Identifying Targets for Antibiotic Use for the Management of Carbapenem-Resistant Acinetobacter baumannii (CRAb) in Hospitals-A Multi-Centre Nonlinear Time-Series Study

Using time-series chest radiographs and laboratory data by machine learning for identifying pulmonary infection and colonization of Acinetobacter baumannii

BACKGROUND

Accurately distinguishing between pulmonary infection and colonization in patients with Acinetobacter baumannii is of utmost importance to optimize treatment and prevent antibiotic abuse or inadequate therapy. An efficient automated sorting tool could prompt individualized interventions and enhance overall patient outcomes. This study aims to develop a robust machine learning classification model using a combination of time-series chest radiographs and laboratory data to accurately classify pulmonary status caused by Acinetobacter baumannii.

METHODS

We proposed nested logistic regression models based on different time-series data to automatically classify the pulmonary status of patients with Acinetobacter baumannii. Advanced features were extracted from the time-series data of hospitalized patients, encompassing dynamic pneumonia indicators observed on chest radiographs and laboratory indicator values recorded at three specific time points.

RESULTS

Data of 152 patients with Acinetobacter baumannii cultured from sputum or alveolar lavage fluid were retrospectively analyzed. Our model with multiple time-series data demonstrated a higher performance of AUC (0.850, with a 95% confidence interval of [0.638-0.873]), an accuracy of 0.761, a sensitivity of 0.833. The model, which only incorporated a single time point feature, achieved an AUC of 0.741. The influential model variables included difference in the chest radiograph pneumonia score.

CONCLUSION

Dynamic assessment of time-series chest radiographs and laboratory data using machine learning allowed for accurate classification of colonization and infection with Acinetobacter baumannii. This demonstrates the potential to help clinicians provide individualized treatment through early detection.

Identification of antibiotic consumption targets for the management of Clostridioides difficile infection in hospitals- a threshold logistic modelling approach

BACKGROUND

This study aims to demonstrate the utility of a threshold logistic approach to identifying thresholds for specific antibiotic use associated with Clostridioides difficile infection (CDI) in an English teaching hospital.

METHODS

A combined approach of nonlinear modeling and logistic regression, named threshold logistic, was used to identify thresholds and risk scores in hospital-level antibiotic use associated with hospital-onset, healthcare-associated (HOHA) CDI cases.

RESULTS

Using a threshold logistic regression approach, an incidence greater than 0.2645 cases/1000 occupied bed-days (OBD; 85th percentile) was determined as the cutoff rate to define a critical (high) incidence rate of HOHA CDI. Fluoroquinolones and piperacillin-tazobactam were found to have thresholds at 84.8 and 54 defined daily doses (DDD)/1000 OBD, respectively. Analysis of data allowed calculating risk scores for HOHA CDI incidence rates exceeding the 85th percentile, i.e. entering critical incidence level. The threshold-logistic model also facilitated performing 'what-if scenarios' on future values of fluoroquinolones and piperacillin-tazobactam use to understand how HOHA CDI incidence rates may be affected.

CONCLUSION

Using threshold logistic analysis, critical incidence levels and antibiotic use targets to control HOHA CDI were determined. Threshold logistic models can be used to inform and enhance the effective design and implementation of antimicrobial stewardship programs.

A Threshold Logistic Modelling Approach for Identifying Thresholds between Antibiotic Use and Methicillin-Resistant Staphylococcus aureus Incidence Rates in Hospitals

The aim of this study was to demonstrate the utility of threshold logistic modelling, an innovative approach in identifying thresholds and risk scores in the context of population antibiotic use associated with methicillin-resistant Staphylococcus aureus (MRSA) incidence rates in hospitals. The study also aimed to assess the impact of exceeding those thresholds that resulted in increased MRSA rates. The study was undertaken in a 700-bed hospital in England between January 2015 and December 2021 (84 monthly observations). By employing the threshold logistic modelling approach, we: (i) determined the cut-off percentile value of MRSA incidence that defines a critical level of MRSA; (ii) identified thresholds for fluoroquinolone and co-amoxiclav use that would accelerate MRSA incidence rates and increase the probability of reaching critical incidence levels; (iii) enabled a better understanding of the effect of antibiotic use on the probability of reaching a critical level of resistant pathogen incidence; (iv) developed a near real-time performance monitoring feedback system; (v) provided risk scores and alert signals for antibiotic use, with the ability to inform hospital policies, and control MRSA incidence; and (vi) provided recommendations and an example for the management of pathogen incidence in hospitals. Threshold logistic models can help hospitals determine quantitative targets for antibiotic usage and can also inform effective antimicrobial stewardship to control resistance in hospitals. Studies should work toward implementing and evaluating the proposed approach prospectively, with the aim of determining the best counter-measures to mitigate the risk of increased resistant pathogen incidence in hospitals.

Control of Healthcare-Associated Carbapenem-Resistant Acinetobacter baumannii by Enhancement of Infection Control Measures

Antimicrobial stewardship and infection control measures are equally important in the control of antimicrobial-resistant organisms. We conducted a retrospective analysis of the incidence rate of hospital-onset carbapenem-resistant Acinetobacter baumannii (CRAB) infection (per 1000 patient days) in the Queen Mary Hospital, a 1700-bed, university-affiliated teaching hospital, from period 1 (1 January 2007 to 31 December 2013) to period 2 (1 January 2014 to 31 December 2019), where enhanced infection control measures, including directly observed hand hygiene before meal and medication rounds to conscious patients, and the priority use of single room isolation, were implemented during period 2. This study aimed to investigate the association between enhanced infection control measures and changes in the trend in the incidence rate of hospital-onset CRAB infection. Antimicrobial consumption (defined daily dose per 1000 patient days) was monitored. Interrupted time series, in particular segmented Poisson regression, was used. The hospital-onset CRAB infection increased by 21.3% per year [relative risk (RR): 1.213, 95% confidence interval (CI): 1.162−1.266, p < 0.001], whereas the consumption of the extended spectrum betalactam-betalactamase inhibitor (BLBI) combination and cephalosporins increased by 11.2% per year (RR: 1.112, 95% CI: 1.102−1.122, p < 0.001) and 4.2% per year (RR: 1.042, 95% CI: 1.028−1.056, p < 0.001), respectively, in period 1. With enhanced infection control measures, the hospital-onset CRAB infection decreased by 9.8% per year (RR: 0.902, 95% CI: 0.854−0.953, p < 0.001), whereas the consumption of the extended spectrum BLBI combination and cephalosporins increased by 3.8% per year (RR: 1.038, 95% CI: 1.033−1.044, p < 0.001) and 7.6% per year (RR: 1.076, 95% CI: 1.056−1.097, p < 0.001), respectively, in period 2. The consumption of carbapenems increased by 8.4% per year (RR: 1.84, 95% CI: 1.073−1.094, p < 0.001) in both period 1 and period 2. The control of healthcare-associated CRAB could be achieved by infection control measures with an emphasis on directly observed hand hygiene, despite an increasing trend of antimicrobial consumption.