Beyond smarts, technology leaders should be passionate about solving their organization’s complex data problems, says Bo Roff-Marsh, CTO of Veda, a Madison, Wisconsin-based company that provides data process automation tools for healthcare organizations.
Roff-Marsh spoke with StrategicCIO360 about today’s data challenges and opportunities, the struggle to find tech talent and the limits of big data.
What technology problems are you passionate about solving?
I’m passionate about solving complex data problems and developing solutions that are scalable, especially in healthcare. It’s no secret that most of the health technology that’s being developed today is focused on improving the patient experience. However, the industry can’t deliver on its goal of patient-centered care until the myriad problems and inefficiencies that exist within its infrastructure are addressed.
Healthcare’s data issues are longstanding and tough to solve. The data is “messy” by nature, often riddled with human errors and inconsistencies. The industry’s disparate IT systems, platform customizations and even some standardization efforts, like HL7, have also contributed to the ongoing struggle to derive valuable data insights and achieve data connectivity.
At my company, we’re applying AI and machine learning to some of the industry’s toughest data issues—like healthcare’s provider roster data processing problem. We’re helping healthcare organizations save millions of dollars and making it easier for patients to access care.
Some may not realize that this roster data issue is due to the astronomical amount of provider data that health plans are processing at any given time—manually. A shared platform between payers and providers doesn’t exist. In 2021 alone, my company’s technology, which sits between provider and payer systems and acts as a translator, saved U.S. health plans approximately 45,000 hours in administrative back-office tasks. The plans were able to deliver better patient care and a more streamlined experience for all.
What have been the biggest challenges as CTO during a time of accelerated digital transformation across industries, especially healthcare? Any lessons learned?
Over the course of the pandemic, the technology industry has faced major resource issues, especially startups. The shortage of tech talent has been one of the biggest challenges for CTOs across industries, as the demand for technology professionals has exponentially increased. The challenge of closing the technology talent gap has also gone hand-in-hand with the need to develop scalable solutions. In the past, I’ve faced difficulty finding the right talent and resources to help our business scale and commercialize.
Another challenge is the perception that government regulations prevent us from advancing technology in healthcare. However, we’ve seen that the needs of the healthcare system and today’s advanced technology’s capabilities very much align, and that it’s possible to apply advanced technology and remain compliant. AI and machine learning in particular have the potential to make a huge impact on the healthcare industry, through streamlining administration tasks, diagnosis and treatment, among other aspects of healthcare.
While working within a highly regulated industry is challenging, I’ve also found opportunity here. By putting ourselves in our clients’ shoes—they’re the ones ultimately responsible for compliance—we’ve become better able to “walk the walk” and successfully straddle both the healthcare and tech industries’ varying levels of regulation.
How will technology like automation and AI help to fix the woefully outdated infrastructure in the healthcare industry over the next few years?
There are many elements of today’s outdated healthcare infrastructure that need to be addressed. One of the largest issues—and least well known—that can benefit from AI, is around data processing and data quality as in provider roster data, claims data and the network data. The data that provider organizations send to their contracted insurance companies must be processed in a way that enables the plans to communicate provider changes to patients in a timely manner. AI is capable of making this process not only happen faster, but with much greater accuracy.
This is just one of numerous examples of how technology can help update healthcare’s infrastructure. Anywhere there are hands on keyboards dealing with data—which is a lot of places in our industry—AI has a role to play. In addition to streamlining administrative processes like provider data management, there are of course clinical applications as well.
Do you have any advice or best practices for other CTOs and IT professionals to keep in mind as they look to automate their organization’s processes?
The stage of the business must be aligned with the processes that technology leaders are implementing internally. For example, I don’t recommend trying to implement enterprise-level solutions when your startup is only in its infancy. It’s essential for a company’s strategy, financial status and product stage to all be aligned before looking to implement major process-driven technology within your organization, such as automation.
We’ve seen that automating business processes can reduce costs, streamline operations and free up talent to focus on higher-valued work. There’s no doubt it’ll continue to be a critical element for companies, but they must evaluate where their business stands before implementing automation.
What is your top prediction for big data management?
Over the last decade, “big data” has been used as an all-encompassing buzzword throughout enterprise technology. However, big data has become much more normalized across the industry, especially during the pandemic. Over time, I think the hype around big data will continue to fade. It’s also not always necessary.
We’ve seen AWS automate extensive big data collections—it’s just not that challenging to manage anymore. Most of the time, big data isn’t even needed for solving healthcare industry problems. For many healthcare data solutions, all that’s needed is the right amount of data to run machine learning processes.
All in all, I think more companies will delve deeper into solutions that don’t require big data. Ultimately, they just need enough data to fulfill an effective use case.
