BackgroundHealth Care Providers (HCPs) report that manual techniques of intravascular fluid resuscitation are commonly used during pediatric shock management. The optimal pediatric fluid resuscitation technique is currently unknown. We sought to determine HCP test-retest reliability (repeatability) and inter-subject variability of fluid resuscitation performance outcomes to inform the design of future studies.MethodsFifteen consenting HCPs from McMaster Children’s Hospital, in Hamilton, Canada participated in this single-arm interventional trial. Participants were oriented to a non-clinical model representing a 15 kg toddler, which incorporated a 22-gauge IV catheter. Following a standardization procedure, participants administered 600 mL (40 mL/kg) of saline to the simulated child under emergency conditions using prefilled 60-mL syringes. Each participant completed 5 testing trials. All testing was video recorded, with fluid administration time outcome data (in seconds) extracted from trial videos by two blinded outcome assessors. Data describing catheter dislodgement events, volume of saline effectively delivered, and participant demographics were also collected. The primary outcome of fluid administration time test-retest reliability was analyzed by one-way analysis of variance (ANOVA) and intra-class correlation (ICC), with good reliability defined as ICC > 0.70.ResultsDifferences in HCP fluid administration times are attributable to inter-subject variability rather than intra-subject variability based on one-way ANOVA analysis, F (14,60) = 43.125; p < 0.001. Test-retest reliability of subjects was excellent with ICC = 0.97 (95% CI: 0.95-0.99); p < 0.001.ConclusionsFindings demonstrate excellent test-retest reliability of HCP fluid resuscitation performance in a setting involving a non-clinical model. Investigators can justify a single evaluation of HCP performance in future studies.Electronic supplementary materialThe online version of this article (doi:10.1186/1756-0500-7-724) contains supplementary material, which is available to authorized users.