Discussion: Evaluating Research Questions, Hypotheses, And Quantitative Research Design

Changes in staff confidence and attributions for challenging behaviour after training in positive behavioural support within a forensic medium secure service

Bronwen Daviesa*, John Griffithsa, Kim Liddiarda, Kathy Loweb and Lauren Steada

aCaswell Clinic, Glanrhyd Hospital, Bridgend, UK; bLearning Disability Services, Glanrhyd Hospital, Bridgend, UK

(Received 9 October 2014; accepted 2 July 2015)

Positive behavioural support (PBS) is a non-aversive approach to preventing and managing challenging behaviours. Seventy-nine qualified and unquali- fied nursing, psychology and occupational therapy staff were trained in using PBS. To measure the effectiveness of the training, confidence in managing challenging behaviour and attributions for causality, control and stability were measured before and after the training. To measure confidence, an adapted version of the Confidence in Coping with Patient Aggression Instrument was used. Attributions were measured using the Challenging Behaviour Attributions Scale and the Causal Dimension Scale II. There was a significant increase in confidence after training. In addition, there were significant changes in attributions relating to causality and stability of challenging behaviour, particularly for qualified staff. The results suggest that confidence and attributions are affected positively by training in PBS within a medium secure forensic mental health setting.

Keywords: training; positive behavioural support; attributions; confidence; violence and aggression; antisocial behaviour

Introduction

Challenging behaviour within the NHS remains a well-recognised issue that brings into question factors such as causation, risk, intervention and associated outcomes (NHS, 2014). Managing challenging behaviour has been a demanding aspect of care provision within forensic services. Behaviour can be described as challenging when it is:

Of such an intensity, frequency or duration as to threaten the quality of life and or the physical safety of the individual or others and is likely to lead to responses that are restrictive, aversive or result in exclusion. (Royal College of Psychiatrists, 2007, p. 10)

*Corresponding author. Email: Bronwen.Davies2@wales.nhs.uk

© 2015 Taylor & Francis

The Journal of Forensic Psychiatry & Psychology, 2015

Vol. 26, No. 6, 847–861, http://dx.doi.org/10.1080/14789949.2015.1072574

 

 

Positive behavioural support (PBS) is a framework of assessment and positive interventions aimed at preventing and managing challenging behaviours (Department of Health, 2014). The research exploring the effectiveness of PBS in managing behaviours that challenge has largely occurred within learning disability populations in the United Kingdom and child and adolescent populations (schools) within the United States of America (e.g. Curtis, Van Horne, Robertson, & Karvonen, 2010; McClean et al., 2005). Being based on behavioural approaches, its utility is clearly much broader than the current contexts it is being employed in (Allen, James, Evans, Hawkins, & Jenkins, 2005). This article focuses on the implementation of PBS within a medium secure forensic mental health service in south Wales (see Griffiths & Wilcox, 2013).

Confidence with respect to challenging behaviour is a key issue in the effectiveness of carer support. Thackrey (1987) defines confidence in managing challenging behaviour as the ‘self-attributed ability, preparation, and comfort in safely and effectively intervening psychologically and physically with the aggressive service user for purposes of self-preservation and therapeutic inter- vention’ (p. 58). Increasing staff confidence in managing challenging beha- viour acts to increase levels of proactive and therapeutic intervention whilst ensuring the safety of both service users and staff alike (Martin & Daffern, 2006). When employing the ‘therapeutics for aggression’ training programme, Thackrey (1987) found that confidence in managing aggression not only increased after training but was maintained eighteen months post intervention. However, where confidence has been measured in forensic or mental health staff populations, the training provided has tended to focus on physical inter- ventions and legal considerations as opposed to positive behavioural approaches (e.g. Martin & Daffern, 2006; McGowan, Wynaden, Harding, Yassine, & Parker, 1999; Thackrey, 1987). Moreover, confidence was measured a number of months after training (McGowan et al., 1999) or measured more generally, rather than as a specific outcome measure of training (Martin & Daffern, 2006). The training administered within the Thackrey (1987) paper did include ‘principles of psychological assessment and intervention’; however, very little detail was given on what these involved and whether or not punitive approaches were advocated.

Data on the impact of PBS training on staff confidence in managing chal- lenging behaviour have been gathered from learning disability staff samples. Lowe et al. (2007) found significant increases in staff reported confidence in dealing with challenging behaviours after training. These gains were main- tained or further increased over time for both qualified and unqualified staff. Similar results were demonstrated by Tierney, Quinlan, and Hastings (2007), who showed increases in self-efficacy after training about understanding and responding to challenging behaviour within learning disability services; these improvements were maintained at three-month follow-up.

848 B. Davies et al.

 

 

Another method used in assessing the effectiveness of training in PBS is the measurement of attributions for causality, control and stability of challenging behaviour and the degree to which they change after training. Attribution theory was brought to the fore by the work of Weiner (1980) who recognised the impact of attributions on care-giving behaviour. He identified three key dimensions of ‘Locus’, ‘Stability’ and ‘Controllability’. Locus is the degree to which carers attribute challenging behaviour as being due to factors within the individual, such as personality or mental health, or factors external to the individual, for example being reprimanded, unsettled or noisy environment. Stability is the degree to which carers believe that behaviour is either stable or changeable over time. Controllability is the degree to which carers believe the behaviour is within the control of the individual or not, for example if someone is acutely psychotic, their behaviour may not be seen as controllable by them. Weiner found that help- ing behaviour was lowest when behaviour was viewed as stable, internal and controllable, engendering feelings of disgust and anger. In contrast, when beha- viour was viewed as external to person, changeable and uncontrollable, helping behaviour was elicited as well as positive affect, such as sympathy. Evidence supporting Weiner’s work is found in studies examining carers working in learn- ing disability, mental health, homeless and forensic services, and with individuals who self-harm (e.g. Forsyth, 2007; Leggett & Silvester, 2003; Markham & Trower, 2003; Meddings & Levey, 2000; Stanley & Standen, 2000); Urquart Law, Rosthill-Brooks, & Goodman, 2009.

In considering care provision for those viewed as dangerous by society, it is logical to assume that care staff will be influenced by societal views. These may impact on their attributions relating to challenging behaviours and the nature of their engagement with this client group. MacKinlay and Langdon (2009) studied sexual offenders with a learning disability and identified that challenging behaviour, and sexual offending was viewed as internal, stable and controllable by the service user, but less controllable than challenging beha- viour more generally. Barrowclough et al. (2001) studied staff attributions within a low secure service for people with severe mental health problems. They found that staff tended to view the behaviour of service users they were less positively disposed to as more controllable by them, and more stable, which, in turn, was related to a more critical and negative attitude. Quinsey and Cyr (1986) studied clinicians’ perceptions of the dangerousness and treatability of offenders and found a negative relationship between them. They found perceived dangerousness to be positively associated with ratings of responsibility, internality of cause, greater stability of cause and greater controllability of cause. Offenders with external causality were perceived as more treatable. Similarly, Reid and Millard (1995) demonstrated that when care staff, within a high secure hospital, believed there was a high degree of controllability and stability of service users behaviours, they were perceived as less treatable. Leggett and Silvester (2003) found that more aversive tech- niques, such as seclusion, were more likely to be used when staff perceived

The Journal of Forensic Psychiatry & Psychology 849

 

 

behaviours to be more controllable by service users. Sharrock, Day, Qazi, and Brewin (1990) studied care staff within a medium secure forensic service, they found attributions of challenging behaviours to unstable factors was associated with higher levels of optimism and increased helping behaviour. Overall, these results would suggest that the more knowledge staff have about the factors that can cause challenging behaviour, the more likely they are to respond in a positively.

A key element in PBS training is to educate people about the internal and external factors that contribute to challenging behaviour. Research has demon- strated that changes in attributions of causality and control after PBS training are variable. Lowe et al. (2007) found that, after the delivery of a ten-day train- ing package on PBS, changes occurred in causal attributions of challenging behaviour (as measured using the challenging behaviour attributions scale (CHABA)) immediately following the training sessions for both qualified and unqualified staff. However, these gains were short lived, with a return to base- line levels within the one-year follow-up. Similar results were found by Dowey, Toogood, Hastings, and Nash (2007) in evaluating a one-day training course on PBS. This study showed that training successfully increased staff’s use of accu- rate causal attributions for challenging behaviour suggesting that similar results can be reached within a shorter training time. Again, these studies have been undertaken within learning disabilities services where the PBS approach is more widely utilised. In contrast, McKenzie, Sharp, Paxton, and Murray (2002) and Tierney et al. (2007) found no significant changes in staff attributions for chal- lenging behaviours after training in PBS. A review of the studies concluded that training in challenging behaviour did have an effect on attributions, particularly when it involved discussions around the causes of challenging behaviour (Williams, Dagnan, Rodgers, & McDowell, 2012). This, they concluded, was a key factor in the training’s success and studies where this was not an integral part of training reported no significant changes in attributions (McDonnell et al., 2008).

Although the impact of PBS-based training on increasing confidence and modifying attributions for challenging behaviours has been shown in studies within learning disability services, albeit with some variability in success, it has not been validated within mental health secure services. This novel application of the PBS model within forensic mental health services in this study aimed to assess the potential benefits for staff both in terms of modifying causal attributions and increasing reported confidence in managing challenging behaviours after training in PBS. The hypotheses were:

(1) Confidence would increase after staff training in PBS. (2) Attributions for external causes of challenging behaviour would

increase and that there would be reductions in attributions of internal causality, stability and personal control.

850 B. Davies et al.

 

 

Method section

Participants

Participants were staff members currently employed at a medium secure service in the South Wales region. PBS training was implemented on the acute wards, and so staff from these areas were invited to attend the training. In total, 79 participants took part in the training, see Table 1 for information regarding the role and gender of participants.

Instruments

Staff members’ confidence in dealing with challenging behaviour was assessed using an adapted version of the Confidence in Coping with Patient Aggression Instrument (Thackrey, 1987). The adaptation of this instrument was undertaken by a clinical psychologist who was also responsible for writing and delivering the PBS training. It included questions related to staff confidence in the delivery of PBS, as opposed to physical interventions which the original ques- tionnaire was designed for, and changed the language from ‘aggression’ to ‘challenging behaviour’ to encompass the broader range of behaviours dis- played by service users. The scale was shorter than the original and comprised of eight questions (as opposed to ten) rated on a seven-point likert scale to assess participants’ perceived level of confidence in understanding and responding to challenging behaviour. The instrument has good face validity and shows good internal consistency with Cronbach’s alpha at .88. In the development of the original instrument, extensive piloting was done by Thackery (1987), testing the items on 236 professionals to ensure their validity. Further validity testing would need to be completed on this current version of the scale.

The CHABA (Hastings, 1997) assesses changes in causal attributions, which may interact with other variables to determine staff responses to chal- lenging behaviour. The scale comprised 33 items requiring participants to rate the determinants of challenging behaviour on a five-point rating scale ranging

Table 1. Information relating to participants.

N = 79 Qualified (N = 48) Unqualified (N = 31)

Male 11 24 Female 37 7

Nurse 33 27 Student nurse 2 Occupational therapist 5 4 Clinical psychologist 4 Assistant psychologist 4

The Journal of Forensic Psychiatry & Psychology 851

 

 

from ‘Very Unlikely’ with a score of −2, to ‘Very likely’ with a score of 2. This measure has acceptable levels of reliability with Cronbach’s alpha values between .65 and .87. The author identifies that there is no validity data avail- able for the CHABA due to lack of external validation criteria; however, this measure is widely used within learning disability research (e.g. Lowe et al., 2007; Tierney et al., 2007). The definitions of the variables measured within the CHABA are listed below:

Learned – overall learning – learned positive and negative Learned negative – avoid something – e.g. difficult or uninteresting tasks,

disliked person Learned positive – to gain something – e.g. attention/tangible item Biomedical – Physical illness/need (e.g. hunger or thirst), medication Emotional – Mood and emotion Physical environment – Noise, lack of space, crowded, too light Stimulation – Boredom, lack of activity or interaction The final measure employed was the causal dimension scale II (CDS)

(McAuley, Duncan, & Russell, 1992) to determine to what extent participants attributed challenging behaviour to four factors: locus of causality, stability, exter- nal control and personal control. Reliability coefficients range between .60 and .92 for the four factors. Confirmatory factor analysis provided evidence of construct validity and support for the four-factor model within this scale (McAuley et al., 1992). This measure has been widely used in attributional research in a number of contexts (e.g. Ball, 2013; Boisvert & Faust, 1999). The measure comprised 12 items with a scale ranging between 1 and 9 between two poles.

For example: This reflects an aspect of the service user 9 8 7 6 5 4 3 2 1

This reflects an aspect of the situation The definitions of the variables measured within the CDS are listed below:

Locus of causality

The degree to which the behaviour was caused by the service user or the envi- ronment/ others. A lower score reflects an environmental cause, and a higher score reflects the locus of causality being viewed as within the service user.

External control

The degree to which others have control over the individual’s behaviour. A lower score reflects less environmental/other persons control, and a higher reflects more control attributed to the environment or others.

Personal control

The degree to which the behaviour was controllable by the service user. A lower score reflects a view that the behaviour was less controllable by the

852 B. Davies et al.

 

 

service user; with a higher score, the behaviour is viewed as more controllable by the service user.

Stability

The degree to which the behaviour was seen as stable or changeable. A higher score indicates that the behaviour is viewed as less changeable and more permanent, whereas a lower score indicates the view that behaviours can change.

Procedure

The PBS training packages were developed by a Clinical Psychologist employed within the service. These were based on elements of the online PBS training developed by learning disability services within the same health board (ABMU Directorate of Learning Disabilities: BTEC e-learning qualifications in PBS. Learning@NHSWales). This same Psychologist was also responsible for delivering the training, with some assistance from a ward manager. The train- ing packages for the qualified and unqualified participants varied slightly because of the different roles and responsibilities that the respective staff mem- bers hold within the service The training package for the qualified staff was one full day in duration and covered basic teaching and education around PBS as well as practicing skills associated with PBS such as completing a func- tional analysis and identifying primary and secondary prevention strategies. The training for the unqualified staff members was half-a-day and included basic teaching and education around PBS alongside an introduction into antecedent, behaviour and consequence (ABC) charts.

The recruitment process was based on ward managers allocating staff to attend the training days offered; there was an expectation in the service for all staff to attend the training and regular dates were provided to managers via e-mail and ward managers meetings. At the beginning and end of the training, attendees were asked whether they would participate by completing the questionnaire. It was explained that the data would be used to evaluate the training and they gave verbal consent for their data to be used in this study. These self-report questionnaires were completed within the training room in the presence of the facilitators. No participants requested assistance to complete any of the self-report measures. Upon completion of the post measures, partici- pants were verbally debriefed once more about how the questionnaire results were intended to be used.

Ethical approval

As the study did not involve accessing any patients or patient identifiable data, ethical approval was not necessary and this was confirmed by the Research

The Journal of Forensic Psychiatry & Psychology 853

 

 

and Development Department of the NHS Health Board. All participants were informed that we were collecting data before and after the training in order to evaluate it. To protect confidentiality, the pre- and post-questionnaire measures were anonymous and attached together, ensuring they came from the same per- son. Information sheets and consent forms were not used, but participants were informed verbally about the questionnaires and had the choice whether or not to complete them.

Statistical analysis

All data were entered into IBM SPSS version 20 for Windows and checked for accuracy by the first author. Data were described and distribution was checked using the Shapiro–Wilk test. A number of variables on the CHABA and the CDS were not normally distributed, therefore the Wilcoxon signed- rank non-parametric test was used to assess for differences pre- and post- training in these variables. A related t-test was used to calculate whether there was a significant difference between confidence pre- and post-training as this was normally distributed. Baseline and post-training comparisons between qualified and unqualified staff were made using Mann–Whitney U tests.

Results

The means and ranges for pre- and post-training measures and for qualified and unqualified staff are shown in Table 2. In addition, Table 2 presents differ- ences between qualified and unqualified staff before and after training.

Unqualified staff were significantly more confident in working with chal- lenging behaviour than qualified staff (p = .007) at baseline, yet after training, this significant difference was no longer evident, with a large increase in the confidence of the qualified staff. A related t-test showed that confidence in working with challenging behaviour significantly increased after training for both qualified (t (29) = −6.56, p = <.001) and unqualified staff (t (27) = −5.67, p = <.001). Hypothesis 1 was therefore supported, and confidence in working with challenging behaviour increased for both qualified and unqualified staff after training in PBS.

At baseline, and after training, qualified staff attributed challenging behaviour to external causes, such as ‘Physical Environment’ (p = .008) and ‘Stimulation’ (p = .006), significantly more than unqualified staff, though the difference did reduce after training they remained significant. Unqualified staff considered challenging behaviour to be significantly more stable than qualified staff both before (p = .004) and after (p = <.001) training. As a number of the attribution variables were not normally distributed, the Wilcoxon signed-rank non-parametric test was used to assess for differences pre- and post-training. Both qualified and unqualified staff showed significant increases, after training in PBS, in attributing the causes of challenging behaviour to learning, learned

854 B. Davies et al.

 

 

T ab le

2 .

T ab le

sh o w in g th e m ea n an d th e ra n g e sc o re s p re – an d p o st -P B S tr ai n in g an d d if fe re n ce s id en ti fi ed

b et w ee n q u al ifi ed

an d

u n q u al ifi ed

st af f u si n g M an n – W h it n ey

U te st s.

P re -t ra in in g

P o st -t ra in in g

Q u al ifi ed

m ea n

(r an g e)

U n q u al ifi ed

m ea n

(r an g e)

S ig n ifi ca n t

d if fe re n ce

– p − 2 ta il

Q u al ifi ed

m ea n (r an g e)

U n q u al ifi ed

m ea n (r an g e)

S ig n ifi ca n t

d if fe re n ce

– p − 2 ta il

C o n fi d en ce

3 7 .5 2 (2 7 – 4 9 )

4 1 .2 1 (2 3 – 5 0 )

.0 0 7 * *

4 4 .8 0 (3 5 – 5 4 )

4 7 .0 0 (3 9 – 5 6 )

.1 6 4

C H A B A

le ar n ed

5 .1 3 (0 – 1 2 )

4 .3 4 (0 – 11 )

.2 3 7

6 .9 1 (0 – 1 2 )

6 .1 3 (0 – 1 2 )

.2 7 6

C H A B A

le ar n ed

p o s.

3 .1 3 (0 – 6 )

3 .0 7 (− 1 – 6 )

.9 9 6

3 .7 9 (− 1 – 6 )

3 .6 0 (0 – 6 )

.5 4 0

C H A B A

le ar n ed

n eg .

2 .0 0 (− 3 – 6 )

1 .1 9 (− 2 – 5 )

.0 5 1

3 .0 6 (− 2 – 6 )

2 .5 3 (− 1 – 6 )

.1 3 4

C H A B A

b io m ed ic al

1 .8 7 (− 11 – 11 )

1 .7 1 (− 4 – 11 )

.5 4 9

4 .4 7 (− 6 – 1 2 )

3 .9 7 (− 5 – 1 2 )

.4 4 4

C H A B A

p h y si ca l en v.

2 .2 8 (− 1 2 – 11 )

− .0 3 2 (− 1 3 – 11 )

.0 0 8 * *

6 .0 4 (− 11 – 1 6 )

4 .3 2 (− 5 – 1 6 )

.0 4 9 *

C H A B A

em o ti o n al

9 .3 6 (3 – 1 4 )

9 .0 6 (1 – 2 1 )

.5 7 7

1 0 .1 7 (0 – 1 4 )

9 .3 8 (0 – 1 4 )

.4 0 7

C H A B A

st im

u la ti o n

3 .0 9 (− 4 – 11 )

.7 8 (− 7 – 9 )

.0 0 6 * *

4 .2 1 (− 3 – 1 2 )

2 .1 7 (− 4 – 1 2 )

.0 1 2 *

C D S lo cu s o f

ca u sa li ty

1 5 .5 6 (1 0 – 2 6 )

1 5 .6 4 (8 – 2 3 )

.6 8 9

1 3 .8 2 (6 – 2 1 )

1 5 .8 5 (1 2 – 2 3 )

.0 9 8

C D S ex te rn al

co n tr o l

9 .9 8 (3 – 1 8 )

9 .8 6 (6 – 1 4 )

.8 8 5

1 0 .1 8 (4 – 1 5 )

9 .8 5 (6 – 1 5 )

.4 6 3

C D S st ab il it y

11 .0 2 (6 – 1 6 )

1 3 .3 2 (6 – 2 1 )

.0 0 4 * *

1 0 .0 0 (3 – 1 5 )

1 3 .2 2 (8 – 1 7 )

< .0 0 1 * *

C D S p er so n al

co n tr o l

1 4 .0 0 (5 – 2 3 )

1 4 .2 8 (8 – 2 7 )

.9 8 6

1 2 .1 6 (3 – 2 1 )

1 4 .5 9 (1 0 – 1 9 )

.0 1 5

* * p < .0 1 ; * p < .0 5 .

The Journal of Forensic Psychiatry & Psychology 855

 

 

negative, biomedical causes, the physical environment and levels of stimula- tion, as shown in Table 3. However, qualified staff also showed a significant increase in attributing the cause of challenging behaviour to learned positive, while no significant change was noted for unqualified staff in this. Hypothesis 2 was therefore partially supported, with increases in attributions for the exter- nal causes of challenging behaviour for both qualified an unqualified staff in relation to learning, learned negative, the physical environment and levels of stimulation. There was, however, not a reduction in attribution to internal causes on the CHABA, with both qualified and unqualified staff showing an increase of attributing behaviour to ‘Biomedical’ causes, and no significant changes in relation to ‘Emotional’ causes.

With regard to variables measured on the CDS, significant reductions were found in attributing locus of causality of challenging behaviour to the service user (p = .001) and considering challenging behaviour as stable and changeable (p = .026) by the qualified staff but not the unqualified staff. There were no significant changes in scores on the external control or personal control domains for either qualified or unqualified staff. Hypothesis 2 was therefore only partially supported in relation to the reduction of attributing challenging behaviour cause to the service user, and seeing challenging behaviour as less stable, by the qualified staff, but not the unqualified staff. In addition, there were no changes in relation to attributions for external or personal control for either qualified or unqualified staff.

As significant changes were observed with the qualified staff but not unqualified staff in relation to ‘Locus of Causality’ and ‘Stability’, the difference in the amount of change was compared between the two groups to identify if this was significant using the Mann–Whitney U test, due to the

Table 3. The results of Wilcoxon signed-rank test showing changes in attributions for challenging behaviour, as measured by the CHABA and CDS-II, after training in PBS.

Qualified staff Unqualified staff

z n p − 2 tail z n p − 2 tail

CHABA learned −3.782 45 <.001** −2.995 29 .003** CHABA learned pos. −2.770 46 .006** −1.828 29 .068 CHABA learned neg. −3.642 46 <.001** −3.095 30 .002** CHABA biomedical −4.446 42 <.001** −2.741 27 .006** CHABA physical env. −4.474 42 <.001** −4.036 26 <.001** CHABA emotional −1.948 44 .051 −.737 29 .461 CHABA stimulation −2.911 47 .004** −2.488 26 .013* CDS locus of causality −3.267 43 .001** −.262 27 .794 CDS external control −.554 44 .580 −.094 27 .925 CDS stability −2.228 44 .026* −.244 26 .807 CDS personal control −1.932 42 .053 −.824 27 .410

**p < .01; *p < .05.

856 B. Davies et al.

 

 

non-parametric nature of some of the data. The only variable showing a significant difference in the change made was locus of causality. Qualified staff made a significantly greater reduction than unqualified staff in attributing the locus of causality to the service user (U = 362, p = .008).

Discussion

Staff reported feeling more confident in their ability to manage challenging behaviour following PBS training, these results supporting hypothesis one. Prior to training, unqualified staff rated themselves as significantly more confi- dent than qualified staff in dealing with challenging behaviour, this difference no longer existed after training, and improvements were seen for both qualified and unqualified staff in their self-reported confidence. These results reflect those of other findings, largely within learning disabilities services, demonstrat- ing improvements in confidence through the introduction of PBS training (Lowe et al., 2007; Tierney et al., 2007).

Attributions for challenging behaviours are important due to the well- established links between attributions and helping behaviours (Weiner, 1980). All causal attributions, measured using the CHABA, increased significantly for qualified and unqualified members of staff, with the exception ‘Learned Posi- tive’ attributions which significantly increased only for qualified staff, and of ‘Emotional’ attributions which showed no significant change for either group. The second hypothesis was therefore partially supported in relation to the CHABA measures as attribution for external causes did increase however there was no reduction in attribution to internal causes. As the PBS training provided did recognise emotional and biomedical causes as possible ‘slow trig- gers’ for challenging behaviours this could be expected. These findings provide further evidence of training in PBS changing attributions for causality of chal- lenging behaviour, supporting previous research conducted by Grey, McClean, and Barnes-Holmes (2002), Dowey et al. (2007) and Lowe et al. (2007).

The changes in attributions for locus of causality, control and stability as measured by the CDS (McAuley et al., 1992), were only evident in relation to qualified staff, no significant changes occurred for the unqualified staff. For qualified staff, locus of causality for challenging behaviour viewed as being within a service user reduced significantly, indicating a move to attributing the causes of challenging behaviour to external environmental factors. In addition, considering challenging behaviour to be more stable and unchangeable signifi- cantly reduced for qualified staff, indicating more optimism that change is possible amongst this client group. The second hypothesis was therefore par- tially supported by the findings in relation to the CDS measures, but changes were only observed in the qualified staff’s attributions after training.

These findings indicate that training in PBS is effective in changing attribu- tions for challenging behaviour, on a number of scales for both qualified and unqualified staff working in forensic mental health settings. These changes are

The Journal of Forensic Psychiatry & Psychology 857

 

 

likely to have a positive impact on the care of service users as perceived treatability, and helping behaviour is predicted by attributing challenging beha- viour as less stable and less internal (Leggett & Silvester, 2003; Quinsey & Cyr, 1986; Reid & Millard, 1995; Sharrock et al., 1990).

It is notable, however, that unqualified staff made no significant changes in their attributions around stability, control and causality on the CDS. When com- paring the changes made in scores on the CDS scales, qualified staff made sig- nificantly greater changes in relation to locus of causality than unqualified staff. In addition, at baseline and after training, qualified staff showed significantly higher levels of attributing challenging behaviour to ‘physical environment’ and ‘stimulation’ causes; however, both qualified and unqualified staff showed sig- nificant increases in these variables after training. Possible reasons for these variations may include differences in the pre-existing skills and experience of the qualified staff enabling them to consider different causalities and reflect on experiences of change in service users behaviours. Most of the CHABA scales showed significant increases, from baseline, for both qualified and unqualified staff; however, this was not the case for the CDS scales. It is possible the latter part of the training programme (which unqualified staff did not receive), involv- ing conducting behavioural assessment and planning interventions, had a sig- nificant impact on modifying challenging behaviour attributions. This possibly offered qualified staff insight into the causes and stability of challenging beha- viour, and interventions that could facilitate change. Future research may look to provide all staff with the same level of training (one-day training) in order to draw firm conclusions as to why these differences were found.

A limitation of this study is that the only post-training measures were taken on the same day the training was provided therefore there are no actual mea- sures of whether these changes are maintained. In future research, the effects of staff PBS training needs to be followed up in an attempt to assess its poten- tial longevity. This would help to clarify whether the benefits of training are outweighed by the associated financial and time burden of training.

The success of the training to date has led to the training being rolled out to staff across all wards in the service and positive feedback being received from those who have taken part to date. Within the NHS, there is a growing emphasis on minimising the use of ‘hands-on’ methods and medication to manage challenging behaviours, focusing instead on prevention of behaviours and verbal de-escalation techniques, which the PBS model endorses (National Institute for Health & Clinical Excellence, 2005; NHS, 2014). It would be helpful to establish, in future research, whether the training leads to positive changes in practice in preventing and managing challenging behaviour by the employment of PBS techniques. Further research is also needed to determine to what extent these attribution and confidence changes translate into reduc- tions in the frequency and intensity of challenging behaviours.

In conclusion, this study has shown that provision of training on PBS can be effective in increasing staff confidence in managing challenging behaviours,

858 B. Davies et al.

 

 

and changing attributions for causality and stability of challenging behaviours. These changes should translate to positive approaches in the care of service users. Although this study should be viewed as preliminary evidence for the efficacy of PBS staff training within forensic services, further research replicat- ing these findings could lead to the adoption of PBS as a viable primary prevention approach in the management of challenging behaviours. Follow-up studies will determine the effect such training and the implementation of PBS plans have on measurable changes in challenging behaviours.

Acknowledgements The authors would like to thank the staff who participated and the service managers for their on-going support and commitment to the PBS project.

References Allen, D., James, W., Evans, J., Hawkins, S., & Jenkins, R. (2005). Positive behavioural

support: Definition, current status and future directions. Tizard Learning Disability Review, 10, 4–11.

Ball, C. T. (2013). Unexplained sporting slumps and causal attributions. Journal of Sport Behaviour, 36, 233–242.

Barrowclough, C., Haddock, G., Lowens, I., Connor, A., Pidliswyj, J., & Tracey, N. (2001). Staff expressed emotion and causal attributions for client problems on a low security unit: An exploratory study. Schizophrenia Bulletin, 27, 517–526.

Boisvert, C., & Faust, D. (1999). Effects of the label “Schizophrenia” on causal attribu- tions for violence. Schizophrenia Bulletin, 25, 479–491.

Curtis, R., Van Horne, J., Robertson, P., & Karvonen, M. (2010). Outcomes of a school-wide positive behavioural support program. Professional School Counseling, 13, 159–164.

Department of Health. (2014). Positive and proactive care: Reducing the need for restrictive interventions. London: Author.

Dowey, A., Toogood, S., Hastings, R. P., & Nash, S. (2007). Can brief workshop inter- ventions change care staff understanding of challenging behaviours? Journal of Applied Research in Intellectual Disabilities, 20, 52–57. doi:10.1111/j.1468- 3148.2006.00339.x

Forsyth, A. (2007). The effects of diagnosis and non-compliance attributions on therepeutic alliance processes in adult acute psychiatric settings. Journal of Psychi- atric and Mental Health Nursing, 14, 33–40.

Grey, I. M., McClean, B., & Barnes-Holmes, D. (2002). Staff attributions about the causes of challenging behaviours effects of longitudinal training in multi-element behaviour support. Journal of Intellectual Disabilities, 6, 297–312. doi:10.1177/ 1469004702006003037

Griffiths, J., & Wilcox, D. (2013). Positive behaviour support in a medium secure envi- ronment. Mental Health Practice, 16, 24–27.

Hastings, R. P. (1997). Measuring staff perceptions of challenging behaviour: The chal- lenging behaviour attributions scale (CHABA). Journal of Intellectual Disability Research, 41, 495–501. doi:10.1111/j.1365-2788.1997.tb00742.x

Leggett, J., & Silvester, J. (2003). Care staff attributions for violent incidents involving male and female patients: A field study. British Journal of Clinical Psychology, 42, 393–406. doi:10.1348/014466503322528937

The Journal of Forensic Psychiatry & Psychology 859

 

 

Lowe, K., Jones, E., Allen, D., Davies, D., James, W., Doyle, T., … Brophy, S. (2007). Staff training in positive behaviour support: Impact on attitudes and knowledge. Journal of Applied Research in Intellectual Disabilities, 20, 30–40. doi:10.1111/ j.1468-3148.2006.00337.x

MacKinlay, L., & Langdon, P. (2009). Staff attributions towards men with intellectual disability who have a history of sexual offending and challenging behaviour. Journal of Intellectual Disability Research, 53, 807–815.

Markham, D., & Trower, P. (2003). The effects of the psychiatric label ‘borderline personality disorder’ on nursing staff’s perceptions and causal attributions for challenging behaviours. British Journal of Clinical Psychology, 42, 243–256.

Martin, T., & Daffern, M. (2006). Clinician perceptions of personal safety and confidence to manage inpatient aggression in a forensic psychiatric setting. Journal of Psychiatric and Mental Health Nursing, 13, 90–99. doi:10.1111/j.1365- 2850.2006.00920.x

McAuley, E., Duncan, T. E., & Russell, D. W. (1992). Measuring causal attributions: The revised causal dimension scale (CDSII). Personality and Social Psychology Bulletin, 18, 566–573. doi:10.1177/0146167292185006

McClean, B., Dench, C., Grey, I., Shanahan, S., Fitzsimons, E., Hendler, J., & Corrigan, M. (2005). Person focused training: A model for delivering positive behavioural supports to people with challenging behaviours. Journal of Intellectual Disability Research., 49, 340–352.

McDonnell, A., Sturmey, P., Oliver, C., Cunningham, J., Hayes, S., Galvin, M., … Cunningham, C. (2008). The effects of staff training on staff confidence and chal- lenging behaviour in services for people with autism spectrum disorders. Research in Autism Spectrum Disorders, 2, 311–319. doi:10.1016/j.rasd.2007.08.001

McGowan, S., Wynaden, D., Harding, N., Yassine, A., & Parker, J. (1999). Staff confi- dence in dealing with aggressive patients: A benchmarking exercise. Australian and New Zealand Journal of Mental Health Nursing, 8, 104–108. doi:10.1046/j.1440- 0979.1999.00140.x

McKenzie, K., Sharp, K., Paxton, D., & Murray, G. C. (2002). The impact of training and staff attributions on staff practice in learning disability services a pilot study. Journal of Intellectual Disabilities, 6, 239–251. doi:10.1177/1469004702006003034

Meddings, S., & Levey, S. (2000). Staff attributions and management of violent inci- dents in hostels for homeless people. Journal of Mental Health, 9, 481–494. doi:10.1080/09638230020005219

National Institute for Health and Clinical Excellence. (2005). Violence: The short-term management of disturbed/violent behaviour in in-patient psychiatric settings and emergency departments (Guidance no. 25). London: Author.

NHS. (2014). Meeting needs and reducing distress: Guidance on the protection and management of clinically related challenging behaviour in NHS settings. London: Author.

Quinsey, V. L., & Cyr, M. (1986). Perceived dangerousness and treatability of offenders the effects of internal versus external attributions of crime causality. Journal of Interpersonal Violence, 1, 458–471. doi:10.1177/088626086001004005

Reid, D., & Millard, R. (1995). The application of attribution theory to perceptions of dangerousness and treatability. Issues in Criminological and Legal Psychology, 27, 4–9.

Royal College of Psychiatrists. (2007). Challenging behaviour: A unified approach. London: Author.

Sharrock, R., Day, A., Qazi, F., & Brewin, C. R. (1990). Explanations by professional care staff, optimism and helping behaviour: An application of attribution theory. Psychological Medicine, 20, 849–855. doi:10.1017/S0033291700036540

860 B. Davies et al.

 

 

Stanley, B., & Standen, P. (2000). Carers’ attributions for challenging behaviour. British Journal of Clinical Psychology, 39, 157–168.

Thackrey, M. (1987). Clinician confidence in coping with patient aggression: Assessment and enhancement. Professional Psychology: Research and Practice, 18, 57. doi:10.1037/0735-7028.18.1.57

Tierney, E., Quinlan, D., & Hastings, R. P. (2007). Impact of a 3-day training course on challenging behaviour on staff cognitive and emotional responses. Journal of Applied Research in Intellectual Disabilities, 20, 58–63. doi:10.1111/j.1468- 3148.2006.00340.x

Urquart Law, G., Rostill-Brookes, H., & Goodman, D. (2009). Public stigma in health and non-healthcare students: Attributions, emotions, willingness to help with adoles- cent self-harm. International Journal of Nursing Studies, 46, 108–119. doi:10.1016/ j.ijnurstu.2008.08.014

Weiner, B. (1980). A cognitive (attribution)-emotion-action model of motivated behav- ior: An analysis of judgments of help-giving. Journal of Personality and Social Psychology, 39, 186. doi:10.1037/0022-3514.39.2.186

Williams, S., Dagnan, D., Rodgers, J., & McDowell, K. (2012). Changes in attributions as a consequence of training for challenging and complex behaviour for carers of people with learning disabilities: A systematic review. Journal of Applied Research in Intellectual Disabilities, 25, 203–216. doi:10.1111/j.1468-3148.2011.00654.x

The Journal of Forensic Psychiatry & Psychology 861

 

 

Copyright of Journal of Forensic Psychiatry & Psychology is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.