Skip to content

Development of a computer-based model to predict delirium in intensive care unit patients

Development and internal Validation of a Dynamic Machine Learning Model for Predicting Delirium in Intensive Care Unit Patients A Prospective Cohort Study in a Tertiary Care center - NIL

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
CTRI
Registry ID
CTRI/2026/03/107147
Enrollment
600
Registered
2026-03-30
Start date
Unknown
Completion date
Unknown
Last updated
2026-04-27

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Health Condition 1: F05- Delirium due to known physiological condition

Interventions

Intervention1: Nil: Nil

Sponsors

AIIMS JODHPUR
Lead Sponsor

Eligibility

Inclusion criteria

Inclusion criteria: Adult patients aged 18 to 99 years admitted to the intensive care unit. Expected ICU stay of 24 hours or more

Exclusion criteria

Exclusion criteria: Age less than 18 years. Severe baseline dementia or cognitive impairment precluding delirium assessment. Persistent coma or brain death at admission. Refusal of consent.

Design outcomes

Primary

MeasureTime frame
Development and internal validation of a machine learning-based prediction model for delirium during an ICU stay using routinely collected clinical data with assessment of discrimination using the area under the receiver operating characteristic curve and calibration.Timepoint: At ICU admission and daily during ICU stay until discharge or death

Secondary

MeasureTime frame
Incidence & time to onset of delirium during ICU stay assessed using validated delirium assessment toolsTimepoint: Assessed at ICU admission & daily during ICU stay until discharge or death;Duration & subtype of delirium including hypoactive, hyperactive & mixed forms.Timepoint: During ICU stay from onset of delirium until resolution or discharge, every 24 hourly;Identification of important clinical & therapeutic predictors associated with development of deliriumTimepoint: Data collected at ICU admission & daily during ICU stay 24 hourly;Evaluation of predictive performance of different machine learning models using discrimination metrics including area under receiver operating characteristic curve sensitivity specificity and F1 score.Timepoint: At completion of data collection and model development.;Assessment of calibration and clinical utility of the prediction model using calibration plots and decision curve analysis.Timepoint: At completion of model development.

Countries

India

Contacts

Public ContactSreehari R Nambiar

All India Institute of Medical Sciences Jodhpur

pk_bhatia@yahoo.com9829159665

Outcome results

None listed

Source: CTRI (via WHO ICTRP) · Data processed: May 1, 2026