How AI is Driving Patient Financial Experience Excellence

Healthcare is ripe for innovation and disruption through the use of artificial intelligence (AI) and machine learning, and revenue cycle management might be one of the strongest use cases for this type of technology.

Revenue cycle management is a complex and oftentimes manual process. But even when provider organizations tap technology to streamline their revenue cycles, staff are generally hindered by fragmented systems and a lack of enterprise solutions, according to a recent KLAS and Center for Connected Medicine survey of health system leaders.

But revenue cycle management was one of the most cited responses when asked about areas for planned AI use, second only to virtual assistant applications.

Healthcare organizations have attempted to implement AI to streamline key revenue cycle tasks.

“Integrating artificial intelligence into an RCM solution allows us to have data driven approaches to automate each step in the process, whether it's when that first letter is going out, that first contact is happening, or that payment is going into a payment plan. Each step, there's a model supporting that decision which helps a person to take the right action at the right time,” says Greg Allen, senior director of data science and product management at Ontario Systems.

The technology has proven to work with revenue cycle tasks, such as prior authorizations, claim status verifications, and claims denial management, to name a few. However, the use of this technology in revenue cycle management is still maturing. Strong use cases now include innovative AI solutions that optimize accounts receivable management by identifying and directing account follow-up, finding patterns that maximize success. But the potential for AI in revenue cycle management is rapidly expanding.

Perhaps the strongest case for AI in revenue cycle management being used today is the capability to understand people to optimize collections in the era of high deductibles.

AI for patient financial responsibility

Patient financial responsibility is at an all-time high. Kaiser Family Foundation recently reported a 111 percent increase in the burden of deductibles across all covered workers. Meanwhile, over half of patients responding to a 2020 HealthCareInsider.com survey said they worry about out-of-pocket healthcare costs leading to household bankruptcy.

The growing burden of patient financial responsibility is having a major impact on healthcare, even preventing people from accessing the care they need. Revenue cycle leaders are also finding that they do not have the tools to address this growing burden, which is also slowing the entire collections process.

AI technology is poised to help providers accelerate patient collections while ensuring a positive experience for patients.

“Using data to understand people is a very strong use case because when you can understand a person and their financial position, you can help them and work with them all while providing a positive experience when they are in a difficult position,” says Allen.

“Maybe it's to understand when you should be contacting someone or what kind of individualized payment plan they should be provided based on their unique situation.”

But creating consumer profiles to optimize the patient financial experience is very difficult at scale, Allen adds.

“That's why we’ve leveraged technology to be able to do assist,” Allen says.

AI technology can create and analyze personalized consumer profiles at scale—a capability many organizations do not have because of resource restraints. By pulling data from disparate data sources, including third-party sources, the technology can identify patterns of behavior for every patient and direct staff to create a personalized financial experience for them, whether it be calling them in the morning because they work nights, emailing them a secure link to a digital statement because they prefer electronic communications, or even not contacting them at all because they frequently use the patient portal on their own to pay medical bills.

“From a data science perspective, if I can't understand my consumer and closely replicate the world around them, then I can't succeed,” adds Allen.

AI has the potential to increase revenue through collections while improving the patient experience and aligning it more with the customized billing solutions available across other industries, such as banking and retail. These patient experience improvements also carry their own financial benefits, if only provider organizations overcome privacy and security concerns.

Misunderstanding, trepidation, and concern about patient privacy have been barriers to AI implementation in the data-heavy world of healthcare. But they don’t have to be anymore.

Overcoming the challenges of AI implementation

Three-quarters of healthcare insiders are worried about the data security and privacy aspects associated with implementation of AI and machine learning tools, a 2020 KPMG survey shows. However, 86 percent of respondents to the survey said they are taking steps to protect patient privacy as they implement AI. Furthermore, the technology is evolving from “black box” technologies, which lack the ability to understand how decisions are made, into transparent and explainable solutions, providing deeper insight and accountability for the revenue cycle.

“We practice something called ‘explainable AI’ or ‘xAI’ to create auditability into our models and into the decision-making process so that we can reproduce our results and provide feedback, along with a clear explanation for why a decision was made,” Allen says.

“That is incredibly important in order to create widespread adoption across this industry, and others, as well, because we can't have a black box solution or hide behind the veil of proprietary technology. Being able to explain what we're doing in a digestible way is another way we combat the existing lack of transparency.”

Building customer profiles for optimized patient collections and experience carries its own risk since many solutions pull from credit bureaus and other third-party data sources containing personally identifying information (PII). This PII significantly increases the risk of a data breach, putting both providers and patients at risk of a costly security event.

AI technologies can pull from alternative data sources, reducing the risk of data misuse and expanding the methods in determining a patient’s propensity to pay, preferred payment methods, and more.

Future of AI in revenue cycle management

With explainable AI, technology has the potential to foster patient relationships, not take away from the human experience in healthcare. AI empowers human decision-making, enabling technology to take over redundant tasks so staff can get to know patients and their preferences. And revenue cycle management is ripe for this type of innovation.

“By having people spend less time on an individual account, by providing more information and insight in order to be more productive, we are able to save costs,” Allen states. “And it doesn't need to be through the reduction of personnel.”

In addition to privacy and security, many stakeholders fear AI will replace humans in the workplace. However, the technology is actually helping revenue cycle staff work more efficiently by supporting their workflows and allowing them to tackle accounts needing more human intervention.

In a world of falling margins and high-deductible health plans, increasing efficiency in revenue cycle management is paramount. And AI can be leveraged to drive business intelligence to communicate more effectively with patients.

“By incorporating AI into the backend financial aspect of the patient experience, providers have the opportunity to create a better business for themselves while simultaneously improving the patient experience, allowing all parties to benefit without losing sight of their human sides,” Allen states.