Attitudes of clinical psychologists towards clients with personality disorders

Abstract

Previous studies have found that a variety of mental health professionals hold negative attitudes towards clients diagnosed with a personality disorder. These negative attitudes may lead to clients receiving a lower quality of service. Specialist training has been found to improve attitudes towards personality disorders but no empirical studies in Australia have examined this among clinical psychologists. In this study, the attitudes of 81 clinical psychologists towards clients with personality disorders were examined. We were specifically interested in investigating the relationship between recency of specialist training and clinician's attitudes as well as the influence of percentage of personality disorder clients on the clinician's caseload. Results demonstrated that both recency of specialist training and percentage of clients seen were associated with more positive attitudes; however, a higher caseload of clients with personality disorders was the most important predictor of positive attitudes. The implication is that recent participation in specialist training for personality disorders appears to be valuable in improving clinician's attitudes but that more positive attitudes are associated with seeing a greater number of individuals with personality disorders.

Key words:

• attitudes
• clinical psychologists
• personality disorder
• training

Individuals with personality disorders use a high number of services, are at risk of suicide, and have a poor quality of life (e.g., Brassington & Kravitz, 2006; Hulbert & Thomas, 2007). Personality disorders are frequently seen in mental health services at an estimated rate of 20% as inpatients and 11% in the community (Swartz, Blazer, George, & Winfield, 1990). In addition, those diagnosed with a personality disorder are at increased risk for substance abuse, relationship problems, and unemployment (Linehan, 1993).

Research exploring attitudes to personality disorders has mostly been conducted with nurses and prison officers. There is a paucity of research specifically relating to clinical psychologists' attitudes towards patients with personality disorders despite this professional group typically being highly involved in treatment provision (Servais & Saunders, 2007). Hence knowledge to date regarding attitudes towards clients with personality disorders is based on studies which have involved other professional groups than clinical psychologists. Fraser and Gallop (1993) conducted a study with 17 nurses who were treating patients in a group format. Using the Colson et al.'s 1986 Hospital Treatment Rating Scale, the authors found that patients with a diagnosis of borderline personality disorder received significantly less empathy from nurses than patients with other diagnoses. Additionally, the nurses demonstrated more negative attitudes towards this group of patients.

In order to develop a training course for practitioners in Australia, a survey by Cleary, Siegfried, and Walter (2002) examined 229 mental health staff (including 15 psychologists) on experience, knowledge, and attitudes towards patients with borderline personality disorder. The majority of the sample (85%) had some contact with borderline personality disorder clients and 80% of respondents reported finding these clients ‘difficult’ compared with other clients. Cleary et al. reported that 95% of staff said that they would be willing to engage in further training to increase their skills and knowledge in working with clients who have this disorder, particularly as two thirds of staff thought their present way of managing borderline personality disorder clients was unsatisfactory. The importance of adequate specialised training for staff working with clients with personality disorders is highlighted by Deans and Meocevic (2006). Surveys were completed by 65 nurses working in an Australian psychiatric inpatient borderline personality disorder unit. Over one third of participants perceived their clients as ‘manipulative, emotional blackmailers, nuisances and time‐wasters’ (p. 47), and that their knowledge in caring for this population was inadequate. However, the measure used was not tested for reliability or validity.

Clinician's attitudes arise from a complex mix of factors including clinical training, their own values and biases, and ethnicity of clients and clinicians (Mikton & Grounds, 2007). Indeed, Bowers et al. (2003) have highlighted this complexity, suggesting that there are many factors embedded in the shaping of attitudes towards patients diagnosed with a personality disorder.

The Attitude to Personality Disorder Questionnaire (APDQ) was originally developed in a survey of a high security setting (Bowers, McFarlane, Kiyimba, Clark, & Alexander, 2000). The attitudes of 651 nurses working in three high security prisons in the UK towards clients with a personality disorder were examined. Clinicians who had a more positive attitude were female, worked within a specialised personality disorder unit, held higher qualifications, considered they were adequately trained, and were younger than their counterparts. Following the survey, further refinements to the scale were added including moral commitment, beliefs, knowledge, emotional reactions, and relationship factors, and these correlated with positive attitudes towards personality disorders (Bowers & Allen, 2006). The APDQ scale consists of five factors (1) enjoyment (positive feelings towards patients), (2) security (feelings of safety and lack of anxiety), (3) acceptance (absence of anger, irritation, and alienation from patients), (4) purpose (a sense of meaningfulness in working with patients), and (5) enthusiasm (a sense of energy); however, a total score is also used to indicate an overall level of attitude towards clients with personality disorders.

Bowers et al. (2003) examined attitudes in another study with 73 prison officers working with detainees in a ‘Dangerous and Severe Personality Disorder’ unit in the UK, where the most prevalent diagnosis was antisocial personality disorder (45%). Attitudes to personality disorders were collected at baseline, 8 and 16 months using the APDQ (Bowers & Allen, 2006), Staff Attitude to Personality Disorder Follow‐up Interview (Bowers et al., 2005), and the Neuroticism, Extraversion and Openness to Experience Personality Inventory‐Revised (NEO‐PI‐R; Costa & McCrae, 1992). A range of variables including types of treatments the prisoners received, knowing more about prisoners' backgrounds, adequate supervision, and a cohesive working environment were associated with positive attitudes. Negative attitudes as indicated by the total score on the APDQ were associated with perceived prisoner behaviours such as ‘manipulation’, ‘attention seeking’, and self‐harm in addition to the structure of the unit and lack of training. However, the APDQ was only moderately effective in predicting attitudes among staff working with personality disorders over time. A follow‐up study in the same unit using semi‐structured interviews examined 66 prison officers' attitudes towards working with detainees with a personality disorder (Bowers et al., 2005). The primary factors contributing towards negative attitudes were ‘manipulative’ and aggressive behaviours by inmates, self‐harm, and prisoners that were perceived as ‘attention seeking’. Positive attitudes were related to staff having received training about personality disorders in order to increase knowledge and self‐efficacy when working with inmates, in addition to adequate supervision.

Attitudes are open to change (Ajzen, 2005), and this is demonstrated in an Australian study by Krawitz (2004) who investigated change in attitude towards borderline personality disorder clients following a specialised training course. A variety of 418 mental health professionals, including psychologists (14%), reported that the training workshop resulted in increased positive attitudes which were maintained 6 months after training; however, this survey was developed for the study and has no evidence of reliability and validity. Similarly, another study of the effects of a 20‐hr dialectical behaviour therapy training programme delivered to nurses, psychiatrists, and social workers in rural Western Australia (WA) via videoconferencing found a significant increase in knowledge and positive attitudes and decrease in burnout (Egan & Rees, 2003). Again, the questionnaire used was developed for the study and had no evidence of reliability and validity, so results of these two studies are difficult to generalize. Another example of how attitudes can change was supported by research with 94 registered nurses following a 2‐day dialectical behaviour therapy workshop (Hazelton, Rossiter, & Milner, 2006). Results showed an increase in positive attitude towards clients with borderline personality disorder, higher levels of optimism, and more understanding about clients with borderline personality disorder that was maintained 6 months after training. Webb and McMurran (2007) also found training was important in changing attitudes to working with people with personality disorders in a study of 117 community mental health nurses using the APDQ (Bowers & Allen, 2006). The researchers compared attitudes across three groups: prison officers, nurses who worked in a high security setting, and nurses in a specialist personality disorder unit. The results indicated that nurses working in the community were less accepting of personality disorder clients and felt less secure in their work setting compared with the staff working in a personality disorder unit. Those who volunteered to take part in a training workshop were found to report higher levels of enjoyment, more positive attitudes as reflected in higher scores on the total scale and subscales of the APDQ.

In summary, it appears that current experience working with clients who have a diagnosis of a personality disorder may be an important variable in explaining more positive attitudes. Crawford, Adedeji, Price, and Rutter (2010) conducted in‐depth interviews with community mental health workers in the UK who work exclusively with individuals with personality disorders. Contrary to expectations, they found low levels of burnout and positive experiences and attitudes being expressed by the staff. This suggests that efforts to understand clinicians' attitudes to clients with personality disorders must take into account the amount of exposure clinicians have to this client group.

The aim of the current study is to explore clinical psychologists' attitudes to working with patients with a personality disorder in Australia. Clinicians' attitudes are an important factor that contribute and influence the level of service that a client receives (Krawitz & Watson, 1999). Therefore, this research is important as attitudes may impact on treatment (Deans & Meocevic, 2006). To our knowledge, there has been no research to date to specifically examine clinical psychologists' attitudes to working with clients with a personality disorder.

The nature of the study is exploratory as previous research has measured attitudes in other professions, such as nurses and prison officers. Based on the results from other professional groups (Bowers & Allen, 2006; Bowers et al., 2000, 2005; Webb & McMurran, 2007), it is predicted that recency of specialist training in personality disorders will be significantly and positively correlated with positive attitudes. Finally, it is hypothesized on the basis of previous research (Crawford et al., 2010) that the percentage of clients currently seen with personality disorders will be significantly and positively correlated with more positive attitudes.

Method

Research design

In order to address the research questions, a cross‐sectional, correlational design was chosen.

Participants

Clinical psychologists were invited to take part in the Australian Survey of Practicing Clinical Psychologists during a 3‐month period. Participants were recruited via advertisement with professional bodies including the Psychologists Board of WA, The Australian Association for Cognitive and Behaviour Therapy (AACBT), and the Australian Psychological Society (APS). There were 81 surveys completed via an online survey. There were 69 females (85%) and 12 males (15%). Power calculation using G*Power 3.0.10 indicated a necessary sample size of 85 to detect a medium effect.

Measures

Demographic questionnaire

Questions included standard demographic information. In addition, several questions were asked regarding training and practice. Participants were asked to state their age in age ranges (20–30 years; 31–40 years; 41–50 years; 51–60 years, over 60 years) and their number of years clinical experience (1–5 years; 6–10 years; 11–15 years; 16–20 years; more than 20 years). Participants were asked the following questions regarding work with clients with personality disorders: ‘What percentage of clients do you have that are diagnosed with a personality disorder?’ To gain information on how recently the participant had engaged in specialist training for personality disorders, the following question was asked: ‘When was the last time you attended specialist professional development for treating personality disorders?’ (never; in the last 3 months; 3–6 months ago; 6–12 months ago; more than 12 months ago). These responses were then coded as: 0 = never, 1 = >12 months ago, 2 = 6–12 months ago, 3 = 3–6 months ago, 4 = <3 months ago. As such, higher scores were indicative of more recent training.

APDQ (Bowers et al., 2000)

The 35‐item APDQ contains items assessing the attitudes of clinicians when treating clients diagnosed with a personality disorder. The items include positive and negative attitudes, for example a positive item is ‘I feel warm and caring toward PD patients’ and a negative item ‘I feel intolerant. I have difficulty tolerating PD patient's behaviour’. Negative items are reverse scored. The items were rated scale on a 6‐point Likert‐scale ranging from 1 = never to 6 = always. The scale has five factors comprising of (1) enjoyment, (2) security, (3) acceptance, (4) purpose, and (5) enthusiasm (Bowers & Allen, 2006). The total score reflects the global level of attitude towards clients with personality disorders, and the higher the score, the more positive attitudes are towards clients with personality disorders. The total scale score of the APDQ has been used in other research to indicate a global level of attitude towards clients with personality disorders (e.g., Webb & McMurran, 2007). Test–retest reliability of the scale is acceptable ranging from 0.72 to 0.85 for subscale scores. The APDQ has shown good validity by demonstrating predictions relating to general health, performance at work, burnout, and perception of staff at management level (Bowers, 2002; Bowers et al., 2003).

Procedure

The research was approved by the Curtin University Human Research ethics committee. The survey was distributed to registered clinical psychologists in Australia via the psychologists board of WA, the APS, and the AACBT. An information sheet was included explaining the purpose of the study and providing the researcher's contact details. In addition, the participants were informed that the survey was confidential. The participants were rewarded for their participation by inviting them to take part in an optional online raffle.

Analysis

Bivariate correlations were used to test the predicted relationships between variables along with a hierarchical regression analysis to examine how much variance in attitudes the hypothesized variables could account for.

Results

A total of 13 cases were excluded from analysis as the APDQ was not completed. This resulted in a final sample of 81 (94–13) with complete data for the APDQ and ‘percentage of clients seen with a personality disorder’ variables.

The majority of respondents were in the 31–40 year age bracket (37%), with 20% aged 20–30 years, 26% aged 41–50 years, 10% aged 51–60 years, and 7% aged >60 years. In total, 81% of the sample was over the age of 31 years. Regarding workplace, 95% of participants worked in the public sector, and 89% in private practice. Of these, 84% worked in both the public sector and private practice, 12% in public sector work only, and 5% in private practice work only. Level of experience as a clinical psychologist varied, with 39% having 1–5 years of experience, 29% 6–10 years' experience, 14% 11–15 years' experience, 5% 16–20 years' experience, and 12% more than 20 years' experience. The mean percentage of clients seen with a personality disorder was 20%, with a range of 0–90%.

The mean score on the APDQ of the 81 participants who completed it was 20.7 (standard deviation = 4.34). The total APDQ scale had excellent internal consistency in the current study (Cronbach's alpha = 0.93), consistent with Bowers and Allen (2006) (alpha = 0.94). The majority of participants (94%) had engaged in specialist training for personality disorders, with only 6% having never engaged in specialist training. Of those who had completed training, 48.1% had engaged in training within the last 3 months, 24.7% had engaged in training in the last 3–6 months, 12.3% had engaged in training in the last 6–12 months, and 8.6% had engaged in training more than 12 months prior.

As seen in Table 1, there was a significant positive relationship between APDQ scores and percentage of clients seen. There was also a significant positive relationship between the APDQ scores and recency of training (i.e., the more recently the clinician had engaged in training, the more positive their attitude was towards personality disorders).

Table 1. Correlation matrix for demographic variables and APDQ

	APDQ	Gender	Age	Training	% clients
APDQ	–	−.07	.10	.26*	.37**
Gender		–	−.16	−.02	−.00
Age			–	.04	.06
Training				–	.20
% clients					–

Note. ‘Training’ refers to ‘Recency of specialist training’. ‘% clients’ refers to the question ‘What percentage of clients do you see diagnosed with a personality disorder'?

*p < .5, **p < .01, (two‐tailed).

To determine whether attitudes would be predicted by the variables of gender, age, percentage of clients seen diagnosed with a personality disorder, and recency of specialist training, a hierarchal multiple regression analysis was conducted. The order of entry of variables was based on our prediction that recency of specialist training would explain more variance in attitudes than simply the percentage of clients seen with a personality disorder. Data were first checked to ensure that the assumptions of normality, linearity, and homoscedasticity of residuals were met. As none of these assumptions were violated, a hierarchical regression was conducted. On step 1 of the model, age and gender accounted for a non‐significant 1.5% of variance in APDQ scores, R² = .015, F(2, 78) = .603, p = .550. On step 2, percentage of clients seen with a personality disorder was added and accounted for an additional 13.9% of variance in APDQ scores, ΔR² = .139, ΔF(1, 77) = 12.67, p = .001. On step 3, the recency of specialist training variable was added and accounted for a further non‐significant 3.7% of the variance in APDQ scores, ΔR² = .037, ΔF(1, 76) = 3.43, p = .068. Using Cohen's (1988) conventions, a combined effect of this magnitude can be considered a medium effect (f² = .23). Unstandardised (B) and standardised (β) regression coefficients and squared semi‐partial correlations (sr²) for each predictor on each step of the hierarchical regression are reported in Table 2.

Table 2. Unstandardised (B) and standardised (β) regression coefficients, and squared semi‐partial correlations (sr²) for each predictor variable on each step of hierarchical regression predicting APDQ (N = 81)

Variable	B	β	sr^2
Step 1
Gender	−.09	−.06	.00
Age	.04	.09	.00
Step 2
Gender	−.09	−.06	.00
Age	.03	.07	.00
% Clients	.00**	.37	.13
Step 3
Gender	−.08	−.06	.00
Age	.03	.07	.00
% Clients	.00*	.33	.10
Specialist Training	.09	.19	.03

*p < .01, **p < .001.

Discussion

This study sought to investigate the relationship between recency of specialist training in personality disorders and clinician's attitudes towards clients diagnosed with personality disorders. Interestingly, while the recency of participation in specialist training was significantly associated with clinician's attitudes, the most significant unique predictor of clinician's attitudes was actually the percentage of current clients on their caseload diagnosed with personality disorders. This suggests that both variables are related to clinicians attitudes but percentage of clients currently seen who have personality disorders is the most powerful predictor of attitudes towards this client group.

The finding that a higher caseload of clients with personality disorders was associated with more positive attitudes could be due to these psychologists already having an existing positive attitude to this client group. The results are consistent with previous research (Bowers & Allen, 2006; Webb & McMurran, 2007) where it has been demonstrated (also using total scores on the APDQ) that professionals who choose to engage in specialist training and work in a setting with a high percentage of clients diagnosed with a personality disorder had a more positive attitude. However, it is also possible that working with this group of clients may act to dispel negative attitudes and preconceptions about the behaviour of this group of clients. Having actual experience with clients with this diagnosis may enable a more balanced view than otherwise possible.

It was encouraging to discover that the majority of the sample had participated in specialist training. The prediction that there would be a more positive attitude among clinical psychologists who had participated in more recent specialist training for personality disorders was supported, with training 3 months ago or less being correlated with more positive attitudes. This finding is not surprising as those who have engaged in recent training may still recall the information and as a result may feel more knowledgeable and empowered about working with clients with a personality disorder.

The total APDQ mean score was similar to previous research with nurses, and prison officers (Bowers & Allen, 2006), suggesting that professional training background does not specifically impact on attitudes. However, the significant correlation between completion of recent training and higher scores on the APDQ suggest that training is associated with attitudes, and is consistent with other research that has found that improved attitudes to personality disorders are associated with completion of specialist training (Bowers et al., 2003, 2005; Egan & Rees, 2003; Hazelton et al., 2006). Positive attitudes may lead to better job performance, a sense of personal well‐being, a better therapeutic relationship, lower rates of burnout, and fewer sick leave days (Bowers et al., 2003). According to Servais and Saunders (2007), training should aim to reduce bias about personality disorders. However, this study does not provide causal evidence that completion of training lead to more positive attitudes; it may be the case that those clinical psychologists with more positive attitudes choose to engage in more recent training.

The study had several limitations that must be acknowledged. The individuals who responded were already likely to be more motivated and thus potentially have a more positive attitude, and participants may have answered the questionnaire in a socially desirable manner in order to appear more altruistic, positive, and non‐judgmental. These factors may have resulted in an overestimation of positive attitudes; therefore, the results may reflect a more favourable attitude than actually exists. However, as already noted, the mean score on the APDQ was very similar to other studies and so this suggests that the sample was representative.

The study included only a small proportion of clinicians who had never participated in specialist training for personality disorders. Furthermore, the questions asked about recency of training and not frequency, and this variable may have been important to assess. Due to this study being embedded in a larger survey, a number of other demographic variables were not included but would have been helpful in further interpreting the results. For example, years of experience working as a clinical psychologist and also years of experience working specifically with clients with personality disorders would have been useful additional information. The APDQ only assesses attitudes to personality disorders in general, and did not focus on specific types of personality disorder such as borderline personality disorder. Finally, the power analysis indicated that a sample size of 85 was required to detect a medium effect size; therefore, the sample size of 81 participants was slightly short of this.

Future research should examine the attitudes of a larger sample clinical psychologists who have never participated in specialist training, in a balanced ratio of males and females. Furthermore, attitudes to a specific personality disorder such as narcissistic personality disorder could be explored, as most research to date has been with patients diagnosed with borderline personality disorder (Servais & Saunders, 2007). It would also be interesting to explore personality characteristics of clinical psychologists using the NEO‐PI‐R (Costa & McCrae, 1992) personality traits as Linehan (1993) cites that the personality traits of clinicians influence attitudes and treatment outcome for clients. Servais and Saunders (2007) have stated that there is a paucity of research pertaining to clinical psychologists' personality traits and attitudes to personality disorders; therefore, this could be explored in future research. Finally, future research could explore the link between attitudes and behaviour, and whether changes in attitudes change behaviour when working with clients with personality disorders. It would be helpful for future research to determine how exactly to influence change in attitudes. Future research could seek to determine whether an increase in knowledge may be responsible for more positive attitudes after training. It would also be useful to determine if there are other variables that explain attitudes. For example, the theory of planned behaviour (Ajzen, 2005) could be used to guide a more comprehensive investigation of attitudes and how to improve them. Ultimately, it would be beneficial for future research to determine how to influence attitudes precisely and if holding a more positive attitude towards working with clients with personality disorders may be associated with less clinician burnout and more successful treatment.