Find Zebra Project:
I wish I had found this a couple weeks ago, so I could have blogged it on International Rare Disease Day, which is always the last day of February. Find Zebra is, get this, a rare disease discovery engine!
I found out about this through this piece in the Physics Arxiv blog of the MIT Technology Review.
The Rare Disease Search Engine That Outperforms Google. March 18,
MIT found it in Arxiv.
Radu Dragusin, Paula Petcu, Christina Lioma, Birger Larsen, Henrik L. Jørgensen, Ingemar J. Cox, Lars Kai Hansen, Peter Ingwersen, Ole Winther. FindZebra: A search engine for rare diseases. Arxiv. http://arxiv.org/abs/1303.3229 (Submitted on 13 Mar 2013)
International Journal of Medical Informatics, Available online 23 February 2013. DOI: 10.1016/j.ijmedinf.2013.01.005
So it was theoretically available in time for Rare Disease Day, it wasn’t very easily discoverable.
In their own words:
“The term “zebra” is a medical slang for a surprising diagnosis. Physicians are taught since medical school to concentrate on the more common diseases: “when you hear a gallop, you should think about a horse, not a zebra”. “I look for zebras because other doctors have ruled out all the horses.” — Dr. Gregory House”
I especially like this part from the beginning of their Arxiv article.
“The web has become a primary information resource about illnesses and treatments for both medical and non-medical users. Standard web search is by far the most common interface for such information. It is therefore of interest to find out how well web search engines work for diagnostic queries and what factors contribute to successes and failures. Among diseases, rare (or orphan) diseases represent an especially challenging and thus interesting class to diagnose as each is rare, diverse in symptoms and usually has scattered resources associated with it.”
We spend a lot of time in the #HCSM community and Twitter chats talking about issues of online health information credibility, authority, reliability, and usage patterns. Indeed, I’ve spent a large part of my career looking into these concerns. I also have worked closely with some rare disease communities, persons with disabilities, invisible disabilities, and have an interest in those areas as well. One of the largest problems in these communities is often the challenge of seeking a diagnosis, of simply finding out what the heck is wrong, and whether or not you can do anything about it. Patients often spend years or decades going from clinician to clinician seeking answers.
A number of tools have been developed previously to assist doctors with this, usually through developing decision support and identifying differential diagnoses to be excluded. There have been some challenges with physician awareness of these tools and clinician adoption. One reason might be that these tools have typically not only identify alternative diagnoses, but then also recommend specific tests to be requested, and other helpful tips. Concerns have ranged from the idea that the tool is trying to supplant the clinician’s judgment to the opposite idea that clinicians may come to depend on the tool instead of their own judgment. Find Zebra neatly avoids this pitfall by sticking to web searching without making any overt recommendations, while focusing on resources of high utility to clinicians.
Find Zebra clearly states that it is intended for use ONLY by medical professionals.
In their own words:
“FindZebra should only be used by medical professionals. Although the articles indexed by the system have been written by medical
professionals or reviewed by medical associations, it is strongly recommended that, as a patient, you consult you local health care provider. FindZebra does not replace professional health care, and cannot be held responsible for erroneous use of the information provided through the system.”
This is a good thing, for the most part. The biggest reason why it is a good thing is that the databases is constructed explicitly to focus on oddball diagnoses, the unusual, the atypical. It is a bit like the old meme that to a surgeon, every condition looks like something that needs surgery, that to the man with a hammer, everything looks like a nail. No matter what you put into Find Zebra, it is going to come back with suggestions that are not very likely. For combinations of symptoms that are each fairly common, the chances are extremely high that whatever it suggests is wrong. For the typical person who actually has a combination of a few unconnected relatively common conditions, Find Zebra will gladly generate a tidy list of exotic diseases to consider. For patients, the risk is that they may end up with a list of completely irrelevant and somewhat scary new ideas to worry about needlessly.
Still, I am also glad that it is not locked down so that patients and the public cannot get to it. Patients with rare diseases or their families and loved ones often end up becoming quite expert in their condition, and also end up becoming quite active participants in identifying trends and impacts in their care, treatment, and diagnosis, ideally in partnership with their clinical team. When you haven’t gotten a diagnosis, or you’ve gotten several but none of them have actually proven accurate, you tend to get a bit desperate and keep digging, hoping that you’ll luck upon something that will help you find an answer and a solution. The longer this drags on the more likely the family is to look at ever farther ranging possibilities, ever more unlikely scenarios. This is when general web searches can lead to those horrible sites that exist for no other reason than to offer hope in the form of lies and take money from people who haven’t found answers anywhere else.
I know of story after story of patients who actually HAVE found answers through digging on their own, asking questions, joining patient communities, being persistent. For each them there are countless others who haven’t, who didn’t know the right questions to ask, the right people to question, the right sources to search, the right words to use. Even though Find Zebra is not intended for the public, in my opinion at this point in time, I would far rather a patient search the web with tools more like Find Zebra than Google or other general search engines. Here’s why.
Find Zebra searches a custom designed subset of information resources focusing explicitly on quality, authority, credibility, and validity. They are searching Orphanet, NORD (National Organization of Rare Diseases), OMIM (Online Mendelian Inheritance in Man), GARD (Genetic and Rare Diseases Information Center), HON, Genetics Home Reference, and more. There are a few other sources I wish they’d included, and couple I question why they included them, but overall, I’m pretty happy with this set. If a patient is searching here, they are going to be more likely to find information that is actually quality. On the other hand, the information is going to be hard to understand, and health literacy may be a real barrier to how useful this can be.