The approaches considered so far generally take a singular emphasis for intervention: for example, interventions to improve compliance with medication prioritise a medical approach; a cognitive view is adopted in the case of CBT interventions; and social and educational viewpoints are emphasised for self-help programs. It is argued, however, that a single model of intervention is not sufficient, and that there are major advantages to more comprehensive and holistic approaches that respond to multiple areas of intervention need (eg, Kramer et al 2003).

A more holistic approach to prevention focuses on identifying the multiple risk and protective factors for mental health. Risk factors are those that contribute to a person's vulnerability to relapse, whereas protective factors mitigate against relapse by enhancing wellbeing; "risk factors increase the likelihood that a disorder will develop and can exacerbate the burden of existing disorder, while protective factors give people resilience in the face of adversity and moderate the impact of stress and transient symptoms on social and emotional wellbeing, thereby reducing the likelihood of disorders." (Monograph 2000 p13).

Evidence related to the risk and protective factors for relapse of mental illness is an area of much needed research activity. Stress is commonly implicated in exacerbating mental illness (eg, Ventura et al 1989). In particular, longer-term stressors in a person's social and physical environment and lack of opportunities for 'fresh starts' are factors that are likely to increase the likelihood of relapse and impede recovery (Brown et al 1988, Paykel et al 2001). "Risk and protective factors occur through income and social status, physical environments, education and educational settings, working conditions, social environments, families, biology and genetics, personal health practices and coping skills, sport and recreation, the availability of opportunities, as well as through access to health services." (Action Plan 2000 p9).

Lists of potential risk and protective factors are provided in Monograph 2000 (pp15-16), which cautions, however, that it is important to note that while the available evidence shows that these factors are associated with mental health outcomes, the strength of association and level of evidence for 'causation' varies. Consequently, no causal relationship can be assumed for these factors; for some individuals there will be no impact of any particular factor or combination of factors, while for other people a particular factor or combination of factors may have a major impact on their ongoing mental health.

Importantly, Monograph 2000 (p34) points out that it is possible that quite different factors influence the relapse and recurrence of disorder compared with those that influence its onset (Zubrick et al 2000a). For example, a determinant of the onset of conduct disorder in children may be poor fetal growth (Zubrick et al 2000b, Breslau 1995) while determinants of the persistence of this disorder at the time a clinician sees the child may be poor parental monitoring and a deviant peer group (National Crime Prevention 1999). These latter determinants form part of the prognosis for treatment and need to be considered as targets for relapse prevention, while the former determinant, now no longer amenable to treatment, is an early risk factor and target for indicated prevention.
Top of page
Clearly, all the domains of life—environmental, physical, social and emotional—can be a source of both strength and stress. Varying levels of evidence suggest that risk and protective factors are likely to be evident in the areas of accommodation, employment, forms of meaningful activity, harmful alcohol and other drug use, physical health, social relationships, violence, and factors that shape personal resilience.

A word of caution is required regarding the implementation of relapse prevention in terms of applying a risk-based approach. Monograph 2000 presents the following potential problems of such an approach.

Firstly, risk factors have limitations as predictive tools. The extent to which any particular risk factor contributes to an episode of mental illness is not certain, as risk is based on probability. Not everyone who is exposed to a risk factor will have an adverse outcome. Knowledge related to the risks for relapse is not yet well developed and it is important to realise the limitations of our understanding in this area.

Furthermore, the nature of risk varies in terms of the element of choice or voluntariness of the risk factors identified, which affects the extent to which the risk is amenable to intervention. The Canadian Association of Gerontology (see Health Canada 1997 p8) identifies personal risks on a continuum of voluntary choice as follows:

  • Risk pursued as opportunity
  • Freely accepted risk
  • Reluctantly assumed risk
  • Risk with little or no choice
Thirdly, Monograph 2000 states that the process of identifying risk can be biased. Risk is not a neutral concept; decisions regarding what are acceptable or unacceptable levels of risk are subjective.
    "The generalized use of the 'at risk' label is highly problematic and implicitly racist, classist, sexist and a 1990s version of the cultural deficit model which located problems or 'pathologies' in individuals, families, and communities rather than in institutions and structures that create and maintain inequality" (Swadener & Lubeck 1995 p3).
The process of ascribing risk can be disempowering for consumers, as the assessment of risk factors is often carried out by services. Asking people themselves what is important for their wellbeing can generate factors very different to those determined by a service provider. Interventions can be paternalistic and at odds with the concepts of consumer empowerment and participation.
    You can put too much emphasis on relapse prevention and there's the risk of trying to be too overprotective; you can highlight vulnerability to the extent that it kind of restricts people's sense of wellbeing. —Clinician
Risk factors interact, and multiple and persistent risk factors predict more strongly than any individual risk factor (Mrazek & Haggerty 1994, Fraser 1997). Addressing a single risk factor or having a short-term orientation to prevention is likely to be ineffective. These types of simplistic approaches are all too common and derive from and underlie the fragmented, sector-specific nature of many services. Structural barriers can hinder intersectoral approaches and impede more multi-focused, holistic and intersectoral prevention interventions.

If risks are improperly identified, interventions can be targeted at the wrong factors. At best such interventions may be ineffective and a waste of scarce resources; at worst they may exacerbate other risk factors. An example of an intervention that may increase risk is the premature removal of children from their families in response to perceived risk and placing them in foster care or institutions. The stolen generation of Aboriginal children resulted from a biased, misguided, paternalistic and racist decision to remove part-Aboriginal children from the 'risks' associated with growing up within Aboriginal communities and, instead, provide them with the 'benefits' of being assimilated into the 'dominant' culture. The disastrous and multi-generational outcomes of this intervention are now evident.

The overzealous application of a risk approach is all too well understood by parents who have mental illness. These people risk the removal of their children when they become acutely unwell and then may have difficulty getting them back once they are well again. Top of page
    DOCS doesn't consider mental illness, it's outside of their Act, if they get involved the children just disappear. —Consumer

    You learn to make sure they [the children] don't get noticed. —Consumer
Finally, a risk approach focuses on the negative. A risk focus concentrates on weaknesses rather than strengths, and can encourage an approach akin to 'rescuing' those in need. Alternatively, there is a danger that enforced intervention will be seen as desirable or even necessary.

While it is important to reduce risk factors where possible, and enable people to better cope with the risk factors that they cannot modify, a concomitant focus on improving protective factors greatly enhances the effectiveness of interventions and is essential in order to place relapse prevention within a recovery orientation.