Effectiveness-Implementation Evaluation of Acute Kidney Injury Decision Support (SMART-AKI)

Transforming Kidney Care in the Emergency Department: A Prospective Effectiveness-Implementation Evaluation

Study Purpose: This study is testing an artificial intelligence (AI)-powered clinical decision support (CDS) system designed to help emergency department (ED) doctors detect and manage acute kidney injury (AKI) earlier. The goal is to see whether the tool improves patient care, clinician decision-making, and hospital outcomes when used in real-world ED settings.

Study Design:

The AI tool will be gradually introduced at three hospital EDs:

Johns Hopkins Hospital (JHH) Bayview Medical Center (BMC) Howard County General Hospital (HCGH)

Before the tool is activated, it will run in the background to collect baseline data without influencing care. Once implemented, doctors will receive training, and researchers will track how often the tool is used and whether it improves AKI care.

What the Study Measures:

Process Outcomes: Does the tool help doctors identify AKI sooner, avoid harmful medications, and improve decision-making about hospitalization?

Clinical Outcomes: Does the tool reduce the number and severity of AKI cases and improve kidney-related health outcomes?

Implementation Outcomes: Do ED doctors find the tool useful? Does it fit into the ED doctor's workflow without slowing the ED doctor's down?

Expected Impact: If successful, the AI tool could be expanded to other hospitals and used to improve early detection and treatment of AKI, reducing kidney complications and improving patient care nationwide.

Study Overview

• Status: Recruiting
• Conditions: Acute Kidney Injury

• Study Type: Observational
• Enrollment (Estimated): 200000

Contacts and Locations

• Study Contact: Name: Eili Y Klein, PhD; Phone Number: 732-266-7732; Email: eklein@jhmi.edu
• Study Contact Backup: Name: Michael Ehmann, MD; Email: mehmann1@jhmi.edu

Study Locations

• United States
  • Maryland
    • Baltimore, Maryland, United States, 21287
      • Recruiting
      • Johns Hopkins Hospital
      • Contact: Eili Klein, PhD; Phone Number: 732-266-7732; Email: eklein@jhmi.edu

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Emergency Department

Description

Inclusion Criteria:

Participants must meet all of the following conditions to be eligible for the study:

1. Emergency Department Visit

   Patients presenting to one of the three study site emergency departments (EDs):

   Johns Hopkins Hospital (JHH) Bayview Medical Center (BMC) Howard County General Hospital (HCGH) Serum Creatinine Measurement

2. At least one serum creatinine (sCr) test performed during the ED visit.
3. Age: Adults (≥18 years old) at the time of ED visit.
4. Follow-Up Data Available: To be included in analysis for the investigator's secondary effectiveness outcome (new or progressive AKI) patients must have repeat creatinine concentration measurement available within 72 hours

Exclusion Criteria:

1. No ED Serum Creatinine Data: Patients who do not have a measured serum creatinine (sCr) value during the ED visit will not have decision support provided during the ED stay.
2. Patients Discharged Without Follow-Up Data: Patients discharged from the ED who do not undergo a follow-up serum creatinine test within 72 hours will be excluded from analysis of new or progressive AKI outcome assessment.
3. Age <18 Years
4. End-Stage Kidney Disease (ESKD) or Chronic Dialysis: Patients with a documented history of end-stage kidney disease (ESKD), (patients receiving chronic dialysis (hemodialysis or peritoneal dialysis) will will be excluded from analysis of new or progressive AKI outcome assessment).

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
All adult ED patients with creatinine concentration measured during the participant's ED stay

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Time Frame
Number of Patients who receive Guideline-Concordant Kidney Care	From time of decision support provision until departure from the ED, approximately 6 hours

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
New or Progressive Acute Kidney Injury		Within 72 hours of first creatinine measurement in the ED
Perceived usefulness and usability of AKIDS software as assessed by the System Usability Score	Will be assessed using the usability using the System Usability Score, a 10-item Likert scale that yields a single composite score on a scale of 0-100, with 100 being most usable.	Prior to AKIDS implementation, immediately after implementation, and post implementation 6 months

Collaborators and Investigators

• Sponsor: Johns Hopkins University
• Collaborators: Agency For Healthcare Research & Quality
• Investigators: Principal Investigator: Eili Klein, PhD, Johns Hopkins University

Study record dates

Study Major Dates

• Study Start (Actual): March 10, 2026
• Primary Completion (Estimated): December 1, 2028
• Study Completion (Estimated): December 1, 2030

Study Registration Dates

• First Submitted: February 17, 2025
• First Submitted That Met QC Criteria: February 17, 2025
• First Posted (Actual): February 21, 2025

Study Record Updates

• Last Update Posted (Actual): April 20, 2026
• Last Update Submitted That Met QC Criteria: April 15, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Keywords: Clinical Decision Support; Emergency Department; Acute Kidney Injury; Prediction; Machine Learning; AKI
• Additional Relevant MeSH Terms: Urogenital Diseases; Pathologic Processes; Male Urogenital Diseases; Kidney Diseases; Urologic Diseases; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Disease Attributes; Renal Insufficiency; Pathological Conditions, Signs and Symptoms; Emergencies; Acute Kidney Injury
• Other Study ID Numbers: IRB00367884; R01HS027793 (U.S. AHRQ Grant/Contract)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No