HealthMap, an example of information curation

Sophia B. Liu, a crisis informatics researcher from the University of Colorado, gave a talk at the International Conference of Crisis Mappers in Boston describing the concept of curation. She explained how in this day of information overload it is important to find trends or “the signal in the noise”. How do we decide what information is important versus what information is redundant or unreliable? She described how the process of crisis information curation is similar to the work a curator of an art exhibit, or a newspaper editor:  one or more people decide what is important, what stories are told. The curator also provides context by relating the story to time and place.

[picapp align=”left” wrap=”true” link=”term=Health+map&iid=6873483″ src=”http://view4.picapp.com/pictures.photo/image/6873483/special-school-helps-teen/special-school-helps-teen.jpg?size=500&imageId=6873483″ width=”380″ height=”251″ /]Of course this relates to crisis mapping since, as one person in the audience tweeted, “If it’s not on the map, did it happen?” This comment simply points out how responders and governments need information in a digested format in order to make decisions, to see the trends. The Health Map is a wonderful example of information curation but with a specific focus on diesease outbreaks. From their website:

HealthMap brings together disparate data sources to achieve a unified and comprehensive view of the current global state of infectious diseases and their effect on human and animal health. This freely available Web site integrates outbreak data of varying reliability, ranging from news sources (such as Google News) to curated personal accounts (such as ProMED) to validated official alerts (such as World Health Organization). Through an automated text processing system, the data is aggregated by disease and displayed by location for user-friendly access to the original alert. HealthMap provides a jumping-off point for real-time information on emerging infectious diseases and has particular interest for public health officials and international travelers.

Health Map was co-founded by Dr. John Brownstein of Harvard Medical School and Ph.d candidate, Clark Freifeld, a research software developer at the Children’s Hospital Informatics Program.  I had the pleasure of meeting both of them at the ICCM conference. Their team consist of 13 other equally impressive people with backgrounds ranging from biomedical engineering to epidemiology and biostatistics.

To learn more about what information is provided on their system see their HealthMap 3.0 Tutorial. In general, when you view the map you see either pins which represent precisely placed alerts or dots which represent country, state or province-wide alerts.  The dots are color coded based on a mathematical computation that takes into account time-frame, number of alerts and number of sources. You can customize the map to provide specific information that you are interested in, for example, by location or disease.

Users are able to add to information to the map with the “outbreak missing feature”–which can even be done on-the-go with a mobile app. People are encouraged, however, to add links to news articles which helps the team verify the information. And in fact, all reports are reviewed prior to being displayed. This user-generated information has its own category “HM Community News Reports” so that other users can understand the source of the content.

This model of curation is laudable, however, I wonder if it would work for information aggregation of real-time disaster data. The verification system mostly depends on reports from sources such as newspaper articles or Government or UN released reports; but during a fast moving crisis, by the time information is released by a news organization it could already have become overcome by events. There is no doubt, however, that this system is one of the very best for understanding disease outbreaks.

2 responses to “HealthMap, an example of information curation

  1. One thing that is discussed again and again is verification of data. It seems logical to want to ensure that actionable data is accurate. Indeed it is a worthy goal, but as mentioned in this article, nearly impossible in a crisis situation.

    Getting back to the basics though, comes the question I haven’t heard answered much.. what is the motivation for false reports? I few quick answers can be produced but what is the evidence? The populace would be reasonably interested in obtaining aid of some sort and presumably exaggerating their need in order to speed delivery.

    In this case, it seems that we could change the motivation proactively by setting up systems that allow communities to be accountable to one another. Is there any reason not to have transparency of supplies and distribution? Knowing the limitations of crisis response vis materiel would possibly prevent the exaggerated reports and promote more peer to peer support.

    Technologically the means is coming quickly to share information like never before. My conclusion is that street level transparency would go a long way to reduce the motivation for false or exaggerated reporting and build trust. Having enthusiastic supporters who participate in the information sharing changes the ease and scope of what is possible in the recovery period.

  2. I too believe there are many ways to verify information, especially if you have the ability to do cross referencing. But with regard to opening up all info, I honestly don’t know if that’s the answer. People seem to demand supplies even when they don’t need them. One example I heard was during the Tennessee floods people that lived on a hill fervently demanded sand bags for their property, even though they were at no risk of flooding. Would knowledge that the county only had X amount of bags to and to cover Y amount of river-front have stopped that kind of behavior? Maybe.
    Hoarding is another issue. If you publish the amount of supplies and people suspect a shortage, would you get folks trying to game the system? It might also depend on how savvy your population is too. One model might work for the U.S. but might not work in Haiti, for instance.
    In the end, I suspect that you are right. Open information will probably happen anyway as more people are able to gather data themselves and publish it on their own, governments will be forced to be open whether that was their intent or not.
    Thanks for the comment. These are large issue to grapple with.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s