A payer changes what it pays for the same procedure depending on where the work happened, and that one fact sits under everything below. So the plain definition comes first. Where the code sits on a claim form is the next practical question, and the answer is narrower than most people expect. One number covers most of what a typical medical practice bills, though the rest of the core set is worth knowing by sight. Medicare's payment amount moves with the setting through a mechanism you can read in the fee schedule itself. Telehealth split into a pair of entries once the patient's location started to matter, and a visit by video now lands on one or the other. Care delivered inside a patient's own home is a separate case again. A type of bill does a related job on institutional claims, and the two get mixed up constantly. What happens after a wrong entry runs from a same-day rejection to a refund request months later. A biller checks the entry before a claim goes out, and that doesn't take long. CMS keeps the list current, and the update path into your billing system matters as much as the list does. What the code never tells a payer is the part most practices skip. Where these facts come from closes the page.
Why does a place of service code change what a payer pays?
A place of service code changes what a payer pays because it answers who carried the overhead for the visit. Seeing a patient in her own clinic, a physician pays for that room, the staff in it, the gloves, the autoclave and the rent. That same physician, rounding on a hospital floor, pays for none of it, and the hospital sends its own bill for that side of the encounter.
Payers price the professional service around that split. Two payments for one encounter would be double payment, so the setting decides which share of the money rides on the physician's claim. The Centers for Medicare and Medicaid Services publishes the rules behind that arrangement in its coding and billing material, and commercial plans build their own policies on the same logic.
That's the whole reason the field exists. It isn't a filing detail or a statistic somebody collects. Payers read it as a claim about economics, and price the line to match.
What is a place of service code?
A place of service code is a two-digit numeric code on a professional claim that identifies the setting where a service was furnished. CMS maintains the code set. Medicare, state Medicaid programs and commercial payers read the same values, which is what makes the set usable across a practice's whole payer mix rather than one plan at a time.
Three things it isn't, and each confusion shows up in real claims. It isn't a procedure code, so it never says what the clinician did. Nor is it a diagnosis code, which leaves the reason for the visit somewhere else entirely. Provider identity is absent too, so the two digits say nothing about who delivered the care or under what specialty.
What the code does carry is a piece of context a payer can't get anywhere else on the form. Setting is invisible in a procedure code. Without the two digits, a hospital visit and an office visit don't look any different on paper.
Where does a place of service code go on a claim?
A place of service code goes in item 24B of the CMS-1500 claim form, and in the matching service-line segment of the 837P electronic professional claim. That box number matters less than the placement. It's that the field lives on the service line rather than in the claim header.
Line-level placement has a consequence practices run into the first time a provider works two settings in a day. One claim can carry several lines with different entries, because each line describes its own encounter. Physicians who round at the hospital in the morning and see patients in the clinic that afternoon don't need a second claim for the pair.
Institutional claims work differently and don't carry the field at all. Hospitals, skilled nursing facilities and home health agencies bill on the UB-04 or its electronic 837I equivalent, where the facility's identity is reported another way. Only the professional claim asks this particular question.
Which place of service code does a practice use most?
Most outpatient practices use place of service code 11, the office, on more claim lines than every other code combined. Anything delivered in the practice's own space, under the practice's own overhead, lands there by default. Beyond that, the core set is small enough to learn in an afternoon, and a biller who knows these ten by sight catches most errors before a claim leaves the building. Everything else turns up rarely enough that it's faster to look up than to memorize. Practices building their billing function from scratch can start with our medical billing guide.
The core place of service codes a typical outpatient practice touches
Code
Setting it names
When a practice reaches for it
02
Telehealth delivered somewhere other than the patient's home
A video or phone visit with the patient at work, in a clinic or traveling
10
Telehealth delivered in the patient's home
A video or phone visit with the patient at home
11
Office
Routine visits in the practice's own space
12
Home
A clinician physically visiting the patient's residence
19
Off campus outpatient hospital
Hospital-owned outpatient sites away from the main campus
21
Inpatient hospital
Rounding on an admitted patient
22
On campus outpatient hospital
Hospital outpatient departments on the main campus
23
Emergency room in a hospital
Emergency department encounters billed professionally
31
Skilled nursing facility
Visits to a patient in a skilled nursing stay
32
Nursing facility
Visits to a long-term nursing resident
Descriptors on the CMS code set page are worded more formally than the table above, so code 10 reads "Telehealth Provided in Patient's Home" there, and the full list runs well past these ten.
How does a place of service code change a Medicare payment amount?
A place of service code changes a Medicare payment amount by selecting which practice expense value the Medicare Physician Fee Schedule applies to the procedure. Every procedure code on the schedule carries two of them. One is the non-facility value, used when the practice supplies the room and the staff. Its pair is the facility value, used when a hospital or similar site supplies them instead and bills separately.
Office and home settings draw the non-facility value. Hospital settings draw the facility one, which is lower for most services a clinician could also deliver in an office, because the overhead has moved onto somebody else's claim.
The gap between the pair is set per procedure code rather than by a single percentage, so there's no rule of thumb that replaces looking it up. A minor office procedure with disposable supplies can show a wide split. Talk-only visits show almost none. Current values are published by CMS in the fee schedule, and they move by rule year.
Which place of service code applies to a telehealth visit?
A telehealth visit takes place of service code 10 when the patient is at home and code 02 when the patient is anywhere else. Where the patient was sitting decides it, which catches people out, because the instinct is to describe where the clinician was.
Payer policy is the second half of the answer, and it's the half that changes. Some plans want one of those two codes on the line. Others ask for the office code with a telehealth modifier attached, and a few treat audio-only encounters under different terms than video. Reading the plan's current telehealth policy beats carrying last year's habit forward, because the rules have moved more than once since remote visits became routine.
Practices running a mixed schedule should watch one specific failure. A patient who books a video visit and then arrives in person leaves the telehealth entry sitting on the line unless somebody corrects it, and nothing in the record flags the mismatch on its own, so it's on the biller to catch.
Which place of service code applies to care in a patient's home?
Care physically delivered in a patient's home takes place of service code 12. Clinicians drive to the residence, examine the patient there and bill the professional claim with that setting on the line. House calls, home-based primary care and in-home therapy visits all sit in the same box.
Code 12 and code 10 describe different events even though both involve a house, and they aren't interchangeable. One means the clinician was in the home. Its counterpart means the patient was at home while the clinician was somewhere else entirely. Mixing them up is the most common home-care error a biller sees, and it's an easy one to make.
Residential settings other than a private home carry codes of their own, including assisted living and group residential arrangements, so a visit to a patient in one of those isn't automatically code 12. Home health agency billing is a separate matter again, since the agency files an institutional claim while the visiting physician files a professional one.
How does a place of service code differ from a type of bill?
A place of service code differs from a type of bill in which claim it rides on and which party it describes, and the two are never interchangeable. One hospital stay generates both at once, which is exactly why the pair gets confused. Both answer a location question, and neither answers the other's version of it. The physician's professional claim carries the two-digit setting, while the hospital's institutional claim carries a type of bill instead.
Place of service code and type of bill compared on the fields that matter
Attribute
Place of service code
Type of bill
Claim it appears on
CMS-1500 and the 837P professional claim
UB-04 and the 837I institutional claim
Who files it
Physicians and other professional billers
Hospitals, skilled nursing facilities, home health and hospice agencies
What it reports
The setting where the patient received the service
The kind of facility, the classification of the bill and its sequence in a series
Where it sits
On each service line
On the claim as a whole
Keeping the pair apart is the practical takeaway, and it's mostly a staffing question rather than a coding one. Billers working professional claims never file a type of bill, and billers working institutional claims never file a setting code, so a team covering both runs the two workflows separately on purpose.
What happens when a place of service code is wrong?
A wrong place of service code produces one of four outcomes, and only the first one is cheap.
A front-end rejection, where the clearinghouse or the payer bounces the claim before anyone adjudicates it.
A denial after adjudication, which drops the claim into an appeals queue and pushes payment out by weeks.
Payment at the wrong rate, where the claim pays but pays the facility amount instead of the office amount, or the reverse.
A post-payment recoupment, when a payer reviews the paid claim later and asks for the difference back.
Silent underpayment is the one that hurts. Rejections announce themselves the same day, while a line that paid at the facility rate when the office rate applied looks like a clean claim on every report a practice runs.
Denials aren't getting any easier across the board. Experian Health's "State of Claims" survey of 250 healthcare professionals found 41% of providers reporting denial rates of 10% or higher (Source: Experian Health, 2025). Practices working that queue can start with our guide on how to reduce claim denials.
How does a biller check a place of service code before a claim goes out?
A biller checks a place of service code by reading the encounter back against where the patient was, where the provider was and what the payer's policy says. That check takes seconds per line once it's routine. Four sources answer it.
The encounter record, read for the patient's location and the provider's, the pair the code has to match.
The payer's policy for the current plan year, since one plan wants the telehealth code and another the office code with a modifier.
The claim scrubber's edits, which catch a facility code on a line billed from the practice's own clinic.
The practice management system's stored code table, checked against the current CMS set rather than assumed.
Somebody's got to own that check. Plenty of practices hand it to a remote biller in their existing system, and Honest Taskers places HIPAA-trained virtual medical billers at $10.00 to $12.65 an hour, under a Business Associate Agreement signed when the professional will access protected health information. Coding accuracy sits beside this work, and our medical coder guide covers that side.
Who keeps the place of service code list current?
The Centers for Medicare and Medicaid Services keeps the place of service code list current, publishing the code set and the descriptor attached to each value. Payers adopt what CMS publishes rather than writing their own numbering, which is why one set works across Medicare, Medicaid and commercial plans.
Changes reach a practice through two channels, and they don't move at the same speed. Payer bulletins arrive first and describe what a plan expects. Stored code tables inside a practice management system arrive whenever the vendor ships an update, and a stale table quietly offers the old descriptor forever. Anyone who's watched a retired value sit in a dropdown for a year knows how that ends.
Reading it from the source doesn't cost anything. Current descriptors are published by the Centers for Medicare and Medicaid Services at CMS, and checking that page beats trusting a copy somebody pasted into a training document in 2023.
What does a place of service code not tell a payer?
A place of service code tells a payer nothing about what was done, who did it, or whether it needed doing. Each of those answers comes from a different field on the same claim, and the setting code doesn't have an opinion on any of them.
Procedure codes report the service. Diagnosis codes report the reason. Provider identifiers and taxonomy report who delivered the care and in what specialty, while modifiers report the circumstances around it. Medical necessity rests on the diagnosis codes and the documentation behind them, so a correct setting on an unsupported claim still gets denied.
Here's the limitation worth writing down. Status inside a setting is the gap, so the two digits cannot tell a payer whether an admitted-looking encounter was inpatient or observation, or whether the room the patient occupied was billed as a hospital bed. Those questions get answered by the facility's own claim and by the documentation, which means a clean setting code isn't on its own proof that a claim is right.
Where do these place of service code facts come from?
The code set, the descriptors and the fee schedule mechanism come from the Centers for Medicare and Medicaid Services, whose coding and billing material is the published source for all three and the version that governs when a vendor's stored table disagrees. Denial context is Experian Health's "State of Claims" survey of 250 healthcare professionals, fielded in 2025 (Source: Experian Health, 2025). Honest Taskers rates, HIPAA training and Business Associate Agreement practice come from the company's own published service terms and rate card, read in September 2026.
Numbers deliberately left out matter as much as the ones above. No percentage difference between facility and non-facility payment appears anywhere on this page, because that gap is set per procedure code and a single figure would be wrong for nearly every code. Nor is there a count of the codes in the full set. Denial rates specific to setting errors aren't published in a form worth quoting, and competitor coverage of this topic couldn't be read from this environment, so nothing here is described as a gap somebody else left.
Related medical billing guides
Three neighboring jobs sit around this one. Denial work grows into its own desk, telehealth billing needs somebody who watches payer policy rather than habit, and the software holding your stored tables decides how current they stay.
When denial work outgrows the place of service code
Setting errors are one line on a much longer denial list, and they're rarely the biggest one. A practice that starts tracking them soon finds eligibility problems, missing authorizations and coding mismatches sitting in the same queue. At that point the work stops being a checklist item and becomes somebody's job, with its own worklist, its own appeal deadlines and its own reporting. Practices reaching that stage tend to compare providers rather than hire a single person, and companies that publish billing as a named service line are the shortlist that's worth reading first. The firms that publish it are ranked in our list of best virtual medical biller companies.
Who fixes a telehealth place of service code
Remote visits are where setting errors cluster, because the entry depends on a fact nobody writes in the chart, namely where the patient was sitting. A practice running a heavy video schedule needs somebody who confirms that fact at booking, watches each plan's telehealth policy for the year and corrects the line when a patient converts to an in-person slot. That work overlaps with scheduling and intake more than it does with coding, which is why practices sometimes staff it from the telehealth support side instead. Firms naming that support are compared in our list of best telehealth virtual assistant companies.
The software that stores your place of service code list
Every practice management system ships a stored code table, and how quickly that table tracks the CMS set depends on the vendor rather than on anything a biller does. Some systems pull updates automatically. Others wait for a release, and a retired descriptor can sit in a dropdown long enough to teach new staff the wrong habit. Knowing which behavior your system has is worth one question to the vendor, and it decides whether a biller should trust the dropdown or the published set. What these systems hold is covered in our overview of medical billing tools and software.