Science magazine reports:
A far higher number of babies in Colombia have developed microcephaly related to Zika virus infections than previously reported. [...] It now appears that incomplete reporting may explain some of the disparity.
[...] the CDC and Colombia's ministry of health and national institute of health, offers "preliminary information" about 476 cases of microcephaly identified over the last 11 months. In contrast, the latest World Health Organization (WHO) "situation report", with data current to 7 December, said that Colombia had only reported 60 cases
[...] Zika is probably not to blame for all of the 476 cases. The paper reports that 306 of the affected babies were tested for Zika virus infection with the ultrasensitive polymerase chain reaction that detects viral RNA or immune markers. Just under half, 147, had evidence of Zika virus infection.
Full paper: https://www.cdc.gov/mmwr/volumes/65/wr/mm6549e1.htm
WHO report (pdf) http://apps.who.int/iris/bitstream/10665/251905/1/zikasitrep8Dec2016-eng.pdf?ua=1
(Score: 0) by Anonymous Coward on Monday December 12 2016, @09:39PM
And those diseases have nonsensical definitions as well. This has nothing to do with sensitivity and specificity. We are not talking about a test for a disease, we are talking about the definition...
Also, my point wasn't other causes of microcephaly, it is that both the presence of a disease state and presence of a virus are highly uncertain here. We are getting into psychology-level territory with the threats to internal validity.
(Score: 2, Interesting) by shrewdsheep on Tuesday December 13 2016, @04:00PM
Maybe I made myself not clear enough. In the end *all* diagnoses can be viewed as being based on quantiles (e.g. a tumor is diagnosed if tissue growth is in the top quantiles). The level of indirection goes further: a diagnosis can be viewed as a test as to whether more tests should be performed. For *all* diagnoses getting the diagnosis does not mean you have to do anything about it. The diagnosis merely enriches the group with the diagnosis for certain problems. Step by step it is decided whether to investigate further and for this specificity/sensitivity (for giving us information on the next step) are important. At some point you stop when you feel certain enough.
(Score: 0) by Anonymous Coward on Tuesday December 13 2016, @04:09PM
In your universe it is impossible to ever reduce the rate at which a disease is diagnosed, the goal posts will just constantly move so that 2.5% (or whatever) people always must have the disease. Sounds like a great scam!