Data-Driven Performance
By Dan Harvey
Radiology Today
Vol. 20 No. 10 P. 18

Data analytics leads to better business decisions, and also improves patient care and satisfaction.

According to many people in radiology, data are an “empowering force.” But to harness that power, tools need to be deployed to make the data actionable. Actionability involves elements such as accessibility, transparency, comprehensibility, and controllability—ie, management. Data analytics is proving to be the catalyst.

“Analytic insights allow administrators to make better strategic business decisions,” says Darren Selsky, Hologic’s senior director and franchise leader of global marketing. “If one health care system or facility appears to be underperforming, scheduling adjustments must be made to improve operational efficiency.”

Furthermore, as Selsky indicates, garnered insights can inform future purchasing decisions, which can justify the need for additional equipment, workflow tweaks, and issues related to overall capacity. Data can also reveal why some enterprises are underperforming while others are overperforming.

Some of what is now possible may suggest that data analytics in radiology is a new trend. That’s a misconception. Indeed, analytics as applied to radiology is a decades-long endeavor. Rather, what is happening now is that companies are improving their data analytic capabilities, to the benefit of the health care providers. This effort is a summons.

“Currently, facilities are increasingly challenged to operate more efficiently while still delivering the best care possible, even if their resources are limited,” Selsky observes.

A significant challenge, says Chris Meenan, general manager of performance solutions with Philips, is better access to the most appropriate data. “Such data provides the end-to-end perspective,” he says. “This involves placement of pieces of information within departmental infrastructure. Get it into one place and eliminate the ‘noise’ to get a clearer picture.” The process is like a high-tech version of assembling a jigsaw puzzle.

Another challenge is the age of the data. Information from as recently as yesterday is already one day too old. It’s retrospective. What’s necessary, Meenan says, are “real-time” data. Make changes today, he emphasizes.

Vendors understand these challenges, which is why companies such as Hologic, Philips, Life Image, and many others are upgrading their offerings and even fostering industry affiliations.

A New Partnership
Here’s an example: Massachusetts-based Life Image is committed to enabling real-time flow of clinical information to engender informed medical decisions. To push its purpose forward, the company formed a partnership with Canada-based Bialogics Analytics to make medical imaging analytics more efficient.

Life Image developed an “ecosystem” that connects hospitals, physicians, patients, pharmaceuticals, medical devices, telehealth, and EHRs. The interoperable network orchestrates the flow of all clinical information with medical images, creating access to the right information at the right times—and in the right context for optimal patient care.

The new partnership means that provider organizations are now better able to perform comprehensive operational analysis, including elements such as image management and improved procedural and workflow control. It’s a complex undertaking, as it requires the combined leverage of the extant Life Image global network with Bialogics’ business intelligence analytics. The shared goal is to streamline medical imaging analytics to foster enhanced operational capability, reduced health care costs, better patient satisfaction, and, above all, better patient care.

The partnership seemed natural, as Bialogics, like Life Image, seeks to eliminate data silos and “democratize data” while remaining “vendor agnostic.” Bialogics’ analytical platform is a vendor-agnostic analytical platform for medical imaging that integrates with existing enterprise systems. Matthew A. Michela, president and CEO of Life Image, says a partnership was investigated in 2018 and became formalized in early 2019.

“We wanted to help health care clients, especially those involved in radiology, deal with business challenges such as cost reduction, more efficient of use of staff, and equipment resources to foster better patient satisfaction and, more importantly, improved care,” Michela says.

This meant studying what clients do on a daily basis—including image management, procedure management, and workflow—to help them become more productive. “The partnership enables us to reach into an organization and identify existing information across imaging machines, PACS, [and] the variety of hardware and software technology,” Michela says.

Analyzed data included basic items such as scheduling, imaging technology usage, and how personnel spend their time. These data are not normally captured, Michela says. “It’s not what you find in typical EHR systems or financial reports. So, hour by hour, with real-time data, usage of all resources is analyzed to help enable an organization to properly decide best usage.”

Michela explains the kinds of problems this approach can solve with this example: If an organization has three advanced and large MRIs, how is this technology being used and when is it not being used? How long do patients wait? Is there enough staff to work the technology? Resolving these challenges can increase equipment utilization by as much as 25%, reduce patient wait times, and lead to more efficient allocation of technical staff. “Everyone benefits,” he says.

“It’s a combination of the images and a platform to extract underlying data,” Michela explains. Bialogics provides a technology that extracts appropriate data. Essentially, the intellectual and analytic properties make information “more findable and discrete,” he says.

Further, data are rationalized and, subsequently, normalized. Life Image provides the networking and necessary integration. “This represents an opportunity to deal with daily barriers, not just within the radiology department but for everyone involved in the entire process, such as departmental directors and the financial management people, even the hiring people who determine the necessary amount of human resources,” Michela says. “Common sense becomes even more clear through objective measurement.”

Although these are bold claims, Michela concedes that the effects may not be earth shaking—even if the boasts are accomplished. “It’s not going to have the kind of impact of a new imaging technology or a blockbuster new drug,” he says. “It’s a bit more subtle, but improvement in focus areas can have substantial impact.”

Where Next?
The company’s first phase involved formalizing the partnership, educating the marketplace, and putting the approach in place. The next phase involves acceleration of adoption. In the meantime, the following two terms need to be understood:

• Vendor agnostic. “Health care technology, especially in radiology, has been narrowly focused on clinical outcomes via better technology and more precise treatment, but what happens is that technology such as PACS, cloud systems, etc, is not necessarily interoperable with other pieces of the ecosystem,” Michela says.

To use a non–health care metaphor, several years ago, if you created a Word document on a Microsoft computer, you couldn’t see it on an Apple computer. The technologies couldn’t talk to one another. “We’ve worked on creating the interoperability,” he says.

“Vendor agnostic” simply means that the end goal for clinicians is more discernable and accessible medical information for diagnosis and treatment, regardless of whose technology is used. “No clinician should have to worry about who is the equipment manufacturer or what is the hardware or software. That’s our issue,” Michela says. His message is clear: His company takes on that responsibility.

• Democratizing data. This concept involves easier access to data, including the ability to transfer them in such a way that fits the puzzle together, enabling extraction of clinical information that is available to every end user.

Implementation does not require complex interfaces or integrations, saving time and resources with an immediately perceivable return on investment. Frustration levels can be greatly reduced. Users can perform analysis from their desktop without the need to make a formal request and then wait days, weeks, or months for IT-driven reports.

Focus on Breast Health
In winter 2019, Massachusetts-based Hologic launched Unifi Analytics. This web-based business intelligence tool enables breast health facility professionals to better manage technology usage, monitor technologist deployment, make more informed business decisions, and forestall unanticipated downtime with predictive X-ray tube replacement technology—tubes are replaced before failures occur.

Previously, tube failure on a gantry could result in a three-day loss of functionality and the need for patient rescheduling, Selsky points out. This is problematic from a financial standpoint. Some patients who are forced to reschedule may decide to go to another facility, he says. Typically, family members follow. As such, a single tube failure could have a somewhat substantial negative impact that translates into revenue loss. Unifi Analytics is proactive.

“An X-ray tube replacement could cause multiple days of unanticipated downtime and subsequent loss in revenue, in addition to the patient experience of dealing with canceled appointments,” Selsky says. “Facilities can predict when a tube will fail.”

Hologic also educates its clients about the value of analytics. “Insights come from within their organization, which immediately reveal opportunities to optimize productivity, improve workflow, and provide better patient care,” Selsky says.

In addition, Hologic helps facilities compare their own performance—in terms of technology efficiency and technologist accuracy—with similar facilities. “This provides a previously nonexistent industry benchmark,” Selsky says.

Many health care systems include multiple satellite facilities. Their administrators have lacked access to data on how the entire health system is performing. “Unifi Analytics enables them to identify inconsistencies across the health system and rectify these in order to maintain consistent patient experience and facility performance,” Selsky says.

Bridge to a New Approach
In 2017, Royal Philips acquired Analytical Informatics, Inc, a Maryland-based company, to bolster its PerformanceBridge portfolio. Philips sought to help clients increase assets, reduce downtime, improve utilization of resources, and improve radiology department practices.

PerformanceBridge is an encompassing term for a changeable portfolio designed to enhance operational performance. A specific offering is PerformanceBridge Practice, designed to help imaging departments simplify unification of data and reveal actionable information. The endgame, according to participants, is to provide operational and business support to enterprises, particularly those focused on radiology.

PerformanceBridge Practice, Meenan says, provides customers an end-to-end view by pulling data from disparate information sources. This enables clients to adhere to the emerging standard for workflow analytics that places information into a consistent framework. This “bridge” is about four years old, but it originally included only modality data. The upgrade covers the entire process including scheduling, acquisition, informatics integration, results, and patient tracking follow-up recommendations, Meenan says. “It’s about all elements within a facility’s daily operation.”

The software provides a real-time perspective about what is happening. Information is less historical and more proactive. That’s why Meenan feels the upgrade will effect significant change in the radiology field. “It could change the way we think about running our businesses,” he says. “Think about the way that certain apps have changed daily life, such as Google Maps, which enables people to get to where they want to go much faster. It’s much the same thing, as it changes the ability to more quickly react, as far as improved workflow and decision making.”

The route to improvement may become more clearly delineated, as users will see patterns emerge. Such pattern recognition may help them eliminate problems before they’re operationally problematic. Areas of opportunity can be revealed that will help users circumvent once-existing challenges, Meenan says.

“Further, this kind of technology enables users to ask, ‘What’s the best that could happen?’” he adds. “This means that we move away from the reactive approach to the more predictive and proactive focus. This kind of technology is changing the way people have been working.”

Adding AI
There are myriad potential applications for analytics in the health care space. “At Hologic, we’re continuously exploring opportunities to leverage AI and data analytics tools with our products to help our customers make the most of their investments and make better decisions,” Selsky says. “It’s as much about providing better care as it is bringing real efficiencies and cost savings to customers as health care costs continue to rise.”

Analytics is connected to context, and AI combined with analytic/business intelligence software can provide the proper context, he says. “Health care data can be vast,” Selsky says. “AI and analytic/business intelligence software can sift through the proverbial mounds of data to find patterns and translate those patterns into actionable insights that transform the way we deliver health care.”

— Dan Harvey is a freelance writer based in Wilmington, Delaware.