How
AI can help Life Sciences find a needle in the haystack
Efficiency is all about making the optimal use of
your resources. For Life Sciences organizations, efficiency is all about their
sales forces reaching the right physician at the right time with the right
message. Yet, the current tools are woefully inadequate.
The tools at hand
Life Sciences have
traditionally relied on prescription and claims data to learn which physicians
had seen the most relevant patients and to identify prescribing patterns. This
historic data is incredibly useful from a strategic planning perspective but it
does nothing to help with tactical day-to-day work of the sales force.
Retrospective in
nature, claims and prescriptions alone show prescribing decisions that have
already been made. You don’t know if the prescriber has patients that can
benefit from your therapy this week or this month, you are making
an assumption he or she might because they had these
patients in the past. In terms of developing the right message for these
physicians, sales reps also have to guess as to why a particular therapy was
chosen. They have no way of knowing if their assumptions are correct or will
resonate with the prescriber’s thought process.
Downward market
pressures
That’s a lot of
guesswork for the industry looking to increase sales while facing significant
downward pressures in the market. Over the past 3-5 years, Life Sciences
organizations had to operate in a changing managed care environment. Patients
are often restricted access to high-cost branded therapies or their physicians
have to take multiple administrative steps to demonstrate that a patient
should, indeed, be on the therapy they consider the most efficacious. At the
same time, some recent highly differentiated drug launches, such as for high
cholesterol or chronic heart failure, have not performed as well as expected.
Moreover, the size of
an average sales team has been shrinking in many companies. Today, you hardly
ever see a 5,000 member sales force that can visit every prospective physician
every month or quarter. Even an opportunity to speak with physicians has
diminished as some adopt a “no-see” policy or have very little time – as little
as a few minutes – to learn about the treatment options. This means that the
time reps do get with prescribers needs to be spent wisely, with the message
that will help physicians in their clinical workflow, not distract them from
it.
The game changer:
applying AI to clinical lab data
So, how do you
improve sales reps’ efficiency? With diagnostic, or lab, data and Artificial
Intelligence (AI) techniques.
A lab result is the
determining factor in about 70% of prescribing decisions. Used on its own or in
combination with claims and prescription records, lab data allows for a
de-identified patient profile to be created and a patient journey to be mapped
out. Clinical lab data indicates – in real time, as soon as the lab test
results are ready – how a patient is doing, if he or she needs a specific
therapy or is ready for a change in therapy.
This is the first
time ever that Life Sciences can know that Dr. Smith currently has a patient
who is a great fit for therapy A. Not only the physician and timing are right
for the visit, the rep also knows why his drug in particular will benefit this
patient the most. Now, that’s a step up from guesswork in terms of equipping
your sales force with the tools to make them more efficient!
The addition of AI
into the integrated data mix takes the predictive ability of Life Sciences from
here-and-now into the future, maximizing efficiency even further. The ability
to predict the diagnosis before it is made and identify therapy that will
address the condition best based on the patient’s clinical profile has
significant business implications not only tactically, but also strategically.
If a company can take
a large patient population and stratify it by urgency, or likelihood, of
requiring the therapy within three, six or nine months, it can both plan and
act in a highly targeted, efficient way. Importantly, when reps know the
predictive trigger, such as a test or another health event, they can better
educate providers on therapy benefits as well as provide patients with support
materials.
A needle in the
haystack
To illustrate the
importance of the AI approach, let us take any chronic progressive disease that
can be typically managed with a standard-of-care therapy. In that population,
there is a small sub-segment who fails the standard-of-care therapy and will
have to be put on a biologic. There are many patients who clinically look like
they may require a biologic but only very few of them will actually get to that
point.
Let’s take a
hypothetical example where the prescriber target list numbers 50,000. If a
biologic manufacturer uses a historical data-based method, it will be faced
with a very large number of prescribers with patients who might be
candidates for its therapy. In other words, if you have 100,000 candidates for
a biologic within the next 1-2 months, you will need to visit about 50,000
prescribers if you were to speak to all of them. However, if only about 1,000
patients will actually need the therapy, then going through the entire list of
physicians is like looking for a needle in the haystack. If you know which
patients will need the therapy, you will only need to visit
their physicians, which brings the target number to, say, 500. Much more
manageable, efficient and likely profitable.
The right information
makes an organization holistically more efficient. Lab data and AI are
demonstrating that they can be just the type of disruptive innovation that will
change how Life Sciences operate, making them more nimble, efficient and
responsive to the market needs.
Lisa Kerber
https://www.linkedin.com/pulse/how-ai-can-help-life-sciences-find-needle-haystack-lisa-kerber/?trk=eml-email_feed_ecosystem_digest_01-recommended_articles-13-Unknown&midToken=AQGIPog6-r4O6Q&fromEmail=fromEmail&ut=266BHcLsT-Oo01
No comments:
Post a Comment