The Use of Patient Reported Outcome Measures (PROMs) for Evaluating Emergency Admissions

The use of Patient Reported Outcome Measures (PROMs) is one way to measure the effectiveness of health services. PROMs use in emergency admissions to hospitals is limited by the methodological obstacle of having no pre-admission measure of health status. The aim of this thesis was to study the use of retrospective PROMs and of routine measures of population health status to identify a reliable method that would allow the extension of PROMs collection into this important area of health care. A literature review found there was strong agreement (ICC>0.75) between contemporaneous and retrospective PROMs in elective conditions and that population data could be used to estimate baseline health status in some conditions. This was confirmed in the prospective cohort studies conducted in elective patients (ICC>0.82 for disease-specific and ICC>0.62 for generic PROMs). However, matching methods used to explore population values as an alternative to retrospective baseline health status did not provide estimates similar to those obtained from elective patients. These exploratory matching methods did, however, provide insights for further research in emergency admissions. The studies in emergency laparotomy (EL) and STEMI heart attack established the feasibility of collecting retrospective PROMs in these emergency settings: 85% of admissions were eligible, and invitation and participation rates were 85% & 72% for EL and 79% & 91% for STEMI. The response rates to three month follow-up questionnaires were good (EL: 74%, STEMI: 66%) enabling the mean recovery of patients in terms of PROMs to be conducted and the effect of any response biases to be determined. The use of retrospective PROMs can provide a reliable method to collect pre-admission health status in some emergency admissions. It is necessary to establish the generalisability of these findings, investigate possible clinical confounders and explore the extent of unwarranted variation between providers in outcomes.