Effectiveness of AI-Enhanced Automated Appointment Reminders

Executive Summary

Detailed Research

Methodology

Evidence includes realist and systematic reviews of reminder systems, randomized and quasi-experimental trials of targeted reminders, and ML-based no-show prediction implementations. Outcomes primarily include no-show rates, same-day cancellations, and patient/staff perceptions of reminders.

AI/ML components typically involve risk prediction models that identify appointments at high risk of nonattendance, guiding the intensity or frequency of reminders.

Key Studies

Systematic Review of Reminder Systems and Realist Synthesis

• Design: Systematic review
• Sample: Reminder system studies
• Findings: Reviews summarized that automated reminders can reduce no-shows by up to 28% vs no reminders and are generally cost-effective. They note that tailoring timing and modality to patient preferences further enhances effectiveness.
• Clinical Relevance: Establishes effectiveness of reminders

Pragmatic Randomized Study of Targeted Text Message Reminders (Kaiser Permanente, 2022)

• Design: Randomized QI study
• Sample: High-risk no-show patients
• Findings: Among patients identified as high-risk for no-shows by an Epic algorithm, two reminders (vs one) significantly reduced missed appointments and same-day cancellations for primary care and mental health visits. The underlying prediction model achieved high discrimination (AUC ≈0.9) in prior validation.
• Clinical Relevance: AI targeting improves effectiveness

Reducing Disparities in No-Show Rates Using Predictive Models (2023)

• Design: Implementation study
• Sample: Safety-net population
• Findings: This implementation used model-driven telephone outreach on top of standard automated reminders, significantly lowering in-person no-show rates in a diverse safety-net population.
• Clinical Relevance: Addresses health equity

AI and SMS-Based Adherence Interventions (2025)

• Design: Health economics evaluation
• Sample: AI+SMS framework
• Findings: An AI+SMS framework used ML models to identify patients likely to miss or delay care and send targeted reminders, demonstrating feasibility and improved adherence in a health-economics evaluation.
• Clinical Relevance: Economic feasibility demonstrated

Clinical Implications

For osteopathic practices, AI-enhanced reminder strategies can reduce no-shows for OMT and complex visits, improving schedule stability and access without excessive manual staff outreach.

Risk-based targeting helps focus personalized reminders on patients most likely to miss appointments, while still allowing DOs to tailor outreach for vulnerable or low-technology patients.

Limitations & Research Gaps

Many studies predate current LLM-era AI and focus on simpler ML models and automated workflows. Data on long-term sustainability, cost savings in small practices, and patient perceptions in osteopathic settings are limited.

Overuse of reminders can cause annoyance or message fatigue, requiring careful configuration and opt-out options.

Osteopathic Perspective

Timely attendance is essential for continuity and effective OMT; AI-enhanced reminders that reduce missed visits support the osteopathic goal of maintaining regular, hands-on care.

DOs should ensure that reminder content aligns with whole-person care, emphasizing the importance of follow-up for function, not just compliance, and that systems respect patient autonomy and preferences.