Data Science and Qualitative Research for Decision Support in the HIV Care Cascade

NCT06604663 | Enrolling By Invitation

• 2 other identifiers: 2022003414R01AI167694
• Study Type: interventional
• Target: 80,000
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this study is to determine whether clinical prediction algorithms derived using statistical machine learning methods can be used to improve patient outcomes in large HIV care programs in sub-Saharan Africa and elsewhere. There are two main questions to be answered. First, can the prediction algorithms accurately identify those who are at risk for (a) missing scheduled clinic visits and/or (b) treatment failure, evidenced by elevated HIV viral load? And second, can the risk predictions be used in a structured way to (a) improve retention in care and/or (b) reduce the number of patients having elevated viral load? Researchers will develop machine learning prediction algorithms, incorporate the risk prediction information into the electronic health record, provide guidance to clinical health workers on use of the point-of-care interface tools that display risk prediction information, and incorporate feedback from clinic staff to modify and co-develop the protocol for using risk predictions for improving patient outcomes. They will then compare the proportion of patients having missed visits and longer-term loss to follow up, and the proportion with elevated viral load, between clinics that use the information from the risk prediction algorithms and those that do not.

Conditions

Human Immunodeficiency Virus; Treatment Adherence; Treatment Compliance; Patient No Show; Patient Engagement; Patient Dropouts; HIV Viremia

Keywords

Electronic Medical Records; Clinical Decision Support System; Machine Learning; HIV Treatment Interruption; HIV Treatment Failure; Artificial Intelligence; Electronic Health Records; Bayesian statistics; Clinical prediction models

Outcome Measures

Primary Outcomes (4)

• The proportion of scheduled patient visits kept by the patient (1-day)

  The proportion of scheduled patient visits where the patient returns on or before the scheduled visit date. Measured weekly at the clinic level.

  The study has 6 waves (or wedges in the stepped-wedge design). The proportion will be measured weekly for the 4 weeks preceding the first wave of CDSS implementation, and then weekly until 8 weeks after the date of the final wave of CDSS implementation.

• The proportion of scheduled patient visits kept by the patient (7-day)

  The proportion of scheduled patient visits where the patient returns on or before the 7th day following the scheduled visit date. Measured weekly at the clinic level.

  The study has 6 waves (or wedges in the stepped-wedge design). The proportion will be measured weekly for the 4 weeks preceding the first wave of CDSS implementation, and then weekly until 8 weeks after the date of the final wave of CDSS implementation.

• The proportion of patients with suppressed VL among those with measured VL

  The proportion of patients with suppressed VL among those with measured VL. This outcome reflects the fraction suppressed only among those who show up for their VL measurement and is the metric designed for tracking progress toward 95-95-95 goals. This endpoint will be measured monthly at the clinic level. The denominator will be number of patient-level VL measures, and the numerator will be number of occasions where the measured VL is undetectable.

  The study has 6 waves (or wedges in the stepped-wedge design). The proportion will be measured for the month preceding the first wave of CDSS implementation, and then monthly until 2 months after the date of the final wave of CDSS implementation.

• The proportion of patients with suppressed VL among those with scheduled VL measurement, whether or not that measure was taken.

  The proportion of patients with suppressed VL among those with scheduled VL measurement, whether or not that measure was taken. This outcome is designed to reflect the fraction of the overall patient population having suppressed VL and is potentially more relevant as a population-level parameter. This endpoint will be measured monthly at the clinic level. The denominator will be number of VL measurements scheduled, and the numerator will be number of occasions where the VL is measured and is undetectable.

  The study has 6 waves (or wedges in the stepped-wedge design). The proportion will be measured for the month preceding the first wave of CDSS implementation, and then monthly until 2 months after the date of the final wave of CDSS implementation.

Study Arms (2)

Usual Care

NO INTERVENTION

Usual Care at AMPATH involves telephoning clients or care supporters the day prior to their appointment (at some clinics) and/or telephoning or making a home visits after appointments are missed. This will be in place at usual care clinics until the date at which the clinic is randomized to receive the CDSS support.

Clinical decision support, CDSS

EXPERIMENTAL

When a clinic is assigned to receive the CDSS support intervention two components will be enacted to enable proactive outreach that prevents a missed visit. These patients are considered to be in the active, experimental arm. Please seem the section above on Detailed Description for background and details on how this intervention is implemented.

Behavioral: Activation of the CDSS system

Interventions

Activation of the CDSS system BEHAVIORAL

Activation of the CDSS system, whereby outreach workers and clinicians have access to and ability to act upon lists of patients who are at highest risk of missing their upcoming clinical appointment.

Clinical decision support, CDSS

Eligibility Criteria

• Age: 18 Years - 100 Years
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• The study will include adult patients (age 18 and over) receiving HIV care through the AMPATH program in Eldoret, Kenya. There is not a patient-level enrollment process. The primary endpoints will be summarized at the clinic level (e.g., proportion of patients who keep an appointment within 7 days of the scheduled appointment).

You may not qualify if:

• None

Sponsors & Collaborators

• Brown University: lead
• Moi University College of Health Sciences: collaborator
• Moi Teaching and Referral Hospital: collaborator
• National Institute of Allergy and Infectious Diseases (NIAID): collaborator

Study Sites (1)

AMPATH

Eldoret, Kenya

Location

MeSH Terms

Conditions

Acquired Immunodeficiency Syndrome; Treatment Adherence and Compliance; Patient Compliance; No-Show Patients; Patient Participation; Patient Dropouts

Condition Hierarchy (Ancestors)

HIV Infections; Blood-Borne Infections; Communicable Diseases; Infections; Sexually Transmitted Diseases, Viral; Sexually Transmitted Diseases; Lentivirus Infections; Retroviridae Infections; RNA Virus Infections; Virus Diseases; Slow Virus Diseases; Genital Diseases; Urogenital Diseases; Immunologic Deficiency Syndromes; Immune System Diseases; Health Behavior; Behavior; Patient Acceptance of Health Care

Study Officials

• Joseph W Hogan, ScD

  Brown University

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: NONE
• Purpose: HEALTH SERVICES RESEARCH
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: The investigators will evaluate the intervention using a stepped-wedge design whereby clinics are assigned the intervention in sequential order, and in batches. Specifically, the investigators will implement the decision support tool at 5 or 6 clinics per month over a period of 6 months. By the end of 6 months, all clinics will be receiving the decision support tool. There may be some variation in the allocation, owing to readiness of individual clinics to begin using the decision support tools. To ensure balance in allocation, randomization will be stratified by geographic location, clinic size and pre-intervention appointment default rate.

Study Record Dates

• First Submitted: September 9, 2024
• First Posted: September 19, 2024
• Study Start: May 20, 2024
• Primary Completion: January 31, 2026
• Study Completion (Estimated): October 31, 2026
• Last Updated: January 12, 2026

Record last verified: 2025-09

Data Sharing

• IPD Sharing: Will not share

Locations

KE (1)