An observational study in Bacteremia, sponsored by Nantes University Hospital. Completed. Open to participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-10-26.
Sponsored by Nantes University Hospital · Observational
Bacteremia, a common infection associated with high morbidity and mortality, requires prompt and appropriate antibiotic treatment. Numerous studies have shown that the appropriate use of antibiotics helps improve the prognosis of patients, reduce the selection of pathogenic organisms, the emergence of antibiotic resistance, adverse treatment effects, and healthcare costs.
The Nantes University Hospital Center has operated a Biomedical Data Warehouse (BDW) since 2018. The growing availability of data through BDWs thus enables the development of decision-support tools based on machine learning.
The primary objective of the OPTIBact project is to predict the optimal empirical antibiotic for hospitalized patients with bacteremia using a machine learning-based predictive model using data from the Nantes University Hospital's BDW.
As secondary objectives, the study aims:
In the OPTIBACT project, we used data extracted from the Nantes University Hospital BDW, to predict the optimal antibiotic therapy for hospitalized patients with bacteremia using a machine learning-based predictive model.
297 studies on the registry are indexed under Bacteremia; 51 are open to participants now.
This study's enrollment of 8,765 is above the median of 300 across 125 observational studies indexed under Bacteremia.
Browse Bacteremia studies →Nantes University Hospital is the lead sponsor of 839 studies on the registry; 195 are open to participants now.
Counted across the registry records on this site, refreshed daily.
primary care clinic at Nantes hospital
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Adult patients (aged 18 or older at the time of blood collection) who had a positive blood culture at Nantes University Hospital between January 1, 2017, and December 31, 2023, except:
Exclusion Criteria:
Excluded are positive blood cultures
No Intervention: Observational Cohort
Accuracy of the predictive model (molecule)
Accuracy of the predictive model for the empirical antibiotic and its 95% confidence interval (95% CI), estimated using the validation set. Accuracy is defined as the proportion of bacteremia episodes for which the antibiotic predicted by the model is optimal, relative to the total number of episodes in the validation set.
Time frame: From descriptive analysis to the development of the predictive model: 6 months
Comparison of the performances estimated for the predictive model for optimal empirical antibiotic (molecule) with real-world estimates
Comparison of the accuracy, positive and negative predictive values, sensitivity, and specificity estimated for the predictive model for the antibiotic with real-world estimates on the validation sample (overlap of 95% CI).
Time frame: From the development of the predictive model: an average of one year
Expected advantages of the predictive model compared to real-world clinical outcome
Expected advantages of the predictive model compared to real-world clinical outcome: the time , in days, between the date of the positive blood culture and the initiation of optimal empirical antibiotic therapy.
Time frame: From the development of the predictive model: an average of one year
Accuracy of the predictive model (dosage)
Accuracy of the predictive model for empirical antibiotic dosage and its 95% CI, estimated using the validation sample.
Time frame: From the development of the predictive model: an average of one year
Comparison of the performances of the predictive model for optimal empirical antibiotic dosing with standard patient care
Comparison of the accuracy, positive and negative predictive values, sensitivity, and specificity estimated for the dosage prediction model with real-world estimates on the validation sample (overlap of 95% CI), across the entire sample and in subgroup analyses (major pathogen types).
Time frame: From the development of the predictive model: an average of one year
Description of clinical outcomes of two groups of patients who received optimal antibiotic therapy and those who did not.
The following clinical outcomes will be described for two groups of patients, who received optimal antibiotic therapy and those who did not: * Time from the date of the positive blood culture to the initiation of optimal empirical antibiotic therapy, in days, * admission to the intensive care unit within 72 hours, * length of stay, in days, * readmissions, * in-hospital mortality and mortality at day 30, * resistance rates of the various pathogens identified in blood cultures, * sources of infection, * signs of severity, * biological pre-treatment parameters, * Treatments received.
Time frame: From the development of the predictive model: an average of one year
To identify patients and episode characteristics independently associated with resistance among patients with a positive blood culture, stratified by pathogen type.
To identify patients and episode characteristics independently associated with resistance among patients with a positive blood culture, stratified by pathogen type. The factors studied will be prior antibiotic exposure, previous colonisation or infection with resistant bacteria, recent hospitalisation, long term care residence, hospital - VS community acquired infection, unit, age, sex.
Time frame: From the development of the predictive model: an average of one year
No study locations are listed for this record.
Plan to share: No — No there is not a plan to make IPD available
No publications or documents are linked to this record.
Ran from Jan 2017 to primary completion in Dec 2023, 83 months.
How these figures are worked out, and which studies are left out of them.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
This study is completed, as verified in Oct 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.