What Is A/R in Medical Billing?

Accounts Receivable in medical billing is the sum of all charges that have been billed but not yet collected.

A/R includes:

• Claims at the payer awaiting adjudication
• Claims paid by payer with patient responsibility outstanding
• Claims in denial workflow (denied, being appealed)
• Claims aging in collections workflow
• Patient balances aged into self-pay status

A/R does NOT include:

• Charges not yet billed (those are 'unbilled charges' — a separate metric)
• Charges that have been written off (those are no longer receivable)
• Contractual adjustments already taken (those reduced the receivable)
• Estimated future charges from scheduled visits

Understanding what is and isn't in A/R matters because practices often look at the wrong number — total charges minus total payments includes write-offs and unbilled, which paints a misleading picture. The correct A/R number for management decisions is billed but not yet collected.

Why A/R deserves attention. A/R is usually among the largest assets on a physician practice's balance sheet, and days in A/R converts directly into working capital. Illustrative arithmetic, not a measured figure: a practice billing $1,000,000 of charges a year averages $2,740 of charges a day ($1,000,000 ÷ 365); at 35 days in A/R it is carrying roughly $95,900 of billed-but-uncollected charges at any moment ($2,740 × 35). Note the basis — days in A/R is computed on charges, so a working-capital estimate built from collections instead will not reconcile with it.

Days in A/R is the average number of days it takes to collect billed charges.

What is actually published. The only freely-available published target for physician practices comes from AAFP: days in A/R "should stay below 50 days at minimum; however, 30 to 40 days is preferable." That is practice-management guidance — AAFP states no population, sample or data year behind it, so it is a target, not a measured median. No free source publishes an observed median days in A/R for physician practices at all; MGMA collects the figure but its medians are licensed and not public. Treat any specific "industry median" of 35 or 47 days you see quoted without a source as unverified, and read your own number against your own trend and payer mix rather than against a number nobody can show you the sample for.

The standard calculation:

Days in A/R = Total A/R balance ÷ (Total billed charges ÷ Days in period)

Worked example (illustrative figures, arithmetic reproduces). A practice has $300,000 in A/R. In the prior 90 days it billed $1,200,000 in charges. Daily charge average = $1,200,000 ÷ 90 = $13,333/day. Days in A/R = $300,000 ÷ $13,333 = 22.5 days.

Hold the denominator constant and the same practice at $500,000 of A/R is at $500,000 ÷ $13,333 = 37.5 days, and at $700,000 it is at 52.5 days — the first inside AAFP's preferable range, the second past AAFP's stated floor of below 50.

Alternate calculations. Some sources use 'allowed amounts' instead of billed charges, especially when contractual write-offs are large. The result is more conservative but less commonly reported. Some sources report 'days in A/R over 90' separately to distinguish the aged portion.

Days in A/R is the single best summary metric for revenue cycle effectiveness because it captures the cumulative result of every other process: charge lag, claim quality, denial work, accounts receivable follow-up, and patient collections all flow into it. A practice with 25-day A/R is doing many things right. A practice with 60-day A/R is doing many things wrong.

A/R is segmented into aging buckets based on days since claim submission. Each bucket carries a different operational meaning and a different action.

On the share each bucket "should" hold, we state the blank. No freely-available primary source publishes a target distribution of A/R across aging buckets, or a target share of A/R over 90 days, for physician practices. The percentages that circulate as benchmarks — "under 15% over 90 days" and similar — could not be traced to a published sample, so this page does not present them as one. What is diagnostic is your own distribution and its movement quarter over quarter, which is why the pattern reading below matters more than any single threshold.

Why the distribution matters. The shape of the distribution tells you what kind of problem the practice has. Examples:

• High 0–30 share, low aged share: Healthy. Claims are flowing, follow-up is happening, aged claims aren't accumulating.
• High 0–30 share, high 91+ share: Current claims work is good, but the practice abandoned aged claims long ago. There's recoverable revenue stuck.
• Low 0–30 share, high 31–60 share: Submission is slow (charge lag) — claims aren't being submitted fast enough to populate the early bucket.
• High 91–120, low 120+: Practice writes off at exactly 120 days. Operational discipline exists but the threshold is wrong — many of those claims could be recovered.
• Even distribution across all buckets: No one is working A/R systematically. Claims age uniformly because nothing happens to them.

The distribution is more diagnostic than the headline number.

Claims do collect at decreasing rates as they age. The tables of "collection probability by aging bucket" that circulate in this industry — 95–98% at 0–30 days, 25–40% past 180 — are quoted constantly and could not be traced to any published study or sample, so this page does not reproduce them. What can be stated without inventing a number is the mechanism, and the mechanism is enough to act on.

Aged A/R loses revenue for four concrete reasons, not because of a curve:

1. Appeal deadlines close. Every payer sets its own window to file a first-level appeal after the remittance date, and the contract or the plan document is the only place it is stated. Once it passes, the dollars are gone regardless of the merits.
2. Timely-filing windows close. A claim that has to be corrected and resubmitted — or that was sent to the wrong payer — is racing a separate, usually shorter clock that also varies by payer and by contract.
3. Documentation gets harder to retrieve. Records requests, prior-auth evidence and encounter detail all cost more staff time to assemble months later than they did the week of the encounter.
4. Patient balances decay into bad debt. A patient balance that has sat through several statement cycles without contact is materially harder to collect than a fresh one, and past a practice's own bad-debt threshold it is usually written off rather than worked.

The practical implication is unchanged by the missing data. Days in A/R is not only a cash-timing metric — the aged tail of A/R is where the write-offs come from, so working the tail is a revenue decision rather than a scheduling one. The right way to size it for your practice is not a published curve but your own history: pull twelve months of closed claims, group them by the aging bucket they were in when they finally paid or were written off, and you have a recovery curve built on your payers and your contracts. That is the only version of this table anyone can legitimately show you.

Six factors push A/R days up. Each one is fixable, and each one is measurable in your own data — we give the mechanism and the measurement rather than a dollar figure, because the published per-practice cost estimates that circulate for these are not traceable to any sample.

1. Charge lag. Every additional day between encounter and submission adds a day to A/R. A practice at 6-day charge lag carries 4 more days of A/R than one at 2 days. Measure it: median days from date of service to claim submission, by provider. Multiply the excess days by your average daily charges to size the working capital it ties up.

2. Denial backlog. Unworked denials sit in A/R indefinitely and age straight past the appeal deadlines above. Measure it: count and dollar value of denied claims with no action logged, bucketed by days since remittance and flagged against each payer's appeal window.

3. Payer-specific deterioration. When a payer slows adjudication, A/R aging follows, and an aggregate KPI hides it for as long as the rest of the book stays healthy. Measure it: days in A/R and over-90 share broken out by payer, monthly — a single payer's trend line is the early warning the blended number cannot give you.

4. Patient balance accumulation. Patient responsibility takes longer to collect than payer responsibility and decays faster without an active workflow, so it drifts toward bad debt. Measure it: patient A/R aging tracked separately from insurance A/R, plus your own patient-pay rate and bad-debt write-off rate by statement cycle. We do not publish a national figure for the share of practice revenue that is patient responsibility, because we could not source one.

5. Eligibility errors. Claims returned or denied for coverage issues stay in A/R through the whole rework cycle. Measure it: count of eligibility-related rejections and denials, and the median age at which each one finally resolves against your baseline for a clean first submission.

6. Prior auth gaps. Claims for services performed without a confirmed authorization get denied and then sit pending appeal. Measure it: authorization-related denial dollars and the median days from denial to final resolution.

Diagnosing high A/R requires breaking it down by category — a single 'high A/R' number doesn't tell you what to fix.

The fastest path to lower A/R days, in priority order:

1. Charge lag discipline. Same-business-day charge entry standard. Reduces A/R by 1–3 days immediately. Effort: low (process change). Impact: large.

2. Daily ERA review. Denials worked within 24 hours of identification, not weekly or monthly. Prevents 60-day accumulation. Effort: medium (workflow change). Impact: large.

3. Aged A/R triage at 31, 46, 61, and 91 days. Every claim reviewed weekly with action assigned. Prevents accumulation in over-90 bucket. Effort: medium (staffing model change). Impact: large.

4. Patient balance workflow. Statements within 7 days of payer payment, payment plans for larger balances, soft-touch calls before bad debt. Effort: medium-high (system + staffing). Impact: very large — and the one whose size you should measure on your own book rather than from a published figure.

5. Eligibility verification before every visit. Prevents rejection-cycle additions to A/R. Effort: low (workflow). Impact: large — size it from your own eligibility-related denial count.

6. Prior auth tracking. No claim submitted without confirmed auth. Effort: low-medium (scheduling integration). Impact: medium-large.

These six are ordered by the ratio of impact to effort, not by a promised result. We do not publish a typical days-reduction or dollar figure for implementing them, because no traceable source supports one and the honest answer depends on where your current number actually comes from — a practice whose A/R is high because of charge lag and one whose A/R is high because of an unworked 90+ tail respond to entirely different work at entirely different speeds.

Insurance A/R and patient A/R behave differently and need different workflows.

Insurance A/R.

• High volume, smaller balance per claim
• Adjudication windows that are knowable per payer rather than universal. Medicare is the one that is actually published: a clean claim sits behind a payment floor of 13 days for electronic claims (so the earliest payment date is the 14th day after receipt) and 26 days for paper, and must be paid or denied within a payment ceiling of 30 calendar days from receipt or the contractor owes interest (Medicare Claims Processing Manual, Chapter 1 §§80.2.1.1–80.2.1.2, retrieved 17 September 2026). Commercial and Medicaid windows are set by contract and by state prompt-pay law — look yours up rather than assuming a national average
• Recovery driven by claim quality, denial work, payer follow-up
• Workflow tools: ERA review, denial worklists, payer-specific follow-up

Patient A/R.

• Lower volume, larger balance per account
• Pay cycles measured in statement cycles rather than adjudication windows, and slower than insurance A/R even when the workflow is working
• Recovery driven by statement cycles, payment options, soft-touch calls
• Workflow tools: patient billing software, online payment portal, payment plans
• Bad-debt threshold is a policy you set, not an industry constant — write it down, apply it consistently, and measure your own median days to full collection against it

Different metrics matter. For insurance A/R, the key metrics are days-in-A/R, denial rate, aged A/R. For patient A/R, the key metrics are patient pay rate (% of patient responsibility collected), bad debt rate (% written off), and patient receivable cycle time.

Why splitting them matters. A blended A/R number hides which half is failing. Patient A/R needs different tools, different staffing and different KPIs than insurance follow-up, and a practice that reports only the blended figure will keep resourcing the half that is already working.

Best practice. Track and report insurance A/R and patient A/R separately. Set separate aging targets, separate dollar targets, and separate workflows. Many PM systems make this difficult by default — invest in segmentation.

Most practices have an aged A/R backlog they've never systematically worked. Recovering it is one of the highest-ROI projects in revenue cycle.

Typical backlog characteristics.

• Claims aged 90–365 days
• Denied claims that were never appealed
• Patient balances aged into self-pay
• Underpaid claims (paid below contracted rate, never pursued)
• Claims sent to wrong payer (still recoverable if within timely-filing window)

Recovery approach.

Phase 1 (Days 1–30): Triage.

• Pull all claims aged 90+ days
• Categorize: payer name, denial reason, dollar value, appeal window status
• Identify the 'pursuable' subset (within appeal/timely filing window)

Phase 2 (Days 31–90): Active recovery.

• Work pursuable claims by dollar value, highest first
• Standard playbook: review denial reason, gather supporting documentation, file appeal or correct-and-resubmit
• Track recovery rate by category

Phase 3 (Days 91+): Document and write off.

• Claims outside pursuable window: document write-off reason, retain in compliance file
• Patient balances over 120 days without contact: third-party collection referral or write-off
• Generate aging-cleanup report for ownership

On recovery rates, we state the blank. The "25–40% of pursuable dollars" figure that circulates for aged-A/R cleanup could not be traced to any published sample, so this page does not present it as a benchmark or as a result you should expect. What determines your actual recovery is knowable before you start: how much of the backlog is still inside an appeal or timely-filing window, how it splits between insurance and patient balances, and how much of it is underpayment rather than non-payment. Phase 1 above exists to produce exactly those three numbers — run it and you have a sized estimate built on your own claims instead of someone else's average.

Sizing the decision. A backlog cleanup has fixed triage overhead regardless of size, so below some threshold the work costs more than it returns. Compute that threshold rather than inheriting one: estimated pursuable dollars × your own historical recovery rate on appeals, against the staff hours the triage and the appeals will take.

Backlog recovery is a different engagement from ongoing follow-up. Recurring A/R follow-up works the current book so a backlog never forms; a legacy recovery project is a one-time, time-boxed pass at claims that have already aged past normal workflow. They have different scopes, different economics and different exit conditions — do not buy one expecting the other.

Patient A/R is the workflow most often left without an owner. Best practices:

1. Communicate cost before the visit. Real-time eligibility shows deductible status, copay, and likely out-of-pocket. Deliver this to the patient at scheduling — not at check-out, not after the visit.

2. Collect at the point of service when possible. Copays, known deductibles, prior balances. Even 50% point-of-service collection on patient responsibility is dramatically more recoverable than 0%.

3. Statement cycles within 7 days of payer payment. Don't wait until month-end. The faster the statement, the higher the pay rate.

4. Online payment portal with mobile support. Card, ACH and mobile wallets. Removing a payment method the patient already has set up removes a reason to defer; measure the lift on your own pay rate rather than from a vendor's figure.

5. Payment plans above a stated balance threshold. Short interest-free plans, with the threshold written into policy rather than decided case by case.

6. Soft-touch calls at 30 and 60 days. Before any aggressive collection escalation. Most balances are paid after a polite reminder call.

7. A written bad-debt threshold. Whatever number you choose, apply it consistently and document it. Either third-party collection referral or write-off — not indefinite holding.

8. No surprise billing. No Surprises Act compliance: good-faith estimates for self-pay, balance billing protections, dispute resolution process. Compliance failure has financial penalty AND drives bad debt.

The percentages usually attached to this list — "75–85% of patient responsibility with a workflow, 50–65% without" — could not be traced to a published sample, so we do not repeat them. Set your own baseline first: patient-pay rate and bad-debt write-off rate for the last twelve months. Then implement the list and measure the same two numbers again. That comparison is yours, and it is the only one that will survive a question about where the figure came from.