Patient Related Outcome Measures (Feb 2020)
Can Patient Reported Outcomes (PROs) from Population Surveys Provide Accurate Estimates of Pre-Admission Health Status of Emergency Hospital Admissions?
Abstract
Esther Kwong,1 Gary Abel,2 Nick Black1 1Department of Health Services Research & Policy, London School of Hygiene and Tropical Medicine, London WC1H 9SH, UK; 2College of Medicine and Health, University of Exeter, Exeter EX4 4PY, UKCorrespondence: Esther KwongDepartment of Health Services Research & Policy, London School of Hygiene and Tropical Medicine 15-17 Tavistock Place, London WC1H 9SH, UKTel +44 207 958 8285Email [email protected]: The use of PROs for assessing the outcomes of emergency hospital admissions requires a means of estimating patients’ pre-admission health status. A possible alternative to asking patients to recall how their health was before the incident causing admission is to use estimates derived from matched samples from population surveys. Our aims were to explore the impact of different methods of matching and to compare the results with estimates based on retrospective reporting.Methods: First, elective hip arthroplasty patients were matched to respondents to the General Practice Patient Survey using age, sex, socio-economic status and number of comorbidities. The impact of restricting matching for locality and specific co-morbidities was explored. Second, the best matching method was applied to emergency admissions for laparotomy and for percutaneous coronary intervention (PCI) after acute myocardial infarction. Data were stratified by patient characteristics. Differences in mean EQ-5D scores between the patients and matched population respondents were tested using t tests.Results: Modifying the most basic form of matching by also taking locality and the specific comorbid conditions into account made no significant difference to the mean EQ-5D score for hip arthroplasty patients. Even using the most detailed matching possible, patients’ mean EQ-5D score was significantly different to that of the general population for all three cohorts. The difference was greatest for elective hip arthroplasty (0.22 v 0.64), less so for emergency laparotomy (0.56 v 0.72) and least for PCI (0.79 v 0.71). This reflects hip arthroplasty patients having a long-standing condition characterised by pain and limited mobility, whereas the other two cohorts may have enjoyed reasonable health until an unexpected acute episode led to their emergency admission.Conclusion: Routine PRO data acquired from population surveys cannot be used as an accurate alternative to retrospectively reported PROMs by patients during their emergency admission episode.Keywords: patient-reported outcome measures, health status, health-related quality of life, self- report, retrospective, hospital admission, general population patient survey
