Self-Pay Balances After an Emergency Department Visit

Self-pay balances in emergency medicine start from a worse position than in any other specialty. The patient did not choose the group, does not recognize its name on the statement, met the physician once for twenty minutes, and has no future appointment that gives them a reason to sort it out. Everything about the relationship works against collection, and none of it is the patient being difficult.

Most balance programs are built for practices where the patient is coming back. A primary care office can raise a balance at check-in. A surgical practice can settle it before the next stage. The conversation has a natural moment.

An emergency group has no such moment. It has a statement, a phone number, and a patient who is now home, possibly still unwell, and who may not connect the name on the envelope with the night they spent in an emergency department.

So the default program is statements followed by more statements followed by a collection agency, which recovers some money and costs the group goodwill it may need later.

What works better is not a harder collection process. It is making the first contact recognizable, making paying easy at the moment the patient is willing, and making a plan available before the balance becomes something the patient has decided to ignore.

The population is larger than the admitted patients

The scale of this problem follows directly from where emergency patients go after the visit. According to CDC data from the National Hospital Ambulatory Medical Care Survey, 11.5% of emergency department visits result in hospital admission.

That means the overwhelming majority of emergency encounters end with the patient going home the same day. Those are the accounts an emergency group is billing, and they are the ones with no downstream visit attached to them.

That single fact should shape the program. The group is not managing a small tail of complicated accounts. It is managing a very large number of one-time relationships, most of which will be resolved or lost in the first few weeks after the visit.

Volume at that scale also rules out certain answers. A program that depends on staff reaching each patient by phone will run out of staff long before it runs out of accounts, which is why so many groups skip straight to statements and then to an agency. The middle of that process is where the recoverable money actually sits.

Most of these patients are not refusing to pay

It helps to be accurate about who is on the other end of the statement, because the wrong assumption produces the wrong program.

KFF analysis of government data estimates that people in the United States owe at least $220 billion in medical debt, with approximately 14 million people, or 6% of adults, owing over $1,000 in medical debt and about 3 million people, or 1% of adults, owing more than $10,000.

That is a large population carrying manageable amounts and a much smaller one carrying amounts they cannot resolve at all. An emergency group’s self-pay file usually looks like the first group, which means the balances are collectable if the process makes paying simple and offers terms before the account goes cold.

It also argues against escalation as the primary tool. Pressure works on people who are choosing not to pay. Most of this file is people who intend to pay, do not have it this month, and have not been offered a way to spread it.

The practical version is that the plan offer should come early rather than as a last resort, and it should be presented as the normal option rather than as an exception the patient has to ask for.

Make the first contact recognizable

The reason so many emergency balance programs underperform is unglamorous. The patient does not know who is contacting them.

The group’s name is not the hospital’s name. The patient remembers the hospital. So the first contact has to establish, in a sentence, that this is the physician group that treated them at that hospital on that date, before it says anything about money. Contacts that fail this test get treated as a scam call, and a patient who has decided a caller is a scam does not answer the second one.

The second obstacle is the contact information itself. Demographics captured during an emergency visit are collected fast, from someone who is unwell, and frequently from a family member. Phone numbers are wrong, addresses are old, and the responsible party may not be the patient. A program that assumes the registration data is right will spend its budget contacting people who cannot be contacted.

So the first step is verification rather than collection. Confirm the contact route, correct it in the record where the patient tells you it is wrong, and find out whether the person you are speaking to is the one who handles the household’s bills.

That work is repetitive, it does not need a person, and it is what makes everything after it work. It also produces something durable, which is a corrected record for the next encounter.

The rules the conversation has to respect

Emergency billing carries protections that shape what the group may ask for, and getting this right is both compliance and credibility.

CMS publishes patient-facing guidance on medical bill rights covering surprise billing and balance billing protections, including protections that apply to emergency services. Patients increasingly know these exist, and a balance conversation that ignores them sounds like a group hoping the patient will not ask.

The operational consequence is that the process has to know which portion of a balance is genuinely the patient’s responsibility before it asks for it. That is a question about the account, not about the patient’s willingness, and it has to be settled first.

A related rule is that the group should be able to answer a dispute without the patient having to escalate. When someone says they believe the bill is wrong, the contact should capture the specific objection, route it to the person who can look at the claim, and stop the collection sequence while that happens. A dispute that keeps receiving automated reminders becomes a complaint.

None of this makes the program softer. A group that is precise about what it is owed, and visibly correct about what it is not, gets paid more reliably than one that sends the same statement four times.

Where the payment actually happens

The last mile is where balance programs quietly lose money, and it is the easiest part to fix.

The moment a patient agrees to pay is short. If paying requires calling back during business hours, finding a card, and reading numbers to a person, a meaningful share of that agreement evaporates. A link delivered while the patient is still holding their phone converts at a different rate than a promise to call back.

Payment plans need the same treatment. Terms offered during the conversation, set up immediately, confirmed in writing, and tracked afterward so a missed installment produces a friendly contact rather than a silent default. Inside athenaOne the plan, the receipts, and the payment history live on the patient record, so anyone who picks up the account later sees the same history the patient remembers.

Automation is well suited to all of it. Verifying contact routes, delivering a recognizable first message, offering plan terms, sending links, confirming enrollment, and following up on a missed installment are repeatable steps with clear outcomes.

What stays with people is the exception work. Hardship conversations, disputes with real substance, and the accounts where the right answer is charity care or a write-off rather than a plan. Those deserve a person, and the whole point of automating the rest is that a person is available for them.

Key Takeaways

Design for the patients who went home, since only about one in nine emergency visits ends in admission and the rest have no follow-up appointment.
Open the first contact by naming the group, the hospital, and the date of service, because an unrecognized caller gets treated as a scam.
Verify and correct contact details before collecting, since registration data captured during an emergency visit is frequently wrong.
Offer plan terms early and as the normal option, because most of the file is people who intend to pay and have not been offered a way.
Settle what the patient actually owes under balance billing protections before asking for it.
Pause the sequence when a patient disputes a bill, capture the specific objection, and route it to someone who can review the claim.
Deliver a payment link during the conversation rather than asking the patient to call back, and set up plans on the spot.

An emergency group is collecting from people with no reason to remember it, no next appointment, and no obvious way to ask a question. Fix the recognition problem, fix the contact data, offer terms early, and make paying take seconds. That is most of the recoverable money, and none of it requires being harder on patients.

Sources