If at first you don't succeed… sue

BMJ 2005; 331 doi: (Published 20 October 2005) Cite this as: BMJ 2005;331:950
  1. Kieran Walsh (bmjlearning{at}, clinical editor
  1. BMJ Learning

    In times past if you failed an exam at medical school you had a “long dark tea time of the soul.” You asked yourself why you spent all that time in the college bar, the snooker hall, or in classical civilisation lectures. You vowed not to leave it all to the last minute for the repeats. But these days—in the United States at least—if you don't succeed you can always sue.

    Firstly, you and your lawyers can look at the validity of the exam. If you pass the exam does that mean that that you're ready to work as a doctor? If you fail does that mean that you're not ready? If the answer to either of these questions is no then you may have a case. For example, if many of the questions in the final medical exam test knowledge of basic sciences then the exam may not be a valid test. A doctor may not know the ins and outs of the Krebs cycle, but he or she may still be able to give sensible and evidence based dietary advice to patients. And what about reliability? Test-retest reliability is a good place to start. How many students who passed their final exams in May would still pass them if they were taken again in September? If large numbers of your students would fail in September, then your exam is not reliable and your students are probably crammers.

    So if you are setting an exam, how do you avoid these pitfalls? Firstly, make sure that the exam tests what it's supposed to test. If it is a test of whether candidates would make good general physicians then ensure that the questions deal with common medical problems that would be seen on an average take. So most of the questions should be on chronic bronchitis rather than porphyria, and they should test candidates' knowledge of the diagnosis and treatment of chronic bronchitis rather than its pathophysiology or epidemiology. And if you want to stop the crammers concentrate on continual assessment rather than a single high-stakes exam at the end. It is strange that some people are calling for doctors to do an exam to stay on the medical register when universities are increasingly realising that the best way of deciding who should get on the register is by regular assessment.

    At BMJ Learning we offer formative assessment; it doesn't count towards an exam or a degree, and you can repeat the modules as often as you like. One of our most recent modules is on the diagnosis and treatment of common skin tumours in primary care. It is weak on paraneoplastic pemphigus but strong on basal cell carcinoma and squamous cell carcinoma and malignant melanoma. There is little hard evidence on the effectiveness or otherwise of online learning in medicine, but one of the few areas where online learning has been shown to be effective is in dermatology and more specifically in the early detection of skin cancer.1 You are welcome to have a look at our module on


    View Abstract

    Sign in

    Log in through your institution