Background

The National Health Service (NHS) Race and Health Observatory (RHO) was founded in April 2020 on the basis of providing an evidence-based approach to tackling racial disparities in health outcomes in the NHS, making policy recommendations and providing tangible support in the interests of equity. Racial equity in mental health has been an ambition of the RHO from inception; the work in this field is undertaken by the Mental Health Advisory Group (MHAG). The MHAG is responsible for commissioning research into racial and ethnic disparities in mental health, and in this regard, improving access to psychological therapies (IAPT) became a key focus. We report here key findings of our commissioned research, which revealed that psychological interventions are inequitably delivered to ethnic communities compared to the White population through IAPT on a number of parameters.

A rapid evidence review

An earlier review¹ showed clear evidence of the disparity in outcomes for Black and ethnic minority communities with mental health conditions, with demonstrable differences in access, experience and outcomes compared to the White population. The review of academic research spanning a 10-year period found that ethnic minority groups experienced distinct inequalities in mental health support provision and in gaining access to the IAPT programme in England, which started in 2008 and has now been rebadged as ‘NHS Talking Therapies for anxiety and depression’. Starting from humble beginnings,² IAPT was accessed by 1.2 million people in the 2021/2022 financial year and has the ambitious target of providing a service to nearly 2 million people by the end of the 2023/2024 financial year. However, the RHO's rapid review found that general practitioners were less likely to refer ethnic minority patients compared to White British people. Barriers were also faced by patients who delayed or avoided seeking help for health problems due to their fear of racist treatment from NHS healthcare professionals.

Ethnic inequalities in IAPT

In the background to this and the anecdotal evidence of persisting difficulties that ethnic patients had reported with IAPT services, the RHO's MHAG commissioned further research through the National Collaborating Centre for Mental Health (NCCMH) and published the findings³ recently. The IAPT programme is delivered by over 10 000 staff, with 80% reporting their ethnicity as ‘White or White British’, just marginally less than the 83% reported by the wider England and Wales adult population. There is less gender equality, with 80% of therapists reported as female, compared to 51% of the adult England and Wales population, according to the 2021 Census.⁴

Utilising nationally aggregated data-sets from NHS Digital, our researchers found that the proportion of referrals that accessed IAPT services dropped in all groups, from 76% in 2015/2016 to 72% in 2021/2022. Between 2015 and 2022, people from ethnic groups categorised as ‘White’ were more likely to be offered therapy after an initial assessment than all other ethnic groups. There were differences within the ethnic groups as defined by Office for National Statistics (ONS) ethnic categories, with ‘Mixed: White and Asian’, ‘Asian: Indian’ and ‘Asian: Other Asian background’ more likely to access services, while people of ‘Mixed: White and Black Caribbean’, ‘Mixed: White and Black African’ and ‘Mixed: White and Asian’ were the least likely to access. Figure 1 shows that between 2016 and 2022, people from ‘White’ ethnic groups were more likely to receive a course of treatment after initial assessment compared with people from ‘Black or Black British’, ‘Asian or Asian British’, ‘Mixed’ or ‘Other Ethnic groups’, year on year. More specifically, trends in the course of treatment rates showed that the ‘Asian: Bangladeshi’, ‘Asian: Pakistani’ and ‘Black: African’ groups were less likely to have a course of treatment than the ‘White’ ethnic group.

Fig. 1 Course of treatment rates for high-level ONS groups (reproduced from IAPT review³). IAPT, improving access to psychological therapies; ONS, Office for National Statistics.

The average depression scores as measured by the Patient Health Questionnaire (PHQ) and average anxiety scores as measured by the Generalised Anxiety Disorder Questionnaire (GAD-7) were highest at the point of entry in the ‘Pakistani’ group, though conversely they are less likely to access services than ‘White British’ groups. In terms of recovery after therapy, outcomes are best for ‘White’ ethnicities, but the gap has narrowed with improvements across for all groups, year on year. Again, there are inter-ethnic differences, with, for instance, the ‘Indian’ data group having much higher recovery rates than the ‘Asian: Pakistani’ and ‘Asian: Bangladeshi’ groups, and drop-out rates were highest among ‘Mixed: White and Black Caribbean’, ‘Black: Caribbean’ and ‘Black: Any Other Black Background’ groups.

There were limited data from the Patient Experience Questionnaire that therapists are encouraged to ask patients to complete. Generally, those from a ‘White: British’ ethnic group were more likely to say that they had received sufficient information on choice of therapies.

One key theme that emerged across focus groups conducted by the research team was cultural sensitivity, including the cultural sensitivity of therapists, and the need for a change in service culture. Therapist characteristics (such as gender, age and ethnicity), cultural sensitivity and the language in which therapy is delivered were considered important for improving IAPT treatment delivery for people from minoritised communities.

Referring to the IAPT Black, Asian and Minority Ethnic Service User Positive Policy Guide (IAPT PPG),^{Reference Beck, Naz, Brooks and Jankowska5} developed to improve IAPT service access, delivery and outcome in the ethnic communities, the study found that the guide was considered helpful but limited resources meant that it was not implemented as recommended, and where it was, there was inconsistency across services – not surprising, since 0% of the 11 commissioners reported ‘good’ or ‘excellent’ knowledge of the guide and 40% had ‘little to no knowledge’, despite all reporting that the IAPT PPG was being implemented in their service.

From the provider end, most clinical leads and commissioners reported having a number of strategic components in place to support better access and mitigate access challenges for people from minoritised ethnic groups although neither discussed explicitly the use of local data to improve services. Disappointingly, almost half of the commissioners (45%) reported that there were no resources or funding for addressing ethnic inequalities in their service, which might predict that gaps in services to ethnic and minority communities will not be addressed, at least in the short term.

Recommendations

The report makes eight overarching recommendations to meet with the challenges for ‘NHS Talking Therapies for anxiety and depression’:

1. (a) Influencing system leaders: Integrated Care Boards (ICBs) and those in leadership roles in mental health services should respond to the inequalities highlighted in the report, use local data to understand the needs of minoritised ethnic communities in their local area and identify the resources and funding that can be used to implement these recommendations and address those inequalities.

2. (b) Training: system leaders should undergo training to improve their understanding of mental health inequalities and of the IAPT PPG, in order to meet the needs of people from minoritised ethnic backgrounds.

3. (c) Best practice: ‘NHS Talking Therapies for anxiety and depression’ should fully implement the IAPT PPG. All ICBs should take full account of the recommendations of the guide when commissioning services, including budgeting for the additional resources needed.

4. (d) Community engagement: designated staff should support services to engage with communities, to co-produce culturally informed pathways into care, and they should engage with and gain understanding of local communities to promote the benefits of psychological therapy, the types of support available and how treatment can be accessed, to facilitate referral.

5. (e) Providing culturally sensitive care: services should ensure that all interventions are provided in a culturally sensitive way with a deeper understanding of sociocultural and discriminatory factors that may have an adverse impact on patients and minoritised communities, ensuring that best outcomes are provided to people whose first language is not English and including effective use of interpreter services.

6. (f) Advancing equity and equality: services should use their own data, alongside nationally available reporting, to develop and evaluate programmes of work with communities and other stakeholders to address access and outcome equities and inequalities.

7. (g) Workforce training and competence: education providers and services, in co-production with people with lived experience who belong to minoritised ethnic communities, should design and deliver ongoing continuing professional development for all ‘NHS Talking Therapies for anxiety and depression’ staff to provide a therapeutic environment that is culturally safe for the communities that clinicians are working with; the training should include a focus on understanding the impact of institutional racism on the experiences of people from minoritised ethnic groups.

8. (h) Workforce representation: services should recruit, train and retain a diverse staff workforce that is reflective of and able to respond to the needs of the local community. National efforts should be made to increase the sociodemographic (gender, ethnicity, age) representation of the workforce (for example, recruiting more men may help to increase engagement and uptake of NHS Talking Therapies for anxiety and depression by men from some minoritised ethnic backgrounds). Services should take ongoing action to address inequalities in opportunities for minoritised staff to progress into senior leadership roles, including positive action to reduce systemic barriers to promotion.