Most AFS interventions preferentially target high-risk patients such as hemopoietic stem cell transplant (HSCT) recipients and acute leukemia patients as well as Candida infections which are easier to monitor than the more common but diagnostically challenging invasive mold diseases (IMD) like Aspergillosis. A recent systematic review showed that the effectiveness of most AFS programs is measured by savings in antifungal drug cost and consumption, a secondary consideration of AFS, rather than by indicators that actually reflect good quality of care such clinical outcomes, effectiveness of antifungal prophylaxis and processes including therapeutic drug monitoring, and speed of the diagnostic work-up.
Like antibiotics, overuse of antifungal drugs is common but largely driven by the high morbidity and mortality of IFD combined with insensitive diagnostic tools. Compared to current methods of clinical audit, semi-automated surveillance using NLP is more efficient and inclusive by avoiding restrictions based on any underlying hematologic condition, and has the added advantage of being potentially scalable.Īntifungal stewardship (AFS) is a growing area of clinical importance in hospitals worldwide that manage patients vulnerable to invasive fungal diseases (IFD). This is the first successful use of applied machine learning for institutional IMD surveillance across an entire hematology population describing process and outcome measures relevant to AFS. A random audit of 10% negative reports revealed two clinically significant misses (0.9%, 2/223). The average turnaround of send-away bronchoalveolar galactomannan of 12 days (range 7–22) was associated with high empiric liposomal amphotericin consumption. Fiberoptic bronchoscopy within 2 days of CT scan occurred in only 54% of episodes. Breakthrough-probable/proven-IMD on antifungal prophylaxis accounted for 60% of episodes with serum monitoring of voriconazole or posaconazole in the 2 weeks prior performed in only 53% and 69% of episodes, respectively. NLP screened 3014 reports from 1 September 2008 to 31 December 2017, generating 784 positives that after review, identified 205 IMD episodes (44% probable-proven) in 185 patients from 50,303 admissions. We used machine learning-based natural language processing (NLP) to non-selectively screen chest tomography (CT) reports for pulmonary IMD, verified by clinical review against international definitions and benchmarked against key AFS measures. This results in antifungal stewardship (AFS) programs preferentially reporting drug cost and consumption rather than measures that actually reflect quality of care. Clinical audit of invasive mold disease (IMD) in hematology patients is inefficient due to the difficulties of case finding.
0 kommentar(er)
