The Rise, Overreach, and Reinvention of Propensity Scoring

Introduction

For more than two decades, healthcare has relied on propensity-to-pay scoring to guide patient-account treatment. It was a meaningful advance, moving patient-pay beyond static billing cycles, generic work queues, and undifferentiated outreach.

But propensity is a directional signal, not a complete revenue strategy. It can estimate who appears more likely to pay. It cannot explain why payment is at risk, determine what might change the outcome, or identify which financial path is most likely to preserve recoverable value.

The limitation is not the score itself, but the market's failure to recognize its boundaries—using a directional signal as though it were a prescriptive revenue strategy.

The Rise and Overreach of Propensity Scoring

Propensity scoring came from consumer credit, lending, collections, and receivables markets, where models were used to estimate behavior, rank recovery opportunities, and prioritize work. Healthcare adopted that logic to bring greater structure and differentiation to patient receivables. [1]

Its overreach began when the signal became a revenue strategy. The governing question became who is likely to pay, rather than what revenue outcome might be possible across all account types and alternative treatment paths, which conditions could be changed, and which action offered the strongest expected value.

The fallout was predictable. As with diagnostic imaging, the quality of the conclusion depends on how deeply the technology can see. An X-ray reveals less than a CT scan; the subject may be the same, but the available insight is not.

Propensity has the same limitation. It predicts likely behavior from the conditions visible to the scoring system, but it does not examine the account deeply enough to identify the economic factors that could reshape that behavior. As a result, lower-propensity accounts are often flagged as weak opportunities rather than evaluated for how they might be repositioned toward higher-probability resolution.

From Propensity to Revenue Engineering

The above limitation defines the next step. Propensity should remain an input, while Economic Intelligence looks beyond payment likelihood to the conditions shaping it—and engineers the financial pathway most likely to improve revenue across every account type.

The principle is familiar in other financial markets: an asset performing poorly under its current structure is not necessarily without value. Changing the terms, reducing the burden, strengthening repayment conditions, or directing the asset through a more appropriate resolution path can materially improve expected value.

Economic Intelligence applies the same logic to patient accounts. It does not accept the initial probability score as the final economic conclusion. It evaluates why the account is likely or unlikely to pay, which conditions can be changed, and whether a different treatment can produce a stronger outcome.

For a high-propensity account, that may mean preserving a simple, low-cost payment path rather than introducing unnecessary financing or extended terms. For a marginal account, it may mean adjusting the initial request, term, timing, or engagement strategy. For a low-propensity account, it may mean identifying missed coverage, financial assistance, hardship treatment, or another pathway capable of converting an otherwise weak recovery opportunity into greater economic value.

Revenue engineering therefore does more than rank accounts. It determines how each account should be structured, treated, and redirected to improve expected value across the full patient population.

The process follows directly from the principles established earlier in the series:

1. Use propensity to estimate likely behavior.
2. Diagnose the economic conditions and revenue opportunities shaping that likelihood.
3. Determine which conditions can be changed to improve expected value.
4. Compare the viable intervention and resolution pathways.
5. Direct the action offering the strongest expected outcome.

Market-Forward: Repositioning the Signal

The market advancement is not a better propensity score, but a more economically complete use of it. Within Economic Intelligence, propensity estimates the account's current likelihood of payment; revenue engineering determines whether the conditions shaping that likelihood can be changed to produce a stronger outcome.

Reference

1. Early healthcare-market development included ARxChange's application of receivables-market principles to healthcare balances, including ranking, sorting, prioritization, valuation, and liquidation-path analysis, along with related scoring, eligibility, estimation, and consumer-risk capabilities introduced by SearchAmerica, Experian, TransUnion, Equifax, and others. See Forbes, "How a (Non-Apple) U.S. Patent Might Just Change the World," July 31, 2012; U.S. Patent Nos. 8,234,209 and 8,442,903; and HFMA, "ARxChange Patents: A Patented Approach to Organizing, Valuing, and Improving Medical Receivables Performance."