What is the Zika virus? Should you be worried about it? And is it really causing severe birth defects in Brazil?
For more information about Zika:
European Center for Disease Prevention and Control: http://ecdc.europa.eu/en/publications/Publications/zika-virus-americas-association-with-microcephaly-rapid-risk-assessment.pdf
Suprisingly on point opinion piece in the Daily Beast: http://www.thedailybeast.com/articles/2015/12/24/doctors-warn-brazilian-women-don-t-get-pregnant.html
To answer the question about the specifics as to why I am so skeptical of Brazil’s reportedly extremely low baseline rate of microcephaly, here’s my response that I copied and pasted from the comments below:
Essentially, there are 2 reasons why Brazil’s baseline rate of microcephaly is implausible.
1. In the medical literature, there are 2 definitions of microcephaly used/advocated for: A head circumference that is either 2 or 3 standard deviations below the mean. For any body measurement that adheres to a normal distribution in the population, we would expect .0014% of the population to fall below 3 SDs of the mean – this is an inviolatable rule of statistics (https://en.wikipedia.org/wiki/68-95-99.7_rule). If head circumference adheres to a normal distribution, then one can calculate that, on average, we would expect ~4200 babies per year in Brazil to have microcephaly (0014% x ~3 million births in Brazil per year). It’s very true that head circumference likely does not strictly adhere to a normal distribution, since profoundly small head circumference (while rare) is still more common than profoundly large head circumference . However, it’s hard to imagine that the distribution of head circumference would be so dramatically skewed such that 4200 microcephalic babies per year becomes just 150.
2. In the medical literature, there is a surprising lack of good data on the empirically measured rates of microcephaly in the population, which could be used to determine if it is normally distributed. In the few primary papers that exist, some of which date to the 1950s, the reported incidence vary 100 fold between different countries. While the rates of individual diseases can vary 100+ fold between countries, microcephaly isn’t an individual disease, but rather a syndrome and/or sign of dozens if not hundreds of different conditions – which are a mix of prenatal infections, maternal drug/toxin exposure, and genetic/chromosomal diseases. It is implausible that a condition with such a huge variety of causes, some of which will be more likely in country A while others will be more likely in country B, would have such a difference in incidence in different parts of the world. In other words, the regional/geographic effects of the incidence of individual causes of microcephaly get averaged out when all causes are considered collectively. So there cannot be a 100 fold variation in the true incidence of microcephaly between countries. With the limited country-specific data that exists, Brazil sits at the extreme low end of incidence.
You could ask, why would I less believe Brazil’s reported baseline microcephaly incidence of 0.005%, than a country like the US with a much higher reported incidence? It’s because it’s much more likely that cases of microcephaly get underreported than overreported, particularly if a newborn technically meets the definition based on head circumference, but has no other obvious physical or developmental problems. If I’m a pediatrician, I may not want to label the patient microcephalic, since that may unnecessarily upset the parents, who will then believe there is something wrong with their child, when there very well may not be – some otherwise normal kids just have small heads.
In addition, as with mandatory reporting of certain infectious diseases in the US, not every doctor takes the time to do the official reporting, or even knows how the reporting process works. So the official count of any disease is lower than reality.
Therefore, irrespective of which specific countries have a low reported rate of microcephaly and which have a high reported rate, the high reported rate is more likely to be accurate. And as mentioned above, Brazil has an unusually low reported rate compared to other countries – thus the reason I don’t believe it.