Awareness, use, attitudes, and implementation challenges of Bayesian dose prediction software in clinical practice: a...

Bayesian dose prediction software supports model-informed precision dosing to improve therapeutic outcomes. This study evaluated clinicians' awareness, usage, attitudes, and perceived barriers related to Bayesian dosing software.

A cross-sectional study was conducted using a validated electronic questionnaire distributed through eight international professional organizations. Descriptive statistics, non-parametric tests, and logistic regression analyses were performed using SPSS version 29.

Of 234 respondents, 190 (81.2%) were eligible and completed the survey. Most were from North America (67.4%), pharmacists (76.3%), and worked in academic hospitals (78.4%). Awareness of Bayesian dosing software was high (82.6%) among respondents, yet only 65% reported its use in clinical practice. The most used software included InsightRx (40.2%), DoseMe (22.5%), and MWPharm (18.6%), mainly for vancomycin (94.1%) and aminoglycosides (50%). The overall median attitude score (IQR) was 4.0 (2.0), indicating positive attitudes, whereas the barrier score was 3.0 (2.0), reflecting high implementation challenges. The most frequently reported barriers were prohibitive licensing costs (65%), lack of institutional support for use and maintenance (50.3%), and limited awareness of its use (49.7%). In multivariable analysis, "region" was the sole predictor of awareness with participants from the Middle East significantly less aware (AOR = 0.14; 95% CI: 0.05-0.42; p = 0.002). Software type was significantly associated with professional role, with pharmacists more likely to select commonly used tools (AOR = 7.2; 95% CI: 1.4-17.9; p = 0.003). A non-significant negative correlation was observed between overall attitude and barrier scores (r = -0.49; p = 0.072).

Despite high awareness and positive attitudes toward Bayesian dosing software, its clinical use remains limited, primarily focused on vancomycin and aminoglycosides, mostly restricted to three commercial tools. Targeted interventions are required to address key implementation barriers related to licensing costs, institutional support, and awareness.