Healthcare and credit-based decisioning
Saw this interesting little piece in Health Data Management today - Hospitals Should Know the Score. An interesting snippet on how important proper credit management is for hosptials - "As much as 30% of patient payments end up being written off as bad debt". How can an EDM approach help solve this? Well one of the most important measures providers can take to improve their financial condition is to ensure they are billing appropriately for the healthcare services they provide—an effort that starts at the front door. During the admissions process, rules-driven applications can perform real-time validation of patient-supplied data against medical records and external data sources. By making certain patients are who they say they are and that addresses and other information are complete and correct, providers reduce their financial risk.
Prediction models and external data sources enable an EDM solution to predict, at admission or pre-registration, what the optimal initial payment request for patients based on their particular financial situation, what the optimal settlement amount is, and how to fashion a payment plan with the highest likelihood of completion. Guidance regarding the initial payment request increases revenue upfront and improves the effectiveness of intake staff. Knowing what patients can and will eventually pay enables financial counselors to help patients who need help and minimize the number who go to collections, while also generating maximum revenue for the provider system. This is similar to a usual business approach except that you don't want to decline care because someone can't pay so much as help them pay for the care they need. Rules can also simplify and speed up admitting by guiding intake specialists through the most efficient process for each incoming patient.
Another part of the solution is to employ rules and models at the point of care to perform identity checks and likelihood to pay. Whenever a patient will incur an out-of-pocket responsibility above a pre-defined threshold (e.g., $25), an EDM solution can inform providers of an optimal strategy to maximize the likelihood of payment. The solution can also check the patient’s information against all rules governing eligibility for charity care, Medicaid, and all types of supplemental funding. Rules would be created and deployed consistently throughout the provider system to comply with both government and organizational patient payment policies, and can be monitored for compliance by staff throughout the organization. This prevents unecessary bad debt being accumulated during a patient's stay.
Once a patient is discharged, an EDM approach can also be used with overdue accounts to improve collection results and minimize recovery costs. Instead of treating all overdue accounts with the same sequence of dunning letters and calls, providers may, in fact, be able to collect more money by doing less. Analytics can identify differences between accounts that affect payment behavior—segmenting first-cycle delinquent accounts, for example, into those likely to self-correct, those likely to be influenced by collections treatments and those unlikely to pay under any circumstances. Providers can use this segmentation to save money by making fewer outbound contacts and thereby also reducing the volume of inbound inquiries such contacts generate. They can make an early referral to a collections agency of those accounts with very high nonpayment scores.
An analysis of delinquent accounts and collections work streams using an experimental design can enable a provider organization to determine which tactics are most effective for which types of accounts. An EDM solution can route accounts in collections to the work stream that will produce the most revenue in the shortest amount of time at the lowest cost.
By applying EDM to the problem of bad debt throughout the patient lifecycle, hosptials can collect more of what they are owed more quickly without having to turn away patients. Fair Isaac is developing an offering, Patient Decision Manager, to do just this and partnering with some leading hospitals to make it work.
Thanks to my colleague Larry Feinstein for help on this one.