The Trauma Audit and Research Network

Data Monitoring

TARN monitors data submitted in two different ways; data completeness and accreditation. Accreditation looks at how frequently key fields used in analysis are recorded. Data completeness looks at the number of submissions made to TARN compared to the number that appear to meet the TARN inclusion criteria in the HES dataset.

In an effort to make the data completeness figures more accurate and also help hospitals improve their data completeness TARN:

  • Provide 3 months of HES data to hospitals as a representative sample that can be used to identify cases that should have been submitted and were not and also identify any cases that are not eligible and tell TARN about them. These cases will then be removed from the denominator. The data for this is available in a report in the Audit section of the TARN website (in the HES / PEDW feedback data category).
  • Display case ascertainment as a rangeThis is shown as a range from the percentage calculated using the original HES denominator to the new value derived following feedback. In the event that no feedback is received the lower bound will be calculated based on the overall reduction seen in the previous years comparison exercise.

A summary of the results of the 2013 comparison exercise is available for download, which showed that 17.1% of the cases identified in the HES dataset as being TARN eligible are in fact not. Some of the most common reasons for this are; incorrect ICD coding, incorrectly recorded LOS and patients with non-traumatic injuries. It also identified some patient groups that are commonly missed when identifying patients to submit to TARN; non-isolated hip fractures in the over 65s, traumatic SDH admitted to medical wards and patients with spinal fractures.

Similar analysis of 2014 feedback showed similar results and that 14.5% of cases were misidentified as TARN eligible.

A brief summary of using the HES dataset to improve data completeness is available.

If you have any questions please contact support.


Data collection

2013 comparison exercise
Improving data completeness