Introduction

Approximately one in three patients with a diagnosis of schizophrenia are considered “treatment-resistant”, defined as having persistent psychotic symptoms despite sequential trials of at least two antipsychotics1. Clozapine is the only evidence-based antipsychotic for treatment-resistant psychotic disorders2. However, despite substantiative empirical evidence for its clinical efficacy, clozapine remains underutilised2,3,4. At a patient level, the underutilisation of clozapine in treatment-resistant psychosis (TRP) is associated with decreased interpersonal and occupational functioning, decreased quality of life and increased caregiver burden and hospitalisation5,6. While clozapine underuse is likely to be multifactorial, one purported factor is clinician perception7,8. Specifically, studies have suggested that clinicians may overemphasis the negative aspects of clozapine treatment, particularly in comparison to patient views9. Systematic reviews exploring broader barriers for clozapine underuse have identified three broad domains, namely patient-related (e.g. mandatory haematological monitoring), clinician-related (e.g. prescribing experience) and administrative (e.g. service fragmentation) barriers10,11. Furthermore, previous surveys of psychiatrists working in adolescent and old age services within the United Kingdom (UK) have highlighted practical difficulties in prescribing and monitoring clozapine12,13. In contrast, the barriers to clozapine use in early intervention in psychosis (EIP) services to date has been relatively unexplored.

In the UK, EIP services provide a range of psychological, social, and pharmacological interventions to promote recovery from a psychotic episode14. Increasing evidence have shown that TRP is prevalent in almost a quarter of people with first episode psychosis (FEP) in EIP services5,15. Recently, greater emphasis has been placed on improving access to clozapine for patients with TRP across EIP services in England16,17. This is based on evidence that a delay in offering clozapine reduces the likeliness of treatment response18. Despite the aforementioned EIP service priority within the UK, prescription rates of clozapine in EIP services remains very low3. Health care providers have a critical role in improving clozapine utilisation rates in EIP services. We therefore sought to identify their views on perceived barriers and potential facilitators to clozapine use.

Methods

This was a mixed-methods study using a cross-sectional survey of clinicians in EIP services across England to determine the attitudes towards clozapine use in EIP services and the perceived barriers and facilitators to clozapine treatment.

Survey Instrument

An online survey was developed by a multidisciplinary team consisting of psychiatrists, pharmacists, psychologists and occupational therapists following guidelines for designing electronic survey designs19 and has previously been described20. In brief, the survey comprised of 15 items divided into three areas: respondent demographics (7 items) (e.g. what organisation do you work for?); attitudes toward clozapine use in EIP services (5 items) (e.g., do you feel patients prefer taking clozapine compared to other antipsychotics?); and perceived barriers and facilitators to clozapine use in EIP services (3 items) (e.g. what would be helpful in facilitating clozapine initiation in your EIP service?). Clinicians were asked to score six purported barriers to clozapine use within EIP services on a scale of 1 to 7 (1 representing the least important and 7 representing the most important). The potential barriers were selected based on a review of the literature9,10,21,22,23. Clinicians were then asked to categorise clozapine prescribing rates in patients eligible within their EIP service as either: (i) ‘overprescribed’ (ii) ‘adequately prescribed’, (iii) ‘under prescribed’ or (iv) ‘unclear’. Clinicians also answered two qualitative questions (i) what they thought were the current barriers to clozapine use and (ii) how clozapine offer can be improved in EIP services. The survey was completed online using Qualtrics XM (Provo, Utah, USA). A pilot study of 60 clinicians was initially conducted within a large mental health provider (South London and Maudsley NHS Foundation Trust) to assess item range, variance, content and clarity. The final version of the survey took a median of 7 minutes (interquartile range 5 to 11 minutes) to complete (survey available on request).

Recruitment and ethical considerations

Clinicians were recruited through EIP services across England. One hundred and nine EIP services in England were contacted via email and 35 services disseminated the survey to their clinicians. The survey was accessible for 12 weeks. Ethical approval was obtained from King’s College London (MRSU-20/21-25297) and individuals gave informed consent prior to beginning the survey.

Analysis

Quantitative analysis was conducted using R24. Survey items concerning the importance of perceived barriers for clozapine use in EIP services were described using means and standard deviations. A Mann-Whitney U test was used to compare identified barriers between prescriber’s and non-prescribers. The threshold for statistical significance was p ≤ 0.05. Bonferroni correction was performed to adjust for multiple comparison. Categorical responses were summarised using percentages. Open-ended responses regarding barriers and facilitators to clozapine use were analysed thematically25. Thematic categories were constructed inductively from response data by two independent researchers (EO and BM). They then met to compare themes and subthemes and derive consensus. Following creation of these themes and subthemes, the two researchers coded each response independently then met to discuss any discrepancies and derive consensus.

Results

Sample characteristics

E-mail invitations were sent to 35 EIP services, comprising approximately 700 clinicians. Of these, 193 accessed the survey and 123 (approximately 18% of all eligible clinicians) completed the survey from the 35 EIP services and were included in the analysis. Respondents consisted of 63 females (67%) and the mean age was 44 years. The most common professions were nurses (50%), doctors (19%) and psychologists (9%). In total, 31 (32%) of responders were prescribers. Participant characteristics are displayed in Table 1.

Table 1 Characteristics of Survey Respondents.
Full size table

Clinician attitudes and clozapine barriers

Based on their experience of working in an EIP service, 38% of clinicians responded that clozapine was ‘adequately prescribed’ and 37% of clinicians responded that clozapine was ‘under prescribed’ and 2.3% felt that it is ‘overprescribed’. A further 23% of respondents were ‘unclear’.

The most important perceived barrier according to clinicians was ‘patient concerns with side effects’ (mean = 5.7, SD = 1.6), followed by ‘monitoring requirements’ (mean = 5.4, SD = 1.7) and ‘clinician concerns with side effects’ (mean = 4.4, SD = 1.7) (see Fig. 1). The least important barrier was ‘clinicians’ difficulty in identifying suitable patients’ (mean = 5.7, SD = 1.5). There was no significant difference between prescribers and non-prescribers for identified barriers.

Fig. 1: Perceived barriers to clozapine treatment.
figure 1

Mean item scores and standard deviations reported. A higher score indicates a more important perceived barrier according to clinicians in EIP services (score ranges: 1–7).

Full size image

Qualitative results

Barriers

Three major themes were identified regarding perceived barriers to clozapine use (illustrative quotes in Table 2). The most frequently mentioned theme was administrative barriers (n = 35, 41%), with the subthemes of blood monitoring, clozapine initiation, patient information, resources and treatment pathways. The second most common theme was patient-related barriers (n = 33, 39%), with the subthemes of adherence, blood monitoring, family and carer attitudes, patient attitudes, side effects and stigma. Finally, for clinician-related barriers (n = 17, 20%), the subthemes were clinician attitudes, formulation, prescriber confidence, side-effects, treatment pathway and TRP identification.

Table 2 Representative quotes from thematic analysis on barriers to clozapine access in EIP services.
Full size table

Facilitators

Eight major themes were identified regarding how to facilitate clozapine use in EIP services (illustrative quotes in Table 3). The most frequently mentioned theme was resources (n = 31, 26%), followed by patient information (n = 24, 20%), training (n = 16, 14%), clinical pathways (n = 15, 13%), monitoring (n = 11, 9%), guidance for clinicians (n = 9, 8%), side effects (n = 8, 7%) and perspectives on clozapine treatment (n = 4, 3%). For resources, subthemes were facilitated clozapine initiation, community resources, dedicated service, monitoring, staffing and TRP identification. For patient information, subthemes were clozapine benefits and risks, patient-friendly information, testimonials and counselling. For training, subthemes were benefits and risks, family and carer training, accessibility, improving clinician knowledge and confidence. For clinical pathways, subthemes were TRP identification, timely and proactive monitoring of non-clozapine treatment response, communication between services, timely and proactive clozapine offer and a dedicated service. For monitoring, subthemes were optimised monitoring, point of care (POC) blood testing and antipsychotic adherence. For the theme guidance for clinicians, subthemes were clozapine initiation and TRP identification. For the theme side effects, subthemes were physical health, proactive management and interventions. For the theme perspectives on clozapine treatment, subthemes were addressing stigma and clinician attitudes. Illustrative quotes can be found in Table 3.

Table 3 Representative quotes from thematic analysis on improving clozapine access in EIP services, Treatment-resistant psychosis (TRP).
Full size table

Discussion

We undertook a national survey of clinicians practising in EIP services in England to examine the perceived barriers and potential facilitators to clozapine use. Over a third of clinicians reported that clozapine was under-prescribed. Based on thematic analysis, we identified 17 perceived barriers that could be broadly grouped into administrative, clinician-related and patient-related themes. Clinicians considered patient concerns about side effects as the most substantial barrier to clozapine, followed by monitoring requirements. Facilitators for improved clozapine use in EIP services included more guidance for clinicians, followed by optimised monitoring and education and training interventions.

To the best of our knowledge, this is the first study to have explored the perceived barriers and facilitators of clozapine use exclusively in EIP services. Consistent evidence has demonstrated widespread underuse and variation of clozapine prescribing and this is seemingly more prevalent in EIP services3. Barriers to improved clozapine treatment have been previously studied and can be broadly divided into patient, clinician and administrative10,11,26,27, which our survey also identified. A previous review on the barriers to the use of clozapine highlighted insufficient prescriber training; concern over rare, but serious side effects (including agranulocytosis and myocarditis); frequent blood tests; relatively high administrative burden; and side effects (including weight gain, sedation, and hypersalivation) as prominent barriers10. Another review on prescriber and institutional barriers and facilitators of clozapine similarly identified side-effects, monitoring requirements and clinician prescribing experience as major determinants of under-prescription11. We found that the barriers to clozapine use specifically in EIP were similar to the aforementioned systematic review, where the authors argued that the lack of practical knowledge with prescribing clozapine was the most likely cause of the negative beliefs surrounding the medication10.

Thematic analysis revealed perceived facilitators to improved clozapine use in EIP services. These were increased clinician training, addressing clinician attitudes towards clozapine and improved resources (including use of POC testing). Requests for education and training interventions to promote safe and effective use of clozapine have been made for years23,28,29. Nevertheless, there is currently no consistent approach to the training of clinicians for clozapine management. The implementation of training and learning from experienced clozapine prescribers are likely to increase clinician confidence11,20,22,30. Consistent with the broader literature, addressing clinician attitudes towards clozapine was considered an important facilitator by EIP clinicians11. Indeed, previous studies have demonstrated that clinicians tend to overstate patient reluctance with clozapine use31. For example, in a clinician survey by Hodge and Jespersen (2008), 52% of respondents estimated that patients were unhappy with the mandatory blood tests, compared to only 19% of the patients when directly asked9. Moreover, clinicians often state concerns with adherence to clozapine, despite evidence suggesting that adherence is comparable to other antipsychotics32. Consistent with existing literature, our study identified improved resources, and in particular use of POC testing, as potential facilitators to improved clozapine use10. Encouragingly, recent work within the UK has demonstrated the effective use of POC testing, including capillary clozapine levels and neutrophil monitoring33,34. The implementation of education and training interventions across EIP services in England alongside wider use of POC testing are potentially cost-effective methods of addressing clinician-related barriers to clozapine. Addressing administrative and structural barriers to improved clozapine, such as ensuring access to local clozapine outpatient clinics for patients, should also be a priority as this would likely have a synergistic effect on other interventions, such as pharmacy support and POC testing. Of note, in the Netherlands, the discontinuition of regular blood tests and institutional pressure (such as regular auditing or mandating detailed action plans for centres with low prescription rates) yielded the second highest recorded increase in clozapine use35,36. Notably, since these changes by the Dutch Clozapine Collaboration Group, the average outpatient prescription rates of clozapine in the Netherlands have increased by 88% between 2004 and 2019. Moreover, organisational changes such as education and quality improvement initiatives have led to significant increases in clozapine use in clinical services in New Zealand and New York37,38. Similar interventions in EIPs in the UK could conceivably lead to similar improvements.

Clinical implications

This study highlights several key areas of potential intervention to potentially improve clozapine treatment access for individuals with TRP in EIP services. Our study suggests that ensuring service provision is sufficient (including an accessible clozapine clinic or equivalent service) may remove a key administrative barrier to initiating and maintain clozapine treatment. Access to expertise, such as a “hub and spoke” model, may also help clinicians with clozapine management3. This is supported by evidence that most clinicians view clozapine prescribing as a burden39. Reducing the perceived cumbersome nature of clozapine monitoring by minimising the degree of physical health monitoring (whilst maintaining safety standards) through evidence-based guidelines alongside use of POC testing could reduce barriers to clozapine use, improve patient satisfaction and overall increase clozapine utilisation40. Our data highlights the potential benefit of ensuring professionals working within EIP services have the opportunity to develop their skills, knowledge and confidence in managing and prescribing clozapine treatment through education and training interventions, such as through e-learning20. This may also increase clinician confidence when communicating with patients and their families and caregivers around clozapine treatment, however studies are required to confirm this. Addressing misconceptions around clozapine use through training and patient testimonials is equally important. For example, delegating clozapine monitoring to advanced nurse practitioners did not improve clozapine prescription rates in a randomised trial in the Netherlands41. While the study was likely underpowered to detect a difference, the authors suggested that the lack of improvement may have been due to misconceptions about clozapine’s therapeutic potential among prescribing psychiatrists. In addition to education and training for clinicians, respondents suggested potential benefit of similar interventions for patients, family and carers. Future studies should investigate service user views on facilitators to improved clozapine use in EIP services.

Strength and limitations

To our knowledge, this is the first study to explore the perspectives of EIP clinicians on barriers and facilitators for increased clozapine use in EIP services. Further strengths of this study include the wide geographical coverage, and inclusion of a range of healthcare professionals. Another strength is the inclusion of both non-prescribers and prescribers. While prescribers decide treatment plans in collaboration with the patient, non-prescribers, such as mental health nurses, are often assigned the key role of coordinating patient care. Care coordinators maintain contact with patients, produce care plans and typically attend medical reviews with clients and prescribers. As such they are well-placed to flag which patients may be good candidate for a trial of clozapine.

A major limitation of our study is that we were unable to ascertain a precise number of the total number of clinicians from which we sampled. Our study is also limited by the low response rate of around 17%, increasing the risk of bias and potentially limiting the generalizability of our findings. Low response rates with surveys of clinicians is a common phenomenon42. However, we received responses from 35 of the 109 EIP services from all regions in England, indicating reasonable geographical coverage. A further important limitation is the absence of a patient perspective. Future work should seek the views of patients regarding the barriers and facilitators for improved clozapine treatment in EIP services.

Conclusion

In our study, the main perceived barriers to clozapine use in EIP services by clinicians are patient concerns and physical health monitoring. National policymakers should consider the development of clinical training, improving access to resources and guidelines, and consider the streamlining of physical health monitoring to minimise barriers and facilitate the use of clozapine within EI services. National policymakers, public health agencies and NHS Trusts should consider these findings to address the underutilisation of clozapine in EIP services across England and internationally to improve patient outcomes in treatment-resistant psychosis.