A quality improvement project at an Iowa City Veterans Affairs clinic reported that annual screening for diabetic kidney disease rose from 41% in June 2023 to 66% by March 2024, reaching and passing the project's 65% target. The measure counted Veterans with diabetes as up to date only when they had both a urine albumin-to-creatinine ratio and a serum creatinine test in the previous year.
Two interventions were introduced in sequence. A national electronic health record reminder was followed by a package designed specifically for the resident clinic, including incentives, gamification, workflow improvements, audit and feedback, and benchmarking.
Because the project was non-randomized and used sequential interventions, the study cannot show that the tailored package, or any one element, caused the rise. There was no concurrent control group to provide a separate comparison over the same period.
Why the clinic used a tailored approach
The prospective, non-randomized project took place in the Iowa City VA resident primary care clinic, which serves approximately 3,000 Veterans, including about 750 with diabetes. At the start of the project, 777 Veterans had diabetes listed on their problem list, while 604 met the electronic quality-measure criteria for the primary outcome.
The project aimed to move screening from 41% to 65% over eight months using a Lean quality improvement approach that included stakeholder interviews, process mapping and gap analysis. The authors hypothesized that earlier national efforts had been less effective in the resident clinic because they were not designed around trainees' status, the temporary nature of residency and residents' time demands.
The primary outcome was a monthly measure of annual screening status. A separate weekly process measure tracked eligible Veterans who had albuminuria testing within 1 day before or 7 days after a primary care appointment.
The larger gain followed the tailored phase
During the 12 months before the academic year running from July 2022 through June 2023, median screening performance was 38%. The run chart showed 44%, or 268 of 604 Veterans, by the end of October 2023. An upward trend began in November 2023, and the 65% target was reached by March 2024.
The weekly process measure showed a sharper shift. Its baseline mean was 16%, based on 74 of 477 Veterans across 20 data points from May through September 2023. After the intervention sequence, the new average was 48%, based on 171 of 359 Veterans.
The authors reported little to no impact from the national reminder before the multifaceted phase. They associated the larger improvement with the tailored phase and described the later 48% weekly-screening average as a new baseline three times higher than the earlier 16% level. They suggested that incentives, order-menu changes, audit and feedback, and education may have been important.
The process-control analysis recorded multiple special-cause signals, the study's label for unusual patterns. They began on November 6, 2023 and remained after all interventions were fully implemented in the week of February 19, 2024. One signal persisted from December 25, 2023 to January 8, 2024.
What the project leaves unanswered
The primary outcome was assessed with a median-centered run chart and visual analysis because the monthly observations were autocorrelated, meaning successive measurements were related. The process measure was assessed with a p-chart using three-sigma control limits and special-cause rules from the Institute for Healthcare Improvement. No confidence interval or p-value was reported for the primary outcome.
The findings come from one resident primary care clinic, so it is uncertain whether the same results would appear in other resident clinics, non-resident clinics or other VA sites. The sequential design also cannot rule out broader changes over time or other local circumstances contributing to the improvement.
The project did not collect balancing or downstream outcomes such as renal-protective medication prescribing and nephrology referrals. It therefore does not show whether better screening changed renal-protective prescribing, referrals or patient kidney outcomes.
The next test is replication
Whether the same result would hold elsewhere remains an open question. The study's single-clinic, sequential design means that replication and closer testing of individual components would be needed to judge generalizability and identify what sustained the improvement.
Paper data and sources
Original title: Improving diabetic kidney disease screening in a VA resident primary care clinic: a quality improvement initiative.
Authors: Kathie Zhang, Benjamin Reed Griffin, Jonathan B Chapman, Matthew D Soltys
Journal/Repository: BMJ open quality
Status: Peer-reviewed
First online: 2026-08-20
DOI: 10.1136/bmjoq-2026-004266
Original paper