3. RELAPSE PREVENTION AND AVOIDANCE

Relapse can occur in any moment of the evolution of the disorder but there are some specific moments where it is more likely such the periods of transition from psychiatric institutions to community housing. 'Critically timed' psychosocial interventions (Thornicroft,2001) are proposed and have been tested in randomized trials.

3.1. Detection of high-risk subjects

If, in schizophrenia, we focus on the problem of primary prevention, we have to settle for a genetic counselling and some basic mental hygiene, insofar as there is little clinical evidence of any truly effective preventive measures. However different authors different research findings on children at risk have identified some vulnerability markers, so pessimism on this subject has diminished somewhat.
Schizophrenia prevention could work to lessen stressful conditions, or increase defence and coping mechanisms; but mainly, it could focus on actions applicable from birth or even before, to inhibit the expression of the illness in those prone to it. But these are non-specific and expensive strategies.

Currently, efforts are centred on identifying groups having an attribute that predicts very high risk for schizophrenia. However, until recently, the only reliable marker for schizophrenia was having a schizophrenic parent, since 10 - 16% of a schizophrenic parent's children develop the illness. However, it would not be justifiable to engage in wide-ranging prevention projects, given that 86 - 90% of the cohort is not at risk.
Studies on high-risk individuals are based on genetics, development psychology, studies on attention and information-processing, and measurements of intra-family processes. They include prospective studies, following a cohort over time to identify attributes of individuals or families existing before onset of the illness. It is possible that such a marker could reflect a pathophysiological or psychopathological process that contributes to the development of schizophrenia, which could possibly have aetiological implications.
There are antecedents and early warning signals of dysfunction that identify children and adolescents at risk. Various studies, covering conception to two years, have shown that some of these individuals are subject to identifiable stressful circumstances, and show early delusional symptoms.


Regarding whether these markers identify persons with a specific risk of developing schizophrenia, or merely any psychopathology, most studies show little specificity, except for a one that showed higher cognitive and attention deficits in the children of schizophrenics.
The detection of children at risk would make it possible to work with them in order to modify some vulnerability factors. In a prospective study, it was found that in children with schizoid personalities, their psychosocial adjustments was somewhat worse than other children who attended a child psychiatry clinic; as a group, they tended to be more solitary, lacking in empathy, hypersensitive, with odd ways of communicating, and often with limited interests. As adults, fewer of them had had heterosexual experiences, and more of them had sought psychiatric help at some time. Although the majority developed schizophrenic spectrum disorders, the risk of developing schizophrenia was small (Wolff, 1991)
Various studies indicate that the more serious the mother's illness, the worse her interaction with her child; it has also been shown that low socio-economic status is correlated with poor mother-child relations. Stress and the woman's risk behaviours can produce childbirth complications and create neurointegrative abnormalities, so that the child may have a difficult temperament, and the stressed mother may treat him inadequately.

3.2. Prevention strategies

Possible prevention strategies are partly based on findings (e.g. early signs of neurointegrative disorder and alteration in parent-child relations) linked to psychiatric disorders in the mother, or problems during pregnancy and childbirth. In these cases, prevention focuses on improving prenatal care, and stimulating a more favourable parent-child relationship.
Another prevention strategy centres on children with attention deficit disorder aiming at detecting families at risk in order to help their children, improving alterations in communication, affective style, and expression of emotions.
It is also known that programmes aimed at lessening negative attitudes towards psychopharmacological medication and those that aim to reduce 'expressed emotion' within the family can lower relapses.
In Scandinavia, Johannessen et al. (Johannessen, Larsen, McGlashan, & Vaglum, 2000) carried out a campaign to inform the public of the early signs of schizophrenia, aimed at increasing early intervention and reducing the duration of untreated psychosis (DUP), which had positive results. Alanen et al. (Alanen, Lehtinen, Lehtinen, Aaltonen, & Räkköläinen, 2000), in Finland, successfully used an integrated model of early treatment of schizophrenia (primarily psychotherapeutic and dynamic-systemic approaches), working intensely with families and making housecalls. This programme reduced the country's annual incidence of schizophrenia from 24.6 per 100,000 (in 1985-89) to 10.4 per 100,000 between 1990 and 1994, when the system was in place. Of the patients who had been formerly hospitalised, 40% were treated on an outpatient basis. The rate of long-term schizophrenic patients in hospital fell to zero in a few years and remained there afterwards.
Although there were some pioneers, such as Sullivan (Sullivan, 1927), only in recent years has there been interest in early intervention, and a more optimistic attitude, encouraged by the Scandinavian countries and by the International Society for the Psychological Treatment of the Schizophrenias and other Psychoses (ISPS).
The Early Psychoses Prevention and Intervention Centre (EPPIC) Programme, in Australia, proposed a similar programme.
Klosterkotter et al (2001) shows results of the "Cologne/Bonn Early Recognition - CER" project on schizophrenia.. At re-examination at an average of 9.6 years later, 79 of 160 patients had subsequently developed a schizophrenic. Best prediction values with a high positive predictive power and a low rate of false-positive predictions were achieved for 10 symptoms and symptom complexes mainly out of the group of thought, speech and perception disturbances.

McGorry (McGorry, 2001) pointed out that until 1960, dynamic psychotherapy dominated the treatment of psychoses, but later fell into disfavour, and personnel were no longer trained as much in these techniques. The trend swung towards a 'dehumanising and inefficient' behaviourism, to which cognitive techniques were later added as a compromise. However, there has recently been resurgence in interest, because the efficacy of a dynamic psychotherapeutic approach has been shown in certain kinds of cases (Hogarty, Kornblith, & Greenwald, 1995; Hogarty et al., 1997).
Birchwood and Spencer (2001) contend that even if the early detection and treatment of early signs appears to confer protection from relapse, the active ingredients of the pharmacological and psychological based treatment studies are as yet unclear.

4. SCOPE AND LIMITS OF AN EVIDENCE BASED APPROACH IN MENTAL HEALTH

The above review on psycosocial interventions in schizophrenia is mainly based in empirically based studies. But this approach has, besides its obvious advantages, some important shortcomings

The language of medicine is at once scientific, moral and political. These three languages make up a social body - which has come to be known as the " body of medicine ". Medicine is a "practical", "operational science". But as Gracia say, it is no longer quite as easy as it was a few decades ago to defend the scientific nature of medicine. Admittedly anatomy is a science, just like physiology, biochemistry or microbiology. But none of these sciences strictly identifies with medicine, even if the doctor has to be familiar with all of them to do his job. " The doctor is in fact a social agent like any other, night watchman or street cleaner. The difference, it is true, is that the doctor needs complex scientific training to carry out his work effectively. But although he requires scientific knowledge, the doctor is not a scientist".
Be that as it may, the assessment of efficacy and efficiency of treatment are presently taking on growing importance for medical practice. An "evidence-based" movement has appeared in medicine as a regulative idea, a a method and a socio-political endeavour (Henningsen, 2000). Scientifically proven therapeutic measures or "Empirically Supported Treatments" are proposed through techniques such as randomized controlled trials, the meta-analysis and the "Consumer Reports" studies. However these procedures have advantages, disadvantages ( Henningsen, 2000). They have also important ethical implications (Helmchen, 2001) since moral neutrality is a myth when referring to the incompatible ethical positions inherent in clinical and research practices (Miller, 2001)
Even if Psychiatry was one of the first medical specialties to use the tools of evidence-based medicine this approach so far has been applied more often to pharmacological than to psychological treatments, but Cochrane collaboration systematic reviews and other forms of review (i.e.The Patient Outcomes Research Team programme in Baltimore) have begun to appear(Thornicroft, 2001).This growing interest has led to the formation of task forces to define, identify, and disseminate information about empirically supported psychological interventions (Sanderson, 1998;Chambless, 1998 ;Barlow, 1999). The American Psychological Association Task Force on the Promotion and Dissemination of Psychological Procedures proposes some characteristics of empirically supported treatments (O'Donohue, 2000 ):they involve skill building, have a specific problem focus, incorporate continuous assessment of client progress, and involve brief treatment contact, requiring 20 or fewer sessions. To be "well-established," treatments for specific disorders must be shown efficacious in at least two independent randomized clinical trials. However, the task force recognizes that these findings are in part an artefact of sociological factors present in contemporary psychotherapy.

4.2. The limits of an evidence-based approach.

Traditional clinical methods of assessing the effectiveness of psychological treatments (such as intelligence testing, projective tests, or "objective" personality tests such as the MMPI-2), are rarely used in empirically supported treatments have come under attack (O'Donohue, 2000).
On the other hand, many psychiatrists have reservations about the evidence-based medicine' approach because of perceived limitations in methodology gaps in interpreting the available evidence and neglect of individual patient uniqueness in quantitative research thru manualized treatment procedures (Beutler, 2000).
Furthermore, Cochrane also sheds light on psychological and practical obstacles which must be overcome before public health care systems can utilize new scientific results. The settings of psychotherapy randomized controlled trials are highly artificial naturalistic psychotherapy and studies should be complemented by efficiency studies and evaluation of whole health care systems if they pretend to be relevant to practice (Mundt, 2001). Finally, empirically supported psychological treatments are not been effectively disseminated to the mental health professionals who deliver them and thus are not readily available to the public who requires them (Barlow, 1999, Goldfried, 1999).
Therapists complain that therapy research has only a remote resemblance to what goes on in actual clinical practice. There is a need of training of staff to implement new psychological treatments, addressing professional barriers that may limit uptake, and investigations of the 'minimum effective dose' or the key active ingredients of the intervention (Lehman & Steinwachs, 1998; National Institute for Mental Health, 1998).
To overcome these difficulties some authors propose to make more naturalistic studies and other plead to ad criteria deriving from mental health policy and economics (Buchkremer, 2001). In this sense, Barlow (Barlow, 1999) offers a way to overcome the problems of rigid manuals as well as those associated with forcing clinicians to adhere to theories and practices that are outside of their interest, experience, and expertise.
New models of research have also been proposed . Margison (Margison, 2000) supports a model of professional self-management 'practice-based evidence', as a complementary paradigm to improve clinical effectiveness in routine practice via the infrastructure of "Practice Research Networks". For the prediction of courses of treatment response Lutz et al (Lutz, 2001) combines a dose-response model with growth curve modeling to determine dose-response relations for well-being, symptoms, and functioning. Barkham (Barkham, 2001) argues for a core outcome measure (the "Clinical Outcomes in Routine Evaluation-Outcome Measure") to provide practice-based evidence for the psychological therapies to complement the evidence-based practice paradigm. Kendall et al (Kendall, 1999) proposes "normative comparisons", a procedure for evaluating the clinical significance of therapeutic interventions, consisting of comparing data on treated individuals with that of normative individuals. Mundt and Backenstrass emphasize the importance of more detailed psychopathology (thru data that can be expected from neurosciences) that can then be matched to specific psychotherapy tools (Mundt, 2001) . In addition to scientific criteria, those deriving from mental health policy and economics are also important (Buchkremer G, Klingberg S, 2001).