Abstract: FR-PO384
Developing a CKD Screening Algorithm for the Primary Care Setting
Session Information
- CKD: Risk Factors for Incidence and Progression - I
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Chronic Kidney Disease (Non-Dialysis)
- 301 CKD: Risk Factors for Incidence and Progression
Authors
- Cheng, Andy Yi-Ming, University of Pittsburgh, Pittsburgh, Pennsylvania, United States
- Jhamb, Manisha, University of Pittsburgh, Pittsburgh, Pennsylvania, United States
- Demoise, David, University of Pittsburgh, Pittsburgh, Pennsylvania, United States
- Kohli, Amar R, University of Pittsburgh, Pittsburgh, Pennsylvania, United States
Background
CKD in the US not only has a significant prevalence, but also has an alarming projected incidence. Current guidelines for CKD screening are discordant, and few studies have been performed in evaluating various screening algorithms. This retrospective cohort study aims to propose a potential CKD screening algorithm for the primary care setting.
Methods
Adult patients aged 18-69 years seen for primary care visits at a tertiary hospital primary care clinic in Western PA were identified. Those with CKD, defined as two eGFR values of less than 60mL/min/1.73m2 per MDRD calculation spaced at least 90 days apart in the EHR, were eligible for study inclusion. Characteristics including age ≥55, African American race, and comorbid DM or HTN were abstracted from the EHR. Statistical analyses, including percentage of CKD patients identified and number of patients needed to screen, were then performed using a reverse screening cohort of 650 patients to assess four potential screening algorithms.
Results
Known comorbid DM or HTN (81.8%) outperformed age ≥55 (69.7%) or African American race (49.5%) in identifying the largest percentage of CKD patients. Inclusion of age ≥55 to known comorbid DM or HTN yielded a greater percentage of CKD patients compared to addition of African American race to known comorbid DM or HTN (Table 1). A comprehensive screening algorithm including all four aforementioned variables mildly increased percentage of CKD patients identified, but at the expense of a moderate increase in number of patients needed to screen (Table 1).
Conclusion
A screening algorithm consisting of age ≥55 or known comorbid DM or HTN appeared to optimize percentage of CKD patients identified with number of patients needed to screen. Given that the proposed screening algorithm was derived via a reverse screening approach originating from a patient population with 100% CKD prevalance, further study applying the algorithm to a blanket primary care population is necessary to fully assess the algorithm's efficacy.
Proposed Screening Algorithms & Associated Statistical Analyses
Screening Algorithm Components | % of CKD Cases Accounted (Number of Patients) | Number of Patients Needed to Screen (95% CI) |
DM or HTN | 81.8% (532) | 5.5 (4.7-6.6) |
DM or HTN or Age ≥55 | 92.0% (598) | 12.5 (9.9-16.9) |
DM or HTN or African American Race | 87.2% (567) | 7.8 (6.5-9.8) |
DM or HTN or Age ≥55 or African American Race | 94.6% (615) | 18.6 (14.0-27.2) |
Funding
- Private Foundation Support