Comorbidity Adjustment

CMS maintains a list of comorbidity subgroups: clusters of ICD-10-CM diagnosis codes associated with higher resource use in home health. A period earns a low comorbidity adjustment when at least one reported secondary diagnosis falls into a qualifying subgroup. It earns a high adjustment when two or more secondary diagnoses fall into subgroup combinations that CMS has identified as interacting, meaning the conditions together predict higher costs than either alone. Everything else gets no adjustment. The tier feeds the case-mix group, so it moves the case-mix weight and the period's payment.

The adjustment is calculated from the secondary diagnoses reported on the claim, which can carry up to 24 secondary diagnosis codes. Those codes must originate in the clinical record: conditions documented by the certifying practitioner, active in the plan of care, or affecting the patient's care and outcomes. This is where the adjustment is won or lost. A referral packet that lists only the acute problem, coded as-is, will miss the diabetes, heart failure, and neurological conditions that legitimately qualify. Coding from the complete record, and querying the physician when documentation is ambiguous, is what separates accurate comorbidity capture from systematic undercoding.

The comorbidity subgroup lists and interactions are not static. CMS recalibrates them in the annual home health final rule, and the CY2026 rule updated the subgroups using CY2024 data. Codes move in and out of subgroups, and interaction pairs change, so a diagnosis profile that earned a high adjustment one year can earn low or none the next without any change in the patient. Coding teams should review the updated subgroup lists each year, and finance teams should re-model expected comorbidity tier distribution rather than assuming last year's mix carries forward.

The most expensive mistake is undercoding: documented chronic conditions that never make it onto the claim, which suppresses the tier and understates patient complexity everywhere downstream, from payment to risk adjustment to quality benchmarking. The opposite failure is coding conditions that appear nowhere in the practitioner's documentation, which inflates payment the record cannot defend and creates audit exposure under program integrity review. Watch also for diagnosis lists that are copied forward across episodes without validation, and for coding workflows that stop at the number of diagnoses OASIS displays rather than using the claim's full capacity for supported codes.