Trust peers

The below states the aims, methodology and outcomes behind how trust peers are created. 


Aim:

Identify which NHS Trusts/Organisations are ‘alike’ based on various characteristics


Method:

2-stage approach was used:

1. First a supervised model (a sophisticated algorithm) was built to capture the relationship between all the available variables and each Trust’s productivity, as measured by their overall cost per WAU. This provided a set of coefficients (multipliers) that show the ‘importance’ of each variable. For a more accurate set of coefficients a decision tree approach called Gradient boosting was used to take into account relationships between variables. Gradient boosting gives a much better approximation of cost per WAU, and it is easy to extract the importance of each metric from the resulting rule tree.

2. We then used the resulting multipliers in a weighted Euclidian distance estimation, inputting data for every variable for every Trust, to see which Trusts are closest to one another in terms of the factors that predict productivity. Every Trust was ranked with regard to its suitability as a peer for every other Trust. We chose the top 10 ranked peers as a suggested set of peers.

 

Variables included in model Development:

 • Demographic data such as age ranges, deprivation indices and urbanity
• Economic data such as operating expenditure and market forces factors (MFF) payment index
• Geographic data such as eastings, northings, number of attendances and number of sites
• Performance data such as delayed transfer of care (DTOC) rate and staff absence rate
• Logistic data such as percentage of emergency admissions and student staff rates
• Clinical data such as case mix at HRG level, degree of specialism and a disease prevalence metric
•Estates data such as PFI and Major Trauma sites

 

Data sources:

 • Deprivation - NHS Digital, HES
• Patients - NHS Digital, HES
• Attendance - NHS Digital, HES
• FTE Staff - NHS Digital, Workforce Statistics
• Diagnosis - NHS Digital, HES
• Night beds - NHS England, Unify data collection - KH03
• Day beds - NHS England, Unify data collection - KH03
• Age 0-5 - NHS Digital, HES
• Age 60-74 - NHS Digital, HES
• Number of procedures - NHS Digital, HES
• Operating expenses - Monitor, Transparency data
• Urban - NHS Digital, HES
• Sites - Organisation Data Service, ODS
• Sea distance - ODS to obtain Trusts postcode and doogal.co.uk to calculate distance
• MFF payment index - NHS England Publications
• StudentStaff - DH
• CCGs - NHS Digital, HES
• Non-type 1 AE department - NHS Digital, HES
• Absence Rate - NHS Digital, Workforce Statistics
• DTOCs - NHS England, Unify2 Data Collection - MSitDT
• Degree of Specialisation - HES databases

 

Primary Outcome:

Lists of 10 peers obtained for each NHS Trust.

 

Secondary output:

Output imported onto Model Health System dashboard to support benchmarking and opportunity estimations.

 

Feedback and Knowledge Base