While considering propensity to pay scoring is not a new idea, there are new approaches that have higher accuracy and effectiveness. The new approach uses models that leverage a patient’s transaction history and can augment it with third party data to go beyond simple credit scoring to understand the patient’s payment patterns and ability to pay.
High medical bills and high deductible insurance plans challenge many patients, causing them to ask, “Who should I pay, and how am I going to keep up?” Well, there’s good news. Recent advancements in artificial intelligence (AI) are making it easier by using transactional and behavioral interactions to provide the right payment approach at the right time.
These new techniques and models improve as they are provided with more data. They also require a solution that is good at taking multiple and often disparate data sources to create a single “source of truth.” When you obtain information from patient history along with other third-party sources, both proprietary and public, it allows the models to provide a significant improvement and predictability not found with traditional credit scores. The models typically start by leveraging two years of historical transaction data that is derived from adjudicated claims and payment information. The models then provide a score and create patient “segments” to help providers track the outcomes against their current experience or base line. This allows providers to gauge the level of improvement and ROI based on each segment of patient.
Patients are separated into three categories: will definitely pay, will definitely not pay and will maybe pay. The key is to focus on the third group and develop a payment plan that fits the patient’s needs. Depending on the patient’s score, providers can ask for the payment in advance, look for a third party sponsorship or financial assistance, or assign to charity.
Many health systems see the benefit of propensity-to-pay models but struggle to integrate it into their existing workflow. Additionally, some systems that do have propensity-to-pay models base their scoring solely off credit scores alone. While this can be helpful, it is not as accurate or predictive as desired, which is why it is essential to use the patient’s historical data for the most accurate prediction.
Health systems should be conscious of the fact that every patient is different and has varying financial situations. There is not a “one-size-fits-all” solution when it comes to propensity-to-pay payment plans. Proper segmenting and scoring can prevent premature outsourcing and unnecessary high vendor fees that lack return on investment (ROI). Additionally, using the scoring to focus resources can remove excessive contact with patients that have a high likelihood of paying. Without leveraging the insights from the models, providers end up using unnecessary time, resources and money and can negatively impact the payment itself.
While many health systems defer integrating propensity-to-pay models due to the initial challenges they may face, the benefits are measurable across revenue and patient satisfaction levels. Prioritization of the work can be focused on expected outcomes along with scoring factors to provide increased efficiency in the workflow. Also, determining a patient’s propensity-to-pay score allows providers to have an open and honest conversation with the patient about payments before the procedure. This conversation shows compassion for the patient and sets them up for payment success rather than failure. It also opens the opportunity to set up a payment plan the patient is able to follow and afford, in turn reducing bad debt. When patients can make payments on time, health systems are able to improve collections while spending less time on needless follow-ups, which saves time, money and reputation.
Health systems and providers will see improvements across their revenue cycle process as well as patient satisfaction levels by implementing propensity-to-pay models into their workflows. It is essential to leverage patient historical data as well as third party sources to determine likelihood to pay with greater accuracy.
Remember that each patient is different and avoid using a one-size-fits-all mindset when considering patient payment approaches. Leveraging a provider that has both data and modeling expertise to implement propensity-to-pay scoring will help acquire up-to-date information, assign an accurate score and ensure the patient’s individual needs are considered throughout the process. The goal of propensity to pay is to allow you to free up resources and work more efficiently to help increase patient outcomes.
If you would like to discuss any of the ideas discussed above, please feel free to contact me. I’d be happy to to show you how the benefits of propensity-to-pay modeling and scoring to help your organization.