Critical care delivery models in emergency departments: a systematic review of the literature and meta-analysis of related outcome effects.

backgroundCritical care delivery in the emergency department (ED) may improve selected patient- and system-relevant outcomes. However, it remains unclear which organisational models to deliver critical care in the ED (CC-ED) have been implemented and what their outcome effects are.

methodsWe conducted a systematic review and meta-analysis including studies describing organisational approaches to deliver CC-ED. Delivery models were categorised and described qualitatively. For studies reporting comparable outcomes, random-effects meta-analysis was performed to estimate pooled effects and 95% confidence intervals. Outcomes included mortality, intensive care unit (ICU) admission rates, ICU length of stay (LOS), hospital LOS, and cost of care.

resultsOf 4,967 records, 67 were included into the qualitative assessment and 11 studies into the quantitative analysis, respectively. We identified five models to deliver CC-ED. These consisted of dedicated critical care areas in the ED (n = 49/67, 73.1%), placement of critical care staff in the ED (n = 5/67, 7.5%), deployment of critical care teams to the ED (n = 5/67, 7.5%), telemedical support of ED staff by critical care teams (n = 5/67, 7.5%), and implementation of protocols to accelerate ICU admission (n = 3/67, 4.5%). Mortality was not different versus usual care in any of the CC-ED models. CC-ED models had variable effects on the ICU admission rate, hospital and ICU LOS. Dedicated critical care areas in the ED were associated with lower ICU admission rates [OR 0.83 (0.76-0.91), p < 0.001], and shorter hospital LOS [mean difference - 0.31 (-0.59 to -0.03) days, p = 0.03], but longer ICU LOS [mean difference 0.45 (0.31-0.6) days, p < 0.001]. Reported cost of care did not differ between CC-ED and usual care.

conclusionsWe identified five different organisational model categories to deliver CC-ED. Dedicated critical care areas such as ED-ICUs was the model category most frequently published. Our quantitative meta-analysis suggests that compared with other CC-ED delivery models, dedicated critical care areas in the ED may reduce both ICU admission rates and hospital LOS of critically ill ED patients.