Variation in the intensity of hematuria evaluation: a target for primary care quality improvement

Abstract Background Hematuria is a common clinical finding and represents the most frequent presenting sign of bladder cancer. The American Urological Association recommends cystoscopy and abdomino-pelvic imaging for patients aged more than 35 years. Nonetheless, less than half of patients presenting with hematuria undergo proper evaluation. We sought to identify clinical and nonclinical factors associated with evaluation of persons with newly diagnosed hematuria. Methods We performed a retrospective cohort study, using claims data and laboratory values. The primary exposure was practice site, as a surrogate for nonclinical, potentially modifiable sources of variation. Primary outcomes were cystoscopy or abdomino-pelvic imaging within 180 days after hematuria diagnosis. We modeled the association between clinical and nonclinical factors and appropriate hematuria evaluation. Results We identified 2455 primary care patients aged 40 years or more and diagnosed with hematuria between 2004 and 2012 in the absence of other explanatory diagnosis; 13.7% of patients underwent cystoscopy within 180 days. Multivariate logistic regression revealed significant variation between those who did and did not undergo evaluation in age, gender, and anticoagulant use ( P P  = .036, P  = .028, respectively). Addition of practice site improved the predictive discrimination of each model ( P Conclusions Patients with hematuria rarely underwent complete evaluation. Although established risk factors for malignancy were associated with increasing use of diagnostic testing, factors unassociated with risk, such as practice site, also accounted for significant variation. Inconsistency across practice sites is undesirable and may be amenable to quality improvement interventions.