mHealth
In 2011, World Health Organization (WHO) defined mobile health (mHealth) as the: “medical and public health practice supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants (PDAs), and other wireless devices” (34). mHealth technology has enormous potential for assisting elderlies in self-management of health, chronic disease management, and living healthy such as assistance for persons with dementia through mobile-app (35). mHealth encompasses receiving healthcare services in various ways, such as voice and short messaging service (SMS), mobile apps, third and fourth generation mobile telecommunications, and mobile sensors integrated with the internet of things (IoT). Today, mhealth services for various age groups, including elderlies(36-39) are emerging as a new field in the healthcare industry and research(18).
Theoretical models
The process of technology acceptance by users has been studied for many years and various theoretical models on technology acceptance have been developed over time. While some theories, such as technology acceptance model (TAM) and theory of reasoned action (TRA), laid the foundation for understanding users’ technology acceptance behavior, other theories, such as unified theory of acceptance and use of technology (UTAUT) and UTAUT2, advanced our understanding of this process. Although TAM and TRA provided much needed insights into the major universal factors that are likely to affect a user’s perceptions of a particular form of technology, UTAUT2 is recognized by researchers (see for example: 40, 41-46) as one of the leading theoretical models with its strong empirical support for the acceptance of various forms of technology by end-users. Given the widespread and global acceptance of UTAUT2, this study is grounded by this theoretical model. We believe that the UTAUT2 model provides us with necessary foundation for this study and allows us to drive a more simplified model that is more appropriate in the context of our study.
The original UTAUT (47) theory was incepted in 2003 to explain users’ intention to adopt new technology and their subsequent use behavior. It was developed by integrating the previous eight competing but dominant theories on technology adoption and its use. UTAUT was designed to provide a more unified view of understanding factors affecting new technology acceptance. According to UTAUT, the core constructs that influence individuals’ intention to use technology are performance expectancy, effort expectancy, and social influence (47). UTAUT also asserts that facilitating conditions are important in forming individuals’ actual use behavior, which together with intention can explain the actual technology use behavior (47). This unified view was later extended to UTAUT2 by including three additional factors – hedonic motivation, price value, and habit. These additional constructs are shown to provide insights for enhanced understanding of both behavioral intention and the actual use technology use behavior.
Since its inception, UTAUT-based research has thrived and made significant impact in both research and practice in the past decade. A review of extant literature suggests the utilization of UTAUT theory in a wide area of research including education settings (44, 45), eHealth literacy (48), healthcare settings (49-51), and many other areas of technology use. Despite the wide use of UTAUT in Information Systems (IS) research, there are several limitations with UTAUT in explaining individuals’ technology adoption behavior regardless of technological and demographical contexts. For example, UTAUT model does not consider cultural factors even though some studies suggest a significant role of culture on technology adoption (52). In addition, UTAUT theory also does not take into consideration the aspect of trust in determining behavioral intention and technology use behavior formation. Noting some of these limitations with the utilization of UTAUT, Venkatesh et al. (53) called for a paradigm shift of UTAUT extensions in technology acceptance and use. The authors of UTAUT further recommended the use of this model using the theoretical notion of contextualization because “context has become one of the important theoretical lens” in IS research (53). Thus, in this study, we grounded our foundation using UTAUT2 as shown in Figure 1 and contextualized the theory in the context of mHealth services use by the elderly population in Hong Kong.
H1: Performance expectancy is positively associated with an individual’s behavioral intention to use mHealth services.
H2: Effort expectancy is positively associated with an individual’s behavioral intention to use mHealth services.
H3a: Social influence is positively associated with an individual’s behavioral intention to use mHealth services.
H4: Facilitating conditions is not important in determining behavioral intention to use mHealth services.
H5: Habit is positively associated with an individual’s behavioral intention to use mHealth services.
H6a: Hedonic motivation is positively associated with an individual’s behavioral intention to use mHealth services.
H7a: Price value is positively associated with an individual’s intention for the use of mHealth services.
Performance Expectancy
Performance Expectancy refers to "the degree to which an individual believes that using the system will help him or her to attain gains in performance" (47). In the context of this study, we define PE as the degree to which an individual believes that using mHealth services will enhance the condition of his or her health. PE has been found as one of the strongest predictors (see for example: 51, 54, 55) of individuals’ behavioral intention (BI) to adopt and use technology. PE is also found as a powerful determinant in the intention to adopt new healthcare related technologies. For instance, the study by Hoque and Sorwar (24) on the adoption of mHealth in the context of Bangladesh reported that PE has a direct effect on behavioral intention. Their study indicated that one unit of change in performance expectancy could cause more than 0.31 units of change in behavioral intention. Similarly, Woldeyohannes and Ngwenyama (56) studied the influence of PE on BI for mHealth adoption and reported PE is very relevant for mHealth adoption considering the ability to meet time-demand with accurate content and reliable performance. Apart from general users, when studied on health professionals such as physicians, PE also affected BI significantly (57). Therefore, based on the evidence found in existing literature, we proposed the following hypothesis.
H1: Performance expectancy is positively associated with an individual’s behavioral intention to use mHealth services.
Effort Expectancy
Venkatesh et al. (47) defined effort expectancy (EE) as “the degree of ease associated with the use of the system”. We define EE in the context of this study as the degree to which an individual believes that receiving mHealth services are easy without needing significant efforts. EE has been reported to have a significant positive impact on technology use behavioral intention (see for example: 44, 51, 54). Researchers also identified this construct as a key factor that directly influences users' intention in using healthcare systems. For example, Sun et al. (58) reported a significant relationship between these two constructs when studied for mobile health monitoring systems. A significant predictor of BI was found in the study by Woldeyohannes and Ngwenyama (56) when the users were asked about the level of difficulty they might find in using mHealth. Hoque and Sorwar (24) also found EE as a determinant of intention in the Bangladesh context. Moreover, Breil (59) studied the acceptance of mHealth applications among people with hypertension and EE to be a significant predictor of BI. Based on these findings, we proposed the following hypothesis:
H2: Effort expectancy is positively associated with an individual’s behavioral intention to use mHealth services.
Social Influence
Venkatesh et al. (47) defined social influence (SI) as “the degree to which an individual perceives that important others (e.g., family and friends) believe they should use the new system." In this study, we define SI as the degree to which an individual believes that his or her decision to receive mHealth services is influenced by significant others. Existing research (see for example: 60, 61) suggest SI a significant contributor to individuals’ behavioral intention decision when pertaining to technology use. In the healthcare context, SI has a strong impact on users' intention to adopt new technology. For example, it is widely found significant in the developing countries where the family members generally influence the elderly users to use mHealth (see for example: 24, 62). Similarly, Kijsanayotin et al. (51) reported that SI positively influences individuals’ health information technology adoption intention in Thailand’s community health centers. Furthermore, Sun et al. (58) found significant positive influence of SI on individuals’ adoption intention of mobile health services. Thus, relying on the findings in existing literature, we posited the following hypothesis for the relationship between an individual’s social influence and his or her behavioral intention to use mHealth services.
H3a: Social influence is positively associated with an individual’s behavioral intention to use mHealth services.
H3b: Social influence positively influences an individual’s trust belief in mHealth services.
Facilitating Conditions
Facilitating condition (FC) describes users’ perceptions about potential (situational or environmental) conditions that either facilitate or hinder for taking certain action(s). The potential conditions can be influenced by either internal or external or both factors. In the original UTAUT study, Venkatesh et al. (47) refer to it as "the degree to which an individual believes that an organizational and technical infrastructure exists to support the use of the system". To fit the context of this study, we defined FC as the degree to which an individual believes they have the necessary knowledge, resources, and supports for using mHealth services. As we study this technology use by older population, we believe FC will play a critical role and a dominating determinant of BI, given that this group of population is specifically known to lack behind and inexperience in using different technologies (see for example: 63). FC has been found to be a determinant in BI for using different technologies. Venkatesh et al. (47) have shown the importance of facilitating condition in shaping individuals’ behavioral intention. This construct is also found to influence healthcare technology use behavior. For example, Dwivedi et al. (64) conducted a cross-country study on adopting mHealth services in the context of USA, Canada, and Bangladesh, and reported facilitating condition significantly influence behavioral intention to utilize the technology. Cimperman et al. (26) also reported a similar finding in investigating the elderlies’ acceptance behavior regarding telehealth services. However, Venkatesh et al. (47) stated that the influence of facilitating conditions on behavioral intention diminishes and it is non-significant in predicting intention if both performance expectancy and effort expectancy are presence in the model. Several other studies (24, 55, 65) also found non-significant relationships between these two constructs in the presence of performance expectancy and effort expectancy constructs. Thus, based on the evidence in the existing literature, we proposed the following hypothesis regarding the relationship between facilitating condition and behavioral intention to use of mHealth services.
H4: Facilitating conditions is not important in determining behavioral intention to use mHealth services.
Habit
Since the incorporation of habit (HA) in the extended UTAUT model, a growing number of studies are focused on understanding the impact of habit on technology adoption decision. Habit can be defined as individuals’ unconscious and automatic past behavior that is frequently repeated. Researchers have attributed habit for overriding intentional action (66). Venkatesh et al. (54) described habit as “the extent to which people tend to perform behaviors automatically because of learning”. Habitual behavior is known as a repetition of past behavior in the future given all other conditions being equal (67). In the context of this study, we define habit as an individual’s belief about his or her frequent use of mHealth services and one’s natural reliance of this technology for healthcare needs. Existing research (see for example: 47, 68, 69) suggests that consumers’ habits play a significant impact on technology use, both directly and as a behavioral intention path to affect their behavior. In fact, “the learning process at the early stages of IS adoption will be complemented once users gain the necessary knowledge about the goals of using the IS” (70). Evidence in existing literature suggests a significant positive and direct relationship between consumers’ habits and their behavioral intentions to use technology (see for example: 67, 71). In fact, some reported habit is more significant for elderly population for forming their technology related behavioral intention (72). Furthermore, several recent studies (see for example: 43, 70) linked habit with individuals’ behavioral intention to adopt healthcare-related technologies. As it can be safely assumed that “habitual previous behavior in a given context will predict behavioral intentions in the same context” (67), we believe individuals’ habit of using technology for personal needs will influence their behavioral intention to use mHealth services for similar needs. Hence, the following hypothesis was proposed.
H5: Habit is positively associated with an individual’s behavioral intention to use mHealth services.
Hedonic Motivation
Venkatesh et al. (54) defined hedonic motivation (HM) as “the fun or pleasure derived from using technology”. Research has shown strong relationship between this construct and behavioral intention to acceptance and use of technology. For example, Brown and Venkatesh (73) reported hedonic motivation to play an important role in determining technology acceptance. Alalwan et al. (74) found HM to have significant positive influence on consumers’ intention to adopt Internet banking. To understand the impact of user acceptance of mobile technology in healthcare, Sudbury et al. (75) reported that HM has significant impacts on BI in the healthcare context. Gao et al. (76) found that hedonic motivation significantly influences individuals’ behavioral intention to accept wearable technology for healthcare purposes. In a recent study, Talukder et al. (65) also validated the efficacy of the UTAUT2 for healthcare technology adoption context and found HM significantly influences individuals’ behavioral intention to accept wearable healthcare technology by elderly population. Therefore, relying on the evidence from existing literature, we posited the following hypothesis regarding the relationship between hedonic motivation and behavioral intention to use mHealth services by elderly population.
H6a: Hedonic motivation is positively associated with an individual’s behavioral intention to use mHealth services.
Hedonic motivation is also known to influence people’s habits. Although most technology adoption studies suggest strong relationship between hedonic motivation and intention to use technology, a few studies investigated interrelationship between hedonic motivation and habit. Habit is found as a mediator between hedonic motivation and intention to use technology (77) such that hedonic motivation influences people habit, which in turn influence behavioral intention. Furthermore, Chiu (78) found that hedonic motivation leads to the building of Habit of technology use. Based on this evidence in existing literature we believe that hedonic motivation is an important antecedent of people’s formation of habit and, thus, posited the following hypothesis.
H6b: Hedonic motivation is positively associated with Habit in the use of mHealth services.
Price Value
Price value (PV) is an outcome of a cost-benefit decision as individuals go through a process of cost and benefit analysis to gauge their perception about the value of certain products or services. The tradeoff between consumers’ benefits (i.e., efficiency, convenience, quality, etc.) and costs (i.e., monetary expenses, difficulty of use, sacrifice, etc.) perceptions determine their value perceptions, which then further influences their decision. Thus, researchers defined perceived value as “consumers’ cognitive tradeoff between the perceived benefits of the applications and the monetary cost for using them” (54). Aw et al. (79) referred to it as “consumers’ perception toward subjective worthiness of service consumption”. In the context of this study, we define perceived value as the degree to which an individual believes that the use of mHealth services will provide more benefits than the costs of using the technology. A number of studies have shown that users’ perception of value derives from people’s intention to use and continue to use technologies and technological services. For example, Mehta et al. (80) reported a direct and positive influence of price value in individuals’ behavioral intention to adopt e-learning technology. Aw et al. (79) also reported a positive relationship between perceived value construct and consumers’ intention to use ridesharing services. Many other studies (54, 81) have also suggested a similar relationship between these two constructs. As mHealth is a cost-effective medium of receiving healthcare service, the researchers infer that it is a strong determinant of behavioral intention to continue to use the technology (68, 82, 83). Thus, the following hypothesis was posited.
H7a: Price value is positively associated with an individual’s intention for the use of mHealth services.
Although there is no study to date shows a direct relationship between price value and habit, studies on the formation of people habit suggest that people habit can be influenced by increasing or decreasing value or reward. For example, Loewenstein et al. (84) reported that increasing incentives help in forming habits with regards to healthy food consumption. Thus, we hypothesized the following relationship between price value and habit.
H7b: Price value is positively associated with an individual’s habit in using mHealth services.
Trust
As technology is integrated into every aspect in our lives, technology trust is considered one of the most important facilitators of people’s willingness to use technologies (85) and, thus, received attention by researchers and practitioners across different disciplines. Technology trust is defined as “the belief that specific technology has the capability, functions, or features to do for one what one needs to be done” (86). In the context of this study, we define technology trust (TR) belief as the degree to which an individual believes that mHealth technology has the capability to provide adequate and responsive help to fulfill their healthcare needs. Although UTAUT2 has been widely used to explain different technology adoption behaviors including healthcare technology adoptions, one of the criticisms for UTAUT2 is that it lacks a trust component. Technology adoption decisions are often facilitated by individuals’ trust on the specific technology. For example, a number of studies (87-89) can be found reporting trust on Internet technology is an essential facilitator in people engaging in online transactions. Thus, it is important that we explore the relationship between trust and behavioral intention in studying mHealth technology adoption.
The influence of technology trust in technology adoption decision is well cited in existing IS literature. Many studies (see for example: 90, 91, 92) can be found focusing on understanding the trust belief and its influence on technology adoption decision. Lankton et al. (93) studied the differences between human-like trust and system-like trust and found that individuals’ trust related to specific system is an important antecedent for intention to continue to use the system. Technological trust is even more important in healthcare as more and more healthcare services are being delivered using technology that requires patients’ interaction, engagement, and disclosure of their sensitive health and personal information. Lack of trust in healthcare technology can result to many adverse effects on patients’ health. Studies suggest that lack of trust on healthcare technology leads to non-disclosure of necessary health information (94-96), which are critical in providing appropriate lifesaving care for patients. Zulman et al. (97) studied technology trust and use of healthcare resources by elderly population and reported distrust on Internet technology is responsible for the avoidance to use the technology as a health resource. Greater Trust is believed to affect behavioral intention positively. A very recent study by Alam et al. (82) also focuses on healthcare technology trust in a developing country. Their findings suggest that trust is appositively associated with behavioral intention to use healthcare technologies and one of the most significant predictors of mHealth apps adoption intention in Bangladesh. In a similar study, Meng et al. (98) also found that trust in mHealth services is essential for mHealth technology adoption by elderly users. Therefore, based on this evidence, we propose the following hypothesis for technology trust and behavioral intention to use mHealth services.
H8: Technology trust is positively associated with an individual’s behavioral intention to use mHealth services.
As mHealth services are provided and received remotely, many factors can hinder or facilitate their successful delivery and reception. For example, the quality of service delivered is often time serve as factor that helps to gain trust which in turn determines continuous use of the technology. In addition, use of mHealth services requires clients to disclose and exchange their personal and sensitive data over the Internet using different devices, such as tables, cellphone, etc. The concern for security and privacy is an important factor in situations where data are exchanged between hosts and clients. Thus, given the complex nature of the trust construct, it is imperative that we dig deeper to understand how individuals’ perceptions are formed for this construct. Thus, we include two additional constructs (Service Quality and Government Policy) that will serve as antecedent factors. We believe these two additional factors will provide important insights into the formation of mHealth services users’ trust belief.
Service Quality
Although delivering quality service has been an essential component in the success of technology adoption, the construct, service quality (SQ), has found little attention in IS literature. The prevalent use of IT for improving customer satisfaction garnered much attraction on the topic of IT service quality in recent years. The perception of service quality is the consequence of individuals’ evaluation of the quality of service received using technology. This concept emerged from marketing literature where customers are known to engage in assessing service they receive from vendors. Consumers’ perception of service quality results from the comparison of customers’ prior expectation and their perception of actual experience of service performance (99). This construct has been extensively studied in the marketing as well as consumer behavior literature, where service quality is commonly defined as individuals’ judgement about overall superiority of service experience (100). In the context of this study, we define service quality perception as the degree to which mHealth services can meet the needs of its users. A number of research studies have found relationship between individuals’ perceived service quality and their behavioral intention. Zeithaml et al. (101) studied the behavioral consequences of service quality on behavioral intention and reported important relationship between the two constructs. Similarly, DeLone and McLean (102) reported using their IS success model that service quality influences intention to use. Several other studies (103, 104) reported indirect relationships where service quality influences satisfaction, which in turn influences behavioral intention. Perceived serviced quality has also been studied in the healthcare context. For instance, Akter et al. (105) studied mHealth continuance intention in Bangladesh and found that service quality perception significantly influences the continuance of mHealth services use. Dagger et al. (106) conducted an in-depth study on service quality in the healthcare context and reported that perceived service quality has significant influence in individuals’ intention to utilize healthcare services. Based on these evidences, we posited the following hypothesis.
H9a: Service quality perception positively influences an individual’s behavioral intention to use mHealth services.
Individuals often rely on their experience with the quality of service they receive to gain a level of trust in the technology. Positive service quality experience increases their trust, whereas negative experience with service quality lowers their trust toward the technology. Evidence in existing literature also suggests a strong relationship between perceived service quality and individuals’ trust beliefs. For example, a large number of studies (107-110) reported that service quality positively influences in shaping individuals’ trust beliefs. In the context of healthcare, using an empirical study, Chang et al. (111) found service quality influences patient trust in medical services. Similarly, Akter et al. (105) studied mHealth services continuance use behavior and found that service quality not only influences behavioral intention but also influences trust beliefs, which in turn influences individuals mHealth services continuance intention. Thus, based on these findings, the following hypothesis is proposed.
H9b: Perceived service quality positively influences an individual’s trust belief.
Government Policy
Government policy (GP) is known to shape the direction of citizens’ product use. For example, if government of a country imposes sanction on using a certain type or brand of technology (e.g., US sanctions on using Huawei devices), it will prevent citizens from using that technology. On the other hand, if the government pass laws and regulations to promote certain product or services, citizens will find it easy to use them. Thus, government policies play a critical role in diffusing certain technology in society by making favorable policies and environments. In investigating technology adoption issues, Ejiaku (112) reported that various government policies play an important role in influencing citizens’ technology adoption. Using SWOT analysis, Sharma and Sehrawat (113) concluded that environmental factors, such as lack of government policy, are responsible for adoption intention of cloud technology. In effort to study technology adoption in developing countries, Dasgupta et al. (114) found that environmental factor, such as government policies, have a significant impact on information technology adoption decisions. A recent study by Wang et al. (115) reports that government policy, such as use promotion, is one of the most important factors in users’ continuance intention in using technology. Government policy is expected to enhance the acceptance of various healthcare technology, such as the use of electronic health record (EHR) and mHealth services. Acknowledging the importance and direct impact of government policy on technology adoption in healthcare, Middleton (116) and many other researchers have called for government policy change to facilitate technology adoption in the healthcare section. Llewellyn et al. (117) reported that healthcare policies act as a significant barrier or facilitator for promoting hospital and community-based services. Many studies recommended government policymakers to pass policies to increase mHealth adoption. For example, Hoque et al. (24) and Chen et al. (118) proposed to make government policies to maximize mHealth services adoption. Based on these evidences, we posited the following hypothesis for government policy and behavioral intention to continue to use mHealth services.
H10a: Favorable government policy positively influences an individual’s behavioral intention to use mHealth services.
Government policy is also known to influence people’s trust toward technology or service. Government policy often time shape whether individuals view certain technology or service as safe or risky. Especially if the technology requires people to disclose their sensitive personal information. In studying human resource information systems, Lippert & Swiercz (119) suggest that policy is an important factor in establishing technology trust, and the level of trust among users guide their decisions whether to use or not to use the technology. The absence of an appropriate policy leads to uncertainties that motivate individuals to avoid technologies. Lu et al. (120) studied facilitating conditions of trust for wireless technology and found that policies play a key role in establishing users’ trust, which in turn influences intention to use the technology. Although we found no study examining the relationship between government healthcare policies and healthcare technology trust and technology adoption, given the sensitive nature of health information, we believe healthcare policies promoted by government are a critical antecedent and have more impact on shaping users’ healthcare technology trust beliefs. Thus, we posited the following hypothesis.
H10b: Favorable government policy is positively associated with an individual’s trust belief.
In addition to the service quality and government policy, we believe social influence will also serve as an important antecedent factor for forming individuals’ trust beliefs toward mHealth services. Existing research suggests that social influence greatly influences individuals’ trust beliefs toward specific technologies. For example, Beldad et al. (121) argued that users are more inclined to trust a specific technology if they see widespread use and significant others expect them to use the technology. In studying the formation of trust in social network technologies, Chang et al. (122) found that social influence is a significant antecedent of technology trust and social influence positively influences individuals’ trust beliefs toward the technology. Thus, we posited the following hypothesis.
H1: Performance expectancy is positively associated with an individual’s behavioral intention to use mHealth services.
H2: Effort expectancy is positively associated with an individual’s behavioral intention to use mHealth services.
H3a: Social influence is positively associated with an individual’s behavioral intention to use mHealth services.
H3b: Social influence positively influences an individual’s trust belief in mHealth services.
H4: Facilitating conditions is not important in determining behavioral intention to use mHealth services.
H5: Habit is positively associated with an individual’s behavioral intention to use mHealth services.
H6a: Hedonic motivation is positively associated with an individual’s behavioral intention to use mHealth services.
H6b: Hedonic motivation is positively associated with Habit in the use of mHealth services.
H7a: Price value is positively associated with an individual’s intention for the use of mHealth services.
H7b: Price value is positively associated with an individual’s habit in using mHealth services.
H8: Technology trust is positively associated with an individual’s behavioral intention to use mHealth services.
H9a: Service quality perception positively influences an individual’s behavioral intention to use mHealth services.
H9b: Perceived service quality positively influences an individual’s trust belief.
H10a: Favorable government policy positively influences an individual’s behavioral intention to use mHealth services.
H10b: Favorable government policy is positively associated with an individual’s trust belief.
Figure 2 shown above is the extended UTAUT2 model that we propose as the second research model for this study. Based on the findings of our first two (UTAUT2 and Extended UTAUT2) research models, we propose the third simplified research model for this study. We refer to the simplified model as Health Technology Service Acceptance (HTSA) model, which is shown in Figure 3.
H3b: Social influence positively influences an individual’s trust belief in mHealth services.
H5: Habit is positively associated with an individual’s behavioral intention to use mHealth services.
H6b: Hedonic motivation is positively associated with Habit in the use of mHealth services.
H7b: Price value is positively associated with an individual’s habit in using mHealth services.
H8: Technology trust is positively associated with an individual’s behavioral intention to use mHealth services.
H9b: Perceived service quality positively influences an individual’s trust belief.
H10b: Favorable government policy is positively associated with an individual’s trust belief.