Updates to 2021 E/M Office Coding Guidelines
In July 2019, the American Medical Association (AMA) published the “CPT Evaluation and Management (E/M) Office or Other Outpatient (99202-99215) and Prolonged Services (99354, 99355, 99356, 99417) Code and Guideline Changes,” which were put into effect on January 1, 2021. These guidelines were incorporated into the 2021 CPT Code Book.
In the first few months of using these new codes, it was quickly found that some clarifications and modifications needed to occur. To address this issue, the AMA has issued a new document that includes some “technical corrections” to the guidelines. The full revised document can be found at https://ama-assn.org/system/files/2019-06/cpt-office-prolonged-svs-code-changes.pdf.
The changes are summarized as follows:
- Medical decision making is revised in the following ways:
- Clarifying when reporting a test that is considered, but not selected after shared decision making.
- Providing a definition of “Analyzed” for reporting tests in the data column.
- Clarifying the definition of a “unique” test.
- Clarifying what is meant by “discussion” between physicians, and other qualified health care professionals and patients.
- Providing a definition of major vs minor surgery.
- Clarifying when reporting a test that is considered, but not selected after shared decision making.
- Clarification around which activities are not counted when reporting time as a key criterion for code level selection.
The changes and their application in the practice of Maternal Fetal Medicine are:
- A “test,” if considered but not selected, can be counted in the data as a test “ordered,” for the purpose of determining the level of data (page 9, paragraph 1).
Example: Dr. Adamson recommends that Abigail have a fetal karyotype by genetic amniocentesis to investigate concerns about fetal abnormalities noted on a detailed fetal anatomy survey performed last week. . Abigail declines the procedure and associated laboratory testing.
This counts as 1 test “ordered” in the Data category. It will be essential to spell out that recommendation in the progress note to support the test “ordering.” In addition, the recommendation for the amniocentesis would support a moderate level of Risk, even though the procedure was declined.
- The AMA has clarified the difference between “interpreting” and “analyzing” a test. Ultrasounds and other imaging services require “interpretation,” while most laboratory tests are “analyzed”—that is, data points are considered. (page 6, paragraphs 3-5).
Example: During an encounter, Dr. Boston orders and performs a follow-up ultrasound and orders 3 laboratory tests for her patient Betty, which are all performed in the in-office laboratory.
Because the ultrasound requires interpretation and is being billed by the physician, it does not count as a data element. However, the laboratory tests do count as 3 data elements, since they are analyzed, but not interpreted. This is true even if the laboratory services are billed by the physician.
Other points of clarification include:
a. Reviewing multiple results of the same test (e.g. serial blood glucose levels) is considered 1 unique test.
b. For the purpose of data reviewed and analyzed, pulse oximetry is not a test.
- The definition of a “unique” source has been clarified (page 6, paragraph 5).
Example: Dr. Connors evaluates Connie, who has been seen previously during her pregnancy by her obstetrician and endocrinologist (for her Type 2 diabetes). Dr. Connors reviews materials from both providers.
This counts as 2 prior external notes. It is important that the two separate providers (sources) be identified in Dr. Connors’ note to count as data elements.
- When performing a “Discussion of management or test interpretation,” the exchange must be interactive. It must be direct and not through intermediaries, such as staff or trainees. (page 7, paragraph 1).
Example 1: Dr. Davis wants Donna to see a pediatric cardiologist, because of concerns about fetal heart abnormalities. Her nurse, Daphne, calls the specialist’s office and schedules an appointment with him. This does not count as “discussion of management or test interpretation.”
Example 2: Dr. Davis wants Donna to see a pediatric cardiologist, because of concerns about fetal heart abnormalities. She calls Dr. Ellis, the cardiologist, and discusses Donna’s case with him for 7 minutes.
This does count as “discussion of management or test interpretation.”
Other points of clarification include:
a. The discussion can be asynchronous (not in person) and does not have to be verbal. It can be via secure email, provided it is an interactive communication.
b. It does not need to occur on the same day as the in person visit, if Medical Decision Making is the method of code selection being used. The AMA does recommend that it be initiated and completed in a short period of time (within a day or two).
c. If the discussion occurs on some other day, it can only be counted once and only when it influences the decision making of the encounter.
d. If time is the method by which the E/M code is being selected, the “discussion” must occur on the date of the face-to-face visit.
- Additional information is provided concerning “minor” vs. “major” surgery, as well as “elective” vs. “emergency” surgery (page 8, paragraph 1). The definitions are determined by the clinician and are consistent with the meaning of the term when used generally by trained clinicians. They do not necessarily correlate with global package classifications.
- When using time to define the level of service, the following do not count (page 3, paragraph 2):
· The performance of other services reported separately
· Travel
· Teaching that is general and not limited to discussion that is required for the management of a specific patient
Example 1: Dr. Farrell documents that she spent 45 minutes working on behalf of her patient Frieda. During that period, 20 minutes are spent in the performance of an ultrasound and writing the report.
When selecting a level of E/M service, only 25 minutes can be counted.
Example 2: Dr. George documents that he spent 45 minutes working on behalf of his patient Giselle. However, 10 minutes of that time were spent discussing general information about the patient’s condition with his two residents.
Since 10 minutes were allocated to educating the residents, only 35 minutes can be counted when selecting the service level.