Impact of an AI-Based Chatbot on Implant Patient Management: RCT

Rationale and Background:

Dental implant therapy is highly successful, but peri-implant diseases, primarily caused by inadequate plaque control, remain a major cause of implant failure. Conventional oral hygiene education methods, such as verbal instructions and printed materials, often fail to ensure long-term patient compliance. Artificial intelligence (AI)-based chatbots, powered by large language models, offer a new way to provide continuous, personalized education and motivation to patients. Although their potential has been explored in orthodontics, little is known about their role in peri-implant maintenance.

Study Objective:

This randomized controlled clinical trial investigates whether an AI-based chatbot developed by dental experts can improve oral hygiene compliance and peri-implant health compared to traditional educational methods.

Study Design:

A total of 110 participants were enrolled between June 2024 and June 2025 in private dental practices in Catania, Italy. The study was approved by the local ethics committee (Protocol ID: 60/2024/CL-PAR) and conducted in accordance with the Declaration of Helsinki (2016 revision). Eligible participants were adults (≥18 years), systemically healthy or with controlled conditions, attending regular supportive periodontal therapy, and having at least one osseointegrated implant. Exclusion criteria included active periodontal disease, heavy smoking, systemic immunosuppression, and pregnancy.

Interventions:

Participants were randomly assigned to two groups:

Chatbot group: Received oral hygiene education and motivational support through an AI-based chatbot specifically developed for implant maintenance.

Control group: Received standard oral hygiene instructions through printed materials and sessions with a dental hygienist.

After the study, the chatbot was made available to the control group.

Outcome Measures:

The primary outcome was the Plaque Index (PI) at implant sites. Secondary outcomes included bleeding on probing (BOP), suppuration (SUP), probing pocket depth (PPD), and radiographic marginal bone loss (MBL).

Expected Impact:

This study aims to validate the use of AI-based educational tools as a complementary strategy for improving patient compliance, plaque control, and peri-implant health. The findings could support the safe and effective integration of AI chatbots into clinical dental practice and patient education programs.

Inclusion criteria

• Systemically healthy or with controlled systemic conditions;
• Good periodontal health on either an intact or reduced periodontium;
• Regular adherence to supportive periodontal therapy (SPT);
• Presence of at least one osseointegrated dental implant in the maxilla or mandible, following transmucosal healing for 2-4 months;
• Daily access to a smartphone.

Exclusion criteria

• Age < 18 years;
• Untreated or active periodontal disease;
• Currently undergoing orthodontic therapy;
• Patients with cognitive deficits;
• Smoking habit (more than 10 cigarettes per day);
• History of heavy alcohol consumption;
• Use of immunosuppressive medication;
• Pregnancy or lactation.