99213 vs 99214: Which Level Can You Defend?

On those facts alone it is a 99214. "2 or more stable, chronic illnesses" is the moderate descriptor for problems addressed, and prescription drug management is a CPT-listed example of moderate risk. That is two of the three elements at moderate, and two of three is all CPT requires. You do not need the data element, and you do not need a complicated visit — the moderate bar is a description of the clinical work, not a reward for difficulty.

The CPT descriptor for 99214 reads "30-39 minutes of total time is spent on the date of the encounter." Thirty minutes is the floor to reach the level; at 40 minutes the visit is a 99215. But time is an alternative to MDM, not an additional requirement — if your MDM is moderate, a 12-minute visit is still a 99214, and you do not need to record the time at all. Only document total time if time is the basis you are using.

Generally yes, provided a decision is documented — but be aware this is the most disputed square on the whole table, and your MAC's wording matters. CPT itself never defines "prescription drug management"; the phrase appears only as a moderate-risk example, so every working rule on it is contractor or specialty-society interpretation rather than CPT text. Noridian and Novitas both state that it "does not require a new drug, a new dosage, or a discontinuation of a current prescription," while adding that "an encounter documented as only a prescription refill without documentation of a problem addressed would not suffice." CGS is the most permissive: "A decision to maintain the current medication is sufficient. Credit is given as long as the documentation clearly indicates that decision-making took place in regard to the medication(s)." AAFP sets the highest bar, wanting the refill's purpose, any dosage adjustment, and a brief management statement linking the drug to the problem addressed. Document to the AAFP standard and you are safe everywhere; document to the CGS standard outside CGS's jurisdiction and you are exposed.

No, not on its own. Noridian and Novitas use identical language: "An order for an over the counter (OTC) drug does not count as prescription drug management. However, managing the patient's prescription drugs in connection with adding an OTC or supplement would show prescription drug management." So telling a patient to start a proton pump inhibitor off the shelf is not the moderate-risk element; adjusting their prescribed regimen around that addition is. Note too that prescription drug management supports only the <em>risk</em> element — you still need a second element at moderate, from problems or data, to reach 99214.

No. CPT addresses this directly twice over: "Tests ordered are presumed to be analyzed when the results are reported. Therefore, when they are ordered during an encounter, they are counted in that encounter," and separately, "ordering a test is included in the category of test result(s) and the review of the test result is part of the encounter and not a subsequent encounter." One unique test ordered is one data element, and reviewing that same result later does not create a second. This is the classic double-dip and one of the easiest errors for an auditor to spot. Two related limits: a test whose professional component you report separately cannot be counted as a data element at all, and a clinical laboratory panel such as a basic metabolic panel is a single test, not one per analyte.

No, and the national data makes that concrete. 99214 has been the most-billed established office visit for years, and its share has been rising: on CMS Medicare utilisation data, 99214 grew from 48.4% of established office visits in 2019 to 53.8% in 2024, while 99213 fell from 42.0% to 36.0%. The inflection is 2021 — the year the MDM and time rules took effect. A distribution weighted toward 99214 is now the norm rather than an outlier. What draws scrutiny is a distribution the panel cannot explain, combined with notes that look identical to one another. CMS's CERT programme puts 99214's improper payment rate at 5.0%, and 62.5% of those errors are incorrect coding rather than missing documentation — meaning the note existed and simply did not carry the level. The defence is never the percentage; it is that each individual note shows two moderate elements.

Yes, and MACs treat it as a medical necessity failure rather than a coding slip. Palmetto GBA's guidance is blunt: "Cloning occurs when medical documentation is the same from beneficiary to beneficiary. It would not be expected that every patient had the exact same problem, symptoms and required the exact same treatment. This cloned documentation does not meet medical necessity requirements for coverage of services rendered due to the lack of specific, individual information." It goes on to say identification of cloned documentation "will lead to denial of services for lack of medical necessity and recoupment of all overpayments made." CMS separately notes that HHS-OIG has identified record cloning and copy-and-paste among EHR features improperly used to facilitate fraud, waste and abuse. Carrying structure forward is fine; carrying the assessment forward unchanged is what fails.

Yes. CPT lets you select the level of each E/M service by either the level of MDM or the total time for that encounter, and the choice is made per encounter. Most clinicians code by MDM by default and fall back to time on the occasional visit where the counselling and coordination were long but the decision making was not complex. The only place it is not a free choice is prolonged services: 99417 may be reported only when the primary service was selected using time alone.

No. Since 1 January 2021 the codes require only "a medically appropriate history and/or examination," and CPT states plainly that the extent of history and physical examination is not an element in selecting the level. The treating clinician decides what is medically appropriate. Padding a note with a ten-system ROS adds no support to a 99214 and, when it is obviously templated, actively undermines the note's credibility.