From 100 hours to 10: Tech streamlines cancer data collection, insights

Data challenges of QOPI certification

The QOPI certification process requires oncology practices to demonstrate quality performance and to participate in a site survey, in which an oncology professional visits the practice to ensure compliance with the QOPI Certification Standards.

The first part of the QOPI process typically involves abstracting patient charts, in which the practice must gather specific, de-identified clinical data to submit to ASCO. The organization offers two data abstraction rounds per year, for which practices select patient charts to report on based on chart selection criteria. The criteria include data such as dates for diagnosis, first office visit and recent office visits.

According to Arpan Patel, M.D., clinical quality officer of Wilmot Cancer Institute at the University of Rochester Medicine, doing this abstraction manually is extremely time-consuming.

"We would spend about 100 hours a year going through people's charts and everything," he shared.

That meant staff were often inundated with QOPI data abstraction in addition to their day-to-day work. And despite the hours spent on the task, the process yielded a fairly limited data sample to assess performance, Brian Shields, director of performance improvement at UCHealth, noted.

Though automating the process was an obvious answer to these challenges, Shields explained that developing homegrown tools is not feasible for most cancer centers, primarily due to time and resource constraints.

"We have a fairly small set of data analytics resources that are available to us for oncology-specific analytics, but to take on building out all of this in-house would've taken us a really long time," he said.

How the network helps meet QOPI requirements

To alleviate the burden of meeting QOPI requirements, both Wilmot Cancer Institute and UCHealth Anschutz Cancer Pavilion turned to ConcertAI's CancerLinQ network. The company acquired the real-world oncology data platform from the American Society of Clinical Oncology in 2023.

The platform brings together clinical data from EHRs across 900 cancer centers and oncology practices. It leverages AI to standardize and harmonize clinical data and provide insights for care delivery and research.

For Wilmot Cancer Institute and UCHealth, automating the data abstraction process for QOPI certification was the primary reason for joining the network and implementing the platform.

Patel noted that joining the network has cut down the time required for data abstraction from about 100 hours to 10 hours or less. This allowed the cancer center to reallocate the staff and resources to more direct patient care activities.

The platform has also helped identify cancer care gaps at UR Medicine and across its regional cancer care sites.

"As a chief quality officer for our cancer institute, I can now see across the board very, very quickly," he said. "And I think that time to data realization is really key because in the past, we manually reviewed [the data], [and] it would take half a year to see all that data and act on it. And the problem is, in quality improvement or patient care in general, if you're making choices on data that's six months old, you're already obsolete, and you're missing the boat."

Shields echoed Patel, noting that identifying patient care gaps in real time has been a valuable benefit of joining the network and gaining access to the data platform. For instance, as a result of the platform's real-time chart abstraction and data analytics, UCHealth identified advanced care planning as an area that needs greater attention.

"Just having those conversations with our patients, documentation of those desires and our patients' wishes in advance, and proactively addressing advanced planning concerns," he said.

UCHealth also identified gaps in smoking status assessment and documentation and is launching a specific quality improvement effort to address them.

Overall, the platform gives the cancer center real-time visibility into 20 care quality metrics, many of which are required for QOPI certification, Shields said.

"One of the things that was attractive to us was just the origins of CancerLink coming directly out of ASCO originally, and that tight integration with ever-evolving standards and performance measures, et cetera," he added. "And so, that was extremely advantageous from our perspective, really helping us guide our understanding of our performance against those measures."

Benchmarking benefits & clinical trial support

Chart abstraction and health system-specific data insights weren't the only advantages of joining the CancerLinQ network. Patel and Shields highlighted the platform's benchmarking benefits.

Because the platform pulls information from cancer centers nationwide, participants can benchmark themselves against national quality metrics and use the data to enhance their outcomes.

"The benchmarking piece was huge," Shields said. "If we built this in-house, we wouldn't be able to benchmark against ourselves and identify which of our facilities and regions are performing well and which ones have opportunities. But to be able to benchmark against academic medical centers across the country -- that's a big piece."

Patel agreed, adding that the network is not only enabling national benchmarking at the organizational level but also guiding clinical decision-making at the individual level. If a provider is considering new treatment options or making an unconventional adjustment, they can use the platform to see if other providers have done the same and what their outcomes were, he explained.

Additionally, the CancerLinQ platform offers other features, including a clinical trial matching capability that has proved valuable for both centers. The TriaLinQ feature automates the process of matching eligible patients with a clinical trial, helping to boost enrollment.

"We're just slowly rolling it out right now," Patel said. "But for a large cancer center that sees 16,000 patients a year, sometimes it's challenging to screen and see what clinical trials that we have available for our patients. So this specific [feature], I've been really, really excited about."

The capability will allow cancer center clinician leaders and nurse navigators to review charts across the organization and identify patients for clinical trials without having to manually review individual patient charts to determine eligibility, he added.

For cancer centers looking to enhance patient outcomes and gain and maintain key quality certifications, the devil is undoubtedly in the data.

"Healthcare is very, very fragmented, and one of the largest gaps that we're seeing across the nation is where data's captured, how to integrate it, how to display it," Patel said.

This gap can be closed, or at the very least narrowed, through technologies like the CancerLinQ network that gather disparate data and provide clinicians with the necessary insights, he added.

Anuja Vaidya has covered the healthcare industry since 2012. She currently covers healthcare IT and innovation, including artificial intelligence, digital healthcare, EHRs and interoperability.