver valid data?

Even if there is expertise at the end of the screening process, where are the required paediatricians, psychologists, child psychiatrists, and other allied health professionals to come from? We already have a serious shortage of well-trained, appropriately qualified and accredited practitioners in Australia; there are too few places available in graduate training programs to fill the need for service providers, and a situation where it is not possible for the health workforce to keep up adequately with population growth.

Assessment and diagnosis

Emeritus Professor Allen Frances from Duke University in the USA has commented publicly on the perils of diagnosis (Frances, 2012; Hall, 2012). Health professionals are seriously concerned about the methods and measurement tools that are used for diagnosing psychological disorders. There are challenges in every step of the process and particular risks in diagnosing very young children. Individual differences in growth and development are notable at this age; labels can give rise to negative consequences for children, many, if not most of whom, will grow out of early difficulties. Problems change over time, diagnosis is not static. Degrees of aggression, oppositional behaviour, anxiety, impulsivity etc. come and go across childhood stages. If emotions and behaviour are poorly managed in families they can become of concern, but this is about family and parental health and child-rearing practices, which are critical to take into account during assessment and in facilitating positive adjustment.

The diagnostic manual which holds sway in child psychiatry has problems with reliability and validity; consensus on diagnosis is often poor, and the diagnostic categories on which it is based are always open to question and change. In considering 3-year-olds, these concerns are exacerbated by the difficulties of measuring childhood behaviours, judging their severity, their persistence, and their relevance as symptoms of disturbance in young children.

The suggestion that children attracting a diagnosis might be medicated is of major concern. There are very few psychotherapeutic drugs that have proven their worth for paediatric use: most of those in use are borrowed from adult medications whose value and safety has not been demonstrated for children. Side effects are common and can be serious.

Are such campaigns cost-effective in the long term and do they reduce risk for later mental health difficulties? While there is some continuity in adjustment problems from childhood to adolescence and beyond, prediction of later mental health from developmental characteristics in young children is imperfect (Najman et al., 2007; Prior et al., 2000). Thus, the potential for disappointing results and wasted resources in this program is considerable.