Over the last century, prevention has become an increasingly important part of modern healthcare. For example, the decline in infectious diseases is due more to preventive strategies such as immunisation and hygiene than to Alexander Fleming's discovery of penicillin. Life style changes and prophylactic medication have reduced coronary heart disease over the last decades. There is now an increased focus on how to prevent mental disorders.1

The arguments supporting a preventative approach to depression are compelling. While effective treatments for depression are available, the experience of a depressive illness is both distressing and associated with poorer physical health,2 occupational dysfunction3 and increased mortality.4 In most countries depression is usually treated by general practitioners, who do not feel sufficiently resourced to provide optimal treatment.5 Even in the unlikely event of optimal treatment being delivered to all affected patients, cost-effectiveness models suggest only 35–50% of the overall burden of illness would be alleviated.6

The standard definition of universal prevention is an intervention directed towards the general population which prevents the onset of new disease.1 Proving the effectiveness of universal prevention requires substantial statistical power; approximately 60 000 participants must be recruited to a trial to demonstrate a 15% rate reduction over 1 year.7 Given such methodological challenges, it is perhaps not surprising that studies to date have tended to examine either selective prevention (targeting high risk groups) or indicated strategies (targeting sub-syndromal symptoms).8

One of the major problems in attempting to prevent mental disorders is the diversity of potential risk factors, many of which are difficult or impossible to alter, such as genes and major life events. In addition, each factor has only a partial impact, so there is a limit to the maximal prevention potential when any particular risk factor is addressed. There are modifiable risk factors for depression in adolescents and adults, for example, physical inactivity,9 obesity,10 smoking,11 and according to the Wilkinson hypothesis, social inequality. Prevention strategies can also focus on promoting protective factors through the use of cognitive behavioural techniques.8

There are many practical and economic reasons why the workplace may be the ideal setting for prevention programmes. From a theoretical point of view, the prospect of using the workplace to help prevent mental illness brings together the two opposing paradigms in occupational mental health. According to the ‘toxic work-stress’ paradigm, the focus of any workplace mental health intervention should be on revealing the harmful components of the work environment with the intention of removing the exposure or buffering its effect. According to the ‘work-is-good-for-your-health’ paradigm, initiatives such as partial sickness absence which promote work participation should be considered prevention programmes in their own right.

In this month's edition of the journal, Ahola et al present the results of a randomised controlled trial testing the effectiveness of a workplace-based prevention programme whereby a resource enhancing group intervention was delivered over 4 days, which helped participants develop skills relevant to handling change and certain stressors at work. The intervention universally reduced the total symptom load of depression.

The findings presented in Ahola et al's study12 are encouraging and provide important evidence for prevention strategies in the workplace. However, from a methodological perspective, it should be noted that the study was not blinded, and the recruitment process might have created expectations that the intervention would be effective, leading to a possible Hawthorne effect. Particular caution is also needed in the interpretation of the sub-group analyses presented by the authors, particularly as it is not clear whether they were planned a priori, an unusually low cut-off score was used to define the presence of depression at baseline and there is no indication of tests for interaction to justify stratified analyses. Such approaches increase the risk of type 1 errors. Despite these concerns, an attempt to examine objective measures of symptom reduction within a randomised controlled trial, rather than just before and after user satisfaction, represents a considerable advance on much of the literature evaluating workplace training programmes.13

Despite the importance of this study and its results, the question of whether effective universal prevention of depression can be implemented in the workplace remains unanswered. While Ahola et al's intervention did not prevent incident new cases of depression, which is the usual definition of universal prevention, the authors were able to show universal symptom reduction, which is closely related. It is exciting to think universal prevention is possible in the context of the workplace. We would welcome a blinded replication of this study. The causal mechanisms also need more investigation.

A shift towards the prevention of mental health problems among workers represents a new era of workplace mental health with many exciting opportunities ahead.