Population-based trend analysis of voiding cystourethrogram ordering practices in a single-payer healthcare system before and after the release of evaluation guidelines.

ABSTRACT

INTRODUCTION: While voiding cystourethrogram (VCUG) is a widely-accepted test, it is invasive and associated with radiation exposure. Most cases of primary vesicoureteral reflux (VUR) are low-grade and unlikely to be associated with acquired renal scarring. To select patients at greatest risk, in 2011 the American Academy of Pediatrics (AAP) published guidelines for evaluation of children ages 2 - 24 months with urinary tract infections (UTIs). Similarly, in 2010 the Society for Fetal Urology (SFU) published guidelines for patients with hydronephrosis. Herein a prospectively-collected database was queried through the Institute of Clinical Evaluative Sciences (ICES), exploring trends in VCUG ordering within the Ontario Health Insurance Program (OHIP), which guarantees universal access to care. MATERIAL AND METHODS: A dedicated ICES analyst extracted data on all patients younger than 18 years in Ontario, Canada, with billing codes for VCUG and ICD-9 codes for VUR, from 2004-2014. The baseline characteristics included patient age, gender, geographic region, specialty of ordering provider and previous diagnoses of UTI and/or antenatal hydronephrosis to determine the indication for ordering the test. Of these, patients were subsequently incurred OHIP procedure codes for endoscopic injection or ureteral reimplantation. Patients who had a VCUG in the setting of urethral trauma, posterior urethral valves, and neurogenic bladder were excluded. RESULTS AND DISCUSSION: Trend analysis demonstrated that the total number of VCUGs ordered in the province has decreased over a decade (Figure 1), with a concurrent decrease in VUR diagnosis. On multivariate regression analysis, the decrease in VCUG ordering could not be explained by changes in population demographics or other baseline patient variables. Most VCUGs obtained per year were ordered by pediatricians or family physicians (mean 2,022+523.8), compared with urologists and nephrologists (mean 616+358.3). Interestingly, while the rate of VCUG requests decreased, the annual number of surgeries performed for VUR (endoscopic or open) did not show a significant reduction over time. CONCLUSIONS: We present a large population-based analysis in a universal access to care system, reporting a decreasing trend in the number of cystograms and differences by primary care versus specialist providers. While it is reassuring to see practice patterns favorably impacted by guidelines, it is also encouraging to note that the number of surgeries has remained stable. This suggests that patients at risk continue to be detected and offered surgical correction. These data confirm previous institution-based assessments and affirm changes in VCUG ordering independent of variables not relevant to the healthcare system, such as the insurance status.