A preliminary evaluation of the CBT Decision Making Questionnaire for Anxiety and Related Disorders (CDMQ-A)

ABSTRACT

Objective

Cognitive-behavioural therapy (CBT) is effective for the treatment of anxiety and related disorders (ARDs). Despite this, the use of best-practice CBT in clinical practice is low. While training and assessment strategies have been developed to improve this science-practice gap, both within the educational and clinical training space, many of the assessment techniques developed to enhance the use of best practice CBT remain impractical to use in busy training settings and are prone to bias.

Method

The current study presents a preliminary evaluation of the CBT Decision-Making Questionnaire for Anxiety and Related Disorders (CDMQ-A). The CDMQ-A contains vignettes covering seven diagnostic categories, each followed by three questions, resulting in a 21-item questionnaire designed to assess CBT decision-making in the treatment of ARDs in adult patients. A sample of expert (N = 7) (M_age = 42.14; SD = 5.64; 57.1% female) and provisionally registered psychologists (N = 104) (M_age = 30.76; SD = 8.32; 82.7% female) completed the measure.

Results

Experts indicated that the vignettes demonstrated satisfactory face and ecological validity. Results indicated that the CDMQ-A can effectively discriminate between experts and provisionally registered psychologists with the expert sample scoring significantly higher than the provisionally registered psychologists t(10.63) = 6.9, p = .01; d = 1.74).

Conclusions

Implications for training and clinical practice are discussed.

KEY POINTS

What is already known about this topic:

1. Cognitive-behavioral therapy is effective for the treatment of anxiety and related disorders.

2. Despite evidence, the use of cognitive-behavioral therapy in clinical practice is low.

3. Techniques available to assess and enhance the use of cognitive-behavioral therapy remain impractical to use in busy training settings and are prone to bias.

What this topic adds:

1. The current study presents a preliminary evaluation of the CBT Decision Making Questionnaire for Anxiety and Related Disorders.

2. Results indicated that the questionnaire can effectively discriminate between experts and provisionally registered psychologists.

3. The development and use of such tools have the potential to have significant implications for the dissemination and implementation of evidence-based practice, particularly within busy educational settings.

KEYWORDS:

• Clinical decision-making
• anxiety
• psychometric properties
• dissemination
• CBT
• clinical training

Cognitive-behavioural therapy (CBT) is effective for the anxiety and related disorders (ARDs), demonstrated both in efficacy trials (Hunsley et al., 2014; Norton & Price, 2007; Olatunji et al., 2010) and in effectiveness studies (Hunsley et al., 2014; Stewart & Chambless, 2009). Outcomes from CBT are also durable with patients maintaining their improvements for many years post-treatment (Wootton et al., 2015). As a result, several key international bodies recommend that CBT be the first line of treatment for the ARDs (National Institute for Health and Care Excellence, 2011; National Research Council and Institute of Medicine, 2009) and significant public funding is available for the provision of CBT to those with ARDs both nationally (Barrington, 2006) and internationally (Clark, 2011). Finally, as an evidence-based psychological intervention, CBT has been identified as one of the key therapeutic modalities taught in professional training programmes throughout the world (Barrington, 2006; Hipol & Deacon, 2013; Kazantzis & Munro, 2011; Weissman et al., 2006).

Widespread dissemination of CBT does not necessarily lead to effective implementation or adherence in clinical practice (Patel et al., 2002) and research demonstrates that the use of CBT in clinical practice is poor, both by registered practitioners in the community (Cook et al., 2010; McCausland et al., 2020; Robertson et al., 2020; Young et al., 2001) and within training programs (Weissman et al., 2006). This observation has been made across mental health disorders more generally (Berry et al., 2009; Mussell et al., 2000) and ARDs more specifically (Ehlers et al., 2009; Goisman et al., 1999; Wang et al., 2005). The consequence of this is that many individuals with ARDs may not be receiving an evidence-based intervention (Cook et al., 2010; Goisman et al., 1999; Young et al., 2001), or are receiving an evidence-based intervention in a modified or suboptimal manner (such as through limited number of sessions or in a manner that fails to align with theoretical underpinnings or evidence-based guidelines), which has been shown to reduce the effectiveness of the intervention (Broman-Fulks et al., 2004; Deacon et al., 2012; Schmidt et al., 2000; Shafran et al., 2009; Smits et al., 2008). In fact, when specifically considering ARDs, research suggests that when not receiving an evidence-based intervention, clients are instead receiving such interventions as supportive counselling (Ehlers et al., 2009) or complementary and alternative treatments (Wang et al., 2005). There is a growing body of literature that suggests that barriers to dissemination are varied, but likely include such factors as clinicians beliefs and knowledge or training in evidence-based practice (Shafran et al., 2009).

To ensure adherence to evidence-based intervention, it is important to develop strategies to monitor effective dissemination and implementation of CBT in clinical practice (Beidas et al., 2013; McHugh & Barlow, 2010). This appears particularly important within the training setting, where developing clinicians are first acquiring the knowledge and skills to effectively assess and treat ARDs. Clinical judgment, or the ability to make decisions through the integration of empirical evidence, observation and client data, has been shown to be impacted by experience (Ruscio & Stern, 2006; Spengler et al., 2009), which trainee clinicians do not typically possess. However, this has been debated in the literature by some who suggest that judgment cannot be expected to improve given the ambiguous nature of activities undertaken by psychologists (Dawes, 1994) and that experience may in fact worsen clinical judgment (Wedding, 1991).

Assessments of practical understanding typically include the use of short answer clinical vignettes (Myles & Milne, 2004) and case reports (Barnfield et al., 2007; Keen & Freeston, 2008; McManus et al., 2010). Of clinical vignette tasks, only one standardized task exists to the authors knowledge. The Video Assessment Task (Myles & Milne, 2004) presents short video clips and asks clinicians to answer questions about symptoms, problem assessment and treatment. Initial examinations suggest very good inter-rater reliability for identification of symptoms (r = 0.97), problem identification (r = 1.0) and naming of appropriate CBT strategies (r = 0.94; Myles & Milne, 2004). Importantly however, the use of videos and expert examination of answers makes this technique particularly time and resource intensive.

Assessments of practical application of knowledge typically include the use of objective structured clinical examinations (OSCEs) which are routinely used in medical training settings (Epstein, 2007; Muse & McManus, 2013; Sholomskas et al., 2005). In recent years, the use of OSCEs as an assessment of clinical competence has increased (Kaslow et al., 2009; Pachana et al., 2011; Roberts & Norris, 2020). Research within the medical setting has indicated that OSCEs are as reliable as assessment of interactions with real clients (Epstein, 2007) and its utility as an assessment tool within the field of psychology is likely significant (Fairburn & Cooper, 2011). However, such techniques remain time intensive, and therefore potentially impractical to use as an assessment of clinical decision-making in busy training settings or programmes with large cohorts of students. In psychology, limited research exists to support the reliability and validity of this technique, however preliminary assessments suggest that the use of this technique as an assessment tool is considered to be a valid and realistic measure of competence in psychology training by students and staff alike (Hung et al., 2012; Sheen et al., 2015; Yap et al., 2012). Within medical training settings, these techniques have demonstrated good reliability (Tudiver et al., 2009; Wass et al., 2001).

Clinical practice assessments include assessor-rated treatment sessions and therapists’ self-assessments. Techniques that require observation of live or recorded sessions, such as the transdiagnostic Cognitive Therapy Scale-Revised (Blackburn et al., 2001) and the disorder specific Multicenter Collaborative Study for the Treatment of Panic Disorder-Global Competence Item (Huppert et al., 2001), are an effective means of assessing comprehension, knowledge application, and allow for the provision of feedback on clinician strengths and areas of development. However, these are also particularly time and resource intensive and therefore may be impractical to use in busy training settings or for assessment of larger scale dissemination efforts.

Self- and assessor-rated measures of CBT competence exist, such as the Cognitive Therapy Self-Rating Scale (Bennett-Levy & Beedie, 2007), the Cognitive Therapy Adherence and Competence Scale (Barber et al., 2003) and the Manual-Assisted Cognitive Behaviour Therapy Rating Scale (Davidson et al., 2004), however are particularly prone to bias in the form of both under- and over-estimation of skill by trainees (Brosan et al., 2008; McManus et al., 2012) or through the use of profession specific “buzz words”, such as “cognitive restructuring”, which may allow trainees to identify correct answers based on recognition without necessarily having an understanding of the theoretical underpinnings. Taken together, whilst important developments, there appears a need for a means of evaluating knowledge application or clinical decision-making, particularly within the busy training setting, that is time efficient and not susceptible to bias.

In an effort to overcome issues of previously developed tools, Carpenter et al. (2016) developed and evaluated a 24-item questionnaire (the Assessment of Clinical decision-making in Evidence-based treatment for Child Anxiety and Related and Disorders; ACE CARD). The ACE CARD was designed to assess comprehension and clinical reasoning ability utilized by trainees when working with children with anxiety disorders. The tool consists of 12 clinical vignettes separated into two parallel forms that training clinicians complete using a four item multiple-choice response format. Vignettes describe typical therapeutic situations experienced at different time points when working with anxious youth. Participants are asked to select the multiple-choice response, using a vignette matching style approach, that most reflects the CBT model of treatment.

An initial psychometric evaluation of the ACE CARD found that this tool is sensitive to clinical experience (Carpenter et al., 2016). For example, experts performed significantly better than trainees and the questionnaire was able to accurately distinguish between these populations (Carpenter et al., 2016). This research marks a preliminary but significant step forward in our ability to understand clinical decision-making when working with paediatric patients with ARDs. However, there is currently no research that has looked at the assessment of clinical decision-making when working with adult patients using a brief and easy to administer format, representing a key gap in current dissemination and implementation efforts.

The availability of such a tool may allow training providers to assess CBT clinical decision-making in trainees and allow them to implement appropriate remediation strategies early in the clinician’s development. Therefore, the aim of this study was to develop and evaluate the initial psychometric properties of the CBT Decision-Making Questionnaire for Anxiety and Related Disorders (CDMQ-A), a 21-item questionnaire that measures CBT decision-making in the treatment of the ARDs. It is expected that the CDMQ-A will effectively distinguish between expert and trainee clinicians, demonstrate high levels of reliability and validity, and high levels of sensitivity and specificity.

Method

Participants

A sample of clinicians with expertise in the treatment of ARDs (N = 7) (M_age = 42.14; SD = 5.64; 57.1% female) and provisionally registered psychologists (N = 104 (M_age = 30.76; SD = 8.32; 82.7% female) completed the measure. To be considered an expert, participants were required to have 1) a qualification equal to or greater than a Master’s degree in Clinical Psychology; 2) possess more than 5 years’ post-graduate experience; 3) indicate their clinical specialization as being within the ARDs; and 4) indicate CBT as their main theoretical orientation. Experts were identified through national and international university clinical training programmes and specialist CBT clinics for ARDs. Experts were then invited to participate in this study by email, sent by the authors. As this part of the study used a convenience sample, it is not possible to report participation rate. The expert clinician sample consisted of 4 women and 3 men with a mean age of 42.14 years (SD = 5.64), the majority of whom worked in a specialized clinic treating ARDs. Of the sample, 5/7 (71.4%) reported their highest level of qualification as holding a PhD and 2/7 (28.6%) reported having a Master’s degree in Clinical Psychology. The sample on average reported 12 years of post-qualification clinical experience (SD = 5.45).

The provisionally registered psychologist (trainee) sample consisted of 86 women and 18 men with a mean age of 30.76 years (SD = 8.32). To be included in the trainee sample, participants were required to 1) be currently enrolled in a professional postgraduate psychology training program; 2) be located in Australia; 3) have received training in CBT as part of their coursework. Of the sample 32/104 (30.8%) were enrolled in a PhD or combined PhD and master’s programme, 62/104 (59.6%) were enrolled in a Masters of Clinical Psychology programme and 10/104 (9.6%) were enrolled in a Masters of Professional Psychology programme. Participant characteristics for the expert sample and provisionally registered sample are outlined in Table 1.

Table 1. Participant characteristics for expert sample (N = 7) and trainee sample (N = 104).

	Expert sample		Trainee sample
Age (M, SD)	42.1 (5.6)		30.8 (8.3)
Gender (Female) (n, %)	4	57.1%	86	82.7%
Highest level of completedtraining (n, %)
PhD	5	71.4%	-	-
Masters	2	28.6%	-	-
Bachelor’s degree	-	-	104	100%
Course enrolled in (n, %)
PhD or PhD / Masters combined	-	-	32		30.8%
Master of Psychology (Clinical)	-	-	62		59.6%
Master of Professional Psychology	-	-	10		9.6%
Years of experience (M, SD)	12 (5.4)		-		-

As could be expected, results of an independent-sample t tests revealed that the two groups significantly differed in age [t(109) = −3.56, p = <.001]. The trainee sample was significantly younger (M = 30.76, SD = 8.32) than the expert sample (M = 42.14, SD = 5.64), representing a large effect (d = 1.74). One-way chi-square revealed no significant differences in gender χ (1) = 2.79, p = .09 between the two groups.

Measures

Demographics questionnaire

The demographic questionnaire asked participants to indicate their 1) gender; 2) age; 3) highest level of training; 4) registration status; 5) years of clinical experience; 6) theoretical orientation; and 7) clinical specialization (if fully registered as a psychologist).

CBT decision-making questionnaire for anxiety and related disorders (CDMQ-A)

The CDMQ-A was developed by both authors who have extensive experience in the assessment of ARDs, CBT for the ARDs, and the supervision of clinical psychology trainees delivering CBT. Both authors equally contributed to the development of the tool, with content drawn from typical presentations seen within university training clinics in Australia. An initial set of eight vignettes, each reflecting one separate diagnostic category was developed. A set of three questions for each vignette were then developed in order to assess clinical decision-making skills, utilizing a four-item multiple-choice response format (see, Figure 1 for a sample vignette). Additional case information was included in each of the three questions that followed the initial case vignette. Vignettes and associated questions were designed to broadly assess decision-making regarding assessment, cognitive and behavioural interventions. Additional questions were designed to assess knowledge of theoretical models, the importance of an evidence-based assessment, case formulation, intervention for subclinical symptoms, and stages of change. Together, questions were designed to assess key skills required to deliver a CBT intervention for common manifestations of ARD symptoms. Importantly, whilst diagnostic classification is utilized here as one factor that contributes to treatment decision-making, it is acknowledged that clinicians must utilize a formulation driven approach to clinical decision-making. Where possible, individual factors, such as cause and maintaining factors, are noted in the provided case information to reflect the importance of the case formulation approach

Figure 1. Sample clinical vignette and one corresponding questions / response options from the CDMQ-A.

.

Accurate responses were informed by scientific literature, evidence-based CBT treatment manuals and a CBT case formulation approach. Incorrect responses were derived from observations of common decision-making errors that trainees often make during their training, observed by the authors. The CDMQ-A underwent multiple reviews by authors and were then evaluated by six international clinicians with expertise in the ARDs, who were asked to rate vignettes and questions within each diagnostic category on face and ecological validity. Face validity was assessed by asking experts to identify, on a 10-point likert scale, how effective each measured what it purported to measure. Ecological validity was assessed by asking experts to identify, on a 10-point likert scale, the degree to which each case reflected clients seen in their clinical practice. Following expert feedback and review, one vignette and the associated three questions were discarded due to concerns with measurement of decision-making and ecological validity.

In the final version of the CDMQ-A, vignettes and questions related to 7 diagnostic categories were retained: 1) obsessive-compulsive disorder (OCD); 2) panic disorder (PD); 3) post-traumatic stress disorder (PTSD); 4) body dysmorphic disorder (BDD); 5) social anxiety disorder (SAD); 6) generalized anxiety disorder (GAD); and 7) agoraphobia (AG). Three-questions were presented for each vignette, resulting in a final 21-item measure. The questionnaire is scored by totalling correct responses to each diagnosis, with total scores ranging from 0 to 7. To achieve a perfect score, all three items linked to each vignette must be answered correctly. The CDMQ-A is available from the corresponding author upon request (see, Figure 1 for a sample vignette).

Procedure

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (Western Sydney University Human Research Ethics Committee; H12072) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study. The study was conducted in two parts. Part one consisted of online questionnaire completion by a sample of expert clinicians. Experts were identified and invited to participate in the online questionnaire via direct email which included a link to the participant information sheet, consent form, demographics questionnaire and the CDMQ-A (including additional questions relating to face and ecological validity for each vignette). The online questionnaires appeared in fixed order and took approximately 30 minutes to complete.

Part two of the study consisted of online questionnaire completed by provisionally registered psychologists (i.e., trainees). Participants were identified and invited to participate via email through postgraduate psychology training programmes and publically available list serves, and via advertisement on relevant social media websites. Emails and advertisements contained the study link which opened to the participant information sheet, consent form, brief demographic questionnaire and the CDMQ-A. The online questionnaires appeared in fixed order and took approximately 20 minutes to complete.

Data analysis

Assumption testing was undertaken prior to analyses. Due to small and unequal group sizes, non-parametric tests were used as appropriate. Independent-sample t tests and 2 × 2 chi-square tests were conducted to assess group differences between correct and incorrect responses provided by trainees and experts on each vignette. Reliability of the CDMQ-A was assessed with Cronbach’s alpha. Effect sizes were calculated and interpreted as suggested by Cohen (1992). Differences between group means were reported as: Small = .20, medium = .50, and large = .80 (Cohen, 1992). Effect size of correlations were interpreted as: Small = .10, medium = .30, and large = .50 (Cohen, 1992). Sensitivity and specificity were determined using a receiver operating characteristic (ROC) analysis. ROC analysis results were analysed in line with recommendations by (Šimundić, 2009); good = 0.70–0.80, very good = 0.80–0.90, excellent = 0.90 = 1.0. All analyses were conducted using IBM SPSS Statistical Software version 25.

Results

Validity

Consistent with prior research (Carpenter et al., 2016; Simpson et al., 2010), expert clinicians rated vignettes on both face and ecological validity on a 10-point Likert scale, as recommended by Nevo (1985). All vignettes included in the final questionnaire were reported as having acceptable face (M = 6.9; SD = 2.31; range: 0–10) and ecological validity (M = 7.41; SD = 2.24; range: 2–10), where acceptable was defined as receiving a score on this scale greater than six.

An independent-sample t test was conducted to investigate group differences on the total CDMQ-A between experts and trainees. Results indicated a statistically significant difference [t(10.63) = 6.9, p = .01]. The mean score on the CDMQ-A was higher in the expert sample (M = 6.29, SD = .76) than in the provisionally registered sample (M = 4.01, SD = 1.69), representing a large effect (d = 1.74). Differences between correct and incorrect responses provided by the experts and trainees on each vignette were examined using 2 × 2 chi-square. Results indicated that experts and trainees differed significantly on the vignettes related to OCD (χ² (1) = 6.03, p = .01), PD (χ² (1) = 8.18, p = .004), and GAD (χ² (1) = 5.29, p = .02). Overall, trainees were found to have made errors 61.5% of the time on OCD specific vignettes as compared to experts who made errors only 14.3% of the time; 55.8% of the time on PD specific vignettes as compared to experts who did not make any errors; and 44.2% of the time on GAD specific vignettes as compared to experts who did not make any errors.

Reliability

Cronbach’s alpha was .71, indicating acceptable internal consistency.

Sensitivity/specificity

Receiver operating characteristic (ROC) analyses were conducted in order to determine the diagnostic sensitivity and specificity of the CDMQ-A. The area under the curve (AUC) was .89 (95% CI: .79 – .97), suggesting very good diagnostic accuracy. A cut-score of 6 provided the best balance between sensitivity and specificity (sensitivity = .86; specificity = .24). The positive predictive power was .55 and the negative predictive value was .99.

Discussion

The CDMQ-A was developed to evaluate clinical decision-making in trainees relevant to psychological assessment and intervention for adults with ARDs. The aim of the present study was to evaluate the initial psychometric properties of the CDMQ-A in a sample of expert clinicians and provisionally registered psychologists. It was hypothesized that the CDMQ-A would effectively distinguish between expert and trainee clinicians, demonstrate high levels of reliability and validity, and high levels of sensitivity.

Expert clinicians reported acceptable face and ecological validity for the CDMQ-A items. Results found a statistically significant difference between results on the CDMQ-A by experts and provisionally registered psychologists and evidence of a significant association between expertise and total score on the CDMQ-A. Scores from the expert sample and trainee sample were significantly different on three of the seven disorder specific vignettes (GAD, OCD, and PD) and thus these vignettes may have the most utility in identifying competency in the treatment of ARDs, however further research is required.

The CDMQ-A demonstrated acceptable reliability (α = .71) and very good utility as a means of distinguishing between expert and provisionally registered psychologists (AUC = .89). While a finding of high specificity suggests that the CDMQ-A was able to correctly identify those who engage in accurate clinical decision-making, low specificity suggests that the tool had greater difficulty identifying incorrect clinical decision-making. Given this, the use of this measure should be used with caution accordingly. Further research is required to better understand specificity. However, it will be important to examine the convergent validity of the measure in future studies. Future studies may also wish to examine divergent validity. As this is the first study to investigate the psychometric properties of the CDMQ-A, it is important that future research replicate these findings. However, the results of the current study are consistent with studies that have used a similar methodological approach, when applied to a child anxiety population. For example, Carpenter et al. (2016), found that the ACE CARD was able to distinguish between an expert and trainee sample, had good internal consistency, and moderate utility as a means of distinguishing between expert and provisionally registered psychologists. Taken together, these studies provide preliminary support for the use of a vignette-based tool to assess clinical decision-making in clinical psychology trainees. To the authors’ knowledge, the CDMQ-A is the first tool of its kind to assess clinical decision-making with an adult anxiety population using a vignette approach.

The CDMQ-A represents an important step forward in the evaluation of clinical decision-making in psychology trainees specifically when working with adult patients with ARDs. Administration of the CDMQ-A allows training institutions to evaluate the application of knowledge to intervention. When deficits are identified through incorrect responses, additional training and resources can be provided specific to these domains, with the aim of further supporting the development of clinical decision-making skills in a closely supervised setting. Given that research has highlighted that clinical decision-making improves with experience (Ruscio & Stern, 2006; Spengler et al., 2009), which training clinicians do not as yet possess, a tool which has the ability to identify and remediate errors in clinical decision-making early in training is likely to further support effective implementation efforts. While the results are preliminary, the tool has several important implications for dissemination and implementation of evidence-based practice for the ARDs, identified as an area of need (Beidas et al., 2013; McHugh & Barlow, 2010). First, given that this tool is brief, easy to administer (paper and pencil format or online) and score, and requires minimal staff input, the CDMQ-A is likely more practical to use in busy training settings, meeting a key area of identified need (Muse & McManus, 2013). Second, by requiring the application of knowledge to assessment and treatment decision-making, this tool is designed to extend upon existing measures of knowledge assessment. Third, through the utilized question response format and removal of “buzz words” (such as “exposure hierarchy” or “cognitive restructuring”), the tool arguably provides a better assessment of comprehension of theory underpinning evidence-based clinical decision-making, and thus may not be prone to the bias experienced by other similar measures (Brosan et al., 2008; McManus et al., 2012). Fourth, while it requires further investigation, the tool may be able to evaluate the effectiveness of these training efforts by examining change after coursework devoted to the treatment of ARDs. Finally, the CDMQ-A may be used as a template for the development of decision-making tools for other common adult psychological disorders, such as depressive disorders, suggested as an area of need more broadly (McHugh & Barlow, 2010; Muse & McManus, 2013).

While overall findings of the present study provide preliminary support for the psychometric properties of the CDMQ-A, a number of limitations should be noted. Firstly, generalizability of the findings may be limited due to the sample including only provisionally registered psychologists practicing in Australia. It remains unknown whether cultural variations would impact the utility of this tool in other English-speaking countries. Future research may wish to further investigate the psychometric properties of this tool in both Australian and international contexts. Secondly, this study used only a small expert sample. Whilst past research has also utilized a similar sample size (Simpson et al., 2010), future studies may wish to administer this questionnaire to large samples of training, generalist, and expert clinicians. Thirdly, clinical decision-making is complex and the CDMQ-A acts only as one brief measure of clinical decision-making and does not evaluate the application of this clinical decision-making on assessment or intervention in clinical practice with real patients. The goal of the tool is to identify trainees who may be particularly lacking specific skills in CBT decision-making, which may require remediation, rather identifying those whose skills are superior. Future research may wish to compare the CDMQ-A to other practice-based measures of clinical decision-making such as the Therapy Process Observational Coding System (McLeod et al., 2013, 2015). Finally, as only one measure of clinical decision-making was developed in this study, administration of this questionnaire to the same population at various time points may yield practice effects. Importantly, future research should seek to examine the questionnaires sensitivity to change following training and potentially develop equivalent questionnaires measuring the same or similar clinical decision-making competencies.

Overall, findings of the present study build on the developing literature on the importance of clinical decision-making in psychological practice and provide preliminary support for the psychometric properties of the CDMQ-A. The development of tools that are brief, easy to administer, score and require minimal staff input have the potential to have significant implications for the dissemination and implementation of evidence-based practice, particularly within busy educational settings. Future research efforts should be made to continue to refine such tools, including with other clinical presentations, and evaluate its use in various settings and diverse samples.

Acknowledgments

The authors would like to acknowledge Prof Tanya Meade for her review of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the University of New England.