Health systems analytics in the Thames Valley & Wessex neonatal network: Simulating neonatal patient pathways
Neonatal care for preterm and sick babies is organised into local areas around the country, called Operational Delivery Networks (ODNs). By working as a network, the aim is to ensure that all babies have the care they need in the most appropriate place, as close to home as possible.
Approximately 60,000 babies are born within the network each year, with approximately 1 in every 10 of these babies requiring some form of neonatal care. Of these babies, approximately 12% of them will require a transfer to another unit.
There are 3 types of neonatal units, and there are set guidelines stating the sort of care that each type of unit should be able to provide (often based around the gestational age/weight of the baby).
With the ever changing world of the NHS, the network needs to be prepared on the possible impact on transfers and patient care, if there was to be any changes to the type of care that a unit can provide.
The network also needs to ensure that each unit is current running at optimal cot levels, and the best quality care is being received by all babies.
CLAHRC Wessex Data Science Hub worked with the Thames Valley & Wessex Neonatal Operational Delivery Network to develop a simulation models of the flow of patients through the network, including the transfers of babies between units. The models mimic the flow of patients through the network, including transfers of more complex cases to suitable units, transfers back to local units of babies as their condition improves, transfers due to capacity constraints and transfers into and out of the network. The model accurately captures the patterns of arrivals and importantly accounts for the unpredictable variability seen in when patients arrives and how the duration of neonatal treatment that they require.
The model is controlled from an Excel interface, in which changes can be made to:
- Arrival rates
- Numbers of beds per unit (including by type eg ICU beds)
- The complexity of cases a unit can take
- The preferred transfer routes
- Provide details of potential new unit(s)
This allows the network managers to explore the impact of potential changes theoretically.
As an example of the types of scenarios that can be tested using the model these results consider re-designating one of the units to only be able to take the least serious cases. This is expected to have an impact on the capacity of the unit being re-designated, the other unit’s capacity requirements and the numbers of transfers required. The table below gives a sample of the results for this scenario. Such results demonstrate the potential impact of such a change and if the change were to take place allow for appropriate planning to adjust the numbers of cots available in advance of the change.