PT  - JOURNAL ARTICLE
AU  - Booth, John
AU  - Bryant, William
AU  - Pope, Rebecca
AU  - Sebire, Neil J
TI  - 64 The use of graph analytics to enhance the understanding of hospital admissions for Paediatric renal transplants
AID  - 10.1136/bmjpo-2023-GOSH.18
DP  - 2023 Dec 01
TA  - BMJ Paediatrics Open
PG  - A10--A10
VI  - 7
IP  - Suppl 2
4099  - http://bmjpaedsopen.bmj.com/content/7/Suppl_2/A10.1.short
4100  - http://bmjpaedsopen.bmj.com/content/7/Suppl_2/A10.1.full
SO  - BMJ Paediatrics Open2023 Dec 01; 7
AB  - Aim The interactions between patients and Health Care Providers (HCPs) within a hospital admission are complex and difficult to navigate using traditional analytical approaches when applied to Electronic Patient Record (EPR) data. The collaboration between HCPs, is not clear and could serve as an important signal to assess quality of care and patient safety. The objective of this paper is to demonstrate whether graph analytics can be applied to EPR data to distinguish different HCPs collaboration and interaction network structures with individual patients undergoing the same procedure.Method We apply graph analysis to the HCPs interactions with an individual patient over the period of a complete hospital admission during which each patient underwent a renal transplant. Collaboration between HCPs was defined by the interaction of HCPs with the same patients within a 60-minute window. Separate sessions were defined by whether a patient was in the Intensive Care Unit (ICU) or in a standard hospital ward. A report is generated that explorers the daily interaction and collaboration network structures within an admission.Results HCPs and patient interactions for 120 hospital admissions where the principal procedure was Allotransplantation of kidney in the period May 2019 to June 2023 were extracted. 2,300 individual graphs were constructed to analyse each patient’s HCP’s interactions within a 60-minute window per day of admission. Varying types of admission were identified, by sex and age group of the patient, length of stay and with or without an ICU visit.Conclusions Graph analysis provided a novel way to augment the numeric data to visualise differing collaboration and interaction network structures which can be used to gain a deeper understanding of the complex nature of Paediatric patient care.