External Validation and Comparison of a General Ward Deterioration Index Between Diversely Different Health Systems.

OBJECTIVES: Implementing a predictive analytic model in a new clinical environment is fraught with challenges. Dataset shifts such as differences in clinical practice, new data acquisition devices, or changes in the electronic health record (EHR) implementation mean that the input data seen by a model can differ significantly from the data it was trained on. Validating models at multiple institutions is therefore critical. Here, using retrospective data, we demonstrate how Predicting Intensive Care Transfers and other UnfoReseen Events (PICTURE), a deterioration index developed at a single academic medical center, generalizes to a second institution with significantly different patient population. DESIGN: PICTURE is a deterioration index designed for the general ward, which uses structured EHR data such as laboratory values and vital signs. SETTING: The general wards of two large hospitals, one an academic medical center and the other a community hospital. SUBJECTS: The model has previously been trained and validated on a cohort of 165,018 general ward encounters from a large academic medical center. Here, we apply this model to 11,083 encounters from a separate community hospital. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The hospitals were found to have significant differences in missingness rates (> 5% difference in 9/52 features), deterioration rate (4.5% vs 2.5%), and racial makeup (20% non-White vs 49% non-White). Despite these differences, PICTURE's performance was consistent (area under the receiver operating characteristic curve [AUROC], 0.870; 95% CI, 0.861-0.878), area under the precision-recall curve (AUPRC, 0.298; 95% CI, 0.275-0.320) at the first hospital; AUROC 0.875 (0.851-0.902), AUPRC 0.339 (0.281-0.398) at the second. AUPRC was standardized to a 2.5% event rate. PICTURE also outperformed both the Epic Deterioration Index and the National Early Warning Score at both institutions. CONCLUSIONS: Important differences were observed between the two institutions, including data availability and demographic makeup. PICTURE was able to identify general ward patients at risk of deterioration at both hospitals with consistent performance (AUROC and AUPRC) and compared favorably to existing metrics.

Risk Factors for Klebsiella Infections among Hospitalized Patients with Preexisting Colonization.

Klebsiella commonly colonizes the intestinal tract of hospitalized patients and is a leading cause of health care-associated infections. Colonization is associated with subsequent infection, but the factors determining this progression are unclear. A cohort study was performed, in which intensive care and hematology/oncology patients with Klebsiella colonization based on rectal swab culture were enrolled and monitored for infection for 90 days after a positive swab. Electronic medical records were analyzed for patient factors associated with subsequent infection, and variables of potential significance in a bivariable analysis were used to build a final multivariable model. Concordance between colonizing and infecting isolates was assessed by wzi capsular gene sequencing. Among 2,087 hospitalizations from 1,978 colonized patients, 90 cases of infection (4.3%) were identified. The mean time to infection was 20.6 ± 24.69 (range, 0 to 91; median, 11.5) days. Of 86 typed cases, 68 unique wzi types were identified, and 69 cases (80.2%) were colonized with an isolate of the same type prior to infection. Based on multivariable modeling, overall comorbidities, depression, and low albumin levels at the time of rectal swab collection were independently associated with subsequent Klebsiella infection (i.e., cases). Despite the high diversity of colonizing strains of Klebsiella, there is high concordance with subsequent infecting isolates, and progression to infection is relatively quick. Readily accessible data from the medical record could be used by clinicians to identify colonized patients at an increased risk of subsequent Klebsiella infection. IMPORTANCE Klebsiella is a leading cause of health care-associated infections. Patients who are intestinally colonized with Klebsiella are at a significantly increased risk of subsequent infection, but only a subset of colonized patients progress to disease. Colonization offers a potential window of opportunity to intervene and prevent these infections, if the patients at greatest risk could be identified. To identify patient factors associated with infection in colonized patients, we studied 1,978 colonized patients. We found that patients with a higher burden of underlying disease in general, depression in particular, and low albumin levels in a blood test were more likely to develop infection. However, these variables did not completely predict infection, suggesting that other host and microbial factors may also be important. The clinical variables associated with infection are readily available in the medical record and could serve as the foundation for developing an integrated risk assessment of Klebsiella infection in hospitalized patients.

Deep learning to detect acute respiratory distress syndrome on chest radiographs: a retrospective study with external validation.

BACKGROUND: Acute respiratory distress syndrome (ARDS) is a common, but under-recognised, critical illness syndrome associated with high mortality. An important factor in its under-recognition is the variability in chest radiograph interpretation for ARDS. We sought to train a deep convolutional neural network (CNN) to detect ARDS findings on chest radiographs. METHODS: CNNs were pretrained on 595 506 radiographs from two centres to identify common chest findings (eg, opacity and effusion), and then trained on 8072 radiographs annotated for ARDS by multiple physicians using various transfer learning approaches. The best performing CNN was tested on chest radiographs in an internal and external cohort, including a subset reviewed by six physicians, including a chest radiologist and physicians trained in intensive care medicine. Chest radiograph data were acquired from four US hospitals. FINDINGS: In an internal test set of 1560 chest radiographs from 455 patients with acute hypoxaemic respiratory failure, a CNN could detect ARDS with an area under the receiver operator characteristics curve (AUROC) of 0·92 (95% CI 0·89-0·94). In the subgroup of 413 images reviewed by at least six physicians, its AUROC was 0·93 (95% CI 0·88-0·96), sensitivity 83·0% (95% CI 74·0-91·1), and specificity 88·3% (95% CI 83·1-92·8). Among images with zero of six ARDS annotations (n=155), the median CNN probability was 11%, with six (4%) assigned a probability above 50%. Among images with six of six ARDS annotations (n=27), the median CNN probability was 91%, with two (7%) assigned a probability below 50%. In an external cohort of 958 chest radiographs from 431 patients with sepsis, the AUROC was 0·88 (95% CI 0·85-0·91). When radiographs annotated as equivocal were excluded, the AUROC was 0·93 (0·92-0·95). INTERPRETATION: A CNN can be trained to achieve expert physician-level performance in ARDS detection on chest radiographs. Further research is needed to evaluate the use of these algorithms to support real-time identification of ARDS patients to ensure fidelity with evidence-based care or to support ongoing ARDS research. FUNDING: National Institutes of Health, Department of Defense, and Department of Veterans Affairs.

Predicting Intensive Care Transfers and Other Unforeseen Events: Analytic Model Validation Study and Comparison to Existing Methods.

BACKGROUND: COVID-19 has led to an unprecedented strain on health care facilities across the United States. Accurately identifying patients at an increased risk of deterioration may help hospitals manage their resources while improving the quality of patient care. Here, we present the results of an analytical model, Predicting Intensive Care Transfers and Other Unforeseen Events (PICTURE), to identify patients at high risk for imminent intensive care unit transfer, respiratory failure, or death, with the intention to improve the prediction of deterioration due to COVID-19. OBJECTIVE: This study aims to validate the PICTURE model's ability to predict unexpected deterioration in general ward and COVID-19 patients, and to compare its performance with the Epic Deterioration Index (EDI), an existing model that has recently been assessed for use in patients with COVID-19. METHODS: The PICTURE model was trained and validated on a cohort of hospitalized non-COVID-19 patients using electronic health record data from 2014 to 2018. It was then applied to two holdout test sets: non-COVID-19 patients from 2019 and patients testing positive for COVID-19 in 2020. PICTURE results were aligned to EDI and NEWS scores for head-to-head comparison via area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve. We compared the models' ability to predict an adverse event (defined as intensive care unit transfer, mechanical ventilation use, or death). Shapley values were used to provide explanations for PICTURE predictions. RESULTS: In non-COVID-19 general ward patients, PICTURE achieved an AUROC of 0.819 (95% CI 0.805-0.834) per observation, compared to the EDI's AUROC of 0.763 (95% CI 0.746-0.781; n=21,740; P<.001). In patients testing positive for COVID-19, PICTURE again outperformed the EDI with an AUROC of 0.849 (95% CI 0.820-0.878) compared to the EDI's AUROC of 0.803 (95% CI 0.772-0.838; n=607; P<.001). The most important variables influencing PICTURE predictions in the COVID-19 cohort were a rapid respiratory rate, a high level of oxygen support, low oxygen saturation, and impaired mental status (Glasgow Coma Scale). CONCLUSIONS: The PICTURE model is more accurate in predicting adverse patient outcomes for both general ward patients and COVID-19 positive patients in our cohorts compared to the EDI. The ability to consistently anticipate these events may be especially valuable when considering potential incipient waves of COVID-19 infections. The generalizability of the model will require testing in other health care systems for validation.

Changes in the Association Between Diagnostic Testing Method, Polymerase Chain Reaction Ribotype, and Clinical Outcomes From Clostridioides difficile Infection: One Institution's Experience.

BACKGROUND: In Clostridioides difficile infection (CDI), the relationship between clinical, microbial, and temporal/epidemiological trends, disease severity and adverse outcomes is incompletely understood. In a follow-up to our study from 2010-2013, we evaluate stool toxin levels and C. difficile polymerase chain reaction (PCR) ribotypes. We hypothesized that elevated stool toxins and infection with ribotype 027 associate with adverse outcomes. METHODS: In 565 subjects at the University of Michigan with CDI diagnosed by positive testing for toxins A/B by enzyme immunoassay (EIA) or PCR for the tcdB gene, we quantified stool toxin levels via a modified cell cytotoxicity assay (CCA), isolated C. difficile by anaerobic culture, and performed PCR ribotyping. Severe CDI was defined by Infectious Diseases Society of America (IDSA) criteria, and primary outcomes were all-cause 30-day mortality and a composite of colectomy, intensive care unit admission, and/or death attributable to CDI within 30 days. Analyses included bivariable tests and logistic regression. RESULTS: 199 samples were diagnosed by EIA; 447 were diagnosed by PCR. Toxin positivity associated with IDSA severity but not primary outcomes. In 2016, compared with 2010-2013, ribotype 106 newly emerged, accounting for 10.6% of strains, ribotype 027 fell from 16.5% to 9.3%, and ribotype 014-027 remained stable at 18.9%. Ribotype 014-020 associated with IDSA severity and 30-day mortality (P = .001). CONCLUSIONS: Toxin positivity by EIA and CCA associated with IDSA severity but not with subsequent adverse outcomes. The molecular epidemiology of C. difficile has shifted, which may have implications for the optimal diagnostic strategy for and clinical severity of CDI.