Leveraging electronic medical records to evaluate a computerized decision support system for staphylococcus bacteremia

Abstract

Infectious disease specialists (IDS) improve outcomes of patients with Staphylococcus bacteremia, but immediate IDS access is not always guaranteed. We investigated whether a care-integrated computerized decision support system (CDSS) can safely enhance the standard of care (SOC) for these patients. We conducted a multicenter, noninferiority, interventional stepped-wedge cluster randomized controlled trial relying on the data integration centers at five university hospitals. By this means, electronic medical records can be used for part of the trial documentation. We analyzed 5056 patients from 134 wards (Staphylococcus aureus (SAB): n = 812, coagulase-negative staphylococci (CoNS): n = 4244) and found that the CDSS was noninferior to the SOC for hospital mortality in all patients. Noninferiority regarding the 90-day mortality/relapse in SAB patients was not observed and there was no evidence for differences in vancomycin usage among CoNS patients. Despite low reported usage, physicians rated the CDSS's usability favorably. Trial registration: drks.de; Identifier: DRKS00014320; Registration Date: 2019-05-06.

Fig. 1. Recruitment CONSORT flowchart.

Due to the stepped-wedge design, each ward took part in the SOC phase at the beginning of the trial and switched to the CDSS phase at a randomized crossover point in time, thus contributing patients from both phases to the analysis. Numbers of patients are displayed for each type of bacteremia separately.

Fig. 2. Differences in endpoint probability between the SOC and CDSS phases.

The gray boxes show the area of noninferiority (=5% of the SOC phase probability). Overlapping CIs indicate that noninferiority could not be established, while the CIs left of the gray box indicate noninferiority. Confidence intervals: 90% for noninferiority hypotheses and 95% for superiority hypotheses.

Fig. 3. Usability rating of the HELP CDSS.

Physicians rated 10 statements on the CDSS’s usability, using a scale from “strongly agree” to “strongly disagree.” Each bar shows the percentage of physicians who agreed with each category for each statement.

Fig. 4. Impact of the CDSS on clinical decisions and patient safety.

Physicians rated 10 statements on their clinical decisions and patient safety, using a scale from “strongly agree” to “strongly disagree.” Each bar shows the percentage of physicians who agreed with each category for each statement.

Fig. 5. Schematic representation of the HELP CDSS.

Rough overview of the decision algorithm that was the basis for the HELP CDSS.

Fig. 6. Data extraction process.

Starting with FHIR data available in each DIC, each step of the data extraction required iterative adaptions until the local analysis result was ready to be sent to the central analysis site.