Health anxiety in medical students

Abstract

matched medical and law students (60 in each group), and found that medical students attended to their health and somatic symptoms more than law students but there was no significant difference in the rate of hypochondriasis. In January, 1996, a short illness behaviour questionnaire* (the Whitley index, a version of the Illness Behaviour Questionnaire) and the Short Health Anxiety inventory were distributed to all 306 clinical medical students at University of Oxford Medical School. No personal identifiers were asked for. During this time, a comparison group was surveyed which comprised all the students studying any subject except medicine at two Oxford University colleges. 183 (59·8%) medical students replied. 148 (48·4%) completed the questionnaire. 110 (non-medical) students completed questionnaires. Data on file for a general Oxford population was also used. An analysis of variance was carried out with SPSS. On total scores for the health anxiety questionnaires, medical students scored nonsignificantly lower than the other students and the nonstudents (p>0·1). An (uncorrected) item analysis for the 18 items on the health-anxiety inventory found significant effects for three items; in two of these, medical students were lower than the other two groups (awareness of bodily sensations or changes; perceived risk of illness; table). On perception of the likely effectiveness of medical treatment in the event of illness, non-medical students perceived treatment as more likely to be effective than the other two groups. The only other difference was that medical students were less likely to avoid seeking medical help. These results question the widely held view that medical students are more likely than others to have excessive anxiety about their health. Two factors may contribute to the impression that this is a common reaction to medical training. First, the phenomenon of trainee doctors who falsely believe themselves to have an illness may receive more attention from and be more memorable to both tutors and the student’s physician than the same phenomenon in a geology student. Second, the early claims of high prevalence (more than 70%) may have resulted in selective attention to the phenomenon and a lowered threshold for its recognition.