Facility Variation of Follow-Up Imaging Practices Among U.S. Veterans for Urinary Stone Disease

Abstract

Purpose:

To assess variation in surveillance imaging practices in various clinical contexts among patients with urinary stone disease (USD) from a large cohort of U.S. Veterans.

Materials and Methods:

We identified adults age ≥18 years within the Veterans Health Administration Corporate Data Warehouse with an index surgical procedure related to USD or an index emergency department (ED) visit with a primary diagnosis of USD between 2010 and 2018. We then explored three clinical periods for use of surveillance imaging: 30–180 days and 180–540 days post-surgery and 90–540 days post-ED visit. We estimated multilevel logistic regression models to identify factors associated with receipt of surveillance imaging. We calculated median odds ratios (MORs) to quantify the amount of variation in imaging procedures by facility.

Results:

Among 35,248 patients who met the inclusion criteria, 10,095 underwent a surgical procedure, and 25,153 had an ED visit. Probabilities of obtaining postoperative imaging 30–180 days post-surgery by facility ranged from 8.4% to 58.5%; from 6.8% to 35.4% for 180–540 days post-surgery; and from 6.1% to 20.5% for 90–540 days post-ED. The greatest between-facility variation occurred 30–180 days post-surgery (MOR: 1.81 [95% confidence interval], 1.66–2.04). There was also wide variation in the type of post-surgical imaging. Combination imaging, X-ray, and ultrasound were the most common imaging modalities obtained in the 30–180 day period for percutaneous nephrolithotomy, shockwave lithotripsy, and ureteroscopy, respectively.

Conclusions:

Among patients with USD, the use of surveillance imaging after surgical procedure and in the post-ED visit setting remains low, with substantial variation across facilities.