People living with psychosis: The good news and the bad news

The second Australian survey of psychosis (Morgan et al., 2011, 2012) provides detailed, comprehensive information about the lives of adults with psychotic disorders living in urban Australia. These data can tell us what is working well and where the gaps in services are, and assist in developing policy and setting priorities for the future. Ideally, the massive efforts of the researchers and participants, and the substantial funding, will result in a better quality of life for people with psychotic disorders.

But will this happen? The Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) produced more than 80 publications, but had little impact on patient care (Lieberman and Stroup, 2011). The second Australian survey of psychosis included a 3-hour interview, cognitive tests, physical measures and blood tests including DNA. The survey was population based, randomly selecting and interviewing 1825 people with psychosis. It went beyond symptoms and medications to cover broader domains including social participation, housing, employment, physical health along with engagement with general practitioners (GPs), mental health services and non-governmental organisations (NGOs). Further, the 2010 survey builds on the findings of a previous survey in 1997–1998 (Jablensky et al., 1999), so changes over time can be identified. It would be a wasted opportunity if the survey generated publications and conference presentations but did not impact on the day-to-day lives of people with psychotic disorders. A sophisticated response might also involve community organisations and agencies concerned with housing, employment and social inclusion, rather than being limited to mental health services.

The two most recent issues of the Journal contain a series of papers describing some of the main findings and these are also captured in a summary report: ‘People living with psychotic illness 2010’ (Morgan et al., 2011). This report will hopefully be found scattered around desks, intranets and hard drives in mental health services, NGOs, psychiatry training programs and government offices. This editorial will highlight some of the most striking findings, with a focus on how these results might be translated into policy and practice. I should acknowledge that this editorial is not objective: I am a member of the Technical Advisory Committee for the survey, and was director of the Adelaide site.

Early intervention?

There has been vigorous debate, much in the pages of this Journal, about the most effective ways to help people in the early stages of psychotic disorders. The survey contributes hard data, drawn from a randomised national sample rather than being limited to people attending a specific service. Almost two thirds (65%) of survey participants were under the age of 25 years when the psychosis began, and 39% were less than 20 years old, supporting the need for a youth focus. On the other hand, the remaining one third of people experienced an onset after the age of 25 years, beyond the reach of many youth-oriented early intervention services. These findings suggest that either early psychosis services should not have restrictive age-based exclusion criteria, or that services for people over 25 years should also have the capacity to provide specific interventions for people experiencing their first episode of psychosis. For most, there was plenty of time for early intervention: 42% of participants reported that the psychosis developed over a period of more than 6 months, whilst for 29% the onset of psychosis occurred over a period of 1 to 6 months.

Overall, 17% of participants used the internet to access information about mental health. Most likely, these would be predominantly younger people. Access to services in the early stages of illness might be improved if there was better information about pathways to care available on the internet. There is an expanding opportunity for services to use new technology and social media to provide resources, communicate with service users, and deliver treatment programs for people with psychotic disorders (Killackey et al., 2011).

The good news: outcomes are better

Comparisons between the 2010 survey and the first Australian survey of psychosis in 1997 suggest that the course of psychotic illness may have improved over time. In 1997, 24% of people had continuous chronic illness with deterioration over time, compared to only 11% in 2010. In both surveys, about 30% had multiple episodes with partial recovery between episodes, but the proportion with good inter-episode recovery improved from 21% to 29%.

There may well be differences between the two samples reflecting differences in participant selection, and the demographics of the regions may have changed. However, another explanation for this improvement is that the changes in mental health services over this period resulted in better outcomes. In 2010, there was less use of involuntary detention and the rate of admission to hospital declined. More people had case managers, used NGO services, and attended rehabilitation or day programs.

There were other changes over the same period. The use of atypical antipsychotics increased from 37% to 78%, with the proportion of people taking clozapine rising from 12% to 20%. The controversy about whether the atypical antipsychotic drugs are any better than the typical antipsychotics is unresolved, but there is little doubt clozapine is superior to any other medication in schizophrenia and schizoaffective disorder. Finally, there could even have been changes in the nature, or proportions, of the underlying disorders presenting with psychosis over the 12–13 years between the two surveys.

The bad news: tobacco, alcohol and drugs

The bad news is that the high rates of smoking amongst people with psychosis have not changed over the last 13 years, and the prevalence of drug and alcohol disorders has increased. In both 1997 and 2010, two thirds of people with psychosis were current smokers. In contrast, only about a quarter of Australian adults smoke (Scollo and Winstanley, 2008). In 2010, people with psychotic disorders averaged 21 cigarettes per day and about 40% had high or very high levels of nicotine dependence. More males smoked (71% compared to 59% of females) and rates of smoking were slightly higher in younger people (under 35 years) compared with those between 35 and 64 years of age.

Smoking continues to be a major factor contributing to the poor physical health of people with psychosis, and participants had higher rates of respiratory disease (asthma in 30% and respiratory problems in 18%) than the general population. Most participants had an income of less than half the national average, so smoking has a substantial financial impact, reducing the money available for clothing, transport and other necessities. More than one in five participants (28%) had at some point run out of food in the previous year.

Recent research shows that smoking may also contribute to social exclusion. Smokers tend to cluster socially with other smokers, and to be on the periphery of social networks (Christakis and Fowler, 2008). The high rates of smoking amongst people with psychotic disorders may therefore have social consequences, with these people tending to be stigmatised and excluded from social engagements because of both their smoking and their mental illness.

Public health campaigns to reduce smoking have been effective in the general population but have had no impact on people with psychosis. There is an urgent need to review current policies and programs designed to help people with severe mental illness address their tobacco addiction and develop more effective approaches. The excuse that people with psychotic disorders ‘do not want to quit’ is not valid as 31% of smokers said they had tried to give up in the previous year.

Whilst the high rates of smoking are unchanged, alcohol-related disorders have increased substantially. In 1997, 38% of men and 18% of women with psychosis met criteria for a diagnosis of alcohol abuse or dependence at some time in their lives. By 2010, 58% of men and 39% of women met these criteria. Comparable figures for the general population are 35% for men and 14% for women. As with smoking, there are substantial adverse effects in terms of physical health, finances and social function.

Substance use has also escalated, with 63% of men and 42% of women in 2010 meeting criteria for abuse or dependence on cannabis or other illicit drugs. In 1997–1998, only 36% of men and 16% of women were abusing or dependent on cannabis and other drugs. In the general population, 12% of men and 6% of women meet these criteria. As with tobacco, people with psychosis were heavy users, with 58% of those using cannabis in the past year using it at least once a week and 38% using it daily or almost daily. Amphetamine use was also high, with 40% having ever used and 12.5% having used in the last year. A quarter had ever used hallucinogens, 23% had used ecstasy, 16% had used heroin and 13% had used cocaine, but only 2–4% had used these substances in the previous year.

Despite these very high rates of drug and alcohol use, only 13% of people with psychosis were participating in relevant treatment programs. The lack of effective services for people with comorbid psychosis and drug and alcohol disorders is well known but, as with smoking, there is a gap between intentions and reality. Integrated services have been shown to be more effective (Drake et al., 2007), but in Australia mental health and drug and alcohol services tend to be separate, requiring people with comorbid conditions to negotiate two complex services simultaneously. It has been suggested that a ‘no wrong door’ approach where patients receive integrated, effective services regardless of where they first present would reduce this problem. The proposed federal government Partners in Recovery initiative should help provide better coordination of existing services for people with severe mental illness.

More bad news: cardiometabolic disorders

The cardiovascular and metabolic health of people with psychoses was extremely poor. More than half of participants (54.8%) met criteria for metabolic syndrome, indicating substantial risk of cardiovascular disease. Three quarters were overweight or obese. Nine out of 10 women had abdominal obesity. Almost half of the participants were hypertensive. Only 3% engaged in more than very low or low levels of exercise and most did not eat adequate amounts of fruit and vegetables. The combination of high rates of smoking, obesity, lack of exercise and poor diet means that people with psychoses are at very high risk of early death. Notably, all of these risk factors are preventable.

Effective treatment for cardiometabolic disorders was unusual. Only half of those who knew they had hypertension were taking antihypertensive drugs, and about 40% who knew they had elevated cholesterol or glucose were receiving medication for these conditions. Despite this treatment most still had abnormal results when tested during the survey.

Similar results have been reported previously in overseas studies and smaller local studies. However, there is a striking discrepancy between the large number of publications reporting the rates and consequences of obesity, diabetes and related conditions in people with severe mental illness, and the small number of papers reporting on interventions. Many public mental health services have introduced monitoring programs, and some have begun lifestyle programs to encourage fitness and weight loss. These programs are usually run by nurses, allied health clinicians or NGOs, and do not include medical services to provide effective identification and treatment of metabolic conditions. This is analogous to setting up a cardiology outpatient service with nurses and dieticians but no doctors.

We generally assume that GPs are responsible for the identification and treatment of physical disorders, including cardiometabolic disorders, in people with mental illness. Most participants in the survey (88%) had seen a GP in the last year. On average, they had visited a GP nine times, compared to a national average of five times. In addition, most reported having had a physical examination, waist or weight measurement, blood pressure measurement and a blood test in the last year. There seems to be ample monitoring but inadequate treatment. Research exploring factors impacting on the treatment of physical disorders in people with severe mental illness is urgently needed. It is not clear why so many people are not being treated for cardiometabolic conditions, or what needs to be done to improve the quality of care.

Different strategies and models of care should be developed, implemented and evaluated. For example, a personalised medicine approach could include, where appropriate, switching away from antipsychotic medications associated with weight gain to those that are weight-neutral. The contribution of mood stabilisers and antidepressants should also be considered and where possible less obesogenic medications could be substituted. At a service level, this would involve raising awareness amongst prescribers of the relative risk of weight gain with different drugs, and education about switching between the various drugs safely.

It has also been proposed that psychiatrists should upgrade their skills in monitoring and treating cardiometabolic conditions and should prescribe drugs for hypertension, diabetes and elevated cholesterol when necessary. In this model, psychiatrists would be responsible for both the psychiatric and the cardiometabolic health of their patients.

There are also some possible models involving co-location. Physical medicine services for people with mental illness could be provided by GPs employed by the mental health service and working at a general practice clinic located within the mental health service. Alternatively, mental health outpatient clinics could be co-located as in the same large, one-stop GP super clinic. All clinicians would use the same case records and contribute to integrated treatment plans.

What are non-governmental organisations doing?

NGOs receive substantial funding to assist people with severe mental illness. The survey provided an opportunity to assess whether these services were reaching people with psychotic disorders, and what kind of services were being delivered. About 30% of participants had used NGO services in the previous year. NGOs provided the majority of group rehabilitation programs: about 22% of participants had attended an NGO program in the previous year whilst 14% had attended rehabilitation groups organised by public mental health services. The most popular NGO rehabilitation programs involved independent living skills, healthy living, fitness, stress management, social skills and creative activities. Despite the obvious need, only a very small proportion of people using NGO programs in the previous year had attended a group addressing smoking, alcohol or drug use.

In contrast, most case managers were employed by the public sector. Of the 1825 participants, 62% had a case manager from the public sector and 20% had an NGO case manager. The NGO case managers were seen more often: 64% of NGO case managers were seen once a week or more compared with 28% of public sector case managers. Despite these differences, 85% of participants with public sector case managers and 89% with NGO case managers were somewhat or very satisfied with the service they received.

Depression, anxiety and suicidality

The focus on psychotic symptoms can sometimes mean that comorbid depression and anxiety receive less attention, although they can make a substantial contribution to disability. Whilst for some participants depression was part of their psychotic disorder (e.g. those with bipolar or schizoaffective diagnoses), many had suffered trauma which might be expected to produce or exacerbate anxiety and depression. The association between childhood adversity and psychosis is increasingly recognised, with a recent meta-analysis showing that people with psychosis are 2.72 times more likely to have experienced childhood adversity than controls (Varese et al., 2012). In our survey, over half (57%) of the participants had experienced a distressing or traumatic childhood event, with 16% reporting having been sexually abused. Their adult lives were also not free of trauma, with 25% having been assaulted in the last year and 39% becoming a victim of an event such as break-in, robbery or assault.

Given these high rates of both childhood and recent trauma, along with the stresses of living with a psychotic illness, it is not surprising that 60% of participants had symptoms of anxiety and 54% had symptoms of depression in the year preceding the survey. Half had attempted suicide at some time in their lives and 11% had suicidal ideation when interviewed.

About 37% were taking antidepressants and 18% were taking anxiolytics. Fewer had received psychological therapies: 30% of all participants were engaged in counselling, psychotherapy or group therapy in the previous year, and 22% had received cognitive behavioural therapy.

What did participants say were their biggest challenges?

Participants nominated their biggest challenges over the next 12 months as financial problems, social isolation and lack of employment. Their GPs generally had a good understanding of their patients’ life situation; they also listed these three issues as the main challenges for their patients with psychotic disorders.

Sixty per cent of participants had a net weekly income between $250 and $399, and a further 20% had incomes below $250 per week. About 70% had not attended any recreational activities in the previous year, 22% felt socially isolated, and 13% had no friends. Employment has the potential to improve both income and socialisation, but there had been disappointingly little change in the rates of employment of people with psychosis between 1997 and 2010. In 2010, only one in three participants had been in paid employment during the previous year, with one in five having worked in the week prior to the interview. Some vocational services were available. About 20% of participants had received some form of assistance to find employment in the previous year, and about 10% had an employment support worker or other help to keep employment.

People with better education and training are generally more employable, and many people with psychotic disorders are educationally disadvantaged. About 18% of those surveyed had problems with basic literacy, and only 31% had completed year 12 (compared with more than half of the general population). On the positive side, 47% did have a post-school qualification, which should improve their employment prospects.

Stable accommodation is essential to being able to work, yet 5% were homeless at the time of the interview and 12% had been homeless during the previous year. There was considerable dependence on public housing, with 27% renting public housing and a further 23% on waiting lists.

Waghorn et al. (2012), describing these data in more detail, conclude that more effective vocational interventions for people with psychotic disorders are urgently needed. These interventions need to be underpinned by stable housing, good social support and the opportunity to undertake further education and training.

Children of parents with psychosis

The complexity of family relationships means that it is quite difficult to collect accurate information about parenting roles and responsibilities. The survey found that 56% of women and 26% of men with psychosis had children, and 24% of women had dependent children living with them. Only 5% of men lived with dependent children, perhaps reflecting the fact that 60% of men had never married or lived in a de facto relationship, compared with 33% of women.

The proportion of women with dependent children had doubled since the 1997 survey. Although many parents were doing well, some parents were impaired in their ability to care for their children. In addition, 17% of fathers and 25% of mothers had children with significant medical or behavioural problems. It is essential that clinicians providing care for people with severe mental illness also look out for the children by asking about parenting responsibilities and ensuring that these parents and their children are well supported.

Strengths and weaknesses of the survey

The second Australian survey of psychosis provides a unique collection of cognitive, physical health and DNA data within a national epidemiological framework, collected contemporaneously with clinical, service utilisation and sociological data. It includes a full diagnostic assessment using a validated diagnostic instrument for psychosis, administered by mental health professionals. It is not restricted to one diagnostic category, such as schizophrenia or bipolar disorder. Besides providing very rich data, the survey, under the guidance of Jablensky and Morgan (Jablensky et al., 1999; Morgan et al., 2011) has also provided training and experience in psychiatric epidemiology to two generations of researchers. Australian researchers continue to play an important role in the international effort to understand more about psychosis, and ultimately to find more effective interventions.

The response rate among the 4189 people contacted for interview from those screening positive for psychosis and randomised was 44%. People who are itinerant were harder to locate and those who do not speak English were excluded. More commonly, potential participants who did not consent were so unwell that they lacked capacity, or for other reasons (ranging from paranoia to lack of interest) refused to take part. The under-representation of people with more severe disorders is an ongoing bias in research in psychosis. Service level data such as rates of involuntary treatment is available, but collecting more specific data is generally not possible due to lack of consent.

There was a noticeable increase in the complexity of the bureaucratic requirements of ethics committees and health services between the two surveys, such that researchers involved in both found the second survey much more difficult to set up. Nevertheless, substantial reforms to mental health services are under way, and a third survey in 2020 could assess the effectiveness of these new services and show us, objectively, how the lives of people with psychosis have changed since 2010.