By Christopher Draven, Manager of Automation, Blue Cross and Blue Shield of Kansas City.
When RPA entered the landscape as a fully realized capability and toolset, it wasn’t long before healthcare executives began getting excited about all the dollars they could save, especially in the realm of automated claims processing. However, this often-repeated corporate initiative (think hotkey macros, auto-adjudication configuration, and business rule engines) rarely yields the expected results.
The automation of claims processing is nothing new. Executives typically see the work as simple and repetitive. Finding ways to quickly identify a claim and get it processed sounds easy enough. We are in the era of artificial intelligence and machine learning, after all.
So why not start with claims?
Claims Processing is Complicated
Training a claims processor, even to handle straightforward claims, can take weeks. Even then, newly trained employees need time to ramp up their “production.” Understanding the process requires knowledge of accumulators, claims history review, business or client-specific exceptions, claim edits, processing limits, and how claims flow through work queues. Additionally, claim types that are easy to process are typically low-volume claims because straightforward claims are usually handled through system configuration.
RPA solutions are most successful when there are few exceptions, and where clear business rules exist. As any claims processor understands, you never know where a claim will take you.
Healthcare Processes Where RPA Delivers
Luckily, there are plenty of areas and healthcare processes where RPA can drive realized savings at a lower development cost.
Revenue Cycle Management Activities:
- Claims Filing: Repetitive and time-consuming, entering the same data over and over again into payer portals, Health Information Exchanges, or clearinghouses, is an area where providers and payers can both realize benefits.
- Patient Collections: The initial stages of most collection efforts are relatively routine – send reminders according to an established timeframe. An automated solution would generate a list of past-due accounts, queue up collection notices based on thresholds (to the point of printing the letters or sending them through to the mailroom to process), check the status of those accounts based on set timeframes, and refer/forward the account as necessary.
- Pre-registration: Getting a patient properly set up in an EHR is a manual process that impacts customer service to other waiting patients and causes the front-office to slow down unnecessarily. While first-time patients may require additional attention, a returning patient should flow through quickly. Automating the pre-registration check-in process could consist of a scheduled bot running an eligibility verification check on scheduled patients, checking to see if all necessary information is present within the patient record, and reaching out via email or text message to send reminders.
- Member Enrollment: For payers, enrolling new group members is a daily event. The use of a standardized and electronic enrollment form allows for data to flow directly into source systems. However, that may not always be an option. Paper forms can be scanned in, read by a bot with OCR, and processed accordingly.
- Provider Network Updates: Constantly updated demographics, credentialing changes, network contracting agreements, participation status, and inclusion of CMS mandated provider updates can make keeping provider directories a challenge. A bot can access the data from all of the source systems, internal and external, and compile that information into a single repository.
- Release of Information: Often called “Copy Services,” ROI companies are paid by providers to handle the regulatory burden of properly, and in a timely manner, releasing PHI to requesting parties. After working as a developer for an ROI company for a few years, I learned that only one step in the entire ROI process needs human intervention.
This process typically requires such companies to log into the client systems, ensure the request is legitimate and duly authenticated, digitally print the requested records, and then transmit the requested patient information to the requestor.
Aside from the compliance need to complete the authorization review, the remaining steps can be automated. In fact, many EHR vendors are beginning to offer this as a paid-for solution. Providers have cheaper options to ensure a properly handled Release of Information request.
- Fax-Based Processes: Faxes, in general, especially faxes that use consistent templates, are an excellent area for process improvement through RPA. Training bots to read standard documents or forms, from high-volume sources, will instantly provide increased workforce capacity.
Healthcare as a business sector is rife with fax-based processes. Look for opportunities to eliminate fax-based processes in medical management and utilization review, patient admission notices to payers, claim forms, appeals/reconsiderations, patient lab results, or even specialist consult notes.
Even automating one of these fax-based processes will drive real savings for a healthcare organization (payer or provider).
The First Rule of RPA Club
While a “big fish” project that automates 5,000 claims a day, reduces thirty FTE and saves millions of dollars sounds like an excellent investment, the first rule to successfully implementing and sustaining an RPA capability in your organization is to understand where RPA is the best solution.
RPA is most successful when the focus is on implementing a digital workforce to drive down administrative costs, increase workforce capacity, and reduce system inefficiencies. Claims processing as a component of an overall RPA strategy is promising but isn’t the best first step.
If you liked this article, please sign up to RPA Today! Registrants will receive our free weekly RPA newsletter updating you on the most recent developments in the Robotic Process Automation, Intelligent Automation and AI space. In addition to news updates, we will also provide feature articles (like this one) with a more in-depth examination of RPA issues for end users and their enterprises.