AIMS AND METHOD
To investigate the attitudes that psychiatrists hold towards people with mental illness. Each member of the Royal College of Psychiatrists in the UK was sent a questionnaire based on previous research in this area, supplemented with relevant questions on management.
2813 of 6524 questionnaires were returned (43%). Psychiatrists’ attitudes compared favourably with those of the general population. Among other findings, they believed that the risk of dangerousness was overemphasised, that misdiagnosis of schizophrenia in Black people is common, and that polypharmacy and the use of antipsychotic medication above British National Formulary levels occurs too often.
Psychiatrists’ attitudes are substantially more favourable towards people with mental illness than those of the general population with individual, but important, exceptions. Some aspects of psychiatric management, especially of antipsychotic medication, may undermine this, however. Comparison with other groups, e.g. general practitioners, nurses and social workers, would be useful in planning how to reduce the stigmatisation of people with mental illness.
Stigmatisation of people with mental illness, especially schizophrenia, seriously affects their lives by its effects, for example on job prospects and relationships. The Royal College of Psychiatrists’ anti-stigma campaign, Changing Minds: Every Family in the Land, was established to change this. It was a 5-year national campaign that aimed to challenge perceptions of mental health problems and increase knowledge of mental issues. Initially it was agreed that we needed to look at our own attitudes, particularly as there has been serious criticism of these (Breggin, 1991). The Schizophrenia subgroup of the campaign developed a questionnaire to assess attitudes of psychiatrists’ to mental illness, especially schizophrenia.
The questionnaire used items from two survey instruments - the Community Attitudes to Mental Illness (CAMI), developed by Taylor & Dear (1981), and one developed at the start of the campaign by the Campaign Management Committee and its Scientific Advisory Committee, carried out by the Office for National Statistics (ONS) for the general public (Crisp et al, 2000). These were supplemented with questions about schizophrenia and its management, developed by the group. Information was requested about respondents’ gender, grade, years since qualification and percentage of case-load with schizophrenia.
The questionnaire was piloted by sending it to psychiatrists in the Southampton area. The response rate was excellent, with 25 of 26 returned, and a range of comments was received that were considered in the redesign of the questionnaire. The questionnaire was estimated to take an average of 8 minutes to complete. The revised questionnaire was then sent to UK members of the Royal College of Psychiatrists, who were asked to return it to a Post Office Box with anonymity assured. The questionnaire was designed using the Teleform computer program for automated data input (produced by Cardiff Software, Inc.), which reads standardised input forms into a database developed for statistical analysis using the Statistical Package for the Social Sciences version 10.
The data from the questionnaire were categorical and analysed by cross-tabulations of the attributes and assessing association using the χ 2 test. All results reported were significant at a level of P<0.05.
Altogether, 2813 questionnaires were returned (43% of the total). All were included, although some had uncompleted questions: 57.4% were male and 36.8% female (5.8% did not disclose their gender); 65% were consultants, 22% trainees and 7% other grades (6% unknown). Slightly more females than males and more consultants (65%: 55%) responded than other grades, compared with the characteristics of the membership as a whole.
Among the statements from the CAMI (Table 1), it was said that mental hospitals are not viewed as outdated means of treating people. Psychiatrists generally believe that less emphasis should be placed on protecting the public from people with mental illness (59.2%), but some (27.5%) disagree. However, a large majority (95.2%) believes that people with mental illness are far less of a danger than most people suppose, with just 2% disagreeing.
The responses drawn from the ONS survey (Table 2) confirmed that most psychiatrists thought that ‘someone with schizophrenia’ tended not to be dangerous to others, but did tend to be unpredictable and perhaps hard to talk to. However, they were certainly not to blame for their condition or able to pull themselves together. They would improve with treatment, but were ambivalent about whether they would eventually recover and whether they ‘ feel the way we all feel at times’.
A small majority of the psychiatrists who responded to the question (53.5%) believe that causes of schizophrenia are a balance of both social and biological factors, compared with primarily biological factors (46.1%) (Table 3). Few, however, think that gene therapy will eradicate it. Misdiagnosis of schizophrenia in Black people is considered common (47.9%), with only a quarter considering it uncommon. Although most believe that discussing contents of delusions and hallucinations is appropriate (90.6%) and that family therapy is effective (80.7%), there is still a sizeable group who do not or have no opinion. Many think that antipsychotic medication is prescribed above British National Formulary (BNF) limits too often and that polypharmacy is too common. When telling patients and carers that they have schizophrenia, psychiatrists talked about diagnosis, causation and family relationships, but less about leisure activities, financial matters and accommodation.
There were few differences between male and female respondents, but with a trend for the more ‘liberal’ position in women. Women were a little more likely to agree that misdiagnosis in Black patients was more common (49.6%: 46.2%), over-prescribing was too common (46.3%: 43%), there was too much emphasis on protecting the public (62.7%: 57%), and disagree with the statement that they would not want to live next door to someone who has been mentally ill (85.4%: 79.1%). They were, however, more pessimistic about patients with schizophrenia recovering (24.4%: 28.2%) and a bit more concerned about them being babysitters (60%:61.4% agreeing, 16.5%: 12.7% disagreeing).
Proportion of caseload with schizophrenia
There were a number of differences between the groups (0%, 1-10%, 11-30%, over 30%), increasing with difference in higher caseload. Higher caseloads correlated with increased perception of a major role in ensuring optimal functioning in areas of work, leisure, financial matters, and accommodation; strongly agreeing with the need to discuss contents of delusions and hallucinations and effectiveness of family; decreased perception of misdiagnosis in Black people as common; disagreeing that prescription of antipsychotics (above BNF limits) occurs too commonly (a similar number agree but there are fewer ‘no opinions’); agreement that polypharmacy occurs too often; more strongly agreeing in telling patients about their diagnosis and enjoying working with people with schizophrenia (6.8%: 9.4%: 17.1%: 28.2% strongly agree: 65.1%: 70%: 69.3%: 46.9% agree).
There was no difference in responses to CAMI statements. However, there was some difference with the ONS questions. Someone with schizophrenia was viewed as less likely to be unpredictable the higher the caseload of patients with schizophrenia and less likely to be hard to talk to, comparing those with no patients on caseload with the others.
There were many differences between the different groups: consultants favouring causation as primarily biological more often than as a balance of biological and social factors (50.8%: 48.6%), compared with trainees (58%: 41.8%) and others (65%: 35%). Consultants were more likely to say that gene therapy will not eventually eradicate schizophrenia: consultants 63.5%: trainees 74.1%: others 64.4%. Concerning discussion of content of delusions and hallucinations: trainees were more likely to strongly agree (38.6%: 48.7%: 39.4%). Misdiagnosis in black people is viewed to occur more commonly in the ‘ other’ group: 47.3%: 44.1%: 63.3% agree - 25.5%: 30%: 13.5% disagree; prescription above BNF limits is thought to be more common by trainees: 43.3%: 48.1%: 45.1% agree: 32.8%: 39% 31.6% disagree. Similar differences were seen with polypharmacy: 65.1%: 74.6%: 65.4% and not routinely prescribing antiparkinsonian drugs with antipsychotics: 60.5%: 70%: 47.9%. Consultants and even more, ‘other’ grades slightly less often or less strongly agree with talking about diagnosis to patients: 91.1%: 95.7%: 84.9%. Consultants are more likely to talk about finance and accommodation. Enjoyment of working with patients with schizophrenia is slightly higher in trainees: 78.2%: 82.7%: 77.5%.
With the CAMI statements, mental hospitals were agreed to be outdated by 26% of consultants, 13% of trainees and 19% of the others. Sixty-seven per cent of consultants (24% no opinion), 80% of trainees (14%) and 75% of the others (23%) do not agree that a woman (or man) would be foolish to marry someone who has suffered from mental illness. Consultants (63%) are more likely (compared with 55% and 49%) to agree that most women formerly in a mental hospital can be trusted as babysitters. Ratings for the ONS survey were as follows: deemed unpredictable: 41%: 32%: 41%; hard to talk to: 38%: 29%: 32%; feel the way we do at times: 42%: 50%: 39%.
Years since qualification
There were many differences between the different groups. Those with 10-20 years since qualification favoured causation as primarily biological more often than as a balance of biological and social (49.9%: 49.8%), compared with less than 10 (43.6%: 55.9%) and over 20 (43.6%: 56.1%). Gene therapy will not eventually eradicate schizophrenia: <10 years 76.2%: 10-20 years 72.9%: > 20 years 55.7%. Concerning avoiding discussion of content of delusions and hallucinations, those qualifying over 20 years ago were more likely to agree (2.5%: 3.5%: 9.9%). Misdiagnosis in Black people was viewed to occur more commonly in the >20 years group: 43.8%: 43.9%: 53.2% disagree. Prescription above BNF limits was thought to be more common by those more recently qualified: 50.4%: 43.7%: 42.2% but there was no difference with polypharmacy. Those qualified longest are less likely to routinely prescribe antiparkinsonian drugs with antipsychotics: 69%: 67.3%: 53.9%. Those qualified longer were less likely to strongly agree about giving diagnosis to patients: 60.8%: 57.8%: 45.2%, although overall they agreed with doing so almost as often. Longer qualification correlated positively with a trend towards giving more information to patients and carers. Enjoyment of working with patients with schizophrenia was sustained throughout all groups.
For the CAMI statements, mental hospitals are agreed to be outdated by 12.5% of recently qualified, 18.5% of 10-20 years and 30.2% with over 20 years (disagreement in 81.6%: 74.1% and 70.3%). Eighty per cent of 0-10 years, 73% 10-20 years and 65% longer qualified do not agree that a woman (or man) would be foolish to marry someone who has suffered from mental illness with no opinion expressed in 16%: 20% and 25%.
With the ONS survey, the only difference in ratings was as follows: unpredictable: 31%: 37%: 44% (neither one nor other - 47%: 39% 34%), i.e. patients considered less unpredictable with time since qualification of the psychiatrist.
Comparison of psychiatrists with the general population
Community attitudes to mental illness questionnaire
Most of the population (66%) disagrees with the statement that one of the main causes of mental illness is a lack of self-discipline and willpower, but psychiatrists express this more strongly (96.7%). There were similar proportions (92% in both) agreeing overall that virtually anyone can become mentally ill, but the general population held this view more strongly (73%: 50.3%).
Eighty per cent of psychiatrists disagree with the statement that there is something about people with mental illness that makes it easy to tell them from normal people compared with 64% of the general population. They are also more inclined to believe that less emphasis should be placed on protecting the public from people with mental illness (59.2%: 34%), although substantial numbers in both groups disagree with this (27.3%: 35%). Both groups tend to agree that people with a mental illness are far less of a danger than most people suppose (95.1%: 64%) and that residents have nothing to fear from people coming into their neighbourhood to obtain mental health services (71.4%: 61%). Similarly, agreement that no one has the right to exclude people with mental illness from a neighbourhood is greater with psychiatrists (90.5%: 71%).
Most women who were once patients in a mental hospital can be trusted as babysitters according to 60% of psychiatrists, but only 19% of the general population. Both groups do not generally agree that a woman would be foolish to marry a man who has suffered from a mental illness even though he seems fully recovered (70.9% psychiatrists: 64% general population). However, this is held more strongly by the general population (18%: 42%). The general population are much more likely to agree (40%: 22%) and less likely to disagree (34%: 69%) that mental hospitals are outdated than psychiatrists.
Stigma campaign ONS survey
Psychiatrists were much less likely to think of someone with schizophrenia as being dangerous to others (5% compared with 66.3%), being hard to talk to (36.5%: 59%) and unpredictable (77%: 40.5%). They were more likely to believe that they were not to blame for their illness (97.3%: 67.5%), and would improve if given treatment (93%: 51%). There is a broader range of opinion about whether they feel the way we do (45%: 11.5%) compared with feeling different (31.8%: 57.9%). They were as likely to think that they cannot improve the way they feel (57.9%: 66.1% compared with 2.5%: 11.5% can pull themselves together). Psychiatrists were less likely than the general population to respond that a person with schizophrenia will never recover fully (22.9%: 51%), with 50.5% compared with 27% taking a neutral position and 26.6% thinking they would compared with 13% of the general population.
The response rate to the questionnaire was 43%, and so the generalisability of the results to all psychiatrists cannot be assumed. However, the findings are worthy of comment and of comparison with general population figures. Inevitably, they reflect psychiatrists’ own perception of their attitudes and may well be distorted in terms of what they perceive to be professionally desirable responses. These attitudes should also reflect their understanding of the evidence base where it exists.
The differences in responses between groups of psychiatrists by grade, case-load and years qualified (but not much by gender) might have relevance to training and clinical governance. More biological beliefs in causation are more common in consultants and those qualified between 10-20 years, but generally the sample slightly favours a balance between social and biological cause, to primarily biological causation. This reflects the continuing debate in the profession.
Misdiagnosis of schizophrenia in Black people is believed to be common, especially by those qualified over 20 years and those not in consultant or trainee posts, but not so much by those with higher caseloads of patients with schizophrenia. This may be surprising in view of research studies, which have suggested such misdiagnosis to be uncommon (Harrison et al, 1988), but accords with the views of many patient groups and some recent research (Hickling et al, 1999). It is possible that such studies using standardised instruments are seen by psychiatrists as not being typical of ‘ normal’ clinical practice where such misdiagnosis may be more common.
Discussion of delusions and hallucinations is deemed no longer to be avoided (except by a group of 10% of those qualifying more than 20 years ago) and is consistent with the increased evidence for psychological treatments of schizophrenia (Sensky et al, 2000). Consultants and those longer qualified or with higher case-loads of people with schizophrenia give increased information about work, leisure, finances, etc. to patients than others. Family work is generally agreed to be effective in schizophrenia, but there remain a minority who are unconvinced. Prescription of medication above BNF levels, and polypharmacy, is agreed to be too common, especially by trainees and those more recently qualified and still there are substantial numbers of psychiatrists, especially consultants, prescribing antiparkinsonian drugs routinely with new prescriptions of antipsychotics (although this question was asked in such a way as to suggest a dichotomous answer, which may have distorted the results). Each of these findings has implications for basic training and continuing professional development. The findings about antipsychotic and antiparkinsonian prescribing are particularly disappointing and need to be a focus for change, possibly through the development of clinical governance procedures over the next few years.
Mental hospitals are not considered outdated, though less so by consultants and those qualified longer. It is possible that ‘mental hospital’ has been interpreted as meaning acute psychiatric unit, wherever sited, which may explain the very positive view from the majority of respondents compared with the general population.
In terms of attitudes, psychiatrists do seem to generally hold non-stigmatising views in comparison with the general population, although as in any group there are exceptions. The emphasis on protection of people from those with mental illness is believed to be excessive by most, as they are considered ‘far less of a danger than most people suppose’. Most psychiatrists do not agree that a man or woman would be foolish to marry someone with mental illness but the longer qualified, the less likely is an opinion on this to be offered. Similarly, most believe that ‘most women who were once patients in a mental hospital can be trusted as babysitters’. In considering ‘someone with schizophrenia’, most psychiatrists thought that they were not likely to be dangerous to others, but might be hard to talk to and unpredictable. This latter view was less likely where caseload of people with schizophrenia was higher, suggesting that direct personal experience improves understanding. Finally, psychiatrists are more optimistic about recovery than the public.
In summary, the attitudes of psychiatrists towards people with mental illness, especially schizophrenia, are generally positive compared with those of the general population. Individually and collectively, they are wellplaced to take leading roles in combating stigmatisation. Some findings, e.g. that many psychiatrists think that Black people are commonly misdiagnosed, warrant further discussion and investigation. Some aspects of management, especially the use of antipsychotic medication, need to be addressed by individuals themselves, and through training and clinical governance. The samples of psychiatrists and the general population were not matched, for example for age and gender, although these factors did not appear to have a major influence on the results. Responses to questionnaires are based on individuals’ perceptions of their attitudes, and may not be reflected in how they are perceived by others or in their clinical practice.
Finally, and fortunately, an overwhelming majority of psychiatrists working with people with schizophrenia enjoy doing so. This survey highlights areas for improvement in training and personal development plans, and for reducing stigmatisation. Using such surveys in other groups, e.g. general practitioners, nurses, social workers and politicians, might improve the focus and impact of stigma campaigns.
The Mental Health Group at the University of Southampton was commissioned by the campaign to do the survey with funding from Sanofi Synthélabo pharmaceuticals. Acknowledgements to Rajnikant Mehta for statistical support and the Department of Health and the Changing Minds Steering Committee for use of their general population survey results.
- © 2004 Royal College of Psychiatrists