Fewer adverse events as a result of the SAFE or SORRY? programme in hospitals and nursing homes. part i: primary outcome of a cluster randomised trial

Background: Patient care guidelines are usually implemented one at a time, yet patients are at risk for multiple, often preventable, adverse events simultaneously.

Objective: This study aimed to test the effect of the SAFE or SORRY? programme on the incidence of three adverse events (pressure ulcers, urinary tract infections and falls). This paper describes Part I of the study: the effect on the incidence of adverse events.

Design: A cluster randomised trial was conducted between September 2006 and November 2008. After a three-month baseline period the intervention was implemented followed by a nine-month follow-up period.

Settings: Ten wards from four hospitals and ten wards from six nursing homes were stratified for institute and ward type and then randomised to intervention or usual care group.

Participants: During baseline and follow-up, patients (≥18 years) with an expected length of stay of at least five days, were asked to participate.

Methods: The SAFE or SORRY? programme consisted of the essential recommendations of guidelines for the three adverse events. A multifaceted implementation strategy was used for the implementation: education, patient involvement and feedback on process and outcome indicators. The usual care group continued care as usual. Data were collected on the incidence of adverse events and a Poisson regression model was used to estimate the rate ratio of the adverse events between the intervention and the usual care group at follow-up.

Results: At follow-up, 2201 hospital patients with 3358 patient weeks and 392 nursing home patients with 5799 patient weeks were observed. Poisson regression analyses showed a rate ratio for the development of an adverse event in favour of the intervention group of 0.57 (95% CI: 0.34-0.95) and 0.67 (95% CI: 0.48-0.99) for hospital patients and nursing home patients respectively.

Conclusion: This study showed that implementing multiple guidelines simultaneously is possible, which is promising. Patients in the intervention groups developed 43% and 33% fewer adverse events compared to the usual care groups in hospitals and nursing homes respectively. Even so, more research is necessary to underline these results.

Trial registration: clinicaltrials.gov, number NCT00365430.