Using EMR Data and AI to Develop a Passive Digital Marker to Predict Postoperative Delirium

Indiana University

Status: Enrolling

Conditions: Delirium in Old Age

Treatments: Other: PDM Alert

Study type: Interventional

Funder types: Other; NIH

Identifiers

NCT06216483

15767

K23AG071945 (U.S. NIH Grant/Contract)

Details and patient eligibility

About

This study will be a pilot randomized controlled trial that will determine the effect of an indicator of delirium risk, which will be delivered either via a paper form or via the electronic medical record, on (1)postoperative delirium incidence; and (2)the development of long term cognitive decline and dementia.

Full description

Patients will be recruited from a large general, minimally invasive, colorectal, and oncologic, university-based surgical practice. English speaking individuals over the age of 65 with planned major abdominal surgery (encompassing abdominal wall, gastrointestinal and hepatobiliary procedures for benign and malignant diseases) will be screened for eligibility following an initial visit with their surgeon. Inclusion criteria will include a required inpatient stay for ≥ 24 hours following surgery and a scheduled preadmission testing and medical risk stratification visit with an internal medicine provider. Patients will be excluded if they have severe mental illness, hearing and visual impairment, alcohol intoxication or related delirium, central nervous disorder (e.g., stroke, traumatic brain injury), and pregnant or nursing status. Once enrolled, subjects will be randomized using a computer-generated scheme (permuted blocks of four and six) to the PDM or usual care, stratified according to type of surgery (hepatobiliary, gastrointestinal, abdominal wall).

Enrollment: 60 estimated patients

Sex: All

Ages: 65+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

65 years old Planned inpatient major surgery requiring inpatient stay ≥ 1 day

Exclusion criteria

• Baseline cognitive impairment or a diagnosis of Alzheimer's Disease/Alzheimer's Disease and Related Dementias (AD/ADRD)
• Alcohol or drug withdrawal
• Prisoner status
• Unable to communicate with research staff due to sensory impairments
• Not fluent in English
• Currently has a personal safety attendant

Trial design

Primary purpose: Diagnostic

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: None (Open label)

---

60 participants in 2 patient groups, including a placebo group

No PDM Alert

Placebo Comparator group

Description:

No PDM alert will be placed on subject's record

Treatment:

Other: PDM Alert

PDM Alert

Experimental group

Description:

PDM alert will be placed on subject's record

Treatment:

Other: PDM Alert