9 . Nonverbal Communication in Doctor - Suicidal Patient

Current techniques of repeated suicide risk assessment are not reliably predictive, and research into new methods is needed. The judgment of clinicians relies partly on nonverbal signs such as facial expressions. If differences in patients and/or interviewer's facial expressions appeared between subjects who were to make subsequent attempts (Repeaters) and those who were not (NonRepeaters), this could lay the foundations for new ways of prediction. Fifty-nine patients admitted to the Geneva University Hospitals after a suicide attempt were video-recorded during an interview with a psychiatrist. After the interview, the therapist was asked to assess the suicide risk on a 4-point scale. At 24-months follow-up, we identified 10 Repeaters, who were matched with 11 of the 48 Non-Repeaters, with respect to gender, age, and number of previous suicide attempts. To code the doctor's and patient’s facial behaviour, we used Ekman and Friesen’s “Facial Action Coding System” (FACS) and we analyzed her behavioural differences with both groups. Results indicated an average activation of all coded units, peri-ocular activation, and duration of her gaze straight at the patient, which were all significantly higher, distinguishing correctly 81.8 to 90.9 % of the patients. By contrast, the doctor's written predictions were erroneous: only 22.7 % of the patients were correctly classified. This fact reflects the doctor's perception of risk, without awareness. To analyze the structure of the interactive behaviour, we used THEME. Different types of behavioural patterns were found to occur exclusively by either repeaters or non-repeaters and significant differences were found in the complexity of patterns between groups. They give us a sense of the nonverbal communication