This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Constantly increasing health care costs have led countries and health care providers to the point where health care systems must be reinvented. Consequently, electronic health (eHealth) has recently received a great deal of attention in social sciences in the domain of Internet studies. However, only a fraction of these studies focuses on the acceptability of eHealth, making consumers’ subjective evaluation an understudied field. This study will address this gap by focusing on the acceptance of MyData-based preventive eHealth services from the consumer point of view. We are adopting the term "MyData", which according to a White Paper of the Finnish Ministry of Transport and Communication refers to "1) a new approach, a paradigm shift in personal data management and processing that seeks to transform the current organization centric system to a human centric system, 2) to personal data as a resource that the individual can access and control."
The aim of this study was to investigate what factors influence consumers’ intentions to use a MyData-based preventive eHealth service before use.
We applied a new adoption model combining Venkatesh’s unified theory of acceptance and use of technology 2 (UTAUT2) in a consumer context and three constructs from health behavior theories, namely threat appraisals, self-efficacy, and perceived barriers. To test the research model, we applied structural equation modeling (SEM) with Mplus software, version 7.4. A Web-based survey was administered. We collected 855 responses.
We first applied traditional SEM for the research model, which was not statistically significant. We then tested for possible heterogeneity in the data by running a mixture analysis. We found that heterogeneity was not the cause for the poor performance of the research model. Thus, we moved on to model-generating SEM and ended up with a statistically significant empirical model (root mean square error of approximation [RMSEA] 0.051, Tucker-Lewis index [TLI] 0.906, comparative fit index [CFI] 0.915, and standardized root mean square residual 0.062). According to our empirical model, the statistically significant drivers for behavioral intention were effort expectancy (beta=.191,
Our research highlighted the importance of health-related factors when it comes to eHealth technology adoption in the consumer context. Emphasis should especially be placed on efforts to increase consumers’ self-efficacy in eHealth technology use and in supporting healthy behavior.
Constantly increasing health care costs have led countries and health care providers to the point where health care systems must be reinvented. At the same time, technological development has paved the way for new ways to monitor health and well-being and made it possible for societies to start moving health care toward a more personalized and preventive direction. New digital and mobile technologies point to a future in which consumers will be more involved in the management of their health, generating data that will benefit service providers, helping them to create more targeted, preventive, and personalized solutions [
Digital and mobile technologies available for consumers—such as pedometers, hearth-rate measurement instruments, and global positioning system–trackers—are empowering consumers to analyze bodily and mental functioning, something that was once the privilege of health professionals [
The possibility that consumers would take more responsibility for their own health using preventive services has created a new promise of cheaper, better, and more efficient electronic health (eHealth) tools [
For individuals to take more responsibility for their health, it is essential to find ways to liberate the health-related data that organizations have in their possession about consumers’ behavior. It is also important to find incentives for consumers to take active actions with this data. As noted in the study by Kim and Park [
The concept of MyData can be defined as a new approach in personal data management and processing that seeks to look at data management from the consumer’s perspective, and look at personal data as a resource that the consumer can access and control [
To reach their fullest potential and nationwide adoption, it is crucial to understand the consumer perspective on these new health care solutions. A better understanding of health consumers’ intentions and behavior would aid the development and implementation of effective and efficient strategies [
Recently, eHealth has received a great deal of attention in social sciences, in the domain of Internet studies. However, only 3% of these studies focus on the acceptability of eHealth to consumers, making consumers’ subjective evaluations an understudied domain [
Technology acceptance is a relatively mature research area and has received plenty of attention in previous research [
In this paper, we combine the key factors of UTAUT2 and health behavior theories to our research model and apply the quantitative structural equations modeling (SEM) approach to analyze the relationships between the variables of the framework. Data for the study was collected using a quantitative questionnaire survey. The survey was a part of a large national research program called Digital Health Revolution (DHR), coordinated by the University of Oulu.
The paper is composed as follows. In section 2 we introduce the previous research and theories that provide the basis for our empirical model. Section 3 presents the methodology and empirical results. Section 4 provides a discussion on the managerial and theoretical implications and the limitations of the study, as well as providing some suggestions for future research.
In the study by Venkatesh et al [
UTAUT has been successfully adapted and tested in a wide range of contexts such as e-learning [
As noted earlier, the intention to use preventive eHealth services is similar to the intention to engage in health protective behavior. It is therefore crucial to apply theories of health behavior to the study of the acceptance and use of preventive eHealth services. Similar findings have been stated by Riley et al [
The HBM was developed from social, psychology, and behavioral theories to help understand why individuals do or do not engage in health-related actions. The basic assumption behind the model is that an individual will either choose to engage in a particular health-related action or not based on the desire to avoid an illness and the belief that a particular action will prevent the illness [
The original HBM consists of four basic factors that influence an individual’s health motivation and intention to take preventive action: the perceived susceptibility (to a negative health condition), the perceived severity (of a possible negative health condition), the perceived benefits (of a particular action preventing the negative health condition), and the perceived barriers (to taking a preventive health action) [
Thus, according to the HBM, for an individual to engage in preventive health behavior, she or he must have an incentive to take the action, feel threatened by current behavioral patterns, and believe that the change will lead to valued outcomes at acceptable costs. The HBM was later extended with a self-efficacy factor by combining it with SCT.
Self-efficacy is determined as the extent to which one believes that one is able to perform a behavior that leads to a valued outcome. The HBM was initially developed to predict the intention to engage in simple health behavior such as one-time immunization or screening tests. Thus, changing lifelong habits such as eating, drinking, or exercising is a far more difficult process and requires the confidence that one is able to make the change before an intervention is possible [
Protection motivation theory is a widely used model for disease prevention and health promotion. Originally developed to explain the effects of fear appeals on health attitudes and behavior, the model is very similar to the HBM. It combines similar factors: severity, vulnerability, response cost, response efficacy, and self-efficacy. According to protection motivation theory, the intention to take preventive health action is formed as a result of two cognitive processes: (1) the individual will evaluate the possible threats (considering severity and vulnerability) of getting an illness and compare them with the intrinsic and extrinsic rewards of a certain negative behavior and (2) the individual will evaluate her or his ability to cope with the threat (response efficacy, self-efficacy, and response cost). As a result of the two processes, the protection motivation will be formulated, which again acts as a force to formulate the intention to take the action [
SCT has been successfully adapted to the study of the intention to engage in health protective behavior. According to SCT, six basic determinants influence health-related behavior: knowledge of the risks and benefits of health-related actions, perceived self-efficacy, outcome expectations about the costs and benefits of health-related habits, individual goals, and the perceived sociostructural barriers to and facilitators for an individual making the change and achieve her or his goals. Self-efficacy is the central part of SCT in that it influences behavior both directly and via other determinants. According to Bandura [
Due to the objective of this study to investigate MyData-based preventive eHealth service acceptance in a consumer context, UTAUT2 will be adapted as the basis for our research model. As most organizations do not yet provide data for individuals in a format that would be useful and practical from the viewpoint of health data analytics and new health services [
The proposed model for consumer acceptance of future MyData-based electronic health (eHealth) services.
To take account of health behavior factors, the three health behavior theories that are considered complement the UTAUT2 model with three health protection motivation constructs: self-efficacy, threat appraisals, and perceived barriers (see
Threat appraisals on the other hand play an important role in creating the motivation to take action to avoid a negative health outcome or to improve one’s poor health condition [
Finally, a wide variety of perceived barriers can have a significant negative influence on behavioral intention to use those systems. In the acceptance of new technologies, consumers have been found to have perceived psychological risks that prevent the adoption of new technologies [
In the context of preventive eHealth services, the use of the technology will provide benefits for an individual in preventing her or him from falling ill [
Effort expectancy has been found to have a significant effect, especially for the elderly consumers’ acceptance of preventive eHealth services [
Social influence can occur as information about the benefits of using preventive eHealth services from a health professional’s advice or through media education and also as encouragement from friends or relatives. Social influences can also provide a reminder or trigger for a motivated person to take action. Social influence is determined here as encouragement or reminders from important others or media channels to promote the use of preventive eHealth services [
The ease of access to the preventive eHealth services is an important factor, especially in the early stages of adoption, before the quality of the information has been determined [
Facilitating conditions have been found to have a significant positive effect on consumers’ behavioral intention to use mobile technologies [
Hedonic motivation is determined as the fun or pleasure that a consumer derives from using a technology [
Habit is determined as the extent to which an individual believes performing a behavior (eg, using mobile apps to track exercise) to be automatic as a result of learning during past behavior [
Self-efficacy is defined as one’s confidence in one’s ability to successfully perform a behavior that leads to a valued outcome. The definition overlaps with the perceived behavioral control factor in the TPB. Self-efficacy is an important factor because it influences an individual’s aspirations and goals in general. If one has high self-efficacy, one will set higher goals and will have higher expectations of achieving those goals [
Thus, a person with high self-efficacy will be likely to believe that using preventive eHealth services will generate better health outcomes compared with a person with low self-efficacy.
Following this, the following can be hypothesized:
Another significant aspect of self-efficacy is that it negatively influences cognitive barriers. If one has high efficacy, one will view obstacles as surmountable and will continue on the path to achieving one’s goals [
The stronger the threat appraisals are, the stronger the motivation for an individual to take part in healthy behavior [
Perceived barriers are defined in the HBM as potential negative aspects that would be expensive, dangerous, unpleasant, inconvenient, or time-consuming when taking a particular health action [
According to the HBM, barriers have a significant effect on an individual’s intention to take health protective actions [
The research model is composed of 16 constructs. The constructs were measured with multiple, reflective items on a 5-point Likert scale. Most of the measurement items were adapted from prior research to preserve content validity. The only exception is the construct
Data for the study was collected using a quantitative, Web-based questionnaire survey. The survey was a part of a large national research program, DHR, coordinated by the University of Oulu. The goal of this multidisciplinary research program is to enable the utilization of data about the individual as part of personal, preventive services, which in turn will improve citizens’ opportunities for self-management of their well-being. The aim of this study was to investigate what factors influence consumers’ intentions to use a MyData-based preventive eHealth service before use. The sample frame was the faculty and the staff of the University of Oulu, a total of 2852 people. The link to the survey was sent via email. The email included both the link to the Web-based survey in WebPropol survey service and a cover letter explaining the research phenomenon and research context, the purpose of the survey, and the use of the data, as well as encouragement to answer. The email was sent to the sample frame in March 2015, and the link for the survey was open for 1 week. We had a total of 855 respondents out of 2852 (29.98% response rate) who voluntarily chose to answer the survey. The demographic distribution of the respondent group is shown in
The gender distribution of the respondent group was female dominated with over two-thirds of the respondents being women. Almost 70% (579/855) of the respondents belonged to the two youngest age groups.
Empirical analyses were made using SEM with Mplus software, version 7.4. Estimations were made using maximum likelihood based on covariance matrix. First, we ran traditional SEM for the research model. That was not statistically significant, and thus, we tested possible heterogeneity in the data by running mixture analysis.
We found that heterogeneity of the data was not an issue in this case. Thus, we moved on toward model-generating SEM [
The demographic distribution of the respondents.
Characteristics | n (%) | |
Male | 305 (35.7) | |
Female | 550 (64.3) | |
Total | 855 (100) | |
18-25 | 352 (41.2) | |
26-35 | 227 (26.5) | |
36-45 | 119 (13.9) | |
46-55 | 107 (12.5) | |
56-65 | 48 (5.6) | |
66 and over | 2 (0.2) | |
Total | 855 (100) |
Construct reliabilities.
Infriska | Persimpb | Changec | Techanxd | Techrske | Severitf | Vulnerag | Techuseh | Healthbei | EEj | BIk | |
AVEl | 0.669 | 0.590 | 0.707 | 0.508 | 0.592 | 0.718 | 0.528 | 0.655 | 0.488 | 0.692 | 0.684 |
Squared AVE | 0.820 | 0.768 | 0.841 | 0.713 | 0.769 | 0.847 | 0.762 | 0.809 | 0.699 | 0.832 | 0.827 |
CRm | 0.852 | 0.733 | 0.853 | 0.679 | 0.736 | 0.944 | 0.637 | 0.838 | 0.582 | 0.833 | 0.885 |
aInformation risk.
bPersonal impediments.
cResistance to change.
dTechnology anxiety.
eTechnology risk.
fSeverity.
gVulnerability.
hTechnology use self-efficacy.
iHealthy behavior self-efficacy.
jEffort expectancy.
kBehavioral intention.
lAVE: average variance extracted.
mCR: composite reliability.
A model-generated structural equation model.
Structural equation model; beta and
Predicted construct and predictor constructs | Beta | ||
Effort expectancy | .191 | <.001 | |
Self-efficacy | .449 | <.001 | |
Threat appraisals | .416 | <.001 | |
Perceived barriers | −.212 | .009 | |
Self-efficacy | −.650 | <.001 |
The hypotheses and their judgments based on the SEM results are shown in
The tested hypotheses.
Hypothesis (H) | Description | Judgment |
H1 | Performance expectancy will influence behavioral intention positively. | Rejected |
H2 | Effort expectancy will influence behavioral intention positively. | Accepted |
H3 | Social influence will influence behavioral intention positively. | Rejected |
H4 | Facilitating conditions will influence behavioral intention positively. | Rejected |
H5 | Hedonic motivation will influence behavioral intention positively. | Rejected |
H6 | Habit will influence behavioral intention positively. | Rejected |
H7 | Self-efficacy will influence behavioral intention positively. | Accepted |
H8 | Self-efficacy will influence perceived barriers negatively. | Accepted |
H9 | Threat appraisals will influence behavioral intention positively. | Accepted |
H10 | Threat appraisals will influence perceived barriers negatively. | Rejected |
H11 | Perceived barriers will influence behavioral intention negatively. | Accepted |
Contrary to prior research applying UTAUT in an eHealth context [
Regarding the health protection motivation constructs, the results were the opposite as all three constructs—threat appraisals, self-efficacy, and perceived barriers—were found to have a significant influence on behavioral intention. In addition to the fact that our study focused on the intention to use future services instead of actual service use, there are other possible reasons for the poor performance of UTAUT2 constructs. Due to the focus on preventive eHealth services, the respondent group in our study had a proportion of healthy people who did not have a chronic illness. Our study also had younger participants compared with many other studies [
Our study contributes to the health information technology literature in two ways. First, we provide a research model that combines the standard UTAUT model with health protection motivation constructs, thus bridging the gap between technology adoption and health behavior theories. Second, our study focuses on increasing understanding of the factors influencing consumers’ eHealth technology acceptance instead of focusing on the acceptance of health care professionals.
Our research highlights the importance of the two health protection motivation constructs. Both threat appraisals and self-efficacy were found to be significant determinants of the intention to use future preventive eHealth services. Both technology use self-efficacy and healthy behavior self-efficacy were found to be significant components of the construct. This result suggests that to promote the use of new preventive eHealth services, emphasis must be placed on both technology use and healthy behavior education. Increased awareness of eHealth technology use and healthy behavior will also have a positive effect as it will diminish the effects of perceived barriers, which were found to have a significant negative impact on intention to use a service.
Typically, the importance of effort expectancy has been found in the studies on elderly people’s intention to use eHealth services [
There are also some limitations that must be considered when considering the generalizability of our results. The research context of future technologies and services that have not been developed yet to their fullest potential poses some limitations. First, the quantitative survey was based on a hypothetical use situation, and thus, the target group did not experience the actual use of the service. Second, the construct of
The above limitations lead us to the following suggestions for future research avenues. First, future research should test the acceptance of these preventive MyData-based eHealth services in an actual-use context and investigate how the direct factors perform in that context. Second, future research should also consider a wider coverage of consumers with a nonacademic background to see the relationships between the considered variables in a larger demographic setting. The importance of inclusion of different demographic segments in eHealth research and promotion activities was also stressed in the study by Kontos et al [
Our study contributes to the exploration of the factors influencing the consumers’ intention to use MyData-based preventive eHealth services before use. We combine the key factors of UTAUT2 and health behavior theories to our research model and apply the quantitative SEM approach to analyze the relationships between the constructs of the research model. Our results indicate that UTAUT2 constructs perform poorly. Only effort expectancy had a significant effect on the intention to use. Contrary to UTAUT2 constructs, all constructs adapted from health behavior theories—threat appraisals, self-efficacy and perceived barriers—were found to have a significant effect on the intention to use. These results suggest that to promote the adoption of preventive eHealth services among consumers, it is essential to invest in increasing the general awareness of healthy behavior and in the expertise of using eHealth technologies. From societal perspective, this implies increasing investment in health and technology-related education. From service perspective, higher probability of new preventive eHealth service adoption could be achieved, for example, by increasing consumer involvement in the creation of services through collaborative practices.
Summary table of eHealth acceptance studies.
Questionnaire items.
average variance extracted
comparative fit index
composite reliability
Digital Health Revolution
electronic health
health belief model
mobile health
root mean square error of approximation
social cognitive theory
structural equations modeling
technology acceptance model
Tucker-Lewis index
theory of planned behavior
Unified theory of acceptance and use of technology
The study was carried out as a part of the DHR project, funded by Tekes—the Finnish Funding Agency for Innovation.
None declared.