Shannon Werb, Chief Information Officer, Virtual Radiologic
Revolutionary Technological Solutions
Deep learning is one technology that will assist clinicians in what we like to think of as “assisted diagnostics.”There are examples of similar initiatives that are focused on replacing specialists—replacing radiologists, but we don’t believe this will happen for the foreseeable future. That said, the role of the specialist will change, probably moving from the primary diagnostic source to a diagnostic source that is assisted by computerized algorithms in a sense, this is happening today when you look at the application of computer-aided detection in mammography.
“The use of NLP really becomes the foundation to help radiology move from a fee-for-service to a pay-for-performance or quality-based reimbursement model”
It is dangerous to think that radiologistswill be replaced. So, we view it as critical that vRad lead the deep learning initiative and bring the appropriate leadershipto market and help drive appropriate change.
A second technology—expanding on the idea of big data/analytics—is the use of Natural Language Processing (NLP) to extract qualitative information from diagnostic reports. In the end, the actual work product of most specialists, especially the radiologist, is the diagnostic interpretation documented in the report. Unfortunately, once the report is created, the valuable information, the actual key indicators that can drive care of the patient, get “lost” in the dictated text.In other words, it is humanly impossible to identify and extract insight from the millions of dictated, text-based radiology reports.
However, by leveraging NLP, we can extract this text-based, qualitative information cost effectively and efficiently—and integrate it into a data and analytics strategy to help drive an outcomes-based conversation about the broad need and use of diagnostic services within radiology. The use of NLP really becomes the foundation to help radiology move from a fee-for-service to a pay-for-performance or quality-based reimbursement model.
New Role of IT’s at Virtual Radiologic
The United States healthcare system recently completed moving from an aging/retired diagnostic coding system (ICD-9) to an enhanced and much expanded system based on ICD-10 codes. Not only have the number of codes increased, but the level of specificity required to justify reimbursement for any given procedure has also increased substantially. This requires clinicians to provideand appropriately document a more complete clinical history, with greater detail than ever before. To make matters more complicated, the United States is the only country that connected the use of diagnostic codes to reimbursement for services. So as if it wasn’t complex enough, the financial risk has now definitely elevated the conversation.
Any ICD-10 initiative really only partially involves technology, or the technical solutions to documenting, aggregating, capturing, and reporting clinical information to support ICD-10 coding. There weremany additional requirements, and vRad’s technology team played core and critical rolescreating, managing and delivering an ICD-10 solution that was clinically accurate, cost-effective, scalable—and user-friendly. The IT team was able to ensure a seamless transition to ICD-10 for our practice, as well as the thousands of healthcare facilities we serve.
For example, we were tapped to provide resources to organize internal stakeholders, ensure we had appropriate internal and external communications, organize training efforts, make sure we had alignment from our legal team, create confidence in our sales teams (both internal account management and our field-based teams) and, of course, create the technical solutions by modifying the workflow to support new documentation requirements.
vRad delivered on all of our ICD-10 initiatives because we remained focused on providing the best possible patient care. At the end of the day, we understood that more information is better, especially when it is structured to help improve the time and accuracy and information needed to treat patients.
Role of CIO as an Ombudsman between IT and Business
The United States healthcare environment has many federal mandates. When it comes to data privacy and security, HIPAA (Health Insurance Portability and Accountability Act of 1996) is at the center of it all. HIPAA requires that healthcare providers name a Privacy Officer and Security Officer. vRad fulfilled this requirementby selectingour compliance director (an attorney) as our Privacy Officer that sits in our Office of the General Counsel, andthe CIO as the Security Officer.
This is an appropriate balance of power and responsibility; the Privacy Officer helps interpret the regulation and builds the enforcement policies for the organization (collaborating with the Security Officer), and the Security Officer is responsible for the tactical implementation of the controls such regulation/policies required to be successful. In addition, my role as CIO is to educate our organization about risk and the importance of why we make the investments we make in security. I communicate that everyone has a role to play in security. This has been part of the culture for some time, given the nature of what vRad does—data protection and security is part and parcel of the CIO’s job in any healthcare provider, which is what we are first and foremost.
Many organizations have the CIO or CISO/CSO fill both roles; however, the relationship between these two is important to help the business understand, prioritize, and implement controls needed to protect patient privacy and data security. Many organizations have already implemented dedicated CISO/CSO positions; this is inevitable in most organizations as they posture towards security.