Monday’s release of the ABS report on youth mental health has generated predictable reactions of shock and outrage. Drawing on data from the 2007 National Survey of Mental Health and Wellbeing, the Mental Health of Young People, 2007 report highlighted the fact that one in four Australians aged 16-24 years had a mental disorder in 2007, but only a quarter of them accessed mental health services (including GP treatment of mental disorders).

In The Australian, headspace CEO Chris Tanti declared this “a shocking state of affairs”. On Crikey, Richard Farmer described the ABS report as “depressing and distressing”.

However, although we certainly should be concerned about mental disorders in young people, these prevalence rates and treatment access rates need to be seen in perspective.

First, the prevalence figures are significantly boosted by relatively high rates of substance use disorders, particularly harmful use of alcohol. A superbly timed paper by Nicola Reavley and colleagues from Orygen Youth Health has just been published in the Australian and New Zealand Journal of Psychiatry. It emphasises that the gulf between prevalence and help-seeking in young people is largely due to higher rates of substance use disorders and low rates of help-seeking for these disorders, particularly by young men. In most cases, the substance involved is alcohol, and the disorder is harmful use, not dependence.

It does not take much to qualify for an ICD-10 diagnosis of harmful alcohol use, which is defined as a pattern of alcohol use that is causing damage to health. The damage can be physical or mental, and it does not have to be severe. Reavley et al noted that just over half of the cases of substance use disorders were mild. So essentially binge drinking with a relatively low level of physical or mental harm is the main problem.

This is not to say that binge drinking is a trivial issue. It is a significant factor in suicide and self-harm (and injuries more broadly, including those caused by car crashes and assaults). It increases the risk of s-xually transmitted diseases and unplanned pregnancies. It contributes to interpersonal conflict and crime and absenteeism from work and education. It is very important to try to reduce binge drinking.

However, conceptualising it as a psychiatric disorder is not helpful for the large proportion of young people who sometimes drink too much. Binge drinking does not have to be habitual to be harmful — one drinking session can have catastrophic consequences. So focusing on identification and treatment of everyone who qualifies for an alcohol use disorder is not the solution (besides, how many young male binge drinkers would be receptive to treatment?).

Instead, we need to continue with population-based strategies such as restricting the availability of alcohol and challenging marketing practices that target young people and encourage excessive consumption. Unfortunately, binge drinking is normative in some demographic groups, and it needs to be tackled at a cultural and subcultural level (among other levels). Fortunately, binge drinking often dissipates as young people mature and settle more firmly into employment and long-term relationships. Even diagnosable cases of harmful alcohol use have high rates of so-called spontaneous remission — which is the main reason prevalence rates are so much lower in older age groups, as shown in graph 1.2, “Prevalence of selected mental disorders”, in the ABS report.

The other key issue in relation to prevalence and treatment rates of mental disorders in young people is the fact that diagnosis is not the same as treatment need, contrary to the usual interpretation. According to Professor Scott Henderson (one of the architects of the original NSMHW) and colleagues, having a disorder as assessed in the NSMHW does not automatically mean needing treatment:

Our view is that having symptoms, even at case level, is necessary but not sufficient to justify treatment … it is irrational to suggest that one in five adults need treatment for a case-level mental disorder.

Elsewhere, more emphatically, he declared: ‘SYMPTOMS ≠ NEED’.

Leading US psychiatric epidemiologists have similarly argued that prevalence rates in surveys do not directly represent treatment need. Robert Spitzer, a key player in the development of the DSM (Diagnostic and Statistical Manual of Mental Disorders) published a paper  with the title: Diagnosis and need for treatment are not the same. According to Regier and colleagues: “most episodes of mental illness are neither severe nor long-lasting”.

Similarly Professor Gavin Andrews has emphasised that many cases of mental disorders are transient, and Henderson and colleagues have documented the fact that a significant proportion are non-disabling. Furthermore, people with serious problems are more likely to seek treatment than those with mild problems: the NSMHW revealed a clear dose-response between severity of disorders and rates of treatment, with people with severe disorders being much more likely to access treatment. There would also be a gradient of treatment-seeking based on duration and disability. So there is a strong bias towards treatment of the cases that most need treatment.

An important issue related to duration is that the one-in-four figure refers to 12-month period prevalence (how many people have had a disorder at some point in the past 12 months) not point prevalence (how many people have the disorder at one specific time). Unfortunately, the prevalence rates reported by the ABS will inevitably be misrepresented as point prevalence rates in statements like: “At any one time, one quarter of all young Australians have a mental illness”. The actual point prevalence rates would be significantly lower.

In the longitudinal component of the landmark US Epidemiologic Catchment Area (ECA) study, although 28% of the adult population have a mental or addictive disorder during the course of a year, only 16% have a disorder at any point in time. Unfortunately, prevalence studies like the NSMHW and the ECA lend themselves to misinterpretation.

Finally, one finding in the ABS report that is likely to be largely ignored is that young people living in the most disadvantaged areas are more likely to have a mental disorder (22% compared with 16% in the least disadvantaged areas). Disadvantage also contributes to hazardous and harmful use of alcohol and other drugs (including, of course, tobacco, the drug that causes most harm in Australia), and many other health and social problems.

Tackling poverty and social disadvantage has greater potential to improve the wellbeing of young people than a specific focus on reducing diagnosable mental disorders.

UPDATE: The ABS corrected its report on 15 October 2010, reporting that there was no significant difference in mental disorder rates between the most and least disadvantaged areas compared (22% and 26% respectively). However, there is considerable evidence from Australia and elsewhere that socioeconomic status is an important determinant of mental health.