How independent clinics can trace claim denials back to the front desk, the exam room, or the payer—and test a fix in 30 days.
“Patients deserve to know who is responsible for their
care.” — Willie
Underwood III, MD, MSc, MPH, president of the American Medical Association
At 8:04 on Tuesday morning, an independent clinic checks in
a patient for a knee injection. The insurance screen comes back green. The
physician documents the side, the indication, and the procedure. Weeks later,
the payer denies the claim. The insurance was active. The plan also required a
referral, and nobody had confirmed it.
At 10:20, the same clinic sees a second patient. This time
the coverage checks are complete. The procedure note says “knee injection.” The
order and the note disagree about which knee. The coder sends a query. The
physician sees it three weeks later, between an inbox message about a refill
and another about an abnormal result.
Same hallway. Same biller. Two different failures. In the
first case, the information was missing before the visit. In the second, the
people who knew the answer had moved on to their next patient by the time
anyone noticed the contradiction.
These are illustrative scenarios. If you run a
practice, you can probably supply the sound effects. The phone is ringing. The
printer is out of labels. Someone is looking for a fax that three people swear
they sent.
By Friday afternoon, both cases can look like “billing
problems.” That diagnosis has all the precision of “feels bad.”
As a physician and founder of OnnX, I think the denial
queue is the most candid employee in the practice. It tells you where a
handoff failed, where a payer may have made a mistake, and where you keep
paying people to rediscover information that someone could have confirmed
sooner.
An efficient appeals team matters. So does asking why the
same appeal needs to exist every Thursday.
Seven things to try before buying another billing tool
- Pull
a recent set of first denials. Include unpaid and unresolved claims.
An export of successful appeals shows only the cases your team had time to
pursue.
- Read
the claim backward. Start with the remittance. Work through
submission, coding, the clinical note, and intake until you find the
earliest point where somebody could have acted.
- Record
the payer’s reason and your team’s finding separately. A payer’s
explanation is evidence. The underlying cause may still be uncertain.
- Rank
problems by frequency, expected dollars, and staff minutes. A frequent
low-dollar error and a rare high-cost exposure deserve different
responses.
- Give
one team one check to run at one moment. “Improve eligibility” is an
aspiration. “Resolve this plan’s referral exceptions two business days
before scheduled procedures” is work someone can do.
- Track
where the burden moves. If denials fall while appointments get
postponed and physician inboxes fill, count all three outcomes.
- Review
results with the people who do the work. The front-desk employee who
knows why the portal is misleading may have a better explanation than the
dashboard.
Where medical billing actually breaks
Revenue cycle management, or RCM, begins when
a visit is scheduled and continues through final payment or a documented
resolution. An independent practice may run it across an electronic health
record, a practice management system, a clearinghouse, several payer portals,
and an arrangement that still involves a fax machine.
The debate usually starts with a reasonable disagreement.
The front desk says coverage, demographics, and authorization determine whether
the claim gets a fair hearing. The clinical team says the record determines
whether the service and coding can be defended. Both teams can point to a stack
of examples. Both stacks can be real.
The mistake is turning those observations into a universal
order of operations. A wound care office with repeated documentation queries
faces a different bottleneck from an imaging practice with frequent location
mismatches on authorizations. A family medicine group that discovers inactive
coverage during check-in has a different opportunity from a specialty practice
losing money on a correctly paid drug.
Before the visit
Insurance eligibility generally answers whether
enrollment appears active for the relevant date. A practice may still need to
check the correct product, network, benefits, referral requirements,
authorization requirements, and any restrictions that apply to the intended
service or location. Which checks matter depends on the patient’s plan and the
service.
A familiar failure is the payer routing mix-up. The
card carries a recognizable brand. Staff select the payer entry they used last
month. The product on the card routes differently, and the clearinghouse
rejects the claim. To the practice, two entries looked interchangeable. To the
clearinghouse, they were two different destinations.
That deserves a better fix than a reminder to “pay closer
attention.” Put the plan or product detail beside the routing instruction where
staff make the selection. Give them a way to flag ambiguity. Assign someone to
update the rule when the payer changes its instructions. A sticky note can
start the experiment. It should eventually have a date and an owner.
Prior authorization has a similar handoff. The person
who secures the approval may have worked from an order that later changed. If
the site, service, dates, or units change, someone needs to know whether the
approval still applies. A green checkmark can outlive the order it was meant to
approve. A useful status says what was approved, where the confirmation lives,
and who checks it when the plan changes.
During the encounter
Clinical documentation improvement, or CDI,
helps the record capture what the clinician actually found, assessed, and did.
In an outpatient clinic that can be as straightforward as identifying the side
treated, making the indication easy to find, and reconciling a mismatch between
the order and the procedure note.
A longer note can still leave the reader hunting for the one
line that explains the service. A template that opens with three screens of
copied history may make the record feel complete while burying the assessment
from today. If your coder has to do detective work for a fact the clinician
could have confirmed in five seconds, the template has made the wrong task
easy.
The ethical limit matters. Capture the specificity that the
encounter supports. Sometimes the available information supports an unspecified
diagnosis code. Treating every unspecified code as a defect invites false
precision. Make it easy to document known details and preserve uncertainty when
it belongs in the record. Read
the official ICD-10-CM guidelines.
After the visit
Charge capture links the service performed to the
claim. Claim scrubbing checks the assembled claim for known issues
before transmission. A clearinghouse acknowledgment tells you something about
the transaction. The payer’s adjudication tells you how the claim was
processed. A payment can still differ from what the practice expected.
Here is where three numbers often get treated as one. First-pass
acceptance tells you how often claims pass a defined submission checkpoint.
Initial denial rate tells you how often claims or lines are denied on
first adjudication, depending on the unit you choose. Expected versus actual
payment tells you whether the amount paid matches the amount you reasonably
expected. It is entirely possible to improve one and see another deteriorate.
A claim can be accepted by the clearinghouse, adjudicated
without a denial, and still be financially disappointing. If your dashboard
calls all three outcomes “clean,” your dashboard has a very agreeable
personality.
Recent News and why it matters at the clinic level
The 2027 Medicare Physician Fee Schedule final rule
was published on November 9, 2026. CMS finalized conversion factors of $32.63
for clinicians who are not qualifying alternative payment model participants
and $32.95 for qualifying participants. Those are 2.31% and 1.83%
decreases, respectively, from the corresponding 2026 factors. A conversion
factor is one component of payment. The impact on a practice depends on the
services it bills and other elements of the fee schedule. Read
the CMS final rule materials and the
CMS rate-setting fact sheet.
One provision brings the original debate into focus. CMS
finalized a payment policy for certain separately identifiable office or
outpatient evaluation and management visits furnished on the same day as a 0-day
global procedure. Under the policy as CMS describes it, the highest-valued
service is paid at 100%, with the other applicable service or services
paid at 80%. This is a payment change with a defined scope. It is easy
to imagine it showing up on a practice report as apparent underpayment if the
expected-payment logic still uses the prior rule. Read
the CMS policy explanation.
A clinic owner can put this news to work without becoming a
federal rule specialist. Ask the billing lead to identify frequently billed
Medicare services that could be affected, confirm the applicable instructions,
and update expected payments by date of service. When the first remittances
arrive under the new policy, spot-check a few. If payment differs from
expectation, trace whether the mismatch is in the rule, the claim, the payer’s
processing, or your model of what should have happened.
On November 10, the Association of American
Medical Colleges published its summary of the fee schedule and Quality
Payment Program changes. That same day, the AMA reported on new policy
supporting physician-led oversight of clinical AI. The AMA action is an
association policy position. It provides a relevant principle for clinics
weighing clinical AI alongside administrative automation. A tool can surface
missing information. A named person still needs the authority to decide what
the practice documents, submits, or does next. Read
the AAMC policy summary and the
AMA report on AI oversight.
Expert opinion round-up
These are three physicians’ published positions, followed by
my application to independent practice operations.
Willie Underwood III, MD, MSc, MPH — Keep responsibility
visible
The AMA president emphasized that AI can assist
physicians and that patients should know who is responsible for their care. The
AMA’s policy calls for physician-led use of tools in clinical decision-making,
alongside appropriate oversight and evidence. Read
Dr. Underwood’s comments.
In an independent practice, I’d apply that advice this way.
Decide which tasks are administrative, which need coding judgment, and which
require the clinician. Let software flag a discrepancy. Let the billing team
resolve routine routing questions. Send a meaningful conflict about what
happened in the encounter to the person who can verify it. For an AI-generated
note, make sure the signer can find and correct the underlying facts before
signing.
Try this at your next demo. Ask the vendor to show you what
happens when the order, draft note, and charge disagree. Then ask who receives
the alert, how a correction is recorded, and how the team knows the claim has
been cleared. That five-minute demonstration can teach you more than an hour of
slides.
Bobby Mukkamala, MD — Measure relief where the work
happens
Dr. Mukkamala, AMA immediate past president, argued
that prior authorization reform should deliver tangible relief for patients and
physicians. In his June 2026 commentary, he described physicians’ skepticism
about insurer commitments and the harm that excessive requirements can cause. Read
Dr. Mukkamala’s commentary.
In an independent practice, I’d apply that advice this way.
Give authorization its own short list of measures. Track time from order to
submission, time from submission to decision, and appointments delayed while
authorization is pending. Add the number of portal or phone touches. A payer
can respond quickly after submission while the clinic spends days gathering the
pieces required to submit. A single turnaround number hides which part needs
attention.
Jan K. Carney, MD, MPH — Watch the impact on access
Dr. Carney, president of the American College of
Physicians, warned that sustained Medicare payment pressure threatens
patient access and called for more stable physician payment. ACP also raised
concerns about the effects of reducing payment for certain same-day services. Read
Dr. Carney’s remarks.
In an independent practice, I’d apply that advice this way.
When reviewing a service line, pair financial measures with an access
measure. If your proposed fix changes how visits are scheduled, ask how
many people will need a second trip, whether there are transportation barriers,
and what clinicians think about the care implications. A spreadsheet may make
an extra appointment look cheap. The patient may have to arrange another ride,
take time off work, and pay another cost share depending on the situation.
Statistics worth putting in context
The 2025 AMA Prior Authorization Physician Survey,
released in May 2026, gathered responses from 1,000 practicing physicians.
The AMA reported an average of 40 prior authorizations per physician per
week and 13 hours of combined physician and staff time. 94%
of respondents said prior authorization contributes to burnout. Only 33%
believed the insurer pledge assessed in the survey would make a meaningful
difference. These are survey findings. Your specialty, payer mix, and staffing
shape the local burden. Read
the AMA’s survey findings.
A 2022 HHS Office of Inspector General report found
that 13% of reviewed Medicare Advantage prior authorization denials met
Medicare coverage rules. It also found that 18% of reviewed payment
denials met Medicare coverage and plan billing rules. Both percentages describe
reviewed denials in that study. They should never be read as percentages of all
requests or all claims, or as estimates of current rates across every payer.
They demonstrate why a root-cause review needs room for payer processing
errors and disputes over coverage. Read the OIG report.
Under the CMS Interoperability and Prior Authorization
Final Rule, decision timeframes beginning in 2026 include 72 hours for
expedited requests and seven calendar days for standard requests for
covered medical items and services under the applicable provisions. The
decision-timeframe requirement excludes qualified health plan issuers on
federally facilitated exchanges. The rule excludes drug prior authorization
from these provisions, and its electronic requirements have their own scope and
timeline. Check the payer and request type before treating either number as a
deadline for a specific case. Read
the CMS prior authorization fact sheet.
In one real case in the OIG report, a Medicare
Advantage plan denied a hand MRI because the patient had yet to get an X-ray.
Reviewers found the MRI medically necessary and the plan reversed the denial on
appeal. The case is a reminder to keep evidence accessible even when a denial
arrives with a tidy explanation. Read OIG case D221.
Suppose you find 30 referral-related denials, each
taking 18 minutes of follow-up. That is nine staff hours in an
illustrative month. If those cases represent $9,000 in expected allowed
amounts, $9,000 is the amount at risk before you know how much will
ultimately be recovered. If your prevention check requires 12 hours of staff
time to run every month, the direct time savings alone fall short. The check
might still be worthwhile if it protects collections or prevents avoidable
delays. You need both sides of the arithmetic.
Three insights your denial report can hide
The highest-volume issue may be the wrong first project
A general rule to fix the most frequent cause sounds
sensible. It can also send a clinic after the wrong problem. Imagine 40 easy
demographic corrections and three losses on a costly administered drug. If each
demographic correction takes a minute and usually pays on resubmission, while
the drug losses reflect a repeatable mismatch between acquisition cost and
expected reimbursement, the financial priority may be the drug workflow.
That drug claim might be paid exactly as the contract
permits. The practice still loses money if it acquires the drug for more
than it expects to collect. The fix could involve making contract and
acquisition information visible to the people responsible for planning and
operations, while keeping clinical treatment decisions patient-centered. A
standard denial report might miss the entire pattern because the claims paid.
Before you rank problems, name what you are trying to
protect. It might be cash, physician time, a timely visit, or all three. When
those priorities conflict, write down who makes the decision.
Better numbers can hide a worse day for patients
A scheduling team that postpones every unresolved
authorization may reduce submitted authorization denials. That result could
look impressive on a weekly slide. It tells you little about how many patients
got care on time.
The same thing can happen on the back end. If you increase
the number of claims held before submission, first-pass acceptance among
released claims may improve while cash slows. Look at the size and age of the
hold queue. Ask who is reviewing the oldest items and what happens near a
filing deadline. In a small office, the person who owns the queue may also
cover reception on Mondays. The workflow has to account for the actual Monday.
Choose a balancing measure before launch. For
authorization checks, use delays or postponements. For a documentation prompt,
use physician interruptions or time to sign. For a scrubber, use clean claims
held and time to release. Your team should be able to tell whether it prevented
work or simply moved the work to someone who is harder to see.
The payer’s reason code and the real cause can diverge
A denial labeled “authorization” may reflect a missing
approval, the wrong location on an approval, or a valid approval that was never
linked to the claim. It might also reflect payer processing. The intervention
differs every time. Training the front desk for all four scenarios is an
efficient way to hold a meeting and an inefficient way to solve the problem.
Compare the submitted claim, order history, payer response,
and authorization record. When the evidence conflicts, make uncertain an
available finding. Set a date and an owner to revisit the case. If you demand
certainty from the first review, the team may learn to choose whichever
category makes the report easiest to close.
A 30-day denial prevention pilot
Step 1 — Pick one corner of the practice
Choose one location, a small service family, and a payer or
plan product that appears often enough to study. You might look at referral
requirements for scheduled musculoskeletal procedures or missing documentation
on a particular imaging service. Avoid starting with every payer and every
visit type. You are trying to see whether a specific action makes a specific
outcome better.
Write the scope so another person can reproduce it. Include
the service dates, claim or line unit, payer product, and how you will handle
voided and duplicate submissions. Name a physician sponsor, an
operational owner, and a backup. A backup makes the pilot more informative. If
the process only works when its creator is present, you have found a dependency
worth fixing.
Pick a decision date. Four weeks is enough to learn about
workflow in many small pilots. Claims can take longer to mature. Agree upfront
which results you will review at day 30 and when you will revisit adjudication
or collections.
Step 2 — Build a baseline from records you already have
Pull clearinghouse acknowledgments, billing work queues,
remittances, and the relevant scheduling records from an approved environment.
Add a small selection of paid claims for comparison. Review cases in date order
or use a repeatable sample. Avoid cherry-picking the stories everybody
remembers. They are good material for a meeting. They can be a poor estimate of
prevalence.
Give each case an internal reference, a service date, a plan
or product, a stage, an initial outcome, and an expected allowed amount when
you can estimate it responsibly. Add the first point of detection, the earliest
point of prevention, and the evidence that supports the finding.
Keep shared operational summaries at the aggregate level and keep patient
information in authorized systems.
A small clinic can start with a few dozen cases if volume
and time are limited. Report the count. A change from six denials to three has
a different evidentiary weight from a change from 600 to 300. The small sample
can still justify trying an inexpensive adjustment if you can reverse it
easily.
Step 3 — Sit with the people who do the work
Trace a few cases together. Ask the scheduler to show you
what the screen said at booking. Ask the authorization coordinator how a
changed order appears. Ask the coder where the relevant clinical fact is
expected to live. Ask the biller what they see before and after transmission.
There is usually a moment when somebody says, “I thought you
could see that.” Pay attention to that sentence. It can explain a month of
denials. Perhaps a note attached in a payer portal cannot be seen in the
practice management system. Perhaps an authorization number sits in a free-text
field that never travels with the claim. Perhaps the biller can see the claim
but lacks access to the information required to check it.
Step 4 — Add the smallest check that could work
For an intake problem, test a pre-visit exception list for
one plan and one procedure family. Show the scheduled service, product,
referral or authorization status, and owner in the approved queue. Give
staff time to resolve the exception. Have a clinician-directed path for cases
in which the administrative status conflicts with the urgency of care.
For a clinical documentation problem, start with a short
prompt inside the existing note or charge workflow. Ask for the missing detail
where it is known. If the note conflicts with the order, route the conflict for
resolution. Keep the question neutral. A prompt that offers the financially
attractive answer first can shape the record in ways you will regret.
For a submission problem, use a deterministic rule when
the condition is unambiguous. The expected payer identifier is a better
candidate for a rule than a judgment about clinical necessity. Run a new rule
silently against a small set first. See what it would have flagged, what it
would have missed, and how many correct claims it would have slowed.
Step 5 — Create an exception path before the first
exception
Write down who acts when a portal is down, the payer gives
inconsistent answers, an authorization expires after a reschedule, or the chart
contains conflicting information. Include where the action is recorded and when
somebody escalates it.
Give staff a way to override a billing hold with a
documented reason and appropriate approval. Keep the decision about clinical
urgency with the clinician. A claim hold and a care delay belong
in different workflows. A rule can stop a file from transmitting. It should
never quietly make a patient’s clinical decision.
Measure how often the exception path gets used. A rule that
generates overrides all week may be poorly targeted or using incomplete data.
Treat overrides as feedback. Review a sample for both false alarms and real
problems.
Step 6 — Keep the scorecard small
Use one primary outcome, one process measure, and one
or two balancing measures. For a referral pilot, the primary outcome could be
first-adjudication referral denials among eligible claims. The process measure
could be the proportion of relevant appointments with referral status verified
before the visit. The balancing measures could be staff minutes and
authorization-related delays.
Use the same definitions before and after the change.
Separate results by payer product when there are enough cases. Record volume
and service mix. A smaller procedure load can change the raw number of denials
without making the process any better.
Set a point at which you will adjust or pause. For example,
if the workflow starts adding substantial delays or the tool repeatedly flags
correct cases, review the rule before broadening it. If the first week goes
badly, capture why. The failure may reveal a missing data element or an
unrealistic staffing assumption.
Step 7 — Decide what earns a second month
At day 30, bring the physician sponsor, practice manager,
front-desk lead, and billing lead together. Review results, a few exceptions,
and one case that surprised the team. Ask where the work moved. Keep the
intervention if the evidence supports it and the burden is acceptable. Revise
it if the signal is promising and the failure is fixable. Stop it if it creates
problems that outweigh the benefit.
Keep a short decision record with what changed, what
you measured, what surprised you, and when you will review mature
claims. When a payer changes its rules or a new staff member joins, the
record explains why the check exists.
Tools, metrics, and resources you can use
Start with the tools you already pay for. Your practice
management system usually has a claims or aging export. Your clearinghouse
can help locate transaction acknowledgments and submission rejections. Electronic
remittance advice, including the 835 transaction when available, lets you
review adjudication and adjustments. Your EHR contains the order,
clinical record, and often the queue where a prompt should live. An approved
internal tracker can tie the review together. Maintain patient-level records
only in approved systems with appropriate access controls.
These measures serve different purposes.
- First-pass
acceptance. Divide claims accepted at your named checkpoint on initial
submission by initial submissions. Decide whether you mean clearinghouse
acceptance or payer acceptance and name it in the report.
- Initial
denial rate. Divide first-adjudication denials by initially
adjudicated claims or lines, using the same unit in both places. State how
you handle partial denials.
- Days
in accounts receivable. Use one consistent accounting definition and
pair the number with aging buckets. A better average can coexist with an
old, stubborn balance.
- A/R
over 90 days. Divide the applicable receivable balance older than 90
days by the total applicable receivable balance. State which receivables
are included and how credits or write-offs are handled.
- Expected
versus actual payment. Compare remittances against a validated
expectation for the relevant service and payer. Keep patient cost sharing,
contractual adjustments, and missing payments distinguishable.
- Staff
minutes per exception. Time the actual work across roles. Waiting for
a decision and working on a case are both worth measuring, in different
fields.
- Authorization-related
patient delays. Count affected appointments and, where possible, the
elapsed delay and ultimate disposition. Check whether an apparent
improvement comes from a shift in scheduling or cancellation patterns.
For reference material, keep official coding guidance,
applicable payer manuals and contracts, fee schedule addenda, and portal
confirmations where authorized users can find them. Track the policy source and
effective date. If the office depends on an unwritten exception known by one
staff member, ask that person to walk through it and document the evidence
together.
Pitfalls and what failure looks like
The universal hard stop. A clinic decides no claim
should leave without a field that is only relevant for certain plans. Its
first-pass acceptance improves among the claims it releases. The hold queue
grows, the oldest claims sit untouched, and one person is stuck overriding flags.
Repair the rule by narrowing it to the conditions that make the field
necessary. Review the first batch it held, including the cases that would have
paid.
The perfect dashboard with no owner. Red squares
multiply. The morning huddle spends ten minutes reviewing them. Everyone agrees
they are red. Put the alert in an existing work queue and name the role
expected to act. Check that the person has the information and authority to
close it.
The scrubber that inherits an obsolete policy. A
payer updates an instruction. The clinic’s software continues to enforce last
year’s version. Keep a source, effective date, and review owner for every rule
that can stop work. When the original evidence disappears, send the rule back
for review.
The documentation project that asks physicians to write
more. A leader expands a template to cover every imaginable scenario.
Clinicians learn where the defaults are and click through faster. Review
whether the added fields capture the specific information the coder needs.
Remove fields that simply invite copy-forward behavior.
The polished pilot story. A team shares the
successful cases and omits the overrides, delays, and untouched claims. The
next location inherits a tool that works in a demo. Share a failed case
alongside the best one. State what failed and what the team changed. At OnnX,
the question I want our early validation work to answer is whether an upstream
check saves more work than it creates. That requires keeping the inconvenient
cases in view.
Myth busters for independent practices
Myth 1 — Active coverage means the visit is covered. Active
enrollment answers one question. Benefits, network, referrals, authorization,
and coverage criteria can raise others. Build checks around the actual service
and plan.
Myth 2 — A longer note is easier to defend. A useful
note makes the relevant assessment, treatment, and rationale easy to find.
Review clarity and accuracy. Word count is a poor quality measure for a chart.
Myth 3 — Every denial is somebody at the clinic’s fault.
Payer errors and policy disputes belong in your analysis. Preserve evidence,
track the deadline, and use the appropriate correction or appeal channel.
Myth 4 — Every claim that passes the scrubber will get
paid. A scrubber checks what it has been configured and equipped to check.
Compare payment with validated expectations and review the cases that sail
through submission and still go wrong.
Myth 5 — AI is the next step whenever the manual process
hurts. First watch someone do the work. A missing field, contradictory
rule, or inaccessible screen may explain the pain. Once you can describe the
recurring failure, choose the least burdensome intervention that addresses it.
Legal, practical, and ethical considerations
HIPAA and vendors. Before using cloud or AI tools
that handle electronic protected health information on the practice’s behalf,
determine the vendor’s role and whether a business associate agreement is
required. Review risk, permissions, access logs, retention, subcontractors, and
incident handling with the appropriate team. Test with synthetic cases where
feasible. A signed agreement is one element of a broader privacy and security
program.
Clinical integrity. Design prompts around what the
care team can substantiate. Preserve the source record and use appropriate,
dated corrections or addenda when required. Involve a coding professional when
billing rules are unclear. If ambient documentation records patient conversations,
have counsel and the practice’s privacy lead review applicable recording,
consent, and retention requirements before rollout. Give clinicians a way to
verify drafts and correct errors where they work.
Patient financial communication. Give patients a
clear explanation when coverage or authorization is unresolved, explain the
next step, and provide a contact. Have appropriate staff check applicable
contract terms and federal or state requirements before moving an unpaid amount
to patient responsibility. Review estimates, notices, and financial assistance
procedures with qualified advisors. A denial is a claim outcome. It does not,
by itself, establish who owes the amount.
Access and fairness. Review whether workflow changes
make care harder to obtain for people who rely on interpreters, have limited
portal access, or need time to arrange transportation. Make it clear how
time-sensitive cases receive clinical review. A patient should be able to reach
a person when the system gets stuck. The value of a workflow includes how it
handles people who have the hardest time using it.
Future outlook
Electronic prior authorization can shorten some handoffs and
make them easier to trace. Under the CMS interoperability rule, applicable API
requirements generally begin in 2027, with scope and timing defined by
the rule. An electronic response can make a status easier to retrieve. A clinic
still has to know which order it belongs to, whether the order has changed, and
who acts on the response.
I expect practical progress from tools that let people
resolve exceptions inside their normal workflow, show the evidence behind an
alert, and send a correction back to the relevant system. A smarter note draft
or faster claim check matters when staff can trust the fields they see and have
the authority to fix them. Every additional login has to earn its place in a
Tuesday morning.
Frequently asked questions
Should our clinic fix eligibility or documentation first?
Start with a defined set of claims and compare frequency,
dollars at risk, staff time, and patient impact. For
recurring referral mistakes, work with scheduling and intake. For recurring
clinical queries, examine the note and charge workflow. Pick the problem you
can define, measure, and address with a clear owner.
Is a rejection the same as a denial?
A rejection commonly means a transaction failed an
intake or validation step. A denial commonly means the payer adjudicated
the claim or line and declined payment under a reason it reports. Check the
terms used by your systems. The difference matters because the correction and
follow-up paths may differ.
How should we test a claim scrubber?
Start with historical examples and a silent trial. Include
correct claims and genuine errors. Count false positives, missed errors,
and time added. Ask the vendor to show the rule source, the override process,
and the audit history. A pilot succeeds when the team can resolve flagged
issues efficiently and send valid claims through on time.
Can a small clinic run this without an analyst?
Yes. One billing lead and an operational partner can review
a small, consistent sample. Start with a single payer product or service family
and a weekly review of exceptions. Use an approved system for patient
information and share only aggregate findings outside the authorized workflow.
Can an ambient scribe solve documentation denials?
An ambient scribe can help draft a record. Its value for
billing depends on the quality and relevance of the resulting note, clinician
verification, and the connection to coding and charge capture. Test whether
queries or rework decrease and whether review time or correction burden rises.
Measure the outcome that motivated the purchase.
What should we do when the payer appears to be wrong?
Check the policy and effective date, compare the submitted
claim with the payer’s response, and gather the relevant evidence. Assign the
case to someone who can verify the available correction or appeal options and
the associated deadlines. If you find a repeated pattern, log the cases
together so the practice can evaluate whether broader escalation is warranted.
References
- Your
Billing Denial Has a Plot Twist: The Problem May Have Started Before the
Claim Existed — A companion article on tracing a denial back to the
first broken handoff.
- Healthcare
Doesn’t Have an AI Problem. It Has an Information Problem. And Your
Billing Department Is Paying for It. — A companion article on
preserving payer rules, workflow context, and staff knowledge.
- AI-Driven Medical
Billing Explained | Faster Reimbursements, Fewer Denials — Overview of
how AI fits into medical billing and EHR workflows.
- CMS-1848-F
— CMS’s November 9, 2026 publication record and downloads for the 2027
Medicare Physician Fee Schedule final rule.
- Calendar
Year (CY) 2027 Medicare Physician Fee Schedule Final Rule — The CMS
fact sheet explains the conversion factors and payment policy for certain
same-day services.
Final Thoughts
A practice learns more from tracing one preventable denial
all the way back than from adding another column to its aging report.
Get Involved
If your billing team stopped fixing the same mistake
tomorrow, where in your practice would it show up first?
Comment with one recurring handoff you would change,
using a general example without patient or confidential details.
Repost this article to help another physician or clinic
owner ask where the next denial can be prevented.
About the Author
Dr. Daniel Cham is a physician, medical technology
consultant, and founder of OnnX, an AI-powered medical billing SaaS
designed to help small and midsize clinics eliminate unnecessary billing
intermediaries. His work spans healthcare management, practice operations, and
medical billing, with a focus on practical ways to reduce administrative burden
and improve the flow of information across care teams.
Editorial Note
This article is for general education. Consult qualified
legal, clinical, coding, or compliance professionals for advice on a specific
practice, payer policy, or patient situation.
Keep the Conversation Going
I share practical lessons from healthcare operations,
medical billing, and building technology for independent practices. If you want
to put one of these ideas to work, start with a case your team already
understands and see what it teaches you about the rest of the process.
Connect: LinkedIn | Website | YouTube| Spotify
P.S. You’ll find a free download in the Featured
section of my LinkedIn profile. No signup required.
#RevenueCycleManagement #MedicalBilling #DenialPrevention
#PriorAuthorization #IndependentPractice #PhysicianEntrepreneur #HealthcareAI
#PracticeManagement #ClinicalDocumentation #OnnX
