In the parable of the blind men and the elephant, each man describes what the elephant must look like based on their limited experience of touching different parts of the animal. This ancient tale of the Indian subcontinent, which dates to at least 500 BCE, illustrates the limitations of individual experience and suggests that by combining different perspectives and understanding the limitations of our own perceptions, we gain a more comprehensive understanding of the world around us.
Applying this logic to real world data (RWD) within the clinical ecosystem highlights the difficulty of making decisions based on limited data and understanding.
RWD is a deep and wide source of information that life sciences commercial teams can use to understand challenges in the market and the patient journey. Data from medical and pharmacy claims, electronic health records and specialty datasets provides a view of hundreds of millions of patients, pharmacies, payers, HCPs, and sites of care. When joined up, these data can create a compelling portrait of opportunities within the marketplace – and help to identify obstacles.
In reality, some key information is missing. The recent Clarivate report, The Challenges of Working with RWD, notes that because U.S. claims- and EHR-based RWD like the Clarivate real world data product capture a percentage of the claims-generating population, but not the entire population, it is essentially a large sample of the claims-generating population. Depending on how those claims are obtained, the sample may not be representative of the full patient population, and certain population segments may be over or underrepresented.
This is where advanced analytics come in, helping us to fill in the blanks and gain an accurate picture of the marketplace.
Applying advanced analytics
Several advanced analytics tools are at the disposal of data scientists and regularly used, explained Hemanth Nair, Clarivate’s Senior Director of Real World Data Analytics. Bootstrapping, for example, uses samples to draw inferences about a population. It is a resampling technique used to estimate the uncertainty or variability of a statistic or model parameter. It is particularly useful when the underlying distribution of data, such as claims data, is unknown.
Time series is another approach to supplement information in the face of missing data. It refers to a sequence of data points collected over successive time intervals. Time series data is recorded in a chronological order, typically at regular intervals, such as hourly, daily, monthly, or yearly. Analysis using this method can uncover patterns, trends, and relationships that may exist over time. This is especially useful when looking at data from the Centers for Medicare & Medicaid Services (CMS) data, Nair explained, as these have a six-month lag, making it difficult to provide timely data about disease prevalence.
Clarivate used these techniques to help a top 25 medtech company that needed monthly facility-level volumes for more than 15 cardiac procedures/devices to compare internal sales data to the total estimated use. In addition to leveraging the internal data sets that Clarivate has, Nair’s team filled in the gaps with these advanced analytics techniques and created a deployment pipeline that could both rapidly and flexibly handle data nuances.
A little bit more of the elephant
Clarivate data from medical and pharmacy claims capture key health information for over 90% of the U.S. patient population, but Nair is not cavalier about the amount of data his team has and knows that it is critical to fill in the gaps.
“As data scientists,” said Nair, “Clarivate has a lot of integrity around our data and our analytics, but no big data scientist is ever going to say that everything is clean and clear cut. If and when they do, that’s when people should start questioning it.”
The inclusion of advanced analytics enables RWD to go further than ever before, creating a complementary and robust data set that describes the market for a particular drug. Questions such as where the patients are, and who and where the prescribing physicians are, can now be answered with pinpoint accuracy.
Using advanced analytics, pharmaceutical and medical device companies gain a comprehensive picture of market dynamics and how they are likely to evolve.
“We have a big chunk of the elephant,” said Nair. “And by applying advanced statistics, we get an even better picture of the whole elephant.”
Download the full report on Challenges of working with real world data and key ways to overcome them here. Learn more about Clarivate real world data products and solutions here.
