We usually meet traditionalist
doctors who would laugh when patients appear to know what ails them after consulting
the internet. But modern medicine is no longer laughing: “big data” which is
the back bone of Google is getting to be the gold standard in the practice of modern medicine.
How Big
Data Is Transforming Medicine by Bernard Marr
When we
visit our doctor or go into hospital, we have faith in the knowledge that the
healthcare professionals involved are treating us according to proven
scientific methods, otherwise known as evidence-based medicine (EBM). This
means they’re prescribing drugs or selecting treatment methods that have proven
successful in clinical research.
Although
the term ‘evidence-based medicine’ only dates back to the early 1990s, the
concept itself is much older. Controlled trials were routinely being conducted
as early as the 1940s, and clinical knowledge and expertise was already being
disseminated in medical journals and textbooks long before that. (In fact, the
oldest medical journal still running today, The New England Journal of
Medicine, was founded in 1812. Even older, the first official clinical trial
was conducted in 1747, into the treatment of scurvy in sailors.)
Clinical
trials and studies are all about conducting research into disease and
conditions, and the various treatment methods that may ease symptoms or
eradicate the illness altogether—they explore which treatments work best for
which illnesses and in which groups of patient. All around the world, EBM
is the established standard for the provision of healthcare.
But, in the age of big data, that might be about to change.
Clinical
trials work by testing new treatments in small groups at first, looking at how
well the treatment works and to identify any side effects. If a trial proves
promising, it is expanded to include larger groups of people. Often the trial
will include comparing the new treatment to other treatments by separating
patients into different groups, each trialing a different treatment. This is
usually done by a process called randomization, where patients are assigned to
the various groups randomly.
In order
to safeguard participants and improve reliability, clinical trials have to meet
rigorous scientific standards. However, that’s not to say there is no risk of
methodological flaws, or that the small-ish populations used in clinical trials
always generalize well outside of a particular study. This is where big data
can help. By mining the world of practice-based clinical data—i.e.
actual patient records—for information on who has what condition and what
treatments are working, we could learn a lot about the way we care for
individuals.
One
company, California-based cognitive computing firm Apixio, has firmly set its
sights on enabling healthcare providers to learn from practice-based evidence
to individually tailor care. As Apixio CEO Darren Schulte explains, “We can
learn more from the practice of medicine and refine our
approach to clinical care. This gets us closer to a ‘learning healthcare
system’. Our thinking on what actually works and what doesn’t is updated with
evidence from the real-world data.”
A
staggering 80% of medical and clinical information about patients is formed of
unstructured data, such as written physician notes, consultant notes, radiology
notes, pathology results, discharge notes from a hospital, etc.
Schulte,
a physician who was Apixio’s Chief Medical Officer before being appointed CEO,
says, “If we want to learn how to better care for individuals and understand
more about the health of the population as a whole, we need to be
able to mine unstructured data for insights.”
Electronic
Health Records (EHRs) have been around for a while but, with the data stored
across a number of different systems and formats, they are not really designed
with analysis in mind. So before Apixio can even analyze any data, they first
have to extract the data from these various sources (such as GP surgeries,
hospitals, government Medicare records, etc.). Then they need to turn that
information into something that computers can analyze. Clinician notes can come
in many different formats—some are handwritten and some are in a scanned PDF
file format—so Apixio uses OCR (optical character recognition) technology to
create a textual representation of that information that computers can read and
understand.
The data
can then be analyzed at an individual level to create a patient data model, and
it can also be aggregated across the population in order to derive larger
insights around the disease prevalence, treatment patterns, etc.
Schulte
explains, “We create a ‘patient object’, essentially a profile assembled using
data derived by text processing and mining text and coded healthcare data. By
creating this individual profile and grouping together individuals with similar
profiles, we can answer questions about what works and what doesn’t in those
individuals, which becomes the basis for personalized medicine.”
Getting
healthcare providers and health insurance plans to share data can be a
challenge, which Apixio overcomes by making sure they offer real value in
return for access to data. As Schulte says, “Unless you solve a real critical
problem today, none of these organizations will give you access to any real
amount of data.” Thus, the focus needs to be on tangible results and solving
problems as opposed to adding to all the hype around big data. Schulte confirms
this: “CIOs at hospitals don’t often see a lot of problems actually being
solved using big data. They see a lot of slick dashboards which are not very
helpful to them. What’s helpful is actively solving problems today.”
Another
big challenge when it comes to patient health data is security, especially
after some high-profile health data breaches. In 2014, medical records
accounted for 43% of all data stolen and the healthcare sector has seen the
biggest increase in data theft since 2010 (far more so than business or
government sectors). Schulte refers to data security as “table stakes”, meaning
it is an essential requirement for anyone wanting to operate in the healthcare
big data arena and confirms, “For every new contract we have to demonstrate our
security.” Patient data must be encrypted at rest and in-transit, and Apixio
never exposes personal health information (PHI) unless access is absolutely
needed by Apixio staff.
Can
practice-based medicine oust evidence-based medicine as the gold standard in
healthcare provision? Perhaps not. But there’s no doubt that we’re on the verge
of exciting new ways to understand, treat and prevent diseases. As Schulte puts
it, “We’re in a new world in terms of the way healthcare is going to be
practiced, based upon these data-driven insights.” A future which combines both
evidence-based and practice-based medicine is likely to produce the best
outcomes for patients—and, at the end of the day, that’s all any healthcare
professional wants.
