Abstract
The ENEIS study aimed to estimate the frequency of adverse events (AE) in public and private hospital, both for events leading to hospitalisation and for those identified during hospitalisation, to assess their preventability rate and to describe their main actives errors and systemic failures.
This was an incidence study on hospitalised patients observed over a maximum of 7 days. The sample was randomised using a three-stage cluster stratified process. In each selected ward, a nurse detected adverse events using 17 criteria, and a physician, in collaboration with the practitioner in charge of the patients and using the medical record, assessed the causality and preventability of AE and their active failures. A root cause analysis was subsequently conducted in a subsample of these AE.
8,754 patients, from 292 wards in 71 hospitals, were observed over a total of 35,234 days of hospitalisation, 17,104 in medical wards and 18,129 in surgical wards. Among the 450 AE (247 in surgery and 203 in medicine), 40% were considered preventable. In the course of the 7 day follow-up per unit, at least one AE was observed in 66% of surgical units and in 58% of medical units. They occurred above all in vulnerable or frail patients, 4 years older than the mean age of the sample.
Adverse events were the cause of 3.9% (CI 95% [3.4-5.6]) of overall hospitalisations. Two thirds resulted from medical care provided in the community (GPs), and the others were readmissions following previous hospitalisation. Among these AE, 47% were considered as preventable. Medication was implicated in nearly 40%.
The incidence density of AE in the course of hospitalisation was 6.6 to 1000 days of hospitalisation (CI 95% [5.7-7.5]). Among these, 35% were considered preventable. Invasive procedure, in particular surgery interventions, were the source of half of AE occurring during hospitalisation.
The active errors related to preventive and diagnostic care were more frequently preventable than to therapeutic care. Systemic failures were present in 37 out of the 45 AE analysed (82.2%) . Among a total of retrieved 145 failures, the team failures were most frequently and closely related to the occurrence of AE.
Each year at national level 125,000 to 200,000 hospitalisations and 120,000 to 190,000 AE during hospitalisation can be considered as preventable. The perioperative period and geriatric patient management should deserve closer attention. Above all, a risk-management culture based on sensitisation, training, error and event analysis are needed urgently and on a wide scale.
