Patient propensity to pay

Predict which patients will pay before you mail another statement

Patients now carry more of their own bills, and your patient financial services team works a growing list of self-pay balances. Most teams send every account the same statements on the same schedule.

That wastes calls on patients who would have paid anyway, and it reaches patients who need help too late. A score for each account lets your team change the order and the offer.

How patient propensity to pay works

Step 1: Bring the billing history together

Pull patient billing, payment history, account age, payer and plan details, and remits, the remittance records that show what insurance paid and what the patient owes, from the systems you already run.

Step 2: Build and compare models

H2O Driverless AI runs automatic feature engineering and ranks candidate models on held out months of your own data.

Step 3: Explain each score

Reason codes and Shapley values show which factors drove each score, so your team can check it before acting on it.

Step 4: Set the next step

Your team maps score bands to actions: no outreach, a reminder, a payment plan offer, or a financial assistance screening.

Step 5: Monitor and retrain

Retrain as payment patterns change, and track model monitoring results so the score stays current.

Keeping the score to billing, not to care

The score guides billing outreach only. It never affects access to care, and your team decides every action on an account.

Review how the model scores across patient groups before it goes live, and keep reviewing after. Reason codes and model documentation from H2O Driverless AI show which factors drive each score, so your compliance team can question them.

Protecting patient data

H2O supports HIPAA and HITECH requirements, and protected health information never has to leave your boundary. You choose how it runs: on-premises, air-gapped, private cloud or cloud.