Big Data Helps Provide Insight Into Cancer

Collecting, processing and analyzing medical data—and putting it to work to fight disease—is a daunting task.

Different IT platforms, hospital coding systems and software tools introduce an array of obstacles and roadblocks that undermine insight into the ways to achieve the best possible treatment methods and outcomes—especially in different settings and places. Of course, when the topic turns to a life-threatening disease such as cancer, the complexity and stakes rise significantly.

One company tackling this challenge is CancerLinQ, a wholly owned nonprofit subsidiary of the American Society of Clinical Oncology (ASCO). The organization accesses, shares and codes rich clinical data to help physicians measure and improve care for millions of patients worldwide. Simply put: it draws on data and records from oncology practices located throughout the United States to support its CancerLinQ system, which uses big data to provide deep insights and, ultimately, improve cancer care.

The company claims that by tapping into information from only 3 percent of the cancer patient population, it can deliver answers about the other 97 percent. But that requires pulling data from public health sources, electronic health records, data warehouses, lab systems, practice management platforms and more, says Rich Ross, chief solutions officer for CancerLinQ. It also requires standardizing data and using the right systems to conduct the analysis.

“We are attempting to fill a previous hole in knowledge and rapidly improve the quality of care for people with cancer,” Ross reports.

Extracting Data From a Source System

The company turned to data integration solutions provider Jitterbit to introduce a robust and agile cloud-based platform that ties into an analytics system built on SAP HANA for Healthcare. “The Jitterbit solution helps us extract data from a source system behind the firewall of a practice or office and then remotely map the data so we can transform it into a consumable format for CancerLinQ,” Ross explains. Data travels to the company’s platform via a Secure File Transfer Protocol (SFTP) connection.

CancerLinQ began implementing the Jitterbit solution in January 2015, and it is now fully operational.

The initiative faced a number of challenges, including bridging variations in the way electronic health records and reports are stored at the practitioner offices. The company had to build a knowledge library—essentially a common set of tools, procedures and best practices for managing the process.

It also had to build glossaries and descriptions so that it could map all the data efficiently and without error. That led to a meta-glossary that has essential rules built into it.

The company is now embarking on its mission of transforming all the data into actionable information. Ross says that the platform will deliver deeper insights into how doctors and other health care professionals can provide personalized treatment “by essentially matching each patient’s care with quality standards and with other patients like them.”

The end goal, he adds, is to bring together structured and unstructured data—including doctors’ notes and pathology reports—to draw new and deeper insights. “We’re hoping to facilitate the automated collection of data so that it is possible to share knowledge, transform patient care and deliver better outcomes,” Ross concludes.