The patient financial services (PFS) department continues to play an essential role in helping health care organizations deliver a positive patient experience, manage business processes and sustain growth. The move to accountable care organizations (ACOs) and other value-based models will not diminish the need for PFS, but it does place more visibility on the department to operate as efficiently as possible.
Intelligence, automation, machine learning and big data are buzzwords we hear all the time now in health care. That’s because these concepts will have a powerful impact in how care is delivered over the next few years. For starters, in patient accounts management, they will automate traditionally manual work that will free up time for staff to focus on more complex issues or opportunities that will impact the bottom line.
Artificial intelligence, machine learning and patient financial services
Artificial intelligence (AI) and machine learning are terms that are used broadly in in a variety of industries, and increasingly in health care and patient financial services. AI is the broadest concept, with machine learning considered to be a subset, along with robotic process automation (i.e., keystroke emulators).
Simply put, machine learning allows the system to correct itself and gradually become “smarter” over time as the technology extracts and processes new data. This means that new data can generate new actions, which opens a door of opportunity in the revenue cycle as it continues to evolve. Machine learning is typically managed by data scientists, who select a data set and apply different algorithms depending on the type and application, which then derive a possible set of actions. The data in these scenarios is typically considered “big data,” meaning that it is an extremely large set of data from various sources.
Increasing efficiency of patient financial services
Applying these concepts to PFS unveils new opportunities to increase overall staff efficiency, giving them more time to focus on the accounts that are most complex. Some examples include:
- Prioritization of work – Based on factors such as the financial impact or age of an account, machine learning can apply prioritization to determine which work items should be prioritized.
- Limit denials – Based on denial codes and the subsequent actions to address them, the system can identify patterns, flag errors or expose duplicate charges before they are billed. That enables more proactive and preventative process changes.
- Automate tasks – This level of intelligence could automate certain tasks or processes in account management that previously had to be handled by one or more resources.
The technology to fully realize the vision behind these capabilities is still in-flight across the health care industry. In the meantime, organizations should examine their current systems for gaps or efficiency gains. These adjustments typically require applying new rules to the system, and while that won’t provide the same flexibility as machine learning, it still could drive better outcomes. Here are some initial recommendations:
1. See where it makes sense to use exception-based workflows
Are there opportunities to limit how much users are reviewing accounts? Reducing the number of items to review (even by just a few) can save time and direct focused attention on more complex issues.
2. Review clinical workflows and direct applicable work upstream where appropriate
Assign edits to the originating departments, including the clinical teams, so they become aware of common issues and can potentially address or prevent them before they reach the revenue cycle. This is not to say that the revenue cycle should drive clinical processes, but that the departments should be working together.
3. Revisit team assignments for over- or under-specialization
Determine how and why team assignments are structured, whether it be by allocation of resources, skill-sets or status quo. Depending on the process, re-allocation of work and/or resources may be all that’s needed to optimize performance.
While the impact may vary, these recommendations will deepen an organization’s understanding of existing workflows and hidden gaps, and may even lead to gains in PFS process improvement. Moreover, these recommendations will make it easier for organizations to adapt new intelligence-based capabilities as they emerge.
Efficiency in PFS is still extremely important in today’s health care environment. Fortunately, opportunities for machine learning technology are gaining traction across the industry and could unlock enormous promise for health care organizations, staff and patients alike. Ultimately, they make it easier to explore the benefits of intelligence-based technology in the transition to value-based care.
Join Cerner in celebrating National Patient Account Management Week, Oct. 14-20, 2018! Share this blog on social media and tag Cerner in the post. Click here to learn more about how Cerner’s integrated billing, claims and contract management solutions enable your organization to verify expected reimbursements and manage any variances and denials right within your workflow.