A Trial of AI-Powered Text Message Outreach on Well-Child Visit Completion

NCT06698640 | Completed Results

• 1 other identifier: HEDIS-OPT-2024-001
• Study Type: interventional
• Target: 2,821
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this clinical trial is to evaluate whether conversational AI-powered text message outreach with appointment scheduling assistance improves well-child visit completion rates in Medicaid-enrolled children aged 0-21 years compared with automated text messages alone or traditional passive outreach. The main questions it aims to answer are: Does automated SMS outreach improve well-child visit completion rates compared to traditional passive outreach? Does conversational AI-powered scheduling assistance lead to higher completion rates than automated SMS alone? Researchers will compare three groups: Control Group: Participants receive traditional passive outreach (mailed reminders). Automated SMS Group: Participants receive standardized automated text message reminders. Automated SMS + Conversational AI Scheduling Assistance Group: Participants receive automated text messages plus AI-powered appointment scheduling assistance that contacts primary care providers directly to book appointments on behalf of families. Participants will: Be randomized into one of three study groups at the household level. Receive outreach from June 9-July 14, 2025. Have well-child visit completion ascertained through administrative claims through December 31, 2025. This study tests whether concrete scheduling support - rather than reminders alone - drives preventive care utilization in pediatric Medicaid populations.

Conditions

Well Child Visit; Gaps in Care

Keywords

gaps in care; preventive care; text messaging

Outcome Measures

Primary Outcomes (1)

• Number of Participants With Well-Child Visit Completion

  Binary indicator of whether each participant completed at least one well-child visit by December 31, 2025, ascertained through administrative claims using HEDIS technical specifications

  Up to 7 months post-randomization (June 1 - December 31, 2025)

Secondary Outcomes (2)

• Staff Time Per Successfully Scheduled Appointment

  Up to 7 months post-randomization (June 1 - December 31, 2025)

• Human Escalation Rate for Automated Scheduling Attempts

  Up to 2 weeks from initiation of first automated scheduling attempt

Study Arms (3)

Control

OTHER

Participants received standard health plan outreach consisting of periodic mailed reminders. Families retained access to all standard appointment scheduling methods including telephone calls to provider offices.

Other: Traditional Passive Outreach

Automated SMS

ACTIVE COMPARATOR

Participants received standardized automated text message reminders at predetermined intervals (initial contact, 2-week follow-up, 4-week follow-up). Messages provided general information about visit importance and instructions to call providers to schedule. Messages did not include interactive scheduling or conversational capabilities.

Behavioral: Automated SMS

Automated SMS + Conversational AI Scheduling Assistance

EXPERIMENTAL

Participants received automated text message reminders with response options. When families expressed interest, structured follow-up texts collected appointment preferences (preferred dates, times, locations, number of children needing visits). An AI-powered telephone scheduling system (GPT-4o, OpenAI) then contacted the child's primary care provider directly to book appointments on behalf of families. The system disclosed its automated nature at call initiation and escalated to human staff when clinic policies were incompatible with automated scheduling or technical failures occurred.

Behavioral: Automated SMS + Conversational AI Scheduling Assistance

Interventions

Automated SMS BEHAVIORAL

Participants received standardized automated text message reminders at predetermined intervals (initial contact, 2-week follow-up, 4-week follow-up). Messages provided general information about visit importance and instructions to call providers to schedule. Messages did not include interactive scheduling or conversational capabilities.

Automated SMS

Automated SMS + Conversational AI Scheduling Assistance BEHAVIORAL

Participants received automated text message reminders with response options. When families expressed interest, structured follow-up texts collected appointment preferences. An AI-powered telephone scheduling system (GPT-4o, OpenAI) then contacted the child's primary care provider directly to book appointments on behalf of families. The system disclosed its automated nature at call initiation and escalated to human staff when clinic policies were incompatible with automated scheduling or technical failures occurred.

Automated SMS + Conversational AI Scheduling Assistance

Traditional Passive Outreach OTHER

Participants received standard health plan outreach consisting of periodic mailed reminders. Families retained access to all standard appointment scheduling methods including telephone calls to provider offices.

Control

Eligibility Criteria

• Age: 0 Years - 21 Years
• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64)

You may qualify if:

• Medicaid beneficiary aged 0-21 years
• Had not completed a well-child visit in 2025
• Enrolled through the managed care organization with authorization for health plan outreach
• Valid mobile phone number capable of receiving SMS

You may not qualify if:

• Do-not-contact designation
• Inactive phone service
• Documented request for no contact from healthcare provider or health plan
• Prior SMS well-child visit outreach earlier in the 2025 measurement year

Sponsors & Collaborators

• Waymark: lead

Study Sites (1)

Waymark

San Francisco, California, 94115, United States

Location

Limitations and Caveats

Single managed care organization setting in Virginia limits generalizability of effect estimates to other states, provider networks, or HEDIS bonus structures. English-only delivery may have excluded linguistically diverse families. Participants and clinical staff were not blinded. Six-month follow-up may not capture longer-term effects. Visit completion was a binary HEDIS indicator and does not reflect visit quality.

Results Point of Contact

• Title: Sanjay Basu, MD PhD
• Organization: Waymark

sanjay.basu@waymarkcare.com

415-577-5796

Publication Agreements

• PI is Sponsor Employee: Yes

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: DOUBLE
• Who Masked: INVESTIGATOR, OUTCOMES ASSESSOR
• Purpose: PREVENTION
• Intervention Model: PARALLEL
• Sponsor Type: INDUSTRY
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: November 18, 2024
• First Posted: November 21, 2024
• Study Start: June 1, 2025
• Primary Completion: December 31, 2025
• Study Completion: March 17, 2026
• Last Updated: June 22, 2026
• Results First Posted: June 22, 2026

Record last verified: 2026-05

Data Sharing

• IPD Sharing: Will not share

Locations

US (1)