OR WAIT null SECS
Lots of vision is discussed — but little real-world investment.
Artificial intelligence (AI) is the latest buzzword concept in healthcare, with one report predicting AI could save the healthcare industry $150 billion by 2026 through clinical innovations alone.
But when it comes to using AI to change healthcare revenue cycle processes, the industry is experiencing a “white couch phenomenon”: lots of vision being talked about on sofas on stages at conferences, but little real-world investment.
That’s not to say AI isn’t making a difference in the revenue cycle. At Northside Hospital in Atlanta, revenue cycle leaders leverage machine learning to predict payment for a majority of claims with a high degree of accuracy. It’s a capability that empowers leaders to determine average days to pay and estimate cash flow more effectively.
But most hospitals and health systems still have significant opportunities to hone their use of analytics before supercharging their revenue cycle capabilities with AI innovation. They also continue to face challenges related to IT security. Healthcare organizations are still considered an easy target for data breaches, in spite of investment in digital records, with 176.4 million patient records compromised between 2010 and 2017.
How can healthcare revenue cycle leaders develop the right foundation for AI in revenue cycle? Here are three things to consider.
Strengthen root cause analysis of claim rejections. One of the newest innovations in revenue cycle is the use of machine learning to predict which claims will be denied before the claim is even submitted to the payer. This approach works by:
However, careful analysis of existing data can also help stop denials before they start by zeroing in on lapses in processes that leave organizations vulnerable to automatic rejection of a claim. These include processes in patient registration and billing.
For example, it’s not uncommon for a claim to be rejected simply because the patient’s name isn’t on the claim. How does this happen? Usually, it’s because staff who are in charge of claims submission are so overloaded, they don’t take the time to proofread the claim before sending.
Data entry errors at patient registration are a primary culprit for automatic rejection, like a patient identification number that is incorrect by one number or inaccurately inputting the patient’s name, birth date, address or ZIP code. While is it easy to address these errors, they strangle cash flow by delaying payment, often by weeks.
An analysis of the organization’s most common reasons for denial will point to opportunities to shore up processes that impact the revenue cycle at both the front end, when scheduling patients for service, and the back end, when submitting claims.
Consider that registration and eligibility errors account for 23.9% of denials. Investing in software that verifies eligibility and patient demographic information in real time at the point of registration significantly reduces the risk of denials. When data entry errors have occurred, this software prompts staff for corrections before the registration process is complete. It’s a solution that not only helps protect cash flow, but also supports a positive patient financial experience by avoiding situations where a patient must call the billing office after the appointment to help correct vital information.
Meanwhile, performing a second eligibility check just before submitting a claim for processing helps avoid situations where the patient’s insurance coverage has changed from the time of scheduling to the point of claims submission—another common cause of denials.
Use data analysis to prioritize patient collections. Propensity-to-pay scoring uses predictive analytics to gauge the likelihood that patients will pay their accounts once insurance has been applied. This tool uses an algorithm to determine not only whether patients have the financial means to pay their out-of-pocket costs, but also whether they will pay, taking into account factors such as income, payment history, medical transaction history (e.g., What is the patient’s past payment behavior for your facility?), and more. Performing this analysis at the point of registration empowers staff to:
Propensity-to-pay scoring also enables healthcare revenue cycle teams to more tightly focus collection efforts for maximum return. For example, if a patient’s score indicates a high likelihood of timely payment, why invest time and money in pursuing the balance due? Instead, staff should direct follow-up efforts toward patients with median-to-low scores. Similarly, when a patient is highly unlikely to pay, revenue cycle leaders may wish to write off the amount due rather than pay the expense of outsourced collection.
Propensity-to-pay solutions aren’t new. One survey found just 14% of healthcare organizations use advanced modeling tools to predict propensity to pay. Such solutions provide an important opportunity for healthcare organizations to use data analysis to transform their collections approach.
Develop highly visual performance dashboards to drill down into performance. The use of highly visual dashboards to share progress against key performance indicators (KPIs)—such as point-of-service cash collections, clean claim rate, and denial write-offs as a percentage of net patient service revenue—can facilitate data storytelling that engages staff in making needed improvements. The more informed and engaged team members are, the better positioned they will be to:
The ideal dashboard will enable users to drill down to varying levels of detail according to their needs, from executive summaries to snapshot views of day-to-day operations to root cause analysis at the charge and line level.
The technological advancements taking place in healthcare revenue cycle are exciting, but leaders need a firm analytic foundation before investing in shiny new tools like AI. Strengthening analytic capabilities is a solid step toward elevating performance while preparing for the new tools and approaches that will define revenue cycle departments of the future.
About the Author:
Eric joined the SSI Group (SSI) as the chief technology officer to lead SSI’s long-term technology vision. He brings nearly 30 years of experience in the software industry with the last 10 in healthcare technology. Prior to joining SSI, he served as the chief technology officer at Nextech and Surgical Information Systems (SIS), where he focused on SaaS, on-premise EMR and practice management solutions as well as inpatient and ambulatory surgery providers from large hospital networks to surgery centers. Eric graduated from The University of Toledo with a bachelor’s degree in Business Administration and Computer Systems.
Get the best insights in digital health directly to your inbox.