Unlocking AI Chatbot Potential in Healthcare: Trust-Enhanced DeLone & McLean IS Success Model

Background: Healthcare chatbots have emerged as a promising application of artificial intelligence in healthcare, offering potential benefits in accessibility, efficiency, and patient engagement. However, despite their growing adoption, limited research has examined the factors that determine their success from the user's perspective. Objective: This study aimed to evaluate the success of a health chatbot service by applying the updated DeLone and McLean Information Systems Success Model augmented with a trust construct, examining the effects of information quality, system quality, service quality, and trust on intention to use, user satisfaction, and net benefits. Methods: An online survey design was employed, utilizing a structured questionnaire with 28 items measuring seven constructs on a seven-point Likert scale. Data were collected electronically from residents of Saudi Arabia between July and September 2024 using convenience sampling. Eligible participants were adults aged 18 years or older who had previously used the health chatbot service. A total of 321 valid responses were obtained. Partial Least Squares Structural Equation (PLS-SEM) was conducted using SmartPLS 3.3 software for measurement and structural model analysis. Results: The measurement model demonstrated acceptable reliability and validity, with composite reliability values exceeding 0.90 and average variance extracted values above 0.70 for all constructs. Structural model analysis supported eight of ten hypotheses. Trust exhibited the strongest effect on intention to use (β = 0.359, p < 0.001), followed by system quality (β = 0.234, p < 0.001) and information quality (β = 0.147, p < 0.01). Intention to use significantly predicted user satisfaction (β = 0.620, p < 0.001) and net benefits (β = 0.278, p < 0.001). User satisfaction demonstrated a strong positive effect on net benefits (β = 0.610, p < 0.001). The model explained 67.6% of the variance in intention to use, 72.7% in user satisfaction, and 71.4% in net benefits. Conclusions: Trust emerged as the most influential factor affecting intention to use the healthcare chatbot service, underscoring its critical role in user acceptance of health chatbot services. Information quality, system quality, and service quality exerted small to moderate effects on behavioral outcomes. These findings suggest that healthcare organizations deploying chatbot services should prioritize building user trust alongside ensuring high system and information quality to maximize user satisfaction and realized net benefits.