Abstract

Introduction Renal tract imaging after urinary tract infection (UTI) has been widely recommended but clinical practice varies substantially among paediatricians.

Aim To describe changes in knowledge and reported ordering practices of paediatricians in response to an evidence based summary about prevalence of abnormalities and test performance of renal tract imaging, in the setting of UTI in children.

Methods 354 paediatricians were randomly selected from a register of Australasian physicians and surveyed 14 months before, and concurrent with, a summary of a relevant systematic review. Respondents' estimates were dichotomised and labelled as correct when within 5% of the evidence-based value. Frequency of correct responses was compared using McNemar's test for paired proportions.

Results Response rate for the return of both surveys was 61% (215/354). Provision of the evidence summary significantly improved knowledge of the frequencies of associated renal tract abnormalities (vesicoureteric reflux and kidney damage), with an increase in correct responses of about 30% post summary (p<0.001 for reflux and damage). Prior to the summary, clinicians underestimated the sensitivity of all imaging tests for the diagnosis of renal damage and reflux by about 30%, with an increase in correct responses of 30–50% for all tests after the summary (p<0.001 for all). In contrast, reported imaging practices for all tests showed no significant change in practice after receipt of the evidence summary.

Conclusions Provision of evidence based information on rates of abnormality and test sensitivity improved knowledge but did not result in any significant change in reported practice. Properties of diagnostic tests conventionally thought to modify use, sensitivity and likelihood of detecting abnormalities, did not influence test ordering practices.