Abstract
Aims:
Studies investigating mental health professionals’ attitudes towards people with mental illness are scarce and there is a lack of comparative studies including both patients’ and mental health professionals’ attitudes. The aim of the present study was to investigate mental health staff’s attitudes towards people with mental illness and compare these with the attitudes of patients in contact with mental health services. A further aim was to relate staff attitudes to demographic and work characteristics.
Methods:
A cross-sectional study was performed including 140 staff and 141 patients. The study included a random sample of outpatients in contact with mental health services in the southern part of Sweden and staff working in these services. Attitudes were investigated using a questionnaire covering beliefs of devaluation and discrimination of people with a mental illness.
Results:
Negative attitudes were prevalent among staff. Most negative attitudes concerned whether an employer would accept an application for work, willingness to date a person who had been hospitalized, and hiring a patient to take care of children. Staff treating patients with a psychosis or working in inpatient settings had the most negative attitudes. Patient attitudes were overall similar to staff attitudes and there were significant differences in only three out of 12 dimensions. Patients’ most negative attitudes were in the same area as the staff’s.
Conclusions:
This study points to the suggestion that mental health care staff may hold negative attitudes and beliefs about people with mental illness with tentative implications for treatment of the patient and development and implementation of evidence-based services. Since patients and staff in most respects share these beliefs, it is essential to develop interventions that have an impact on both patients and staff, enabling a more recovery-oriented staff–patient relationship.
Introduction
Negative attitudes towards and discrimination against people with mental illness are still highly prevalent in the general population (Angermeyer & Dietrich, 2006). This is viewed as one of the most serious obstacles to successful treatment, rehabilitation and inclusion of people with mental illnesses in society. Unemployment, loss of income (Sharac, McCrone, Clement, & Thornicroft, 2010), not seeking care or delayed care (Andrews, Henderson, & Hall, 2001), a limited social network (Livingston & Boyd, 2010; Thornicroft et al., 2009), impaired self-esteem (Ilic et al., 2011), isolation and loneliness are for many people with a mental illness the consequences of stigma and discrimination. Stigma also affects disease progression and recovery (Perlick et al., 2001).
Studies on public attitudes towards people with mental illness show that the public’s perceptions have not changed significantly in recent decades (Angermeyer & Dietrich, 2006; Sorenson, 1994). Negative expectations, fear and a reluctance to assist have been demonstrated (Wahl, 1999). Public perceptions of persons with mental illness as dangerous and unpredictable and difficult to talk to are also common (Crisp, Gelder, Rix, Meltzer, & Rowlands, 2000). The impact of national anti-stigma campaigns has so far not been thoroughly evaluated and the results of these campaigns are viewed as equivocal, and it is largely unknown to what extent attitudes towards people with a mental illness are related to actual behaviour (Angermeyer & Dietrich, 2006). The internalization of negative stereotypes about mental illness develops early in life, which has a special relevance for a person who later in life develops a serious mental illness, and may lead to self-stigma, which could result in refraining from applying for work and avoiding contact with mental health care and social contacts (Rüsch, Angermeyer, & Corrigan, 2005).
The major part of research on attitudes towards mental illness has focused on the general public’s attitudes and beliefs, and the views of professionals in contact with people with mental disorders such as staff in mental health care have only recently been highlighted. However, there is evidence that people with mental illness feel patronized, humiliated and punished in contact with services and that patients point out mental health staff as one of the groups that are the most stigmatizing (Thornicroft, Rosem, & Kassam, 2007). Studies on mental health staff’s attitudes have mainly focused on the prevalence of stereotypes and the desire for social distance from people with mental illness. A recent review of these studies revealed that a majority constituted comparisons between staff and the general public’s perceptions (Schulze, 2007). One of the main findings from this review is that results from three-quarters of the studies (n = 16) show that beliefs from mental health providers do not differ from the general public or are more negative. This is in contradiction with a hypothesis that professionals’ knowledge of mental illness and regular contact with people with a mental illness would result in more favourable attitudes. However, it seems that this does not act as a protective factor when it comes to the prevalence of stereotypes regarding people with a mental illness or a greater willingness to interact with these people (Lauber, Nordt, & Rossler, 2006; Nordt, Rossler, & Lauber, 2006).
Schulze’s (2007) review did not, however, reveal any studies comparing staff and patients’ attitudes towards people with mental illness. The aim of the present study was therefore to investigate beliefs of devaluation and discrimination in a sample of mental health professionals and to compare attitudes in this respect with a sample of patients in contact with mental health outpatient services. A further aim was to relate staff attitudes to a number of demographic and working situation characteristics of the staff.
Methods and subjects
Design and subjects
A cross-sectional study was performed in order to investigate and compare attitudes towards people with mental illness among staff and patients in contact with mental health services. In total 280 staff in 14 units at selected services in the south-west of Sweden were invited to participate in the study and 140 completed the questionnaire. The response rate was 50%. Background characteristics of the staff are presented in Table 1. Most participants were female (71%), nurses or assistant nurses (81.9%) and working in outpatient care (58.4%). Most of the respondents were working with patients suffering from non-psychotic disorders (73.6%). The length of employment in psychiatric care ranged from less than one year to 44 years. Due to ethical considerations, no information was gathered concerning patients or staff who declined to participate, thus no analysis with regard to the representativity of the participants was accomplished.
Background characteristics of mental health care staff (N = 140).
The study included outpatients in contact with mental health services at eight of the units from which staff were invited to participate. A random selection of 20% out of a total population of 1,195 patients was invited to take part in the study by their key worker. Inclusion criteria were experience of outpatient care, understanding of and ability to read the Swedish language and being at least 18 years of age. In total, 239 patients were randomly selected for participation, of whom 37 declined, resulting in an external dropout rate of 15.5%. Sixty-one patients who had agreed to participate did not complete the questionnaires, thus the internal dropout was 30.2%. The final sample consisted of 141 patients who agreed to participate: a response rate of 59%. Most patients were women and lived alone in their own apartments. In terms of diagnosis, 50.9% of the subjects had an affective disorder, 20.2% had a diagnosis of schizophrenia, 11.4% had an eating disorder and 17.5% had other diagnoses. With regard to work, 31.4% were engaged in some form of competitive employment while 35.2% were on sick leave or receiving an old age pension (Table 2).
Sociodemographic and clinical characteristics of the patients (N = 141).
The present study was part of a study with a focus on health and health promotion, including a number of instruments not further explored in this paper (Jormfeldt, 2007; Svedberg, 2007). The study was reviewed by the Regional Research Ethics Committee of Lund University, Sweden, and permission was obtained from the head of psychiatric primary care in the county where the units included in the study were situated. No conflicts of interest concerning this study have been detected.
Procedure
For the investigation among staff, a letter with written information about the study, a schedule for providing informed consent, the questionnaire and a prepaid response envelope was sent to the invited participants’ work address. For the patient investigation, a key person among the staff at the eight units was appointed to perform the selection of patients. Participants were chosen randomly from the case register at the unit by the appointed key person. The key person also ensured that each participant received identical information and invitation to participate in the study, based on a detailed written short manual. All participants were provided with oral and written information about the purpose and structure of the study, after which they gave their informed consent in writing. The questionnaire was distributed to the patient by the key person, along with a prepaid return envelope.
Instruments
Attitudes in terms of perceptions of devaluation and discrimination were investigated by using a questionnaire developed by Links (1987), the 12-item Perceived Devaluation–Discrimination Questionnaire containing 12 statements about negative beliefs among the general population towards people with mental illness. Response categories are: 4 = ‘strongly agree’, 3 = ‘agree’, 2 = ‘disagree’, 1 = ‘strongly disagree’. The scoring for positively worded items are reversed in creating a scale score. After reversing, items are summed to an overall score where a higher score indicates more negative beliefs (range 12–48). Reliability and validity of the Swedish version of the instrument has been tested with satisfactory results (Björkman, Svensson, & Lundberg, 2007). Reliability for the present study was .91 for staff and .88 for patients (Cronbach’s α).
Statistics
Responses to the attitude questionnaire were dichotomized into the categories ‘agree’ (score 3–4) and ‘do not agree’ (score 1–2). Age of staff was dichotomized using the median age (47) as the cut-off point, and working years was dichotomized using median years of working experience (16). Analyses of differences between staff and patients and between subcategories of staff on an item level were made using the χ 2 test. Analyses of differences in overall score were made using the student’s t test. Alpha level was set at p < .05. The statistical software used was SPSS version 15.
Results
Frequencies of staff partially or totally agreeing to the statements regarding beliefs about people with mental illness, and comparisons with patient ratings, are presented in Table 3. In general, negative attitudes towards people with a mental illness were predominant among staff, with a majority having a negative attitude in six out of the 12 items and more than one-third in all 12 items. The most negative attitudes concerned whether an employer would pass over an application for work in favour of another applicant without mental illness (75.6%), whether most young women would be reluctant to date a man who had been hospitalized for a mental illness (67.4%) or would hire a former mental patient to take care of their children (66.4%). Patients of these services in most instances shared these negative attitudes and beliefs with the staff (Table 3). In three items, staff have a significantly more positive attitude: entering a mental hospital is a sign of failure (49.6% vs 61.7%, p = .043), whether people think less of a person who has been in a mental hospital (44.9% vs 62.1%, p = .004) and that opinions of people who had been hospitalized would be taken less seriously (49.6% vs 61.4%, p = .048). There was no significant difference between staff and patients in the sum score of the scale.
Frequencies of patients and staff partially or totally agreeing to 12 statements of beliefs about people with a mental illness (staff N = 140, patients N = 141).
χ 2 test.
Comparisons of attitudes among subcategories of staff are shown in Table 4. Staff below the median age showed significantly more negative attitudes concerning the acceptance of a former patient as a teacher of young children (p = .029) and if people would think less of a person who has been in a hospital (p = .02). However, there was no significant difference in the sum score of the scale. Staff primarily caring for people with a psychosis held more negative attitudes in three instances: if a former mental patient is as trustworthy as other people (p = .011), if employers would hire a former mental patient (p = .008) and if opinions of a person having been in a mental hospital would be taken less seriously (p = .046). There was also a significant difference in the sum score of the scale, staff caring for psychosis patients showing more negative attitudes (31.8 vs 29.7, p = .045).
Differences in beliefs about people with mental illness with regard to staff characteristics (frequencies of staff totally or partially agreeing are presented).
χ 2 test.
The most frequent differences were shown between staff working in outpatient and inpatient mental health care. Staff in inpatient services held significantly more negative attitudes in seven out of the 12 attitudes rated (Table 4). The largest differences concerned if people would accept a fully recovered mental patient as a teacher of young children (p = .001) and if people in the community would treat a former mental patient just as they would treat anyone else (p = .001). There was also a significant difference in the overall score of the scale (29.1 vs 31.8, p = .004). Analyses of differences with regard to staff’s gender, years of working in mental health care and nursing staff vs other staff showed no significant differences in attitudes.
Discussion
There are several limitations of this study. The final response rate of 59% of the random sample selected might have induced a bias in the results. However, it was not possible for ethical reasons to investigate if responders were representative of the intended sample, and therefore comparisons could not be made with dropouts in terms of social and clinical background characteristics. Furthermore, we used a diagnostically rather heterogeneous sample as a result of the random procedure of all patients in contact with the services. The main reason for using this procedure was that we were primarily interested in how attitudes appeared in a sample representative of these services, and not in specific diagnostic differences in these respects. The issue of social desirability is always at hand in attitude studies. However, we do not believe that this influenced the results in any major way since all measures were self-report questionnaires and all participants were guaranteed confidentiality.
A main finding from the present study was that negative beliefs about people with mental illness are prevalent among mental health care staff. Although not very intensively studied, this is in line with earlier studies showing that negative beliefs or attitudes are widespread among mental health care staff, and similar to those of the general public (Schulze, 2007). Beliefs were also related to some staff and work setting characteristics. Younger staff held more negative beliefs, a finding supported by other studies (Jorm, Korten, Jacomb, Christensen, & Henderson, 1999). Work setting had an impact on beliefs insofar as staff mainly treating psychosis patients and staff working in inpatient settings held more negative beliefs compared to staff treating other diagnostic groups and staff working in outpatient settings, respectively. Most differences were found in comparisons between staff in inpatient and outpatient settings. This might be a reflection of the fact that these subgroups of staff to a greater extent have contacts with people with more severe, long-term and recurrent illness, which might induce attitudes where they think less of people with mental illness and view them as less trustworthy, and less capable of acquiring or maintaining a job. If such negative beliefs of possibilities for people with a severe mental illness to acquire a job are reflected in actual rehabilitation work and in treatment planning, this may reduce ambitions of recovery and work on behalf of the patient and instead induce pessimism and hopelessness in the individual. This may also on a service level prevent the implementation of evidence-based interventions in this field such as supported employment, where several reviews have shown that more than half of patients engaged in this intervention get a job in the open labour market (Bond, Drake, & Becker, 2008). Consequently, an implication of negative beliefs in staff may on a more general level be an important obstacle to the implementation of an evidence-based practice in various intervention domains.
This is a study of self-reported attitudes and a crucial question is whether and how these attitudes translate into actual behaviour on part of the patients and staff. The patients’ negative attitudes may be contextualized as internalized stigma or self-stigma. To result in a behavioural response, it requires that these attitudes seem relevant on a personal level, that the person has a group identification with those focused by these attitudes, and furthermore that these attitudes are perceived as legitimate. Only in this case will result be a self-discrimination process with its potential and social and psychological outcomes (Corrigan, 2004; Rüsch et al., 2009). A similar process is relevant for the translation of negative attitudes among staff into actual behaviour towards staff. These stereotypical attitudes need on an affective level to be viewed as accurate in order to evoke prejudices that may lay the ground for derogatory and discriminating behaviour towards the patients (Corrigan & Watson, 2002a, 2002b).
In general, these findings and similar findings from other studies are in contrast with the hypothesis that better knowledge of various aspects of mental illness and daily contact with people with mental illness would lead to more positive beliefs. However, it is apparent that better knowledge and frequent contacts do not always prohibit or protect against stigmatizing and stereotyped beliefs about people with mental illness (Lauber, Anthony, Ajdacic-Gross, & Rossler, 2004). Mental health professionals are also as citizens exposed to early socialization processes that may induce an internalization of stigmatizing attitudes and discriminatory behaviour. Mental health professionals are also exposed to what has been labelled ‘associative stigma’ or ‘stigma by proxy’, that is they are working in a discipline with low status and are less valued than staff working with other patient groups in the health care system. It is still unclear to what extent this situation may carry over in actual behaviour in their daily work with patients (Halter, 2008). It is, however, necessary to be aware that this phenomenon exists and may affect the well-being of patients. It has been suggested that interventions empowering patients may be a way of action to reduce perceptions of stigma and discrimination among persons with mental illness (Corrigan, 2004). In a similar way it may also be necessary to intervene among mental health care staff in order to empower them and provide a consciousness of how stigmatizing attitudes may impact on their work with patients.
To our knowledge this is the first study comparing beliefs and attitudes of mental health staff and people with mental illness in contact with mental health services. Our second main finding was that patients’ and staff’s beliefs about what most people think of those who have mental illness in most aspects did not differ. There were significant differences in three aspects, all relating to having been hospitalized, where staff held more positive views. These results confirm that anticipated stigma or self-stigma is widespread among people with a mental illness (Fung, Tsang, & Corrigan, 2008), with all the implications this may have for help-seeking behaviour, self-image and restrictions in social life (Rüsch et al., 2005; Rüsch, Corrigan, Todd, & Bodenhausen, 2010). These findings are also in correspondence with earlier studies showing that attitudes in these respects are similar to those of the general population (Link 1987; Link, Cullen, Struening, Shrout, & Dohrenwend, 1989).
Conclusion
In summary, this study points to the importance of an awareness that mental health care staff may hold negative attitudes and beliefs about people with mental illness, and that this may have implications for the treatment of the individual patient and for the development and implementation of evidence-based services. Furthermore, since patients and staff in most respects seem to share these beliefs, it is essential to develop interventions that have an impact on these beliefs in both patients and staff, enabling a more recovery-oriented staff–patient relationship.
