Background: A cost-effectiveness model that accurately represents disease progression, outcomes, and associated costs is necessary to evaluate the cost-effectiveness of interventions for chronic kidney disease (CKD). Study Design: We developed a microsimulation model of the incidence, progression, and treatment of CKD. The model was validated by comparing its predictions with survey and epidemiologic data sources. Setting & Population: US patients. Model, Perspective, & Timeframe: The model follows up disease progression in a cohort of simulated patients aged 30 until age 90 years or death. The model consists of 7 mutually exclusive states representing no CKD, 5 stages of CKD, and death. Progression through the stages is governed by a person's glomerular filtration rate and albuminuria status. Diabetes, hypertension, and other risk factors influence CKD and the development of CKD complications in the model. Costs are evaluated from the health care system perspective. Intervention: Usual care, including incidental screening for persons with diabetes or hypertension. Outcomes: Progression to CKD stages, complications, and mortality. Results: The model provides reasonably accurate estimates of CKD prevalence by stage. The model predicts that 47.1% of 30-year-olds will develop CKD during their lifetime, with 1.7%, 6.9%, 27.3%, 6.9%, and 4.4% ending at stages 1-5, respectively. Approximately 11% of persons who reach stage 3 will eventually progress to stage 5. The model also predicts that 3.7% of persons will develop end-stage renal disease compared with an estimate of 3.0% based on current end-stage renal disease lifetime incidence. Limitations: The model synthesizes data from multiple sources rather than a single source and relies on explicit assumptions about progression. The model does not include acute kidney failure. Conclusion: The model is well validated and can be used to evaluate the cost-effectiveness of CKD interventions. The model also can be updated as better data for CKD progression become available.
All Science Journal Classification (ASJC) codes