Publication, Part of National Diabetes Audit
National Diabetes Audit 2021-22, Report 1: Care Processes and Treatment Targets, Detailed Analysis Report
Audit, Survey, Other reports and statistics
Change to mapping used for integrated care board (ICB) analysis
An issue was found with the mapping used to define integrated card boards (ICB) in this report resulting in larger than expected cohorts for each ICB. This has been corrected in this report and has resulted in slight changes to the results of analyses broken down by ICB.
21 December 2023 00:00 AM
Health inequalities in diabetes, 2017-18 to 2021-22
Core20PLUS5
Core20PLUS5 is a national NHS England approach to inform action to reduce healthcare inequalities at both national and system level. The approach defines a target population – the ‘Core20PLUS’ – and identifies ‘5’ focus clinical areas requiring accelerated improvement.
Receiving all 8 care processes - Contributing factors
Logistic regression was performed on all cohorts from 2017-18 to 2021-22 for both type 1 and type 2 diabetes to examine the relation between the outcome variable (receipt of all 8 care processes) and patient characteristics.
The multivariable models have been explored to quantify the contributions of different personal characteristics to the receipt of care processes. The models have a weak predictive ability with a c statistic around 0.6, meaning the patient characteristics in the model do not greatly contribute to whether a person receives all 8 care processes or not. Therefore findings from these models should be treated with caution.
The models below show a variation in receiving care processes associated with age, sex, duration of diabetes, deprivation and ethnicity. The personal characteristic that has the greatest adverse influence is young age.
Notes:
1. See Statistical terms in Additional information for explanations of logistic regression and c statistic.
2. For the logistic regression models, type 2 diabetes only includes people with a diagnosis of type 2 diabetes. See Definitions in Additional information for details of diabetes types.
3. The strength of association is based on odds ratio which is plotted on the x-axis in the forest plots. For explanation of odds ratio see Statistical terms in Additional information. See Appendix for forest plots.
Type 1 diabetes
- Younger age (when compared to reference group 40-49 years)
- Living in a more deprived area (when compared to reference group second most deprived quintile) (1)
- Older age (when compared to reference group 40-49 years)
- Asian, black or mixed ethnicity (when compared to reference group white ethnicity) (2)
- Living in a less deprived area (when compared to reference group second most deprived quintile)
- Diabetes duration of 1-4 years and 5-9 years (when compared to reference group diabetes duration of 15 years and over) (3)
Notes:
1. Except in 2018-19.
2. Only Asian and black ethnicity was associated with an increased likelihood in 2017-18.
3. Only a diabetes duration of 1-4 years was associated with an increased likelihood in 2018-19.
Type 2 diabetes
- Younger age (when compared to reference group 40-49 years)
- Living in a more deprived area (when compared to reference group second most deprived quintile) (1)
- Female sex (when compared to reference group male sex)
- Diabetes duration of less than a year (when compared to reference group diabetes duration of 15 years and over)
- Older age (when compared to reference group 40-49 years)
- Asian, black or mixed ethnicity (when compared to reference group white ethnicity) (2)
- Living in a less deprived area (when compared to reference group second most deprived quintile)
- Diabetes duration of 10-14 years (when compared to reference group diabetes duration of 15 years and over)
Notes:
1. Except in 2017-18 and 2018-19.
2. Only Asian and black ethnicity were associated with an increased likelihood in 2017-18. Mixed ethnicity was associated with a reduced likelihood in 2017-18.
Notes:
1. The strength of association is based on odds ratio which is plotted on the x-axis in the forest plots. See Appendix for forest plots.
2. For explanations of odds ratio and reference groups see Statistical terms in Additional information.
Frailty
Frailty describes a dynamic state of increased vulnerability to adverse health outcomes resulting from loss of physiological reserve. Please see the Methodology section for further details on frailty.
Table 1: People with a valid HbA1c reading, by diabetes type and by whether they were with severe frailty, low HbA1c or both, England and Wales, 2021-22
| Diabetes type | Number of people with a valid HbA1c reading | Percentage with severe frailty (%) | Percentage with HbA1c ≤ 53 mmol/mol (%) | Percentage with HbA1c ≤ 53 mmol/mol and severe frailty (%) |
| Type 1 | 216,510 | 1.9 | 23.0 | 0.4 |
| Type 2 and other | 3,123,120 | 5.1 | 49.3 | 2.8 |
Table 2: People with type 2 and other diabetes, a valid HbA1c reading and prescribed insulin, sulphonylurea or both, by whether they were with severe frailty, had low HbA1c or both, England and Wales, 2021-22
| Drugs prescribed | Number of people with a valid HbA1c reading | Percentage with severe frailty (%) | Percentage with HbA1c ≤ 53 mmol/mol (%) | Percentage with HbA1c ≤ 53 mmol/mol and severe frailty (%) |
| Insulin | 431,750 | 8.9 | 18.3 | 2.0 |
| Sulphonylurea | 614,140 | 4.9 | 22.7 | 1.4 |
| Both | 88,310 | 7.9 | 11.6 | 1.2 |
Notes:
1. Prescription data for Semaglutide (GLP-1 receptor agonist), Ertugliflozin (SGLT2 inhibitor), Empagliflozin with linagliptin (SGLT2 inhibitor with DPP-4 inhibitor), and Saxagliptin with dapagliflozin (DPP-4 inhibitor with SGLT2 inhibitor) has not been included within the NDA drugs data collection. This does not affect the above analysis and work is ongoing to ensure all glucose-lowering medication are collected and included in future publications.
2. The methodology underlying this analysis has been updated to align it with other areas of the report. Therefore the figures presented in this report are not comparable with those previously published.
- In 2021-22, as in 2020-21, the frailty analysis shows substantial numbers of very frail people in whom treatment does not seem to have been relaxed in line with NICE guidance (1).
- 2.0% of people with type 2 and other diabetes who were prescribed insulin were with severe frailty and had an HbA1c ≤ 53 mmol/mol.
- 1.4% of people with type 2 and other diabetes who were prescribed sulphonylurea were with severe frailty and had an HbA1c ≤ 53 mmol/mol.
Notes:
1. NICE guidance - Blood glucose management.
Frailty and receipt of care processes
Frailty and treatment target achievement
Notes:
1. As part of disclosure control, the percentage of people with type 1 diabetes and severe frailty in Wales achieving all 3 treatment targets is not calculated as the rounded denominator is 20 or less. For more information on disclosure control see the Disclosure control section in Methodology.
2. The methodology underlying this analysis has been updated to align it with other areas of the report. Therefore the figures presented in this report are not comparable with those previously published.
- People with type 1 diabetes and moderate or severe frailty were more likely to receive all 8 care processes than those with no frailty recorded. 44.4% of people with type 1 diabetes in England and 25.0% of those in Wales with moderate or severe frailty received all 8 care processes; and 33.8% of people with type 1 diabetes in England and 15.6% of those in Wales with no frailty received all 8 care processes.
- People with type 2 and other diabetes who were severely frail were almost as likely to receive all 8 care processes than those with no frailty recorded. 38.7% of people with type 2 and other diabetes in England and 30.2% of those in Wales with severe frailty received all 8 care processes; and 47.1% of people with type 2 and other diabetes in England and 30.0% of those in Wales with no frailty received all 8 care processes.
- 20.1% of people with type 1 diabetes and moderate or severe frailty in England and 15.4% of those in Wales achieved all 3 NICE recommended treatment targets.
- 38.7% of severely frail people with type 2 and other diabetes in England and 31.4% of those in Wales achieved all 3 NICE recommended treatment targets.
Learning Disability
Learning disability and receipt of care processes
Learning disability and treatment target achievement
Notes:
1. SNOMED codes in the GP data are used to identify diagnosis of a learning disability. SNOMED CT is a clinical terminology (Systematized Nomenclature of Medicine (Clinical Terms)).
2. See the section Learning disabilities in Methodology for details of standardisation.
- In people with type 1 diabetes, those with a learning disability were as likely as others to receive all key care processes, but less likely to achieve their treatment targets. Only 18.3% of people with type 1 diabetes and a learning disability achieved all 3 treatment targets in England compared to 22.4% of all people in the NDA with type 1 diabetes in England. Only 14.1% of people with type 1 diabetes and a learning disability achieved all 3 treatment targets in Wales compared to 15.9% of all people in the NDA with type 1 diabetes in Wales.
- In people with type 2 and other diabetes, those with a learning disability were less likely to receive all key care processes, but more likely to achieve their treatment targets. Only 38.6% of people with type 2 and other diabetes and a learning disability received all 8 care processes in England compared to 47.9% of all people in the NDA with type 2 and other diabetes in England. Only 18.9% of people with type 2 and other diabetes and a learning disability received all 8 care processes in Wales compared to 30.2% of all people in the NDA with type 2 and other diabetes in Wales.
Severe mental illness (SMI)
SMI and receipt of care processes
SMI and treatment target achievement
Notes:
1. SMI refers to people with psychological problems that are often so debilitating that their ability to engage in functional and occupational activities is severely impaired. SNOMED codes in the GP data are used to identify diagnosis of a severe mental illness. SNOMED CT is a clinical terminology (Systematized Nomenclature of Medicine (Clinical Terms)).
- A greater percentage of people with type 1 diabetes and a SMI in England received all 8 care processes when compared to all people with type 1 diabetes in the NDA in England (37.3% of those with a SMI compared to 35.2% of all people in the NDA in England). A lower percentage of people with type 1 diabetes and a SMI in Wales received all 8 care processes when compared to all people with type 1 diabetes in the NDA in Wales (11.9% of those with a SMI compared to 15.7% of all people in the NDA in Wales).
- A lower percentage of people with type 2 and other diabetes and a SMI received all 8 care processes when compared to all people with type 2 and other diabetes in the NDA (41.0% of those with a SMI compared to 47.9% of all people in the NDA in England; and 22.5% of those with a SMI compared to 30.2% of all people in the NDA in Wales).
- Conversely, for those with type 1 diabetes and a SMI, a lower percentage of people achieved all 3 treatment targets when compared to all people with type 1 diabetes in the NDA (16.4% of those with a SMI compared to 22.4% of all people in the NDA in England; and 8.3% of those with a SMI compared to 15.9% of all people in the NDA in Wales).
- For people with type 2 and other diabetes, there is not a great difference in the percentage of people achieving all 3 treatment targets between those with a SMI and all people with type 2 and other diabetes in the NDA (35.5% of those with a SMI compared to 35.7% of all people in the NDA in England; and 29.9% of those with a SMI compared to 28.4% of all people in the NDA in Wales).
Last edited: 12 April 2024 3:37 pm