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

“The Fornell–Larcker criterion stipulates that the square root of AVE for each latent variable should be greater than the correlations with other variables [135].”

“To further justify sample adequacy, a post hoc power analysis was conducted using G*Power 3 [134] for an F-test (R 2 increase) with 10 predictors.”

“However, in most management, health informatics, and IS studies using PLS-SEM, a sample size between 200 and 300 is generally considered robust [129].”

“A study shows that trust exists when one party has confidence in the reliability and integrity of an exchange partner [119].”

“The primary objective of this study is to evaluate the success of the Healthcare Chatbot Service from the user’s perspective by using the D&M ISS Model as a guiding framework for this research [52].”

“This approach was chosen because the target population was not centrally registered and could only be reached via online platforms [123,124].”

“In this study, all three endogenous variables exhibited VIF values greater than 4 and tolerance values less than 0.25, indicating the potential presence of multicollinearity [140,141].”

“Effect size measures the impact of each predictor on the dependent construct [138].”

“…vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of depression and anxiety [7,31,32].”

“…Modeling (PLS-SEM) is commonly determined using the “10-times rule,” which advises that the minimum sample should be ten times the largest number of indicators used to measure any single construct, or ten times the largest number of structural paths directed at any endogenous construct [127,128].”

“Research tends to affirm the significant impact of system quality on user satisfaction across various types of IS, including healthcare applications [52,63,76,79,82,92–95].”

“Consequently, many studies over the past decade have conceptually and empirically explored factors influencing users’ trust [121].”

“This modification underscores the necessity of incorporating service quality when evaluating IS, as previously advocated by several scholars [72].”

“Information quality is the degree to which an information system produces outputs that are appropriate for user decision-making [74,75].”

“…on users’ behavioural intentions and acceptance studies [35], we nonetheless produce little, if at all, clear understanding about how core information system attributes contribute to user satisfaction, usage, and overall system success when engaging with AI-powered healthcare services [14,48–51].”

“It is defined as “an estimate of the probability that a person will use the application” [105].”

“This non-probability design supports feasibility in an otherwise hard-to-reach population, but it also constrains external validity and limits generalization beyond the study sample [125,126].”

“However, in the context of healthcare services, omitting trust would be a critical limitation, as it has long been fundamental to healthcare delivery, underpinning users and provider relationships and shaping reliance on healthcare outcomes [53,54].”

“Path coefficients: bootstrapping (5000 bootstrap samples; 321 bootstrap cases; no sign changes) [136,137].”

“The adaptability and robustness of the D&M ISS Model have also been demonstrated in several health IT success projects in both developing and developed nations [63,66–69].”

“In this vein, as chatbots become increasingly embedded in healthcare workflows, establishing trust from both patients and providers is essential for adoption and meaningful use [55,56].”

“…services over other AI technological alternatives, particularly in reshaping healthcare service delivery models by enhancing accessibility, personalization, and efficiency [5,8,9], while simultaneously recasting patient education, interaction, and engagement with health information [10–12].”

“The domain has been recognized in the extant literature as a significant factor influencing users’ behavioral intentions [99,100] and actual use of digital systems [101].”

“While prior studies have linked increased user satisfaction to higher levels of information quality [82,85,87,111], this study diverges from that established relationship.”

“However, the measurement of net benefits should be contextualized according to the specific system and level of impact under examination [113], recognizing that user perception and satisfaction have emerged as pivotal proxies for evaluating the actual value realized from system use [110,114].”

“…vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of depression and anxiety [7,31,32].”

“Net benefits refer to the extent to which the system contributes to the success of target subjects [52,110].”

“…on users’ behavioural intentions and acceptance studies [35], we nonetheless produce little, if at all, clear understanding about how core information system attributes contribute to user satisfaction, usage, and overall system success when engaging with AI-powered healthcare services [14,48–51].”

“This model has been extensively tested in the context of e-commerce systems [62], e-government systems [63,64], and technology adoption in tourism [65].”

“User satisfaction is a crucial metric for measuring the impact of IS benefits, as well as a key aspect of implementing new technologies [117].”

“Prior studies articulated that trust functions as a central mechanism that introduces perceptions of safety and credibility beyond what quality dimensions alone can ensure [57–59].”

“…services over other AI technological alternatives, particularly in reshaping healthcare service delivery models by enhancing accessibility, personalization, and efficiency [5,8,9], while simultaneously recasting patient education, interaction, and engagement with health information [10–12].”

“While prior studies have linked increased user satisfaction to higher levels of information quality [82,85,87,111], this study diverges from that established relationship.”

“…this vein, evaluating information quality involves assessing factors such as completeness, accuracy, uniqueness, relevance, timeliness, precision, comprehensibility, conciseness, and informativeness that pertain to relationships with users’ acceptance and intention to use IS-based services [76,77].”

“…vaccine literacy and mitigate vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms…”

“Users with stronger intentions to use the system are more likely to develop deeper, ongoing interactions with the service, which results in increased satisfaction [94,106–109].”

“In this vein, Chatbots, also known as conversational agents or virtual agents, represent a type of AI application designed to simulate human-like conversations with users in natural language, providing information or services in specific domains [2–4].”

“Evidence from the literature asserts that when IS-based health services deliver accurate, complete, timely, and relevant information that meets users’ informational needs, users are more likely to positively engage with the services [83–87].”

“Research has affirmed the critical importance of trust in driving intention to use healthcare and digital health technologies [80,106,122].”

“The adaptability and robustness of the D&M ISS Model have also been demonstrated in several health IT success projects in both developing and developed nations [63,66–69].”

“Indices and factor loading thresholds were set according to established standards [130–133].”

“Consequently, the updated D&M ISS model posits that three core components—service quality, information quality, and system quality—collectively influence system utilization and user satisfaction, thereby explaining the success of the IS platform [73].”

“…vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of depression and anxiety [7,31,32].”

“Users with stronger intentions to use the system are more likely to develop deeper, ongoing interactions with the service, which results in increased satisfaction [94,106–109].”

“The D&M ISS Model provides a practical framework for evaluating the effectiveness of health IS [70].”

“While prior studies have linked increased user satisfaction to higher levels of information quality [82,85,87,111], this study diverges from that established relationship.”

“It has been shown that system quality plays a central role in shaping how users perceive and interact with a website, as their continued use often depends on how reliable, responsive, and user-friendly they find the system to be [88].”

“The domain has been recognized in the extant literature as a significant factor influencing users’ behavioral intentions [99,100] and actual use of digital systems [101].”

“In the extant literature, chatbot applications in the health services domain are neither novel nor marginal; rather, they have gained ground in the existing body of literature, reflecting a sustained and evolving scholarly interest in their potential and applications [13–15].”

“Users may acknowledge a chatbot’s technical quality yet still require trust before engaging in repeated use or acting upon its health-related recommendations [60].”

“However, prior D&M ISS-based studies show that user satisfaction and net benefits are closely related, with user satisfaction often functioning as an important antecedent of net benefits [142–144].”

“Prior studies articulated that trust functions as a central mechanism that introduces perceptions of safety and credibility beyond what quality dimensions alone can ensure [57–59].”

“Because prior research suggests that gender can shape technology-related perceptions and adoption behavior, the observed relationships may not be identical in a more gender-balanced sample [146,147].”

“…vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of depression and anxiety [7,31,32].”

“…has affirmed the relationship between service quality and user satisfaction on the grounds of users’ general judgment of an entity’s excellence or superiority of the services that arises from comparing their expectations before experiencing the service with their actual experiences [102–104].”

“…informed by Technology Acceptance Model (TAM) and Unified Theory of Acceptance and Use of Technology (UTAUT) perspectives, and focuses primarily on users’ behavioral intentions [36–44], leaving limited explanation on how healthcare chatbots function as effective service systems for use in practice.”

“…in promoting public health messages to enhance vaccine literacy and mitigate vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30],…”

“Information quality, as described earlier, also affects user satisfaction [80–82].”

“Research tends to affirm the significant impact of system quality on user satisfaction across various types of IS, including healthcare applications [52,63,76,79,82,92–95].”

“Others have gone so far as to demonstrate benefits and utilization of AI-powered chatbots in various healthcare environments [5,35,45–47], but how effectively it is utilized to provide appropriate service support to the users remains an unresolved concern in the extant literature.”

“The adaptability and robustness of the D&M ISS Model have also been demonstrated in several health IT success projects in both developing and developed nations [63,66–69].”

“…vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of depression and anxiety [7,31,32].”

“Recognized as a type of IS, the healthcare chatbot service can be assessed using the D&M ISS Model [61].”

“Because prior research suggests that gender can shape technology-related perceptions and adoption behavior, the observed relationships may not be identical in a more gender-balanced sample [146,147].”

“Others have gone so far as to demonstrate benefits and utilization of AI-powered chatbots in various healthcare environments [5,35,45–47], but how effectively it is utilized to provide appropriate service support to the users remains an unresolved concern in the extant literature.”

“…appear to affirm the growing role of chatbots in promoting public health messages to enhance vaccine literacy and mitigate vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring…”

“…appear to affirm the growing role of chatbots in promoting public health messages to enhance vaccine literacy and mitigate vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring…”

“…been to convey the message that chatbots have the toolkit to assist users with various healthcare services yet remains less conclusive regarding the extent to which users find them useful for delivering context-sensitive healthcare service advice, which needs to be thoroughly evaluated [14,33–35].”

“…vaccine literacy and mitigate vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of…”

“In this vein, as chatbots become increasingly embedded in healthcare workflows, establishing trust from both patients and providers is essential for adoption and meaningful use [55,56].”

“Based on the D&M ISS Model, the extant literature has established that users are more likely to adopt and continue using a system if they perceive the information provided to be of high quality [5,52,78,79].”

“Research tends to affirm the significant impact of system quality on user satisfaction across various types of IS, including healthcare applications [52,63,76,79,82,92–95].”

“…services over other AI technological alternatives, particularly in reshaping healthcare service delivery models by enhancing accessibility, personalization, and efficiency [5,8,9], while simultaneously recasting patient education, interaction, and engagement with health information [10–12].”

“…in promoting public health messages to enhance vaccine literacy and mitigate vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30],…”

“Recent empirical evidence from digital health implementations consistently validates the positive relationship between system quality and user acceptance [90,91].”

“However, prior D&M ISS-based studies show that user satisfaction and net benefits are closely related, with user satisfaction often functioning as an important antecedent of net benefits [142–144].”

“…with prior empirical evidence affirming that satisfaction functions as a direct predictor and motivator for sustained system engagement [52,94,116,118], this study posits that users who experience high satisfaction with healthcare chatbot services through positive interactions, perceived…”

“…vaccine hesitancy [16–19] improving diabetic patient education [20,21], providing emotional support for patient [22,23], assisting in remote mental health monitoring [24–26], aiding tobacco cessation [27–30], enhancing drug compliance, and alleviating symptoms of depression and anxiety [7,31,32].”

“…of Artificial Intelligence (AI), characterized by unprecedented advances in machine learning algorithms and large language models, has pervaded many decision-making domains [1], enabling AI systems to imitate and augment human cognitive functions in reasoning, learning, and problem solving [2–4].”

“Recent empirical evidence from digital health implementations consistently validates the positive relationship between system quality and user acceptance [90,91].”

“…this vein, evaluating information quality involves assessing factors such as completeness, accuracy, uniqueness, relevance, timeliness, precision, comprehensibility, conciseness, and informativeness that pertain to relationships with users’ acceptance and intention to use IS-based services [76,77].”

“The domain has been recognized in the extant literature as a significant factor influencing users’ behavioral intentions [99,100] and actual use of digital systems [101].”

“[150] confirmed that system quality does not significantly influence behavioral intention to use online services, suggesting that technical attributes become less salient when trust and service interactions dominate adoption decisions.”

“…with prior empirical evidence affirming that satisfaction functions as a direct predictor and motivator for sustained system engagement [52,94,116,118], this study posits that users who experience high satisfaction with healthcare chatbot services through positive interactions, perceived…”

“…services over other AI technological alternatives, particularly in reshaping healthcare service delivery models by enhancing accessibility, personalization, and efficiency [5,8,9], while simultaneously recasting patient education, interaction, and engagement with health information [10–12].”

“While AR technology has been developed for various applications in healthcare and education, its use in a blended-reality approach is considered effective for administering CBT, particularly for SAD [28–32].”