CancerLinQ: Big Data for Quality Benchmarking




We are well on our way to creating the most robust source of oncology real-world evidence ever assembled for discovery and comparative-effectiveness research.
— Robert S. Miller, MD

At this year’s ASCO Quality Care Symposium, CancerLinQ’s Vice President and Medical Director, Robert S. Miller, MD, shed light on CancerLinQ’s current and future value in the oncology community.1 Dr. Miller opened by explaining to the audience that CancerLinQ is an instrument for quality improvement, and it is live now in oncology care practices around the country. He described CancerLinQ as a tool that is growing in size and complexity, which visualizes data more deeply than an electronic health record (EHR), creating new cohorts of patients by which to explore trends and data.

“We are well on our way to creating the most robust source of oncology real-world evidence ever assembled for discovery and comparative-effectiveness research,” said Dr. Miller.

Keeping Information Fresh

Dr. Miller gave a brief description of CancerLinQ’s software, which is connected at the backend of the electronic health records system. The clinical data flow into the CancerLinQ cloud via daily feeds, with incremental updates to keep the information fresh. He stressed that before the aggregated data can be viewed, they are de-identified for patient protection. Additionally, patient-level data are processed through the CancerLinQ quality measures platform to allow individual practices to see their performance on standardized quality metrics and compare it to that of all participating practices. CancerLinQ has been live for a little more than a year, during which 80 practices and cancer centers have signed business arrangements with ASCO’s Web-based tool.

Implementation Process

Dr. Miller then summed up CancerLinQ’s implementation philosophy: having minimal financial and technical impact on all of the challenges faced by today’s oncology practices. It takes about 4 to 6 months to onboard a practice into CancerLinQ, plus an additional 3 steps that add more value to the data collection. The implementation process is fully guided by ASCO support systems.

According to Dr. Miller, CancerLinQ also began assembling a coalition of other sister societies and organizations in the broader oncology community, creating a richer environment of shared value.

Merit-Based Incentive Payment System

Given the growing emphasis on quality measurement and reporting, CancerLinQ seeks to become a vital component in the ongoing efforts to add more value to oncology care. Central to electronic quality reporting is the Merit-Based Incentive Payment System (MIPS). Dr. Miller noted that the Merit-Based Incentive Payment System is a comprehensive set of measurers, but he distilled the challenge of this electronic quality reporting: “A practice must report on at least 50% of patients eligible for each measure and have a 20-case minimum. You cannot do this manually, so you need electronic solutions such as CancerLinQ. And another point that is sometimes missed is that it is not just Medicare. It’s all payers.”

Additional Roles

Dr. Miller explained that CancerLinQ is positioned to be an extension of ASCO’s established Quality Oncology Practice Initiative (QOPI), mentioning that the main difference between QOPI and CancerLinQ is that QOPI is a purely manual program, whereas CancerLinQ is fully automated, firing against the electronic health-care data. “However, CancerLinQ gathers all charts and data, and in that way, I believe it’s more representative of the care being rendered, more in the spirit of what Centers for Medicare & Medicaid Services has done in terms of its Quality Payment Program,” he explained.

Up to this point, Dr. Miller discussed CancerLinQ as a tool for quality improvement and care for subscribing oncology practices. However, he said that CancerLinQ is now beginning to develop its database for hypothesis-based research for third-party access. To that end, Dr. Miller noted the launch of CancerLinQ Discovery, an extension of the analytic database that is being developed in partnership with AstraZeneca. Future collaborations with government agencies, medical specialty societies, and additional life sciences companies are also underway.

Dr. Miller briefly described CancerLinQ Discovery as a nimble tool that takes key structured data elements and curates them to ensure they exist in established form. Moreover, CancerLinQ Discovery will use natural language processing and manual processes to extract additional data and provide them in structured form. Dr. Miller admitted that natural language processing isn’t ready for prime time.

“But in our context, we’re using the technology to surface elements within clinical notes, path reports, and imaging reports, and we are focusing on certain priority domains. And we are doing this by embedding human curators. I hope when I present on CancerLinQ next year, the natural language processing data-entry issue will be solved,” said Dr. Miller.

Third-Party Access

According to Dr. Miller, third-party access was an issue that was considered at length due to its privacy concerns over patient data. Several rock-solid privacy safeguards were installed; for instance, the CancerLinQ Research and Publications Committee must approve all queries, and clinical data never pass out of CancerLinQ control.

After discussing other system safeguards, Dr. Miller explained the real-world evidence, which is an intrinsic component of CancerLinQ being embraced by such agencies as the U.S. Food and Drug Administration, which will be using real-world evidence in certain regulatory decisions. He continued by enumerating the real-world evidence opportunities that will help CancerLinQ transform the way clinical research is performed.

Dr. Miller concluded his presentation by reminding the audience that CancerLinQ is a tool for quality assurance and compliance issues, one that helps oncologists visualize clinical data in new and exciting ways.

“Soon CancerLinQ will be a tool for oncologic research and discovery. Our Research & Publications Committee is working through the details, and we expect to be seeing your research proposals filing in during the third quarter of this year,” said Dr. Miller. ■

Disclosure: Dr. Miller reported no potential conflicts of interest.

Reference

1. Miller RS: CancerLinQ: Big data for quality benchmarking and point of care insights. 2017 Quality Care Symposium. Presented March 3, 2017.



Advertisement

Advertisement



Advertisement

click me