Back to search

McMaster University

Designing Human- and AI-Assisted Health Coaching for Preventive Behaviour Change

Abstract

dc:description.abstract

Background: Effective support for health behaviour change remains a critical challenge in chronic disease prevention, particularly among young adults navigating postsecondary education and life transitions. Health coaching approaches, such as motivational interviewing (MI), show promise for promoting healthier behaviours in this population. More recently, artificial intelligence (AI) has created new opportunities and challenges for this support. Objectives: This thesis examined how health coaching interventions can be designed, delivered, and evaluated to support preventive health behaviour change through six studies spanning MI and AI-assisted health coaching. Methods: Six studies were conducted: a systematic review and meta-analysis, a pilot randomized controlled trial (RCT), a secondary analysis of exit survey data, a qualitatively driven convergent mixed-methods secondary analysis of MI session records, a scoping review, and an experimental study of AI-generated health coaching responses. Results: A systematic review and meta-analysis of 33 RCTs in community-based adult populations demonstrated that MI produced small but statistically significant reductions in weight (−1.11 kg), BMI (−0.42 kg/m²), waist circumference (−1.13 cm), and body fat (−1.55%). The MOTIVATE pilot RCT evaluated a six-month MI intervention among 101 university students and met all feasibility thresholds for recruitment, retention, data completion, and participant satisfaction. The exit survey analysis identified high acceptability and perceived improvements in health behaviours, while the mixed-methods session-record analysis identified evolving priorities, contextual barriers, and adaptive strategies. A scoping review of AI-assisted health coaching prompts identified substantial variability in prompt design and evaluation, and an experimental study demonstrated that structured prompting improved the quality and fidelity of AI-generated responses (83% vs. 6% acceptable). Conclusions: These studies advance understanding of health coaching for preventive behaviour change across human- and AI-assisted contexts. Findings suggest that behaviour change is complex and individualized, and that effective health coaching extends beyond information provision, with the quality of coaching playing a central role.

Degree

thesis:*
Department dc:contributor.department
Health Research Methodology
Year dc:date.issued
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Incze, Taylor
Advisor dc:contributor.advisor
  • Anderson, Laura N.

Subjects

dc:subject × 7

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
McMaster University
Base URL
macsphere.mcmaster.ca/server/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Incze, Taylor. Designing Human- and AI-Assisted Health Coaching for Preventive Behaviour Change. 2026. https://hdl.handle.net/11375/33872