In an article in 2006, I wrote a systematic review arguing that pre-adolescent children with anxiety disorders were a neglected group. Reference Cartwright-Hatton, McNicol and Doubleday1 Scanning the epidemiological literature for prevalence of anxiety in the pre-adolescent age group, it became clear that anxiety disorders are the most common psychiatric disorder of childhood. Prevalence rates ranged from 2.6 to 41.2%, with a rate of just over 3% arising from a large, carefully conducted British study. Reference Ford, Goodman and Meltzer2 Of the studies that also reported prevalence rates for other psychological disorders, all seven found childhood anxiety to be more common than childhood depression, and four of six found childhood anxiety to be more common than disorders of behaviour. And, if anxiety is the most common psychiatric disorder of childhood, then might it actually be the most common serious and chronic health condition of childhood? Asthma is widely thought to be the most common, serious physical health condition of childhood (in high-income countries, at least), and estimates of its prevalence, according to a recent, large, multinational study, range from 2.8 to 37.6% in 6-7 year olds, Reference Asher, Montefort, Björkstén, Lai, Strachan and Weiland3 which is comparable to the estimates for anxiety disorders that I unearthed for the systematic review. Like anxiety disorders, asthma covers a range of severities and, happily, is outgrown by many children. Like anxiety disorders, it is a cause of significant morbidity and mortality. Unlike childhood anxiety disorders, it is taken seriously, widely treated and extensively researched.
Despite the fact that, even according to conservative estimates, 1 in 30 children has a diagnosable anxiety disorder, it seems likely that the majority of children never receive any kind of treatment for their condition. We do not know much about access to treatment for these individuals, but in a 2-year period of recruiting children with anxiety disorders to a treatment trial in a large British city, I managed to find just six children who had made it to child and adolescent mental health services (CAMHS). The remaining 70 were self-referred by exasperated parents, many of whom had previously tried, and failed, to get support for their child. This is supported by findings from a UK-based survey of CAMHS use, which showed that just 17.7% of 5-15 year olds with emotional disorders had accessed these services. This figure was even lower (across disorders) for younger children, and was approximately the same proportion (again, across disorders) as had received help from ministers of religion and practitioners of alternative medicine. Reference Ford, Goodman and Meltzer4 I am not going to dwell on the reasons for these difficulties. It is, I suspect, a combination of underrecognition of symptoms by parents, underrecognition of the consequences of these by primary care staff and underresourcing in CAMHS.
But what is all the fuss about? Should we be bothering to treat these children? The answer to this question lies in two parts, which I shall address in turn. First, what happens if we do not treat them? Second, what happens if we do?
What happens if we do not treat childhood anxiety?
The good news is that a significant proportion of children with anxiety disorders will grow out of it. Several large-scale, longitudinal, cohort studies, such as the New York Longitudinal Study, Reference Thomas, Chess, Plomm and Dunn5 the Dunedin Study Reference Silva and Stanton6 and the Great Smokey Mountains Study, Reference Costello, Angold, Burns, Stangl, Tweed and Erkaali7 have attested to this. However, each of these studies also suggests that a large proportion of these children will remain anxious into adolescence and early adulthood, although the nature of their anxiety diagnosis may change, and this may be less likely for some anxiety disorders than others.
Moreover, we are starting to see that children with anxiety disorders are at risk of a range of other psychiatric conditions. Longitudinal studies suggest that children with anxiety disorders are at particularly high risk of subsequent depression and probably also substance misuse, when they reach adolescence and discover that they can self-medicate. Regardless of whether children maintain their anxiety disorder or develop other diagnoses, we know that they are at risk of a range of related difficulties, including underperformance at school, relationship problems and, less widely known, markedly increased levels of suicidality. Anxiety disorders are also costly to families, and accordingly, to society: a recent Dutch study Reference Bodden, Dirksen and Bögels8 showed that having a child with an anxiety disorder imposed health-related costs on families (missed work, extra childcare, etc.) that were over 20 times higher than for non-anxious children.
What happens if we do treat childhood anxiety?
This issue of the Journal reports a trial of a cognitive-behavioural therapy (CBT) based intervention for anxiety in 7-12 year olds. In this study, over 70% of those receiving the intervention were free of their primary anxiety disorder at 6-month follow-up. Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell9 Shockingly, the first randomised trial to attempt to treat youth with anxiety disorders using a psychological intervention did not appear until 1994. This produced pretty pleasing results: using CBT with its 9- to 13-year-old sample, two-thirds were free of their primary diagnosis at the end of treatment. Reference Kendall10 Since then, there have been a few dozen trials, mostly using variants of CBT, mostly with fairly heterogeneous groups of children with anxiety disorders. A recent systematic review of these Reference Reynolds, Wilson, Austin and Hooper11 concluded that CBT was effective for the treatment of anxiety disorders in children and adolescents. When effect sizes for studies treating adolescents and children were compared, results were more favourable for the former (a ‘very large’ effect size), but there was still a moderate effect size for pre-adolescents. Unsurprisingly, results were also weaker in studies that employed an active control group, but these still showed a positive effect for CBT. Likewise, a Cochrane review of pharmacotherapy for childhood anxiety disorders Reference Ipser, Stein, Hawkridge and Hoppe12 reported positive effects for medication. Across 22 studies, medications for anxiety (largely selective serotonin reuptake inhibitors) resulted in a ‘response’ in 58.1% of those treated and 31.5% of those given a placebo. It was noted, however, that the review found no controlled evidence for benzodiazepine medications, which are still prescribed for anxiety in children and adolescents.
Unfortunately, we still know little about what happens to treated children in the longer term. Most recent studies have reported follow-up to 6 or 12 months, and these usually show maintenance or slight improvement in outcomes for treated participants. Very few trials have followed up participants for any longer, but those that have (for example Kendall et al reassessed children 7.4 years after treatment) Reference Kendall, Safford, Flannery-Schroeder and Webb13 have shown that treatment gains were maintained. Assuming that this study is representative, it seems that longer-term reductions in anxiety are possible, although, for obvious ethical reasons, the control group were not followed up during this period.
Similarly, there is precious little research examining the impact of treatment on the other difficulties that anxiety is associated with. So, for example, we do not really know whether treating childhood anxiety improves school performance or interpersonal relationships, or whether it reduces the risk of subsequent depression or substance misuse (although the 7.4-year follow-up study described above suggests that it might). There are good theoretical reasons to think that it will, but we cannot at present, be sure.
Conclusion
Despite the gaps in our knowledge, the case for treatment is pretty clear. Thirlwall et al Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell9 (this issue) show that a fairly light-touch intervention produced outcomes equivalent to those reported by the systematic review of CBT cited above. This intervention consisted of just four face-to-face sessions and four telephone calls with a therapist. Reference Thirlwall, Cooper, Karalus, Voysey, Willetts and Creswell9 Moreover, this interesting study produces some engaging evidence that therapists need not be highly skilled or experienced to achieve these results: therapists with no previous clinical experience and very little training performed as well as a group with moderate levels of experience (it should be noted that all were well-supervised and operating closely to a detailed manual).
There are a lot of children with anxiety disorders out there, and while they are not being treated, they are storing up a lot of trouble for the future. We have the means to help them, and we should be doing so.
Acknowledgement
S.C.-H. was supported by an NIHR Career Development Fellowship during the preparation of this manuscript. The views expressed are those of the author and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health.
eLetters
No eLetters have been published for this article.