1066 — Implementation of a Pilot Randomized Pragmatic Trial of a Screening Program for Chronic Kidney Disease among Non-Diabetic Hypertensive Veterans
Lead/Presenter: Anna Rubinsky,
All Authors: Peralta CA (San Francisco VA Health Care System; University of California, San Francisco)
Rubinsky AD (San Francisco VA Health Care System; University of California, San Francisco)
Frigaard M (San Francisco VA Health Care System; University of California, San Francisco)
Rolon L (San Francisco VA Health Care System; University of California, San Francisco)
Seal K (San Francisco VA Health Care System)
Tuot D (University of California, San Francisco)
Powe N (University of California, San Francisco)
Chao S (San Francisco VA Health Care System)
Voora S (San Francisco VA Health Care System; University of California, San Francisco)
Shlipak MG (San Francisco VA Health Care System; University of California, San Francisco)
We describe the implementation of a pilot, cluster-randomized pragmatic trial to evaluate the feasibility and effectiveness of a "triple marker" chronic kidney disease (CKD) screening program among non-diabetic, hypertensive veterans seen in primary care at one VA hospital.
The trial uses VA electronic health records (EHR) to identify participants, deliver interventions and ascertain outcomes. Eligible patients included non-diabetic, hypertensive veterans age 18-80 with active primary care at the San Francisco VA Medical Center and no diagnosis of CKD. Participants were randomized in clusters determined by primary care provider (PCP) into three arms, including usual care and two incrementally intensified treatment strategies implemented by PCPs: (1) screen for CKD followed by patient and provider education on CKD management, or (2) screen-educate plus a clinical pharmacist-led CKD management program. Triple-marker laboratory testing for CKD (creatinine, cystatin C and albumin-to-creatinine ratio) was ordered by study staff for all participants in the intervention arms who had an upcoming appointment with their PCP.
Among 2,012 eligible patients mailed information letters about the study, 133 opted-out within 30 days and an additional 60 were excluded due to letters returned as undeliverable. The remaining 1,819 patients were randomized within 41 participating PCP clusters. Two PCPs left the practice after randomization but before the interventions were implemented, potentially affecting 179 patients. During the 12-month study period, 58% (701/1,215) of participants randomized to an intervention arm had a scheduled PCP appointment. Of these patients, 52% (367/701) completed the triple-marker testing after it was ordered, resulting in 79 (11%) new CKD cases. Of the 35 CKD cases identified among patients randomized to the screen-educate plus pharmacist management arm, only 6 (17%) were referred by PCPs to the pharmacist. No adverse events have been reported to date.
This study achieved high rates of participation by providers and patients and high rates of screen-detected CKD. Limitations include potential contamination from provider turnover, enrollment limitations due to the mass mailing approach of informed consent, and lower-than-expected rates of completed CKD testing among patients with tests ordered and of pharmacy referrals. Future pragmatic trials should consider randomization by clinic rather than by provider and "rolling" enrollment of patients with individual informed consent at PCP appointments or waived consent for very low risk interventions.
"Real world" lessons from this VA trial can inform design of pragmatic trials that address identification and management of CKD.