Communication skills knowledge, understanding and OSCE performance in medical trainees: a multivariate prospective study using structural equation modelling.

Abstract

To test the stability of medical student communication skills over a period of 17 months as exhibited by performance in objective structured clinical examinations (OSCEs) and to determine the strength of prediction of these skills by initial levels of knowledge and understanding. This is a prospective study using a 2-wave cohort. Medical undergraduates (n = 383) from 2 years intake (1996 and 1997) were followed through the first 3 years of a medical curriculum. The study procedure involved the objective structured video examination (OSVE) conducted at formative and summative examinations during the first year. Two OSCE measures were employed: expert examiners and simulated patients completed the Liverpool Communication Skills Assessment Scale (LCSAS) and the Global Simulated Patient Rating Scale (GSPRS), respectively. The OSCE data were collected at Level 1 and 17 months later at Level 2 examinations. The measurement model followed prediction. A causal model using latent variables was fitted with Level 2 OSCE performance regressed on Level 1 OSCE and OSVE marks. Expert and simulated patient OSCE data were fitted separately and combined to determine strength of model fit according to professional and patient opinion of student skills. The overall fit of the models was acceptable. Communication skills performance showed a high level of stability. Some negative effect of cognitive factors on future skills performance was found. Early development of communication skills shows stable performance following an introductory course. Knowledge of communication skills has a small but significant influence on performance, depending on the time of testing. New assessments of cognitive factors are required to include both tacit and explicit knowledge.