HealthcareFree
Clinical Documentation Gap Review Checklist
Gaps counted by element across every chart, so a rate that is flat across one template points at the template rather than the people.
River's documentation improvement review reads encounter documentation against the codes actually submitted and reports which required element was absent, counted by element rather than by chart. Every finding carries its denominator, the providers it appeared under, and the note template the encounter was written in. It does not re-code anything. It reports that the documentation does not contain what the submitted code requires, and leaves the coding determination with your certified coder. That boundary is not a hedge, it is the only defensible place for it.
That unit of analysis is the whole difference. A chart audit produces a per chart verdict and a recommendation to train somebody. Aggregating by element produces a failure rate per element per template, and a rate that is flat across everyone using the same template is a template defect. Training cannot fix it, because there is nowhere in the note to put the answer. One field beats nine conversations every time, and it is measurable next quarter.
This looks for services billed without support, which is audit exposure. Finding services performed and never billed is the opposite direction and a charge capture and coding review does that. Most practices need both, run separately, because one hunts revenue and the other hunts liability. An element that was never the clinician's to record belongs to the patient intake form rather than to the note. This is for practice managers, compliance leads and coders who already know the answer and need it counted before anyone else will act on it.
Most of these checklists audit a retired rule
Since January 2023, for most visit families, the level is selected on medical decision making or on time, and the Claims Processing Manual is explicit that history and exam no longer impact visit level selection. It adds that when time is used, the full time must be completed, and that documentation volume should not be the primary influence on the level billed. Most checklists ranking for this search still count exam bullets, which is auditing against a rule that stopped applying.
Larkmead Family Medicine had 3,180 encounters reviewed by a vendor the previous quarter, which returned 3,051 findings. Every one of them was an incomplete review of systems or an exam below the old bullet count. Not one of those changes a visit level under the current rule. The review it actually needed found 679 findings that do change one, and the vendor had checked for none of them at all. Three of the four were fields the note templates did not contain.
Of those 679, the largest single group was 412 encounters where the level was selected on time and no total time appeared anywhere in the note. Eight of nine clinicians failed it between 34 and 41 percent, which is flat, so it was the template. A teaching physician presence statement was absent on 61 of 74 resident visits, because that template has no field for one to go in at all.
How it works
Send both sides
The encounter documentation and the codes that were actually submitted against those encounters.
Name the templates
Which note template each encounter was written in, and which clinicians use which one.
Count by element
Failure rate per required element, with the spread across everybody on the same template.
Change the template
Flat rates become fields to add. Outliers become one conversation with one person.
What you get
- Failure rate per element, with its denominator, not a verdict per chart
- Spread across the clinicians on each template, which separates a defect from a person
- Findings that no longer affect the visit level marked as exactly that
- The fix written as a field to add, its position, and whether it is required
- Every coding determination left with your certified coder, with no exception made
- Encounter volume behind each field, so you know what a template change is worth
Common questions
Does it decide whether the code was right?
No, and that line does not move. It reports that the documentation does not contain an element the submitted code requires, which is a documentation fact. Whether a different code fits is a coding determination for a certified coder. Where the gap has already surfaced as denials, a denial root cause analysis reads it from the remittance side.
How do you know it is the template and not the provider?
By the spread. At Larkmead, total time was missing on 34 to 41 percent of encounters for eight of nine clinicians using the same note, which is flat. Independent interpretation ranged from 2 to 68 percent, and one clinician accounted for 71 of 88 misses. Flat is a field. An outlier is a person.
Why do you say the usual checklists are out of date?
Because most of them still count history and exam elements, and those stopped selecting the office visit level. The manual says so directly and adds that documentation volume should not drive the level. A report full of review of systems findings is measuring compliance with a rule that no longer applies to the thing it claims to protect.
Do history and exam still need to be documented?
Yes, when the code descriptor calls for a medically appropriate history and examination, and for other payment systems and conditions of participation. What changed is that they no longer determine the level. So they belong in the note and not in the level calculation, which is a distinction most checklists never draw.
What about incident-to and supervision?
That is one of the recurring elements, and an easy field to add. Only the supervising practitioner may bill an incident-to service, so a note that does not name who supervised, or names someone other than the billing NPI, has a gap. Larkmead had it on 118 of 486.
Is this the same as a charge capture review?
Opposite direction. This finds services billed that the documentation does not support, which is exposure. A charge capture and coding review finds services performed and never billed, which is revenue. Run them separately, because a single pass that does both tends to quietly net one against the other.
How do we know the template change worked?
Run the same count next quarter on the same elements. That is the point of counting by element instead of by chart: the rate is comparable across periods, and a field that was added either moved it or did not. Change one template at a time or the next measurement cannot attribute anything.
Clinical Documentation Gap Review Checklist
Fill in the form and your workspace opens with the work already underway.