The TEAM instrument for measuring emergency team performance: validation of the Swedish version at two emergency departments

Abstract

Abstract Background The Team Emergency Assessment Measure (TEAM) questionnaire is designed for rating the non-technical performance of emergency medical teams during emergencies, e.g., resuscitation or trauma management. Originally developed in Australia it has today been translated and validated into eleven languages, but a Swedish version is lacking. The aim was therefore to cross-culturally translate and evaluate the reliability and validity of the TEAM questionnaire in a Swedish health care setting. Methods The instrument was forward and backward translated and adapted into a Swedish context according to established guidelines for cross-cultural adaptation of survey-based measures. The translated version was tested through 78 pairwise assessments of 39 high-priority codes at the emergency departments of two major hospitals. The raters observed the teams at work in real time and filled in the questionnaires immediately afterwards independently of each other. Psychometric properties of the instrument were evaluated. Results The original instrument was translated by pairs of translators independently of each other and reviewed by an expert committee of researchers, nurses and physicians from different specialties, a linguist and one of the original developers of the tool. A few adaptations were needed for the Swedish context. A principal component factor analysis confirmed a single ‘teamwork’ construct in line with the original instrument. The Swedish version showed excellent reliability with a Cronbach’s alpha of 0.955 and a mean inter-item correlation of 0.691. The mean item-scale correlation of 0.82 indicated high internal consistency reliability. Inter-rater reliability was measured by intraclass correlation and was 0.74 for the global score indicating good reliability. Individual items ranged between 0.52 and 0.88. No floor effects but ceiling effects were noted. Finally, teams displaying clear closed-loop communication had higher TEAM scores than teams with less clear communication. Conclusions Real time observations of authentic, high priority cases at two emergency departments show that the Swedish version of the TEAM instrument has good psychometric properties for evaluating team performance. The TEAM instrument is thus a welcome tool for assessing non-technical skills of emergency medical teams.