Teaching medical students and doctors how to communicate risk

The need for doctors to have proficient communication skills is well recognised,¹ but teaching students how to communicate risk to patients seems to have received little attention in the undergraduate medical curriculum. Primarily it is statisticians who teach the concept of probability. This ultimately translates into the communication of risk that informs the clinical consultation. Although students need to appreciate the basics of statistical methods and know the different ways to convey risk,² it is particularly important that they have the opportunity to practise these skills under safe conditions and receive constructive feedback.¹,³

The recently established graduate entry programme at St George's Hospital Medical School in London has an integrated curriculum across all years.⁴ The course is delivered by using problem based learning whereby students use “triggers” from a problem case or scenario to identify their own learning issues.⁵ It has been suggested that students are more motivated by such an approach, where the practical problems they experience act as a stimulus for learning.⁶ Despite the best efforts of teachers, however, students tend to see medical statistics as inherently mathematical and irrelevant.⁷ Within the graduate entry programme integrating the teaching of medical statistics with communication skills has enabled students to see the relevance of medical statistics, in particular the need to communicate risk effectively to patients.

Various techniques can be used to raise students' awareness about the problems that may occur when communicating risk, not only for the patient but also the doctor. Qualitative expressions of probability, such as “unlikely,” “a chance,” “occasionally,” and “probably” are used all the time in clinical medicine to describe risk. A useful approach is to ask students individually to rate such expressions as a numerical proportion,⁸ placed in relation to an event occurring, such as a child being born with Down's syndrome. The results of this exercise enable students to compare their own interpretations with those of their peers. Typically, the interpretation varies widely,⁹ and some students find it impossible to ascribe a single value to a qualitative expression of probability. Ideally this awareness raising exercise encourages students to think carefully about how their patients might interpret such descriptive statements of probability, and clarify if necessary.

Role playing consultations between patient and doctor plus small group discussions play a valuable part in the students' general education.¹⁰ These teaching methods develop the skills of listening, presenting ideas, decision making, and working as part of a team. They help the student appreciate that one of the biggest problems that clinicians face is how to interpret results from epidemiological studies or clinical trials in ways that have meaning to the individual patient sitting in front of them. The patient will either develop a disease or not, and it is not obvious how the patient's view of their outcome is altered by any risk the clinician ascribes. After feedback the key message that students take away is to tailor information to what the patient wishes to know and to verify his or her understanding.

Video recordings of patients (or actors) in a clinical consultation are a valuable way to raise awareness.³ Students can observe patients' concerns and suggest where the clinician could have enhanced his or her communication. We have used a video of women describing their experiences when undergoing screening for Down's syndrome and open neural tube defects in pregnancy. After watching the video the students can explore different ways of discussing risk with patients and presenting test results in ways that are helpful. One of the issues discussed was the difficulty for patients in appreciating the magnitude of a risk in relation to everyday events (see table on bmj.com). Furthermore, the way in which information was presented may influence subsequent decisions.¹¹ If the probability of having a child with Down's syndrome was framed negatively—as a 20% risk of an affected child—women were more likely to have an amniocentesis than if the risk was framed positively—an 80% risk of no abnormality.

We believe that integrating medical statistics with communication skills in this way helps students appreciate the relevance of probability by learning it in context while also developing skills in communicating risk. Edwards et al have shown with general practice registrars that these approaches and benefits may also be relevant to postgraduate learning and skill acquisition.¹²