## Ebola numbers

The ebola epidemic is a big deal. It's killing people, and it's going to kill many more.

One of the things people in goverment do these days when confronted with a problem is turn to the STEM experts and ask them "What's going to happen?" Sometimes, we can give good answers. But often, those answers are miscommunicated. One current example is this washington post article. The article talks about how the reproduction number for ebola is 1.5-2, and needs to come down.

The reproductive number use useful, but alone, not informative. We also need to know the serial interval of transmissions, which appears to be about 15 days (shorter than the 3 weeks sometimes quoted). This serial interval is the bright spot in the omnious dark cloud on the horizon. Influenza has a serial interval of only 3.6 days!

The CDC has done some more extensive modelling work, but you can get the basic idea with some highschool math and the formula

$\text{cases} \approx \mathcal{R}_0^{t/s}$

where $\mathcal{R}_0$ is the basic reproduction number and $s$ is the serial interval.

With a serial interval of 15 days, it will take 2 - 3 months for the Ebola epidemic to grow from 10,000 to 100,000 cases, and 4-6 months to hit 1 million. If Ebola had Flu's serial interval, it would only take 24-40 days!

This makes things more of a slow, crushing disaster than most of what we have recently experienced. Let's put the numbers into context. The plague of Justinian(540 AD) we guess killed maybe 50 million people, and the Black Death killed maybe 100 million. Right now, it doesn't look like we're going to hit those numbers. But the Indonesian tidal wave killed 1/4 million people in 2004, and that kind of number is well-within reach. This will dwarf the impact of tornadoe disasters. The deadlyest earthquake in the last century only killed 300,000 (Haiti, 2010), while it seems quite possible that Ebola will hit 500,000 in early 2015. But on the other hand, this is still a small number compared to the 2.5 million Americans that die every year! It will take a full year from now before the Ebola epidemic starts competing with that number.

We have vaccines against Ebola already in clinical trials, and every indication is that vaccines will work against Ebola. The big question for us in the USA is how fast we can scale up production of our best vaccinees. Of course, if we fail to get a vaccine, the world may be a very different place in 2 years. Another risk is if Ebola evolves to have a faster serial interval or easier transmission, in which case we might be talking about 10 or 100 times greater impacts.