Newly graduated Australian psychologists’ interest and confidence in psychological approaches for psychosis: what role can psychology training providers play in increasing access to psychological interventions for psychosis?

ABSTRACT

Objective

Despite recommendations in international guidelines, implementation of psychological therapy for psychosis is limited. Clinician level factors (i.e. attitudes, confidence, and training) are significant barriers to implementation, but have not been investigated within the context of Australian professional psychology graduates. The current study aimed to examine predictors of interest and confidence in working therapeutically with people with psychosis in recent graduates of Australian psychology postgraduate training courses.

Method

An online survey was completed by 108 recent graduates from Australian postgraduate psychology courses. Participants provided data on hours of training, practicum experience, stigma, therapeutic optimism and levels of interest and confidence in working therapeutically with people with psychosis.

Results

Participants reported low levels of training and practical experience in psychological therapy for psychosis. Participants with higher training hours and who had delivered an intervention to someone with psychosis during training reported significantly more confidence in their ability to work with people experiencing psychosis. However, stigma was the only significant predictor of interest; psychologists with higher stigmatised attitudes were less interested in working with psychosis.

Conclusions

Increased opportunities for practicum experience with people with psychosis and interventions that address stigma in professional psychology training may support increased interest and confidence in providing psychological interventions for psychosis.

KEY POINTS

What is already known about this topic:

1. Psychological therapy (particularly CBT) is a recommended, evidence-based intervention for psychotic symptoms.

2. Despite this, most people experiencing psychosis do not get offered psychological therapy.

3. Clinician factors such as training and attitudes are often barriers to implementing psychological therapy for psychosis.

What this topic adds:

1. In Australia, psychologists are well placed to deliver psychological interventions to people with psychosis, however no research has examined factors that might predict psychologists’ interest and confidence in this work.

2. Recently graduated psychologists with higher stigmatised attitudes to people with psychosis are less likely to have an interest in this work. Psychologists who have had more teaching hours and relevant practicum experience are more confident in working with people with psychosis.

3. Increased opportunities for practicum experience with people with psychosis and interventions that address stigma in professional psychology training may support increased interest and confidence in providing psychological interventions for psychosis.

KEYWORDS:

• Cognitive behavioural therapy
• implementation
• psychologist
• psychosis
• schizophrenia
• training

Research shows that 1% of the population will experience a psychotic disorder in their lifetime, with the associated symptoms causing significant distress for the individual and their families (National Institute for Health and Care Excellence [NICE], 2014). Many individuals experience persistent symptoms over long periods of time despite pharmacological intervention (Leucht et al., 2017). International guidelines recommend that cognitive behavioural therapy for psychosis (CBTp) is offered to individuals who are at risk of developing psychosis, during first or subsequent episodes of psychosis, or during the recovery period for those with persistent symptoms (Galletly et al., 2016; NICE, 2014). CBTp is an evidence-based psychotherapy that aims to reduce the distress associated with psychotic symptoms and improve functioning, through intervention at the cognitive and behavioural level. Specifically, CBTp typically involves collaborative development of a shared formulation to inform understanding of the development and maintenance of the psychotic symptoms, psychoeducation, challenging unhelpful thoughts and beliefs that mediate emotional and behavioural responses, normalisation and acceptance of the psychotic experiences, development of adaptive coping skills, relapse prevention planning, and management of comorbid conditions such as social functioning impairments and substance abuse (Morrison, 2017)

A number of meta-analyses have found CBTp to be effective for managing positive symptoms of psychosis, with effect sizes ranging from 0.34 to 0.44 (Turner et al., 2020; Wykes et al., 2008). While effect sizes are in the small range, particularly when compared to active control treatments, CBTp is currently the most efficacious psychological treatment for psychotic symptoms. Recent refinements and innovations in CBTp protocols are also finding increased effect sizes in robust treatment trials (Craig et al., 2018; Freeman et al., 2021). Despite being recommended in international guidelines and having robust empirical evidence, implementation of psychotherapy for psychosis is poor. In the context of Australia, data from the second Australian national survey of psychosis showed that only 10% of people living with psychosis reported receiving evidence-based CBTp during a year of treatment in specialist mental health services (Harvey et al., 2019).

Due to the low provision of CBTp, research has investigated potential barriers impacting implementation. The findings have identified that barriers may exist in three main areas, including client factors (i.e., symptomology and phase of illness; Switzer et al., 2019), service-provision factors (i.e., lack of referral pathways, limited availability of CBTp clinicians; Avasthi et al., 2020), and clinician factors (i.e., lack of confidence, knowledge, and training; Avasthi et al., 2020). The clinician and service-provision factors have been found to hold more predictive value than client factors for the implementation of CBTp (Greenwood et al., 2018; Harvey et al., 2019). A systematic review of 26 studies exploring CBTp implementation in the United Kingdom (Ince et al., 2016) found that clinician barriers such as a lack of training were most frequently mentioned as obstacles to implementation, followed by negative beliefs and attitudes regarding psychotherapy for psychosis and a lack of confidence in delivering the interventions. Similarly, a cross-sectional survey gathering quantitative and qualitative data on Australian and Canadian mental health clinicians’ (Psychologists, Psychiatrists, Nurses, Occupational Therapists and Social Workers) attitudes towards delivering CBTp, found that training in CBTp and social norms (i.e., access to a supervisor who endorsed and supported implementation of CBTp) were significantly associated with the delivery of CBTp (Lecomte et al., 2018).

An attitude that has been highlighted as an important factor in predicting the implementation of CBTp, is the perception of the intervention’s effectiveness, also known in the literature as Therapeutic Optimism. Therapeutic Optimism refers to clinicians’ self-reported, specific expectancies regarding patient outcomes in a clinical setting (Byrne et al., 2006). The research conducted by Lecomte et al. (2018) found two themes. The first theme related to the attitude that their clients were too ill to benefit from CBTp intervention, and the second related to the efficacy of the intervention; that it was too brief or other interventions (mainly psychopharmacology) were better than CBTp. Quantitative data from Lecomte et al. (2018)’s survey, also found that positive attitudes towards CBTp were significantly associated with clinician intention to deliver the intervention. In addition to the individual attitudes identified, collective attitudes (i.e., social norms), held by colleagues and supervisors were also found to be a significant factor in the participants’ intention to deliver CBTp (Lecomte et al., 2018).

Another important attitude to consider in this area is stigma. Stigma refers to the devaluation of, and negative attitudes towards an individual (Stier & Hinshaw, 2007). Stigma surrounding mental illness, held both by the general population and healthcare professionals, has been consistently found to be a barrier to accessing appropriate healthcare treatment (Knaak et al., 2017). Additionally, research has found that the stigma held towards different mental health care diagnoses is variable, with psychotic disorders being found to be perceived more negatively than other mental health diagnoses (Wood et al., 2014).

Professional psychologists are well placed to be at the forefront of delivering psychological interventions to people experiencing psychosis and have generally been considered the profession with the expertise to do so, however research has not yet specifically explored the interest and confidence of psychologists regarding the use of psychological approaches in psychosis, particularly in an Australian context.

Aims

The current research aimed to examine how training and attitudes predict recently graduated Australian psychologists’ interest and confidence in psychological interventions for psychosis.

Hypotheses

Based on the previous research, it was expected that hours of teaching in psychological therapy for psychosis, direct client practicum hours with clients experiencing psychosis, access to a supervisor with specialist knowledge of psychological therapy for psychosis, personal exposure to someone experiencing psychosis, clinician’s stigma, and clinician’s optimism regarding psychological therapy for psychosis, would predict graduates’ interest and confidence in working with people who experience psychosis.

Method

Design

The study was cross-sectional in design. Participants completed an anonymous survey with questions targeted at understanding their interest and confidence in working psychologically with people who experience psychosis, as well as factors that may predict these outcomes (e.g., training and experience).

Participants

An a priori power analysis was calculated using G*Power (Faul et al., 2007). The sample size for the current study (n = 118) was considered adequate given the analysis indicated that 92 would be the required sample to detect a medium effect size (ƒ² = 0.15) using standard alpha (α = .05), power of .80, and 5 predictor variables in a linear regression analysis.

Participants were recently graduated psychologists from Australian postgraduate professional psychology programs. To be included in this study, participants must have completed an accredited postgraduate professional psychology degree from a university within Australia in 2021.

Participants were recruited through contacts at postgraduate psychology courses that provide an accredited postgraduate course in professional psychology (Master of Professional Psychology, Master of Psychology (Clinical), Doctorate in Clinical Psychology) and through advertisements through the Australian Psychological Society, and on social media. Participants were recruited between January 2022, and August 2022.

Measures

Demographic measures

Demographic information included gender, age, degree type undertaken (e.g., Master of Clinical Psychology, Master of Professional Psychology or a Doctorate of Psychology), location (state) postgraduate course was undertaken, hours of training in psychotherapy for psychosis, intervention delivery experience, hours of direct client contact with people experiencing psychosis on practicum placements, and access to specialist supervision during placement.

Interest and confidence

Participants’ interest and confidence in working with people who experience psychosis was measured using single item, 10-point Likert-type scales, ranging from 1 (would not see if I had a choice) to 10 (interested in specializing in the area) for the interest scale and 1 (not at all confident) to 10 (very confident) for the confidence scale. These scales were based on items from a study investigating similar variables (Koder, 2008).

Social distance scale

Stigma regarding people who experience psychosis was measured using the Social Distance Scale (SDS; Link et al., 1987), which measures the participant’s willingness to interact with people with mental illness across a variety of situations. For the current study, questions were altered from referencing mental illness in general, to specifically someone who experiences psychosis. The scale has 7 items, measured on a 4-point Likert scale, ranging from 0 (definitely willing) to 3 (definitely unwilling). Possible scores range from 0 to 21, with higher scores representing a greater desire to distance oneself from persons who have psychosis. The SDS has good construct validity and reliability overall (α= .85; Fernandez et al., 2015), and within the current sample (α= .83).

Therapeutic optimism scale

Clinician attitudes regarding psychological interventions for psychosis were measured using the Therapeutic Optimism Scale (TOS; Byrne et al., 2006), which is a self-report measure of clinician optimism in respect to the expectations a clinician has about their patient’s treatment outcomes. For the current study, questions were altered from referencing mental disorder in general, to specifically refer to someone who experiences psychosis. The scale has 10 items (4 of which are reverse scored), measured on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Possible scores range from 10 to 50, with higher scores representing a greater therapeutic optimism. The TOS has acceptable construct validity and reliability overall (α = .68; Byrne et al., 2006), and within the current sample (α = .77).

Procedure

Ethical approval to conduct the research was obtained from the University of the Sunshine Coast Human Research Ethics Committee (HREC; S211659). Participants were directed to an online survey (Qualtrics^TM) and were provided information about the study. They were then asked to provide digital informed consent prior to participating in the study. Those who did not provide consent exited the platform while those who did proceeded to the survey. Participants were able to exit the survey at any point up until the submission of their survey responses.

Statistical analyses

Statistical Package for the Social Sciences (SPSS; Version 27) program was used for all statistical analyses. To assess whether training hours, intervention delivery, personal experience, stigma, and therapeutic optimism were associated with interest and confidence, a multiple regression analysis was conducted using the Enter method. To determine if there was any difference on any variables between those participants enrolled in a Master of Professional Psychology or Master of Clinical Psychology, an exact sampling distribution for Mann–Whitney U test (Dinneen & Blakesly, 1973) was applied for the scaled variables, and a 2 × 2 Chi Square analysis was applied for the categorical variables. Participants who had completed a different degree were excluded from this analysis due to the low numbers in this group. A non-parametric Mann–Whitney U test was considered the most appropriate for this analysis due to the univariate issues with normality in the continuous variables (i.e., hours of training, confidence, interest, stigma, and therapeutic optimism). Due to these issues with univariate normality, descriptive statistics are also reported using median and interquartile range.

Results

The participants included in the study ranged in age from 23 to 62 years (M = 30.79, SD = 8.1). Figure 1 shows the flow of participants through the study, 205 consented to take part, and 118 were included in the final analysis. Table 1 shows the sociodemographic characteristics of participants included in the analysis.

Figure 1. Participant flow-chart.

Table 1. Sociodemographic characteristics of participants (N = 118).

Characteristic
	n	%
Age
23–32	86	72.90
33–42	18	15.30
43–52	11	9.30
53–62	3	2.50
Gender
Female	102	86.40
Male	14	11.90
Non-Binary	2	1.70
State
QLD	33	28.00
VIC	31	26.30
NSW	26	22.00
WA	12	10.20
ACT	9	7.60
TAS	4	3.40
NT	2	1.70
SA	1	0.80
Degree
Master of Clinical Psychology (MCP)	54	45.80
Master of Professional Psychology (MPP)	48	40.70
Combined PhD/Master of Clinical Psychology	2	1.70
Other^a	14	1.90

^aIncluded the following degree types: Master of Educational and Developmental Psychology (n = 6); Master of Community Psychology (n = 2); Master of Counselling Psychology (n = 2); Doctor of Philosophy (Clinical) (n = 3).

Preliminary analysis

Eleven outliers with implausible data were removed from the dataset prior to the final analysis. Visual inspection of histograms and p-P plots was conducted to confirm the assumption of approximate normal distribution of model residuals. Due to non-normal model residuals for the confidence variable, a square root transformation was applied to this variable. Collinearity was not considered problematic as all Variance Inflation Factor (VIF) scores were ≤10, Tolerance values were ≥ .02. No values for Cook’s distance were above one. Five cases had missing data. Missing data was handled with Listwise deletion, due to being only a small portion (>5%) of the total sample size (van Buuren, 2018).

Descriptive statistics are presented in Table 2. The results indicated that 74.6% (n = 88) of respondents had not delivered an intervention to people experiencing psychosis during their training. Participants reported a median of 3/10 for their level of confidence in delivering interventions to people with psychosis, where lower scores indicated less confidence. However, median interest levels were 5.5/10, where higher scores indicated more interest in working with people with psychosis in their careers. Participants received on average (median) three hours of training in psychological interventions for psychosis during their postgraduate studies. Participants mean score on the social distance scale indicated that the sample had low average levels of stigma towards people with psychosis, with the overall score falling below the midpoint of possible scores. Therapeutic Optimism was also high in our sample, with the mean score falling well above the midpoint.

Table 2. Descriptive statistics and significance tests across MPP and MCP degrees (N = 114).

Variable (Possible range)	Overall		MPP		MCP		Z Value	P Value
	M	IQR	M	IQR	M	IQR
Hours of Training	3	4	2	4	3	5	−1.89	.058
Interest (1–10)	5.5	4	6	2	5	4	−2.04*	.041
Confidence (1–10)	3	3	3	2	4	4	−1.92	.055
Social Distance (0–21)	9	6	9	6	9	6	−.64	.525
Therapeutic Optimism (10–50)	40	6	40	6	40	6	−.81	.418

M = Median; IQR = Interquartile Range; MPP = Master of Professional Psychology; MCP = Master of Clinical Psychology; Z value = Range of Mann–Whitney U test across the 5 datasets; *indicates a significant Z value.

There was a significant difference in participant level of interest in delivering psychological interventions to people with psychosis across the different degree types. Median interest scores were significantly higher in Master of Professional Psychology (MPP) participants than in Master of Clinical Psychology (MCP) participants. There was no significant difference in intervention delivery, hours of training, or levels of confidence, personal experience, social distance or therapeutic optimism between degree types.

Main analyses

Interest

As shown in Table 3, the model accounted for significant variability in interest levels of working with people experiencing psychosis, R² = .19, F(5, 106) = 5.075, p < .001. By Cohen’s (1988) conventions, a combined effect of this magnitude is medium (ƒ² = .23). Of the independent variables, stigma was a significant predictor; a one-point increase in stigma led to a .216 decrease in interest scores. That is, increased stigma (as measured by desire to be socially distanced from people with psychosis) was associated with decreased interest in wanting to work with individuals who experience psychosis. All other independent variables did not contribute significantly to the variance in the level of interest in working with people with psychosis.

Table 3. Multiple regression analysis predicting interest from IVs.

Interest	B	95% CI		SE B
		LL	UL		β	R^2	∆ R^2	P
Model						.19	.16	<.001
Constant	3.910	−.53	8.35	2.24				.084
Personal experience	.117	−.33	.56	.05	.05			.746
Intervention delivery	.081	−.98	1.14	.53	.02			.879
Training hours	−.015	−.10	.08	.05	−.03			.603
Social Distance	−.216	−.34	−.09	−.34	−.34			<.001
Therapeutic Optimism	.080	−.02	.18	.16	.16			.108

Model = “Enter” method in SPSS Statistics; B = unstandardised regression coefficient; CI = confidence interval; LL = lower limit; UL = Upper limit; SE B = Standardised error of the coefficient; β = standardised coefficient; R² = coefficient of determination; ∆ R² = Adjusted R².

Confidence

As shown in Table 4, the model accounted for significant variability in participant confidence in their ability to work with people experiencing psychosis, R² = .32, F(5, 106) = 10.130, p < .001. By Cohen’s (Cohen, 1988) conventions, a combined effect of this magnitude is large (ƒ² = .47). Of the independent variables, intervention delivery and training hours were the strongest predictors. That is, people who had delivered an intervention to someone with psychosis during their training and who had more hours of training in psychological approaches for psychosis, showed increased confidence in their ability to deliver psychological interventions to people with psychosis. All other independent variables did not contribute significantly to the variance in confidence scores.

Table 4. Multiple regression analysis predicting confidence from IVs.

Confidence	B	95% CI		SE B
		LL	UL		β	R^2	∆ R^2	P
Model						.32	.29	<.001
Constant	.762	−.16	1.69	.47				.105
Training hours	.026	.01	.04	.01	.23			.007
Intervention delivery	.415	.20	.64	.11	.33			<.001
Personal life	.079	−.01	.17	.05	.14			.092
Social distance	−.018	−.04	.01	.01	−.13			.155
Therapeutic optimism	.012	−.01	.03	.01	.10			.257

Model = “Enter” method in SPSS Statistics; B = unstandardised regression coefficient; CI = confidence interval; LL = lower limit; UL = Upper limit; SE B = Standardised error of the coefficient; β = standardised coefficient; R²= coefficient of determination; ∆ R² = Adjusted R².

Discussion

Our study is the first to explore predictors of interest and confidence in the delivery of psychological interventions in psychosis in recently graduated psychologists in Australia. The findings provide insights into the current landscape of training in psychotherapy for psychosis for Australian psychologists as well as examining potential barriers to implementation (through examining variables that predict interest and confidence in working with this population).

Findings indicate that the majority of psychologists (75%) in this study finished their postgraduate training without having an opportunity to work therapeutically with people with psychosis. Furthermore, out of those participants that did have an opportunity to work with this population, only 15% (n = 18) of the whole sample had access to a supervisor familiar with working with people with psychosis. Additionally, respondents received an average of only three hours of training on interventions for psychosis during the course of their postgraduate degree. Considering Master of Clinical Psychology and Master of Professional Psychology courses in Australia typically include 6 and 12 coursework courses respectively, each averaging 150 learning hours each, these hours are low. The median level of interest (M = 5.5/10), and confidence (M = 3/10) in delivering interventions to people with psychosis, were also relatively low. Interestingly, therapeutic optimism was high in our sample, with the mean score falling well above the midpoint. Further, the sample had low levels of stigma towards people with psychosis, with the overall score falling below the midpoint of possible scores.

Importantly, the results indicated that the higher the psychologist’s desire for social distance from people experiencing psychosis (i.e., stigma), the lower levels of interest they held for working with this population in the future. Additionally, psychologists who had provided an intervention to people with psychosis during their training and those who had more teaching in intervention delivery for people with psychosis had significantly higher levels of confidence in the delivery of psychological interventions for psychosis.

Interestingly, our findings contrast somewhat from previous research (Eisen et al., 2022; Sivec et al., 2020) in that training hours were not found to significantly change the levels of interest. This also contrasts with previous findings in other areas of psychology specialisation, for example increased training in working with older adults led to increased interest in work in this area (Koder, 2008). This could be explained by the low teaching hours reported in our sample (M = 3), as those studies findings significant improvement in interest levels included a minimum of 12 hours of training (Eisen et al., 2022). These findings did however align with previous research that has found that specific teaching in CBTp increased clinicians’ confidence in their ability to work with this client group (Eisen et al., 2022; Sivec et al., 2020).

The findings also indicate that delivering an intervention to someone with psychosis had no effect on interest levels, meaning that individuals who had the opportunity to work with this population group did not report significantly different levels of interest in working with this population group in the future. A possible explanation for this, is that only 62.10% (n = 18) of those who delivered an intervention in our sample had access to a supervisor who they described as having expertise in the area. This is in keeping with previous findings that social norms can be an implementation barrier, with clinicians more likely to have interest in implementing an intervention if they are within a culture that values and supports this (Lecomte et al., 2018). Delivering an intervention to someone with psychosis is however a large contributor to newly graduated psychologists’ confidence in their ability to deliver psychological interventions to this population group. This aligns with previous findings in in other areas of psychological practice that also experience implementation issues, such as Geropsychology (Koder, 2008).

The findings also indicate that the more stigma the psychologist held towards individuals with psychosis, the less interest they had in working with this population group in the future. This finding aligns with previous research conducted in barriers to mental health services overseas, that found stigma to be a barrier for individuals with mental illness to access appropriate healthcare services (Gronholm et al., 2017; Knaak et al., 2017). However, previous literature has predominantly focused on stigma impacting access to healthcare services due to the client factors (e.g., delayed help-seeking, discontinuation of treatment, etc). However, our study extends this, highlighting that stigma is also a barrier for access to appropriate healthcare, through reducing the likelihood of clinicians willing to work with people experiencing psychosis (Gronholm et al., 2017; Knaak et al., 2017). Therefore, the findings here suggest that stigma is an important target in improving Australian psychologists’ interest and confidence in delivering psychotherapy for psychosis.

Finally, psychologist’s positive attitudes towards psychotherapy for psychosis (i.e., therapeutic optimism) were not found to be a significant predictor of interest and confidence in working with individuals experiencing psychosis, which is in contrast to previous findings (Ince et al., 2016; Koder, 2008; Lecomte et al., 2018). One explanation for this could be that our sample had high levels of therapeutic optimism, with a small amount of variance. Additionally, it could also be that our findings indicate that therapeutic optimism has less of an impact on newly graduated psychologists, and that training hours, experience and stigma play a more important role for this specific group.

Study strengths and limitations

The findings here should be interpreted alongside consideration of some shortcomings in the study design. Firstly, the cross-sectional design impacts our ability to make causal inferences (Savitz & Wellenius, 2022), as we cannot rule out that people with more interest and confidence in the area seek out placements that allow them to work with people experiencing psychosis, or seek out courses with more teaching in the area. Further, the sample was not completely random, as they self-selected to participate, thus, it is possible that our participants had more favourable views and more interest in working with people experiencing psychosis and are not completely representative of the whole cohort. It is also possible that participants may have had a positive reporting bias to some of the measures used in the study due to perceptions about the agenda of the researchers, particularly the measures of stigma and effectiveness of psychological therapy. Further, the participants were largely from Queensland, Victoria and New South Wales, and therefore these results may not adequately capture the current landscape of newly graduated psychologists from the other states in Australia.

Implications

Although Australian postgraduate courses need to cover many clinical areas, psychosis (which, despite being a relatively low prevalence disorder, contributes significantly to the burden of disease in Australia) appears to be underrepresented in regards to both the hours that are being allocated to it in teaching, and the opportunities to work with this client group on placements. Given that teaching hours and practicum experience were significant predictors of confidence in delivering psychological interventions for psychosis, increasing the training hours and opportunities to implement interventions for psychosis on placements within Australian Postgraduate Psychology programs may be an important target in improving the poor implementation rates of psychotherapy for psychosis within Australia. Furthermore, due to the significant impact that stigma was found to have on our sample’s level of interest in working with people experiencing psychosis, it may also be beneficial to include an anti-stigma intervention that will reduce the level of stigma in newly graduated psychologists (Reddyhough et al., 2021), thus potentially increasing implementation through increased interest in this area of work.

Conclusion

There is currently limited research investigating Australian psychology graduates’ attitudes regarding interventions for psychosis, and this was the first study to provide insight into the current landscape. Results revealed that hours of training and opportunities to provide intervention to people with psychosis during the training program were low. The results also indicate that these areas significantly predicted higher levels of confidence in working with this population. The results also found that psychologists with higher levels of stigma were less interested in working with people experiencing psychosis throughout their career.

The findings have important practical implications, as the present study highlights the importance of ensuring Australian postgraduate programs provide training on interventions for psychosis, as well as opportunities to deliver interventions while on practicums to ensure that the psychology graduates have increased confidence in their ability to work with the population group. Additionally, the findings also highlight the importance of reducing stigma through incorporating education programs into the postgraduate programs.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of this study are available from the corresponding author, RB, upon reasonable request.