Can Significant Event Audit identify key factors that might reduce emergency admissions?
Problem
Emergency hospital admissions are stressful for patients, expensive, and it is thought that a large number could be avoided. There is limited evidence for effective interventions to reduce emergency admissions, which have increased by 2% a year over the past decade. Significant Event Audit (SEA) involves a systematic attempt by the practice team to investigate, review and learn from a single event. SEA is used widely in general practice, but not routinely for analysing emergency admissions. We performed a study in 20 general practices to test whether SEA is acceptable for analysing emergency admissions for avoidable factors, to improve this approach if indicated and to gather data needed to run a larger study.
Approach
We provided training for each practice to carry out SEAs in a standard way. Three local hospitals randomly identified patients with an emergency admission. Practices approached those patients for consent for the study, and looked at (1) details of what happened before the admission (2) the people and systems that were involved (3) what might have been done differently to avoid the admission and (4) developed action points to promote change in practice that might help reduce future unplanned admissions. We ran three focus groups of practice staff and patients who had an admission to determine the acceptability of SEA to staff and patients.
Findings
19 of 20 practices completed the study, and completed 132 SEAs (26% of selected cases). On average 4 doctors, one nurse and one manager attended SEA meetings. In 13% of admissions that practices discussed there was something that could have been done differently, but it was thought that only half (6.5%) of these admissions could be avoided. Patient factors were present in 94%, Practitioner factors in 59%, and Systems factors in 53% of these admissions. Themes from the focus groups included: the value of SEA in principle; the terms ‘significant events’ and ‘avoidable admissions’ being challenged as attributing blame ahead of critical enquiry; the consent process excluding important patient groups; having randomly selected cases working less well, with reduced ownership and engagement.
Consequences
SEA alone is not a suitable intervention to reduce admissions in a future RCT as the effect size is likely to be small; despite this SEA is a suitable tool for analysing emergency admissions for avoidable factors. The SEA template is useful, but should use different terms with less negative connotations, and allow practices more choice over the cases for selection to increase engagement and learning. Patient consent requirements should allow for third party consent, and for some clinician review even without consent as important cases were excluded. Recommendations include the renaming of SEA to include a term with less negative connotations.