1. Introduction
The rise of web-based technologies, aided by a wide range of products and delivery services, has facilitated the development of mobile health, attracting consumers from offline hospitals to online visits. Although mobile health is still flourishing, several issues have emerged due to inefficient patient–physician communication or physical presentation in the virtual realm [1]. To reduce the ensuing perceived risks, mobile health livestreaming with integrated social media features has been developed [2]. On livestreaming platforms, qualified healthcare celebrities are able to sign up for a channel where they can spread health knowledge and communicate with consumers in real time. Consumers can freely enter or exit the mobile health livestreaming platform at any time. When consumers have a health consultation need, they can use the platform to ask the streamer and other consumers who are watching. The livestreaming also provides links to paid consultations, including text consultations, phone consultations, and video consultations, with the aim of streamlining the ordering process and offering special discounts to consumers [3]. In this circumstance, consumers have access to more healthcare information and also can engage in the live health consultation experience with content contribution [4].
Consumer engagement for communication platforms that cater to their health needs and interests is increasing rapidly, which in turn is leading to a surge in mobile health livestreaming platforms [5]. In China, livestreaming platforms have gained immense popularity, with the country rapidly becoming the global leader in the livestreaming market [6]. The healthcare industry adopted this trend with several mobile health livestreaming platforms jumping on the bandwagon. According to a report by IResearch [7], 27.5% of livestreaming merchants in 2021 come from the medical and cosmetic medicine sector in China. Particularly during the COVID-19 pandemic, these mobile health livestreaming platforms proved to be an effective channel for virtual health consultation and served as a medium to promote epidemic prevention knowledge to the public. By offering a livestreaming platform, online health platforms can attract and retain a substantial number of consumers. For instance, Health 160 (
Given the fundamental importance of consumer traffic to the success of livestreaming, it is critical to comprehend the level of consumer engagement. Online consumer engagement encompasses subsequent behaviors that arise from an individual’s cognitive and affective needs, such as entertaining, watching, consuming, awarding, and commenting [9]. Despite the academic attention given to consumer engagement [10,11,12,13], empirical studies remain insufficient in understanding the cognitive and affective mechanisms of consumers in mobile health livestreaming. Furthermore, current investigations related to mobile health livestreaming primarily employ qualitative methods without adequate reliability and validity tests [14,15]. While the market for mobile health livestreaming has grown, academia has not kept up with this trend. Several sustainable development studies call for attention to new health promotion initiatives using ICTs, as they are conducive to addressing COVID-19 global crises and achieving the sustainable development of business and well-being [16,17].
While livestreaming has been utilized for several years in fields such as gaming and education, it has not been commonly used by most health consumers. However, due to the COVID-19 pandemic, mobile health livestreaming has rapidly developed as an alternative for patients seeking consultations. This shift in behavior may result in differences between the decision-making processes of consumers engaging in live mobile health consultations and those who prefer offline visits. Although engagement has been extensively studied in various healthcare segments, such as the online health community [18], online medication purchasing [19,20], and online health consultation [21,22], limited studies have explored engagement in the emerging field of mobile health livestreaming. Therefore, this study aims to investigate consumer decision making using the dual-process theory regarding cognitive and affective mechanisms. This study also examines the perceived vicariousness, perceived synchronicity, arousal, and affinity, which provides insights into mobile health livestreaming consultation behavioral decisions. The significant growth of the health consumer population has also promoted the formation of a consumer-centered health service industry chain and has played a crucial role in upgrading intelligent healthcare services. Consequently, an enhanced understanding of mobile health livestreaming consumer engagement can benefit the healthcare industry.
What are the characteristics of mobile health livestreaming that engage consumers in a post-crisis period? How do mobile health livestreaming features influence consumer engagement through cognitive and affective mechanisms? How do affective mechanisms influence the relationship between cognitive mechanism and consumer engagement? Further investigation is needed. Therefore, to answer these research questions, we constructed a validated assessment model to identify the driving force of consumer engagement in mobile health livestreaming and to determine the cognitive and affective mechanisms of consumer engagement in China. The paper is structured as follows: the following section presents the literature review and theoretical basis of this study. The methodology describes the research design. The results are then presented. Finally, the discussion and the conclusions are highlighted.
2. Literature Review and Theoretical Basis
2.1. Mobile Health Livestreaming and Functional Features
Mobile health livestreaming is a subset of online health consultation. Unlike asynchronous health consultation technology, it relies on a set of features created by real-time interactive technology, such as live imaging, information distribution, and caching [23]. The combined application of these features enables synchronized communication between physicians and patients, improves the quality of patient–physician interaction, and meets the individualized needs of patients for medical assistance [24,25]. Consequently, researchers have posited that real-time medical consultations can engage the consumer and motivate their adoption and purchase behavior in a more social and interactive way [26,27,28]. It provides a more synchronized, intelligent, and vicarious channel for consumers to receive consultations about their conditions on a website or platform.
Figure 1 illustrates that the process involved in mobile health livestreaming and traditional online health consultation differs significantly in the pre-, during-, and post-consultation stages. In traditional online health consultation, consumers typically pay for a designated specialist either per visit or per hour before starting free asynchronous or paid synchronous one-on-one communication. The interaction is usually limited to one-on-one communication between the consumer and the healthcare provider. In contrast, mobile health livestreaming allows an interaction to be built on free group interactions, ranging from consumer to consumer and from consumer to healthcare provider. The advantage of mobile health livestreaming is that the consumer can first experience the interaction before deciding whether to move to a private, paid consultation. During the livestreaming, consumers can continuously watch or participate in physician–patient interactions on the live stream, obtain health guidance simultaneously, and interact with other consumers with similar health concerns. Mobile health live streamers are also inclined to be constantly responsive to consumers’ various disease concerns, providing comprehensive treatment guidance and physical demonstrations to consumers.
Mobile health livestreaming provides a unique and engaging experience compared to traditional online health consultations. The synchronous nature and sense of vicariousness in mobile health livestreaming allow consumers to receive health guidance and interact with healthcare providers more intelligently. Additionally, the visual and audio cues in livestreaming create a sense of arousal and affinity, allowing consumers to feel as comfortable during the health consultation as if they were physically present with the healthcare provider. These factors can affect consumers’ concentration on mobile health livestreaming, potentially enhancing their engagement and trust in the platform.
2.2. Dual-Process Theory
Our theoretical framework evolves from dual-process theory, which originated in the field of social psychology and states that human decision-making behavior are primarily determined by the interplay of cognitive and affective mechanisms [29,30,31]. Cognitive mechanisms reflect the process of rational thinking, in which the consumer actively and selectively shapes their personal behavior from a complex mass of information, inspired by their own motivations. Since rational thinking is a logical operation in the brain, the working of cognitive mechanism is a time-consuming process. Comparatively, affective mechanisms represent fast and automatic intuition in human decision-making behavior. Intuition is hidden deep in consumer’s minds and dictates their unconscious activity. In reality, consumers tend to be irrational and find it difficult to disengage from their decision-making style, and this is where the affective mechanism plays an essential role.
Moreover, the cognitive mechanism and the affective mechanism not only operate independently but also interact with each other. Specifically, when intuition comes into play, consumers’ deliberate behavior may be shaped. Thus, the affective mechanism is commonly considered to be moderator of cognitive mechanism, and when these two mechanisms work cooperatively, consumer decision-making processes exhibit a state that integrally reflects individual intentions. As shown in the existing literature, the philosophy of dual-process theory is well established in various contexts, covering mental health [32], self-regulation [33,34], trust [35], and consumer reviews [36]. In addition, scholars have identified the potential of dual-process theory in healthcare. An integration of the two mechanisms would provide a valuable and different explanation of healthcare. For example, Keech and Hamilton examined a dual-process framework to explore how stress management interventions drive behavior change in university students. Since the dual-process model can help us to effectively explore factors that are conducive to the sustainable development of engagement behavior, this study used dual-process theory to reveal consumer engagement decisions in medical consultation livestreaming [37].
2.3. Mobile Health Livestreaming Engagement
In this paper, our objective is to investigate mobile health livestreaming engagement, which is considered to be a crucial benefit for commercial companies utilizing online platforms [38]. In general, engagement represents consumers’ behavioral manifestation towards a brand or company, and its significance extends beyond mere usage and purchase. In the health context, we define mobile health livestreaming engagement as the level of interaction between physicians and patients and the connection of patients with the platform or e-patient-initiated activities.
Previous research has revealed that engagement encompasses activities with all parties on the platform [39]. To measure consumer engagement for mobile health livestreaming, we adopted the framework proposed by Liu et al. [40], which includes four dimensions: entertainment, search, evaluation, and purchase. According to mobile health livestreaming, the entertainment dimension can be assessed through metrics such as video views, likes, and shares [9]. The search dimension means consumers use mobile health livestreaming to search for health-related information, which can be assessed through search queries and time spent watching livestreaming [41]. Third, the evaluation dimension means that consumers evaluate the quality and reliability of health-related information presented in mobile health livestreaming, which can be assessed through metrics such as consumer feedback, ratings, and reviews [42]. Fourth, the purchase dimension measures the degree to which mobile health livestreaming consumers purchase health-related products or services [43,44]. Thus, by exploring engagement, we aimed to capture a broader range of consumer behaviors and attitudes, including their level of satisfaction with the platform, their intention to continue using the platform, and their likelihood to recommend the platform to others.
2.4. Cognitive Mechanism and Mobile Health Livestreaming Engagement
To understand consumer engagement in the mobile health livestreaming environment, it is crucial to examine the cognitive mechanism involved in decision making [45]. Decision making is often driven by rational analysis, as proposed by the dual-process theory. Thus, it is important to investigate how the conscious rational system operates in shaping consumers’ understanding of mobile health consultation services and how this understanding influences their behavioral decisions. In previous studies on livestreaming, cognitive factors such as perceived vicariousness and perceived synchronicity have been proposed as mechanisms for consumer engagement [46]. Giertz et al. also suggested that cognitive antecedents play a significant role in consumer engagement in livestreaming [47]. Xue et al. specifically focused on the cognitive state and explored the effect of live interactions on social commerce engagement [48].
In this study, vicariousness and synchronicity are two key cognitive factors that distinguish mobile health livestreaming from traditional online health consultations. Meanwhile, perceived intelligence, which refers to the ability of the platform to understand, learn, and provide feedback on consumer needs [49], can be affected by vicariousness and synchronicity in the mobile health livestreaming environment. Firstly, vicariousness refers to the extent to which consumers’ prior experiences resonate with the experiences presented by a live streamer [46]. In the context of mobile health livestreaming, consumers can immerse themselves in the services provided by healthcare professionals and observe other consumers asking questions and receiving guidance, just as they once did offline [50]. When consumers enter a live healthcare streaming room, engaging presentations by streamers, who may use their bodies or demonstrations, can alleviate concerns about the service quality [51] and information credibility [52] among consumers. Therefore, vicariousness is a critical intelligent characteristic of mobile health livestreaming, indicating its role in influencing perceived intelligence. Therefore, we propose:
Perceived vicariousness is positively related to perceived intelligence.
Secondly, synchronicity refers to the real-time nature of mobile health livestreaming, allowing consumers to interact with healthcare providers immediately [45,48]. Synchronous information can facilitate information exchange and create a shared understanding among consumers, enhancing the platform’s trust and perceived intelligence. Additionally, synchronous information can facilitate the timely resolution of issues and the provision of feedback, which can further enhance platform’s interactivity, responsiveness, and service capability [53]. Some scholars point out that consumers connected by much synchronous information will be more likely to perceive platform intelligence in a virtual environment [54,55]. However, no specific contributions on mobile health livestreaming content were found. With previous evidence, it is likely to assume that a similar relationship may occur in mobile health livestreaming, as suggested by the hypothesis:
Perceived synchronicity is positively related to perceived intelligence.
Moreover, perceived intelligence may influence consumer engagement with the mobile health livestreaming platform. Consumers with higher perceived intelligence are more likely to view the platform as a reliable source of information, boosting their confidence and motivation to engage actively [56]. Higher perceived intelligence can also lead to the perception of the platform as more personalized and tailored to their needs, increasing their engagement levels [57]. Further, livestreaming consumers who perceive the platform to be more intelligent are inclined to actively participate in the consultation process, including asking questions and providing feedback, which can result in better health outcomes and a more collaborative and practical consultation experience [48]. In addition, consumers with a higher perception of intelligence may switch to alternative platforms or providers, increasing consumer engagement in the overall mobile health livestreaming market environment [58]. Thus, we propose:
Perceived intelligence is positively related to engagement.
2.5. Affective Mechanism and Mobile Health Livestreaming Engagement
The affective mechanism has been recognized as another effective means of influencing consumer engagement. In the context of livestreaming, affective mechanism refers to the extent to which consumers tend to engage in livestreaming spontaneously and instinctively [59]. Affective mechanism describes a series of emotions that arise from environmental stimuli and how consumers unconsciously develop a set of psychological preferences. Unlike conscious rational analysis, emotions with environmental triggers drive consumers to act in certain ways unconsciously, while consumers cannot clearly explain the triggers afterwards. The literature has confirmed the influence of affective mechanism on consumer beliefs and actual behavior in livestreaming. For instance, using a systematic meta-analysis, Jeyaraj found that affect in the affective mechanism was positively associated with consumers’ willingness to engage [60].
Considering affective mechanism, mobile health livestreaming has two representative features in its business model: arousal and affinity. Firstly, arousal, which refers to consumers feeling that livestreaming elicits excitement and pleasure, is one of the main differences between traditional online health consultation and mobile health livestreaming [61]. When a person is more aroused, for instance through visual or auditory stimulation and traditional belief challenges, they are more likely to better concentrate and perform tasks more effectively, especially tasks that require sustained attention and cognitive effort [62]. For consumers seeking health advice, arousal creates a utilitarian atmosphere between the patient and the physician in mobile health living streaming rooms, thus more effectively capturing consumer concentration [63]. Thus, we propose:
Arousal is positively related to concentration.
Secondly, affinity refers to consumers feeling attracted by livestreaming [64]. The affinity of live streamers allow consumers to interact with a minimal feeling of subjective distance. Live streamers often communicate with consumers using intimate body language and a relaxed speaking style, which in turn creates a comfortable and pleasant atmosphere for interaction. The fondness and comfort derived from affinity can increase immersion perceptions and willingness to focus on current remote interactions [65]. Specific to this study, when consumers enter the mobile health livestreaming room, pleasure stimuli from visual and auditory sources make consumers more prone to experience a sense of immersion, such as vividness and humor in the streaming media or the fantasy and pleasant experience of watching medical celebrities/experts, which further drives consumer concentration on stream media [66]. Consequently, affinity increases consumers’ concentration. Thus, we hypothesize:
Affinity is positively related to concentration.
Moreover, it is worth noting that consumers who concentrate on mobile health livestreaming are more emotionally engaged in the live content. When the livestreaming begins, consumers immediately receive a pop-up invitation and may enter the session, reducing the cost of seeking a health consultation. Even if consumers miss the live consultation, a video review option allows them to revisit the historical interaction information between the health provider and consumers. This phenomenon is evidenced by the fact that the more concentration the consumer has, the higher their level of engagement [67,68]. Therefore, the psychological benefits of livestreaming functionality facilitate consumer engagement in the live stream.
Concentration is positively related to engagement.
Research also revealed that the affective mechanism moderates the effects of cognitive mechanism on consumer engagement. For example, after examining a consumer dataset from a social commerce platform, Farivar et al. identified the moderating role of affective mechanism between trust risk and consumer engagement [69]. In addition, we infer that affective concentration may be crucial for the impact of cognitive intelligence on mobile health livestreaming engagement. Consumers may increasingly join mobile health livestreaming out of affective attachment, rather than just because they are seeking information or advice [70]. Therefore, when the engagement behavior in livestreaming is repeated, subsequent behavior will be gradually controlled by spontaneous affective mechanism.
The relationship between perceived intelligence and engagement is moderated by concentration.
Based on the literature review outlined above, Figure 2 depicts the conceptual model for this study. A conceptual model was constructed using perceived vicariousness (PV), perceived synchronicity (PS), arousal (AR), and affinity (AF) as independent variables; perceived intelligence (PI) and engagement (EG) as dependent variables; and concentration (CC) was examined both as a dependent variable and as a mediating variable.
3. Methodology
3.1. Overview of Research Design
An investigative procedure was designed in order to test our proposed model, based on the method of partial least squares structural equation modeling (PLS-SEM) [71], which outlines the multiple necessary stages covered in this study. To evaluate the validity and reliability of the research findings, the methodology section provides a detailed description of the construct measures, sample and procedures methods, and analysis methods used in the study. The construct measures were developed and validated using three processes [72]: firstly, the creation of items based on prior research; secondly, item refinement by a panel of experts to ensure clarity and validity; and thirdly, the pretesting of the construct by a focus group to ensure questionnaire reliability and quality. The survey was then distributed to actual respondents who participated in mobile health livestreaming, with the unit of analysis being the individual consumer who conducts the interactive decision-making process. After data cleaning, 499 valid answers were collected. Finally, we used the PLS-SEM method to evaluate the validity and reliability of the research findings.
3.2. Construct Measures
All measurements in the questionnaire were adapted from prior validation studies. The items and sources used for the measurements are listed in Appendix A Table A1. Perceived vicariousness and perceived synchronicity were adapted from Li et al. [46]. Perceived intelligence was adapted from McLean et al. [49]. Arousal was assessed using the items from Tong et al. [61]. Affinity was measured according to Franke et al. [64]. The items of concentration were adapted from Eldenfria and Samarraie [67], while those of engagement were measured from Wongkitrungrueng and Assarut [8]. A seven-point Likert scale was applied, as the most accurate and the easiest to use [73], with the anchors of “1, strongly disagree” to “7, strongly agree”. The study also involved relevant control variables for subject characteristics, including gender, age, education, monthly income, experience of mobile health livestreaming usage, frequency of mobile health livestreaming usage, and breadth of mobile health live platform usage, following Chen et al. [31].
A forward–backward translation method was used to design the questionnaire to ensure that respondents’ answers were error-free [74]. First, a language researcher translated all the items into Mandarin, and the other language researcher translated the items written in Mandarin back into English. Afterwards, a mobile health livestreaming streamer and five consumers who watched mobile health livestreaming were asked to verify whether the translation was easy to understand. Meanwhile, to ensure the reliability and effectiveness of the scale, it was necessary to invite experts from relevant fields to review the scale [75]. In this study, three professors involved in consumer behavior research were invited to review the question items. After confirming its suitability, a preliminary trial was conducted on 35 university students to assess the quality of the instruments and further improve the feasibility and logistics of the data collection process. Reliability and validity were verified to ensure the suitability of the questionnaire for the research questions. The ethical issues and considerations associated with the study included obtaining informed consent from study participants, protecting participants’ privacy and confidentiality, and complying with relevant ethical guidelines and regulations.
3.3. Sample and Procedures
Following a literature review of mobile health and a sample population by Yan et al. [76], consumers of mobile health and social media platform content were considered as the target group. We specifically focused on Haodf, Ping An Health, Chunyu Physician, WeDoctor, and Douyin, all of which have a significant online health market base in China and are typical cases of platforms that have enabled mobile health livestreaming functionality. In addition to online health platforms, social media platforms were also within the scope of the study. Douyin, which is one of the largest short-form video communication software in China, is equipped with a livestreaming feature that engages many consumers who maximize social attributes in live commerce. These platforms reflect mobile health livestreaming functionality and electronic consumer engagement without being influenced by the cross-platform user experience.
We distributed an anonymous online questionnaire on Wen Juan Xing (
We determined the required sample size for this study using G*Power software (Version 3.1). Following Franque et al. [77], we considered a significance level of 0.05, an effect size of 0.416, and a power of 100%, yielding a minimum sample size of 120 participants. Of the total participants, most respondents fell into the age group of 31–40 (42.28%), and a higher proportion of female participants (54.71%) was observed compared to males (45.29%). About 50.50% of the participants had an undergraduate degree. Following Kim et al. [78], the distribution of education and sex essentially met the characteristics of mobile platform users in this study. The most common monthly household income category was between CNY 6000 and 9000 (47.09%). In terms of the experience of mobile health livestreaming usage, the top response was 1–2 years (41.48%). The majority of the participants reported the frequency of mobile health livestreaming usage was less than five times per month (49.50%). The largest breadth of mobile health livestreaming platform usage was one (46.69%), which is basically in line with the characteristics of emerging market development.
To verify the nonresponse bias in this study, the hypothesis based on the following extrapolation was used: responders who responded less readily are comparable to non-respondents [79]. The less-readily responding group was defined as those who answered the questionnaire later. To recruit participants, we conducted the research through two rounds. The first round consisted of early responders who completed the questionnaire, while the second round consisted of late responders who also completed the questionnaire. The responses of the two groups were compared to evaluate their responses toward all measurement items. The first round yielded 238 consumers, which provided an initial understanding of the research framework. Three weeks later, we released the second round survey and obtained 261 consumers. A multivariate analysis of variance (MANOVA) was then conducted to confirm that there were no significant differences between the earlier and late groups. In addition, to ensure the quality of the data obtained, attention tests were incorporated into the main study to detect the presence of automatic clickers and inattentive respondents who may not have paid sufficient attention to the survey [80]. The use of extrapolation methods to validate nonresponse bias enhances the validity of the study’s findings. The inclusion of two rounds of respondents and the use of MANOVA to compare the responses of the two groups further strengthen the reliability of the results obtained in this study.
3.4. Analysis Methods
PLS-SEM is widely recognized as an important analytical tool in the social sciences due to its broad statistical scope and its validation as a non-experimental research technique for evaluating measurement and structural models [81]. Therefore, this study employs the PLS-SEM method to analyze data obtained from the mobile health livestreaming field and the impact of various cognitive and emotional factors on consumer engagement in mobile health livestreaming. First, partial least squares (PLS) was performed to evaluate our model, as it is ideally qualified to measure sophisticated structural models. Second, the structural equation model (SEM) was the preferred method, allowing researchers to program the relationships between multiple independent and dependent variables simultaneously [82].
The PLS-SEM analysis is often superior to other analytical techniques for several reasons. Firstly, when the primary goal is to identify potential associations between variables but not the strength of these associations, PLS-SEM is considered the most appropriate method. Secondly, PLS-SEM can handle uncertainty related to factors and avoid excluding solutions, making it a reliable method. Finally, PLS-SEM is a robust method that ensures convergence and reduces statistical identification issues. This study utilized the SmartPLS software version v.3.3.3 to perform PLS-SEM analysis. SmartPLS can handle complex structural models while eliminating the need to adhere to strict data settings such as residual distributions and large samples. Moreover, SmartPLS can streamline the modeling of reflective structures and effectively represent reliability and validity. In the context of mobile health livestreaming, PLS-SEM can help to uncover the relationships that may not be immediately apparent and to identify the important factors that influence consumer behavior.
4. Results
4.1. Measurement Model
The assessment of the measurement model includes reliability and validity. Reliability is assessed by evaluating Cronbach’s α, composite reliability (CR), Rho_A, and indicator loadings. Table 2 demonstrates that all Cronbach’s α and composite reliability values are above 0.7, ensuring the structure’s internal consistency [83]. After the calculation, all item loadings and Rho_A are above 0.7, is an acceptable threshold.
Validity testing involves convergent validity and discriminant validity. Convergent validity was tested by evaluating the extracted average variance (AVE), the square root of AVE, and cross-loadings. AVE can be used to measure the explanatory power of the variables in the model for the measured constructs. According to the standard introduced by Fornell and Larcker [84], the AVE of all structures must be more significant than 0.5, showing an adequate convergent validity. As shown in Table 3, the correlation coefficient is less than the square root of AVE, implying favorable discriminant validity between the structures.
Table 4 shows satisfactory discriminant validity, as each item loading is larger than the cross-loadings of any other structure in the same line.
In this study, the risk of common method bias (CMB) existing in the self-reported data was minimized by following a couple of procedures. First, the questionnaires were randomly distributed in two sessions over one month with participants from different online health platforms and social media platforms in China. This randomized arrangement pattern may reduce CMB. Secondly, the standardized procedure proposed by Liang et al. [85] was employed to extract a common method factor. By comparing the loadings of the common method factor with the substantive structure, it could be determined whether CMB would cause problems in the study [86]. Appendix A Table A2 shows the results after comparison, indicating that the risk of CMB was lower because the average methodological factor loadings (0.049) were significantly lower than the average substantive factor loadings (0.802). In addition, the majority of the method factor loadings were significant. Therefore, it is reasonable to assume that CMB was not a major issue in this study.
4.2. Structural Model
In our study, R-squared (), effect size (), and predictive correlation () were used to assess the quality of the structural model. An value of at least 0.25 was considered acceptable, and an acceptable value was equal to or greater than 0. The effect size was classified into three categories: small (0–0.02), medium (0.02–0.15), and large (0.15–0.35). Results showed that our model accounted for 21.1% of the variance in perceived intelligence, 29.9% of the variance in concentration, and 35.7% of the variance in engagement. The Stone–Geisser test was also applied to obtain values of 0.114, 0.299, and 0.357 for perceived intelligence, concentration, and engagement, respectively, which confirmed the excellent predictive performance of the structural model.
Table 5 displays the results of the structural model, which found that perceived vicariousness had a significant positive effect on perceived intelligence (β = 0.332, p < 0.001), as did perceived synchronicity (β = 0.195, p < 0.001), supporting H1 and H2. The study also revealed that arousal (β = 0.547, p < 0.001) and affinity (β = 0.228, p < 0.001) had significant effects on concentration, supporting H4 and H5. Furthermore, perceived intelligence (β = 0.245, p < 0.001) and concentration (β = 0.461, p < 0.001) were found to significantly impact engagement, supporting H3 and H6.
4.3. Moderation Effects
The result of moderation effects are shown in Table 5, indicating that concentration positively moderated the relationship between perceived intelligence and engagement, supporting H6a. To test this effect, we followed the two-stage approach for moderation analysis recommended by Liang et al. [85]. We also performed slope tests to further investigate the nature of this moderation effect. The analysis revealed significant slopes for both low and high levels of concentration, indicating that engagement increases more rapidly with perceived intelligence at high levels of concentration, while engagement increases only slightly with perceived intelligence at low concentration levels. Our findings on the moderating effect are illustrated in Figure 3.
4.4. Mediation Effects
Following the guidance proposed by Zhao et al. [87], we evaluated the mediating effects between the measurement variables. Table 6 demonstrates the mediated path test that perceived intelligence mediates the impact of perceived vicariousness (β = 0.081, p < 0.001) and perceived synchronicity (β = 0.048, p < 0.01) on engagement and that concentration mediates the impact of arousal (β = 0.253, p < 0.001) and affinity (β = 0.105, p < 0.001). In addition, the 95% CI from the 5000 bootstrap samples was more significant than 0, indicating that mobile health livestreaming features influence e-patients’ engagement through perceived intelligence and concentration.
5. Discussion
We present an exploratory framework based on dual-process theory to uncover the factors influencing consumer engagement in mobile health livestreaming. Four features that distinguish mobile health livestreaming from traditional online health communities are identified, and their effectiveness in activating cognitive and affective mechanisms is tested. It was found that the affective mechanism significantly moderated the relationship between cognitive mechanism and consumer engagement. This study yields three main findings, which are described as follows.
First, the finding of this study reveals that perceived intelligence plays a crucial role in the cognitive mechanism for predicting consumer engagement. This finding is consistent with previous research on personalized engagement marketing, which has suggested that perceived intelligence directly influences consumer engagement decisions [49]. Specifically, the mobile health livestreaming platform can clearly demonstrate its intelligence level to consumers through perceived vicariousness and perceived synchronicity. These livestreaming features provide a good channel for the remote communication of commercial subjects and enable consumers to assess whether perceived intelligence meets their preferences. The perceived intelligence of the mobile health livestreaming platform is the core element that drives consumer engagement, and it is associated with acquiring information about service functions and levels. Consumers strive to meet their health needs, and the level of intelligence demonstrated by the health information owners, who are integrated with the livestreaming platform and health experts, has an impact on the level of consumer engagement in mobile health livestreaming. Thus, the information asymmetry between consumers and mobile health livestreaming providers is significantly reduced, lowering the likelihood of the public suffering unsustainable health-care service losses and economic costs.
Second, our study found that concentration plays a crucial role in the affective mechanism for influencing consumer engagement. As stated by Eldenfria and Al-Samarraie [67], online environmental stimuli can trigger concentration behavior, and our research supports this finding. We discovered that the arousal features lead to increased consumer concentration on mobile health livestreaming, which is supported by previous research [62]. With the increasing support of new health promotions using ICTs, consumers have access to many arousal experiences through the interaction with mobile livestreaming systems. These experiences create interesting impressions that capture consumers’ attention and promote their emotional concentration on the livestreaming content. This arousal effect is subtle, but it significantly influences consumers, ultimately leading to affective concentration. Moreover, we found that affinity drives the concentration on mobile health livestreaming, which is consistent with the findings of Chen and Lin [66]. Compared to other single-interaction methods, mobile health livestreaming is preferred by consumers because it reduces the time and effort required to find suitable health services. As consumers develop a connection with the content and the experts presenting it, they become more invested in the experience and more likely to participate in the livestreaming session.
Furthermore, concentration strengthens the impact of perceived intelligence on mobile health livestreaming engagement. This result is consistent with the interaction hypothesis of the dual-process theory [69]. The formation of concentration enhances the need for cognitive evaluation before engagement, strengthening the awareness of judging the perceived mobile health livestreaming intelligence. When individuals concentrate on the content of mobile health livestreaming, they can better understand the information presented by health experts. This concentration can lead to greater identification with perceived streamers’ intelligence, allowing consumers to better follow and engage in mobile health livestreaming. For example, in the mobile health livestreaming room of Haodf Online, consumers can receive “free appointment coupons”, which allow consumers to receive an additional free appointment opportunity after the physicians’ appointment quota has been fully utilized, which increases consumers’ interest and attracts more consumers to engage in mobile health livestreaming.
6. Conclusions
6.1. Implications
This is the first study to consider a consumer engagement model from the perspective of dual-process theory to distinguish the cognitive mechanism and affective mechanism of engagement decision making in mobile health livestreaming. Our research provides several significant theoretical and managerial contributions.
In this study, we contribute to the literature on mobile health livestreaming by proposing a dual-process model that investigates the effects of mobile health livestreaming features on consumer engagement from the perspectives of cognitive and affective mechanisms. While some studies have explored health promotion using information and communication technologies (ICTs), little is known about the factors that drive consumer engagement in the mobile health livestreaming context [88]. We identify perceived intelligence as a cognitive factor and concentration as an affective factor, both of which contribute to online consumer engagement and the increasing popularity of mobile health livestreaming [49,89]. Our findings demonstrate a significant relationship between mobile health livestreaming features and consumer engagement, mediated by perceived intelligence and concentration.
Secondly, our findings contribute to the research field of mobile health livestreaming through an exploration of mobile health livestreaming features that may activate cognitive and affective effects on consumers. In contrast to traditional mobile health communities, we identified four mobile health livestreaming characteristics and examined the impact of these four characteristics on the activation of cognitive and affective effect that may induce consumer engagement. From the perspective of cognitive mechanism, previous studies have identified various cognitive effects that encourage engagement by transmitting clues [90]. Our study further examines cognitive effects by hypothesizing that perceived vicariousness and perceived synchronicity are essential factors that may lead to perceived intelligence. From the perspective of affective mechanism, arousal and affinity were first proposed and conceptualized in mobile health livestreaming. They were proven to activate affective effects due to their positive association with concentration. Furthermore, this study reveals a moderating role of concentration between perceived intelligence and consumer engagement in mobile health livestreaming.
Thirdly, our study on cognitive mechanism provides practical contributions for the sustainable development of mobile health livestreaming, which requires creating and disseminating more intelligence signals to enable consumers to evaluate health services. We identified perceived vicariousness and perceived synchronicity to significantly influence consumers’ perceived intelligence of health services, leading to engagement. Therefore, for platforms still needing livestreaming, a practical implication is to consider technological means to further release the intelligence signals of health services. For example, intelligent recommendation systems can be used in mobile health platforms to recommend better health services that fit consumers’ preferences. This function can improve the level of platform intelligence to help consumers identify higher levels of perceived intelligence. Additionally, platforms that have already adopted livestreaming should expand the advantages of communication and interaction through more enriched IT artifacts. For example, platform service providers can also use virtual reality technology to allow consumers to experience different health scenario, providing a more immersive health service experience and intuitively enabling consumers to feel the intelligence signals of health services.
Fourthly, this study provides a practical viewpoint that the affective concentration mechanism is another sustainable driving force for consumer engagement. Consumers typically develop concentration during their engagement in mobile health livestreaming activities. In other words, unconscious affective processes lead to consumer engagement. Similarly, it is recommended to use two mobile health livestreaming features to activate affective mechanism. One of these features is affinity, which suggests that streaming media for health services should be both interesting and lively. For example, humorous interactive games can be interspersed during health services to stimulate the communication atmosphere in the live room. The characteristics of arousal also affect consumer engagement. Mobile health livestreaming streamers should improve language skills, terminology, and overall style. Consumers seek the health knowledge as comfortably as possible during livestreaming. Thus, providing accessible sustainable and humanized health consultation services may attract more consumers [17].
6.2. Limitations and Future Research Agenda
This study is subject to several limitations and offers future directions for the study to continue in depth. Firstly, the platforms selected in this study, such as Chunyu Physician and WeDoctor, may pose potential problems such as the incomplete characterization of livestreaming. Therefore, future research could explore livestreaming characteristics, such as livestreaming technology and livestreaming culture, which influence consumer engagement to varying degrees [91]. In addition, future research could apply empirical methods more strictly, such as machine learning models, econometric models, and natural experiments, to analyze the subtle effects of mobile health livestreaming features on different types of consumer engagement. In future studies, transaction-level data should also be encouraged to examine consumer engagement across different live mobile health platforms if conditions permit.
Secondly, this study focused on consumers, but health experts and platforms are also integral parts of livestreaming activities. Future research could approach practical issues in mobile health livestreaming from multiple perspectives. For example, researchers can extract multidimensional variables from live room conversation transcripts to identify their effect on the efficiency of consumer engagement or employ game theory to discuss the balance among health experts, consumers, and platforms.
Thirdly, although our study filled a gap in the literature by considering perceived intelligence and concentration, the explanation for the differences in mobile health livestreaming features still needs to be more comprehensive. As technologies become more prevalent, it is important to understand how they may affect consumer engagement in mobile health livestreaming. The impact of technological advancements, such as the integration of virtual and augmented reality in mobile health livestreaming, could also be examined to supplement the conclusions of our study.
Conceptualization, F.L. and X.W.; methodology, X.W.; software, X.W.; validation, F.L., Q.Z., S.L. and X.W.; formal analysis, X.W.; investigation, S.L. and Q.Z.; resources, F.L.; data curation, X.W. and Q.Z.; writing—original draft preparation, X.W.; writing—review and editing, X.W. and Q.Z.; visualization, X.W. and Q.Z.; supervision, F.L.; project administration, F.L.; funding acquisition, F.L. All authors have read and agreed to the published version of the manuscript.
The study was performed under the Declaration of Helsinki and was authorized by the Ethics Committee of the Institute of Education and Economics of the University of International Business and Economics, No. IEER20220177.
Informed consent was obtained from all subjects involved in the study.
Data are available upon special request from the corresponding author.
The authors declare no conflict of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 1. Comparison between traditional online health consultation and mobile health livestreaming.
Demographics of respondents (n = 499).
| Demographics | Characteristics | Frequency | Percent (%) |
|---|---|---|---|
| Sex | Male | 226 | 45.29% |
| Female | 273 | 54.71% | |
| Age | 20 s | 72 | 14.43% |
| 30 s | 120 | 24.05% | |
| 40 s | 212 | 42.48% | |
| 50 s | 95 | 19.04% | |
| Education | High school certificate or below | 47 | 9.42% |
| Technical school | 113 | 22.65% | |
| Undergraduate degree | 252 | 50.50% | |
| Master’s or higher degree | 87 | 17.43% | |
| Monthly income, CNY | <3000 | 34 | 6.81% |
| (3000, 6000) | 184 | 36.87% | |
| (6000, 9000) | 235 | 47.09% | |
| >9000 | 46 | 9.22% | |
| Experience of mobile health livestreaming usage | Under 1 year | 166 | 33.27% |
| 1–2 years | 207 | 41.48% | |
| Over 2 years | 126 | 25.25% | |
| Frequency of mobile health livestreaming usage | Less than 5 times per month | 247 | 49.50% |
| 5–10 times per month | 191 | 38.28% | |
| More than 10 times per month | 61 | 12.22% | |
| Breadth of mobile health livestreaming platform usage | 1 | 233 | 46.69% |
| 2 | 168 | 33.67% | |
| ≥3 | 98 | 19.64% |
Reliability and validity.
| Variables | Items | Ladings | Cronbach’s α | CR | Rho_A | AVE |
|---|---|---|---|---|---|---|
| Perceived vicariousness (PV) | PV1 | 0.864 | 0.838 | 0.901 | 0.872 | 0.752 |
| PV2 | 0.839 | |||||
| PV3 | 0.897 | |||||
| Perceived synchronicity (PS) | PS1 | 0.915 | 0.916 | 0.940 | 0.934 | 0.798 |
| PS2 | 0.883 | |||||
| PS3 | 0.876 | |||||
| PS4 | 0.899 | |||||
| Perceived intelligence (PI) | PI1 | 0.861 | 0.889 | 0.923 | 0.889 | 0.750 |
| PI2 | 0.851 | |||||
| PI3 | 0.889 | |||||
| PI4 | 0.862 | |||||
| Arousal (AR) | AR1 | 0.855 | 0.798 | 0.881 | 0.803 | 0.713 |
| AR2 | 0.859 | |||||
| AR3 | 0.818 | |||||
| Affinity (AF) | AF1 | 0.944 | 0.907 | 0.942 | 0.907 | 0.844 |
| AF2 | 0.868 | |||||
| AF3 | 0.942 | |||||
| Concentration (CC) | CC1 | 0.833 | 0.753 | 0.859 | 0.755 | 0.669 |
| CC2 | 0.825 | |||||
| CC3 | 0.796 | |||||
| Engagement (EG) | EG1 | 0.971 | 0.990 | 0.991 | 0.990 | 0.935 |
| EG2 | 0.968 | |||||
| EG3 | 0.968 | |||||
| EG4 | 0.969 | |||||
| EG5 | 0.975 | |||||
| EG6 | 0.963 | |||||
| EG7 | 0.959 | |||||
| EG8 | 0.962 |
Note: CR, composite reliability; AVE, average variance extracted.
Discriminant validity: Fornell–Larcker criterion.
| Items | (1) PV | (2) PS | (3) PI | (4) AR | (5) AF | (6) CC | (7) EG |
|---|---|---|---|---|---|---|---|
| (1) PV | 0.873 | ||||||
| (2) PS | 0.465 | 0.869 | |||||
| (3) PI | 0.537 | 0.495 | 0.916 | ||||
| (4) AR | 0.526 | 0.456 | 0.351 | 0.855 | |||
| (5) AF | 0.468 | 0.350 | 0.475 | 0.482 | 0.865 | ||
| (6) CC | 0.530 | 0.579 | 0.438 | 0.540 | 0.528 | 0.880 | |
| (7) EG | 0.607 | 0.534 | 0.545 | 0.519 | 0.476 | 0.590 | 0.857 |
Note: on the diagonal is the square root of AVE; below the diagonal is the correlation.
Item loadings and cross-loadings.
| PV | PS | PI | AR | AF | CC | EG | |
|---|---|---|---|---|---|---|---|
| PV1 | 0.864 | 0.034 | 0.154 | 0.046 | 0.097 | 0.037 | 0.046 |
| PV2 | 0.839 | 0.072 | 0.082 | 0.009 | 0.047 | 0.060 | 0.062 |
| PV3 | 0.897 | 0.063 | 0.200 | 0.037 | 0.061 | 0.003 | 0.069 |
| PS1 | 0.016 | 0.915 | 0.046 | 0.042 | 0.262 | 0.087 | 0.157 |
| PS2 | 0.022 | 0.883 | 0.056 | 0.103 | 0.301 | 0.010 | 0.148 |
| PS3 | 0.014 | 0.876 | 0.010 | 0.119 | 0.217 | 0.028 | 0.107 |
| PS4 | 0.013 | 0.899 | 0.096 | 0.079 | 0.222 | 0.018 | 0.129 |
| PI1 | 0.170 | 0.094 | 0.861 | 0.299 | 0.201 | 0.190 | 0.072 |
| PI2 | 0.143 | 0.023 | 0.851 | 0.108 | 0.232 | 0.234 | 0.101 |
| PI3 | 0.178 | 0.059 | 0.889 | 0.188 | 0.195 | 0.134 | 0.108 |
| PI4 | 0.165 | 0.044 | 0.862 | 0.155 | 0.185 | 0.104 | 0.005 |
| AR1 | 0.095 | 0.135 | 0.359 | 0.855 | 0.293 | 0.161 | 0.060 |
| AR2 | 0.086 | 0.198 | 0.390 | 0.859 | 0.298 | 0.276 | 0.097 |
| AR3 | 0.031 | 0.208 | 0.321 | 0.818 | 0.184 | 0.175 | 0.202 |
| AF1 | 0.087 | 0.063 | 0.112 | 0.031 | 0.944 | 0.042 | 0.167 |
| AF2 | 0.164 | 0.039 | 0.103 | 0.116 | 0.868 | 0.075 | 0.539 |
| AF3 | 0.137 | 0.016 | 0.076 | 0.041 | 0.942 | 0.082 | 0.164 |
| CC1 | 0.085 | 0.022 | 0.287 | 0.096 | 0.337 | 0.833 | 0.093 |
| CC2 | 0.081 | 0.116 | 0.238 | 0.131 | 0.271 | 0.825 | 0.313 |
| CC3 | 0.016 | 0.057 | 0.235 | 0.385 | 0.230 | 0.796 | 0.117 |
| EG1 | 0.002 | 0.382 | 0.182 | 0.188 | 0.041 | 0.213 | 0.971 |
| EG2 | 0.002 | 0.376 | 0.174 | 0.139 | 0.012 | 0.199 | 0.968 |
| EG3 | 0.045 | 0.378 | 0.214 | 0.181 | 0.074 | 0.142 | 0.968 |
| EG4 | 0.022 | 0.330 | 0.212 | 0.147 | 0.021 | 0.175 | 0.969 |
| EG5 | 0.016 | 0.341 | 0.195 | 0.163 | 0.051 | 0.215 | 0.975 |
| EG6 | 0.003 | 0.314 | 0.170 | 0.193 | 0.049 | 0.174 | 0.963 |
| EG7 | 0.008 | 0.303 | 0.237 | 0.127 | 0.037 | 0.202 | 0.959 |
| EG8 | 0.044 | 0.356 | 0.229 | 0.178 | 0.089 | 0.159 | 0.962 |
Note: All items are significant at a p-value of <0.01.
Structural path analysis results.
| Hypothesis | β | Standard Deviation | T Statistics | p Value | Confidence Interval 95% |
|
Supported |
|---|---|---|---|---|---|---|---|
| (H1) PV→PI | 0.332 *** | 0.044 | 7.53 | 0 | [0.244, 0.416] | 0.130 | Yes |
| (H2) PS→PI | 0.195 *** | 0.043 | 4.571 | 0 | [0.112, 0.280] | 0.045 | Yes |
| (H3) PI→EG | 0.245 *** | 0.05 | 4.933 | 0 | [0.148, 0.342] | 0.066 | Yes |
| (H4) AR→CC | 0.547 *** | 0.037 | 14.741 | 0 | [0.471, 0.615] | 0.259 | Yes |
| (H5) AF→CC | 0.228 *** | 0.041 | 5.551 | 0 | [0.145, 0.308] | 0.080 | Yes |
| (H6) CC→EG | 0.461 *** | 0.049 | 9.461 | 0 | [0.363, 0.556] | 0.228 | Yes |
| (H6a) CC × PI→EG | 0.088 *** | 0.017 | 5.075 | 0 | [0.058, 0.126] | 0.032 | Yes |
Note: Perceived vicariousness, PV; perceived synchronicity, PS; perceived intelligence, PI; arousal, AR; affinity, AF; concentration, CC; engagement, EG; *** p-value < 0.001.
Mediation analysis.
| Mediation Paths |
β | Standard Deviation | T Statistics | p Value | Confidence Interval 95% | Supported |
|---|---|---|---|---|---|---|
| PV→PI→EG | 0.081 *** | 0.020 | 4.049 | 0.000 | [0.045, 0.123] | Yes |
| PS→PI→EG | 0.048 ** | 0.016 | 3.048 | 0.002 | [0.021, 0.083] | Yes |
| AR→CC→EG | 0.253 *** | 0.029 | 8.643 | 0.000 | [0.196, 0.312] | Yes |
| AF→CC→EG | 0.105 *** | 0.028 | 3.818 | 0.000 | [0.056, 0.163] | Yes |
Note: Independent variable, IV; moderator, M; dependent variable, DV; perceived vicariousness, PV; perceived synchronicity, PS; perceived intelligence, PI; arousal, AR; affinity, AF; concentration, CC; engagement, EG; *** p-value < 0.001; ** p-value < 0.01.
Appendix A
Questionnaire items.
| Construct | Measurement Items | Reference |
|---|---|---|
| Perceived vicariousness |
PV1: During a live stream, I can feel what the streamer is trying to say about the recommended health solutions and their guidance experience. |
Li et al. [ |
| Perceived synchronicity |
PS1: During a live stream, the platform processes my comments inputs very quickly. |
Li et al. [ |
| Perceived Intelligence |
PI1: During a live stream, the streamer had the ability to identify and respond to viewers’ health needs. |
McLean et al. [ |
| Arousal |
AR1: The mobile health livestreaming room excites me. |
Tong et al. [ |
| Affinity |
AF1: If there is no live broadcast, I would miss it. |
Franke et al. [ |
| Concentration |
CC1: I do not think of anything other than interacting with streamers. |
Eldenfria and Samarraie [ |
| Engagement |
EG1: I spend more time on mobile health livestreaming. |
Wongkitrungrueng and Assarut [ |
Common method bias measurement.
| Construct | Indicator | Substantive Factor Loading |
R12 | Method |
R22 |
|---|---|---|---|---|---|
| Perceived vicariousness (PV) | PV1 | 0.714 | 0.510 | 0.012 | 0.000 |
| PV2 | 0.851 | 0.724 | −0.001 | 0.000 | |
| PV3 | 0.857 | 0.734 | 0.013 | 0.000 | |
| Perceived synchronicity (PS) | PS1 | 0.797 | 0.635 | 0.044 | 0.002 |
| PS2 | 0.752 | 0.566 | 0.045 | 0.002 | |
| PS3 | 0.691 | 0.477 | 0.039 | 0.002 | |
| PS4 | 0.781 | 0.610 | 0.042 | 0.002 | |
| Perceived intelligence (PI) | PI1 | 0.786 | 0.618 | 0.044 | 0.002 |
| PI2 | 0.758 | 0.575 | 0.042 | 0.002 | |
| PI3 | 0.761 | 0.579 | 0.043 | 0.002 | |
| PI4 | 0.727 | 0.529 | 0.040 | 0.002 | |
| Arousal (AR) | AR1 | 0.619 | 0.383 | 0.047 | 0.002 |
| AR2 | 0.656 | 0.430 | 0.050 | 0.003 | |
| AR3 | 0.576 | 0.332 | 0.044 | 0.002 | |
| Affinity (AF) | AF1 | 0.743 | 0.552 | 0.056 | 0.003 |
| AF2 | 0.657 | 0.432 | 0.050 | 0.003 | |
| AF3 | 0.755 | 0.570 | 0.057 | 0.003 | |
| Concentration (CC) | CC1 | 0.874 | 0.764 | 0.044 | 0.002 |
| CC2 | 0.871 | 0.759 | 0.043 | 0.002 | |
| CC3 | 0.805 | 0.648 | 0.038 | 0.001 | |
| EG1 | 0.941 | 0.885 | 0.071 | 0.005 | |
| Engagement (EG) | EG2 | 0.936 | 0.876 | 0.071 | 0.005 |
| EG3 | 0.938 | 0.880 | 0.071 | 0.005 | |
| EG4 | 0.934 | 0.872 | 0.071 | 0.005 | |
| EG5 | 0.945 | 0.893 | 0.072 | 0.005 | |
| EG6 | 0.928 | 0.861 | 0.070 | 0.005 | |
| EG7 | 0.864 | 0.746 | 0.071 | 0.005 | |
| EG8 | 0.939 | 0.882 | 0.071 | 0.005 | |
| Average | N. A. | 0.802 | 0.643204 | 0.049 | 0.002 |
References
1. Balapour, A.; Reychav, I.; Sabherwal, R.; Azuri, J. Mobile technology identity and self-efficacy: Implications for the adoption of clinically supported mobile health apps. Int. J. Inf. Manag.; 2019; 49, pp. 58-68. [DOI: https://dx.doi.org/10.1016/j.ijinfomgt.2019.03.005]
2. Hilty, D.; Chan, S.; Torous, J.; Luo, J.; Boland, R. A Framework for Competencies for the Use of Mobile Technologies in Psychiatry and Medicine: Scoping Review. JMIR MHealth UHealth; 2020; 8, e12229. [DOI: https://dx.doi.org/10.2196/12229] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32130153]
3. Zhou, M.; Huang, J.; Wu, K.; Huang, X.; Kong, N.; Campy, K.S. Characterizing Chinese consumers’ intention to use live e-commerce shopping. Technol. Soc.; 2021; 67, 101767. [DOI: https://dx.doi.org/10.1016/j.techsoc.2021.101767]
4. Song, S.; Xue, X.; Zhao, Y.C.; Li, J.; Zhu, Q.; Zhao, M. Short-Video Apps as a Health Information Source for Chronic Obstructive Pulmonary Disease: Information Quality Assessment of TikTok Videos. J. Med. Internet Res.; 2021; 23, e28318. [DOI: https://dx.doi.org/10.2196/28318]
5. Gopalsamy, R.; Semenov, A.; Pasiliao, E.; McIntosh, S.; Nikolaev, A. Engagement as a Driver of Growth of Online Health Forums: Observational Study. J. Med. Internet Res.; 2017; 19, e304. [DOI: https://dx.doi.org/10.2196/jmir.7249]
6. Iimedia. 2022–2023 China Live Streaming E-Commerce Industry Operation Big Data Analysis and Trends Research Report. Available online: https://www.iimedia.cn/c400/86233.html (accessed on 20 February 2023).
7. IResearch. 2022 China Online Medical and Health Services Consumption White Paper. Available online: https://www.iresearch.com.cn/Detail/report?id=4057&isfree=0 (accessed on 20 February 2023).
8. Wongkitrungrueng, A.; Assarut, N. The role of live streaming in building consumer trust and engagement with social commerce sellers. J. Bus. Res.; 2020; 117, pp. 543-556. [DOI: https://dx.doi.org/10.1016/j.jbusres.2018.08.032]
9. Dolan, R.; Conduit, J.; Frethey-Bentham, C.; Fahy, J.; Goodman, S. Social media engagement behavior: A framework for engaging customers through social media content. Eur. J. Mark.; 2019; 53, pp. 2213-2243. [DOI: https://dx.doi.org/10.1108/EJM-03-2017-0182]
10. Tarute, A.; Nikou, S.; Gatautis, R. Mobile application driven consumer engagement. Telemat. Inform.; 2017; 34, pp. 145-156. [DOI: https://dx.doi.org/10.1016/j.tele.2017.01.006]
11. Jozani, M.; Ayaburi, E.; Ko, M.; Choo, K.-K.R. Privacy concerns and benefits of engagement with social media-enabled apps: A privacy calculus perspective. Comput. Hum. Behav.; 2020; 107, 106260. [DOI: https://dx.doi.org/10.1016/j.chb.2020.106260]
12. Bitrián, P.; Buil, I.; Catalán, S. Enhancing user engagement: The role of gamification in mobile apps. J. Bus. Res.; 2021; 132, pp. 170-185. [DOI: https://dx.doi.org/10.1016/j.jbusres.2021.04.028]
13. Oakley-Girvan, I.; Yunis, R.; Longmire, M.; Ouillon, J.S. What Works Best to Engage Participants in Mobile App Interventions and e-Health: A Scoping Review. Telemed. E-Health; 2022; 28, pp. 768-780. [DOI: https://dx.doi.org/10.1089/tmj.2021.0176]
14. Xu, J.H.; Cai, Y.; Fang, Z.; Paliyawan, P. Promoting Mental Well-Being for Audiences in a Live-Streaming Game by Highlight-Based Bullet Comments. Proceedings of the 2021 IEEE 10th Global Conference on Consumer Electronics; Kyoto, Japan, 12–15 October 2021; pp. 383-385.
15. Goetzen, A.; Wang, R.; Redmiles, E.M.; Zannettou, S.; Ayalon, O. Likes and Fragments: Examining Perceptions of Time Spent on TikTok 2023. arXiv; 2013; arXiv: 2303.02041[DOI: https://dx.doi.org/10.48550/arXiv.2303.02041]
16. OECD. Available online: https://www.oecd.org/statistics/Measuring-impacts-of-business-on-well-being.pdf (accessed on 28 April 2023).
17. Kocağ, E.K.; Popescu, C.R.G. Coping With COVID-19 While Focusing on Good Health and Well-Being: Vaccination Willingness. Frameworks for Sustainable Development Goals to Manage Economic, Social, and Environmental Shocks and Disasters; IGI Global: Hershey, PA, USA, 2022; pp. 1-15.
18. Tseng, H.-T.; Ibrahim, F.; Hajli, N.; Nisar, T.M.; Shabbir, H. Effect of privacy concerns and engagement on social support behaviour in online health community platforms. Technol. Forecast. Soc. Change; 2022; 178, 121592. [DOI: https://dx.doi.org/10.1016/j.techfore.2022.121592]
19. Henderson, E.M.; Keogh, E.; Rosser, B.A.; Eccleston, C. Searching the Internet for help with pain: Adolescent search, coping, and medication behaviour. Br. J. Health Psychol.; 2013; 18, pp. 218-232. [DOI: https://dx.doi.org/10.1111/bjhp.12005] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23126577]
20. Park, E.; Kwon, M. Health-Related Internet Use by Children and Adolescents: Systematic Review. J. Med. Internet Res.; 2018; 20, e120. [DOI: https://dx.doi.org/10.2196/jmir.7731]
21. Osei-Frimpong, K.; Wilson, A.; Lemke, F. Patient co-creation activities in healthcare service delivery at the micro level: The influence of online access to healthcare information. Technol. Forecast. Soc. Change; 2018; 126, pp. 14-27. [DOI: https://dx.doi.org/10.1016/j.techfore.2016.04.009]
22. Chen, S.; Guo, X.; Wu, T.; Ju, X. Exploring the Online Doctor-Patient Interaction on Patient Satisfaction Based on Text Mining and Empirical Analysis. Inf. Process. Manag.; 2020; 57, 102253. [DOI: https://dx.doi.org/10.1016/j.ipm.2020.102253]
23. Wang, C.L. New frontiers and future directions in interactive marketing: Inaugural Editorial. J. Res. Interact. Mark.; 2021; 15, pp. 1-9. [DOI: https://dx.doi.org/10.1108/JRIM-03-2021-270]
24. Lu, Z.; Xia, H.; Heo, S.; Wigdor, D. You Watch, You Give, and You Engage: A Study of Live Streaming Practices in China. Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems; Montreal, QC, Canada, 21–26 April 2018; pp. 1-13.
25. Wilke, J.; Mohr, L.; Yuki, G.; Bhundoo, A.K.; Jiménez-Pavón, D.; Laiño, F.; Murphy, N.; Novak, B.; Nuccio, S.; Ortega-Gómez, S. et al. Train at home, but not alone: A randomised controlled multicentre trial assessing the effects of live-streamed tele-exercise during COVID-19-related lockdowns. Br. J. Sports Med.; 2022; 56, pp. 667-675. [DOI: https://dx.doi.org/10.1136/bjsports-2021-104994]
26. Leung, L.; Chen, C. E-health/m-health adoption and lifestyle improvements: Exploring the roles of technology readiness, the expectation-confirmation model, and health-related information activities. Telecommun. Policy; 2019; 43, pp. 563-575. [DOI: https://dx.doi.org/10.1016/j.telpol.2019.01.005]
27. Alam, M.Z.; Hu, W.; Kaium, M.A.; Hoque, M.R.; Alam, M.M.D. Understanding the determinants of mHealth apps adoption in Bangladesh: A SEM-Neural network approach. Technol. Soc.; 2020; 61, 101255. [DOI: https://dx.doi.org/10.1016/j.techsoc.2020.101255]
28. Wu, C.; Zhou, Y.; Wang, R.; Huang, S.; Yuan, Q. Understanding the Mechanism Between IT Identity, IT Mindfulness and Mobile Health Technology Continuance Intention: An Extended Expectation Confirmation Model. Technol. Forecast. Soc. Change; 2022; 176, 121449. [DOI: https://dx.doi.org/10.1016/j.techfore.2021.121449]
29. Chaiken, S.; Trope, Y. Dual-Process Theories in Social Psychology; Guilford Press: New York, NY, USA, 1999; pp. 462-483.
30. van Gelder, J.-L.; de Vries, R.E.; van der Pligt, J. Evaluating a dual-process model of risk: Affect and cognition as determinants of risky choice. J. Behav. Decis. Mak.; 2009; 22, pp. 45-61. [DOI: https://dx.doi.org/10.1002/bdm.610]
31. Chen, H.; Chen, H.; Tian, X. The dual-process model of product information and habit in influencing consumers’ purchase intention: The role of live streaming features. Electron. Commer. Res. Appl.; 2022; 53, 101150. [DOI: https://dx.doi.org/10.1016/j.elerap.2022.101150]
32. Coyne, S.M.; Rogers, A.A.; Zurcher, J.D.; Stockdale, L.; Booth, M. Does time spent using social media impact mental health?: An eight year longitudinal study. Comput. Hum. Behav.; 2020; 104, 106160. [DOI: https://dx.doi.org/10.1016/j.chb.2019.106160]
33. Hamilton, K.; Gibbs, I.; Keech, J.J.; Hagger, M.S. Reasoned and implicit processes in heavy episodic drinking: An integrated dual-process model. Br. J. Health Psychol.; 2020; 25, pp. 189-209. [DOI: https://dx.doi.org/10.1111/bjhp.12401] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31876984]
34. Khan, S.K.; Ali, N.; Khan, N.A.; Ammara, U.; Anjum, N. Understanding multiscreening phenomenon for online shopping through perspective of self-regulation and dual process theory: Case of Chinese young generation. Electron. Commer. Res. Appl.; 2020; 42, 100988. [DOI: https://dx.doi.org/10.1016/j.elerap.2020.100988]
35. Han, X.; Du, J.T.; Zhang, T.; Han, W.; Zhu, Q. How online ratings and trust influence health consumers’ physician selection intentions: An experimental study. Telemat. Inform.; 2021; 62, 101631. [DOI: https://dx.doi.org/10.1016/j.tele.2021.101631]
36. Biswas, B.; Sengupta, P.; Kumar, A.; Delen, D.; Gupta, S. A critical assessment of consumer reviews: A hybrid NLP-based methodology. Decis. Support Syst.; 2022; 159, 113799. [DOI: https://dx.doi.org/10.1016/j.dss.2022.113799]
37. Keech, J.J.; Hamilton, K. An integrated dual-process model for coping behaviour. Stress Health; 2022; 38, pp. 591-601. [DOI: https://dx.doi.org/10.1002/smi.3121]
38. Tran, L.T.T. Managing the effectiveness of e-commerce platforms in a pandemic. J. Retail. Consum. Serv.; 2021; 58, 102287. [DOI: https://dx.doi.org/10.1016/j.jretconser.2020.102287]
39. Shahbaznezhad, H.; Dolan, R.; Rashidirad, M. The Role of Social Media Content Format and Platform in Users’ Engagement Behavior. J. Interact. Mark.; 2021; 53, pp. 47-65. [DOI: https://dx.doi.org/10.1016/j.intmar.2020.05.001]
40. Kumar, A.; Salo, J.; Li, H. Stages of User Engagement on Social Commerce Platforms: Analysis with the Navigational Clickstream Data. Int. J. Electron. Commer.; 2019; 23, pp. 179-211. [DOI: https://dx.doi.org/10.1080/10864415.2018.1564550]
41. Baumel, A.; Muench, F.; Edan, S.; Kane, J.M. Objective User Engagement With Mental Health Apps: Systematic Search and Panel-Based Usage Analysis. J. Med. Internet Res.; 2019; 21, e14567. [DOI: https://dx.doi.org/10.2196/14567] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31573916]
42. Cechetti, N.P.; Bellei, E.A.; Biduski, D.; Rodriguez, J.P.M.; Roman, M.K.; De Marchi, A.C.B. Developing and implementing a gamification method to improve user engagement: A case study with an m-Health application for hypertension monitoring. Telemat. Inform.; 2019; 41, pp. 126-138. [DOI: https://dx.doi.org/10.1016/j.tele.2019.04.007]
43. Meng, F.; Zhang, X.; Liu, L.; Ren, C. Converting readers to patients? From free to paid knowledge-sharing in online health communities. Inf. Process. Manag.; 2021; 58, 102490. [DOI: https://dx.doi.org/10.1016/j.ipm.2021.102490]
44. Zeng, Q.; Zhuang, W.; Guo, Q.; Fan, W. What factors influence grassroots knowledge supplier performance in online knowledge platforms? Evidence from a paid Q&A service. Electron. Mark.; 2022; 32, pp. 2507-2523. [DOI: https://dx.doi.org/10.1007/s12525-022-00588-2]
45. Behera, R.K.; Bala, P.K.; Dhir, A. The emerging role of cognitive computing in healthcare: A systematic literature review. Int. J. Med. Inf.; 2019; 129, pp. 154-166. [DOI: https://dx.doi.org/10.1016/j.ijmedinf.2019.04.024]
46. Li, Y.; Li, X.; Cai, J. How attachment affects user stickiness on live streaming platforms: A socio-technical approach perspective. J. Retail. Consum. Serv.; 2021; 60, 102478. [DOI: https://dx.doi.org/10.1016/j.jretconser.2021.102478]
47. Giertz, J.N.; Weiger, W.H.; Törhönen, M.; Hamari, J. Content versus community focus in live streaming services: How to drive engagement in synchronous social media. J. Serv. Manag.; 2021; 33, pp. 33-58. [DOI: https://dx.doi.org/10.1108/JOSM-12-2020-0439]
48. Xue, J.; Liang, X.; Xie, T.; Wang, H. See now, act now: How to interact with customers to enhance social commerce engagement?. Inf. Manage.; 2020; 57, 103324. [DOI: https://dx.doi.org/10.1016/j.im.2020.103324]
49. McLean, G.; Osei-Frimpong, K.; Barhorst, J. Alexa, do voice assistants influence consumer brand engagement?—Examining the role of AI powered voice assistants in influencing consumer brand engagement. J. Bus. Res.; 2021; 124, pp. 312-328. [DOI: https://dx.doi.org/10.1016/j.jbusres.2020.11.045]
50. Ma, X.; Zou, X.; Lv, J. Why do consumers hesitate to purchase in live streaming? A perspective of interaction between participants. Electron. Commer. Res. Appl.; 2022; 55, 101193. [DOI: https://dx.doi.org/10.1016/j.elerap.2022.101193]
51. Lu, B.; Chen, Z. Live streaming commerce and consumers’ purchase intention: An uncertainty reduction perspective. Inf. Manag.; 2021; 58, 103509. [DOI: https://dx.doi.org/10.1016/j.im.2021.103509]
52. Men, J.; Zheng, X.; Davison, R.M. The role of vicarious learning strategies in shaping consumers’ uncertainty: The case of live-streaming shopping. Internet Res.; 2023; ahead-of-print [DOI: https://dx.doi.org/10.1108/INTR-06-2022-0415]
53. Tan, H.; Yan, M. Physician-user interaction and users’ perceived service quality: Evidence from Chinese mobile healthcare consultation. Inf. Technol. People; 2020; 33, pp. 1403-1426. [DOI: https://dx.doi.org/10.1108/ITP-01-2019-0039]
54. Costa, R. da On a new community concept: Social networks, personal communities, collective intelligence. Interface-Comun. Saúde Educ.; 2005; 9, pp. 235-248. [DOI: https://dx.doi.org/10.1590/S1414-32832005000200003]
55. Lee, Y.; Kim, D. The influence of technological interactivity and media sociability on sport consumer value co-creation behaviors via collective efficacy and collective intelligence. Int. J. Sports Mark. Spons.; 2021; 23, pp. 18-40. [DOI: https://dx.doi.org/10.1108/IJSMS-04-2020-0058]
56. Nazir, S.; Khadim, S.; Ali Asadullah, M.; Syed, N. Exploring the influence of artificial intelligence technology on consumer repurchase intention: The mediation and moderation approach. Technol. Soc.; 2023; 72, 102190. [DOI: https://dx.doi.org/10.1016/j.techsoc.2022.102190]
57. Gao, Y.; Liu, H. Artificial intelligence-enabled personalization in interactive marketing: A customer journey perspective. J. Res. Interact. Mark.; 2022; pp. 1-18. ahead-of-print [DOI: https://dx.doi.org/10.1108/JRIM-01-2022-0023]
58. Kumar, V.; Rajan, B.; Venkatesan, R.; Lecinski, J. Understanding the Role of Artificial Intelligence in Personalized Engagement Marketing. Calif. Manage. Rev.; 2019; 61, pp. 135-155. [DOI: https://dx.doi.org/10.1177/0008125619859317]
59. Hu, M.; Chaudhry, S.S. Enhancing consumer engagement in e-commerce live streaming via relational bonds. Internet Res.; 2020; 30, pp. 1019-1041. [DOI: https://dx.doi.org/10.1108/INTR-03-2019-0082]
60. Jeyaraj, A. Models of information systems habit: An exploratory meta-analysis. Int. J. Inf. Manag.; 2022; 62, 102436. [DOI: https://dx.doi.org/10.1016/j.ijinfomgt.2021.102436]
61. Tong, X.; Chen, Y.; Zhou, S.; Yang, S. How background visual complexity influences purchase intention in live streaming: The mediating role of emotion and the moderating role of gender. J. Retail. Consum. Serv.; 2022; 67, 103031. [DOI: https://dx.doi.org/10.1016/j.jretconser.2022.103031]
62. Munaro, A.C.; Hübner Barcelos, R.; Francisco Maffezzolli, E.C.; Santos Rodrigues, J.P.; Cabrera Paraiso, E. To engage or not engage? The features of video content on YouTube affecting digital consumer engagement. J. Consum. Behav.; 2021; 20, pp. 1336-1352. [DOI: https://dx.doi.org/10.1002/cb.1939]
63. Li, M.; Mao, J. Hedonic or utilitarian? Exploring the impact of communication style alignment on user’s perception of virtual health advisory services. Int. J. Inf. Manag.; 2015; 35, pp. 229-243. [DOI: https://dx.doi.org/10.1016/j.ijinfomgt.2014.12.004]
64. Franke, T.; Attig, C.; Wessel, D. A Personal Resource for Technology Interaction: Development and Validation of the Affinity for Technology Interaction (ATI) Scale. Int. J. Human–Computer Interact.; 2019; 35, pp. 456-467. [DOI: https://dx.doi.org/10.1080/10447318.2018.1456150]
65. Lim, J.S.; Choe, M.-J.; Zhang, J.; Noh, G.-Y. The role of wishful identification, emotional engagement, and parasocial relationships in repeated viewing of live-streaming games: A social cognitive theory perspective. Comput. Hum. Behav.; 2020; 108, 106327. [DOI: https://dx.doi.org/10.1016/j.chb.2020.106327]
66. Chen, C.-C.; Lin, Y.-C. What drives live-stream usage intention? The perspectives of flow, entertainment, social interaction, and endorsement. Telemat. Inform.; 2018; 35, pp. 293-303. [DOI: https://dx.doi.org/10.1016/j.tele.2017.12.003]
67. Eldenfria, A.; Al-Samarraie, H. Towards an Online Continuous Adaptation Mechanism (OCAM) for Enhanced Engagement: An EEG Study. Int. J. Hum.–Comput. Interact.; 2019; 35, pp. 1960-1974. [DOI: https://dx.doi.org/10.1080/10447318.2019.1595303]
68. Holdener, M.; Gut, A.; Angerer, A. Applicability of the User Engagement Scale to Mobile Health: A Survey-Based Quantitative Study. JMIR MHealth UHealth; 2020; 8, e13244. [DOI: https://dx.doi.org/10.2196/13244]
69. Farivar, S.; Turel, O.; Yuan, Y. A trust-risk perspective on social commerce use: An examination of the biasing role of habit. Internet Res.; 2017; 27, pp. 586-607. [DOI: https://dx.doi.org/10.1108/IntR-06-2016-0175]
70. Ladhari, R.; Massa, E.; Skandrani, H. YouTube vloggers’ popularity and influence: The roles of homophily, emotional attachment, and expertise. J. Retail. Consum. Serv.; 2020; 54, 102027. [DOI: https://dx.doi.org/10.1016/j.jretconser.2019.102027]
71. Urbach, N.; Ahlemann, F. Structural Equation Modeling in Information Systems Research Using Partial Least Squares. J. Inf. Technol. Theory Appl. JITTA; 2010; 11, 2.
72. Heart, T. Who Is out There? Exploring the Effects of Trust and Perceived Risk on SaaS Adoption Intentions 2010. ACM SIGMIS Database DATABASE Adv. Inf. Syst.; 2010; 41, pp. 49-68. [DOI: https://dx.doi.org/10.1145/1851175.1851179]
73. Taherdoost, H. What Is the Best Response Scale for Survey and Questionnaire Design; Review of Different Lengths of Rating Scale/Attitude Scale/Likert Scale. Hamed Taherdoost; 2019; pp. 1-10. Available online: https://ssrn.com/abstract=3588604 (accessed on 20 February 2023).
74. Mohatlane, E.J. Back-Translation as a Quality Control Mechanism in Sesotho Translation. J. Soc. Sci.; 2014; 41, pp. 167-175. [DOI: https://dx.doi.org/10.1080/09718923.2014.11893353]
75. Rubio, D.M.; Berg-Weger, M.; Tebb, S.S.; Lee, E.S.; Rauch, S. Objectifying content validity: Conducting a content validity study in social work research. Soc. Work Res.; 2003; 27, pp. 94-104. [DOI: https://dx.doi.org/10.1093/swr/27.2.94]
76. Yan, M.; Filieri, R.; Raguseo, E.; Gorton, M. Mobile apps for healthy living: Factors influencing continuance intention for health apps. Technol. Forecast. Soc. Change; 2021; 166, 120644. [DOI: https://dx.doi.org/10.1016/j.techfore.2021.120644]
77. Franque, F.B.; Oliveira, T.; Tam, C.; Santini, F.d.O. A meta-analysis of the quantitative studies in continuance intention to use an information system. Internet Res.; 2020; 31, pp. 123-158. [DOI: https://dx.doi.org/10.1108/INTR-03-2019-0103]
78. Kim, Y.; Briley, D.A.; Ocepek, M.G. Differential innovation of smartphone and application use by sociodemographics and personality. Comput. Hum. Behav.; 2015; 44, pp. 141-147. [DOI: https://dx.doi.org/10.1016/j.chb.2014.11.059]
79. Armstrong, J.S.; Overton, T.S. Estimating Nonresponse Bias in Mail Surveys. J. Mark. Res.; 1977; 14, pp. 396-402. [DOI: https://dx.doi.org/10.1177/002224377701400320]
80. Chandler, J.; Mueller, P.; Paolacci, G. Nonnaïveté among Amazon Mechanical Turk workers: Consequences and solutions for behavioral researchers. Behav. Res. Methods; 2014; 46, pp. 112-130. [DOI: https://dx.doi.org/10.3758/s13428-013-0365-7]
81. Puteh, F.; Azman Ong, M.H. Quantitative Data Analysis: Choosing Between SPSS, PLS and AMOS in Social Science Research. Int. Interdiscip. J. Sci. Res.; 2017; 3, pp. 14-25.
82. Cheng, E.W.L. SEM being more effective than multiple regression in parsimonious model testing for management development research. J. Manag. Dev.; 2001; 20, pp. 650-667. [DOI: https://dx.doi.org/10.1108/02621710110400564]
83. Bland, J.M.; Altman, D.G. Statistics notes: Cronbach’s alpha. BMJ; 1997; 314, 572. [DOI: https://dx.doi.org/10.1136/bmj.314.7080.572] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/9055718]
84. 84. Fornell, C.; F. Larcker, D. Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. J. Mark. Res.; 2022; 18, pp. 39-50. [DOI: https://dx.doi.org/10.1177/002224378101800104]
85. Liang, H.; Saraf, N.; Hu, Q.; Xue, Y. Assimilation of Enterprise Systems: The Effect of Institutional Pressures and the Mediating Role of Top Management. Manag. Inf. Syst. Q.; 2007; 31, pp. 58-87. [DOI: https://dx.doi.org/10.2307/25148781]
86. Williams, L.J.; Edwards, J.R.; Vandenberg, R.J. Recent Advances in Causal Modeling Methods for Organizational and Management Research. J. Manag.; 2003; 29, pp. 903-936. [DOI: https://dx.doi.org/10.1016/S0149-2063_03_00084-9]
87. Zhao, X.; Lynch, J.G.; Chen, Q. Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis. J. Consum. Res.; 2010; 37, pp. 197-206. [DOI: https://dx.doi.org/10.1086/651257]
88. Akbar, S.; Coiera, E.; Magrabi, F. Safety concerns with consumer-facing mobile health applications and their consequences: A scoping review. J. Am. Med. Inform. Assoc.; 2020; 27, pp. 330-340. [DOI: https://dx.doi.org/10.1093/jamia/ocz175]
89. Olaniyi, B.Y.; del Río, A.F.; Periáñez, Á.; Bellhouse, L. User Engagement in Mobile Health Applications. In Proceedings of 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining; Washington, DC, USA, 14–18 August 2022; pp. 4704-4712.
90. Sittig, S.; Wang, J.; Iyengar, S.; Myneni, S.; Franklin, A. Incorporating Behavioral Trigger Messages Into a Mobile Health App for Chronic Disease Management: Randomized Clinical Feasibility Trial in Diabetes. JMIR MHealth UHealth; 2020; 8, e15927. [DOI: https://dx.doi.org/10.2196/15927]
91. Lee, J.; Lee, J.-N. How purchase intention consummates purchase behaviour: The stochastic nature of product valuation in electronic commerce. Behav. Inf. Technol.; 2015; 34, pp. 57-68. [DOI: https://dx.doi.org/10.1080/0144929X.2013.853837]
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Li, Siheng 1 ; Zhao, Qun 3 1 Institute of Education and Economy Research, University of International Business and Economics, Beijing 100029, China
2 School of Business, Renmin University of China, Beijing 100872, China
3 School of Information Resource Management, Renmin University of China, Beijing 100872, China