User feedback

Going further, service users have questioned the emphasis given to medication wedded, as it almost invariably is, to a biomedical framing of their experiences (Reference Rogers, Day and WilliamsRogers 1998). They have complained that side-effects such as loss of motivation, sexual dysfunction, weight gain, drowsiness and restlessness (problems not always prevented by, and often associated with, atypical antipsychotic medication) are more troubling than extrapyramidal side-effects (Reference Day, Kinderman and BentallDay 1997) and have argued that medication can actually impede or prevent healing, learning and recovery (Reference Statsny and LehmannStatsny 2007). Consequently, service users consistently call for alternative treatments that make antipsychotic medication less central and just one of many possible treatment choices (Reference PodvollPodvoll 2003; Reference Agar-JacombAgar-Jacomb 2006; Reference GrayGray 2006). These calls have resonated with a growing interest among mental health professionals exploring alternatives to current provision within the context of greater patient choice (Reference Read, Mosher and BentallRead 2004; Reference Lewis, Tarrier and DrakeLewis 2005).

New ways of thinking

Finally, critiques of the philosophical assumptions underpinning the biomedical approach to the treatment of schizophrenia have begun to gain traction. For example, the canonical view of the nature of schizophrenia suggests a disease of the brain (Reference Hyman, Lieberman and WeinbergerHyman 1998). Given this assumption, we might suppose that methodological reductionism (Box 1) would be the most useful approach to understanding schizophrenia – seeking to know the disorder through garnering more and more information about the function and dysfunction of affected brains. Indeed, this seems to have been the case throughout the history of modern psychiatry (Reference GriesingerGriesinger 1868; Reference Pincus, Henderson and BlackwoodPincus 1993; Reference Moncrieff and CrawfordMoncrieff 2001) and, in particular, during the closing decades of the 20th century, a period that witnessed major technological advances in the neurosciences (Reference Calton, Cheetham and D'SilvaCalton 2009).

However, there are certain problems inherent in this approach. First, attempting to understand a psychological experience solely by studying the brain commits the ‘mereological fallacy’ (Reference Bennett and HackerBennett 2003), that is, the act of ascribing to the constituent parts of a thing attributes that logically apply only to the whole. Hence, attempting to understand the human experience of schizophrenia requires reference to psychological predicates (distress, persecutory thoughts, hallucinations), yet these apply only paradigmatically to the human being as a whole (it is illogical, given our current level of neuroscientific understanding, to talk of a brain hallucinating or feeling distressed). It follows that if the understanding of a human experience such as schizophrenia requires an interest in the whole person, then any attempt to help that person that relies on the use of the above predicates demands a similar approach. However, conventional biomedical treatment for people diagnosed with schizophrenia relies almost entirely on medication designed solely to affect brain function, thereby treating a part of the person and not the whole.

Second, the recent interest in neuroscientific accounts of schizophrenia (Reference Calton, Cheetham and D'SilvaCalton 2009), particularly those emanating from functional neuroimaging, appears, at least to some extent, predicated on the assumption that there is something novel or exciting about demonstrating that the brain is involved in human experiences called mental illnesses. We wonder whether this may reflect a deeply sedimented Cartesian dualism wherein mind and body (brain) comprise two different and ineluctably separate substances (Descartes 1637). In this ordering of the world it would be astounding to show that a mental disorder such as schizophrenia can influence the brain (and vice versa). However, most contemporary philosophers of mind would probably draw back from a dualist approach to consciousness, instead preferring to see it as an embodied phenomenon, attendant on, but not exclusive to, the brain (Reference RoseRose 2006).

A deeper understanding of experience

In effect, we perhaps should not be so ready to stand in awe of prettily coloured images of brains claiming to represent particular states of mind. Of course the brain is involved in those aspects of conscious experience described as mental illnesses, but it does not follow that this experience can or should be wholly accounted for by reference to brain activity alone. Consequently, it might seem a little misguided to place undue emphasis on treating mental disorders such as schizophrenia by using interventions (such as antipsychotic medication) that are designed solely to alter brain activity. Further more, we might imagine that alternative, minimal (or no) medication approaches, which perhaps more readily embrace the idea of treating the whole person (and thus neither commit the mereological fallacy nor reduce human experience simply to brain function), could potentially be at least as effective as antipsychotic medication in tending the distress sometimes associated with schizophrenia.

What follows is a review of approaches to treating schizophrenia using no or minimal antipsychotic medication. Owing to our interest in trying to understand the potential efficacy and utility of the whole-person approach, we include only those studies that immersed their participants in a holistic therapeutic milieu. Therefore, we excluded treatment studies that were based solely on psychodynamic or cognitive–behavioural approaches because, like medication, these are arguably designed to influence specific aspects of human consciousness, such as intra-psychic processes or cognitive schema.

Community care

Modern psychiatry is but a fragment of the history of madness (Reference PorterPorter 2002). The treatment of schizophrenia with antipsychotic medication is, therefore, a modern phenomenon and one standing atop and alongside other, older traditions of thought and practice. For 700 years people who have experienced what modern psychiatry now terms schizophrenia have lived as members of the community in Geel, a city in Belgium (Reference Goldstein and GodemontGoldstein 2003). Prior to (and during) the period in which the use of antipsychotic medication was becoming popular in the West, ‘the mad’ would be housed with local families; these boarders would be assimilated into the family structure, receiving support and care that allowed them to function in the ‘normal’ social world despite the emotional distress that some experienced. Research in the modern era suggests that the Geel foster-family approach works because it stands apart from the biomedical model, with its emphasis on diagnosis and treatment using medication (Reference Pierloot and DemarsinPierloot 1981).

Self-help

Likewise, the ‘moral treatment’ developed at the York Retreat by William Tuke towards the end of the 18th century emanated from outside the ambit of medicine in response to the brutality of contemporaneous treatments (Reference DigbyDigby 1985). This non-medical approach encouraged the exercise of residents' self-control rather than punishment, strove for peace, respect and dignity in all relationships, and emphasised the importance of maintaining usual social activities, work and exercise (Reference WarnerWarner 1997). Contemporary data suggest that mortality at The Retreat was low compared with other asylums and its recovery rate of 54.5% for first admissions also compared favourably (Reference DigbyDigby 1985). It may seem strange to discuss such supposedly outmoded traditions of thought and practice in a journal entitled Advances in Psychiatric Treatment, yet we argue that these approaches, predicated as they were on a gentle and humane engagement with the vagaries of human experiences at the limits, and invoking respect, dignity, collective responsibility and an emphasis on interpersonal relationships as guiding principles, retain an immediacy and relevance for the contemporary treatment of schizophrenia.

The Soteria paradigm

In the late 1960s and early 1970s, there were a number of attempts to create alternatives to orthodox medication and hospital-based treatment for people diagnosed with schizophrenia (Reference CooperCooper 1967; Reference BurstonBurston 2000). These initiatives cohered around a non-medical understanding of schizophrenia as constituting an important and thoroughly meaningful aspect of an individual's life history. Contrary to the received wisdom of the day, the use of anti-psychotic medication was marginalised, with an emphasis instead on enabling individuals to go through their experience of psychosis with minimal interference and high levels of support (Reference Pullen, Campling and HaighPullen 1999). Relatively well-known UK-based initiatives included Kingsley Hall, associated with R. D. Laing and colleagues (Reference Barnes and BerkeBarnes 1971) and Villa 21, associated with David Cooper (Reference CooperCooper 1967). These ventures inspired interest at an international level and catalysed the development of similar programmes in other countries. Among these ‘second-generation’ alternatives was the Soteria paradigm, initially developed by Mosher and colleagues in the USA (Reference MosherMosher 1999) and then replicated, albeit in a slightly modified form, in Switzerland by Ciompi and colleagues (Reference Ciompi, Dauwalder and MaierCiompi 1992). Several detailed expositions of the paradigm are available (Reference WarnerWarner 1997; Reference Mosher, Read, Mosher and BentallMosher 2004a,Reference Mosher and Hendrixb; Reference WatkinsWatkins 2006).

The Soteria paradigm encompasses both theoretical and practical elements that over time have accreted into the ‘Soteria critical elements’ (Reference Aderhold, Statsny, Lehmann, Statsny and LehmannAderhold 2007), which are summarised in Box 2.

At the paradigm's core is the strong assumption that a person diagnosed with schizophrenia should be engaged with at the interpersonal level, with an emphasis on securing shared meanings and understandings of their subjective experience (Reference Mosher, Menn and MatthewMosher 1975). Overarching and all-encompassing theories of disease, professionally acquired belief systems and practices, and the chemical alteration of consciousness by antipsychotic drugs are all considered barriers to this process (Reference Mosher, Read, Mosher and BentallMosher 2004a). To this end the ethos of the original Soteria house repudiated the conventional biomedical approach (and in fact distanced itself from any consistent ideological framework). The treatment facility was a suburban house staffed by non-professionals trained to tend to people at the limits of human experience without succumbing to the usual assumptions and prejudices (Reference Mosher, Reifman and MennMosher 1973). The facility used contextual constraints to engender a defined and predictable social environment within which so-called interpersonal phenomenology could be practised (Box 3). Finally, and perhaps most importantly for the purposes of this critique, the use of antipsychotic medication was marginalised, and ordinarily such drugs were not prescribed for at least the initial 6 weeks of treatment (Reference Mosher, Menn and MatthewMosher 1975).

All of the above would perhaps be of limited interest to modern psychiatry were it not for the systematic and sustained empirical evaluation of the paradigm's efficacy that has been undertaken over the almost 40 years since its inception (Reference Calton, Ferriter and HubandCalton 2008).

Testing the paradigm

Three randomised controlled trials of the efficacy of the Soteria paradigm have been conducted: two in the USA (Reference Bola and MosherBola 2003) and one in Switzerland (Reference Ciompi, Dauwalder and MaierCiompi 1992). The 223 participants (179 in the USA study, and 44 in the Swiss) were diagnosed with first- or second-episode schizophrenia-spectrum disorders. These studies compared people treated in Soteria communities with people who received treatment as usual (TAU), meaning hospital admission and treatment with antipsychotic medication. These have generated a considerable secondary literature, with 76 citations currently identifiable (Reference Calton, Ferriter and HubandCalton 2008). The US trials found that at 6-week follow-up there were significant and similar improvements in global psychopathology in both the Soteria group and the TAU group, even though only 24% of the Soteria group had received any antipsychotic medication (and only 16% of these had received drug treatment for longer than a week). Multivariate analysis of 2-year follow-up data found significantly greater improvements in global psychopathology and ‘composite outcome’ (an eight-item assessment including employment and social functioning; Reference Bola and MosherBola 2002) significantly more participants living independently, and significantly fewer readmissions for the Soteria group compared with the TAU group (Reference Bola and MosherBola 2003). In addition, only 34% of the Soteria group had taken antipsychotic medication (continuously or intermittently), compared with 95% of the TAU group (43% of the Soteria participants had taken no antipsychotic medication at all). Only 59% of the Soteria group had received psychiatric treatment, compared with 100% of those in treatment as usual (Reference WarnerWarner 1997).

The Swiss trial reported outcomes on a range of measures, including the Brief Psychiatric Rating Scale, housing and employment status, global outcome, global autonomy, relapse rate and average medication dose (Reference Ciompi, Dauwalder and MaierCiompi 1992). At 2-year follow-up, TAU and Soteria participants demonstrated similar levels of psychopathology and functioning, with the Soteria group using significantly lower doses of antipsychotic medication (25% less during the acute treatment phase and 50% less overall). Both the US and Swiss studies reported longer stays in the Soteria communities than in hospital, but additional costs were offset by the much lower prevalence of antipsychotic use, not to mention the reduced ‘personal’ costs of not using medication. It must be borne in mind that there is no evidence to suggest that the three iterations of the Soteria paradigm cost more than treatment as usual, and some limited evidence to show that they cost less (Reference Ciompi and HoffmannCiompi 2004).

The Finnish collaborative studies

From 1967, Yrjo Alanen and colleagues in Turku, Finland, developed a primarily psychosocial approach to the care of people diagnosed with early schizophrenia. This need-adapted or integrated approach is fast becoming the standard treatment for schizophrenia in Finland (Reference Mosher, Read, Mosher and BentallMosher 2004a) and is based on several core principles (Reference Alanen, Lehtinen and RäkköläinenAlanen 1991), some of which are given in Box 4. It could be argued that this represents a holistic approach to care and, although antipsychotic medication was used, the original model advocated withholding drugs for an initial 3-week assessment period to enable psychosocially informed recovery.

The Acute Psychosis Integrated Treatment Project, which began in 1992, specifically addressed the issue of medication (Reference Lehtinen, Aaltonen and KoffertLehtinen 2000). It involved six study centres (all of which stuck to the need-adapted model), three of which employed a no or low-dose antipsychotic drug approach after an initial 3-week antipsychotic-free period (experimental group), with the remainder using antipsychotics at recommended therapeutic doses (control group). The study enrolled 106 participants (67 experimental and 39 control – the majority diagnosed with schizophrenia). Whole-cohort analysis after 2 years showed that 41% had spent less than 2 weeks in hospital over the study period, 52% had experienced no psychotic symptoms in the previous year and 40% had Global Assessment Scale (GAS) scores of 70 or more (suggesting relatively good global functioning) (Reference Endicott, Spitzer and FleissEndicott 1976). Employment data showed that 47% were working, which is interesting given that US data have suggested that only 15–20% of people diagnosed with schizophrenia are working 2 years after admission (Reference Mosher, Read, Mosher and BentallMosher 2004a). In the experimental group, 43% had never taken antipsychotic medication, whereas the corresponding figure for the control group was 6%. The experimental group had received significantly less hospital treatment and had experienced fewer psychotic symptoms during the previous year (Reference Lehtinen, Aaltonen and KoffertLehtinen 2000).

Further study

An iteration of the need-adapted approach conducted in Sweden reported results similar to those achieved with treatment as usual, but at about half of the direct economic costs (Reference Cullberg, Levander and HolmqvistCullberg 2002, Reference Cullberg, Mattsson and Levander2006). Jaakko Seikkula and colleagues in Oulu, Finland, have further refined the need-adapted approach (Reference Mosher, Read, Mosher and BentallMosher 2004a). Their open-dialogue family and network approach (Reference Seikkula, Aaltonen and AlakareSeikkula 2006) aims to treat people diagnosed with schizophrenia in their own homes. The treatment involves the service user's social network and starts within the first 24 hours of initial contact, with the general aim of generating a constructive dialogue with the person and their family in an effort to find personally meaningful understandings of their experiences. People diagnosed with schizophrenia and treated using this version of the need-adapted approach had significantly fewer relapses and residual psychotic symptoms, were more likely to be employed, spent significantly less time in hospital and used antipsychotics significantly less often than people exposed to treatment as usual (Reference Seikkula, Aaltonen and RasinkangasSeikkula 2003).

US studies

From the 1950s until the late 1970s a series of studies concerning the treatment of schizophrenia with or without antipsychotic medication was conducted in the USA (Reference Wirt and SimonWirt 1959; Reference Pasamanick, Scarpetti and DinitzPasamanick 1967; Reference Schooler, Goldberg and BootheSchooler 1967; Reference Klein and RosenKlein 1973; Reference Carpenter, McGlashan and StraussCarpenter 1977; Reference Goldstein, Rodnick and EvansGoldstein 1978; Reference Rappaport, Hopkins and HallRappaport 1978; Reference May, Tuma and DixonMay 1981). Most were placebo-controlled immediate assignment studies that used only antipsychotic medication in their experimental cohorts (and thus did not attempt to treat the whole person). For the purposes of this review only two bear further scrutiny: the studies conducted by William Carpenter and colleagues at the US National Institute of Mental Health in Maryland (Reference Carpenter, McGlashan and StraussCarpenter 1977) and by Maurice Rappaport and colleagues at the St Agnews State Hospital in California (Reference Rappaport, Hopkins and HallRappaport 1978). Both are of note because, unlike the others, they used specially designed therapeutic (albeit hospital-based) milieus that emphasised self-understanding, social adaptation, tolerance of bizarre behaviour, interpersonal support and a willingness to give considerable time and attention to a person going through crisis.

In the Carpenter study, 49 people diagnosed with ‘good-prognosis schizophrenia’, a record of adequate prior work and social functioning, and a short history of illness, were arbitrarily assigned by their psychiatrist to treatment with or without antipsychotic medication (Reference WarnerWarner 1997). There were no differences between the two groups with regard to their prognostic ratings and initial clinical characteristics. At 1-year follow-up the people assigned to the no-medication arm spent significantly less time in hospital (108 days compared with 126 days) and were significantly less likely to be readmitted (35% v. 45%) or be treated with drugs (44% v. 67%) during post-discharge follow-up (Reference Carpenter, McGlashan and StraussCarpenter 1977).

In the Rappaport study, 80 young men with a diagnosis of acute schizophrenia were randomly assigned on admission to either antipsychotic medication or a placebo, with both groups being exposed to the same therapeutic milieu. At 3-year follow-up, 73% of the medicated group had been readmitted to hospital, compared with 8% of those originally given a placebo and thereafter never medicated. The authors concluded that ‘anti-psychotic medication is not the treatment of choice, at least for certain patients, if one is interested in long-term clinical improvement’ (Reference Rappaport, Hopkins and HallRappaport 1978). This tradition of research stimulated a long-standing debate concerning the ethics of medication-free research into the treatment of schizophrenia, which continues to this day (Reference Rothman and MichelsRothman 1994; Reference CarpenterCarpenter 1997a; Reference Fins and MillerFins 1997; Reference WyattWyatt 1997; Reference Moser, Reese and SchultzMoser 2005; Reference BolaBola 2006). The consensus appears to be that such research is not associated with widespread problems of informed consent or adverse consequences to patients and is ethically justifiable in the search for new treatments (Reference Carpenter, Schooler and KaneCarpenter 1997b; Reference Bola, Lehtinen and CullbergBola 2009).

Personal experience: Tim Calton

The question feels quite easy to answer; I would not engage with psychiatric services and, in fact, would do my best to stay as far away as possible from them. These experiences came about in a particular context and for particular reasons, and I would not want them dishonoured by being forced to understand them as figments of an imagined illness. The one-size-fits-all approach to understanding and treating madness is anathema to me: where is the room for thought about personal crises, the influence of past and current stress, and just different ways of seeing the world in the monolith of psychosis/schizophrenia? Hearing voices did not bother me, though the experience of my thoughts being available to everyone in the room, the lurching inversion of the utterly private into the public, was terrifying beyond belief and led to what I can only describe as a stupor; I was literally paralysed with fear. More insidious was the thought that those people who purported to love and like me were lying, perhaps even plotting against me. It felt like falling off the calendar. I think it speaks volumes about their capacity to tolerate difference, idiosyncrasy, and uncertainty that we were able to find a way through those moments in the abyss together, without recourse to such deadening concepts as ‘psychosis’ and ‘schizophrenia’ and certainly without the need for chemical sanitation. That the people who helped in those moments were psychiatrists is an irony not lost on me (or them). That I was very fortunate to be with people with a capacity to tolerate the extreme nature of the situation is something I reflect on (and am grateful for) every day. I think the lack of treatment choices for people diagnosed with schizophrenia and/ or experiencing psychosis is abject and shames our profession.