Variability between psychiatrists on domains of the mental status examination.

Background: The Mental Status Examination (MSE) is a core element of the psychiatric assessment. To investigate the current level of psychopathological knowledge, the reliability of the MSE can be used as a proxy as it is based on descriptive psychopathology. Methods: Three psychiatrists wrote their MSE based on 27 video recordings. The variability and inter-rater agreement were evaluated using an agreement scale from 1 to 5, made by several psychiatrists. The agreement was analysed by mean values, stratified mean values and Cronbach's alpha. Results: The total agreement had a Cronbach's alpha of 0.87 (p < 0.000). The grand mean of variability was 4.1 (SD = 0.8). The domains with the highest variability were Attitude (Est. = 3.5, SD = 0.9), Affect (Est. = 3.8, SD = 0.8) and Motor activity (3.7, SD = 0.9). The videos with the highest variability were #2 (Est.=3.3, SD = 1,1) and #21 (Est. = 3.6, SD = 1.1). Conclusions: The overall reliability of the MSE based on Cronbach's alpha was good and the mean variability was low. This indicates that the MSE performed by three psychiatrists is reliable. The main reason for variability was due to individual interpretation and discrepancies in literature. The literature-driven variability could possibly be reduced by approaching national tradition to international practise. The individual variability maybe reduced by increasing the opportunities for clinical group rating.