Truly Negative Result
![[Print This Post]](http://www.rantlust.com/wp-content/plugins/wp-print/images/print.gif "[Print This Post]")
A final year MBA student from my alma mater committed suicide last week after having been diagnosed with Niemann-Pick Disease, a fatal genetic disease.
According to Wikipedia, the highest occurrence of NPD is in the Ashkenazi Jewish community, with an incidence of 1 in 40,000. Given this incidence and a test accuracy of 99.9%, the chance of actually having the disease, if tested positive, is just 2.4%. If the test accuracy fell to 99%, the chances drop to 0.25%.
Could this be a case of the test providing a False Positive?
While sad, one would have hoped that an MBA student has a better grasp of conditional probability (if indeed the perceived long odds was the reason for his/her suicide). A famous case is that of Sally Clark who was convicted at least in part on questionable statistical evidence.
http://www.sallyclark.org.uk/RSS.html
http://plus.maths.org/issue21/features/clark/index.html
Not sure if there have been any further updates to that case.
That 1 in 73 million figure was obtained by squaring the probability of one cot death.
The doc was struck off from the Medical Council for the damaging witness but he successfully appealed against the verdict. In addition to Sally Clark he sent three other innocent mothers to jail.
Can you tell me more about this student? I have a child with Niemann Pick and am interested in learbing about cases where adults had the diagnosis. My guess is that the form this student had was not one that Ashkenzi Jews usually get but I may be wrong.
Thank you
Jonathan,
The student was of Indian origin. To be specific, he was a Tamilian from Chennai. I don’t have any more details on the testing either.
The point of this post was not so much the disease but how statistics can be misintrepreted.