Intensive Care Unit clinical outcomes

Our Paediatric (PICU), Neonatal (NICU) and Cardiac (CICU) Intensive Care Units provide medical treatment to children who are critically ill. Our specialist multidisciplinary service is one of the largest for children in the United Kingdom and Europe.

A key outcome measure used in Intensive Care Units (ICU) is the survival rate for patients, measured at the time when they are discharged. Raw mortality rates may be challenging to interpret as patients that are admitted in a sicker condition are at greater risk, and therefore the outcomes need to be ‘adjusted’ to consider the level of severity of the patients who are admitted to the intensive care unit.

The PIM3 (Paediatric Index of Mortality version 3) score is a nationally accepted risk score. This is calculated for every child admitted to ICU, to quantify the risk of death on admission. The PIM3 score is based on medical history, interventions and physiological measurements taken from time of first contact with an ICU doctor up to the first hour after admission. The standardised mortality ratio (SMR) is the ratio of the number of actual deaths compared to the number of expected deaths based on the PIM3 score. The SMR helps us to evaluate if the survival rate is what would be expected according to how sick children are when they are admitted to intensive care.

The table and the funnel plots shown were provided by the Paediatric Intensive Care Audit Network body (PICANet) for admissions to the GOSH ICUs between January 2021 to December 2023.

The funnel plot displays the SMR for each intensive care unit in the UK as a dot. If the actual number of deaths was the same as the predicted number of deaths, then the SMR would be 1, corresponding to the horizontal line in the middle of the funnel plots. The two dotted lines above and below the horizontal lines are called ‘99.9% control limits’. If an intensive care unit’s dot fell outside either of these lines, this would mean that the mortality rate would be very different to what is expected based on the case mix (upper line = much higher mortality, and lower line = much lower mortality than expected). Centre E1 is the combined GOSH Paediatric and Neonatal ICUs (PICU/NICU) and Centre E2 is GOSH Cardiac ICU (CICU). The dots for both E1 and E2 sit within the 99.9% control limits, so based on this threshold the SMR for both of these units is neither much higher nor much lower than expected.

Table 1 shows the SMR values for PICU/NICU (E1) and CICU (E2) with the ‘95% control limits’ for the three years between the start of 2021 and the end of 2023.

There were 3,268 admissions to PICU/NICU. The SMR was 1.06 and the 95% control limits were 0.91 and 1.22. So based on the 95% control limit threshold, the mortality rate was as expected based on how sick the children were when they were admitted.

There were 2,038 admissions to CICU. The SMR was 0.70 and the 95% control limits were 0.52 and 0.91. So based on the 95% control limit threshold, the mortality rate for E2 was somewhat lower than expected based on how sick the children were when they were admitted.

Table 1 Standardised Mortality Ratio (SMR) for PICU/NICU (E1) and CICU (E2) for Jan 2021 to Dec 2023, PIM3 Risk Adjusted (PIM3 Recalibrated 2024)

Figure 1 Standardised Mortality Ratio (SMR) funnel plot for PICU/NICU (E1) and CICU (E2), Jan 2021 to Dec 2023, PIM3 Risk Adjusted (PIM3 Recalibrated 2024)

Emergency readmissions to ICU within 48 hours of discharge is a quality outcome indicator that is monitored to help us evaluate how well the process of discharging children from intensive care is working. For example, if children were discharged from intensive care too early, then the percentage of readmissions would rise.

Table 2 shows the percentage of emergency readmission to PICU/NICU within 48 hours of discharge. Table 3 shows the percentage of emergency readmissions within 48 hours for CICU. These data are displayed by year alongside the national average for that year across all paediatric ICUs in the UK and Republic of Ireland.

Table 2 Percentage of PICU/NICU emergency readmissions within 48 hours of discharge, 2021 to 2023

The overall emergency re-admission rate is small. As with any rare event, variability can be high, which is reflected in our single centre data. The rate of emergency readmissions in 2023 was higher than previous years. We seek to understand each case at our weekly mortality and morbidity meeting. While we have not identified any single theme, we continue to monitor this closely.

Table 3 Percentage of CICU emergency readmissions within 48 hours of discharge, 2021 to 2023

The percentage of children who were readmitted to CICU (E2) within 48 hours was similar to or lower than the national average in each of the three years. This suggests that the discharge processes are working well.

Intubation is the placement of a flexible plastic tube through the mouth to maintain an open airway and provide ventilator-assisted breathing during critical illness. Extubation is the removal of that tube, which should happen in a planned way. Extubation can happen accidentally, which teams seek to avoid. The rate of unplanned extubations is therefore an important measure of care quality in the ICU environment. Unplanned extubation can happen when a child wakes up suddenly and all intensive care units expect to see this happen occasionally. However, if unplanned extubation is happening more often than expected this might mean that the care needs to be reviewed.

Table 4 shows the unplanned extubation rates (calculated per 1000 intubated days) for PICU/NICU (E1) alongside the average rate for all other intensive care units combined.

The unplanned extubation rate is higher than the national average on GOSH PICU/NICU. We monitor this closely and have identified it is mostly due to a higher rate of unplanned extubations in NICU. NICU at GOSH is the only neonatal unit that contributes data to PICANet, so the patient population may not be similar to the wider national PICU population. Nonetheless, we continue to undertake quality improvement work and education on the NICU to reduce this rate over time. Following improvements in tube fixation, sedation and in the techniques used to move babies for procedures and cares, there has been a 50 per cent reduction in the accidental extubation rate on NICU in 2024.

Table 4 PICU/NICU (E1) unplanned extubation rates per 1000 intubated days for all admissions, 2021 to 2023

Table 5 shows the unplanned extubation rates (calculated per 1000 intubated days) for CICU (E2) alongside the average rate for all other intensive care units combined.

The rate of unplanned extubations in CICU (E2) for each of the reported three years shown was lower than the national average. The rate of unplanned extubation in 2023 rose slightly to 3.3 per 1000 intubated days. While this is still lower than the national average (4.4 per 1000 intubated days), it is monitored in CICU. Unplanned extubations can be a sign of the intensive care unit being very busy and this is certainly an issue that CICU is watching closely.

Table 5 CICU (E2) unplanned extubation rates per 1000 intubated days for all admissions, 2021 to 2023

All data is published with the permission of PICANet from Paediatric Intensive Care Audit Network Annual Report 2024 (published December 2024): Universities of Leeds and Leicester.

The full report can be found here at Paediatric Intensive Care Audit Network (PICANet).

Please also see our Cardiothoracic outcomes page for benchmarked data from the National Congenital Heart Audit.

In 2016, a website called Understanding Children’s Heart Surgery Outcomes was launched to help parents and families to make sense of published survival data about children’s heart surgery in the UK and Ireland.

The ICUs are fully engaged in the Trust-wide quality improvement initiatives to improve the care of deteriorating patients on our inpatient wards. The Trust is also working towards development of a high dependency service outside the ICUs. Together, these initiatives may reduce emergency readmissions to the ICUs.