What Scrubbing Is
Claim scrubbing is systematic pre-submission error checking: every claim passes through rule sets — demographic completeness, code validity, code-pair conflicts, payer-specific requirements — before it leaves. The economics are the whole argument: an error caught at scrub costs a minute; the same error caught after denial costs weeks of rework and payment delay.
The Edit Layers
Format and completeness edits catch structural failures: missing identifiers, invalid dates, malformed fields. Clearinghouses provide these by default.
Code-set edits validate codes against current CPT, ICD-10, and HCPCS sets — retired codes and invalid combinations stopped before payers see them.
Bundling and relationship edits apply NCCI (National Correct Coding Initiative) logic and similar rules: code pairs that conflict, mutually exclusive procedures, and modifier requirements for legitimate exceptions.
Payer-specific edits encode individual payers’ quirks — the requirements generic rules miss and your denial history reveals. This layer is where scrubbing stops being commodity and starts being competitive.
Practice-specific edits encode your own recurring mistakes: the provider who forgets a required field, the service line with chronic modifier issues.
Building From Denial History
The best edit set is reverse-engineered from your own failures: mine 6–12 months of rejections and denials, rank causes by frequency and dollars, and write an edit for each recurring preventable cause. Generic edit libraries catch generic errors; your denial history catches your errors.
The Tuning Loop
Edits are living rules: monthly review adds edits for new failure patterns and — equally important — retires zero-yield edits that only add friction. Track edit yield (errors caught per rule); an edit that never fires is noise, and noise slows claims without protecting them. Over-editing is a real failure mode: claims crawling through hundreds of low-value rules serve nobody.
What Scrubbing Cannot Catch
Rules catch what rules can describe. Documentation insufficiency, medical-necessity judgment, and clinical-coding mismatches need human review — scrubbing complements coding review, never replaces it. A claim can pass every edit and still be wrong about what happened in the exam room.
Practical Checklist
- Clearinghouse and code-set edits confirmed current
- NCCI/bundling logic active with modifier awareness
- Payer-specific rules built from your denial history
- Edit yield reviewed monthly; stale rules retired
- Scrub-failure queue worked daily with correction turnarounds
Frequently Asked Questions
Our system has thousands of edits available — should we enable them all? No — enable what maps to your failure history and payer mix. Blanket enablement buries claims in false positives and staff in noise. The discipline is curation: every active edit should be able to justify its existence with caught errors.
Do edits slow down our claims? Passing claims go out same-day as always; failing claims were going to bounce anyway — the edit moved the fix from weeks later to minutes now. Only bloated, untuned rule sets create real drag, which is what yield review prevents.