How to estimate denial rework costs

You need to explain what denials mean for your team’s workload. Start with the cases that need follow-up, then estimate the work. Keep that estimate separate from the amounts still under review.

1. Define a group before assigning a cost

Begin with a consistent population: a specified number of claims and an initial denial rate that uses the same unit and scope. If the volume counts claims but the rate counts lines, the multiplication does not answer a well-defined question.

For this tool, “initial” means the first denial outcome counted under your report’s definition. Do not include each resubmission as if it were a new original claim. If the source system’s definition is unclear, fix that before drawing a conclusion from the estimate.

Use figures from a report with a defined group of claims and a consistent follow-up period. Enter those totals in the calculator; individual records are not needed.

2. Calculate the initially denied population

In our worked example, there are 20,000 claims per month and an 8% initial denial rate. The modeled count is 1,600. The estimate comes from the monthly volume and denial rate you enter.

A percentage is not automatically comparable across periods. Changes in the denominator, the service mix, reporting timing or classification can change the meaning. Record those definitions alongside your report so another reviewer can follow the comparison.

3. Estimate the administrative work separately

Enter one estimated rework cost per initially denied claim. At $21 in the example, 1,600 claims produce $33,600 of administrative cost. Include labor in that per-case estimate, rather than counting the same work a second time.

State what the $21 includes: direct staff time, an external charge or a fully loaded cost estimate. Freed staff time is additional capacity; it is not automatically a reduction in payroll or other cash costs.

If a change frees staff capacity but staffing expense stays the same, report the capacity effect separately from cash savings. Use actual expense changes to support a cash-savings figure.

4. Show the associated allowance without calling it lost

At an average allowed amount of $230, the initially denied group is associated with $368,000 of allowance. That is a scale indicator under the model. It is not a verified outstanding balance or an amount entirely owed by a single payer.

The actual position could involve patient responsibility, other payers, adjustments and later payments. This tool does not know those allocations. For a defined responsibility-based payment check, use the separate variance calculator.

5. Give “resolved” a definition and a time window

The resolution percentage is a supplied assumption. Decide whether your internal measure means a payment, an accepted resubmission or a closed work item. Those outcomes are not interchangeable, and the tool cannot choose between them.

At an assumed 48% resolution rate, the remaining share is 52%. The model therefore gives 832 unresolved claims and $191,360 of associated allowance. These claims are not labeled permanently uncollectible. The result does not predict appeal success.

If a group of claims has not had enough time to resolve, a percentage based on an older group of claims may be misleading. Use comparable follow-up or make the uncertainty visible. Keep the estimate separate from results observed in a follow-up report.

6. Keep additional delay in days

Our example uses 34 additional days. The tool displays that assumption on its own. It does not multiply days by the associated allowance and call the result financing cost, interest or permanent loss.

A separate cash-flow model would need to explain the timing distribution, amounts actually payable, actual payment dates and financing assumptions. Use your cash-flow analysis for that separate question.

7. Write a conclusion the inputs can support

A careful summary might say: “Using 20,000 monthly claims and an 8% initial denial rate, our scenario estimates 1,600 initially denied claims. At the entered $21 rework cost, the associated administrative cost is $33,600. The group is associated with $368,000 of allowance; it is not all confirmed lost. The assumed additional delay is 34 days.”

This separates the workload estimate from the allowance still under review. It also records the assumptions another reviewer would need.

8. Investigate the actual explanation separately

CMS’s remittance guidance describes the roles of adjustment and remark codes. Use the applicable source records to understand why a payment differs. For a recurring pattern, follow the root-cause review to compare affected cases and supporting records.

Try the denial cost calculator ↗

Read the method and rounding policy. Do not enter patient identifiers or upload a claims file. A worked example is enough to learn the model.

Use the same cohort for every input

The initial denial rate, average allowance, rework assumption and resolution assumption must describe the same chosen group. An appeal overturn rate among decided appeals is not automatically a resolution rate for all initial denials. If the assumptions come from different groups, label the model as exploratory rather than presenting it as an operating result.

Keep reading

The denial review guide

Primary sources

• Health Care Payment and Remittance Advice
• Payment, remittance advice and EFT transactions
• Allowed amount