Dataset Information

Prediction of Antibiotic Susceptibility for Urinary Tract Infection in a Hospital Setting.

ABSTRACT: Empiric antibiotic prescribing can be supported by guidelines and/or local antibiograms, but these have limitations. We sought to use data from a comprehensive electronic health record to use statistical learning to develop predictive models for individual antibiotics that incorporate patient- and hospital-specific factors. This paper reports on the development and validation of these models with a large retrospective cohort. This was a retrospective cohort study including hospitalized patients with positive urine cultures in the first 48?h of hospitalization at a 1,500-bed tertiary-care hospital over a 4.5-year period. All first urine cultures with susceptibilities were included. Statistical learning techniques, including penalized logistic regression, were used to create predictive models for cefazolin, ceftriaxone, ciprofloxacin, cefepime, and piperacillin-tazobactam. These were validated on a held-out cohort. The final data set used for analysis included 6,366 patients. Final model covariates included demographics, comorbidity score, recent antibiotic use, recent antimicrobial resistance, and antibiotic allergies. Models had acceptable to good discrimination in the training data set and acceptable performance in the validation data set, with a point estimate for area under the receiver operating characteristic curve (AUC) that ranged from 0.65 for ceftriaxone to 0.69 for cefazolin. All models had excellent calibration. We used electronic health record data to create predictive models to estimate antibiotic susceptibilities for urinary tract infections in hospitalized patients. Our models had acceptable performance in a held-out validation cohort.

SUBMITTER: Hebert C

PROVIDER: S-EPMC7318043 | biostudies-literature | 2020 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of Antibiotic Susceptibility for Urinary Tract Infection in a Hospital Setting.

Hebert Courtney C Gao Yuan Y Rahman Protiva P Dewart Courtney C Lustberg Mark M Pancholi Preeti P Stevenson Kurt K Shah Nirav S NS Hade Erinn M EM

Antimicrobial agents and chemotherapy 20200623 7

Empiric antibiotic prescribing can be supported by guidelines and/or local antibiograms, but these have limitations. We sought to use data from a comprehensive electronic health record to use statistical learning to develop predictive models for individual antibiotics that incorporate patient- and hospital-specific factors. This paper reports on the development and validation of these models with a large retrospective cohort. This was a retrospective cohort study including hospitalized patients ...[more]

PMID: 32312778

Similar Datasets

Project description:BackgroundWe evaluated if flowcytometry, using Sysmex UF-5000, could improve diagnosis of urinary tract infections by rapid identification of culture negative and contaminated samples prior to culture plating, thus reducing culture plating workload and response time. We also evaluated if it is possible to reduce the response time for antibiotic susceptibility profiles using the bacteria information flag on Sysmex UF-5000 to differentiate between Gram positive and negative bacteria, followed by direct Antibiotic Susceptibility Testing (dAST) on the positive urine samples.MethodsOne thousand urine samples were analyzed for bacteria, white blood cells and squamous cells by flowcytometry before culture plating. Results from flowcytometric analysis at different cut-off values were compared to results of culture plating. We evaluated dAST on 100 urine samples that were analyzed as positive by flowcytometry, containing either Gram positive or Gram negative bacteria.ResultsUsing a cut-off value with bacterial count ≥100.000/mL and WBCs ≥10/μL, flowcytometry predicted 42,1% of samples with non-significant growth. We found that most contaminated samples contain few squamous cells. For 52/56 positive samples containing Gram negative bacteria dAST was identical to routine testing. Overall, there was concordance in 555/560 tested antibiotic combinations.ConclusionFlowcytometry offers advantages for diagnosis of urinary tract infections. Screening for negative urine samples on the day of arrival reduces culture plating and workload, and results in shorter response time for the negative samples. The bacteria information flag predicts positive samples containing Gram negative bacteria for dAST with high accuracy, thus Antibiotic Susceptibility Profile can be reported the day after arrival. For the positive samples containing Gram negative bacteria the concordance was very good between dAST and Antibiotic Susceptibility Testing in routine. For positive samples containing Gram positive bacteria the results were not convincing. We did not find any correlation between epithelial cells and contamination.

Project description:Hospital readmissions following severe infections are a major economic burden on the health care system and have a negative influence on patients' quality of life. Understanding the risk factors for readmission, particularly the extent to which they could be prevented, is of a great importance. In this study we evaluated potentially preventable risk factors for 60-day readmission in patients surviving hospitalization for complicated urinary tract infection (cUTI). This was a multinational, multicentre retrospective cohort study conducted in Europe and the Middle East. Our cohort included survivors of hospitalization due to cUTI during the years 2013-2014. The primary outcome was 60-day readmission following index hospitalization. Patient characteristics that could have influenced readmission: demographics, infection presentation and management, microbiological and clinical data; were collected via computerized medical records from infection onset up to 60 days after hospital discharge. Overall, 742 patients were included. The cohort median age was 68 years (interquartile range, (IQR) 55-80) and 43.3% (321/742) of patients were males. The all-cause 60-day readmission rate was 20.1% (149/742) and more than half were readmitted for infection [57.1%, (80/140)]. Recurrent cUTI was the most frequent cause for readmission [46.4% (65/140)]. Statistically significant risk factors associated with 60-day readmission in multivariable analysis were: older age (odds ratio (OR) 1.02 for an one-year increment, confidence interval (CI) 1.005-1.03), diabetes mellitus (OR 1.63, 95% CI 1.04-2.55), cancer (OR 1.7, 95% CI 1.05-2.77), previous urinary tract infection (UTI) in the last year (OR 1.8, 95% CI: 1.14-2.83), insertion of an indwelling bladder catheter (OR 1.62, 95% CI 1.07-2.45) and insertion of percutaneous nephrostomy (OR 3.68, 95% CI 1.67-8.13). In conclusion, patients surviving hospitalization for cUTI are frequently re-hospitalized, mostly for recurrent urinary infections associated with a medical condition that necessitated urinary interventions. Interventions to avoid re-admissions should target these patients.

Project description:BackgroundHealthcare associated infections (HAI) are a major burden for the healthcare system and associated with prolonged hospital stay, increased morbidity, mortality and costs. Healthcare associated urinary tract infections (HA-UTI) accounts for about 20-30% of all HAI's, and with the emergence of multi-resistant urinary tract pathogens, the total burden of HA-UTI will most likely increase.ObjectiveThe aim of the current study was to develop two predictive models, using data from the index admission as well as historic data on a patient, to predict the development of UTI at the time of entry to the hospital and after 48 hours of admission (HA-UTI). The ultimate goal is to predict the individual patient risk of acquiring HA-UTI before it occurs so that health care professionals may take proper actions to prevent it.MethodsRetrospective cohort analysis of approx. 300 000 adult admissions in a Danish region was performed. We developed models for UTI prediction with five machine-learning algorithms using demographic information, laboratory results, data on antibiotic treatment, past medical history (ICD10 codes), and clinical data by transformation of unstructured narrative text in Electronic Medical Records to structured data by Natural Language Processing.ResultsThe five machine-learning algorithms have been evaluated by the performance measures average squared error, cumulative lift, and area under the curve (ROC-index). The algorithms had an area under the curve (ROC-index) ranging from 0.82 to 0.84 for the entry model (T = 0 hours after admission) and from 0.71 to 0.77 for the HA-UTI model (T = 48 hours after admission).ConclusionThe study is proof of concept that it is possible to create machine-learning models that can serve as early warning systems to predict patients at risk of acquiring urinary tract infections during admission. The entry model and the HA-UTI models perform with a high ROC-index indicating a sufficient sensitivity and specificity, which may make both models instrumental in individualized prevention of UTI in hospitalized patients. The favored machine-learning methodology is Decision Trees to ensure the most transparent results and to increase clinical understanding and implementation of the models.

Dataset Information

Prediction of Antibiotic Susceptibility for Urinary Tract Infection in a Hospital Setting.

Publications

Prediction of Antibiotic Susceptibility for Urinary Tract Infection in a Hospital Setting.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets