Chapter 36 — Further Reading

The orientation for this chapter's sources. Almost everything that governs risk-adjusted payment is free, federal, and revised annually — and this is the one area of the book where the model version matters as much as the year. A description of "the HCC model" written three years ago may be describing a different set of categories with different numbering and different coefficients from the one your contract runs on. The habit this list exists to build: get the model version in force for your payment year, from the plan or from CMS, and read the actual tables. Everything else, including this book, is orientation.

Tier 1 — Canonical: the rules and the primary sources

• CMS's Medicare Advantage risk adjustment program materials (the risk adjustment pages of the CMS website, the annual Advance Notice and Rate Announcement for the Medicare Advantage and Part D programs). This is where the model version, the phase-in schedule, the normalization factors, the coding intensity adjustment, and the county benchmarks actually live. The Rate Announcement is published every spring for the following payment year and is the single most load-bearing document in this chapter.
• The CMS-HCC model software and its published mappings. The diagnosis-to-category crosswalks and the coefficient tables are published with the model. These are the only correct source for any question of the form "does this code map, and to what?" — and the answer changes by model version.
• CMS's Risk Adjustment Data Validation (RADV) program pages, and the contract-level RADV final rule published in the Federal Register in January 2023. Read the rule's preamble as well as its text: the discussion of extrapolation and of the fee-for-service adjuster is the clearest statement of both sides of the argument that Case Study 1 summarizes. Check for subsequent rulemaking and for the status of the litigation.
• ICD-10-CM Official Guidelines for Coding and Reporting — Section I.A.15 (the "with" convention), Section I.B, and Section IV (CMS/NCHS, free, annual). Section IV's instruction on reporting conditions that coexist and require or affect patient care, treatment, or management is the binding rule that MEAT operationalizes. Chapter 9 §9.7 is this book's treatment of the convention; the Guidelines are the authority.
• The False Claims Act, 31 U.S.C. §§ 3729–3733, including the qui tam provisions — the statutory basis of most of the enforcement described in §36.8 and Case Study 1.
• The Affordable Care Act's sixty-day overpayment provision and CMS's implementing rules for Parts C and D. Verify the current state of the law; the Parts C and D rule has been litigated.
• The Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) and the Quality Payment Program resource library (qpp.cms.gov) — the authority for everything §36.9 says about MIPS, and the place to get the current year's categories, weights, thresholds, measure specifications, and MIPS Value Pathways. Do not use any secondary description of MIPS, including this book's.
• The Medicare Shared Savings Program regulations (42 CFR Part 425) and CMS's Shared Savings Program pages — benchmarks, attribution methodology, minimum savings rates, sharing rates, the quality performance standard, and the risk score growth cap §36.10 describes.
• The Affordable Care Act's commercial risk adjustment program (the HHS-HCC model), documented in the annual Notice of Benefit and Payment Parameters. Read it if you work in the individual or small-group market; the model is not the Medicare one.
• CPT Category II codes and their reporting guidelines (American Medical Association), including the performance measure exclusion modifiers 1P, 2P, 3P, and 8P. Revised annually with the rest of CPT; check the descriptors every January.

Tier 2 — Attributed: measurement, evaluation, and the honest numbers

• MedPAC's annual Report to the Congress: Medicare Payment Policy, the Medicare Advantage chapter. The best recurring plain-language analysis of coding intensity, risk score differences between Medicare Advantage and fee-for-service, and the effect of chart reviews and health risk assessments. Free, public, and written to be read by people who are not economists. If you read one Tier 2 source, read this one.
• OIG reports and Work Plan items on Medicare Advantage risk adjustment — contract-level audits of individual organizations, and the thematic work on diagnoses sourced only from chart reviews or health risk assessments. Searchable at oig.hhs.gov; the subjects recur, which is itself information.
• The Department of Justice's press releases on False Claims Act resolutions involving Medicare Advantage risk adjustment. The primary public record of what has actually been alleged and resolved. Read these rather than any summary of them, including this chapter's — the theories of liability are stated plainly and the amounts are exact.
• NCQA's HEDIS measure specifications, and the measure stewards' own documentation for whatever measures your contract names. Case Study 2's whole argument is that a measure specification has a history and a current version, and that reading last year's is a real failure mode.
• The ACCORD trial — Gerstein et al., "Effects of Intensive Glucose Lowering in Type 2 Diabetes," New England Journal of Medicine, 2008 — together with the ADVANCE and VADT reports of the same period. The clinical evidence behind Case Study 2, and worth reading in the original if you are ever asked to defend why a quality target is not a clinical instruction.
• The Choosing Wisely recommendations of the American Geriatrics Society on glycemic targets in older adults, and current professional guidance on individualized targets. The clearest short statement of why a single-threshold measure cannot be right for every patient.
• The research literature on exception reporting in the United Kingdom's Quality and Outcomes Framework. One of very few places where the behavior Case Study 2 describes has been measured at national scale rather than speculated about.
• AAPC's Certified Risk Adjustment Coder (CRC) materials and AHIMA's risk-adjustment practice resources — the professional literature for this specialty. Benchmarks and survey figures in this literature are Tier 2: ranges, not facts.
• Simborg, D., "DRG creep: a new hospital-acquired disease," New England Journal of Medicine (1981). Two pages, forty years early, and the best short statement of the general problem this chapter's enforcement landscape exists to manage. Chapter 33's Case Study 2 tells that story; read it alongside this chapter's.

Tier 3 — Illustrative and constructed: this book's own apparatus

• Every coefficient, benchmark, risk score, and dollar figure in this chapter — the 1.555 and 0.752 and 0.428 worksheets, the \$1,000.00 per member per month benchmark, the \$3,888.00 annual difference, the shared-savings settlement, and the panel counts. All constructed teaching figures, labeled where they appear, arithmetically checked against themselves and against nothing in the real world. They exist to teach the structure and they expire on principle.
• Account 10-4471 — the March 14 office note, its problem list, its assessment, and the MEAT grid in Figure 36.3. Constructed, canonical within this book, and unchanged by this chapter: the chapter's finding is about the record, not about the coding.
• Account 22-8891 — Ridgeview Regional Medical Center's admission, used in §36.4 as a lens: the same coded record read by a risk model instead of a DRG grouper. The Chapter 33 file itself is unaltered.
• Figure 36.1's recapture worklist and Figure 36.2's compliance summary — constructed documents built to show what a defensible program's paperwork looks like, including the deletion count that makes it credible.
• Case Study 2's organization — a labeled composite assembled from documented patterns in the pay-for-performance literature. The clinical evidence and guideline history in that case study are real; the practice is not.