A few years ago, a rheumatologist by the name of Jennifer Frankovich was assigned to the treatment of a young girl showing symptoms of kidney failure. Rheumatologists typically diagnose and treat disorders in joints, muscles, and ligaments; but in this case, Dr. Frankovich was skeptical that the young girl’s symptoms were signs that directed attention to her joints, muscles or ligaments. Instead, she was suspicious that the patient’s combination of symptoms were similar to those in past cases of lupus patients with life-threatening blood clots.
Comparing the new patient to previous ones without enough records to affirm her suspicions, Dr. Frankovich took on a special project in search of some answers. To begin, she searched a database of previous lupus patients for ones who had symptoms similar to those in her new patient. The second step in her search was to run a test within the database checking if the patients developed blood clots. After her collected statistical analysis, she changed the minds of everyone who doubted her, showing an inarguable probability of a blood clot in the new patient.
With this newly discovered likelihood, Dr. Frankovich and her team decided to give the patient a drug that would prevent any blood clots. The only trouble is that they are still unsure whether the patient needed that drug. They are only sure that with their limited information, that was the best decision. To prove Dr. Frankovich’s hypothesis correct, a much more advanced and timely clinical trial would need to be done, and that kind of research is rare.
It seems that Dr. Frankovich’s focused analysis saved a patient from blood clots, and so many leaders in the field are currently pushing for such focused research to become more common. With it, doctors with few patients would be able to steer medical care of many patients in better directions.
Luckily, there is a newly growing interdisciplinary field, known as biomedical informatics, that uses computer science to create algorithms for filtering medical databases to search for unknown correlations. Previously, interactions between two commonly prescribed drugs have been found to have dangerous side effects when taken together. Researchers have also noticed that the months or seasons of birth may be correlated to risks for certain diseases. For example, most multiple sclerosis patients are born during the Spring in the Northern Hemisphere, while in the Southern Hemisphere, they are born during the Fall.
Such correlations can change the way that medical decisions are made, and the way that treatment for certain illnesses are carried out. Over time, and with sufficient imperative refinement, this kind of medicine may become the most common kind. After all, research of the past is historical evidence that often benefits the present.
