AR Follow-Up in Medical Billing: Prioritizing Payer Follow-Up

TL;DR

• Working AR oldest-first feels disciplined but wastes touches. Rank claims by timely filing risk, dollar value, payer response window and claim state (no response, pended, denied) instead.
• Use aging buckets (0 to 30, 31 to 60, 61 to 90, 91 to 120, over 120 days) to see the shape of the problem, not as the work order itself.
• Every follow-up touch should end in structured fields: status, status codes, reference number, what the payer needs, who owns the next action and when to touch it again.
• Check status electronically (276/277) or in the portal first, and save phone calls for claims where the electronic answer is missing, generic or contradictory.
• Measure the queue with days in A/R and aged A/R share, plus internal metrics such as touches per claim and first-touch resolution, reviewed by payer.

"Work the 120+ bucket first" is the most common AR follow-up instruction and one of the least useful. The oldest claims are not always the most at risk, the most valuable or the most fixable. A $90 claim at 140 days that the payer already denied for a non-covered service is less urgent than a $4,000 claim at 75 days that the payer cannot find and that is closing in on its filing limit.

AR follow-up in medical billing is the work of finding out why billed insurance claims have not been paid and moving each one to a resolution: payment, a corrected claim, an appeal or a justified write-off. At volume, the hard part is not making calls. It is deciding which claim deserves the next touch and making sure each touch produces something the next person can use.

Why oldest-first breaks at volume

Aging is a reasonable proxy for risk, but only a proxy. A claim's age tells you how long it has been open, not whether the payer has it, whether anything is owed by you, or how close it is to a deadline that cannot be undone.

Oldest-first queues also tend to concentrate effort where returns are lowest. Claims at the far end of aging often include ones already denied, already appealed or already judged uncollectible, so staff can spend their best hours re-confirming bad news. Meanwhile, younger claims that could be fixed cheaply drift into the same state.

The alternative is a scored queue: aging is one input among several, and the queue order comes from the combination.

Building a prioritized AR follow-up queue

A practical priority score combines five factors. The weights depend on your clients and payer mix, but the inputs are fairly standard.

Timely filing risk comes first because it is the one factor that turns a collectible claim into a permanent loss. For traditional Medicare, 42 CFR 424.44 generally requires claims to be filed "no later than the close of the period ending 1 calendar year after the date of service," with limited exceptions. Commercial and Medicaid filing limits are set separately and differ by payer, plan and contract. A claim that is "not on file" with the payer is the classic trap: it looks like a slow claim until someone discovers it never arrived. Our guide to timely filing deadlines covers how to build deadline awareness into the queue.

Payer response windows keep you from wasting touches. For Medicare, the Claims Processing Manual sets a payment floor: HIPAA-compliant electronic claims are not paid before the 14th day after receipt and paper claims not before the 29th, while clean claims must be paid or denied within 30 days. Claims that need development have a 45-day processing standard. For Medicare Advantage, 42 CFR 422.520 sets timing for non-contracted claims and claims from MA private fee-for-service enrollees, but for contracted providers the prompt payment terms are whatever the contract says. Commercial windows depend on state law and contracts. Touching a claim before its window has passed usually produces nothing more than "still in process."

Using aging buckets without letting them run the queue

HFMA's MAP Keys define aged A/R in five mutually exclusive buckets: 0 to 30, 31 to 60, 61 to 90, 91 to 120 and more than 120 days, aged from discharge date for hospitals or date of service for ambulatory and physician settings. HFMA describes its AR-1 key, aged A/R as a percentage of billed A/R, as a "trending indicator of receivable aging and collectability," and its AR-2 key breaks the same buckets out by payer group.

That framing is the right one. Buckets tell you where the inventory is accumulating and which payers are slow. They are a reporting lens. Within each bucket, the scored factors above should decide order. A useful habit is to review bucket movement weekly by payer: if one payer's 61 to 90 bucket is growing while others are flat, that is usually a systemic issue (an enrollment problem, a new edit, a payer backlog) worth one escalation rather than hundreds of individual calls.

Treat denials, pends and no-response claims as different queues

Mixing claim states in one queue is a common source of wasted effort, because each state needs a different person and a different action.

• No response or not on file. The question is whether the payer has the claim. X12's claim status categories distinguish receipt (A1, which "does not mean that the claim has been accepted for adjudication"), acceptance into adjudication (A2) and not found (A4). These need fast action because of filing limits.
• Pended. The payer has the claim and is waiting on something. X12 category P3 means the payer is waiting for information "already been requested from the provider," which makes it an active task with a due date. Other pending categories are usually recheck items.
• Denied. The claim is finalized (X12 F2). The work is denial analysis: correct and resubmit, reconsider, appeal or write off. That belongs with staff who know the payer's appeal rules.
• Paid but short. Underpayments need contract and remittance review, not a status call.

Our explainer on 277 claim status codes explains the category system in more detail. Routing on the code, not on a free-text note, is what makes these queues workable.

What every follow-up touch should capture

A touch that ends with a note like "called UHC, still processing" has to be repeated by the next person. A touch that ends in structured fields does not. At minimum, capture:

The next-touch date is the field most teams skip and the one that matters most for queue health. It should follow from the payer's response window or the deadline the payer gave, not from a default 30-day recall.

Choosing the channel: 276/277, portal or phone

HIPAA defines the health care claim status transaction in 45 CFR 162.1401 as an inquiry from a provider to a health plan "to determine the status of a health care claim," and a response from the plan. Under 45 CFR 162.1403, HHS adopted the CAQH CORE Phase II claim status operating rules for that transaction. In practice, the 276/277 is the cheapest way to sweep a large queue for status changes.

It is not always enough. In the 2024 CAQH Index, CAQH CORE reported that one reason providers rely on manual processes is that "they receive more robust information through these processes than through the ASC X12N 277 Claim Status response transaction." The same report, based on 2023 data, found that conducting a claim status inquiry by phone took medical providers and staff 25 minutes per inquiry, the highest of the administrative tasks it measured. That is the trade-off in one sentence: manual inquiries such as phone calls can return fuller answers, and phone inquiries cost far more time.

A sensible sequence:

1. Run 276/277 inquiries across the queue on a schedule to catch status changes and remove claims that have finalized.
2. Check the portal for claims that need documents, letters or line-level detail the 277 did not return.
3. Call when the electronic response is generic, contradictory or missing (for example, X12 category E2, "Information Holder is not responding; resubmit at a later time"), or when you need something only a representative can do, such as reprocessing or a reference number for an escalation.

Escalation and exception handling

Some claims should leave the standard queue entirely. Define the triggers in advance so they do not depend on individual judgment:

• Repeat touches without progress. A claim with several touches and the same status needs a supervisor or a payer escalation, not another identical call.
• Systemic patterns. The same denial reason or pend across many claims for one payer points to an upstream fix (enrollment, coding, an edit) and a single payer contact.
• Approaching deadlines. Claims within a set number of days of a filing or appeal limit should jump the queue automatically.
• Payer commitments that lapse. If a representative promised reprocessing by a date and it did not happen, escalate with the reference number from the original call.

Keep exceptions visible in their own view. When they sit inside the general queue, they get touched like everything else and never resolved.

Staffing capacity and measurement

Capacity planning starts with the queue, not headcount: claims needing a touch this week, multiplied by the expected touches to resolution, divided by touches per staff hour by channel. Phone touches and electronic touches belong in separate lines, because the time per touch is so different.

For outcome measures, HFMA's net days in A/R is net A/R divided by average daily net patient service revenue, which HFMA calls a "trending indicator of overall A/R performance." MGMA describes days in A/R as total A/R divided by average daily charges, so be consistent within a client and over time; our analysis of days in A/R by practice size goes into why the benchmarks vary. Aged A/R share (the percentage of billed A/R over 90 or over 120 days, using the HFMA buckets) shows whether old inventory is shrinking.

Two operational metrics complete the picture, though they are internal definitions rather than industry standards:

• Touches per resolved claim. Rising touches per claim at one payer usually means poor information quality from that payer or a routing problem on your side.
• First-touch resolution. The share of claims where the first touch produced a definitive status and next action. Low first-touch resolution often points to missing fields or the wrong channel.

Review all of these by payer and by client. An overall number hides the payer that is driving the backlog.

Where automation fits

Automation helps most with the repetitive retrieval work: sweeping status electronically, checking portals and making routine status calls, then writing results back as structured fields. It does not replace the people who decide what to do with an ambiguous answer, write an appeal, handle a payer dispute or manage a client relationship.

SuperDial's claim status workflow automates that retrieval layer. It routes each request through the fastest channel available (EDI, API, payer portal, then a phone call) and returns status, payment details, denial reasons, reference numbers and next actions in a consistent format, so AR teams can spend their time on the exceptions.

A weekly review that keeps the queue honest

A short weekly review by payer catches most problems before they become write-offs:

• Which claims are within 30 days of a filing or appeal limit, and does each one have an owner?
• Which payer's buckets moved in the wrong direction, and is there a single cause?
• Which claims have three or more touches with no change in status?
• Which next-touch dates are already past?
• What share of touches this week ended with every required field filled in?

If the answers come from structured data rather than notes, the review takes minutes. If they do not, fixing the touch record is the first improvement to make.

FAQ

What does an AR caller do in medical billing?

An AR caller contacts insurance payers about unpaid or underpaid claims to find out their status, what the payer needs and how to resolve them. The role usually also includes recording the result, requesting reprocessing and routing denials to the right follow-up path.

What are AR aging buckets?

They are ranges of how long a receivable has been outstanding. HFMA's standard buckets are 0 to 30, 31 to 60, 61 to 90, 91 to 120 and more than 120 days, aged from date of service for physician and ambulatory settings or discharge date for hospitals.

How soon should you follow up on an unpaid insurance claim?

After the payer's normal processing window has passed, and sooner if there is any doubt the payer received the claim. For Medicare, a clean electronic claim is not paid before the 14th day after receipt, and clean claims must be paid or denied within 30 days. Commercial timing depends on state law and your contract.

Should AR follow-up start with the oldest claims?

Not by default. Age matters, but timely filing risk, dollar value and claim state are better predictors of which touch will recover money. Use aging to report and to spot payer trends, and use a scored queue to decide order.

Sources

• Healthcare Financial Management Association, MAP Keys: Indicators of Revenue Cycle Excellence, 2023 Edition: https://www.hfma.org/wp-content/uploads/2023/10/2023-Edition-MAP-Keys.pdf
• Medical Group Management Association, MGMA Stat: Days in A/R holds steady for most practices, but payer pressure persists in 2026 (July 30, 2026): https://www.mgma.com/mgma-stat/days-in-a/r-holds-steady-for-most-practices-but-payer-pressure-persists-in-2026
• eCFR, 42 CFR 424.44, Time limits for filing claims (current as of October 27, 2026): https://www.ecfr.gov/current/title-42/section-424.44
• eCFR, 42 CFR 422.520, Prompt payment by MA organization (current as of October 27, 2026): https://www.ecfr.gov/current/title-42/section-422.520
• eCFR, 45 CFR 162.1401, Health care claim status transaction: https://www.ecfr.gov/current/title-45/section-162.1401
• eCFR, 45 CFR 162.1403, Operating rules for health care claim status transaction: https://www.ecfr.gov/current/title-45/section-162.1403
• Centers for Medicare & Medicaid Services, Medicare Claims Processing Manual, Chapter 1: General Billing Requirements (Rev. 13725, issued April 10, 2026): https://www.cms.gov/regulations-and-guidance/guidance/manuals/downloads/clm104c01.pdf
• X12, Claim Status Category Codes (code list 507): https://x12.org/codes/claim-status-category-codes
• CAQH, 2024 CAQH Index Report: From Transactions to Trust: https://www.caqh.org/hubfs/Index/2024%20Index%20Report/CAQH_IndexReport_2024_FINAL.pdf
• SuperDial, Automated Claim Status Verification: https://www.superdial.com/workflows/claim-status
• SuperDial, RCM Companies: https://www.superdial.com/rcm-companies