It shouldn't take years to diagnose a rare disease.

I want to introduce you to Emily[1]. She is 11 years old. She and her Mom are buying school supplies when her face starts to tingle. She tells her Mom, but she seems okay, so they continue shopping. On the way home Emily reports that she can’t see very well. Mom turns around to see her daughter’s face so swollen that her eyes are almost shut. Emily has never had an allergic reaction before; they go to the ER. She is treated with IV fluids and steroids to decrease the inflammation. The swelling gets worse overnight and finally decreases after another 24 hours. She starts sixth grade with that episode behind her. She feels back to normal, and forgets it ever happened.

In December Emily goes to the dentist for her cleaning. She has no cavities, takes her new orange toothbrush, and heads home. A few hours later, she feels a little winded. Two hours later she is really having trouble breathing. Her Dad calls 911, and they fly off to the hospital. Three days later, after a breathing tube and a brief stint in the intensive care unit, she is released home. She is back to normal.

Three days later she is seen by an allergist to try to determine the cause of these two episodes. She has scratch testing, which only reveals a mild allergy to pollen. The doctor provides her with an epipen to use if she has another attack.

Now Emily is 13. She has had four more attacks like the first two, she sees the allergist regularly, and has used 3 epipens. Three of these events have landed her in the hospital. It is February and she gets the stomach flu. No one else in her family does. For three days she experiences nausea, vomiting and abdominal pain. She misses school again. Her grades are starting to suffer, and she constantly feels afraid that an attack will come.

Five years pass. Instead of thinking about college she is trying to avoid attacks.

Here is what Emily knows:

  • If she plays in the snow, she gets an attack. No sledding. No skiing. No snow ball fights. She stays inside on cold days
  • When she goes to the dentist she gets an attack. She only sees the dentist once/year, and so far the allergist has been unable to prevent the episodes, no matter what drugs he tries. Dentist = hospitalization
  • She gets the stomach flu more than anyone she knows. She is so prone to it that she is often the only person in her family to get sick. She has seen a gastroenterologist a few times, but that has been no help
  • She lives in fear of her period. When puberty first started for Emily, she was in the hospital 6 times in 6 months. Her doctor tried putting her on birth control pills to see if that would help. That month, she was in the hospital twice.

Have you noticed what Emily doesn’t know?

Her diagnosis.

She has been sick for seven years now. If her luck doesn’t change, she might not be diagnosed for another seven years.

There MUST be a way to help Emily. She doesn’t have allergies. She doesn’t get the stomach flu more often than everyone else. She has a rare disease.

The doctors she is seeing are perfectly smart, well-trained people. But medical school was a while ago, and chances are they didn’t learn about her disease then anyway. They have never seen Hereditary Angioedema before, and probably never will again.

The good news for Emily is that she lives in a time of data collection.

I do not like it that my least favorite big box store knows I buy chocolate-covered raisins. I do not like the pop-up car insurance ads that appear on my computer after I research my next car.

But in the field of health care? We might actually save a life. We might help an Emily live to her fullest potential.

With the advent of electronic medical records, and the use of them being mandated, we are confronted with a new era. We are seeing a confluence of the ability to collect vast amounts of data, aggregate it and analyze it. There are now commercially available health databases that store information on over 500 million people.

At Vencore, we have been using sophisticated analytics for decades in the defense community. We are applying our data mining and systems integration capabilities to the field of healthcare. In particular, we are finding patients who have a rare disease but are not diagnosed. It turns out that if we can find “bad guys” we can find sick people. We have developed methods to find people who look like Emily – based on the diagnoses they carry, and the specialists they have seen – and point to a possible diagnosis.

We have the capability to analyze terabytes – or petabytes or exabytes – of data in an integrated fashion. In addition, we can use our systems integration tools to integrate data. We can fit different components, or sub-systems, together seamlessly. We are building products that enable claims, EHR, radiology and lab data, for example, to interface in a meaningful way. This provides a far more holistic view of the patient’s journey than was previously possible.  

It is of course not reasonable to expect one provider to be able to do all of that. Remember, one exabyte of data is thousands of times larger than all of the information stored in the National Library of Medicine.

We are leveraging the expertise of data scientists coupled with the knowledge of medical professionals to engineer a novel approach to the problem of rare diseases. I used to be part of a large internal medicine practice. At my busiest, I had 2500 patients. I never ONCE saw a patient with any of the rare diseases we are targeting. I will readily admit that I might not have known it if I had. Now I feel like I can impact the lives of hundreds of millions of patients. I see the power of what we are doing stemming from the union of these two fields.

Ten years ago, Emily would have languished for several years more until someone figured out what she had. That someone is often the patient – it is not uncommon for people with rare diseases to diagnose themselves through internet research. The problem with waiting twelve years for a diagnosis is that Emily might die first. Sometimes the throat swelling is so severe that people can’t breathe at all. Even if she doesn’t die, she isn’t exactly living the life we all want for our children.

We have to diagnose these patients sooner.

We have to get them into proper treatment.

If we can find patients who look like Emily, and let their doctors know, then they could get on treatment and avoid years of unnecessary hospitalizations.

If we can help find the right patients for drug trials, then we could get life-altering drugs to market more quickly, and more cost-effectively.

We have to acknowledge that we live in an age of data. So far we have used big data mostly to make money – through advertising, investing and social media. There is value to that of course. But wouldn’t it be better if we could also use technology to improve our health? To help people suffering for years without a diagnosis? To help them in a way that respects their privacy? We can use these new data-based tools – the right way – to help doctors provide the highest quality of care. There are estimated to be 55 million people with rare disease in the United States and the European Union alone. Let’s start finding them.

Tara Grabowsky, MD
Chief Medical Officer