ABSTRACT
We present an ensemble transfer learning method to predict suicide from Veterans Affairs (VA) electronic medical records (EMR). A diverse set of base models was trained to predict a binary outcome constructed from reported suicide, suicide attempt, and overdose diagnoses with varying choices of study design and prediction methodology. Each model used twenty cross-sectional and 190 longitudinal variables observed in eight time intervals covering 7.5 years prior to the time of prediction. Ensembles of seven base models were created and fine-tuned with ten variables expected to change with study design and outcome definition in order to predict suicide and combined outcome in a prospective cohort. The ensemble models achieved c-statistics of 0.73 on 2-year suicide risk and 0.83 on the combined outcome when predicting on a prospective cohort of [Formula: see text] 4.2 M veterans. The ensembles rely on nonlinear base models trained using a matched retrospective nested case-control (Rcc) study cohort and show good calibration across a diversity of subgroups, including risk strata, age, sex, race, and level of healthcare utilization. In addition, a linear Rcc base model provided a rich set of biological predictors, including indicators of suicide, substance use disorder, mental health diagnoses and treatments, hypoxia and vascular damage, and demographics.
Subject(s)
Carcinoma, Renal Cell , Kidney Neoplasms , Veterans , Humans , Veterans/psychology , Retrospective Studies , Cross-Sectional Studies , Prospective Studies , Suicide, Attempted , Machine LearningABSTRACT
INTRODUCTION: Recent evidence has shown that oral antibiotic therapy is not inferior to IV antibiotic therapy in the treatment of complicated Staphylococcus aureus infections. Therefore, oral antibiotic therapy is now frequently prescribed in clinical practice due to cost benefit, ease of administration, decreased complication rate, and lack of need for IV access. In vitro susceptibility testing for ß-lactam oral antibiotics is not routinely performed as the guidelines provided by the Clinical and Laboratory Standards Institute (CLSI) recommend using oxacillin and cefoxitin as surrogate markers. Hence, oral antibiotic susceptibilities for cephalexin and dicloxacillin are not reported and implied based on oxacillin and cefoxitin. The objective of the current study was to determine whether susceptibilities among S. aureus isolates are predictable when comparing commonly used IV and oral beta-lactams. METHODS: Cefazolin, cephalexin, dicloxacillin, and oxacillin broth microdilution minimum inhibitory concentrations (MICs) were determined for 100 clinical isolates of methicillin-sensitive S. aureus by broth microdilution following CLSI guidelines. RESULTS: Among these isolates, median MICs for cephalexin were eight-fold higher than cefazolin MICs and median MICs for dicloxacillin were four-fold less than oxacillin MICs. Ten percent of more strains studied had a major or very major error in its susceptibility reporting when cephalexin was compared to its surrogate marker oxacillin. DISCUSSIONS/CONCLUSIONS: The variations in MICs observed compounded with the dosing and pharmacokinetic differences of oral versus IV ß-lactam suggests that establishing breakpoints for oral ß-lactam antibiotics is necessary to ensure adequate therapy is selected for the treatment of complex S. aureus infections.
Subject(s)
Methicillin-Resistant Staphylococcus aureus , Staphylococcal Infections , Humans , Cefoxitin/pharmacology , Cefoxitin/therapeutic use , beta-Lactams/pharmacology , beta-Lactams/therapeutic use , Cefazolin/pharmacology , Cefazolin/therapeutic use , Staphylococcus aureus , Dicloxacillin/therapeutic use , Anti-Bacterial Agents/pharmacology , Anti-Bacterial Agents/therapeutic use , Oxacillin/pharmacology , Oxacillin/therapeutic use , Staphylococcal Infections/drug therapy , Microbial Sensitivity Tests , Cephalexin/pharmacology , Cephalexin/therapeutic use , Monobactams/therapeutic useABSTRACT
Many countries that had early access to COVID-19 vaccines implemented vaccination strategies that prioritized health care workers and the elderly. As barriers to access eased, vaccine prioritization strategies have been relaxed. However, these strategies are still an important tool for decision makers to manage new variants, plan for future booster shots, or stage mass vaccinations. This paper explores the impact of vaccine prioritization strategies using networks that represent communities with different demographics and connectivity. The impact of vaccination is compared to non-medical intervention to reduce transmission. Several sources of uncertainty are considered, including vaccine willingness and mask effectiveness. This paper finds that while prioritization strategies can have a large impact on reducing deaths and peak hospitalization, selecting the best strategy depends on community characteristics and the desired objective. Additionally, in some cases random vaccination performs as well as more targeted prioritization strategies. Understanding these trade-offs is important when planning vaccine distribution.
ABSTRACT
INTRODUCTION: Community-acquired urinary tract infections (UTIs) caused by extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli have limited oral therapeutic options and pose significant clinical challenges. The goal of this study was to evaluate the in vitro synergy between CFM and AMC against ESBL E. coli with aims to identify an oral treatment option for UTIs. METHODS: Minimum inhibitory concentrations (MICs) of CFM in the presence of AMC were determined for 46 clinical isolates by placing a CFM Etest on a plate with AMC impregnated in the agar. Isolates with CFM MIC ≤1 µg/mL in the presence of AMC were considered susceptible to the CFM and AMC combination. Five isolates were then selected for further testing using time-kill analysis in the presence of CFM, AMC, and CFM with AMC. Time-kill curves were plotted to determine synergy over 24 h. RESULTS: AMC improved the activity of CFM against ESBL E. coli isolates by 128-fold in the Etest analysis with 85% of tested isolates being susceptible to the combination. A fourfold or greater reduction in CFM MIC was exhibited in 44 of 46 (96%) isolates when in the presence of AMC. Synergy and bactericidal activity between CFM and AMC were exhibited in each of the five isolates tested by time-kill analysis. DISCUSSION/CONCLUSION: This study found that AMC improves the activity of CFM against ESBL E. coli and that this antibiotic combination has potential as an oral therapeutic option to treat ESBL E. coli UTIs.