Katherine Newman-Taylor

Background: Best practice guidelines recommend CBT for everyone with psychosis, yet high rates of non-engagement and early drop-out result in many not receiving ‘full dose’ therapy. For those who do engage, effect sizes are typically modest. Most people with psychosis have an insecure attachment style, characterised by predictable and problematic patterns of beliefs, emotion regulation and interpersonal behaviours, including help-seeking. Attachment-congruent adaptations to CBT could improve retention and recovery outcomes for people who are also insecurely attached. Aims: This is the first evaluation of attachment-congruent adaptations to CBT for psychosis. We assessed impact of therapy for the first five people (aged 30–50) to complete therapy and measures. Method: We used a single case ABC design and measured paranoia, voices, help-seeking, coping, and personal goals over baseline, intervention, and follow-up phases. We predicted that adapted CBT would lead to improved outcomes which would be maintained at follow-up. Trait measures of voices, paranoia, attachment, and recovery are used to describe the sample. Results: Participants presented with complex needs, including paranoia, voices, and attachment insecurity. Baseline phases showed high and variable levels of psychosis. All showed improvements in paranoia, voices, willingness to seek help, ability to cope, and/or personal recovery goals over the intervention phase. Gains were largely maintained or improved further at follow-up. Conclusions: Attachment-congruent adaptations to CBT are likely to benefit people with psychosis who present with the additional cognitive, affective and behavioural complexities characteristic of insecure attachment, which threaten access and impact of recommended treatments if not addressed.

The UK National Institute for Health and Care Excellence (NICE) recommends CBT for people with psychosis, yet many do not access therapies promptly, and duration of untreated psychosis predicts poorer outcomes. In addition to systemic service issues, it is likely that insecure attachment, common in people with psychosis, constitutes a key barrier to therapy. If we can articulate attachment patterns in terms of interpersonal beliefs and behaviours, we will be better placed to engage people in CBT. We used a cross-sectional design to examine theoretically predicted relationships between attachment style and (1) beliefs about self and others, (2) underlying assumptions, and (3) help-seeking intentions, across the psychosis continuum – in analogue and clinical samples. We recruited a general population analogue sample with elevated levels of non-clinical paranoia ( n =172) and a clinical sample with self-reported psychosis ( n =130). All completed standardised measures of attachment, beliefs about self and others, underlying assumptions and help-seeking. Regression analyses showed that insecure attachment predicted higher levels of negative self- and other-beliefs, and problematic underlying assumptions, and reduced help-seeking intentions, with some differences in patterns across analogue and clinical samples. Attachment is associated with CBT treatment targets and can inform psychological formulation and treatment. Our measure of assumptions may have been inappropriate for the clinical group. We conclude with public health, service level, and clinical implications to improve engagement and outcomes in CBT for psychosis. Key learning aims: (1) Everyone with psychosis should have access to recommended treatments including CBT, but many do not. (2) Insecure attachment is common in people with established psychosis or elevated levels of non-clinical paranoia, and may be a key barrier to accessing therapy. (3) We show that insecure attachment is associated with (1) beliefs about self and others, (2) underlying assumptions, and (3) help-seeking intentions. (4) Attachment style can inform CBT formulation and intervention, and wider service level and public health campaigns designed to improve engagement and outcomes for people with or vulnerable to psychosis.

CBT for psychosis is an established and evolving psychological therapy. Historical controversies about the nature of psychosis persist, and more recent debates about the outcome literature lack precision, muddying the waters further. Based on our experience as clinicians, teachers and supervisors, and following NHS and national lead roles, we describe ten common misconceptions about CBT for psychosis. These include misconceptions about the evidence, the focus of therapy, ‘thinking positively’, and the nature of collaboration and the therapeutic relationship. We refute these misconceptions based on current theory, research, and best practice guidelines. We highlight the need to get out of the clinic room, measure the impact of therapy on personal recovery and autonomy, and meet training and governance requirements. It is essential that clinicians, service leads, and our professional bodies uphold core standards of care if people with psychosis are to have access to high quality CBT of the standard we would be happy to see offered to our own family and friends. Key learning aims: (1) To recognise common misconceptions about CBT for psychosis. (2) To counter these misconceptions theoretically and empirically – to inform ourselves, colleagues and service leads committed to ensuring high quality CBT for psychosis. (3) To highlight statutory and professional body responsibilities to ensure parity of esteem for people with psychosis, who deserve high quality, ‘full dose’ treatments delivered by appropriately trained clinicians, and supported by robust governance systems, just as we would expect for people with physical health conditions.

Background: People at high risk for psychosis access primary care mental health services for depression and anxiety and are unlikely to recover from these affective symptoms. We report the first controlled trial of cognitive–behavioural therapy (CBT) for depression and anxiety, minimally adapted for psychosis risk, in primary care. Aims: To evaluate feasibility, acceptability and signals of efficacy for CBT for depression and anxiety adapted for psychosis risk, designed in collaboration with people with psychosis. Method: A longitudinal controlled trial comparing best practice CBT for depression and anxiety (CBT-BP) with CBT adapted for psychosis risk (CBT-PR), in patients meeting criteria for UK primary care services and who are also clinically high risk for psychosis (trial registration no. ISRCTN40678). Results: Rates of recruitment (55 to CBT-BP, 44 to CBT-PR), completion of measures (90% CBT-BP, 94% CBT-PR) and retention in therapy (75% CBT-BP, 95% CBT-PR) demonstrate the feasibility and acceptability of the adapted therapy. Routine measures of depression and anxiety signal improved clinical and recovery outcomes for CBT-PR. Psychosis and relational measures signal sustained improvement (at 3 months) in the CBT-PR group. No serious adverse events were reported. Conclusions: Primary care mental health services present a unique opportunity to identify and treat people at risk of psychosis at a time when they are help-seeking. CBT for depression and anxiety, minimally adapted for psychosis risk, can be delivered in routine services, and is likely to improve clinical and recovery outcomes and reduce psychosis risk. A definitive trial is needed to estimate clinical and cost-effectiveness.

Untreated psychosis can have a devastating impact on individuals and their families. Recommended therapies are underutilised. Attachment theory provides a framework to understand intra‐ and interpersonal responses to threat cues, such as voices and paranoia, which affect recovery. This is the first study to test a coherent set of theory‐driven predictions about the impact of attachment style on people's ability to manage distress associated with psychosis, and seek help effectively. We used a cross‐sectional design to examine the impact of attachment style on emotional regulation, help‐seeking intentions, service engagement, and clinical and recovery outcomes in people with psychosis. We recruited a total of 65 participants with psychosis from community and NHS pathways. Participants completed standardized measures of predictor and dependent variables at one‐time point. Regression analyses showed that insecure attachment (anxious and avoidant) predicted more use of unhelpful emotional regulation strategies, less help‐seeking, and poorer clinical and recovery outcomes, with medium to large effects (p < 0.05). We found no effects for service engagement. Attachment style predicts intra‐ and interpersonal responses to threat in psychosis, and may partially account for the considerable variation in engagement with recommended therapies, and longer‐term outcomes. Routine assessment of attachment at initial service contact would identify people likely to struggle to seek and accept therapeutic interventions, who may in turn benefit from attachment‐congruent engagement efforts.

Purpose: Many people with psychosis do not seek help which delays access to recommended treatments. Duration of untreated psychosis is associated with poor healthcare outcomes and increased risk of relapse. The reasons why people delay accessing treatment remain unclear. This is the first systematic review to synthesise the literature examining professional and non‐professional help‐seeking in psychosis across clinical and subclinical populations. Methods: We searched four databases (APA PsycINFO, APA PsycArticles, Medline and British Library EThOS) to generate a comprehensive account of the quantitative literature. Heterogeneity of measures precluded a meta‐analysis. Results: We identified 19 articles (including 9686 participants) that met criteria for the review. Help‐seeking in psychosis is associated with being female, having a higher level of education, and experiencing more than one symptom. People with psychosis report stigma, poor mental health literacy and lack of family support as key barriers. Clinicians report childhood physical abuse, insecure attachment and severity of psychosis as additional barriers. We also found differences in preferred sources of help across cultures. There is currently no consensus on reliable help‐seeking measures. Conclusions: This is the first systematic review to examine help‐seeking behaviour in psychosis. Assertive and culturally sensitive engagement efforts should be targeted towards people with a history of early adversity, poor mental health literacy, limited social support and more severe psychosis.

Objectives: Early interventions improve outcomes for people at high risk of psychosis and are likely to be cost saving. This group tends to seek help for emotional problems – depression and anxiety – via primary care services, where early detection methods are poor. We sought to determine prevalence rates of high risk for psychosis in UK primary care mental health services and clinical outcomes following routinely delivered psychological therapies. Methods: We used a brief screen designed for settings with low base rates and significant time constraints to determine prevalence of high risk for psychosis in UK ‘Talking Therapies’ services. We examined socio‐demographic characteristics, presenting problems and recovery trajectories for this group, compared with people not at risk of psychosis. Results: A 2‐item screen selected for specificity yielded a prevalence rate of 3% in primary care mental health services. People at elevated risk of psychosis were younger and more likely to report at least one long‐term physical condition. This group presented with higher levels of depression, anxiety and trauma symptoms at assessment and were less likely to have recovered at the end of treatment, compared to people not at risk. Conclusions: Very brief screening tools can be implemented in busy health care settings. The 3% of referrals to UK primary care psychological therapies services at elevated risk of psychosis typically present with more severe symptoms and greater levels of comorbidity and may require augmented interventions to recover fully.