Chapter 15 — Further Reading

A word about currency before the list. This is the one chapter in the book where the publication date of a source is as important as its authority. Anything describing office visit level selection that was written before 2021 describes a framework that no longer governs, and anything describing hospital, observation, consultation, emergency department, nursing facility, or home E/M that was written before 2023 does the same. Check the date on everything, including material published by organizations you trust.

Tier 1 — Verified canonical sources

The CPT Evaluation and Management guidelines, current edition. This is the governing document and it is not long. It contains the definitions of "addressed," "stable," "unique test," "external," "independent historian," "discussion," and every entry in the risk table — all of which this chapter summarizes and none of which it replaces.

If you read one thing from this list, read this. Most of the audit findings described in §15.12 are corrected by four pages of definitions that almost nobody in the field has read straight through.

The CPT medical decision making table, published with those guidelines. The grid itself, with all four levels across all three elements on one page. Print it. Keep it where you work.

CPT Assistant. Subscription. The 2021 and 2023 transitions generated an unusually large body of clarifying guidance, and the ambiguous cases in §15.5–§15.7 have mostly been addressed specifically.

CMS Evaluation and Management Services guidance, including the Medicare Learning Network's educational materials and the Medicare Claims Processing Manual (Publication 100-04), Chapter 12. Free.

CMS guidance on prolonged services and G2212. §15.9's Medicare complication is documented here and the threshold difference is stated explicitly. Read this rather than trusting a summary — including this book's.

CMS guidance on split (or shared) visits, including the current definition of the substantive portion. This definition has moved more than once, and it is the single item in this chapter most likely to be out of date by the time you read it. Verify it directly, every year.

CMS guidance on incident-to services, in the Medicare Benefit Policy Manual (Publication 100-02), Chapter 15. The seven conditions in §15.11 come from here, and the new-problem disqualification that Account 10-4471 illustrates is stated in the source.

Office of Inspector General reports and Work Plan items on evaluation and management services, including reports on level distribution and on time-based selection. Free. The documented basis for both case studies in this chapter.

CMS public utilization data by provider and service. Free. This is the data set that generates outlier analysis, and it is worth knowing that it is public — anyone, including a payer, a journalist, or a competitor, can look up how your organization's E/M distribution compares to its peers.

Payer-specific E/M policies. Several large payers publish their own level-selection review policies and their own approaches to distribution analysis. As with modifiers in Chapter 14, the payer's published policy is operative for that payer's claims.

Tier 2 — Attributed, specifics unverified

AMA educational material on the 2021 and 2023 E/M changes, including the published implementation guides. These are freely available, they are written for clinicians as much as for coders, and they are the best material available for explaining the change to a physician who has not had time to read the guidelines.

AAPC and AHIMA E/M training material and audit tools. Both organizations publish scoring worksheets. Check the copyright date before using one — Case Study 1 in this chapter is about exactly what happens when a pre-2021 worksheet stays in circulation.

Specialty society guidance on E/M in specific specialties. Frequently better than general material, because it works from encounters that resemble yours. Particularly useful for the problems element, where "moderate" looks very different in dermatology than in nephrology.

Published analyses of E/M level distribution before and after 2021. Useful for understanding what happened to the national distribution and therefore what "outlier" means now. Treat specific figures as study-specific.

Commentary on documentation burden and clinician burnout, which was the stated motivation for the 2021 change. Worth reading if you want to understand why the change happened rather than only what it was — and it makes Case Study 1's second cost, the two years of pointless templated content, land harder.

Compliance literature on template design and default values. §15.12 and Case Study 2 both turn on this, and there is a substantial practical literature on how electronic record configuration creates compliance exposure that nobody chose.

Tier 3 — Illustrative and constructed

Every scenario in §15.5 through §15.9, including the four data scenarios and the time examples.

The four readings of the same note in §15.10. Constructed as a teaching device from Account 10-4471's documentation.

Case Study 1 in its entirety — the twelve-physician group, the 1997-era templates, the two years, and the new coder's question. The pattern is ordinary; the organization and figures are constructed.

Case Study 2 in its entirety — the specialty practice, the prefilled time default, the 958 minutes, and the billing supervisor's spreadsheet. The detection method is real and you should run it.

Account 10-4471's leveling in §15.13.

Three things worth doing

Read the E/M guidelines straight through, once. Twenty minutes. It is the highest return on reading time available anywhere in professional coding, and the fact that so few people have done it is why §15.12's audit findings are as common as they are.

Print the MDM table and put it where you work. Not a summary of it. The table.

Then run Case Study 2's query against your own organization. Sum documented total time by rendering provider by date of service for one month, and compare it to the hours in a working day. It needs no charts and no permission, it takes an afternoon, and if it comes back clean you have learned something real. If it does not, you have found the thing before someone else did.