The Practice Under Pressure← All research

What Reviewers Actually Read

Clinical documentation that survives a medical-necessity review.

Merits Research · every figure linked to its source · last verified July 2026

A medical-necessity reviewer is not reading a chart to learn what happened to the patient. They are reading it against a document — a coverage policy that was written and published before the claim was ever submitted, and that lists, in order, the conditions under which the service is paid. For a Medicare claim that document is a Local Coverage Determination or National Coverage Determination. For a commercial claim it is the payer's own medical policy. The reviewer's job is a match: does the record contain the facts the policy requires, or does it not?

That reframes what “good documentation” means. A note can be clinically excellent — fluent, thorough, obviously written by someone who knows the patient — and still fail review, because the reviewer is not grading prose. They are looking for specific, findable elements. The notes that survive are the ones built to be matched: they carry the criteria the policy names, in language close enough to the policy's own that no inference is required to find them.

The useful part is that the criteria are not secret. They are published, and you can read the exact document the reviewer is holding. To make the point concrete, this piece uses one procedure — the epidural steroid injection — and two real, public coverage documents that govern it: a Medicare LCD and a commercial payer's clinical policy. They were written by different organizations for different populations. They ask for the same four things.

Read the document the reviewer is holding

Novitas Solutions — a Medicare Administrative Contractor — publishes LCD L39240, Epidural Steroid Injections for Pain Management. Its “Covered Indications” section states that the injection “will be considered medically reasonable and necessary when the following three (3) requirements are met.”1 The three requirements, quoted from the LCD:

History, physical examination, and concordant radiological image-based diagnostic testing supporting one of the following… Lumbar, cervical or thoracic radiculopathy, radicular pain and/or neurogenic claudication due to disc herniation, osteophyte or osteophyte complexes, severe degenerative disc disease, producing foraminal or central spinal stenosis.1

Novitas LCD L39240 — Covered Indications, requirement 1

Radiculopathy, radicular pain and/or neurogenic claudication is severe enough to greatly impact quality of life or function. An objective pain scale or functional assessment must be performed at baseline (prior to interventions). The same scale must be repeated at each follow-up for assessment of response.1

Novitas LCD L39240 — Covered Indications, requirement 1

Pain duration of at least four (4) weeks, and the inability to tolerate noninvasive conservative care or medical documentation of failure to respond to four (4) weeks of noninvasive conservative care.1

Novitas LCD L39240 — Covered Indications, requirement 1

Four elements are load-bearing here, and each is a documentation instruction, not a clinical opinion: a diagnosis in a named category, supported by concordant imaging; an objective baseline measurement that is repeated at follow-up; a symptom duration of at least four weeks; and failure of conservative care across that same four-week window. A note that lacks any one of them does not meet the LCD, however sick the patient plainly is.

A commercial payer asks for the same four things

None of this is a Medicare peculiarity. Aetna's Clinical Policy Bulletin 0016, Back Pain — Invasive Procedures, sets medical-necessity criteria for interlaminar epidural injections in its own words. It requires that the “pain is radicular in nature (radicular signs may include, but are not limited to, a positive straight leg raise or a dermatomal pattern of sensory loss),” and that the member “has failed to improve after 4 or more weeks of conservative treatments (e.g., rest, systemic analgesics, physical therapy).”2 For a repeat injection, the bulletin is exact about what the record must show:

Additional interlaminar epidural injections, if the initial injection resulted in at least two of the following for at least two weeks: a 50% or greater relief in pain; increase in the level of function/physical activity; reduction in the use of pain medication and/or additional medical services.2

Aetna Clinical Policy Bulletin 0016 — Back Pain: Invasive Procedures

Line the two documents up and the same skeleton appears: an objective sign of radicular pathology, a symptom duration, a documented failure of conservative care, and — for anything past the first injection — a measured, numeric response. Two payers, one public and one commercial, converge on the same four questions. That is the structure worth charting to, because it is the structure the reviewer is reading against.

The same visit, charted two ways

Below is a single fictional encounter — an invented patient, invented numbers, written only to illustrate the gap. The first version is how the visit is often charted: accurate, humane, and unmatched to the policy. The second carries the same clinical reality but is written so a reviewer can find each required element without inferring it. Nothing clinical was added; the facts were made locatable. Each strengthened line is tagged with the criterion type it satisfies.

Fictional illustration — not a real patient record

As often chartedFails the match

S: 58 y/o with ongoing low back pain radiating down the leg. Pain is affecting daily life. Has tried a few things without much luck.

O: Antalgic gait. Tender lumbar paraspinals. MRI reviewed, degenerative changes noted.

A: Chronic low back pain with radicular features.

P: Given severity and failed conservative management, will proceed with epidural steroid injection. Medically necessary.

Same encounter, charted to the criteria

Diagnosis → procedureSymptom durationConservative care tried & failedObjective findings
As the criteria are writtenMeets each named element

S: 58 y/o with L5 radicular pain in a dermatomal distribution down the left leg; onset 9 weeks ago, continuous since.

≥ 4-week duration, stated as a number

Conservative care completed: 6 weeks of physical therapy plus scheduled NSAIDs; failed to improve — pain and function unchanged.

Failure after ≥ 4 weeks of named conservative care

O: Positive straight-leg raise on the left; dermatomal L5 sensory loss. MRI (04/2026): left paracentral L4–L5 disc herniation with foraminal narrowing, concordant with symptoms.

Objective exam findings + concordant imaging

Baseline pain 7/10 on NRS; Oswestry 46%. Same scales to be repeated at follow-up.

Objective baseline scale, repeatable at follow-up

A: Left L5 lumbar radiculopathy due to L4–L5 disc herniation with foraminal stenosis (M51.16).

Diagnosis in a covered category, linked to the procedure

P: Interlaminar epidural steroid injection at L4–L5 under fluoroscopic guidance for the documented radiculopathy above.

Procedure tied to the specific diagnosis

The second note is not longer by much, and it is not more clinically sophisticated. It is simply written for a second reader who was not in the room. Where the first note says “a few things without much luck,” the second names the conservative care and its duration. Where the first says “affecting daily life,” the second records a baseline score on a named scale. Where the first asserts “medically necessary,” the second never uses the phrase — it lets the criteria be met and leaves the conclusion to the reviewer. The word “necessary” persuades no one; the four elements do.

A checklist, taken only from the criteria

The checklist below is not a template of best practices. Every row is drawn from the two quoted documents and nothing else — if a line is here, it is because L39240 or CPB 0016 names it. For any given service the governing policy differs, so the durable habit is not memorizing this list; it is pulling the actual LCD or payer policy and charting to whatever it names.

Criterion typeWhat the record must containFrom
Diagnosis → procedureA diagnosis in a covered category (e.g., radiculopathy), stated explicitly and tied to the ordered service — not a general complaint.L39240 req. 1; CPB 0016
Objective findingsExam signs of the pathology (e.g., positive straight-leg raise, dermatomal sensory loss) plus concordant imaging that supports the diagnosis.L39240 req. 1; CPB 0016
Objective baseline measureA pain scale or functional assessment performed at baseline, on a named scale, to be repeated at follow-up.L39240 req. 1
Symptom durationPain duration of at least four weeks, stated as a specific interval rather than “chronic” or “longstanding.”L39240 req. 1
Conservative care tried & failedThe specific conservative treatments attempted and their duration (≥ 4 weeks), with a documented failure to respond.L39240 req. 1; CPB 0016
Measured response (repeats)For any repeat, a numeric response to the prior injection — e.g., ≥ 50% pain relief, increased function, reduced medication.L39240 req. 6; CPB 0016

Every row is quoted or paraphrased from the two cited documents (sources 1–2). It is not a general standard of care; it is what these two policies require for this one procedure.

Why this is the whole game on a medical-necessity denial

A CO-50 denial — “non-covered because not deemed a medical necessity” — is, almost always, a failed match rather than a clinical disagreement. The service may have been entirely appropriate; the record simply did not carry, in findable form, the element the policy required. That is why the strongest appeal rarely argues about medicine. It quotes the payer's own criterion, then points to the line in the record that satisfies it — the same match the reviewer was performing, run in the provider's favor. See medical necessity for how coverage defines the term.

The reviewer reads the note against the document. The document is public. Read it first, chart to it, and the match takes care of itself — before there is ever a denial to appeal.

This is documentation and coverage analysis, not clinical advice; clinical decisions and their documentation remain the treating provider's judgment.

Sources

  1. 1Novitas Solutions — LCD L39240, Epidural Steroid Injections for Pain Management (CMS Medicare Coverage Database; Covered Indications quoted verbatim; orig. eff. 06/19/2022, rev. eff. 04/16/2026) · rev. Apr 2026
  2. 2Aetna — Clinical Policy Bulletin 0016, Back Pain: Invasive Procedures (interlaminar epidural injection medical-necessity criteria, quoted verbatim) · accessed Jul 2026

Merits turns a denied EOB into a cited appeal letter — the same sourced discipline as this page.

Appeal a denial →