With one in six people in the UK affected by Common Mental Disorders (CMDs) like anxiety and depression every week (McManus et al., 2016), improving access to psychological therapies is seen as a cornerstone to a healthy healthcare system and economy (Clark, 2011).

Since 2008, the Improving Access to Psychological Therapies (IAPT, now rebranded as NHS Talking Therapies) programme has aimed to meet this challenge, and by many metrics IAPT has been a huge success. With millions more in therapy and a near 50% recovery rate (Wakefield, 2021), IAPT has inspired several other countries to adopt similar systems (e.g., Cromarty et al., 2016; Naeem et al., 2017).

However, inequalities in access amongst certain socio-demographic groups are consistently found in mental health services (Spiers et al., 2019), leaving those most vulnerable to developing CMDs (e.g., ethnic minorities, disabled individuals) with the least support (Macintyre et al., 2018).

Within IAPT itself, research on access equity is disappointingly slim with only local studies (often with incomplete datasets) finding inequality of access (Ahmad et al., 2022). Sharland and colleagues (2023) aimed to redress this in a nationwide retrospective cohort study, investigating if access to IAPT was equitable across socio-demographic characteristics and met the probable need within the general population.

Methods

2011 Census data was linked to 2017/18 IAPT records to build a comprehensive socio-demographic dataset of all patients in England who received ≥1 IAPT treatment session (N = 727,802).

The UK Household Longitudinal Study (UKHLS, 2017/18) was used to estimate probable CMD rates in the general population (N = 15,614).

To investigate the impact of socio-demographic characteristics on the likelihood of accessing IAPT when experiencing a probable CMD, logistical regression analysis compared the IAPT dataset to the UKHLS dataset, across several characteristics like age and ethnicity.

Results are presented as percentages with 95% confidence intervals.

Results

IAPT access rates in the general population

Overall, around 2% of adults in England accessed IAPT services in 2017/18. This was considerably lower than the 18% who had a probable CMD based on the UKHLS dataset.

IAPT access rates by socio-demographic characteristic

1) Age

Older adults were the least likely to report a CMD in the general population. Older adults who did report a probable CMD were three times less likely to access IAPT for treatment compared to younger adults (18-24) (4.7% [4.3 to 5.1] compared to 14.2% [12.0 to 17.5]).

2) Sex

Males with a probable CMD were significantly less likely than females to access IAPT (8.5% [7.9 to 9.2] compared to 10.8% [10.3 to 11.4]).

3) Ethnicity

People with Asian ethnicity were the least likely to self-report a CMD in the general population, and also the least likely to access IAPT services even if they had a probable CMD (e.g., 7.2% [6.2 to 8.7] compared to 7.9% [6.1 to 11.5] Black ethnicity).

Although the Mixed Ethnic group were most likely to report a CMD and access IAPT, as a proportion of need within this group, they were still less likely to access IAPT compared to White ethnicity (9.7% [7.6 to 13.5] compared to 10.2% [9.7 to 10.7]).

4) Country of birth

People born outside the UK were nearly half as likely to access IAPT treatment compared to those born in the UK (1.2% [1.2 to 1.3] compared to 2% [2.0 to 2.0]). Even with a probable CMD, people born outside of the UK were significantly less likely to access IAPT treatment compared to those born in the UK (7.2% [6.3 to 8.4] compared to 10.4% [9.9 to 10.9]).

5) English as a first language

For those with a probable CMD whose first language was not English, accessing IAPT was found to be significantly less likely when compared to native speakers (6.9% [5.9 to 8.4] compared to 10.1% [9.7 to 10.6]).

6) Religion

People with no religious affiliation and a probable CMD were significantly more likely to receive IAPT treatment, compared to those with religious affiliation CMD (12.4% [11.6 to 13.3] compared to 9.1% [8.6 to 11.3]).

7) Disability status

Those with a disability were more likely to self-report a CMD, but significantly less likely to access IAPT services with a probable CMD, compared to those without a disability (6.8% [6.4 to 7.3) compared to 11.5% [10.8 to 12.1]).

8) Highest qualification

Having no qualifications was associated with higher reporting of probable CMDs. However, those with no qualifications self-reporting a CMD were less likely to access IAPT services, compared to those having any higher education qualification (6.3% [5.4, to 7.4] compared to 10.3% for both below degree level [9.7 to 11.1] or degree level and above [9.5 to 11.2]).

Conclusions

This study found that access to IAPT services overall was relatively low when compared to the probable rate of CMD in the general population. This may indicate barriers to access more broadly.

More specifically, access to IAPT services was not equitable across all socio-demographic characteristics, even when there was a probable CMD need. Particularly vulnerable groups included:

• Older adults
• Ethnic minorities
• Individuals with disabilities
• English not being a first language
• Being born outside the UK
• Being male
• Lower academic attainment

Strengths and limitations

This unique approach to investigating equity of access to IAPT manages to capture a rich, detailed amount of demographic information across the whole of England. As patient data was taken from national registers, this study offered a large, representative sample of England, adding validity to its findings.

However, by using data from 2017/18, we cannot draw conclusions about equity in IAPT in the present day. For example, the impact of newer NHS initiatives tackling equity (e.g., Beck et al., 2019) or the impact of COVID-19 (e.g., Aragona et al., 2020) on IAPT access rates by socio-demographic characteristics, were not captured by this study.

Using a large sample size from the UKHLS survey data offered one of the best, most feasible ways to investigate probable CMD in the general population. This enabled the researchers to explore inequalities in IAPT access rates amongst socio-demographic characteristics in an unprecedented way. However, this dataset also has some flaws:

• UKHLS only includes private households and excludes households consisting entirely of immigrants before 2014/15. This raises doubts about how representative it is of ethnic minority groups or those living in places such as care homes or prisons.

• The General Health Questionnaire (GHQ-12) was used to “proxy probable CMD” in the UKHLS because it doesn’t have the validity to precisely capture CMDs. This leaves it susceptible to under-/over-estimating CMD rates.

With the absence of data from other mental health services (e.g., secondary care, work-based schemes or private), we also cannot gain a clear picture of socio-demographic inequalities in access to mental health support more broadly.

Finally, with such a rich demographic dataset, this study missed opportunities to analyse intersectionality. Previous studies indicate that having multiple vulnerable socio-demographic characteristics exacerbates inequalities in mental health support (Bhopal, 2020), meaning sub-groups were likely missed from this study.

Implications for practice

As this study has limitations to its datasets, it is not advisable to base policy or practice changes solely on this data. However, the results do corroborate more recent research on inequalities of access to IAPT (e.g., Harwood et al., 2023; Smyth et al., 2022), adding weight to the case for exploring systemic changes to NHS Talking Therapies.

Incentivising services through schemes such as the ‘Quality Premium’ reimbursement scheme and the “Commissioning for Quality and Innovation” initiative have aimed to improve some equity disparities (e.g., access rates of older adults or those with long-term conditions) (NHS, 2018). These performance-related pay schemes financially reward services and commissioning boards who achieve quality care outcomes and reduce inequalities (influenced by the Five Year Forward View NHS mandate), on a local level. This study may encourage commissioning boards to explore opportunities to expand schemes and help tackle a broader range of socio-demographic barriers to accessing IAPT.

This study’s results also add evidence for trying more radical approaches to improving access within IAPT services. Specifically, transformational organisational approaches, in which co-collaboration is used to fundamentally change the way services are developed from the outset, have gained more attention recently (Smith et al., 2023). By ensuring that patients, communities, and carers can co-design services, it could be more likely that services will become more accessible and favourable to people experiencing barriers. The new “Patient and Carer Race Equality Framework” aims to address this challenge for racial discrimination (Dyer et al., 2020). It will be interesting to see if this framework can be used effectively and eventually, more broadly, to overcome other socio-demographic barriers.

Finally, this study opens several interesting avenues for further studies to help explore equity in IAPT:

• Using a qualitative approach would offer a rich, detailed account of the perception of IAPT services and why vulnerable groups do not refer.
• Exploring intersectionality within this approach to ensure sub-groups are not missed.
• Redoing this study using 2021 Census/IAPT data to capture a more recent, compelling picture of the landscape of equity in IAPT today.

Having worked in several IAPT services for the past 2 years, I am truly honoured to have the responsibility of guiding people through therapy, often for their first experience of support. However, I may not have quite appreciated the bravery and challenge of accessing services for people faced with socio-demographic barriers. Sharland and colleagues’ study highlights the need to work harder to reduce these barriers, as only then can IAPT truly say it has met its aims.

Statement of interests

None.

Links

Primary paper

Sharland, E., Rzepnicka, K., Schneider, D., Finning, K., Pawelek, P., Saunders, R., & Nafilyan, V. (2023). Socio-demographic differences in access to psychological treatment services: evidence from a national cohort study. Psychological Medicine, 1-12.

Other references

Ahmad, G., McManus, S., Cooper, C., Hatch, S. L., & Das-Munshi, J. (2022). Prevalence of common mental disorders and treatment receipt for people from ethnic minority backgrounds in England: repeated cross-sectional surveys of the general population in 2007 and 2014. The British Journal of Psychiatry, 221(3), 520-527.

Aragona, M., Barbato, A., Cavani, A., Costanzo, G., & Mirisola, C. (2020). Negative impacts of COVID-19 lockdown on mental health service access and follow-up adherence for immigrants and individuals in socio-economic difficulties. Public Health, 186, 52-56.

Beck, A., Naz, S., Brooks, M., & Jankowska, M. (2019). Black, Asian and Minority Ethnic service user positive practice guide 2019. BABCP.

Bhopal, R. S. (2020). COVID-19: Immense necessity and challenges in meeting the needs of minorities, especially asylum seekers and undocumented migrants. Public Health, 182, 161.

Clark, D. M. (2011). Implementing NICE guidelines for the psychological treatment of depression and anxiety disorders: the IAPT experience. International Review of Psychiatry, 23(4), 318-327.

Cromarty, P., Drummond, A., Francis, T., Watson, J., & Battersby, M. (2016). NewAccess for depression and anxiety: adapting the UK improving access to psychological therapies program across Australia. Australasian Psychiatry, 24(5), 489-492.

Dyer J, Murdoch C and Farmer P. (2020, October 16). Advancing mental health equalities strategy. NHS England.

Harwood, H., Rhead, R., Chui, Z., Bakolis, I., Connor, L., Gazard, B., … & Hatch, S. L. (2023). Variations by ethnicity in referral and treatment pathways for IAPT service users in South London. Psychological Medicine, 53(3), 1084-1095.

Macintyre, A., Ferris, D., Gonçalves, B., & Quinn, N. (2018). What has economics got to do with it? The impact of socioeconomic factors on mental health and the case for collective action. Palgrave Communications, 4(1), 1-5.

McManus, S., Bebbington, P. E., Jenkins, R., & Brugha, T. (2016). Mental health and wellbeing in England: the adult psychiatric morbidity survey 2014. NHS digital.

Naeem, F., Pikard, J., Rao, S., Ayub, M., & Munshi, T. (2017). Is it possible to provide low-intensity cognitive behavioral treatment (CBT Lite) in Canada without additional costs to the health system? First-year evaluation of a pilot CBT Lite program. International Journal of Mental Health, 46(4), 253-268.

NHS (2018). Technical Guidance Annex B Information on Quality Premium. NHS England.

Smith, S. M., Kheri, A., Ariyo, K., Gilbert, S., Salla, A., Lingiah, T., … & Edge, D. (2023). The Patient and Carer Race Equality Framework: a model to reduce mental health inequity in England and Wales. Frontiers in Psychiatry, 14, 1053502.

Smyth, N., Buckman, J. E., Naqvi, S. A., Aguirre, E., Cardoso, A., Pilling, S., & Saunders, R. (2022). Understanding differences in mental health service use by men: an intersectional analysis of routine data. Social Psychiatry and Psychiatric Epidemiology, 57(10), 2065-2077.

Spiers, N., Qassem, T., Bebbington, P., McManus, S., King, M., Jenkins, R., … & Brugha, T. S. (2016). Prevalence and treatment of common mental disorders in the English national population, 1993–2007. The British Journal of Psychiatry, 209(2), 150-156.

Wakefield, S., Kellett, S., Simmonds‐Buckley, M., Stockton, D., Bradbury, A., & Delgadillo, J. (2021). Improving Access to Psychological Therapies (IAPT) in the United Kingdom: A systematic review and meta‐analysis of 10‐years of practice‐based evidence. British Journal of Clinical Psychology, 60(1), 1-37.

Photo credits

• Photo by Danie Franco on Unsplash
• Photo by Marcel Strauß on Unsplash
• Photo by Danie Franco on Unsplash
• Photo by Matt Bennett on Unsplash
• Photo by Alexander Hafemann on Unsplash
• Photo by Jon Tyson on Unsplash