Caitlin Rivers

This morning the World Health Organization released downloadable, machine-readable data (yay!) of the number of Ebola cases at the county level (available here). This release is particularly special, because it includes both the number of cases according to the situation reports and the counts as reported by the patient database. The patient database (also known as a line list) is usually considered the gold standard for outbreak data. Until this release, the public had no data from the database - the situation reports were the only resource.

The good news is that this data is immensely useful for epidemiologists, modelers, public health responders - pretty much everyone involved with Ebola work. The bad news is the situation reports are apparently fairly unreliable. Ideally the two data sources would match up very closely. This is not the case.

It's disappointing that the sitreps are not what we thought they were. But ultimately I'm really glad that this information is being made available, because now we can adjust accordingly. I think we should commend the Ministries of Health and the WHO for releasing this data - I value open data and transparency very highly, even when it brings some surprising results. 
[Graphics below the break]

I've had several people ask me how they can help to fight the Ebola outbreak in West Africa. I have no particular knowledge about the answer to this question, but I felt it was worth looking into. Here is a roundup of things I have found. Feel free to add any additional information or suggestion in the comments section.

Medecins Sans Frontiers (aka Doctors Without Borders) is an NGO that, as far as I can tell, one of the primary organizations involved in fighting this outbreak (and innumerable others). They also have an excellent score (92.34 of 100 overall, 89/100 for financial, 100/100 for transparency) on Charity Navigator. Donate using this site.

MAP International "is responding by providing infectious disease protection suits and supplies." They sent 5,600 protective suits as recently as Aug 6. MAP is a Christian-affiliated organization. They also have a great score on Charity Navigator. They have donation links all over their website.

Samaritan's Purse is an evangelical Christian organization that does medical missions (among other things) in the affected area. Instead of donating money, you can choose from a 'gift catalog' for things like emergency food. Their Charity Navigator score is also >95/100.

According to their website, Medical Teams International is mobilizing community health workers, providing infection control training, and supporting health workers. They appear to be operating primarily in Liberia. Their Charity Navigator score is lower than the others, but still decent at 85.

UNICEF is helping by donating supplies and organizing behavior interventions.  Their Charity Navigator score is 94, and they are actively seeking donations to help fight the outbreak.

ProMED-mail is a mailing list that reports infectious disease outbreaks to 60,000 subscribers daily. They are often the first ones to break news of an outbreak to the global public health community. I donate to ProMED fairly regularly. They are not listed on Charity Navigator. Donate using this link.

EcoHealth Alliance does a lot of great research at the human/animal interface, which is where a majority of emerging infectious diseases originate. A list of the research programs they run is available here, and their donation link is here. Their Charity Navigator score is 90.

OpenStreetMap has an open-source response going. I'm not familiar with their response, so I can't really provide more details on this one.

Contact your local government and encourage them to devote resources to the outbreak. I poked around and wasn't able to find any specific legislation being considered.

I'm sure that can't be it, but I haven't found too many other crowdsourced Ebola projects. This surprises me, because past disasters (e.g. Hurricane Sandy), a lot of projects popped up. If you have any ideas or know of any projects, please email me or leave a comment.

The CDC announced today that a second case of MERS coronavirus in the United States has been identified in Florida. Worldwide there have been 538 cases and 135 deaths since the virus first emerged in humans in 2012. The Florida patient, whose sex has not been released, is a healthcare worker living and working in Saudi Arabia, where a majority of cases have been found. 

On May 1 the patient traveled from Jeddah, Saudi Arabia to London, and then continued on to Boston and Atlanta before reaching his or her destination in Florida. The patient experienced symptoms while traveling, but it is unclear whether he or she was infectious during that time. On May 8 the patient sought care and was admitted at an Orlando hospital. The patient is doing well.

The CDC is working in collaboration with 20 state health departments to contact travelers who shared flights with the infected patient. The incubation period for the virus is estimated to be 14 days at most, so people who shared a flight with the infected but have not yet experienced symptoms are unlikely to become sick. 

The first case of MERS in the US was identified in Indiana the first week of May. A marked increase in cases in April 2014 has epidemiologists worried that the virus had mutated to become more human to human transmissible. However, samples taken from the Indiana case suggest that the virus has not mutated, and that the increase in cases is consequent to better identification of existing cases.

The CDC does expect additional cases to be imported from the Arabian Peninsula, but cautions that risk to the general public is ‘extremely low’.

Read my analysis on what it means here.

One year ago, I complained about the state of public data. I specifically pointed out that over 100 years of Nationally Notifiable Disease Surveillance System data were trapped in PDF, basically unusable. Thankfully, Project Tycho has since come to the rescue by releasing NNDSS data from 1888 to 2013. 

Not only can you retrieve machine readable data from their website, but they have an API! Although wonderful, APIs can sometimes be a bit of a pain to get to know, so I wrote a python wrapper to make life a little easier. I've named it pycho, and you can find it on github here. 

Epidemiologist Paige Bordwine shared with me why she loves epidemiology, and what she wishes people knew about public health. This is an excerpt from an interview on being a 'disease detective' during the fungal meningitis outbreak.

Is there anything else you wish people knew about public health, epidemiology, the work that you do?
It is the most fun job I've ever had. I don't mean fun in the sense that it's good having people that are sick, but it is fun and very personally fulfilling being able to offer them something. Whether it's education, or help them connect the dots on care, or services they can get. 

I went on an Epic Quest today to find publicly-available line listings. I found a grand total of three machine readable ones (one of which is my own), and a couple more non-machine readable ones. 

Update: Epipy is coming right along! I've added multiple plot types, and a few functions for epip analyses. You can learn more at cmrivers.github.io/epipy, or browse the code at github.com/cmrivers/epipy.

Fourteen months after the fungal meningitis outbreak was first recognized, it is finally grinding to a halt. Throughout the outbreak I have been providing intermittent analyses (see the bottom of this post for links), and I also tried my hand at science journalism by interviewing two professionals on the front lines of the outbreak. This will probably be my last update on the topic. As always, this blog is just a hobby, so enjoy these analyses at your own risk.

There are an astonishing 751 cases and 64 deaths across 20 states. Michigan (264 cases), Tennessee (153 cases), Indiana (94 cases) and Virginia (54) cases were the hardest hit. 

One of the most striking features of the outbreak was how variable outcomes were from state to state. Attack rates ranged from just under 14% in Pennsylvania to 18% in North Carolina. 

Author's note: This article was written the week of March 25. I haven't been able to find an outlet for it to be published, so I'm offering it here. It's my first attempt at science communication for a general audience. Please let me know what you think!

Last October an outbreak of fungal infections rocked the East Coast. Dozens - then hundreds - of people were falling ill, and some were dying. Medical professionals struggled to treat a disease about which almost nothing was known. Epidemiologists worked around the clock first to trace the source of the outbreak, then to monitor people who were at risk of developing the disease.

Although the story has since fallen off the news ticker, the outbreak continues to unfold. To date, 733 people in 20 states have been diagnosed with a fungal infection; of those, 53  have died. Roughly 6% of the people who received a contaminated injection have fallen ill with meningitis, stroke, or infections in the soft tissues around the injection sites. The long timeline of the outbreak is unusual. The first cases occurred just days after the patients received a contaminated injection, and yet more cases are still cropping up six months later.

What if I told you a brand new public library is coming to your town. It's going to be really well stocked with great books - but you can't open them, you can only look at the covers. That is the current state of our public data right now. Here's why you should care.