Prescribers’ Information Needs for Antibiotic De-escalation in Nursing Homes

Background: Antibiotics are commonly prescribed empirically in nursing homes (NHs), often by cross-covering prescribers. While there is some understanding of prescribing decision-making, factors influencing post-prescribing decisions to de-escalate therapy remain poorly understood.

Methods: Researchers conducted interviews with 15 prescribers (8 physicians, 7 nurse practitioners) in Wisconsin. Ten interviews informed development of an antibiotic timeout review tool to support nurse-prescriber communication. Five interviews simulated post-prescribing decision-making using four scenarios, with one researcher role-playing a nurse without using the tool. Another probing the respondent to think aloud through their decisions and information needs. Transcripts were analyzed thematically to assess information needs addressed and not addressed in the communication tool.

Results: Our analyses uncovered several key findings: 1) primary providers are rarely notified of prescriptions started by colleagues; 2) prescribers often make initiation and modification decisions without directly assessing the resident; 3) some are uncomfortable changing broad-spectrum prescriptions by colleagues even if they would have prescribed a more narrow-spectrum antibiotic; 4) prescribers need to know whose prescription they are changing, and nurse practitioners appear less comfortable modifying physicians’ prescriptions; 5) justification of initial testing and antibiotic orders is often missing, including original symptoms, and intended infection to treat; 6) modification decisions require details on original and current symptoms as well as lab and culture results.

Conclusion: Antibiotic de-escalation in NHs is highly dependent on information availability and presentation due to time pressures and the scarcity of opportunities for synchronous communication in this setting. Thus prescribers are willing to make modification decisions despite information they need being missing. Without high-quality information, prescribers are less comfortable challenging colleagues’ initial prescriptions they disagree with, especially in terms of de-escalation. Structured communication tools that incorporate decision support to aid antibiotic de-escalation can address the missing information that prescribers need.