Internship project: Defining indicators of unplanned readmission to hospital

Opportunity

During the project the scholar will develop a set of indicators to identify if a patient currently admitted to hospital is likely to be re-admitted upon discharge. To develop these indicators the candidate will analyse anonymised historical in-patient admission and clinical data to identify the factors (from patient demographics such as age, gender, etc. to reason for admission, problems and allergies as well as clinical observation data including fluid balance, procedures, vital signs, unplanned activities and complications) that may indicate a patient is likely to be re-admitted to hospital. This data will be analysed to create a 10 point risk score that can be used to determine if a patient currently undergoing care can be discharged with minimal likelihood of an unplanned re-admission.

Project Objective

• To characterise the factors in anonymised in-patient admissions data that indicate a patient has undergone an unplanned re-admission to hospital.
• To categorise these factors into clusters that are indicative of patient risk-level of readmission and create a 10-point patient risk score.
• To evaluate the accuracy of the 10-point patient risk score on historical patient data on unplanned readmission to hospital.