Pattern recognition versus bayesian approach for diagnosis in primary careBMJ 2006; 332 doi: https://doi.org/10.1136/bmj.332.7542.646 (Published 16 March 2006) Cite this as: BMJ 2006;332:646
- John Fletcher, general practitioner (firstname.lastname@example.org)1,
- Robin Fox, general practitioner1
- 1 Bicester Health Centre, Bicester, Oxon OX26 6AT
- Correspondence to: J Fletcher
Our first thought on reading this case was, “That's just too unusual. We can't possibly send every 28 year old woman with a bit of chest pain off for echocardiography.”1 In thinking that, we were using pattern recognition—a common approach to clinical diagnosis. We assumed that a young woman could not have note-worthy ischaemic heart disease.
But is there more to this case? If we were to substitute a 28 year old woman for a 50 year old man in the same scenario, most clinicians would recognise a familiar …