A correlation of clinical performance on written test and standardised patient.

The objective of the study was to determine the relationship between written responses and actual performance on a patient with a sore throat. The written test was in a format of modified essay question. The standardised patient was a medical student who had chronic recurrent pharyngotonsillitis. The study subjects were residents in the Department of Medicine who were not aware of the study objective. The 37 residents were asked to complete the written test and 18 of them were randomly selected for actual performance assessment. The standardised patient saw the selected residents during their usual service hours. The information obtained from the standardised patient, medical records and prescription slips were analysed. The results revealed: 1) Thirty per cent of the information actually performed were not recorded. 2) Thirty-four per cent of the actions described in the written test were not actually performed, most of these actions were unnecessary and superfluous. 3) There was a good correlation between written responses and actual performance on a standardised patient.