Incidence, Persistence, and Steady-State Prevalence in Coding Intensity for Health Plan Payment

Structured AbstractO_ST_ABSObjectiveC_ST_ABSTo define measures of Medicare diagnosis coding intensity that capture the dynamics of changes in coding practices. Study setting and designRetrospective analysis of coding for risk adjustment using observational claims data from Medicare beneficiaries. Data sourcesEnrollment and claims data from 2017-2018 of a random 20 percent sample of Medicare beneficiaries were subset to those assigned to an Accountable Care Organization in 2018. Principal findingsWe decompose prevalence of a diagnosis code into incidence (proportion of beneficiaries that newly have the code) and persistence (proportion of beneficiaries who previously had the code and continue to do so). Together these define steady-state prevalence, the hypothetical long-run prevalence implied by no changes in current rates of incidence and persistence of coding. Steady-state prevalence can help explain why observed prevalence tends to grow over time without continued behavioral change. For example, our measures suggest that the prevalence of the Specified Heart Arrhythmias diagnosis would continue to rise from 18.7% in 2018 to 28.0% without changes in coding practices. ConclusionsResearchers and policymakers can better understand why changes in coding practices can take years to be fully reflected in data and monitor coding behavior by using our proposed measures. Callout BoxO_ST_ABSWhat is known on this topicC_ST_ABSO_LIMedicare incentivizes providers and insurers to code for as many diagnoses per beneficiary as possible, contributing to billions of dollars of Medicare program spending. C_LIO_LIEmpirical research indicates that the frequency of diagnostic coding in Medicare has increased steadily in recent years. C_LIO_LICurrent measures of coding practices for a given diagnosis largely count the prevalence, or the proportion of beneficiaries with a specified code, in one time period. C_LI What this study addsO_LIThis study proposes decomposing coding practices over two years into several measures, which policymakers can use to better monitor coding behaviors and understand how they may change over time. C_LIO_LIThis study defines steady-state prevalence as a novel measure of the ultimate prevalence of a code implied by current coding patterns hypothetically remaining unchanged. C_LIO_LIApplying our measures to Accountable Care Organization coding finds that observed coding prevalence will increase even without further behavioral change for the 10 diagnoses most impactful to 2018 risk adjustment. C_LI