20 years versus 20 minutes: increasing the understanding of delusions from a cognitive behavioural perspective in two older adult community mental health teams

Abstract

Abstract

The last two decades have seen considerable development in our understanding of delusions, with medical perspectives broadening into more holistic bio-psychosocial models. Of particular relevance to services is the increasing evidence base for cognitive behavioural models and treatments. A number of national initiatives have aimed to ensure these developments are reflected in services. Primarily these have targeted adult services, yet delusions are no less common in older adults. Across two older adult services this study searched over 10,000 clinical notes to assess the number of individuals experiencing delusions, explored the perspectives of 27 staff members regarding their clinical approaches to working with delusions, and delivered a brief training on cognitive behavioural models and treatments of delusions. Three service users contributed to the training content. Endorsement of theoretical perspectives regarding delusions were measured pre- and post-training. Results showed over 20% of patients in the service may be experiencing paranoia or delusions, but only a handful were referred for CBT. Of 27 staff participants, less than half had received training on working with delusions, most of which was from as many as 20 years ago. Nearly all reported that a lack of knowledge prevented them from considering psychological perspectives. A quarter of staff across the two teams attended the training. No immediate changes in outcome measures were seen immediately post-training, with a striking diversity in approach reported by different members of staff. All staff qualitatively reported finding the training valuable, although engagement was dampened by high levels of stress and burnout.

Key learning aims

1. (1) To understand the developing cognitive behavioural evidence base for understanding and treating delusions in older adults.

2. (2) To understand the need for this growing evidence base to be reflected in services, including older adult services, not only in structured interventions offered by psychological therapists, but also in patient interactions and care planning by all professions working in multi-disciplinary teams.

3. (3) To understand some of the key challenges in assessing and improving current understanding of cognitive behavioural approaches to delusions within services.

Introduction

Cognitive behavioural approaches to delusions

There have been significant changes in the way we understand psychosis in the last two decades (Health Education England, 2020). Historic understanding of psychosis being a chronic biological illness treatable only via medication has broadened into a more patient-centred understanding of psychosis as an umbrella term for a number of different experiences that exist on a spectrum of severity within the general population much like anxiety and depression (Garety et al., 2001; Johns and Van Os, 2001). Delusions are one key psychotic experience, and evidence-based cognitive models have now been developed that posit a number of common, understandable, and treatable factors that cause and maintain paranoia and delusions, such as sleep dysfunction, negative cognition, high levels of worry, and reasoning biases (Freeman, 2016; Garety and Freeman, 1999).

Based on these models, increasingly bespoke and targeted cognitive behavioural therapies are showing new ways of improving delusions, reducing distress, and increasing quality of life for individuals with psychosis (Bighelli et al., 2018). Most recently, the Feeling Safe Programme, a theoretically driven cognitive therapy, showed the largest treatment effects ever seen for individuals with persistent persecutory delusions when tested in a randomised controlled trial against another psychological intervention (befriending) (Freeman et al., 2021). Work is ongoing to consider whether cognitive behavioural treatments should be considered more frequently without the use of medication at all (Morrison et al., 2020), particularly given concerns about the negative side-effects of anti-psychotics such as weight gain, drowsiness, and movement disorder among many others. Many of these side-effects are also of particular risk to older adults (Leon et al., 2010).

Psychosis and delusions in older adults

Despite psychosis having a peak of onset during adolescence and early adulthood, a considerable proportion of people first experience psychosis later in life, with some estimates suggesting a quarter of first episode psychosis occurs in those aged over 65 (Mitford et al., 2010). This may be for a number of reasons, including similar social and environmental triggers posited for the development of psychosis at any age, as well as more unique cohort stressors such as grief, retirement, loss of independence, and changes to role investments in later life, as well as cohort beliefs that may lead to a lack of disclosure of difficulties until older adulthood (Laidlaw et al., 2004). Psychotic experiences more generally are also common to a number of disorders more prevalent in older adults, such as Parkinson’s disease and stroke (Rootes-Murdy et al., 2022).

Older adults experiencing psychosis typically do not have the same access to care as younger adults (Mitford et al., 2010), yet the cost of caring for older adults experiencing psychosis is simultaneously higher than for younger age adults with psychosis (Gallagher-Thompson et al., 2008). We also know that for other conditions like anxiety and depression, older adults are less likely to seek care, but that when they do, they see recovery rates that often exceed their working age counterparts (Saunders et al., 2021). The number of older adults experiencing and needing treatment for psychosis is expected to increase dramatically given our ageing population, thus a greater focus on how to care for older adults with psychosis is essential. There is also evidence that paranoia and delusions in particular may be even more prevalent and severe in older onset psychosis compared with younger populations with psychosis. This may be due to the overlap with increasing cognitive decline and neurodegenerative diseases that lead to disorders such as dementia (Suen et al., 2019). For example, there is some evidence that delusions and for example, Alzheimer’s disease, may have some shared biological aetiology (Reeves et al., 2012). Conversely, there is very limited evidence to support the use of anti-psychotic medications in late-onset psychosis, suggesting that psychological treatments might be an important avenue of exploration, regardless of whether the psychosis is considered primary or secondary to another disorder (Reinhardt and Cohen, 2015).

Implementation into services

Given the development of cognitive behavioural approaches to delusions, there is recognition of the need to ensure our mental health services incorporate and reflect this evidence base. Most obviously this includes psychological therapists being trained in the delivery of CBT for psychosis, including the more specific interventions like the Feeling Safe Programme. However, when first accessing a service, individuals are typically seen by Community Mental Health Teams (CMHTs) made up of a diverse range of professionals. There are therefore a number of more nuanced ways that incorporating cognitive behavioural approaches into multi-disciplinary teams (MDTs) could be of benefit. For example, being aware of cognitive explanations of delusions could encourage staff to gain helpful additional information at general assessment. This might include asking more detail about the content and distress associated with delusions rather than challenging the unusual beliefs or viewing them as an epiphenomenon of the psychosis that is safest left alone, as might have been encouraged in earlier models of delusions (Berrios, 1991).

A number of national initiatives have begun trying to incorporate more psychological models of delusions into mainstream mental health care. The NHS Long Term Plan (NHS England, 2019b) and NHS Mental Health Implementation Plans (NHS England, 2019a) set out new models for more integrated and personalised community mental health care, including increased access to psychological therapies and upskilling the NHS workforce to deliver better care for individuals with psychosis. More specifically, Health Education England have also commissioned a two-day training on ‘Understanding Psychosis and Bi-polar disorder’ for CMHTs (Health Education England, 2020). The training covers a broad range of social and psychological perspectives on psychosis, including, but not limited to, cognitive behavioural models. This training is available to older adult services although adaptations specific to the needs of older adults with psychosis are not included in the training package.

There is limited in-built evaluation of the training package’s impact beyond brief pre and post ratings of staff knowledge and confidence, and indeed across the literature there is limited recent examination of CMHT staff perception of either their training needs or the real-world impact of training that is provided. There is, however, general evidence that healthcare staff can lack confidence in working with those experiencing psychosis (Copello et al., 2012; Kramarz et al., 2021; Yung et al., 2003) and that given the diverse professional background of staff working in CMHTs, knowledge of empirical psychological models is often not standard (Brooker, 2001). Evidence from some years ago does suggest some positive outcomes from staff training in this area, however (Laube Higson, 2000). For instance, training Community Psychiatric Nurses to deliver psychosocial interventions for psychosis led to a reduction in patients’ positive and negative symptoms (Brooker et al., 1994), with similar studies showing significant overall NHS cost savings for those services (Brooker, 2001; Brooker et al., 1997). On the other hand, despite these positive outcomes, these studies raised questions over whether services had the capacity to have non-psychology staff delivering these interventions, and whether it led to poorer care being delivered in other areas. Certainly, within our service, staff shortages and pressures meant there would be no capacity for staff to be delivering more interventions. A more feasible alternative can instead be to offer training that increases staff knowledge and confidence to approach clinical work while holding cognitive behavioural perspectives in mind to guide staff–patient interactions and care planning, rather than training in the delivery of full interventions.

The current study

The present study took place across two multi-disciplinary Older Adult CMHTs (OA CMHTs) serving one county within an NHS trust. These services provide care as close to home as possible for adults aged over 65 with severe mental health needs. In each service there is also a Step 4 team providing psychological therapies, predominantly CBT. Psychology staff within one of the teams had identified a need for increasing CMHT staff’s understanding of and confidence to work with people experiencing severe paranoia or delusions, particularly from cognitive behavioural perspectives, echoing the need that has also been identified and reflected at a national level.

The present OA CMHTs were not commissioned to receive the HEE Psychosis and Bi-polar training, and if they were to receive it in future, considerable staff shortages would likely preclude many from attending given its two-day length. Consequently, a smaller scale assessment of staff needs and training delivery was considered a key initial step. This project chose specifically to focus on delusions within the context of psychosis given some similar work on hallucinations was currently being undergone in the service by another clinical psychologist. Understanding of how to work with those experiencing delusions had also been identified by the authors and their colleagues as a particular area requiring improvement within the service.

Prior to delivering training it was necessary to understand more objectively the current situation and needs of the service. An audit to identify the number of individuals in the service experiencing paranoia or delusions was therefore conducted, with exploration of how many of these patients are then referred for psychological therapy and how paranoia and delusions were being described and documented in clinical notes. A staff survey was also run, to explore current understanding, knowledge, and confidence regarding working with individuals experiencing paranoia or delusions. This would help to identify areas of focus for the training.

In summary, the aims were as follows: first, to audit the number of individuals within the two teams who over the last year have been experiencing paranoia or delusions, and to identify what proportion of these individuals were referred to the Step 4 Psychological Service; secondly, to assess staff knowledge and confidence in delivering care to individuals experiencing delusions, and their understanding of psychological perspectives on delusions via completion of a staff survey; finally, to design a staff training session based on the identified need and deliver this intervention, as well as gaining feedback from staff on its impact. Given the lack of existing service level data from older adult teams assessing the prevalence of psychotic experiences, rates of referral for psychological therapy, or staff conceptions regarding psychosis, these aims were considered to provide considerable novelty to the existing evidence base.

The project took place after the COVID-19 pandemic where teams were trying to return to face-to-face work. Both teams began experiencing severe difficulties maintaining staffing levels, with significant numbers of staff recruited as locums and without previous experience of working with older adult populations. Adding to this, the training was scheduled during a time where services were unexpectedly experiencing an extended outage of Carenotes, the trust’s clinical notes system, adding to pressures on staff workloads. Given these challenges, it was clear from the start that to engage staff in a training, the session needed to be brief, straightforward, and have a practical focus.

Method

The project did not require NHS ethical approval given it falls within a service evaluation remit. Trust service evaluation/audit approval was sought and received prior to beginning the project.

Audit

Using routinely collected data, information was collated regarding the number of patients within the two teams who were experiencing delusions or paranoia and the number of which were then referred to the Psychology team. CRIS, a service providing deidentified access to clinical notes that is increasingly being used in research (Bowen et al., 2020), was used to search care notes, including free text data, from both teams.

The following search terms were used: ‘delus*’ or ‘paranoi*’ or ‘grandios*’ or ‘unusual belief’ or ‘unusual idea’ or ‘unusual thought’, with notes excluded if ‘no’ proceeded any of the terms.

Given the authors’ observations that clinical notes regarding delusions often described staff attempts to ‘challenge’ individuals regarding their beliefs, a search was also conducted to see how frequently the word ‘challenged’ was included with clinical notes.

Staff survey

All CMHT staff were invited to take part in a survey assessing current levels of staff knowledge regarding psychological understanding of delusions. Demographic data included age, gender, job role, and length of time working in the team. Further questions asked about the numbers of individuals on caseloads experiencing paranoia or delusions, any past training in understanding or working with delusions, and questions about what staff typically do or ask when assessing individuals with delusions. For example, do they aim to challenge delusions and do they assess distress? Finally, staff were asked to list and rank the most helpful treatments for delusions, and perspectives on psychological approaches to delusions were gained. Free text boxes were included on questions where relevant to allow staff to elaborate on their answers.

Training

All CMHT staff from both teams were invited to take part. The lead author liaised with service managers to identify suitable times for the training to take place. The training was devised based on the survey results and the contributions from the service users involved. Two lunchtime training slots were offered, with a recording of the session also distributed to those unable to attend. Slides and a summary sheet were also distributed electronically and put on display in the offices.

The training was devised by the authors and delivered by the lead author online via MS Teams. The session consisted of two parts. The first part concerned rethinking delusions as an experience rather than just a symptom by presenting evidence for its existence on a spectrum of severity across the population and a discussion of how certain types of unusual or false beliefs may be considered normal or acceptable in different contexts. An evidence based cognitive model (Freeman, 2016) of persecutory delusions was presented to illustrate six common, understandable, and easily treatable factors that can contribute to delusions. Evidence of the effectiveness of CBT for psychosis was then presented. Part 2 concerned how CMHT staff can work with individuals experiencing delusions, starting with straightforward ways to engage patients by asking meaningful questions from a cognitive behavioural approach. Common concerns and myths about delusions as well as specific considerations for older adults were then explored.

Measures

Increased ability to consider cognitive behavioural perspectives when caring for individuals with delusions is something that would primarily be expected to develop over time, across in-the-moment staff–patient interactions. This is difficult to measure in a standardised way, a challenge also faced within the HEE commissioned psychosis training. For this project, staff were instead asked to complete four analogue scales measuring endorsement from zero (not at all) to 100 (completely) of purely biological versus psychological perspectives of delusions immediately before and after the training. The statements were: delusions are a symptom of a chronic neurological disease that can be managed but not cured; it’s important to challenge patients’ delusions to help bring them back into reality; delusions always require treatment by medication; there is value in using psychological approaches with people experiencing delusions. At the end of the training staff were also asked to provide feedback on what they found helpful about the training, what could be improved, and any further topics or types of training and support they would like access to in the future.

Service user involvement

Two individuals with experience of delusions and one carer of an individual with delusions from within the service fed into the development of the staff training, one over the telephone and two face-to-face. They were asked questions such as ‘what would you like staff to know or understand about the experience of feeling unsafe or worried around other people?’ and ‘what are some of the things that have been most helpful or unhelpful regarding your interactions with staff?’. Table 1 summarises their input and how this was incorporated in the training.

Table 1. Summary of service user feedback and how this was actioned

Results

Audit

4065 individuals were being held under the two teams between April 2021 and April 2022. Over a quarter of these individuals (n=1161) had clinical notes hitting at least one of the search terms (n=807 for delusion; n=839 for paranoia; n=121 for grandiose; n=19 for unusual belief; n=23 for unusual idea; n=79 for unusual thought, with notes excluded if ‘no’ proceeded any of the terms). Of these, there were only six referrals. There were 166 clinical notes that had the search terms present as well as the word ‘challenged’ within the same note.

A random selection of five individuals who each had five or more notes mentioning paranoia or delusions (the two most common search terms) was made using an online generator, yielding 89 clinical notes, of which five were again randomly selected.

There were five individuals who had five or more notes containing one of the search terms as well as the word ‘challenged’. A random sample of five of these notes was selected. Table 2 presents extracts from these notes.

Table 2. Extracts from clinical notes

Staff survey

Twenty-seven members of staff completed the survey: eight psychology staff, three psychiatrists, five nurses, five occupational therapists, three social workers and one manager. Staff turnover in both CMHTs was evidently high, with half of respondents having been employed in the team for less than 1 year so far. The number of individuals on respondents’ caseloads ranged from 0 to half the entire CMHT area, with the estimated numbers of individuals on caseloads experiencing severe paranoia or delusions ranging from 0 to 10 (an average response of 2, although several staff responded with ‘not sure’).

Half of respondents reported having never had training or teaching on working with paranoia or delusions and for those that had, this was typically limited to what had been within their professional training, e.g. within their general psychiatry training. As such, with the exception for those more recently qualifying, most training was from many years ago.

Nine respondents wrote medication as the most helpful intervention for individuals experiencing delusions, with all others listing psychological or social approaches as the topmost helpful intervention. Despite this, 14 out of the 19 non-psychology staff members reported they may not consider psychological perspectives or referral for psychological interventions due to a lack of knowledge or confidence, with other reasons including the individual being too unwell or not wanting any. Most staff reported being open to talking to a patient about their delusions, but almost 20% reported that they do not typically ask about the distress associated with a patient’s delusion. Sixty per cent (equating to almost all non-psychology staff) responded ‘sometimes’ to a question about whether it is important to challenge a person’s delusion in order to help them see that their belief is not true.

Qualitative comments included multiple concerns that agreeing with a patient’s delusions could increase the patient’s belief and conviction in their delusion and that even just talking to or asking an individual about their delusion may cause distress or aggression. On the other hand, some respondents did feel it important to ask detail about a patient’s delusion in order to understand their perspective and create a care plan. Multiple respondents felt psychological approaches could be helpful with someone who is well enough and who has tried medication, primarily to help them test/check their reality.

Staff training

Roughly a quarter of staff attended the training. Eleven staff members attended the first training session and nine attended the second, including two consultant psychiatrists, four social workers, four assistant psychologists, three clinical psychologists, and seven nurses (two holding senior leadership and management roles). Several additional members of staff watched the recording of the training in their own time. In total, 23 responses were gained for the pre-training questionnaire, and 25 responses for the post-training questionnaire. To help preserve anonymity, pre and post responses were not linked through any form of identification, and several members of staff joined each training session late or had to leave early. As a result, it was not clear how many individuals completed both the pre and post measures and thus how similar the two groups of staff were. Table 3 summarises the results the measures. For question 1 the mean endorsement pre-training was 32.88 (range: 0–71) and post-training was 34.64 (range: 0–90). For question 2 mean endorsement pre-training was 25.13 (range: 0–80) and post-training was 19.68 (range: 0–89). For question 3 mean endorsement pre-training was 32.76 (range 0–94) and post-training was 27.81 (0–78). For question 4 mean endorsement pre-training was 88.08 (range 42–100) and post-training was 85.65 (52–100). Overall, it is clear that on average there was limited change across the group pre–post training, with very high ranges in scores suggesting significant diversity in views and approaches to understanding delusions.

Table 3. Endorsement of statements pre- and post-training

In the post-survey qualitative comments, 22 respondents described the training as relevant and helpful, with the remaining three people not completing this question. Particular parts that respondents reported finding helpful included: historical perspectives for older adults with psychosis, causal and treatment mechanisms in paranoia, summarising the evidence base, encouragement of validation rather than challenging patients, checking whether any reported persecution could be real, and the practical aspects of what to actually say to patients.

In terms of what staff wanted more access to in the future, 17 wanted more training, 14 wanted case formulation sessions, six wanted supervision groups, seven wanted 1:1 consultation with a clinical psychologist, six wanted Q&A sessions and 11 wanted access to resources. Suggested topics included working on insight, working on delusions in the context of severe co-morbid depression or dementia, and full access to training on CBT for psychosis. However, a number of staff also noted that engaging with this kind of training was difficult due to being in a state of ‘trying to keep heads above water’.

Discussion

This project aimed firstly to identify the number of patients across the two services who experience paranoia or delusions, secondly to assess the theoretical approach typically taken by staff when working clinically with delusions, and thirdly to deliver a training incorporating an up-to-date cognitive behavioural understanding of delusions. There was a striking number of patients in the services (over 20%) who were identified to have the search terms present in their notes. While there may be a number of false positives (for example, notes where paranoia or unusual thoughts were mentioned but in a different context or discussed in terms of their absence), none of the randomly selected notes showed evidence of this. A number of these individuals may have been experiencing paranoia or delusions in the context of a dementia rather than, for example, first episode psychosis, but a psychological understanding in this context is arguably no less important (Woods and Clare, 2015). It would be helpful to investigate further whether there are significantly more older adults experiencing these symptoms than previously recognised, or whether there is something about the way paranoia is reported in clinical notes that is misrepresentative or over-inflated.

The randomly selected notes also provided insight into the way delusions are recorded. While one of the notes (note 2 in Table 2) provided a helpful example of the clinician considering the individual’s context and religious background as a way to understand their unusual ideas, most other notes contained little information on the content, context, or impact of the individual’s beliefs. Table 2 also confirmed that people with delusions are sometimes being ‘challenged’ directly on their delusional beliefs despite several examples of this proving unhelpful and distressing to the individual.

This high number of patients with clinical notes mentioning delusions and paranoia was in stark contrast to the number of referrals received in the Psychological Therapies teams. From the present data alone, it is not clear if there were in fact many more patients who would have been suitable and benefited from psychological therapy, but this is certainly one plausible interpretation. We could have chosen to measure whether referrals to Psychology increased following the training. However, it was felt that this outcome can become tokenistic when used as a metric of change and in fact the real aim was to improve staff–patient interactions at all levels.

Results from the survey showed there was definite room for providing further training in understanding psychosis, and feedback on the training further showed this would be welcome and relevant. As is often a challenge (Hardy et al., 2021) the high staff turnover that was apparent means that going forward training either needs to be repeated, or to go into an induction pack with supervisors monitoring completion where possible and relevant.

Interpreting the wider impact of the training was also not straightforward. Both the training and associated measures were required to be brief in order to increase the likelihood of staff engaging and attending. The measures of perspectives on delusions did not show much average group change after the training, perhaps suggesting that a brief intervention is not sufficient for helping to merge perspectives and understanding of delusions. Arguably the much more striking result from the measures, however, was just how different individual responses were. Ranges for each question were stark, with some respondents completely disagreeing with the first three questions (i.e. scoring their agreement at or close to zero), and others almost completing agreeing (i.e. scoring their endorsement above 90%). The smallest range was for question 4, regarding the value of psychological approaches, but even this ranged from 42% to 100% agreement. Given that NICE guidance recommends psychological treatments as a first-line intervention for psychosis alongside medication, it is surprising that some staff members rated the value of psychological treatments so low. While a key strength of multi-disciplinary working is the integration of care from multiple perspectives, the question must be asked about whether consistent, reflective, and evidence-based team working, and patient care, can really be achieved in the face of such stark differences in conceptualisation and approach. Regular team case discussions and supervision should mitigate against these concerns, but for this to succeed all staff must feel valued and empowered to speak up and reflect in these contexts, which may not be the case in every service. Moreover, not every patient can be discussed by the whole team.

There was considerable positive qualitative feedback regarding the helpfulness of the training and the gap staff felt it was helping to fill. Following the training, the lead author was contacted by several members of staff to discuss cases and seek advice on how to work with particular patients with delusions. There is increasing evidence emerging regarding the role of champions within healthcare implementation (Santos et al., 2022); meaning having a staff member within CMHTs emerge as a champion in this way can support staff members to seek advice quickly and easily. With high rates of staff turnover this can be challenging, however. Indeed, the lead author herself was only on short-term placement in the service which ended shortly after the project was completed. However, it was agreed within the team that a clinical psychologist would re-facilitate the training to CMHT staff every 6 months in order to help reduce the difficulties associated with high staff turnover. Continued efforts to involve service users in service delivery is also important for bringing about improvements. This is particularly the case for those with delusions who, if being seen as having chronically lost touch with reality, may otherwise be overlooked and not offered opportunities to participate in service feedback and delivery.

There were also several mentions of burnout and stress in the qualitative comments, adding to the narrative of it being challenging to make change within a context of high stress and understaffing (Paris and Hoge, 2010). Despite this, there were also requests for more help in this area with lots of staff stating they would like further training sessions, potentially with the recognition that enhancing knowledge can often bring about time savings in the long run.

Implications

In this project, the data on numbers of older adults experiencing delusions were not easily available. Many staff were not sure of how many people on their caseload were experiencing paranoia or delusions, and the Psychology team received very few referrals. It would be easy to have initially concluded, then, that this is not an area of particular importance. However, upon searching for more data, it became clear that this was not the case. It was evident that staff are often working with severe paranoia or delusions, but do not necessarily feel confident in doing so. Most significantly, staff approach their clinical work regarding delusions from vastly different theoretical positions. Service leaders need to be aware of these differences and find ways of bringing staff together to provide holistic and consistent care.

Supporting change within CMHTs is a fine balance, however. While 20 minutes of training could never set out to compete with 20 years of professional practice, even training as brief as this felt too much for many staff members to find the time to engage with. At times of high stress, it is more common to rely on well ingrained knowledge from core training than to fish for more recent learning that may not have previously been put into practice. A sustained approach may therefore be required, for example using ideas from nudge theory (Thaler and Sunstein, 2009), in addition to formal training enforcement. Nudges might include having a champion associated with the training and as an expert in the area who is able to offer consultation, training reminders, and resources, and to encourage joint working and case presentations facilitated by both MDT and psychology staff together. As staff start to see the benefits to their learning, incorporation of learning is likely to be reinforced, thereby gradually allowing for wider and more significant change.

Limitations

There were a number of limitations to this study. The most significant was the difficulty in using standardised outcome measures to assess the impact of the intervention. Qualitatively it was clear that staff found the training relevant and helpful, and it did lead to 1:1 follow-up consultations where the lead author was able to provide advice regarding a case, but there was no way to clearly measure long-term changes in staff conceptualisation of delusions.

A further limitation is that with regard to the endorsement of medical versus psychological perspectives, the large ranges combined with not all the same people completing pre and post measures meant the group changes (or lack of) were not necessarily meaningful, and it was not possible to assess change at the individual level for those who did complete both sets, given pre and post responses were not linked.

Thirdly, it was challenging to accurately measure the number of individuals experiencing paranoia and delusions. The search terms may have missed some individuals where paranoia or delusions were qualitatively described without using the named terms, yet also included others whether the terms paranoia or delusions were discussed but in terms of their absence rather than presence. These data can therefore be only seen as an estimate.

Conclusion

In 2001 an article summarising the previous decade of psychological training available to mental health staff described the situation as ‘hectic’, with little consistency across trusts or professions (Brooker, 2001). Twenty years on it is not clear that the situation has vastly improved, evidenced here by the inconsistency in prior training of staff on delusions and the fact that neither CMHT was able to access the wider HEE commissioned training. Delivering a brief training as in this project may be considered to only add to the ‘hectic’ situation, especially given the limited follow-up of measurable outcomes. However, we argue the project provides a timely example of the difficulties of promoting change within CMHTs with consideration of what is still possible within a stretched system. Where larger interventions – such as the HEE training – are not possible, smaller scale nudges towards up-to-date evidence-based practice may be paramount.