PP16 Creating a linked dataset to explore patient outcomes after leaving ambulance care

Background

Most English ambulance services have implemented electronic patient report forms (ePRF), enabling electronic capture of patient information; previously all information was collected using paper forms. This development brings new opportunities, including linking ambulance data to patient information post ambulance discharge and improving ambulance quality measures. This research aims to create a linked ambulance service, A and E, inpatient and mortality dataset.

Methods

EMAS supplied CAD operational and ePRF clinical data from the study period to the University of Sheffield, and patient identifiers (name, date of birth, address, NHS number) to NHS Digital. NHS Digital traced missing NHS numbers, retrieved patient-level HES A and E, HES inpatient, and ONS mortality data, and returned this data to the University of Sheffield. We created a linked ambulance service and subsequent care dataset by linking EMAS clinical information to data from NHS Digital. This allows us to identify subsequent recontacts, A and E attendances, admissions and deaths.

Results

Our sample comprised only calls recorded via ePRF plus Hear and Treat – excluding 38.90% of total calls. 80.06% of patients in our sample were successfully traced by NHS Digital; unsuccessful traces were due to missing or incomplete CAD/ePRF data. Only CAD identifiers were available for hear and treat patients, limiting tracing success. Non-traced patients could not be linked to HES/ONS data. Of the successfully traced patients, 94.61% were linked to at least one record in the HES or mortality data.

Conclusions

We successfully created a linked dataset for 154927 EMAS patients. This can be used to identify patient outcomes after leaving ambulance care and to develop and test new ways of monitoring ambulance service care quality and performance. Missing information was problematic for data linkage and may be more common among patients who were unable to provide personal details on scene e.g. sicker patients or those who died.

http://ift.tt/2wmlJaq