1. Main facts and figures

  • in 2017/18, for all ethnic groups, the majority of referrals for treatment through Improving Access to Psychological Therapies (IAPT) showed an improvement on finishing their treatment
  • women’s IAPT referrals were more likely to have shown improvement than men’s, with the exception of Bangladeshi, Chinese and Black Caribbean women, where rates of improvement were either the same as, or lower than men’s
  • referrals of White patients through IAPT were more likely to show improvement than other broad ethnic groups; referrals for Asian and Other including Chinese ethnic groups were the least likely to show improvement – this was found for both men and women
  • the percentage of referrals showing improvement has increased across all ethnicities between 2015/16 and 2017/18, and the gap between groups has narrowed, with the percentage of referrals showing improvement increasing more for the Asian and Other including Chinese groups than for the White group
Things you need to know

About 1 million referrals start treatment each year through Improving Access to Psychological Therapies (IAPT) services. IAPT is an adult service, but contains a small amount of referrals for patients who are under 18.

IAPT data is based on monthly submissions to NHS Digital by the providers of these services, who are required to record patients’ ethnicities. The proportion of patients with a recorded ethnicity has risen gradually, from about 80% in April 2015 to about 89% in March 2018. This analysis only includes referrals for which a valid ethnicity was recorded.

The data presented is national data. The ‘download the data' section contains data at the local level by clinical commissioning groups, although some data has been left out as the small number of patients in some ethnic groups makes it unreliable.

The definition for ‘no reliable change’ has been revised since the previous publication of this measure page. The new methodology separates out patient referrals which showed no reliable change from those where it wasn’t possible to assess change due to missing data. All data presented here is now based on the new methodology.

What the data measures

This data measures outcomes for all referrals for treatment for anxiety or depression made through the NHS’s Improving Access to Psychological Therapies (IAPT) services. The data is broken down by ethnicity, gender and time. Data showing the change over time for different ethnic groups by gender is available in the download file.

Following treatment, patients’ conditions are assessed as showing either:

  • improvement
  • no change
  • deterioration

Note that these categories do not quite add up to 100% because the percentage of referrals with an unknown outcome is not shown.

Outcomes are assessed using patients’ responses to questionnaires that measured their depression and anxiety over the course of their treatment. These give a clinically reliable basis for determining patients’ outcomes.

A referral must have ended in the financial year and the patient must have attended at least 2 appointments in the course of the referral.

The ethnic categories used in this data

The 16 categories used in this data are those listed in the 2001 Census. The categories are broadly the same as those used in the 2011 Census, with the following exceptions:

  • the 2001 Census categorised Chinese people within 'Other ethnic group'
  • the 2001 Census did not list Gypsy and Irish Travellers or Arab people
  • The 2001 categories are therefore as follows:

White:

  • English/Welsh/Scottish/Northern Irish/British
  • Irish
  • Any Other White background

Mixed/Multiple ethnic groups:

  • White and Black Caribbean
  • White and Black African
  • White and Asian
  • Any Other Mixed/Multiple ethnic background

Asian/Asian British:

  • Indian
  • Pakistani
  • Bangladeshi
  • Any Other Asian background

Black/African/Caribbean/Black British:

  • African
  • Caribbean
  • Any Other Black/ African/ Caribbean background

Other ethnic group:

  • Chinese
  • Any Other ethnic group

2. By ethnicity

Percentage of referrals showing improvement, deterioration or no change following a course of psychological therapy, by ethnicity
Ethnicity Reliably improved No reliable change Reliably deteriorated
% % %
Asian 62.7 28.4 7.5
Bangladeshi 61.7 28.1 8.6
Indian 65.3 27.1 6.4
Pakistani 61.3 29.1 8.3
Asian other 60.4 30.0 8.1
Black 63.9 27.3 7.3
Black African 63.1 27.9 7.5
Black Caribbean 64.5 26.9 7.3
Black other 64.2 27.1 7.2
Mixed 63.2 28.3 7.2
Mixed White/Asian 62.7 28.8 7.2
Mixed White/Black African 64.1 26.8 7.9
Mixed White/Black Caribbean 63.7 27.9 7.3
Mixed other 62.8 28.7 6.9
White 67.2 26.2 5.7
White British 67.3 26.2 5.7
White Irish 65.8 26.7 6.6
White other 65.6 27.0 6.3
Other inc Chinese 61.7 28.8 8.0
Chinese 65.4 27.7 5.9
Any other 61.0 29.0 8.4

Download table data (CSV) Source data (CSV)

Summary

This data shows that for IAPT services in 2017/18:

  • referrals for White patients were more likely to show improvement following therapy (67.2%) than for other ethnic groups
  • the referrals least likely to show improvement were for the Asian (62.7%) and Other including Chinese (61.7%) broad ethnic groups
  • for specific ethnic groups, referrals for the White British and the White Irish ethnic groups were the most likely to show improvement (67.3% and 65.8% respectively) and also the least likely to show no change (26.2% and 26.7% respectively)
  • referrals for Other Asian (60.4%), Pakistani (61.3%), Bangladeshi (61.7%) and Any Other (61.0%) specific ethnic groups were least likely to show improvement
  • 8.6% of referrals for Bangladeshi patients showed deterioration – the highest percentage for any ethnic group
  • referrals for White British (5.7%) and Chinese (5.9%) patients were the least likely to show deterioration
  • referrals for patients from the Other Asian ethnic group were the most likely to show no change (30.0%)

3. By ethnicity over time

Percentage of referrals showing improvement, deterioration or no change following a course of psychological therapy, by ethnicity over time
Reliably improved (%) No reliable change (%) Reliably deteriorated (%)
Ethnicity 2015/16 2016/17 2017/18 2015/16 2016/17 2017/18 2015/16 2016/17 2017/18
Asian 56.7 61.1 62.7 31.7 29.6 28.4 8.4 7.8 7.5
Bangladeshi 53.9 58.9 61.7 32.0 29.7 28.1 10.7 10.1 8.6
Indian 59.2 64.0 65.3 30.7 28.1 27.1 7.4 6.7 6.4
Pakistani 54.9 59.1 61.3 32.5 30.8 29.1 9.1 8.3 8.3
Asian other 55.2 59.1 60.4 32.3 30.8 30.0 8.7 8.2 8.1
Black 59.8 62.5 63.9 29.4 28.7 27.3 8.2 7.4 7.3
Black African 58.8 61.7 63.1 29.6 28.9 27.9 8.5 7.8 7.5
Black Caribbean 60.8 63.3 64.5 29.1 28.5 26.9 7.7 7.1 7.3
Black other 58.6 61.8 64.2 30.0 29.2 27.1 8.7 7.5 7.2
Mixed 58.6 62.4 63.2 30.9 29.1 28.3 7.3 6.7 7.2
Mixed White/Asian 58.2 63.5 62.7 31.5 28.8 28.8 7.4 6.0 7.2
Mixed White/Black African 60.8 61.0 64.1 28.4 30.5 26.8 7.8 6.5 7.9
Mixed White/Black Caribbean 58.8 61.5 63.7 31.2 30.2 27.9 7.0 6.8 7.3
Mixed other 58.0 62.9 62.8 30.9 27.9 28.7 7.5 7.0 6.9
White 63.5 66.0 67.2 28.3 26.9 26.2 6.0 5.8 5.7
White British 63.6 66.1 67.3 28.2 26.9 26.2 5.9 5.8 5.7
White Irish 63.8 65.2 65.8 27.6 26.9 26.7 5.9 6.5 6.6
White other 61.9 63.7 65.6 28.8 28.2 27.0 6.8 6.7 6.3
Other inc Chinese 56.4 59.0 61.7 30.8 30.5 28.8 8.1 8.6 8.0
Chinese 60.6 64.5 65.4 29.4 26.9 27.7 6.6 7.4 5.9
Any other 55.7 57.9 61.0 31.1 31.2 29.0 8.4 8.8 8.4

Download table data (CSV) Source data (CSV)

Summary

This data shows that:

  • the percentage of referrals showing improvement has increased across all ethnicities between 2015/16 and 2017/18
  • the largest increase in percentage of referrals showing improvement has been for Asian referrals; this group had the lowest percentage of referrals showing improvement in 2015/16 (from 56.7% in 2015/16 to 62.7% in 2017/18)
  • the smallest increase has been for White referrals; this group had the highest percentage of referrals showing improvement in 2015/16 (from 63.5% in 2015/16 to 67.2% in 2017/18)
  • the gap between the percentage of White referrals showing improvement and that of referrals for the Asian or Other including Chinese groups has narrowed: from 6.8 percentage points in 2015/16 to 4.5 percentage points in 2017/18 for Asian referrals and from 7.1% to 5.5% for the Other including Chinese group
  • among the specific ethnic groups, the largest increase in percentage of referrals showing improvement has been for Bangladeshi referrals - from 53.9% in 2015/16 to 61.7% in 2017/18 and the smallest increase has been for White Irish referrals - from 63.8% in 2015/16 to 65.8% in 2017/18

4. By ethnicity and gender

Percentage of referrals showing improvement, deterioration or no change following a course of psychological therapy, by ethnicity and gender
Reliably improved No reliable change Reliably deteriorated
Ethnicity Men (%) Women (%) Men (%) Women (%) Men (%) Women (%)
Asian 61.5 63.4 29.3 28.0 7.6 7.5
Bangladeshi 62.8 61.0 28.1 28.0 7.1 9.4
Indian 64.5 65.7 28.1 26.7 5.9 6.6
Pakistani 60.6 61.7 29.0 29.1 8.7 8.1
Asian other 57.2 62.3 32.1 28.9 9.1 7.5
Black 63.6 64.1 28.0 27.0 6.5 7.7
Black African 62.7 63.2 28.9 27.5 6.3 7.9
Black Caribbean 64.5 64.5 27.0 26.9 6.9 7.5
Black other 63.4 64.5 28.6 26.5 6.0 7.6
Mixed 61.8 63.9 29.8 27.6 6.9 7.3
Mixed White/Asian 61.4 63.3 30.4 28.1 7.2 7.2
Mixed White/Black African 61.3 65.2 29.3 25.8 7.9 8.0
Mixed White/Black Caribbean 61.6 64.6 29.6 27.2 7.1 7.4
Mixed other 62.1 63.1 29.7 28.3 6.3 7.2
White 66.5 67.6 27.0 25.8 5.6 5.8
White British 66.6 67.7 26.9 25.8 5.6 5.7
White Irish 65.7 66.0 27.0 26.4 6.1 6.7
White other 64.4 66.0 28.2 26.5 6.1 6.4
Other inc Chinese 59.9 62.6 29.2 28.6 8.9 7.6
Chinese 67.6 64.6 24.5 28.8 6.7 5.6
Any other 58.7 62.2 29.9 28.5 9.3 8.0

Download table data (CSV) Source data (CSV)

Summary

This data shows that for IAPT services in 2017/18:

  • amongst the broad ethnic groups, the highest rate of improvement for women’s referrals was for the White ethnic group at 67.6%; the lowest was for the Asian and Other ethnic groups, at 63.4% and 62.6%
  • similarly, amongst the broad ethnic groups, the highest rate of improvement for men’s referrals was for the White ethnic group at 66.5%, while the Asian and Other ethnic groups had the lowest rates of improvement, at 61.5% and 59.9%
  • women’s referrals were more likely to show improvement than men’s in every ethnic group except for the Chinese, Bangladeshi and Black Caribbean groups
  • for women’s referrals, the specific ethnic groups most likely to show improvement were White British (67.7%) White Irish and White Other (both 66.0%); for men’s referrals, the specific ethnic groups most likely to show improvement were Chinese (67.6%) and White British (66.6%)
  • for women’s referrals, the specific ethnic groups most likely to show deterioration were Bangladeshi (9.4%) and Pakistani (8.1%) for men’s referrals, the specific ethnic groups most likely to show deterioration were Other Asian (9.1%) and Any Other (9.3%) ethnic groups

5. Methodology

Data is submitted by providers of Improving Access to Psychological Therapies (IAPT) services on a monthly basis. This means that information about referrals is updated each month.

Ethnic group is a required field, which means that wherever providers hold or collect this information about a person, it should be included in their submission to NHS Digital.

For all referrals, a Patient Health Questionnaire (PHQ-9) is used, alongside either a Generalised Anxiety Disorder (GAD7) questionnaire or an Anxiety Disorder Specific Measure (ADSM), depending on the recorded diagnosis. ADSMs are questionnaires that are sensitive measures of the severity of particular anxiety disorders. The IAPT Data Handbook recommends relevant ADSMs for:

  • obsessive-compulsive disorder
  • generalised anxiety disorder
  • social phobia
  • health anxiety
  • agoraphobia
  • panic disorder
  • post-traumatic stress disorder

If one of these conditions is recorded, the relevant ADSM should be used to measure change in anxiety during treatment. If the relevant ADSM has not been given at least twice during a course of treatment, or if the condition recorded is not one of those specified in the above list, the GAD7 (IAPT’s generic anxiety measure) is used to assess change in anxiety.

The patient’s first and last scores are compared for both the PHQ-9 and the relevant anxiety questionnaire. A referral has shown reliable improvement if there is a reliable decrease on one or both of the scores on these questionnaires, and no reliable increase in either of them. A reliable change is defined as either a PHQ-9 change of 6 points or more, or one of the following, depending on which questionnaire is asked:

  • GAD7 - a change of 4 points or more
  • Agoraphobia Mobility Inventory - a change of 0.73 or more
  • Social Phobia Inventory - a change of 10 points or more
  • Panic Disorder Severity Scale does not have an agreed threshold so GAD7 is used
  • Obsessive Compulsive Inventory - a change of 32 points or more
  • Impact of Events Scale - a change of 9 points or more
  • Health Anxiety Inventory (Short Week) - a change of 4 points or more

Suppression rules and disclosure control

In order to protect patient confidentiality in IAPT publications, any figures based on a count of less than 5 referrals are suppressed.

Rounding

In order to prevent suppressed numbers from being calculated, all sub-national counts have been rounded to the nearest 5. Percentages are based on unrounded numbers and have been rounded to 1 decimal point at England level and the nearest whole number at Clinical Commissioning Group (CCG) level.

Related publications

Adult improving access to psychological therapies programme

Quality and methodology information

Further technical information

www.digital.nhs.uk/iaptreports

6. Data sources

Source

Type of data

Administrative data

Type of statistic

Official statistics

Publisher

NHS Digital

Publication frequency

Yearly

Purpose of data source

The data is used to to monitor and improve the provision of talking therapies services for the treatment of people with anxiety and/or depression.

7. Download the data

Outcomes for treatment for anxiety and depression - Spreadsheet (csv) 23 MB

This file contains the following: ethnicity, year, geography, gender, outcome, value, denominator, numerator