Written and maintained by CASRAI Editorial Board
Last updated
Before 2021, time only decided an office visit level when the encounter was dominated by counselling and coordination of care, and the time that counted was face-to-face time. Both halves of that rule are gone. Time is now a standalone alternative to medical decision making for most E/M families, the counselling threshold has been abolished, and the quantity being measured is no longer face-to-face time at all — it is total time on the date of the encounter.
The change is larger than it sounds. A great deal of physician work that was previously invisible to the coding system — chart review before the patient walks in, ordering, documenting, coordinating afterwards — now counts, provided it happened on the same calendar date and was performed personally. Equally, several things clinicians assume must count do not.
Verified against the American Medical Association’s CPT Evaluation and Management (E/M) Code and Guideline Changes, effective 1 January 2023 — the E/M Services Guidelines sections “Levels of E/M Services” and “Guidelines for Selecting Level of Service Based on Time,” including the enumerated lists of qualifying and non-qualifying activities — and against CMS Medicare Learning Network booklet MLN006764, Evaluation and Management Services (May 2026 revision), which sets out Medicare’s prolonged-services and split-or-shared rules. CPT codes, descriptions and other data are copyright the American Medical Association.
The Choice Between Time and MDM
The guidelines present two routes and no hierarchy:
“Select the appropriate level of E/M services based on the following: 1. The level of the MDM as defined for each service, or 2. The total time for E/M services performed on the date of the encounter.”
It is or. You pick, per encounter, whichever reflects the encounter most accurately — the AMA’s own E/M guidance frames it exactly that way. You do not need to use the same method for every visit, or for every visit with a given patient. What you cannot do is blend them: establish a level by MDM and then add time to push it higher, or take the higher of the two after computing both and calling that the answer.
Practically, time tends to be the better instrument when an encounter involved a lot of work but little decisional complexity — long counselling, an involved family discussion, extensive records review — and MDM tends to be better when the reverse is true. When an encounter is genuinely well described by both, either is defensible.
Where time does not apply
Two carve-outs matter. Emergency department services do not use time at all, and the guidelines explain why: those services “are typically provided on a variable intensity basis, often involving multiple encounters with several patients over an extended period of time.” And certain time-based families such as critical care “use time differently” — the guidelines direct you to review the instructions for each category rather than assuming the general rule. There are also low-level codes for which MDM does not apply, where the guidelines send the supervision case elsewhere: “if the physician’s or other qualified health care professional’s time is spent in the supervision of clinical staff who perform the face-to-face services of the encounter,” a different code applies.
What “Total Time” Means
“For coding purposes, time for these services is the total time on the date of the encounter. It includes both the face-to-face time with the patient and/or family/caregiver and non-face-to-face time personally spent by the physician and/or other qualified health care professional(s) on the day of the encounter (includes time in activities that require the physician or other qualified health care professional and does not include time in activities normally performed by clinical staff). It includes time regardless of the location of the physician or other qualified health care professional (eg, whether on or off the inpatient unit or in or out of the outpatient office). It does not include any time spent in the performance of other separately reported service(s).”
Four constraints are packed into that paragraph, and each one disposes of a common assumption.
| Constraint | What it excludes |
|---|---|
| Date of the encounter | Work on any other calendar date, however related |
| Personally spent | Anyone else’s time, unless it is a qualifying split or shared visit |
| Requires the clinician | Activities normally performed by clinical staff, even if you did them |
| Not separately reported | Time on anything else you are billing for on that claim |
Location, notably, is not a constraint. Reviewing the chart from home the evening of the visit counts, provided it is the same calendar date.
The third constraint is the subtlest and the most frequently misapplied. The test is whether the activity is one that requires a physician or qualified health care professional — not whether a physician happened to perform it. A clinician who personally rooms a patient and takes their vital signs because the practice is short-staffed has not thereby generated countable E/M time.
The midnight rule
“Total time on the date of the encounter is by calendar date. When using MDM or total time for code selection, a continuous service that spans the transition of two calendar dates is a single service and is reported on one calendar date. If the service is continuous before and through midnight, all the time may be applied to the reported date of the service.”
Continuity is the condition. A service running from 23:15 to 00:40 is one service and all 85 minutes may be applied to the reported date. Forty minutes on Tuesday and forty-five on Wednesday, with a break in between, is not one 85-minute service.
The Two Lists
The guidelines enumerate what counts, and separately what does not. These are the operative lists; they are short, and worth knowing verbatim rather than approximately.
Time that counts, when performed
- Preparing to see the patient (eg, review of tests)
- Obtaining and/or reviewing separately obtained history
- Performing a medically appropriate examination and/or evaluation
- Counseling and educating the patient/family/caregiver
- Ordering medications, tests, or procedures
- Referring and communicating with other health care professionals (when not separately reported)
- Documenting clinical information in the electronic or other health record
- Independently interpreting results (not separately reported) and communicating results to the patient/family/caregiver
- Care coordination (not separately reported)
Time that does not count
- The performance of other services that are reported separately
- Travel
- Teaching that is general and not limited to discussion that is required for the management of a specific patient
Three observations about these lists.
Documentation time counts. This is the largest practical change from the old rules, and the most under-claimed. Writing the note on the date of the encounter is countable E/M time.
“Not separately reported” qualifies three separate entries. Referral and communication, independent interpretation, and care coordination each carry it. The same principle governs the data column of the MDM table — if you bill for it, you cannot also count it — and it is the reason the professional and technical component split has consequences beyond the fee schedule.
The teaching exclusion is narrow, not total. What is excluded is teaching “that is general and not limited to discussion that is required for the management of a specific patient.” Discussion with a trainee about this patient’s management is not excluded by that wording. General instruction is.
Split and Shared Visits
Where two clinicians contribute, the guidelines define both the summing rule and its limit:
“When time is being used to select the appropriate level of services for which time-based reporting of shared or split visits is allowed, the time personally spent by the physician and other qualified health care professional(s) assessing and managing the patient and/or counseling, educating, communicating results to the patient/family/caregiver on the date of the encounter is summed to define total time… Only distinct time should be summed for shared or split visits (ie, when two or more individuals jointly meet with or discuss the patient, only the time of one individual should be counted).”
Overlapping time is counted once. A twenty-minute joint discussion between a physician and a nurse practitioner contributes twenty minutes, not forty.
Medicare then layers its own rule about who bills, which is a different question from how the time is totalled. Per MLN006764, for hospital inpatient, hospital outpatient and emergency department visits, “beginning January 1, 2024, the physician or NPP who provides more than 50% of the total time or the substantive part of MDM should bill for the visit.” For prolonged visits specifically, “the substantive portion is more than 50% of the practitioner’s total time.” Critical care is different again: because those visits do not use MDM, “the substantive portion means more than 50% of the total time spent by the practitioners.”
CMS also requires the identification on the claim: “Use modifier FS for split or shared E/M visits… This tells us that even though you’re submitting the claim under 1 practitioner’s NPI, 2 practitioners performed the visit.” Documentation must identify both clinicians, and the one who performed the substantive portion must sign and date the record. Note that split-or-shared billing is a facility-setting construct — CMS defines the relevant facility setting as “an institutional setting in which payment for services and supplies provided incident to physician or NPP professional services is prohibited under our regulations,” which is why incident-to billing and split-or-shared billing occupy mutually exclusive territory.
Prolonged Time, and the Medicare Divergence
When total time exceeds the range of the highest-level code in a family, prolonged services may be reported. Here Medicare and CPT part company: CMS instructs that for office and outpatient visits “when you select a visit level using time, report the prolonged O/O E/M visit time using HCPCS add-on code G2212,” rather than the CPT prolonged-services code, and its thresholds differ from the CPT ones. The MLN booklet publishes a table of required total times for each combination, and CMS footnotes it: “Total time is all the reportable time, including prolonged time, you spend with the patient on the date of service of the visit.”
Two rules follow. First, prolonged services are available only when you selected the level by time in the first place — there is no prolonged add-on for an MDM-levelled visit. Second, check whether the payer follows CPT or Medicare on which code to use, because billing the wrong one is a clean denial.
Documenting Time So It Survives Review
The guidelines impose one explicit documentation requirement, and it is unambiguous:
“The total time on the date of the encounter spent caring for the patient should be documented in the medical record when it is used as the basis for code selection.”
A total, on the date, in the record. Beyond that minimum, the practices whose time-based claims survive review tend to do three things.
They state the method. A line reading “level selected on total time” removes any ambiguity about which route was used, and prevents a reviewer from testing the note against the MDM table and finding it short.
They record a real number, not a template. A total that reads “40 minutes” on every chart in a sample is the pattern that triggers a probe. Times that cluster suspiciously at the bottom of each threshold band attract the same attention.
They show what the time was spent on. The total is the requirement; a brief breakdown is the defence. “Total time 52 minutes: 25 minutes face-to-face, 12 minutes reviewing outside records, 10 minutes documentation, 5 minutes coordinating with home health” is a claim a reviewer can test and accept. “52 minutes” alone is a number they can only take on trust — and on a records request, trust is not what is being extended.
Frequently Asked Questions
Does time spent the day before the visit count?
No. Total time is on the date of the encounter, by calendar date. Chart review the previous evening does not count, however necessary it was. The single exception is a continuous service spanning midnight, which is reported on one date with all its time applied there.
Can I count time my nurse spent?
No, unless it is a qualifying split or shared visit with another physician or qualified health care professional. The guidelines exclude time in activities normally performed by clinical staff, and clinical staff are defined as those who do not individually report the service.
Does the visit have to be dominated by counselling?
No. That threshold belonged to the pre-2021 rules and no longer exists. Total time is total time, regardless of what proportion was counselling.
If time supports a higher level than MDM, which do I bill?
Whichever you selected the visit on, applied consistently for that encounter. You choose the method that reflects the encounter most accurately and then apply it. Computing both and taking the higher is not the rule the guidelines describe, and a pattern of doing so is visible in aggregate data.
Does time spent on a same-day procedure count?
No — time on separately reported services is expressly excluded. Whether the E/M is separately reportable alongside the procedure at all is the modifier 25 question, covered in the modifier 25 guide. The same exclusion applies to the preventive half of a shared preventive-and-problem appointment; see that guide.
Does a telehealth visit use the same time rules?
The E/M time rules apply to the service being reported, and location of the clinician is expressly not a constraint on countable time. Which code family, place of service and modifier apply is separate — see modifier 95 and 93.
Is MDM safer than time?
Neither is inherently safer; they fail differently. A time-based claim fails when the total is undocumented, templated or unexplained. An MDM-based claim fails when the columns do not independently support the level — see the MDM table overview. Choose the one that actually describes the encounter, and document it as the one you chose.
