Denial Rate Benchmarks: The Good, The Bad, and the Alarming

Answer honestly. Are you worried about your denial rates? If you aren't, you should be.

Most practices are measuring the wrong numbers and comparing their rate to the wrong benchmarks. The current denial rate quoted by nearly every source right now is 11.81%, and it comes from Kodiak Solutions data covering more than 2,100 hospitals. Your practice is not in that data set. MGMA reports a first pass denial rate of 8% for single specialty groups, and that is where the figure sat in 2019 as well.

Even more concerning is what happens after the denial. Kodiak reports a final denial rate of 2.8% for 2024, so most denied claims get paid once somebody works them, which is alarming. KFF found that 80.7% of appealed Medicare Advantage prior authorization denials were overturned in 2024. Payers are denying claims they will eventually pay, and you absorb the difference as staff time, rework cost, and days in A/R. If that does not make you angry, it should.

Benchmarks

Benchmark	Value	Population	Source
Optimal initial denial rate	< 5%	All providers	HFMA KPI guidance
Stated industry average range	5% to 10%	All providers	HFMA KPI guidance
First pass denial rate	8%	Single specialty groups	MGMA DataDive
Average denial rate	16%	5,729 organizations	Allscripts PPD, via MGMA
Initial denial rate, 2024	11.81%	2,100+ hospitals	Kodiak Solutions
Initial denial rate, 2020	10.2%	Hospitals	Kodiak, via Becker's
Final denial rate, 2024	2.8%	Hospitals	Kodiak Solutions
Denials originating in the front office	~50%	5,729 organizations	Allscripts PPD, via MGMA
Denials potentially avoidable	86%	Industry	Change Healthcare, via MGMA
Average cost to rework one claim	$25.20	Industry	MGMA

Note the spread between the 8% MGMA figure and the 16% Allscripts figure. Both describe practices rather than hospitals, and they come from different data sets with different participants. Treat them as a range and pay closer attention to your own trend than to either number.

Why the denial rate alone will mislead you

There is more than one way to count it

HFMA's Claim Integrity Task Force publishes two versions of the initial denial rate. One counts claims, so denied claims divided by total claims. The other counts money, so denied charges divided by total charges. Both use a rolling three-month average.

Those two calculations will not agree, and the reason matters. Your denied claims are not an average sample of your claims. A surgical or imaging claim is worth many times what an office visit is worth, and higher dollar services draw more payer scrutiny. When you count claims, that surgery counts once, the same as a 99213. When you count dollars, it counts for what it is worth.

The practical effect is that the same practice in the same month can look acceptable on claim count and considerably worse on dollars. Neither number is wrong. They answer different questions, and the dollar version is the one that tells you what is actually at risk.

So before you compare yourself to any published benchmark, find out which version it is. A denial rate quoted with no basis attached cannot be compared to anything.

A rejection is not a denial

A rejection fails an edit at the clearinghouse or the payer's front end. It never reaches adjudication, so there is no claim number and no remittance advice. A denial was accepted, adjudicated, and the payer decided not to pay. MAP Key AR-5 counts only claims that produced a remit, so rejections fall outside that calculation entirely. They belong in their own report with their own workflow.

What mis-categorization actually costs

The most expensive denials I have encountered were not denials. Running revenue cycle across 50 facilities and 180 providers, I discovered a batch of information requests shelved by our offshore team as "unworkable denials." They weren't denials at all. They were simple requests for a single missing provider ID. Because nobody answered, the claims expired and hard-earned revenue evaporated.

That did not happen because people were careless. The team didn't understand the request, and there was no system to log them or track them. A denial rate would not have shown any of it. The claims were sitting in a queue nobody was measuring.

Three jobs, three sets of metrics

Most of the confusion here comes from asking one number to do three jobs. The front end metrics reduce how many denials arrive. The denial rate tells you the size of the problem. Worked denial rate, appeal success rate, and A/R aging tell you whether your denials workflow is doing its job once the denials exist. You need all three sets.

Start with prevention

The front end determines most of your denial volume, and the data is consistent about it.

• Registration and eligibility account for nearly 27% of denials.
• Authorization and pre-certification account for 11.6%.
• Roughly 50% of all denials originate in the front office.
• 86% of denials are considered potentially avoidable.

Clean claim rate is the one to lead with. It sits at the last checkpoint before adjudication, it captures the result of everything upstream, and it is the only front-end MAP Key that HFMA attaches a published target to. The other three tell you where a falling clean claim rate is coming from.

Measuring the denials workflow

These tell you how much of the money came back, how long it took, and whether your denials process is working. Prevention lowers the volume arriving in the queue. These tell you if your denials management team and your workflow process is effective.

What good looks like in a 1 to 15 provider practice

No published benchmark is banded by practice size. What follows takes HFMA's published values and lands on the good end of each range rather than the average, because a practice this size has less room to absorb leakage than a health system does.

Metric	Aim for	Basis
Initial denial rate	< 5%	HFMA optimal
Clean claim rate	98%	HFMA published target
Denials resolved within 30 days	85%	HFMA published target
Days in A/R	30	Good end of HFMA's 30 to 40 range
A/R over 90 days	< 10%	HFMA published target
Self pay A/R over 90 days	< 30%	HFMA published target
Net collection rate	97% to 99%	HFMA optimal
Charge capture	Within 3 days	Good end of HFMA's 3 to 5 day range
Late charges	< 2%	HFMA published target
Bad debt	< 5%	HFMA published target

This is a very dependent on specialty, location, and demographics. A practice carrying heavy prior authorization requirements or high dollar procedures will run structurally higher denial rates than a primary care practice with identical processes, and no published benchmark adjusts for that. Keep all of this in mind when you consider your personal benchmarks. These are just a guide. Your own trend over a rolling quarter is more useful than your position against any single number in this table.

These targets are reachable. Across 50 facilities, 180 providers, and more than $150M in reimbursement managed, the programs I have run produced a 37% reduction in Days in A/R, more than $7M in bad debt eliminated, more than $2M in net recoveries, and 97% sustained coding accuracy.

Other measures to track

Don't forget your payer mix

Before you conclude that your practice is underperforming, look at who you bill. Kodiak reports Medicare Advantage initial and final denial rates at more than double those of traditional Medicare. Medicare Advantage insurers denied 7.7% of nearly 53 million prior authorization determinations in 2024, up from 6.4% in 2023. ACA marketplace issuers denied 19% of in network claims in 2024, with individual issuers ranging from 3% to 36%.

This is why the payer level cut is not optional. A single blended denial rate across a mixed book will not tell you much.

Measuring this in your own system

Pull the trailing 90 days, which keeps you consistent with the rolling three month average the Claim Integrity Task Force uses. Run the denial rate on claim counts and on gross charges, and run everything by payer as well as in total.

Three things will inflate the number if your report includes them:

• Contractual adjustments. A contractual write down appears on properly paid claims. AR-5 does not count these.
• Patient responsibility. AR-5 excludes patient responsibility amounts and group code PR from the numerator.
• Rejections. A rejection never produces a remittance, so it cannot appear in an AR-5 calculation on either side.

Confirm how your practice management system flags denials, and which basis it is using, before you trust the report it hands you.

Frequently asked questions

What is a good denial rate for an independent practice?

HFMA's published KPI guidance defines an optimal initial denial rate as below 5% and puts the stated industry average at 5% to 10%. For medical groups, MGMA DataDive reports 8% on first pass for single specialty groups, while the Allscripts Practice Performance Database reports a 16% average across 5,729 organizations. The 11.81% figure quoted in most articles is hospital data from Kodiak Solutions and is not the right comparison for a practice. No published benchmark is banded by number of providers, so a 1 to 15 provider practice should aim at HFMA's optimal values rather than at any industry average, with the understanding that specialty drives real variance.

What is the difference between initial and final denial rate?

Initial denial rate is the percentage of claims denied on first pass. Final denial rate is the percentage still denied after all reworks and appeals are exhausted. Kodiak reported 11.81% and 2.8% respectively for 2024, so most denied claims are eventually paid, but only after rework that MGMA puts at an average of $25.20 per claim.

Do HFMA MAP Keys include target values?

No. MAP Keys publish formula definitions with a numerator, denominator, and data source. They do not publish numeric targets. The target values commonly quoted in the industry come from HFMA's separate KPI guidance and attributing them to a MAP Key is not accurate.

There are other sources of benchmarks available online, but please do your own research before taking them at face value.

Which metrics actually prevent denials?

Remember, everything rolls downhill. Prevention starts at registration. Registration and eligibility account for nearly 27% of denials, authorization and precertification for 11.6%, and roughly 50% of denials originate in the front office. HFMA defines Clean Claim Rate (CL-1), Insurance Verification Rate (PA-3), Service Authorization Rate (PA-4 and PA-5), and Total Charge Lag Days (PB-4) as the front-end measurements. They reduce how many denials arrive. They do not replace the denial and A/R metrics, which measure whether your denials workflow is working once a denial exists. Make sure your front end processes are clean, and that you are running a 97% or better clean claim rate.

What percentage of denials get appealed?

KFF found that 11.5% of denied Medicare Advantage prior authorization requests were appealed in 2024, and 80.7% of the appeals that were filed overturned the denial. In ACA marketplace plans, fewer than 1% of roughly 85 million in network denials were appealed.