HEOR! HEOR! More Data!

As if the volume, variety, and velocity of pharmaceutical data aren’t already flowing at a head-spinning rate, health economics and outcomes research (HEOR) data is adding another layer of complexity - as well as opportunity. 

HEOR groups are finding great value in Big Data. They are mining medical records, clinical trial data, insurance claims information, and more to lend new and deeper insights into the health outcomes of both drugs being developed and those on the market. The growth of IT and analytics tools that allow for more robust data analysis is helping fuel increased use of HEOR data.

HEOR Defined
HEOR has been defined as a scientific discipline that quantifies the economic and clinical outcomes of medical technology. It is used to complement traditional clinical development information (i.e., efficacy, safety, quality) in guiding patient access to drugs. HEOR helps payers, patients, and pharmaceutical companies understand the value of drugs and medical devices. 

The Wild West of HEOR Data 
As the ROI on drug development overall stalls, it’s no longer enough to collect data that proves the efficacy and safety of therapies. Pharmaceutical & biopharmaceutical companies need to prove the economic viability of drug products as well. This puts greater pressure on organizations already overwhelmed with data.

The available HEOR data – both proprietary and public, structured and unstructured – needs to be wrangled for more timely, manageable and useful consumption. This includes more recent contributions to big data in pharma – from social media posts and online patient forums to wearable monitoring device outputs – which have emerged as a substantial source of information on a wide range of medical conditions, treatments, and costs.

Consider: More than 100 million people are sharing health information over the Internet:
in public patient- or disease-focused forums such as patientslikeme.com or curetogether.com
through pharmaceutical or insurance companies’ contact centers
with healthcare providers via email or web portal
through online surveys, among others.

Although the primary reasons for this social sharing have been information, education, and support for patients and their families, a secondary benefit has been the value such information can have on drug and medical device development. Patientslikeme.com has 31 million data points on disease, covering more than 2,500 conditions. The forum says the real-world health experiences of its 400,000+ members helps organizations that focus on its members’ conditions. 

Much of this data on medical conditions, treatments and patient experience is information that isn’t captured by traditional patient-reported outcomes (PRO) instruments. It may be self-reported, or it could be interview responses from patients that are not amended or interpreted by clinicians or others. 

Finding Value in Big Data
Big data approaches can improve clinical and drug decisions by:
Strengthening HEOR
Targeting drugs at specific patient populations
Accelerating drug development
Improving clinical study patient recruitment and retention

Doing so, however, requires actively listening to the available data, integration of unstructured with structured data, and seizing value by operationalizing data.

In one empirical study of clinical trials eligibility criteria for chronic lymphocytic leukemia (CLL) and prostate cancer, unstructured data was “essential to solving” 59% of CLL and 77% of prostate cancer trial criteria. The study found that “structured data alone is insufficient in resolving eligibility criteria for recruiting patients into clinical trials” for the two diseases.

But how can you easily and rapidly source relevant data from disparate databases? What if, as in HEOR, the data sets are wildly different, from highly-structured lab values and medication lists to unstructured data captured in physician notes and online patient forums

Empowering Scientific Decision-Making
What empowers scientists working in pharma or CROs today, who must understand the implicit relationship between data sources across the expanding data universe? 

Scientists and business analysts who have a unified view – including structured, semi-structured, and unstructured information – dramatically reduce their time to insight and improve the quality of their decisions and actions. Arming them with a self-service data source discovery portal helps to quickly identify and link relevant project-based data sets across disparate systems. This moves you faster to better insights, more quickly to better decisions, more effectively to market.

Are you equipped to leverage big data and gain a competitive edge by mining the depths of HEOR data? If so, hear, hear! If not, see here.