Opening Pandora's box: the unpredictability of reassurance by a normal test result

Abstract Objectives: To determine the rate of failure of patient reassurance after a normal test result and study the determinants of failure. Design: Replicated single case study with qualitative and quantitative data analysis. Setting: University teaching hospital. Subjects: 40 consecutive patients referred for echocardiography either because of symptoms (10 patients) or because of a heart murmur (30). 39 were shown to have a normal heart. Interventions: Medical consultations and semistructured patient interviews were tape recorded. Structured interviews with consultant cardiologists were recorded in survey form. Main outcome measures: Patient recall of the explanation and residual understanding, doubt, and anxiety about the heart after the test and post-test consultation. Results: All 10 patients presenting with symptoms were left with anxiety about the heart despite a normal test result and reassurance by the consultant. Of 28 patients referred because of a murmur but shown to have no heart abnormality, 20 became anxious after detection of the murmur; 11 had residual anxiety despite the normal test result. Conclusions: Reassurance of the “worried well”—anxious patients with symptoms or patients concerned by a health query resulting from a routine medical examination or from screening—constitutes a large part of medical practice. It seems to be widely assumed that explaining that tests have shown no abnormality is enough to reassure. The results of this study refute this and emphasise the importance of personal and social factors as obstacles to reassurance. Key messages Obstacles to reassurance relate not only to prob- lems of doctor-patient communication but also to the patient's past experience and social circum- stances Because of the risk of residual anxiety and the epidemiological hazard of a false positive or incon- clusive test result, referral for the test should be avoided if an expert clinical opinion will suffice Clinical skills in recognising anxiety and identifying innocent murmurs should be rein- forced to minimise unnecessary referrals The consequences of possible residual doubt and anxiety must be factored into measurements of cost effectiveness of echocardiography and of other imaging technologies