Simulating Medical Talk

Postgraduate medical assessments now involve an important and well-established focus on effective interaction with patients, and this is often formally assessed as part of various clinical examinations. Through simulated consultations with role-played patients, doctors can be observed in action, enabling examiners to assess not only clinical knowledge, but crucially how that knowledge is deployed and communicated with a patient in a spoken context. However, the pass rates in many of these exams, particularly higher failure rates for those who have done their initial medical training overseas, have led to concerns that linguistic and cultural factors may play a role in doctors' poorer performance in consultations.

Building on a Knowledge Transfer Partnerships study, conducted at King's College London from 2011-2013 under Professor Celia Roberts (PI) and Professor Kamila Hawthorne, this ESRC 'Future Research Leaders' grant enables me to further research these simulated consultations, with a view to better understanding what is happening at an interactional level. A close linguistic analysis is made of video data from simulated patient surgeries, including an analysis of non-verbal features of communication, in addressing where communicative differences or difficulties might be occurring. This has implications for further understanding contemporary clinical examinations and the training and preparation postgraduate doctors can undertake.

I am currently looking at the use of simulation in two key settings:

• General Practice
• Emergency Medicine

This research looks at performance in clinical skills assessment from a linguistic and cultural perspective, with a view to understanding why there are such differential pass rates and giving suggestions on how this issue can be tackled.  The book, available for download here, is both a research report and a guide to the sociolinguistic methodology used. While the findings are based on a research project in partnership with the Royal College of General Practitioners, they are applicable to many other medical settings where standardised examinations of simulated consultations are used. More widely, this research addresses a central paradox in institutional life – how to balance validity in assessments and be fair to a diverse group of candidates in an increasingly diverse society, while maintaining reliability with standardised and universal marking criteria.

It has been widely acknowledged that candidates from overseas fair less well in such examinations. A close look at the interactions which make up these simulated consultations shows that there are complex and subtle differences between passing and failing candidates which cannot be explained simply as ‘language’ and ‘cultural’ differences and put in a box separate from issues of fairness. These structured examinations, unintentionally, contribute to the weight of the assessment on overseas candidates, particularly in how interpersonal effectiveness is judged both explicitly and implicitly.

The research has identified a range of successful candidate strategies which form the basis of a set of e-learning materials, created at the University of Notitngham to be published  by the RCGP. It also suggests that aspects of the exam, notably the more subjective features of interpersonal skills, are not best assessed in highly structured exams.  This area needs to be better defined, using a new analytic language, to debate how and where it could be most effectively and fairly assessed. 

This is an issue for medical teams with very real consequences.  For example the Elaine Bromiley case (Bromiley, 2008) centred around the failure of care due to communication difficulties between ad hoc team members. Two nurses, who knew a tracheotomy needed to be performed, were unwilling to speak up because of the perceived seniority of a surgeon in the team. Such events illustrate that technical skills alone are not sufficient in ensuring patient safety. Building on work by Bezemer et al (2011) on operating theatre teams, a key new research question that can be answered by linguistics is: How are power relationships achieved interactionally in emergency scenarios within ad hoc teams? 

Much of the literature on medical role-play makes the case for their being as authentic as possible. However, evidence from linguistics suggests simulated interactions necessarily demonstrate differences to ‘real-life’, particularly in the audience-design of utterances for an overhearing 3rd party (Seale et al, 2007; Roberts, Atkins and Hawthorne, forthcoming 2014). What evidence is there of audience design in the QMC simulations? Since the team simulations are unlikely to be ‘authentic’, what is it participants gain from the interactions that is helpful?

Data-collection for this part of the study has recently commenced at Nottingham. A range of simulations are being filmed, particularly those focusing on trauma scenarios and team communication skills.