Research supports the use of psychological therapies among people with mild to moderate intellectual disability (ID). One barrier to people with ID accessing psychological treatments is the confidence of mental health practitioners. This article explores the confidence of Australian clinicians in providing therapy to people with ID. One hundred and fifty-two psychologists and counselors in Australia completed a survey exploring self-reported confidence when working with clients who have ID and mental health difficulties. Clinicians were most confident with generic counseling skills, but less confident with elements of assessments and interventions. The use of treatment protocols was endorsed as helpful particularly among those with low confidence. This highlights the need for dissemination of treatment guides and training to help increase clinician confidence.
Rates of mental illness for people with intellectual disability (ID) are high, ranging from 16% (Cooper, Smiley, Morrison, Williamson, & Allen, 2007) to 54% (Gustafsson & Sonnander, 2004). In the last decade, research into psychological treatments for people with ID and comorbid psychopathology has produced promising results (Osugo & Cooper, 2016; Vereenooghe & Langdon, 2013). However, despite the developing evidence base for psychological therapies, access to mental health services among people with ID remains low (Michael & Richardson, 2008). One identified barrier to treatment is the confidence of clinicians when working with this population (Dagnan, Masson, Cavagin, Thwaites, & Hatton, 2014; Rose, O'Brien, & Rose, 2007). The current study aims to consider the level of confidence of Australian practitioners and associations with confidence, with a view to identifying ways to improve access to psychological treatments for people with ID.
In Australia, less than 10% of young people with ID and mental illness accessed treatment over a 14-year period (Einfeld et al., 2006), in contrast to 35% of the general population who accessed treatments for mental health disorders (Slade, Johnston, Oakley Browne, Andrews, & Whiteford, 2009). One factor thought to prevent people with ID accessing mental health services is the limited confidence of clinicians in working with this population (Dagnan et al., 2014; Rose et al., 2007). It is known that for clinicians working with individuals who do not have ID, confidence has a significant impact upon the engagement of clients and their mental health outcomes (Bennun, Hahlwek, Schindler, & Langholtz, 1986; Keijsers, Schaap, & Hoogduin, 2000; Shaw et al., 1999). Heinonen and colleagues (2012) found that lower therapist confidence was predictive of poorer client outcomes for those with anxiety disorders. Furthermore, retrospective evaluation from clients, indicates that patients who improve most on mental health measures perceive their therapists as more confident, competent, and experienced (Bennum et al., 1986; Keijsers et al., 2000).
Specific factors have been identified as related to increased clinician self-efficacy and confidence, namely clinical experience and training. Clinical experience is associated with confidence—the more experience a practitioner has, the greater their level of confidence (Ronnestad & Skovholt, 2003; Bischoff & Barton, 2002). In addition, training and professional development are associated with increased confidence and knowledge (Beidas & Kendall, 2010). Professionals have reported lacking the skills, confidence, and training to work with disability populations (Bouras & Holt, 2004; Rose et al., 2007; Torr et al., 2008). Man and colleagues (2016) explored the clinical competencies and training needs of Australian psychologists working with people with ID. Overall, the majority of psychologists reported limited academic training in working with people with ID and expressed a need for continual and specialized training, which has been found to increase the knowledge, skill, and confidence of practitioners working with people with ID (Murray, McKenzie, Quigley, & Sinclair, 1999; Torr et al., 2008; Werner & Stawski, 2012).
Research by Dagnan and colleagues (2014) has examined how therapist confidence can be measured in a systematic way. The Therapy Confidence Scale—Intellectual Disabilities (TCS-ID; Dagnan et al., 2014) assesses confidence for various therapeutic processes such as building rapport, explaining procedures, and identifying appropriate treatment approaches. Clinicians in the United Kingdom (UK) were most confident with “generic” therapy skills such as providing empathy and listening and forming therapeutic relationships, but less confident using assessments and appropriate interventions (Dagnan et al., 2014). This likely reflects the limited research into effective psychological treatments for people with ID, and the subsequent lack of existing treatment protocols to guide clinicians. Although treatment guides are not without limitations (Duncan & Miller, 2006), they have been acknowledged as important vehicles for the dissemination of evidence-based practices and are used by approximately half of clinicians (Becker, Smith, & Jensen-Doss, 2013).
The current study aims to extend upon the research of Dagnan and colleagues (2014) and Man and colleagues (2016). Based upon findings of limited clinician confidence in the UK (Dagnan et al., 2014) and reports that Australian clinicians lack specialist competencies (Man, Kangas, Trollor, & Sweller, 2016), we sought to directly examine confidence in an Australian clinician sample. Given that little is known about how confident Australian clinicians feel when working with people with ID, the aims of the research were twofold:
Identify how confident Australian clinicians are when working therapeutically with people with ID; and
Identify factors that may be associated with increased confidence.
It was hypothesised that clinicians would report greater confidence using generic counseling skills with clients with ID, but lower confidence with specific elements of assessment and intervention, similar to Dagnan et al., (2014). It was hypothesized that previous experience working therapeutically with clients with ID and specific professional development in relation to working with people with ID would be associated with increased confidence. In addition, given the limited published research on conducting interventions with people with ID, it was expected that clinicians would endorse treatment guides as enabling them to provide better quality therapy and increased confidence in doing so. It was predicted that clinicians with low levels of confidence or limited experience working with this population would particularly endorse the use of treatment manuals.
The questionnaire was completed by 152 clinicians currently working in Australia, including clinicians who had not previously worked with people with ID. The sample had a mean age of 37.93 (Standard Deviation [SD] = 12.46), and 85.5% were female. In Australia psychologists can achieve general registration via two year's supervision post-completion of a 4 year undergraduate degree. Alternatively, these two years can consist of a university based Master's degree in a specialist area (e.g., counseling psychology, clinical psychology). While completing the two years under supervision, psychologists are ‘provisionally' registered. After completing two years post-Master's supervision during which they are known as ‘registrars', a psychologist can be endorsed in a specialist area (e.g., counseling psychology, clinical psychology). Participants were counselors of varying training backgrounds or psychologists who were provisionally registered, generally registered, completing a registrar program, or registered with specialist endorsement. The largest group was psychologists with specialist endorsement (32.9%); clinical psychology was the most common endorsement. The average number of years working as a psychologist or counselor was 9.45 years (SD = 9.35), ranging from less than 1 year to 42 years. Most completed their undergraduate training in Australia (87.5%) and for those with postgraduate training, 83.6% attended in Australia. Overall, the sample was relatively representative of that of Australian psychologists (Australian Health Practitioner Regulation Agency, 2016).
Therapist experience with intellectual disability
Participants were asked whether they had provided therapy to people with ID, the treatments they had used, and whether they had undertaken additional training/professional development in relation to working with people with ID. Knowledge of existing treatment guides, protocols, and manuals was assessed. Participants were asked to rate if they thought using treatment guides when working with clients with ID may improve the quality of therapy and increase clinician confidence. This was rated on a five-point scale from 1 = “not confident” to 5 = “highly confident.”
Therapy confidence scale—Intellectual disabilities (TCS-ID;
Dagnan et al., 2014)
The TCS-ID is a 14 item self-report measure of confidence in working with people with ID, rated on a five-point scale anchored from 1 = “not confident” to 5 = “highly confident.” Clinicians rate confidence for key elements of the therapy process, including therapeutic relationship, assessment, and intervention. The sum of these items provides an overall measure of confidence, with a maximum score of 70. Due to terminology differences between Australia and the UK, the term “learning disability” was changed to “intellectual disability.”
The TCS-ID has good internal consistency (Cronbach's α = .93), high test-retest reliability (r = .83) and a mean corrected item-total correlation of .68 (SD = 0.08, Dagnan et al., 2014). The scale has a single factor structure and is significantly positively correlated with the General Clinical Self-Efficacy Scale (r = .43; Dagnan et al., 2014). The current study revealed good internal consistency (α = .96) and mean corrected item-total correlation ranged from .52–.86.
The research was approved by the University of Technology Sydney ethics committee. Participants were recruited via online advertising through professional organizations, social media, and sharing between professionals. The study was advertised as seeking mental health practitioners to complete a brief online survey about providing psychological therapy to clients with ID. It was stated that clinicians did not need to have experience working with clients with ID. All clinicians consented to participating prior to completing the study. The survey was disseminated using Qualtrics® (2016) survey software.
Descriptive statistics were used to identify sample characteristics and mean ratings for items on the TCS-ID. Cronbach's alpha was calculated to confirm internal consistency on the TCS-ID. Correlations and independent sample t-tests were used to identify differences for confidence on the TCS-ID according to gender, age, and years of experience, and analysis of variance (ANOVA) to compare confidence for professional groups.
A regression model was used to determine the significance of factors associated with confidence on the TCS-ID. Correlational analyses and independent sample t-tests were used to analyze the relationship between clinician confidence and how treatment guides may improve the quality of therapy and clinician confidence. An alpha level of .05 was used for all statistical tests.
One hundred and four participants (68.4%) had provided therapy to at least one client with ID and a comorbid mental illness. Of those, 76.0% used Cognitive Behavioral Therapy (CBT), 31.7% used Acceptance and Commitment Therapy (ACT), 26.9% used Interpersonal Psychotherapy (IPT), 9.6% used Dialectical Behaviour Therapy (DBT), 7.7% used Psychodynamic Therapy, 2.9% used Schema Therapy, and 39.4% used other therapies. (Note: Participants were able to select more than one treatment approach). Other forms of therapy included behavioral interventions, attachment-based therapy, family therapy, social skills training, mindfulness, play therapy, solution-focused therapy, and narrative therapy.
Approximately half the sample (50.7%) had undertaken training/professional development to further their skills in working with clients with ID. Of those, 76.6% attended workshops, 70.1% engaged in peer consultation, 67.5% read books/journal articles, and 27.3% engaged in other forms of professional development. (Note: Participants were able to select more than one form of training/professional development modality). Other ways participants had developed their skills included attending lectures, clinical supervision, and specific workplace training.
Confidence in therapy
Means (M) and SDs for each item on the TCS-ID were calculated (Table 1). Participants were most confident being empathetic towards a client with ID (M = 4.38), listening carefully to concerns presented by a client with ID (M = 4.01), and working with caregivers and important people in the lives of people with ID (M = 3.79). Participants were least confident explaining the results of an assessment to a person with ID (M = 2.84), using assessments in a way that a client with ID would understand (M = 2.85), and identifying therapeutic approaches that would be effective for a client with ID (M = 3.00). The overall sample mean on the TCS-ID was 47.34 out of 70 (SD = 11.90).
ANOVA was used to determine whether there were significant differences in confidence according to professional qualifications. Participants significantly differed in their confidence based on professional group, F(5,145) = 3.02, p = .013. Post hoc comparisons using the Tukey HSD test indicated that psychologists with general registration (M = 46.71, SD=10.69) were significantly more confident than provisional psychologists in an accredited university degree (M = 40.72, SD = 10.31), p = .012. Counselors (M = 53.23, SD = 12.44) were significantly more confident than provisional psychologists with an accredited university degree (M = 40.72, SD = 10.31), p = .023.
Confidence and age were significantly positively correlated, such that increased confidence was associated with older age (Table 2). Confidence and years of experience produced a significant positive correlation, as increased confidence was associated with greater years of experience. No significant difference was found for male (M = 50.23, SD = 13.39) compared to female clinicians (M = 46.85, SD = 11.61), t(150) = 1.23, p = .220. Previous experience working with clients with ID was significantly correlated to confidence such that those who had experience with this population were more confident, and training relating to ID was significantly correlated with confidence such that those who had undertaking ID specific training were more confident providing therapy.
An independent t-test revealed that on the TCS-ID, those who had provided therapy to clients with ID and mental health disorders were significantly more confident (M = 50.76, SD = 11.21) than those who had not (M = 39.94, SD = 9.87), t(1) = 5.74, p < .001. Participants who used a psychodynamic approach were of highest confidence on the TCS-ID (M = 59.00, SD = 9.87), while those who used schema therapy reported the lowest mean confidence (M = 46.33, SD = 20.66) (Table 3). Of those who had provided therapy to people with ID, 36.5% indicated using only one type of therapy, while 63.5% had used two or more types of therapy. An independent t-test indicated that those who used two or more types of therapy were significantly more confident (M = 53.88, SD = 10.57) than those who used only one type (M = 45.34, SD = 10.31), t(102) = 4.00, p < .001.
Participants who developed their skills in working with clients with ID and mental health disorders were significantly more confident on the TCS-ID (M = 51.53, SD = 10.96) than those who had not (M = 43.04, SD = 11.33), t(150) = 4.70, p < .001. Participants who engaged in “other” forms of training had the highest mean confidence, (M = 57.14, SD = 9.58), followed by peer consultation (M = 52.87, SD = 11.12), reading books/journal articles (M = 52.21, SD = 11.17) and attending workshops (M = 52.19, SD = 10.87). Of those who had developed their skills specific to working with people with ID, 23.4% had engaged in one type of training, while 76.6% engaged in two or more. An independent t-test indicated that those who engaged in two or more types of training were significantly more confident (M = 53.15, SD = 10.75) than those who engaged in only one type of training (M = 46.22, SD = 10.18), t(150) = 2.42, p = .018.
A stepwise multiple regression was conducted with mean confidence on the TCS-ID as the dependent variable. No independent variables were highly correlated, with the exception of age and years of experience in the profession which produced a strong correlation (Table 2). Thus, in the regression analysis, age was excluded as an independent variable. Assumptions of normality, linearity, and homoscedasticity were satisfied. Gender and years working in the profession were entered into step one of the model. Experience treating an individual with ID and undertaking training in ID were added in step two. At step one, gender and years of experience contributed significantly to the regression model, F(2,148) = 12.89, p < .001 accounting for 14.8% of variance in confidence. ID specific variables (i.e., previous experience in treating patients with ID and training specific to ID) introduced in step two explained an additional 16.4% of variation in confidence and this change in R2 was significant, F(4,146) = 16.61, p < .001. When all four predictors were added, number of years' experience and ID specific variables were significant predictors of confidence.
Endorsement of treatment manuals
Approximately three quarters of the sample (76.3%) were not aware of any treatment manuals or protocols which existed to guide therapy when working with adults who have ID and mental illness and 68.4% were not aware of any treatment guides for working with children who have ID and mental illness. Overall, respondents endorsed the use of treatment protocols and manuals when working with clients with ID. Seventy-one percent agreed or strongly agreed that using treatment manuals may improve the quality of therapy, while 86.1% agreed or strongly agreed that using treatment manuals may increase confidence when working with clients with ID and mental health disorders (see Figure 1).
Correlational analyses revealed no significant association between confidence on the TCS-ID and whether participants endorsed the view that using treatment protocols may assist to improve the quality of therapy, r = −.15, p = 075. There was, however, a significant negative correlation between confidence on the TCS-ID and whether respondents thought that manuals may improve clinician confidence when working with clients with ID, such that those with lower levels of confidence thought that manuals would assist them to feel more confident, r = −.19, p = .023.
Clinicians who had not provided therapy to clients with ID were significantly more likely to agree with treatment guides helping provide better quality therapy (M = 4.13, SD = 0.87), compared to those who had provided therapy to clients with ID (M = 3.65, SD = 1.01), t(150) = −2.79, p = .006. Those who had not provided therapy to clients with ID were significantly more likely to agree treatment manuals would increase confidence (M = 4.48, SD = 0.65), than those who had provided therapy (M = 3.99, SD = 0.96), t(150) = −3.20, p = .002.
This study aimed to explore the confidence of Australian clinicians when working with people with ID and mental health disorders, and factors that may be associated with increased confidence. It was hypothesized that clinicians would be more confident with general counseling skills rather than elements specific to assessments and interventions, as found in the sample of clinicians surveyed by Dagnan et al. (2014). This was supported by the results, as Australian clinicians were most confident being empathic, listening to concerns, and working with caregivers; and least confident using assessments, explaining assessment results, identifying therapeutic approaches, and using knowledge of the most effective mental health interventions to use. This indicates that clinicians can engage people with ID therapeutically, but are less sure how to effectively use and implement assessments and interventions.
Training and professional development activities, as well as prior experience working with clients with ID, accounted for a significant increase in confidence among clinicians, as predicted. These results, in combination with the findings of Man and colleagues (2016), highlight the need and importance of additional training opportunities for practitioners when working with people with ID. Man and colleagues (2016) identified that only about a third of participants received specialist training in ID as part of their academic training, but that the majority of participants would like further professional development in mental health and ID. Therefore, not only do clinicians want specialist training in ID and mental health, training is associated with increased confidence when working with this population and may serve as a way to reduce barriers to psychological treatments.
Finally, it was predicted that the use of treatment guides would be supported by clinicians because the guides enable them to provide better quality therapy and lead to greater clinician confidence when working with people with ID. Although the majority of participants were unaware of existing treatment protocols, most agreed that using such guides would result in better therapy and greater confidence. This was found particularly for clinicians who had low ratings of confidence on the TCS-ID and for clinicians who had not previously provided therapy to people with ID. These results are not surprising, given that participants were less confident with implementing interventions on the TCS-ID. Thus, treatment guides may be able to provide further direction on improving clinician confidence.
Among clinicians who had provided therapy to clients with ID and comorbid psychopathologies, the most frequently used therapy was CBT. This is consistent with existing literature as more evaluation has been done to support the use of CBT for people with ID than other psychological therapies. Of note is that the confidence ratings for those using CBT was lower than other approaches, such as the psychodynamic approach. However, it is unclear whether the differences between therapeutic approaches were statistically significant, and thus warrant further investigation. Additionally, clinicians who used two or more types of therapy were more confident than those who only used one. This could indicate that confidence may be associated with greater flexibility in the approach of the clinician by using multiple therapeutic techniques to tailor treatment to the individual client. Individuals with ID have specific neurocognitive, learning, and language needs which can vary according to etiology and level of disability (Hronis, Roberts, & Kneebone, 2017). It appears that clinicians with knowledge about a range of therapies and the skills to adapt these to client needs are more confident in providing therapy to individuals with ID. This however, is only one possible hypothesis, as the use of multiple therapy modalities may be related to the number of clients with ID with whom the clinician has worked. As such information was not gathered, further research to investigate this may be warranted.
Overall, the results of this study have important implications for clinicians and researchers working with people with ID. Therapist confidence has been recognized as an important predictor of therapy outcomes (Heinonen, Lindfors, Laaksonen, & Knekt, 2012) as well as a barrier preventing people with ID from engaging in therapy (Dagnan et al., 2014; Rose et al., 2007). This study suggests that low therapist confidence might be increased through engagement in training programs and professional development activities, and by providing clinicians with protocols to use as guides when engaging in therapeutic interventions with people with ID.
Clinical counseling skills such as listening, empathy, and affirming, which develop the therapeutic alliance, can alone lead to increased improved treatment outcomes (Horvath & Symonds 1991; Martin, Garske, & Davis, 2000). For many clients however, more specific and focused interventions are required. The findings indicating that clinicians are least confident using assessments and interventions with clients with ID is not surprising, given that research into effective psychological and cognitive-based treatments for people with ID has lagged behind that for neurotypical individuals. There is some support for the use of CBT for people with ID though many of these use small sample sizes, participants with mild mental health symptoms, and have limited follow up (Osugo & Cooper, 2016). Case series studies, pilot studies, and small sample studies have recently been conducted with third wave therapies (e.g. Acceptance and Commitment Therapy, Dialectical Behavior Therapy, Mindfulness Based Cognitive Therapy) though the conclusions which can be drawn are limited. Therefore, not only is further research needed to evaluate the efficacy of various psychological treatments for people with ID, the dissemination of findings in the forms of treatment guides and training programs is likely to be useful among practicing clinicians.
One limitation of the study was that clinicians working with people with ID may have been more motivated to complete the survey than those who have not worked with people with ID, potentially resulting in higher ratings of confidence than may be true for a random sample of Australian practitioners. This also makes it difficult to compare these responses to the responses in the study by Dagnan et al. (2014) who specifically surveyed clinicians whose work was not primarily with individuals with ID. Additionally, the survey did not ask about behavioral interventions which clinicians may have used, but focused on cognitive and psychodynamic therapies. It would have been interesting to compare whether clinicians who provided behavioral therapies were more confident than therapists using other treatments, given that there is extensive literature examining behavioral interventions for this population in comparison to more cognitive based treatments. Furthermore, as participants could identify more than one therapy type which they used, comparisons for confidence could not be made across the different therapy approaches.
Based on the existing literature and findings of this article, it is evident that research into psychological treatments for people with ID is progressing, but these findings are not being effectively disseminated and adopted by clinicians who still lack the confidence to work therapeutically work this population. Future research should focus upon the development and dissemination of specific treatment resources among practicing clinicians, evaluate the effectiveness of treatment guides, and examine whether the implementation of such guides increase the confidence of practitioners. Additionally, specific professional development resources and training programs should be developed and evaluated. It is hoped that by increasing the access mental health practitioners have to training and treatment resources, we can increase their confidence in working with people with ID and subsequently reduce one of the barriers which prevent people with ID engaging in psychological therapy.