Adaptive process triage (ADAPT) is a triage system developed in Sweden in 2006. It is currently used by four University hospitals in the region of Stockholm and several other hospitals in Sweden. It is introduced in several hospitals in Denmark. ADAPT is based on a trace and trigger tool for vital signs according to the ABCD-principle and a short systematic questionnaire for each chief complaint. The main objective is detection of the seriously ill, guidance of healthcare personnel for patient streaming and to be a communication tool. The triage system is based on lean-principles and modern ideas for patient streaming to very urgent care, urgent care, admitting of the elderly and to 'see and treat'. Following ADAPT each patient are triaged as either red (life-threatening), orange (seriously ill), yellow (ill), green (need of assessment) or blue (fast-track).

The aim of this study is to assess the validity of ADAPT as a triage system.

All patients visiting the ED at Södersjukhuset in July - December 2008 were eligible for inclusion. Only the first visit of each patient in that time period was recorded. All adult patients visiting (excluding gynaecology and obstetrics) were included. Need for admittance to a ward or ICU was used as outcome parameters. Pearson Chi-square test and Fischers' exact test was used for statistical assessment of the distribution in the outcome parameters.

N = 35054. Out of these 4.2% were triaged as red, 15.6% orange, 36.6% yellow, 28.7% and 14.8% blue.

The red group was admitted to a general ward in 77.5%, orange in 55.6%, yellow in 36.3%, green in 14.9% and blue in 2.0%. There was statistical significance (p = .000).

The red group was admitted to ICU in 24.7% of the cases, orange in 4.6%, yellow in 1.1% yellow, green in 0.2% and blue in 0.0%. There was statistical significance (p = .000).

Our study show that ADAPT triage levels are good predictors of the need for admittance to a general ward as well as a good predictor of the need for intensive care.