rcpch logo npda logo hqip logo

National Paediatric Diabetes Audit (NPDA) Workforce Spotlight Report 2024/25

1 Workforce Spotlight Extended Analysis

This report describes the available workforce providing care to paediatric diabetes units (PDUs)1 in England, Wales and Jersey as at 31st March 2025. Some PDUs share workforce and resource; we invited these PDUs to submit jointly. In total, 161 submissions to the Workforce Spotlight Audit were received, covering 100% of PDUs (172/172) who participate in the NPDA.

1 A clinical unit providing diabetes care for children and young people with diabetes. In this report, a PDU can refer to one or more units who submitted jointly.

Results are compared to the 2017/18 Workforce Spotlight Audit for metrics that have been repeated in both audits. Where possible, data from this Workforce Spotlight Audit were linked to PDU-level results from the 2024/25 NPDA Report on Care and Outcomes (referred to as the ‘2024/25 core audit’). Some values from the 2024/25 core audit have been masked due to small numbers, in line with the RCPCH guidance on managing data disclosure risks.

Due to inconsistency in how data was submitted, some responses have been transformed. For example, “approx 5”, “at least 5”, “5 but need to check” are transformed as “5”; “Not known” or other invalid responses are removed.

1.1 Clinic Attendance

60.9% of PDUs (98/161) offered all patients with Type 1 or Type 2 at least four appointments with a consultant in the 2024/25 audit year (Figure 1).

Figure 1. Percentage of PDUs offering at least four appointments with a consultant to all patients in England, Wales, and Jersey in 2017/18 and 2024/25. N.B. Jersey was not included in the 2017/18 Workforce Spotlight Audit.

129 PDUs had a known Was Not Brought (WNB)2 rate. The average WNB rate was 10.5%, ranging from 2.5% to 38.0% (Figure 2). At PDU level, WNB rates were significantly associated with health check completion rates (Figure 3). Due to the nature of the WNB rate, comparison against median HbA1c, which requires patient attendance, was not applicable.

2 Children and young people with diabetes who did not attend and had not cancelled a planned appointment.

Figure 2. WNB rates for all PDUs in England and Wales (no data for Jersey).

The coefficient (\(b\)) shows the magnitude and direction of the association between the predictor and outcome. A 1‑unit increase in predictor (\(x\)) is associated with an average \(b\)‑unit change in the outcome (\(y\)), holding other variables in the model constant.

The confidence interval (\(95\% CI\)) reflects the range of plausible values for the true effect. A narrower CI indicates a more precise estimate, while a wider CI reflects more uncertainty. If the CI includes 0, the result may be compatible with no effect. In the scatter plot, the \(95\% CI\) ribbon represents the uncertainty around the estimated mean outcome \(y\) at each value of the predictor \(x\). The ribbon does not represent the range of individual values, but rather the uncertainty around the model’s fitted line. The ribbon also differs from the reported \(95\% CI\) for a coefficient, which summarises uncertainty in the overall effect estimate (\(b\)) across the dataset.

The p-value (\(p\)) indicates strength of evidence against “no association” (null hypothesis). A small p-value (typically \(p<0.05\)) suggests the observed association is unlikely due to chance alone. However, it does not reflect the magnitude or clinical importance of the effect.

As differences in patient numbers and case-mix may influence outcomes across units, in each model, all models were adjusted for unit-level characteristics including total patients on 31st March, unit’s mean age, mean diagnosis duration, proportion of White patients, and proportion living in most deprived areas. The adjustments helps ensure that any observed associations are less likely to be driven by differences in patient populations between units.

The R-squared (\({R}^{2}\)) indicates the proportion of variation in the outcome explained by the full model. For example, \({R}^{2}=0.10\) means that 10% of variability in the outcome is accounted for by all predictors in the model. On the other hand, partial R-squared (\(partial\ {R}^{2}\)) indicates the proportion of variation in the outcome explained by the predictor of interest, after accounting for other predictors in the model. In this report, \(partial\ {R}^{2}=0.05\) means that the predictor of interest explains 5% of the variation in the outcome, after adjusting for unit size and their case-mix. Higher values indicate better model fit, but it does not necessarily imply clinical importance.

Figure 3. Relationship between WNB rates and health check completion at PDU-level.

Definition of children and young people on 31st March is provided in Glossary.

In England and Wales, the Healthcare Travel Costs Scheme (HTCS) allows eligible families to claim a refund of travel costs when attending hospital appointments. 32.9% of PDUs (n = 53) routinely informed families of funding they may access for costs associated with attending hospital appointments.

1.2 Additional Care Needs

1.2.1 Child Protection

90.7% of PDUs (n = 146) reported data on the number of children and young people with diabetes under their care who were considered a ‘child in need3’, on the child protection register (Wales and Jersey) or had a child protection plan (England)4, and/or considered ‘child in care5’.

3 A child who is thought to need extra help from children’s services to achieve or maintain a reasonable standard of health or development.

4 In England, a child protection plan is a formal written agreement created when a child is assessed to be at risk of significant harm from abuse. In Wales and Jersey, the child protection register is a confidential list of all children who have been identified as being at risk of significant harm.

5 A child who has been “looked after” by the local authority (England and Wales) or the minister (Jersey) for a continuous period of more than 24 hours.

Children and young people managed within a PDU were less likely to be a ‘child in need’, but more likely to be a ‘child in care’ or on a child protection plan/register. This is consistent with the findings of the 2017/18 Workforce Spotlight Audit.

Table 1 provides a breakdown of the reported proportion of children and young people being cared for by PDUs based on data from PDUs who knew their numbers. Where available, for comparison, rates of children and young people from the general population are reported. Figure 4 shows the rate per 10,000 compared to the general population rate and also compared to the 2017/18 Workforce Spotlight Audit.

Country
Category England and Walesa England Wales Jerseya
Children in need
% of services who knew the number 83.1% (133/160) 83.3% (125/150) 80.0% (8/10) 100.0% (1/1)
% of children in services with a known response 2.5% (643/25628) 2.6% (634/24606) 0.9% (9/1022) * (*/44)
Rate per 10,000 children in services with a known response 250.9 257.7 88.1 *
Rate per 10,000 children aged under 18 in the general population - 330.3b 282.6d -
Child Protection Plan / Child Protection Register
% of services who knew the number 84.4% (135/160) 84.7% (127/150) 80.0% (8/10) 100.0% (1/1)
% of children in services with a known response 0.9% (238/26040) 0.9% (223/24933) 1.4% (15/1107) * (*/44)
Rate per 10,000 children in services with a known response 91.4 89.4 135.5 *
Rate per 10,000 children aged under 18 in the general population - 40.6b 41.0def -
Children in care
% of services who knew the number 87.5% (140/160) 87.3% (131/150) 90.0% (9/10) 100.0% (1/1)
% of children in services with a known response 1.0% (286/27473) 1.0% (270/26190) 1.2% (16/1283) * (*/44)
Rate per 10,000 children in services with a known response 104.1 103.1 124.7 *
Rate per 10,000 children aged under 18 in the general population - 67.1c 116.1g -
a To preserve data utility and anonymity, England and Wales summary is shown while masking non-zero <4 patients.
b Children in need (Department for Education, 2025)
c Children looked after in England including adoptions (Department for Education, 2025)
d Children receiving care and support per 10,000 population aged under 18 (StatsWales, 2023)
e Children receiving care and support on the Child Protection Register by category of abuse and gender (StatsWales, 2023)
f Population estimates by local authority, year, sex and age (StatsWales, 2023)
g Children looked after on 31 March per 10,000 population aged under 18 (StatsWales, 2024)
h Full report published by Welsh Government (2024)
Table 1. Proportion of children and young people with diabetes in the child protection system.
Figure 4. Rate per 10,000 of children and young people with diabetes in the child protection system, compared to 2017/18.
Figure 5. Proportion of children and young people with diabetes in the child protection system by country and network.

1.2.2 Mental Health and Neurodiversity

90.7% of PDUs (n = 146) reported data on the number of children and young people with diabetes under their care who had a diagnosis of a learning disability/global developmental delay, autistic spectrum disorder, attention deficit hyperactivity disorder, and/or an eating disorders.

Table 2 provides a breakdown of the proportion of children and young people with one of the above diagnoses being cared for by PDUs based on data from PDUs who knew their numbers.

Country
Category England and Walesa England Wales Jerseya
Diagnosis of global developmental delay (under 5 years) / learning disability (over 5 years)
% of services who knew the number 80.6% (129/160) 81.3% (122/150) 70.0% (7/10) 100.0% (1/1)
% of children in services with a known response 3.1% (766/25108) 3.1% (742/24116) 2.4% (24/992) 0.0% (0/44)
Diagnosis of autistic spectrum disorder (ASD)
% of services who knew the number 81.2% (130/160) 82.7% (124/150) 60.0% (6/10) 100.0% (1/1)
% of children in services with a known response 7.0% (1739/24971) 7.0% (1711/24276) 4.0% (28/695) * (*/44)
Diagnosis of attention deficit hyperactivity disorder (ADHD)
% of services who knew the number 78.8% (126/160) 80.0% (120/150) 60.0% (6/10) 100.0% (1/1)
% of children in services with a known response 3.4% (826/24170) 3.4% (804/23475) 3.2% (22/695) * (*/44)
Diagnosis of eating disorder
% of services who knew the number 89.4% (143/160) 89.3% (134/150) 90.0% (9/10) 100.0% (1/1)
% of children in services with a known response 0.5% (157/28561) 0.5% (146/27242) 0.8% (11/1319) 0.0% (0/44)
a To preserve data utility and anonymity, England and Wales summary is shown while masking non-zero <4 patients.
b No comparable figures were found for the general paediatric population in England, Wales, and Jersey.
Table 2. Proportion of children and young people with diabetes with a mental health or neurodevelopmental disorder.
Figure 6. Proportion of children and young people with diabetes with a mental health or neurodevelopmental disorder by country and network.

1.2.3 Tertiary and Satellite Services

PICU6 and tertiary-level support7 are also defined in Glossary.

6 A specialist unit providing intensive care for children and young people who are experiencing severe medical conditions. Examples of diabetes emergencies include DKA.

7 Supports for children outside of their normal secondary level catchment area. This could include PICU services for diabetes emergencies (e.g., DKA), diabetes provision for children requiring other services such as surgery (day case or inpatient), and specialist outpatient management (e.g., high-level exercise advice, eating disorders management).

26.1% of s (n = 42) offered tertiary-level diabetes support for patients beyond their secondary catchment area (Figure 7). Additionally, 47.2% of s (n = 76) provide peripheral, rural, or cross-site clinics in addition to their main location.

Figure 7. Tertiary-level support provided by PDUs (multiple-choice).

1.3 Safeguarding

69.6% of (n = 112) were able to report the number of safeguarding referrals made in 2024/25. On average, made 5.9 referrals within the audit year (range 0 to 20). Figure 8 shows the team members generally responsible for safeguarding referrals and those that have protected time in their job plans for safeguarding.

Figure 8. Team members responsible for safeguarding and with protected time for safeguarding (can contain multiple choices).

1.4 Out of Hours Support

95.0% of PDUs (n = 153) provided access to 24/7 telephone advice on diabetes management for children and young people, parents, and/or carers. Figure 9 shows which team members are responsible for providing telephone advice within that offer 24/7 telephone advice.

Figure 9. Team members responsible for providing 24/7 telephone advice (can contain multiple choices).

1.5 Transition

93.8% of PDUs (n = 151) had a dedicated transition clinic or service, run jointly with adult diabetes . In comparison, this number was 94.2% (163/173) in 2017/18.

On average, these clinics accepted young people at age 16.2 (Figure 10). 82.0% of PDUs (n = 132) were able to transfer young people into a dedicated young adult diabetes service.

Figure 10. Age young people are accepted into transition clinics or services at PDU-level.

PDUs were asked to described the proportion of their patients transferring via each method, as shown in Figure 11. The most commonly reported way for young people to transfer to adult services was through a gradual process over 1-2 years meeting members of the adult team during the process.

On average, 72.8% young people were transferred through the gradual process. In comparison, 9.4% young people were through direct transfer, 15.5% were seen jointly by paediatric and adult diabetes clinic, and 2.3% were transferred through other methods.

In terms of the most common methodology in each PDU (with ties allowed), the gradual process was the most common in 76.2% of PDUs (n = 125), followed by joint clinic in 15.9% of PDUs (n = 26), direct transfer in 6.1% of PDUs (n = 10), and other methods in 1.8% of PDUs (n = 3).

Popularised by Allen et al. (2021), a raincloud plot shows the distribution, individual data points, and summary statistics in a single figure. It is designed to provide a more complete and transparent view of the data than traditional visualisations such as box plots, violin plots, or bar plots.

The “rain” consists of the raw (typically jittered) data points, with each dot representing a single observation. The “cloud” is a a half‑violin plot that displays the estimated probability density of the data. Finally, the box plot in the middle shows summary statistics, including the median, quartiles, and the range of the data. In this report, the mean is also displayed as an additional marker.

Interactive raincloud plots were created using the npdar R package (Fang, 2026).

Figure 11. Percentage of young people transferred to adult services by differing methodologies.

1.6 Structured Education

The below structured education programmes are defined in Glossary.

100.0% of PDUs (n = 161) delivered a structured education programme for children and young people diagnosed with Type 1 diabetes locally. Of these, 62.7% (n = 101) delivered a quality assured programme. Figure 12 shows a breakdown of structured education programmes delivered.

Figure 12. Percentage of PDUs delivering each structured education programme (multiple-choice).

PDUs that offered a quality assured structured education programme had lower median HbA1c on average (Figure 13). However, in the regression model where unit size and patient demographics were accounted for, the difference between PDUs that did and did not offer structured education programmed was not statistically significant, and there was no evidence of an overall difference between the three groups.

ANOVA compares the outcome across groups using a linear model framework and is mathematically equivalent to linear regression. The interpretation of the coefficient (\(b\)), confidence interval (\(95\% CI\)), and p‑value (\(p\)) therefore follows the same principles.

For comparisons between two groups (e.g., Group 1 vs Group 2), the coefficient (\(b\)) represents the difference in the mean outcome between groups. Specifically, it estimates how much higher or lower the outcome is, on average, in one group compared with the comparison group. The confidence interval (\(95\% CI\)) provides a range of plausible values for this group difference. If the interval includes 0, the data are compatible with no difference between groups. Narrower intervals indicate more precise estimates of the group difference. The p-value (\(p\)) for the coefficient reflects the strength of evidence for a difference between the two groups. A small p-value (typically \(p<0.05\)) suggests that the observed difference is unlikely due to chance alone. As with regression, it does not reflect the magnitude or clinical importance of the difference.

The overall ANOVA result is summarised by the F statistic (\(F\)) and its associated p-value (\(p\)). The F test evaluates whether there is any evidence of a difference in mean outcomes across all groups. Again, a small p-value (typically \(p<0.05\)) suggests that at least one group differs from the others, although it does not specify which groups differ.

As differences in patient numbers and case-mix may influence outcomes across units, in each model, all models were adjusted for unit-level characteristics including total patients on 31st March, unit’s mean age, mean diagnosis duration, proportion of White patients, and proportion living in most deprived areas. The adjustments helps ensure that any observed associations are less likely to be driven by differences in patient populations between units.

The eta-squared (\({η}^{2}\)) indicates the proportion of variation in the outcome explained by the full model. For example, \({η}^{2}=0.10\) means that 10% of variability in the outcome is accounted for by all predictors in the model. On the other hand, partial eta-squared (\(partial\ {η}^{2}\)) indicates the proportion of variation in the outcome explained by the predictor of interest, after accounting for other predictors in the model. In this report, \(partial\ {η}^{2}=0.05\) means that the predictor of interest explains 5% of the variation in the outcome, after adjusting for unit size and their case-mix. Higher values indicate better model fit, but it does not necessarily imply clinical importance.

Interactive raincloud plots were created using the npdar R package (Fang, 2026).

Also see Note 2 for details of a raincloud plot.

Figure 13. Median HbA1c and availability of a quality assured structured education programme.

1.7 Psychology

1.7.1 Psychological Screening

91.9% of PDUs (n = 148) routinely formally screened for emotional wellbeing of either the child and/or family. Of these, the most common method of screening was “Validated questionnaire8” (Figure 14).

8 A questionnaire that has been scientifically tested and shown to produce consistent (reliable) and accurate (valid) results about the condition it is designed to measure (Heale & Twycross, 2015).

Figure 14. Percentage of PDUs screening for emotional wellbeing by each method (multiple-choice).

24.2% of PDUs (n = 39) routinely formally screened for eating disorders (ED). Of these, the most common method of screening was Validated questionnaire (Figure 15).

Figure 15. Percentage of PDUs screening for EDs by each method (multiple-choice).

1.7.2 Access to Psychological Care

72.7% of PDUs (n = 117) routinely offered contact with a psychologist following diagnosis of diabetes, with 43.5% (n = 70) offering contact within 4 weeks of diagnosis (Figure 16). Figure 17 shows the percentage of PDUs offering each psychology intervention.

Figure 16. Time to contact with a psychologist following diagnosis.
Figure 17. Percentage of PDUs offering each psychology intervention.

PDUs that have a local high HbA1c protocol which includes assessment and intervention offered by a clinical psychologist have the same median HbA1c as those who do not (Figure 18).

See Note 2 for details of a raincloud plot.

Figure 18. Median HbA1c and high HbA1c protocols including assessment and intervention by clinical psychologist.

1.8 Funding

1.8.1 Best Practice Tariff (England Only)

88.7% of PDUs in England (n = 133) knew whether they were achieving Best Practice Tariff (BPT)9 payments or not. Of those, 81.2% (n = 108) stated that they were achieving BPT payments, compared to 94.2% in 2017. Of those receiving payments, 50.0% received BPT payments based on current patient numbers.

9 A payment mechanism that is designed to incentivise quality and cost effective care. In paediatric diabetes, this is an annual payment, per patient that receives care meeting the best practice criteria.

Of the PDUs in England not receiving BPT payments or unsure, 18.7% expected to start within 12 months.

On average, 27.2% of BPT payments went directly into diabetes care in PDUs (range 0.0% to 100.0%).

1.8.2 All funding

15.5% of PDUs (n = 25) received at least £3,453 per patient receiving a full year of care in 2024/25, including as part of BPT. PDUs that received at last £3,453 per patient had a lower average HbA1c (Figure 19) and lower health check completion rate (Figure 20). However, the differences between PDUs receiving and not receiving funding were not statistically significant, and there was no evidence of an overall group difference.

See Note 2 and Note 3 for details of a raincloud plot and ANOVA.

Figure 19. Funding at or exceeding £3,453 per patient receiving a full year of care and median HbA1c.
Figure 20. Funding at or exceeding £3,453 per patient receiving a full year of care and health check completion rate.

There is no evidence that PDUs that received at least £3,453 per patient have better access to hybrid closed loop (HCL) systems (Figure 21).

Figure 21. Funding at or exceeding £3,453 per patient receiving a full year of care and HCL uptake.

1.9 Workforce

1.9.1 Core Diabetes MDT

The core diabetes multidisciplinary team (MDT)10 includes: consultants and other doctors, paediatric diabetes specialist nurses (PDSN), dietitians, diabetes educators, clinical psychologists, and administrative staff. The total core diabetes MDT WTE11 was 3.25 per 100 patients (Figure 22), a decrease from 3.40 in 2017/18 (Figure 23).

10 A group of professionals from one or more clinical disciplines who together make decisions regarding recommended treatment of individual patients. An MDT caring for children and young people with diabetes will typically include consultant paediatricians, paediatric diabetes specialist nurses, dietitians, diabetes educators, and psychologists.

11 A unit that indicates the workload of an employed person in a way that makes workloads comparable across various contexts. WTE is typically based on hours worked as a proportion of the contracted hours normally worked by a full time employee in the post. For example, a person working standard hours each day, but only 3 days out of 5, would could as 0.6 WTE.

Definitions of MDT, PDSN, and WTE are also provided in Glossary.

Figure 22. WTE of core diabetes MDT per 100 children and young people with diabetes by country and network.
Figure 23. Overall-level WTE of core diabetes MDT per 100 children and young people with diabetes in 2024/25 vs 2017/18.

Total core MDT WTE per 100 was significantly associated with median HbA1c at PDU-level (Figure 24). When looking at each staff group within the MDT, median HbA1c was negatively correlated with WTE of PDSNs (Figure 25).

See Note 1 for details of linear regression.

Figure 24. Median HbA1c and MDT WTE per 100 children and young people with diabetes at PDU-level.
Figure 25. Median HbA1c and each MDT WTE per 100 children and young people with diabetes at PDU-level.

Total core MDT WTE per 100 patients was not significantly associated with health check completion rate at PDU-level (Figure 26). However, when looking at each staff group within the MDT, health check completion rate was positively correlated with WTE of consultants and doctors (Figure 27).

See Note 1 for details of linear regression.

Figure 26. Health check completion rate and MDT WTE per 100 children and young people with diabetes at PDU-level.
Figure 27. Health check completion rate and each MDT WTE per 100 children and young people with diabetes at PDU-level.

Total core MDT WTE per 100 was significantly associated with HCL uptake, even after the removal of outliers (Figure 28). When looking at each staff group within the MDT, HCL uptake was positively correlated with WTE of consultants and doctors, as well as WTE of PDSNs (Figure 29).

See Note 1 for details of linear regression.

Figure 28. HCL uptake and MDT WTE per 100 children and young people with diabetes at PDU-level.
Figure 29. HCL uptake and each MDT WTE per 100 children and young people with diabetes at PDU-level.

WTE of core diabetes MDT was not correlated with WNB rate (Figure 30).

See Note 1 for details of linear regression.

Figure 30. WNB rate and MDT WTE per 100 children and young people with diabetes at PDU-level.

1.9.1.1 Minimum staffing standards

The minimum staffing standards are outlined in the National CYP Diabetes Network Workforce Standards (NCYPDN, 2024, p. 17).

15.5% of PDUs (n = 25) had core diabetes staffing at or above the recommended level. Table 3 compares the mean WTE per 100 patients to the recommended minimum for the core diabetes MDT, split by clinic size of < or ≥ 150.

PDU with <150 patients
PDU with ≥150 patients
WTE type Minimum standarda Standard met Minimum standarda Standard met
Consultants and other doctors 0.50 17.0% (9/53) 0.50 10.2% (11/108)
PDSNs 1.70 58.5% (31/53) 1.60 56.5% (61/108)
Dietitians 0.80 9.4% (5/53) 0.75 13.9% (15/108)
Psychologists 0.50 15.1% (8/53) 0.40 9.3% (10/108)
Administrative/Data staff 0.80 17.0% (9/53) 0.70 10.2% (11/108)
Total core MDT (excluding diabetes educators) 4.30 17.0% (9/53) 3.95 14.8% (16/108)
a Minimum staffing standards for Core Diabetes MDT (National CYP Diabetes Network, 2024, Table 3)
Table 3. Core diabetes MDT staffing compared to recommended minimum.

1.9.1.2 Paediatric Consultants

99.4% of PDUs (n = 160) had at least one consultant in post, with 87.6% (n = 141) being led by two or more consultants. The mean WTE per 100 patients was 1.48 (range 0.00 to 10.34).

Paediatric consultants typically worked more programmed activities (PAs)12 than allocated within job plans. On average, PDUs had 2.92 consultant PAs per 100 patients (range 0.00 to 9.94) dedicated to the PDU. In comparison, the actual mean PAs worked per week was 3.21 per 100 patients (range 0.00 to 9.94).

12 A 4-hour unit of time (one half day), 10 of which comprise a consultant’s work week. One WTE is equivalent to 10 PAs on average.

Including consultant and non-consultant doctors (see section below), 12.4% of PDUs (n = 20) met the NCYPDN minimum of 0.50 WTE (5 PAs) per 100 patients (2024).

The mean number of years working as a consultant was 11.8 years, and the mean number of years looking after children with diabetes was 13.3 years.

89.1% (393/441) of consultants had specific training in paediatric diabetes before taking on care of children with diabetes. Figure 31 shows the training undertaken by those who had specific training in paediatric diabetes.

Figure 31. Paediatric diabetes training undertaken by paediatric consultants.

Grandparent exemption is defined in Glossary.

1.9.1.3 Non-Consultant Doctors

18.8% of PDUs (n = 30) had a non-consultant doctor, such as a non-consultant staff grade, SAS, or Trust doctor. The mean WTE per 100 patients was 0.07 (range 0.00 to 10.34).

44.2% (19/43) of non-consultant doctors had specific training in paediatric diabetes before taking on care of children with diabetes. Figure 32 shows the training undertaken.

Figure 32. Paediatric diabetes training undertaken by non-consultant doctors.

14.4% of PDUs (n = 23) had a doctor in training contracted to work for their PDU. The mean WTE per 100 was 0.06 (range 0.00 to 1.48).

1.9.1.4 Paediatric Diabetes Specialist Nurses (PDSNs)

100.0% of PDUs (n = 161) had a PDSN in post, with 85.7% (n = 138) having a minimum of 2 WTE. The mean WTE per 100 was 1.76 (range 0.93 to 6.21). 57.1% of PDUs (n = 92) met the NCYPDN recommended minimum PDSN staffing - 1.70 per 100 for PDUs with <150 patients, 1.60 for PDUs with ≥150 patients (2024).

98.1% (656/669) of PDSNs worked in both hospital and community settings, 90.4% (605/669) adjusted insulin doses under protocol, 26.8% (179/669) were nurse prescribers, and 91.0% (609/669) were able to start patients on hybrid closed loop (HCL) systems.

95.0% of PDUs (n = 153) had at least one PDSN who was able to adjust insulin (mean = 3.8). 58.4% of PDUs (n = 94) had at least one PDSN who was a nurse prescriber (mean = 1.1). 98.8% of PDUs (n = 159) had at least one PDSN able to start patients on HCLs (mean = 3.8). 98.8% of PDUs (n = 159) had at least one PDSN who was able to work in both hospital and community settings (mean = 4.1).

The majority of PDSNs were on Agenda for Change (AfC)13 band 6 (Figure 33). 14.3% of PDUs (n = 23) had a PDSN at Band 8 or above.

13 The pay system used for most staff in the NHS, excluding doctors and senior managers.

Figure 33. WTE of PDSN per 100 children and young people with diabetes by country and network.

1.9.1.5 Dietetics

97.5% of PDUs (n = 157) had a dietitian in post. The mean WTE per 100 was 0.55 (range 0.00 to 2.08). 12.4% of PDUs (n = 20) met the NCYPDN recommended minimum dietitian staffing - 0.80 per 100 for PDUs with <150 patients, 0.75 for PDUs with ≥150 patients (2024).

89.6% (258/288) of dietitians worked in both hospital and community settings, 8.0% (23/288) were supplementary prescribers providing adjusting insulin via a clinical management plan and with supervision of an independent prescriber, and 44.4% (128/288) adjusted insulin doses under a local diabetes protocol.

91.9% of PDUs (n = 148) employed at least one dietitian that worked in both hospital and community settings, 11.8% (n = 19) had at least one dietitian that was a supplementary prescriber, and 49.1% (n = 79) had at least one dietitian able to adjust insulin doses under protocol.

The majority of dietitians were on AfC14 band 7 (Figure 34). 82.0% of PDUs (n = 132) had a dietitian at Band 7 or above.

14 The pay system used for most staff in the NHS, excluding doctors and senior managers.

Figure 34. WTE of dietitian per 100 children and young people with diabetes by country and network.

8.1% of PDUs (n = 13) had a dietetic assistant, with a mean WTE per 100 patients of 0.01 (range 0.00 to 0.33).

1.9.1.6 Diabetes Educators

3.1% of PDUs (n = 5) had a diabetes educator in post. The mean WTE per 100 patients was 0.01 (range 0.00 to 0.54).

1.9.1.7 Psychology

79.5% (n = 128) of PDUs had a clinical psychologist in post. The mean WTE per 100 patients was 0.23 (range 0.00 to 0.89). 11.2% of PDUs (n = 18) met the NCYPDN recommended minimum staffing - 0.50 per 100 for PDUs with <150 patients, 0.40 for PDUs with ≥150 patients (2024).

The majority of clinical psychologists were on AfC15 band 8 (Figure 35). 64.0% of PDUs (n = 103) had a clinical psychologist at Band 8 or above.

15 The pay system used for most staff in the NHS, excluding doctors and senior managers.

Figure 35. WTE of clinical psychologist per 100 children and young people with diabetes by country and network.

11.9% of PDUs (n = 19) had an assistant psychologist in post, with a mean WTE per 100 patients of 0.03. 6.3% of PDUs (n = 10) had an another psychological practitioner in post, with a mean WTE per 100 patients of 0.01.

1.9.1.8 Administrative Staff

95.7% of PDUs (n = 154) had an administrative staff member (including data support staff) in post, with a mean WTE per 100 patients of 0.44 (range 0.00, 1.41). 12.4% of PDUs (n = 20) met the NCYPDN recommended minimum staffing - 0.80 per 100 for PDUs with <150 patients, 0.70 for PDUs with 150+ patients (2024).

Excluding data support staff, the majority of administrative staff were on AfC16 band 4 (Figure 36). 52.8% of PDUs (n = 85) had an administrative staff at Band 4 or above.

16 The pay system used for most staff in the NHS, excluding doctors and senior managers.

Figure 36. WTE of administrative staff (excluding data support staff) per 100 children and young people with diabetes by country and network.

1.9.2 Additional Staffing

Paediatric diabetes additional staffing includes: social workers, support workers (play specialists, youth or family support workers) and healthcare assistants. The total additional staffing WTE was 0.15 per 100 (range 0.00 to 2.53) (Figure 37).

Figure 37. WTE of additional staffing per 100 children and young people with diabetes by country and network.

Figure 38 compares additional staffing between 2017/18 and 2024/25. The 2017/18 Workforce Spotlight Audit did not include health care assistants or technology assistants.

Although there was an increase in children and young people in the child protection system in 2024/25 compared to 2017/18 (Figure 4), social workers WTE saw a decrease in 2024/25 compared to 2017/18 (Figure 38), which may reflect the wider workforce challenges in social care.

Figure 38. Overall-level WTE of additional staffing per 100 children and young people with diabetes in 2024/25 vs 2017/18.

Total additional staffing WTE per 100 was not associated with median HbA1c at PDU-level, with most PDUs having low additional staffing WTE per 100 patients (Figure 39). When looking at each staff group, there was no clear relationship with median HbA1c (Figure 40).

See Note 1 for details of linear regression.

Figure 39. Median HbA1c and additional staffing WTE per 100 children and young people with diabetes at PDU-level.
Figure 40. Median HbA1c and each additional WTE per 100 children and young people with diabetes at PDU-level.

Total additional staffing WTE per 100 was also not associated with health check completion rate at PDU-level (Figure 41). When looking at each staff group, there was no clear relationship with health check completion rate (Figure 42).

See Note 1 for details of linear regression.

Figure 41. Health check completion rate and additional WTE per 100 children and young people with diabetes at PDU-level.
Figure 42. Health check completion rate and each additional WTE per 100 children and young people with diabetes at PDU-level.

Total additional staffing WTE per 100 was not associated with HCL uptake (Figure 43). When looking at each staff group, there was no clear relationship with health check completion rate (Figure 44).

See Note 1 for details of linear regression.

Figure 43. HCL uptake and additional WTE per 100 children and young people with diabetes at PDU-level.
Figure 44. Health check completion rate and each additional WTE per 100 children and young people with diabetes at PDU-level.

1.9.2.1 Minimum staffing standards

The minimum staffing standards are outlined in the NCYPDN Workforce Standards (2024, p. 18).

Table 4 compares the WTE per 100 patients to the recommended minimum for the additional staffing.

PDU with any number of patients
WTE type Minimum standarda Standard met
Social workers 0.10 0.6% (1/161)
Support workers 0.60 3.7% (6/161)
Health Care Assistants 0.20 12.4% (20/161)
a Minimum standards for additional staffing (National CYP Diabetes Network, 2024, Table 4)
Table 4. Additional diabetes staffing compared to recommended minimum.

1.9.2.2 Health Care Assistants and Technology Assistants

19.3% of PDUs (n = 31) had a health care assistant or technology assistant in post, with a mean WTE per 100 patients of 0.06 (range 0.00 to 0.69).

For health care assistants, 12.4% of PDUs (n = 20) met the NCYPDN recommended minimum of 0.2 WTE per 100 patients (2024).

1.9.2.3 Support Workers

In this report, the term Support Workers includes Youth Workers, Family Support Workers, and Play Specialists.

22.4% of PDUs (n = 36) had a support worker in post, with a mean WTE per 100 patients of 0.09 (range 0.00 to 2.53). 3.7% of PDUs (n = 6) met the NCYPDN recommended minimum of 0.6 WTE per 100 patients (2024).

1.9.2.4 Social Workers

Only 2 PDUs had a social worker in post, with only 1 meeting the NCYPDN recommended minimum of 0.1 WTE per 100 patients (2024).

1.10 Vacancies

54.0% of PDUs (n = 87) reported at least one staffing vacancy.

Across England, Wales, and Jersey, there were 18.9 consultant vacancies, totalling 50.79 PAs, giving a PA vacancy rate of 2.69 per position. 73.5% of consultant vacancies had been vacant for more than 3 months.

There were 109.0 vacancies for other roles, totalling 69.93 WTEs. This is an average of 0.43 WTE per PDU (or 0.43 WTE per PDU who answered). 95.9% of these vacancies had been vacant for more than 3 months.

Figure 45. WTE of staffing vacancy per 100 children and young people with diabetes by country and network.

Vacancies were not significantly associated with median HbA1c (Figure 46, Figure 47).

See Note 1 for details of linear regression.

Figure 46. Relationship between consultant and doctor vacancies in WTE and HbA1c outcomes at PDU-level.
Figure 47. Relationship between other vacancies in WTE and HbA1c outcomes at PDU-level.

2 Glossary

Agenda for Change (AfC)

The pay system used for most staff in the NHS, excluding doctors and senior managers.

Best practice tariff (BPT; England only)

A payment mechanism that is designed to incentivise quality and cost effective care. In paediatric diabetes, this is an annual payment, per patient that receives care meeting the best practice criteria.

CASCADE

The Child and Adolescent Structured Competencies Approach to Diabetes Education. This is a structured education programme for young people with Type 1 diabetes and their families in the UK (NIHR, 2014).

Child in care (formally referred to as a ‘looked after child’)

A child who has been “looked after” by the local authority (England and Wales) or the minister (Jersey) for a continuous period of more than 24 hours.

Child in need

A child who is thought to need extra help from children’s services to achieve or maintain a reasonable standard of health or development.

Child protection plan/register

In England, a child protection plan is a formal written agreement created when a child is assessed to be at risk of significant harm from abuse. In Wales and Jersey, the child protection register is a confidential list of all children who have been identified as being at risk of significant harm.

Children and young people on 31st March

Total numbers of children and young people with diabetes still receiving care from a PDU on 31st March 2025, and remained in one PDU (i.e., did not appear in two or more PDUs) and did not die during the core audit year 2024/25. This number is used as a proxy for PDU size in many analyses, including but not limited to the core staffing WTE per 100 patients (Figure 22) and additional staffing WTE per 100 patients (Figure 37). In some analyses, however, this number is restricted to fit the question being asked. For example, for the cohort rate per 10,000 patients in the child protection system amongst all PDUs with a known response (Table 1) and cohort rate of patients with a mental health condition amongst all PDUs with a known response (Table 2), total patients are restricted to PDUs who has given a valid response to the question.

DAFNE

Dose Adjustment for Normal Eating. This is a structured education programme for people over 17 years old with type 1 diabetes (DAFNE, 2025).

DEApp

The Diabetes Education App. This is a structured education programme for children and young people with type 1 diabetes delivered via an app.

Grandparent exemption

The recognition of experience and prior practice for established consultants, allowing them to transition into specialised roles without meeting all current training requirements.

Multidisciplinary team (MDT)

A group of professionals from one or more clinical disciplines who together make decisions regarding recommended treatment of individual patients. An MDT caring for children and young people with diabetes will typically include consultant paediatricians, paediatric diabetes specialist nurses, dietitians, diabetes educators, and psychologists.

Paediatric diabetes specialist nurse (PDSN)

A specialist nurse who provides support and education to children and young people with diabetes and their families.

Paediatric diabetes unit (PDU)

A clinical unit providing diabetes care for children and young people with diabetes. In this report, a PDU can refer to one or more units who submitted jointly.

Paediatric intensive care unit (PICU)

A specialist unit providing intensive care for children and young people who are experiencing severe medical conditions. Examples of diabetes emergencies include DKA.

Programmed activity (PA)

A 4-hour unit of time (one half day), 10 of which comprise a consultant’s work week. One WTE is equivalent to 10 PAs on average.

Tertiary-level support

Supports for children outside of their normal secondary level catchment area.

This could include PICU services for diabetes emergencies (e.g., DKA), diabetes provision for children requiring other services such as surgery (day case or inpatient), and specialist outpatient management (e.g., high-level exercise advice, eating disorders management).

SEREN

Structured Education Reassuring Empowering Nurturing. A structured education programme for children and young people with Type 1 diabetes in Wales (CYPWDN, 2026).

Validated questionnaire

A questionnaire that has been scientifically tested and shown to produce consistent (reliable) and accurate (valid) results about the condition it is designed to measure (Heale & Twycross, 2015).

Was not brought (WNB)

Children and young people with diabetes who did not attend and had not cancelled a planned appointment.

Whole time equivalent (WTE)

A unit that indicates the workload of an employed person in a way that makes workloads comparable across various contexts. WTE is typically based on hours worked as a proportion of the contracted hours normally worked by a full time employee in the post. For example, a person working standard hours each day, but only 3 days out of 5, would could as 0.6 WTE.

3 Acknowledgements

NPDA Clinical Lead

  • Professor Justin Warner, Clinical Lead, RCPCH

NPDA Project Team

  • Calvin Down, Head of Audits, RCPCH
  • Zhaonan Fang, NPDA Analyst, RCPCH
  • Tom Kellier, Audits Administrator, RCPCH
  • Amani Krayem, NPDA Manager, RCPCH
  • Humfrey Legge, Audits Senior Analyst, RCPCH
  • Saira Pons-Perez, Audits Senior Analyst, RCPCH
  • Richard Shepherd, NPDA Coordinator, RCPCH

NPDA Project Board

  • Carly Blagojevic, Paediatric Diabetes Specialist Nurse, Newcastle Hospitals NHS Foundation Trust
  • Dr Fiona Campbell, Consultant in Paediatric Endocrinology and Diabetes, Leeds Teaching Hospitals NHS Trust
  • Grace Cuff, Project Manager, NCAPOP, HQIP
  • Kate Cullen, Parent Representative
  • Dr Mark Deakin, Consultant in Diabetes and General Paediatrics, Alder Hey Children’s NHS Foundation Trust
  • Professor Paul Dimitri, Vice President for Science and Research, RCPCH
  • Dr Simon Fountain-Polley, Consultant Paediatrician, Hywel Dda University Health Board
  • Rachel Harris, Paediatric Diabetes Specialist Nurse, Cardiff and Vale University Health Board
  • Emily Hillyer, Paediatric Diabetes Specialist Nurse, University Hospitals Sussex NHS Foundation Trust
  • Dr Neil Hopper, Consultant Paediatrician, South Tyneside and Sunderland NHS Foundation Trust
  • Dr Nway Htun Kyi, Paediatric Trainee, Cardiff and Vale University Health Board
  • Simon Lewthwaite, Parent Representative
  • Dr Fulya Mehta, Children and Young People Diabetes Lead, NHS England
  • Caroline Schmutz, Research Partnerships Manager, Breakthrough T1D
  • Sarah Berry, Research Partnerships Manager, Breakthrough T1D
  • James Smith, Senior Case Manager, NHS England
  • Anthony Walker, Policy Lead, Diabetes UK
  • Rachael Ward-Sample, Associate Director for Quality and Improvement, NCAPOP, HQIP
  • Dr Diana Yardley, Advanced Nurse Practitioner, Oxford University Hospitals NHS Foundation Trust

NPDA Methodology and Dataset Group

  • Francesca Annan, Clinical Specialist Paediatric/Adolescent Diabetes Dietitian, University College London Hospital NHS Foundation Trust
  • Dr Nivedita Aswani, Consultant in General Paediatrics and Diabetes, Sheffield Children’s NHS Foundation Trust
  • Carly Blagojevic, Paediatric Diabetes Specialist Nurse, Newcastle Hospitals NHS Foundation Trust
  • Dr Fiona Campbell, Consultant in Paediatric Endocrinology and Diabetes, Leeds Teaching Hospitals NHS Trust
  • Dr Mark Deakin, Consultant in Diabetes and General Paediatrics, Alder Hey Children’s NHS Foundation Trust
  • Dr Halina Flannery, Clinical Psychologist, University College London Hospital NHS Foundation Trust
  • Dr Simon Fountain-Polley, Consultant Paediatrician, Hywel Dda University Health Board
  • Rachel Harris, Paediatric Diabetes Specialist Nurse, Cardiff and Vale University Health Board
  • Emily Hillyer, Paediatric Diabetes Specialist Nurse, University Hospitals Sussex NHS Foundation Trust
  • Dr Nway Htun Kyi, Paediatric Trainee, Cardiff and Vale University Health Board
  • Dr Evelien Gevers, Consultant in Paediatric Endocrinology and Diabetes, Barts Health NHS Trust
  • Dr Neil Hopper, Consultant Paediatrician, South Tyneside and Sunderland NHS Foundation Trust
  • Dr Diana Yardley, Advanced Nurse Practitioner, Oxford University Hospitals NHS Foundation Trust

HQIP Support

  • Grace Cuff, Project Manager, NCAPOP, HQIP
  • Rachael Sample, Associate Director for Quality and Improvement, NCAPOP, HQIP

Special thanks to the National CYP Diabetes Network for their input into the dataset, in particular the Social Care and T1D working group and the Psychology working group.

4 Bibliography

Allen, M., Poggiali, D., Whitaker, K., Marshall, T., Langen, J. van, & Kievit, R. (2021). Raincloud plots: A multi-platform tool for robust data visualization [version 2; peer review: 2 approved]. Wellcome Open Research, 4(63). https://doi.org/10.12688/wellcomeopenres.15191.2
CYPWDN. (2026). SEREN – structured education reassuring empowering nurturing. https://serendiabetes.com/
DAFNE. (2025). Dose adjustment for normal eating. https://dafne.nhs.uk/
Department for Education. (2025a). Children in Need, Reporting year 2025. https://explore-education-statistics.service.gov.uk/find-statistics/children-in-need/2025
Department for Education. (2025b). Children looked after in England including adoptions, Reporting year 2025. https://explore-education-statistics.service.gov.uk/find-statistics/children-looked-after-in-england-including-adoptions/2025
Fang, Z. (2026). npdar: National paediatric diabetes audit r tools. https://rcpch.github.io/npdar
Heale, R., & Twycross, A. (2015). Validity and reliability in quantitative studies. Evidence-Based Nursing, 18(3), 66–67. https://doi.org/10.1136/eb-2015-102129
National Children and Young People’s Diabetes Network. (2024). Workforce Standards for Children and Young People’s Diabetes Services (pp. 17–18). (Table 3 and Table 4). https://www.cypdiabetesnetwork.nhs.uk/national-network/wp-content/uploads/sites/14/2024/03/Workforce-Standards-for-Children-Young-Peoples-Diabetes-Services.pdf
NIHR. (2014). Structured, intensive education maximising engagement, motivation and long-term change for children and young people with diabetes: A cluster randomised controlled trial with integral process and economic evaluation the CASCADE study. https://www.journalslibrary.nihr.ac.uk/hta/HTA18200
StatsWales. (2023a). Children receiving care and support on the child protection register by category of abuse and gender. (Table ID: nmux6qfpux07; accessed 2026-05-13). https://stats.gov.wales/en-GB/593e0ef5-fa92-4f86-8d3c-dd9cac077d10/filtered/nmux6qfpux07
StatsWales. (2023b). Children receiving care and support per 10,000 population aged under 18. (Table ID: kcsacpfav98k; accessed 2026-05-13). https://stats.gov.wales/en-GB/6789f87f-ee0e-4e75-a487-30a1e1573725/filtered/kcsacpfav98k
StatsWales. (2023c). Population estimates by local authority, year, sex and age. (Table ID: xuxrm0fdsn6g; accessed 2026-05-13). https://stats.gov.wales/en-GB/82d9faea-b515-41cd-aedb-8f4594f66ce5/filtered/xuxrm0fdsn6g
StatsWales. (2024). Children looked after on 31 march per 10,000 population aged under 18. (Table ID: 2rr6gbd4gryd; accessed 2026-05-13). https://stats.gov.wales/en-GB/4eb2c4bd-9dbc-4293-ae71-60031d2cbc28/filtered/2rr6gbd4gryd
Welsh Government. (2024). Children receiving care and support census: On 31 march 2023. (Official statistics in development). https://www.gov.wales/children-receiving-care-and-support-census-31-march-2023-official-statistics-development-html