Page 427 - ISC PROCEEDINGS 21.4

(Venkatesh et al., 2003). In the context of AI-assisted healthcare, switching intention is
not solely determined by personal evaluation but is also strongly influenced by social
relationships. In Vietnam, characterized by a collectivist culture and a cautious mindset,
social influence functions as a mooring factor that reduces individuals’ readiness to shift
from traditional practices to new alternatives (Duy Hung et al., 2024).
Perceived risks in healthcare involve users' apprehensions about potential losses,
including AI system accuracy, reliability, data security, and legal liability (Dai et al., 2025).
Studies confirm it as a major barrier reducing adoption intentions for new medical
technologies (Esmaeilzadeh et al., 2021; Robinson et al., 2023). In the PPM framework,
perceived risks serve as a mooring factor that restrains switching by amplifying
uncertainty, despite push and pull influences.
Switching costs refer to the costs incurred when users change service providers
(Porter, 1998). Prior studies identify switching costs as a key mooring factor in the PPM
framework, negatively influencing switching intentions (Cheng et al., 2019; Chi et al.,
2021). In healthcare, switching costs represent a major barrier, particularly for individuals
with low confidence in new technologies (Lai & Wang, 2015). Thus, the following
hypotheses are proposed:
H2: Mooring factors have a negative effect on the switching intention to AI-assisted
healthcare services.
H2a: Social influence has a negative effect on the switching intention to AI-assisted
healthcare services.
H2b: Perceived risks have a negative effect on the switching intention to AI-assisted
healthcare.
H2c: Switching costs have a negative effect on the switching intention to AI-assisted
healthcare services.
2.3.4. Moderating effect of Mooring factors on the Push and Pull factors
In the PPM framework, mooring factors (social influence, perceived risks, and
switching costs) negatively moderate the effects of push and pull factors on switching
intention to AI-assisted healthcare (Ye & Potter, 2011; Cheng et al., 2022). Specifically,
social influence (e.g., lack of support for the new platform) weakens the impact of push
and pull on switching (Ye & Potter, 2011; Duy Hung et al., 2024). Perceived risks
concerning AI diagnostic accuracy, data privacy, and treatment efficacy attenuate these
effects (Yu et al., 2022). Switching costs likewise delay or offset push and pull influences
(Wu et al., 2017), as seen in healthcare where they hinder adoption of cloud-based
services (Lai & Wang, 2015). Accordingly, the following hypotheses are proposed:
H3: Mooring factors have negatively moderated the impact of Push on the switching
intention to AI-assisted healthcare.
H4: Mooring factors have negatively moderated the impact of Pull on the switching
intention to AI-assisted healthcare.
2.3.5. The effects of independent variables on trust in AI
In the Vietnamese healthcare context, this study proposes performance expectancy
and effort expectancy as two key determinants influencing trust in AI, as these
expectancy constructs are regarded as cognitive antecedents that shape users’ trust in
technology (Oliveira et al., 2014).
Expectancy factors (performance expectancy and effort expectancy) have a
positive impact on trust in AI (Choudhury et al., 2022). This indicates that the extent
to which residents believe that technology can deliver superior performance is directly

426