DRG downgrades, clinical validation denials and medical necessity denials — the denials now filling most queues — share a trait: none of them dispute the care delivered. They dispute how the record reads once the case is closed, which means the argument is settled at a moment most organizations do not review.
During a featured session at Becker’s 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health on Sept. 15, Erica Strick, senior vice president at CorroHealth, made the case for adding one clinical review after coding is complete and before the claim is released. Ms. Strick oversees both clinical documentation integrity and clinical denials, functions that in most organizations sit under different leaders.
Below are four takeaways from her presentation.
Note: Quotes have been edited for length and clarity.
1. Concurrent CDI is not wrong — it is early
When a CDI specialist opens a chart on hospital day two, Ms. Strick said, labs are pending, the working diagnosis may change by Friday and the coded DRG does not yet exist. “It’s like proofreading a novel while the author is still writing the final chapter,” she said. “And the ending is the only part that the payer reads.” A payer’s medical director is reading a finished document, often months later, looking for whether the clinical evidence is there in black and white.
2. Every record has two audiences
The care team asks whether the patient was sick; the payer asks whether the documentation proves the patient met clinical criteria. Concurrent CDI asks whether the documentation was complete while the patient was admitted — a question of completeness, Ms. Strick said. Pre-bill CDI asks a question of defensibility: if the payer’s medical director reviewed this claim tomorrow, would they approve it? The second review assumes someone on the other end is looking for a reason to say no.
3. Target the review
Reviewing every chart is neither realistic nor the point, Ms. Strick said; the goal is exposure, not volume. High-dollar DRGs, high-denial service lines, high-risk payers and clinical validation diagnoses such as sepsis, acute respiratory failure and malnutrition make the list — conditions that are clinically real but evidentially fragile. Pulling last year’s clinical denials sorted by volume and dollars produces the criteria, and machine learning can sharpen the targeting using an organization’s own denial patterns rather than a static list set in committee.
4. Promise fewer denials and the program will lose in a budget meeting.
“Pre-bill CDI will not lower your denial volume,” Ms. Strick said, adding that she wanted to be quoted accurately on that point. Denial volume is driven by payer strategy. What the review changes is the condition of the record when the denial lands. She recommended baselining three measures before implementation: overturn rate on appeal, the share of appeals rated strong at submission and dollars denied versus dollars upheld.
What this means for revenue cycle leaders
Scoring appeal strength also separates a documentation problem from a payer problem. Ms. Strick described a client whose appeals carried the same strength ratings across two payers and drew a 70% to 75% overturn rate from one and 10% from the other — evidence to take into contract negotiations rather than back to the CDI team.
Appeal teams cannot create clinical evidence after the fact, she noted; they can only work with what the record already holds. “Your appeal team doesn’t have to build the case. They just have to present it,” Ms. Strick said. “That is the difference between defending a claim and hoping.”
