Abstract
Rapid molecular diagnostic tests improve antimicrobial stewardship (AMS) by facilitating earlier refinement of antimicrobial therapy. The INHALE trial tested the application of the BioFire FilmArray Pneumonia Panel (Pneumonia Panel) for antibiotic prescribing for hospital-acquired and ventilator-associated pneumonias (HAP/VAP) in UK intensive care units (ICUs). We report a behavioral study embedded within the INHALE trial examining clinicians’ perceptions of using these tests. Semi-structured interviews were conducted with 20 ICU clinicians after using the Pneumonia Panel to manage suspected HAP/VAP. Thematic analysis identified factors reinforcing perceptions of the necessity to modify antibiotic prescribing in accordance with test results and doubts/concerns about doing so. While most acknowledged the importance of AMS, the test’s impact on prescribing decisions was limited. Concerns about potential consequences of undertreatment to the patient and prescriber were often more salient than AMS, sometimes leading to “just-in-case” antibiotic prescriptions. Test results indicating a broad-spectrum antibiotic were unnecessary often failed to influence clinicians to avoid an initial prescription or de-escalate antibiotics early as they considered their use to be necessary to protect the patient and themselves, “erring on the side of caution.” Some clinicians described cases where antibiotics would be prescribed for a sick patient regardless of test results because, in their opinion, it fits with the clinical picture—“treating the patient, not the result.” Our findings illustrate a tension between prescribing guidelines and clinicians’ “mindlines,” characterized by previous experiences. This highlights the need for a “technology plus” approach, recognizing the challenges clinicians face when applying technological solutions to patient care.IMPORTANCERapid molecular diagnostic tests for pathogens and resistance genes may improve antibiotic-prescribing decisions and stewardship. However, clinicians’ desire to protect their patients with antibiotics often overrides more distal concerns about possible resistance selection, limiting the application of these tests in practice. Findings underscore the challenge of changing prescribing decisions based on technical results or guidelines, highlighting factors such as clinicians’ previous experience and “knowledge in practice” as more proximal drivers of these decisions. Implementation strategies for technological solutions to antimicrobial resistance must be “behaviorally intelligent,” recognizing the challenges facing clinicians when making “life or death” prescribing decisions.CLINICAL TRIALSThis study is registered with ISRCTN as ISRCTN16483855.