CDS software improves patient outcomes as well as practitioner performance

Computerized clinical decision support has strong upsides and few to no downsides for both clinicians and patients, according to a systematic literature review conducted at Texas State University’s school of health administration.

Analyzing a small but representative set of articles meeting the review’s criteria, Clemens Scott Kruse, PhD, MHA, MBA, and Nolan Ehrbar, BHA, made several key findings:

  • 61% of 36 articles described occurrences of positive patient medical outcomes, while
    • 8% found no statistically significant difference in medical outcomes between intervention and control groups, and
    • 31% did not report or discuss medical outcomes.
  • 66% of 38 articles documented occurrences of positive practitioner performance, while
    • 13% found no difference in practitioner performance, and
    • 21% did not report or discuss practitioner performance.

Similarly encouraging for clinical CDS proponents: Zero articles turned up reporting negative patient outcomes or negative practitioner performance associated with the use of CDS.

Further, although nine articles did not discuss practitioner performance, only five articles reported no difference in productivity.

The full findings are posted in the August edition of JMIR Medical Informatics.

In their discussion, Kruse and Ehrbar note the dearth of previous CDS literature reviews that included data on patient outcomes. The lack may have owed to limitations in search strategies or the relative newness of the technology in clinical settings, they suggest.

They credit their review methodology for ferreting out studies with data on both outcomes and performance.  

Also curious for its absence in the presently reviewed qualified articles, they add, was alert fatigue.

This phenomenon is often “attributed to medical error in the areas of pharmacy and physician ordering systems, which are common attributes in computerized CDS systems,” Kruse and Ehrbar point out. “Even in clinical trials, alert fatigue is known to be persistent over time. It is interesting that it was not noted, and if it was not noted, it was not controlled for in the studies analyzed.”

Based on the shortfall, they call for additional in-depth literature review to examine such aspects of CDS as whether clinicians avoid the tools due to the possible “piling on” of yet another set of alerts.

“Decision-support tools extend beyond the practitioner to the patient, and some tools are not software-based but based on patient-reported data,” Kruse and Ehrbar conclude. “The implementation of computerized CDS systems can mutually benefit the practitioner and patient, and they show great promise for healthcare in the future.”

The study is posted in full for free.