Abstract
Hospital readmissions have a significant impact on the healthcare system and it has been estimated that the readmission rate in frail elderly can range from 20 to 40%, with at least 50% of these readmissions being preventable. There is also considerable expenditure on potentially avoidable readmissions and with these statistics, there is a significant focus on reducing readmission rates.
The overall purpose of this study was to develop a tool that could be used to identify patients at high risk for hospital readmission and then utilise this tool to undertake a Randomised Controlled Trial (RCT). The RCT involved a more robust Clinical Pharmacist intervention on discharge from hospital to determine if this additional pharmacist support could reduce readmission rates and increase quality of life. This study consisted of two phases.
Phase one was the development of an electronic tool that would identify patients at high risk of readmission to hospital by using data already held within electronic patient health record. This development occurred in multiple stages and was constructed in conjunction with Senior Medical Officers from several areas of the hospital including from the Internal Medicine, Care of the Elderly and the Emergency Departments. The development included identifying, prioritising and weighting the various risk factors for readmission.
The tool was then tested and validated to ensure that it was drawing the appropriate information and identifying appropriate patients. The validation process showed the tool to have a high predictive rate for identifying patients at high risk for readmission and allowed for the identification of patients to recruit into phase two of the project (the RCT).
Phase two, the RCT, was initiated in May 2017 using the above tool. The trial ran for 12 months with the overall aim to investigate whether a more robust Clinical Pharmacist intervention on discharge had any impact on readmission rates, Emergency Department (ED) presentation rates and the patient’s overall Quality of Life. The intent of the RCT was to measure any improvement on the effectiveness of the transition of care from hospital care to primary care and determine if there was a positive impact on outcomes.
The efficiency of the tool in identifying patients at high risk for readmission had the unexpected effect of identifying a considerable number of individuals who were unable to participate due to a level of cognitive impairment (which was one of the exclusion criteria of the RCT). This resulted in 43% of potential participants being ineligible to participate. A total population of identified patients was 392, and of this, 90 patients were recruited and randomised into the study.
The average age of the identified participants was 80.6 years, indicating that many of the high risk and high use patients in this ward are elderly with an increased rate of co-morbidities and medication use.
Comparing readmission rates between the control and the intervention arm of the study indicated that the intervention group were readmitted to the hospital slightly more often than the control group. However, numbers in each arm were too small to have any statistical significance and no statistical evaluation was undertaken on the data.
Comparing ED presentation rates between the control and the intervention arm of the study indicated that the intervention group may have had higher ED presentation rates at 28 days than the control group. However, numbers in each are were too small to have any statistical significant and no statistical evaluation was undertaken on the data.
Comparing quality of life scores between the control and intervention group, demonstrated similar quality of life scores on admission and post discharge, however, again, numbers in each arm were too small to have any statistical significance and no statistical evaluation was undertaken on the data.
While the study was not able to demonstrate a statistical significance in reduction of readmission rates, many lessons and areas for opportunity to improve transition of care and the discharge service were identified.