|Home | About | Journals | Submit | Contact Us | Français|
Schizophrenia is one of the most serious and frightening of all mental illnesses. No other disorder arouses as much anxiety in the general public, the media, and doctors. Effective treatments are available, yet patients and their families often find it hard to access good care. In the United Kingdom, as in many parts of the world, this is often due to poor service provision, but sometimes it is simply down to misinformation. In this review, we clarify the causes and presentation of schizophrenia, summarise the treatments that are available, and try to clear up a few myths.
We searched the online electronic databases Web of Knowledge, the Cochrane Library, and the current National Institute for Health and Clinical Excellence (NICE) guidelines for suitable evidence based material.
The name schizophrenia derives from the early observation that the illness is typified by “the disconnection or splitting of the psychic functions.”w1 Unfortunately, this has led to the misconception that the illness is characterised by a “split personality,” which it is not. Box 1 lists the common symptoms of schizophrenia.
People with schizophrenia typically hear voices (auditory hallucinations), which often criticise or abuse them. The voices may speak directly to the patient, comment on the patient's actions, or discuss the patient among themselves. Not surprisingly, people who hear voices often try to make some sense of these hallucinations, and this can lead to the development of strange beliefs or delusions.
Many patients also have thought disorder and negative symptoms. While negative symptoms may be less troubling to the patient, they can be very distressing to relatives. Psychiatrists often classify schizophrenia into subtypes according to the balance of symptoms that a patient manifests (box 2).
While we often think of schizophrenia as a major departure from normal health, mild symptoms can occur in healthy people and are not associated with illness.1 This has led to the conclusion that schizophrenia reflects a quantitative rather than qualitative deviation from normality, rather like hypertension or diabetes.
We have known for more than 40 years that excessive dopamine transmission in the brain's mesolimbic system plays a key role in schizophrenia. Support for this comes from several strands of evidence (box 3). How does dopamine excess lead patients to believe that their neighbours are talking about them, or that the Central Intelligence Agency (CIA) is after them? Shijit Kapur, a Canadian researcher, has proposed an attractive theory linking dopamine dysregulation to symptom formation.w2 He pointed out that one of dopamine's roles is to provide “salience” by converting a neutral mental representation into one with personal importance that grabs our attention. The theory proposes that, in psychosis, excessive dopamine adds salience to mundane and insignificant thoughts or perceptions. In this way, everyday events—eye contact with a stranger, a trivial sound, or the comments of a newsreader—are given personal importance. Delusions may then develop in an attempt to make sense of a world in which personally important events are going on around the patient all the time.
Systematic reviews show that despite its relatively low incidence (15.2/100000),2 the prevalence of schizophrenia (7.2/1000)2 is relatively high, because it often starts in early adult life and becomes chronic. The incidence of schizophrenia varies; at present it is rising in some populations (such as South Londonw3) but falling in others.3 A comprehensive global survey concluded that schizophrenia accounts for 1.1% of the total disability adjusted life years worldwide and 2.8% of the years lived with disability worldwide.w4
Schizophrenia typically presents in early adulthood or late adolescence. Men have an earlier age of onset than women, and also tend to experience a more serious form of the illness with more negative symptoms, less chance of a full recovery, and a generally worse outcome.4 Systematic reviews show that it is more common in men than women (risk ratio 1.4:12) and is more frequent in people born in cities—the larger the city and the longer the person has lived there the greater the risk.5 It is more common in migrants.6 A large and comprehensive study showed that rates of schizophrenia in African-Caribbean people living in the UK are six to eight times higher than those of the native white population.w5 Rates remain high in the children of migrants, but this is not reflected in increased rates in their home country.w6 Environmental and social factors have been implicated in this increased risk, and intriguingly the risk of schizophrenia in migrants is greatest when they form a small proportion of their local community.7
Schizophrenia is a multifactorial disorder, and the greatest risk factor is a positive family history (fig 11).). While the lifetime risk in the general population in just below 1%, it is 6.5% in first degree relatives of patients,8 and it rises to more than 40% in monozygotic twins of affected people.9 Extended family, adoption, and twin studies show that this risk reflects the genetic proximity between relative and proband.
Recently, some progress has been made in identifying the genes that increase the risk for schizophrenia. In 2002, the deCODE genetics group in Iceland identified a haplotype in the neuregulin 1 (NRG1) genew7 which seemed to double the risk of illness, a result later replicated in Scotland and Wales, South Africa, and China. Other susceptibility genes that have recently emerged are dysbindin (DTNBP1) and DISC1.
It seems likely that many risk genes exist—each of small effect and each relatively common in the general population. Patients probably inherit several risk genes, which interact with each other and the environment to cause schizophrenia once a critical threshold is crossed.
A meta-analysis has shown that patients with schizophrenia are more likely to have experienced obstetric complications, in particular premature birth, low birth weight, and perinatal hypoxia.w8 They are also slightly more likely to have been born in late winter and early spring, possibly reflecting intrauterine viral exposure. These early environmental hazards appear to have a subtle effect on brain development. In adulthood different environmental stressors act—including social isolation, migrant status, and urban life10—and this remains the case even when life events attributable to the incipient psychosis itself are excluded. The way parents raise their children does not seem to have a major impact on future vulnerability, but families do have an important part to play in the course of the illness; patients with supportive parents do much better than those with critical or hostile ones. Collectively, these risk factors point to an interaction between biological, psychological, and social risk factors that drive increasingly deviant development and finally frank psychosis.11w9
We know that stimulants like cocaine and amphetamines can induce a picture clinically identical to paranoid schizophrenia, and recent reports have also implicated cannabis. The evidence that patients with established schizophrenia smoke more cannabis than the general population is overwhelming. Well conducted and comprehensive cohort studies, like that from Dunedin in New Zealand,12 show that early cannabis use—long before psychotic symptoms appear—increases the future risk of schizophrenia fourfold, while a meta-analysis of prospective studies reported a doubling of the risk.13 This effect is robust, even after controlling for any effect of self medication,13 undermining the suggestion that early cannabis use is an attempt to alleviate distress caused by the developing illness. Only a small proportion of people who use cannabis develop schizophrenia, just as only a few of those who misuse alcohol develop cirrhosis. This probably reflects a genetically determined vulnerability to the environmental stressor, a gene-environment interaction. Indeed, variations in the dopamine metabolising COMT (catechol-O-methyltransferase) gene affect the propensity to develop psychosis in people who use cannabis.14
A recent meta-analysis and systematic review has confirmed that patients with schizophrenia have smaller whole brain volumes and larger lateral ventricles.15 Furthermore, these volume changes have greatest impact on grey matter in the frontal and temporal lobes. These deficits appear to be present even at the earliest stages of the illness, though whether they progressively worsen over the course of the illness remains contentious.
Henry grew up in Uganda and came to the UK about three years before he started to become unwell at the age of 25. He was living with his partner and had a young son. He initially became more irritable, argumentative, and more unpredictable; eventually the relationship ended and he moved out. He began to drift socially and at work.
Henry became increasingly preoccupied that he was the victim of a conspiracy that seemed to involve one of his neighbours. Over a period of months Henry became increasingly concerned that the neighbour was playing tricks on him—sending him messages and talking to him through the walls that separated their homes. Henry did not know who to turn to, and eventually he bought a knife for protection. One night Henry decided to confront the neighbour, and he was arrested shortly afterwards. Henry saw a prison psychiatrist who diagnosed paranoid schizophrenia, and he was transferred to hospital. Henry was convicted of threatening behaviour but received a hospital order under the Mental Health Act 1983. He was treated as an inpatient for the next two years with antipsychotic medication and insight oriented cognitive therapy. He is now treated and supervised by his local community forensic psychiatry team. He lives in his own housing association flat and works five days a week in a local supermarket; he was recently offered a promotion and sees his son every week.
Box 4 lists the most common positive symptoms of schizophrenia, and box 5 shows the ICD-10 (international classification of diseases, 10th revision) diagnostic criteria. However, few patients initially present with such florid symptoms. Patients are more likely to have more nebulous symptoms such as anxiety and depression, social problems, or changes in behaviour, particularly difficulties in concentrating or becoming withdrawn from their normal social life. Box 6 outlines useful screening questions for patients presenting in this manner.
If the onset of psychosis is suspected, the patient should be rapidly referred to secondary care (box 7). This will be the local early intervention or home treatment team in many parts of the UK, or the generic catchment area community mental health team. The risk that patients pose to themselves and others must be assessed (table(table)) at this first assessment and this information included in the referral. If the presence of psychotic symptoms is confirmed by a psychiatrist, then after discussion it may be appropriate for the general practitioner to prescribe an antipsychotic. Current NICE guidelines16 recommend considering and offering an oral atypical antipsychotic such as amisulpiride, risperidone, quetiapine, or olanzapine in low doses. The need for hospital admission and even the use of the Mental Health Act will depend mainly on the patient's presentation, the risk assessment, and the availability of good community support. General practitioners can contribute greatly to this decision because of their long term relationship with the patient and family.
John was in his mid-twenties when he was referred to the local early intervention in psychosis service on the advice of his counsellor. He lived with his partner and worked in a local shop at the time. For many years he had misused various illegal drugs including cannabis, amphetamine, LSD, and cocaine.
John's problems began a year or two earlier when he had a panic attack climbing a flight of stairs. He was treated with a β blocker and when this was unsuccessful he was given counselling. In the course of John's counselling sessions he revealed that he had experienced other unusual phenomena, particularly vivid dreams. He felt that he had some degree of control over these dreams though they were accompanied by a sense of not knowing whether he was really asleep.
When John was assessed by the early intervention service he reported that he sometimes thought that he could smell petrol and butane when others could not and that he could hear his phone ring when no one had called. He said he felt that people were murmuring about him, though he could not be sure, and if he checked he found nothing. He was also very worried about his physical health. Finally, he admitted that he had begun to notice unusual coincidences and links between events and people.
Paul met criteria for an “at risk mental state” for psychosis and was offered cognitive behaviour therapy with a clinical psychologist to deal with these symptoms. He declined treatment with an antipsychotic drug.
Nine months later John had made an excellent recovery—most of his symptoms had improved, he was not taking any illicit drugs, and he was still at work and with his girlfriend.
Most general practitioners with a couple of thousand patients on their list will see one or two new cases of psychosis each year. The mean duration of untreated psychosis—the time between full symptoms emerging and starting continuous antipsychotic treatment—is currently around one to two years in the UK.w10 A systematic review and meta-analysis have shown that the longer this period, the worse the outcome.17w11 The idea that reducing the duration of untreated psychosis will be reflected in improved outcome has led to a recent expansion in first episode services in the UK and other countries. Whether or not this proves to be the case,18 patients with psychotic symptoms should be identified and treated as quickly as possible.
An average general practitioner in the UK will look after about 12 patients with schizophreniaw12 and exclusively manage the care of about six. Once a patient has recovered from an acute episode of schizophrenia, current NICE guidelines recommend that they remain on prophylactic doses of antipsychotic for one to two years and continue to be supervised by specialist services. After that time, if they are well and symptom free, the drug dose can gradually be reduced and the patient carefully monitored to detect any signs of relapse; if such signs occur, then the dose must be increased until they disappear. Such a programme of careful monitoring may best be achieved by collaboration between primary and secondary care.
General practitioners are central to ensuring that patients with schizophrenia receive good quality physical health care (fig 22).19 Current NICE guidelines encourage all practices to establish a mental health register and offer regular physical health checks tailored to the needs of the patient. Special attention should be paid to screening for endocrine disorders; hyperglycaemia and hyperprolactinaemia; cardiovascular risk factors such as smoking, hypertension, and hyperlipidaemia; and side effects of medication, particularly neurological, cardiovascular, and sexual ones (box 8).
Some patients will inevitably need to be referred back to secondary care. Guideline criteria for this decision include:
Specialist mental health services in the UK are obliged to structure patient care using the care programme approach.w13 w14 This approach provides a statutory framework that aims to ensure that clinicians comprehensively consider the patient's main areas of need. It demands regular assessment of various aspects of the patient's life including mental and physical health, relationships, accommodation, occupation, finances, etc. Although the framework is cumbersome and proscriptive, its strength lies in this rigid approach to structuring care and reviewing risk. Whether such a framework can be successful given the current attempts of government and primary care trusts to reduce the costs of mental health care remains to be seen.
The first line drug for a patient with a first episode of psychosis is an oral atypical antipsychotic, such as risperidone or olanzapine (fig 33).). Drug companies have emphasised the superior side effect profile of these drugs, but in reality the atypicals have different side effects from typical antipsychotics, and they can be just as debilitating. Well conducted randomised controlled trials have shown that, except for clozapine, they are no more effective than the older typical drugs.21 22 Thus, patients with established illness who already take a typical antipsychotic, who are clinically well, and who have no troublesome side effects should not change to an atypical.16 Clinicians should consider changing patients who take typical antipsychotics and have extrapyramidal side effects to an atypical drug. Intermittent dosing regimens and drug holidays to reduce side effects are not recommended because of the increased risk of relapse. Depot preparations are usually offered to prevent covert non-concordance with treatment and to facilitate dosing regimens. The lowest effective dose of antipsychotic should be used, and the concurrent use of two or more antipsychotics should be limited to specialist services. Anticholinergic drugs should not be routinely prescribed to prevent side effects because of their adverse effects on cognition and memory.
Meta-analysis has shown that clozapine is the best drug for 20-30% of patients who are resistant to treatment.23 Treatment resistance is defined as failure to respond to two or more antipsychotics (one of which should be an atypical) when given at an adequate dose for at least six to eight weeks, and once confounding factors such as concordance failure or substance misuse have been excluded. To prevent agranulocytosis, which occurs in less than 1% of patients taking clozapine, a full blood count must be done regularly. Clozapine is the only antipsychotic that can reduce positive and negative symptoms in patients with treatment resistance, and it should be prescribed as soon as treatment resistance is confirmed.
Several psychological treatments can help ameliorate symptoms, improve functioning, and prevent relapse, although their availability is often limited by a lack of trained therapists. Systematic reviews show that cognitive behaviour therapy can reduce persistent symptoms and improve insight19 24; NICE guidelines recommend that it should be provided for at least 10 sessions over three months. Family therapy provides support and education for families. It aims to improve communication between family members, raise awareness in all people involved, and reduce distress. It can help reduce relapse rates, admission rates, symptoms, and the burden on carers, as well as improve compliance with treatment. Systematic reviews have shown that psychoeducation can reduce relapse and readmission rates and is potentially cost efficient. Other treatments with less robustly established evidence include cognitive remediation therapy and social skills training. Psychodynamic psychotherapy may increase the risk of relapse.w15
The common perception that schizophrenia has a poor prognosis is not true. More than 80% of patients with their first episode of psychosis will recover, although less than 20% will never have another episode.25 While many patients with schizophrenia have a lifelong vulnerability to recurrent episodes of illness, a large proportion will have few relapses and make a good functional recovery. Poor premorbid adjustment, a slow insidious onset, and a long duration of untreated psychosis—together with prominent negative symptoms—tend to be associated with a worse prognosis.17w16 An acute onset, an obvious psychosocial precipitant, and good premorbid adjustment all improve the prognosis.
Thanks to Paul Tabraham and Penny Collins for help preparing this manuscript.
Contributors: Both authors contributed to the conception, planning, drafting and critical revision of the article and approved the final version. MMP is guarantor.
Competing interests: MMP has received travel awards from Pfizer, Janssen-Cilag, and Eli Lily. RMM has received honorariums for speaking at meetings organised by most major producers of antipsychotic drugs, and his research group has received funding from Eli Lilly and Astra Zeneca.
Provenance and peer review: Commissioned; externally peer reviewed.