Screening for chronic kidney disease (CKD) can potentially decrease kidney disease-related morbidity and mortality via early diagnosis and initiation of evidence-based therapies. However, whether patients should be screened for CKD remains highly controversial, as is the optimal target population in whom screening should be implemented. Due to insufficient evidence, statements from professional organizations have either no recommendation for screening for CKD or are discordant on whether they recommend screening in high-risk populations. We will address this critical public health question and evaluate the potential benefits of screening for CKD across different populations by developing an enhanced version of the Cardiovascular Disease (CVD) Policy Model that includes kidney parameters and outcomes, called the CKD Policy Model. The CVD Policy Model is a validated state-transition Markov model of CVD events and mortality in US adults. Markov modeling is an established technique that allows us to synthesize evidence and simulate and quantify the expected benefits of interventions on downstream outcomes. We have access to pooled longitudinal cohorts, comprising over 65,000 individuals followed for up to 30 years with sequential measurements of estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR). We will adapt and enhance the CVD Policy Model for nephrology applications by incorporating categories of eGFR and UACR to model CKD stage transitions in two dimensions. We will determine incident CKD probabilities based on a combination of demographics (age, sex, race/ethnicity) and risk factors, including history of diabetes, history of hypertension, and family history of kidney disease. Having developed and tailored the new CKD Policy Model, we will use a Markov decision analysis to project the impact of CKD screening. The treatment intervention triggered by CKD identification will include three treatments: 1) angiotensin converting enzyme inhibitors and angiotensin receptor blockers (ACEi/ARBs); 2) statins; and 3) blood pressure regimen intensification. The outcomes of the model will be the changing incidence of CVD events, end-stage renal disease, and mortality. The overall goal of this proposal is to establish an evidence-based framework for developing potential CKD screening strategies.
Our specific aims are: 1) to determine individualized probabilities of incident CKD and CKD progression based on patient demographics and risk factors and develop the CKD Policy Model; and 2) to estimate the expected impact of different selective CKD screening strategies on CVD events, incident ESRD, and cause-specific and all-cause mortality using Markov modeling. The results of this research will identify the optimal characteristics to guide patient selection for first CKD screening and the optimal repeat screening frequency, and ultimately inform the design of a future pragmatic CKD screening trial.
Chronic kidney disease (CKD) is an asymptomatic disease at early stages, so screening for CKD with blood and urine tests can improve early detection, increase the utilization of effective medications, and improve blood pressure control. However, guidelines differ on their recommendations of whether and whom to screen for CKD. We will use computer simulation modeling to estimate the potential benefits of CKD screening and treatment on reductions in cardiovascular events, kidney failure, and death.
