OBJECTIVE Evaluate the prevalence of bacterial cultures for patients with new-onset fever or instability obtained that were concordant with a new clinical decision support (CDS) tool within and across sociodemographic characteristics. PATIENTS AND METHODS A retrospective, cross-sectional evaluation assessing the prevalence of CDS-concordant cultures (blood, urine, respiratory cultures) obtained in an academic pediatric and pediatric cardiac intensive care unit 12 months before and after implementation of the new CDS tool. We compared CDS concordance before and after implementation within sociodemographic subgroups: race, ethnicity, primary language, payor type, biological sex, and age. Next, we compared CDS concordance across sociodemographic categories, before and after the intervention (eg, female vs male), using prevalence ratios (PRs). RESULTS We analyzed 2838 cultures (1503 pre and 1335 post). There was an improvement in CDS concordance in 13 of the 20 sociodemographic subgroups when comparing the postintervention with the preintervention period. Comparing across sociodemographic categories, there were no differences in CDS-concordant testing prevalence across race, ethnicity, primary language, or sex. However, we identified differences in CDS concordance by age. Before the intervention, cultures in patients younger than 6 months had 18% lower CDS concordance than patients older than 15 years (reference group) (PR 0.82, 95% CI 0.69–0.98); there was no difference after the intervention. Similarly, cultures in patients aged between 5 and 15 years had 23% lower CDS concordance than patients older than 15 years (PR 0.77, 95% CI 0.63–0.96); this attenuated after the intervention (PR 0.86, 95% CI 0.74–0.99). CONCLUSIONS A new CDS tool was associated with an increase in CDS-concordant bacterial culture practices within most sociodemographic subgroups and enhanced culture ordering practices among younger patients.
Booth et al. (Tue,) studied this question.