|| Checking for direct PDF access through Ovid
This trial evaluates implementation of critical care outreach in a middle-income country.Critical care outreach delivered by a team of intensive care nurses was implemented across general hospital wards in an Iranian university hospital. The order of implementation was randomized with wards stratified by predicted mortality rates. Effectiveness was evaluated using a stepped wedge cluster randomized controlled trial design, comparing outcomes between patients admitted before and after implementation. The primary outcomes were inhospital mortality and cardiopulmonary resuscitation. A nested qualitative study explored challenges to implementation and contextualized the trial outcomes.Between July 2010 and December 2011, 13 wards were sequentially randomized to implement the critical care outreach: 7802 patients were admitted before implementation and 10 880 after implementation. There were 370 deaths (4.74%) among patients admitted before implementation and 384 deaths (3.53%) after implementation. Adjusting for clustering and temporal trends, the odds ratio for mortality was 1.03 (95% confidence interval, 0.68-1.53). Results for other outcomes were broadly similar. Focus groups revealed a lack of endorsement of the intervention by management and ward nurses.This pragmatic evaluation of critical care outreach in a middle-income country did not show a reduction in mortality or other outcomes.