What the study found
Peer comparative inpatient prescribing reports were not associated with a change in hospitalists' prescribing rates of broad-spectrum antibiotics for hospital-onset infections. The authors also noted that additional efforts may be needed to make these reports more useful.
Why the authors say this matters
The authors state that antibiotic overuse is harmful to patients and the health care system, and they suggest that better ways to use prescribing feedback in inpatient settings are needed. They conclude that additional efforts to augment the utility of these reports are justified.
What the researchers tested
This quality improvement study used a stepped-wedge cluster design across 5 acute care facilities in one large health care network. Hospitalists received bimonthly email reports comparing their observed-to-expected ratios for days of therapy, which is a measure of antibiotic use, and the study examined broad-spectrum antibiotics for hospital-onset infections, including those active against Pseudomonas aeruginosa.
What worked and what didn't
The intervention was not significantly associated with lower prescribing rates after accounting for time, clustering of hospitalists, and patient characteristics. In multivariable models, the intervention effect was RR 0.97 (95% CI, 0.91-1.04), while prescribing was higher among clinicians caring for more patients with sepsis and end-stage kidney disease and lower across sequential reporting periods.
What to keep in mind
This summary does not describe any limitations beyond the study being conducted within one health care network. The abstract reports the setting, design, and measured outcomes, but it does not provide further detail about why the reports did not change prescribing behavior.
Key points
- Peer comparative inpatient antibiotic prescribing reports were not linked to lower hospitalist prescribing rates.
- The study examined broad-spectrum antibiotics for hospital-onset infections across 5 acute care facilities.
- The intervention used bimonthly email reports with observed-to-expected ratios for days of therapy.
- In multivariable models, the intervention estimate was RR 0.97 (95% CI, 0.91-1.04).
- The authors conclude that additional efforts to improve the usefulness of these reports are justified.
Disclosure
- Research title:
- Peer comparison reports did not change hospitalist antibiotic prescribing
- Authors:
- L Witt, Radhika Prakash-Asrani, K Ashley Jones, C. Christina Mehta, Zanthia Wiley, Jesse T. Jacob, Hasan Shabbir, Julianne Gent, Chad Robichaux, J Howard-Anderson, Sujit Suchindran, Julia E. Szymczak, Mph Raymund B. Dantes, Scott K. Fridkin
- Institutions:
- Emory University, Emory Healthcare, Emory University Hospital, University of Utah
- Publication date:
- 2026-04-28
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.