This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), 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.
For effective health promotion using health information technology (HIT), it is mandatory that health consumers have the behavioral intention to measure, store, and manage their own health data. Understanding health consumers’ intention and behavior is needed to develop and implement effective and efficient strategies.
To develop and verify the extended Technology Acceptance Model (TAM) in health care by describing health consumers’ behavioral intention of using HIT.
This study used a cross-sectional descriptive correlational design. We extended TAM by adding more antecedents and mediating variables to enhance the model’s explanatory power and to make it more applicable to health consumers’ behavioral intention. Additional antecedents and mediating variables were added to the hypothetical model, based on their theoretical relevance, from the Health Belief Model and theory of planned behavior, along with the TAM. We undertook structural equation analysis to examine the specific nature of the relationship involved in understanding consumers’ use of HIT. Study participants were 728 members recruited from three Internet health portals in Korea. Data were collected by a Web-based survey using a structured self-administered questionnaire.
The overall fitness indices for the model developed in this study indicated an acceptable fit of the model. All path coefficients were statistically significant. This study showed that perceived threat, perceived usefulness, and perceived ease of use significantly affected health consumers’ attitude and behavioral intention. Health consumers’ health status, health belief and concerns, subjective norm, HIT characteristics, and HIT self-efficacy had a strong indirect impact on attitude and behavioral intention through the mediators of perceived threat, perceived usefulness, and perceived ease of use.
An extended TAM in the HIT arena was found to be valid to describe health consumers’ behavioral intention. We categorized the concepts in the extended TAM into 3 domains: health zone, information zone, and technology zone.
Since the advent of the information age, the use of diverse health information technology (HIT) has become widespread in chronic disease management, disease prevention, and health promotion. HIT is “the application of information processing involving both computer hardware and software that deals with the storage, retrieval, sharing, and use of health care information, data, and knowledge for communication and decision making” [
For effective use of collected health-related data in HIT, it is crucial that health consumers have the behavioral intention to measure, store, and manage their own data. The effort put forth by health consumers to measure, store, and manage their own data strongly determines the quality of the data. Therefore, until data collection and storage is fully automated, consumers’ behavioral intention will be the predominant deciding factor in the accuracy and usefulness of such data. Additionally, a better understanding of health consumers’ intention and behavior would aid the development and implementation of effective and efficient strategies. Thus, identifying the factors influencing health consumers’ intention and behavior to measure, store, and manage their own health-related data would enable the development of a theoretical model to successfully describe their intentions and actions. Developing a model requires determining the interrelationships among the factors of health information by integrating various behavior and information technology theories.
First, regarding health behavior theories, health behavior includes any activity undertaken by an individual, regardless of actual or perceived health status, for the purpose of promoting, protecting, or maintaining health, whether or not such behavior is objectively effective in obtaining the intended results [
Next, regarding the theories of information technology, the Technology Acceptance Model (TAM) is the most widely applied model to describe consumer acceptability of information technology. TAM can be adapted by applying various factors involving consumers’ behavior in the context of HIT [
The Health Belief Model (HBM) was one of the first and one of the best-known social cognition models to explain health behavior change [
Several key concepts, such as attitude, behavioral intention, and behavior, from these theories are the same and the paths are very similar; thus, it is possible to synthesize a model by extracting the common key concepts and building the logical paths between them. In 2008, Yun [
The objective of this study was to develop and test a model describing the behavioral intention and the health behavior of consumers of various HITs, including the Internet, smartphones, and social network services. This model, the HIT-driven extended TAM, is called the Health Information Technology Acceptance Model (HITAM).
We recruited study participants from the three largest online health portals in South Korea. They are KorMedi.com, HiDOC, and Kunkang-In (which means Healthy People, from the National Health Insurance Corporation, Republic of Korea). Members of these popular online health portals who used health information through the Internet, smartphones, or social network services were contacted for participation in the study.
Data were collected using an online survey method developed by a private company specializing in online survey research. A survey was posted on these portals from October 21 to December 8, 2011, and in total 728 members replied.
We sought and obtained approval from the institutional review board of the College of Nursing, Seoul National University before collecting the data.
For this study, we developed a structured, self-administered questionnaire titled “We would like to know whether you use Smart HIT in your health management.” The questionnaire was composed of 50 items, measuring general characteristics of the research participants and variables in 10 categories. General characteristics were measured with 6 items, health belief and concerns with 5 items, subjective norm with 5 items, perceived susceptibility with 3 items, perceived seriousness with 4 items, HIT self-efficacy with 6 items, HIT reliability with 5 items, perceived ease of use with 5 items, perceived usefulness with 5 items, attitude with 3 items, and behavioral intention with 3 items. Health status was measured by asking if they had any diseases or comorbidity. (Refer to
Attitude and behavioral intention are two outcome categories. Attitude measures the positive perception of and satisfaction with the use of HIT. Behavioral intention measures the intent and willingness to use HIT. Participants were asked to rate their agreement with the following 3 statements to measure attitude: (1) I am positive about using HIT to manage my health and to search for reliable health information, (2) I think it is beneficial to manage my health and search for reliable health information using HIT, and (3) I am satisfied by and large with the use of HIT to manage my health and search for reliable health information using HIT. The following 3 items measured behavioral intention: (1) I will continue to use HIT to manage my health and to search for reliable health information, (2) I will regularly use HIT to manage my health and to search for reliable health information, and (3) I will recommend use of HIT to other people to manage their health and to search for reliable health information.
We derived items measuring perceived susceptibility, perceived seriousness, perceived threat, and behavioral intention from HBM. Items measuring HIT self-efficacy, HIT reliability, perceived ease of use, perceived usefulness, attitude, and behavioral intention were derived from TAM3. Items measuring health belief and concerns, subjective norm, attitude, and behavioral intention were derived from TPB. We adapted categories of the questionnaire from Yun’s study on the development of a consumer health information-seeking behavior model from 2008 [
Comparison of variables in Yun [
Variables in Yun’s model | Variables in proposed model | Reasons for inclusion, exclusion, or modification |
Health concerns | Health belief and concerns | |
Perceived susceptibility | Perceived susceptibility | |
Perceived seriousness | Perceived seriousness | |
Subjective health-related knowledge | Excluded because it is for the knowledge level only, which is too specific | |
Subjective norm | Added to strengthen the normative beliefs such as the motivation to comply in the theory of planned behavior | |
Internet self-efficacy | HITa self-efficacy | Modified to adapt to HIT because it is the broader term |
Perceived ease of use | Perceived ease of use | |
Perceived usefulness | Perceived usefulness | |
Perceived credibility | HIT reliability | Perceived credibility in Yun’s model is the same concept as the current model’s HIT reliability |
Attitude | Attitude | |
Behavioral intention | Behavioral intention |
a Health information technology.
The measurement tool was a 7-point Likert-type scale ranging from 1 for strongly disagree to 7 for strongly agree. The questionnaire was originally created in Korean, so it was not necessary to translate it. The reliability of the original instrument was indicated by Cronbach alpha = .853. The content of the questionnaire was independently confirmed by a group of HIT experts.
To describe health consumers’ behavioral intention toward HIT services that use computers, the Internet, and smartphones, we developed a structural equation model based on previous research and a literature review. We also analyzed other theories relevant to HITAM synthesis, such as HBM, TPB, TAM, Extended TAM (TAM2, unified theory of acceptance and use of technology, TAM3), and Consumer’s Health Information Seeking Behavior Model.
Extensive monitoring of behavioral change requires a long-term observational study and prohibitively large sample groups. Also, self-reported behavioral intention and users’ action itself are hard to measure. Finally, cross-sectional research methodology provides a challenge, as the correlation between behavioral intention and the actual taking of action has been shown to be weak [
Our structural equation model relied on findings from previous research regarding consumers’ behavioral intention to use HIT.
Conceptual relationships of the relevant models for development of the Health Information Technology Acceptance Model (HITAM). HBM = Health Belief Model, TAM = Technology Acceptance Model, TPB = theory of planned behavior.
Path diagram of the Health Information Technology Acceptance Model (HITAM) Model. HIT = health information technology.
Descriptive statistics of the participants’ general characteristics were analyzed with IBM SPSS version 19 (IBM Corporation, Somers, NY, USA). The structural equation model was fitted with maximum likelihood estimation routines in IBM SPSS Amos 16.
The general characteristics of the 728 research participants are shown in
General characteristics of the participants (N = 728).
Characteristic | n | % | |
|
|||
Male | 372 | 51.1% | |
Female | 356 | 48.9% | |
|
|||
<19 | 10 | 1.4% | |
20–29 | 144 | 19.8% | |
30–39 | 258 | 35.4% | |
40–49 | 150 | 20.6% | |
50–59 | 101 | 13.9% | |
>60 | 65 | 8.9% | |
|
|||
Clerical | 223 | 30.6% | |
Professional | 113 | 15.5% | |
Homemaker | 86 | 11.8% | |
Student | 75 | 10.3% | |
Self-employed | 58 | 8.0% | |
Manufacturing | 33 | 4.5% | |
Government official | 24 | 3.3% | |
Other | 116 | 15.9% | |
|
|||
< Middle school | 10 | 1.4% | |
High school | 177 | 24.3% | |
College | 135 | 18.5% | |
University | 299 | 41.1% | |
> Graduate school | 107 | 14.7% | |
|
|||
Yes | 361 | 49.6% | |
No | 367 | 50.4% | |
|
|||
>1000 | 37 | (5.1) | |
1001–2000 | 107 | (14.7) | |
2001–3000 | 186 | (25.5) | |
3001–4000 | 159 | (21.8) | |
4001–5000 | 118 | (16.2) | |
<5000 | 121 | (16.6) |
The descriptive statistics of the variable scores and the reliability coefficients of the measuring tool in each category are provided in
Descriptive statistics of the latent variables and the reliability coefficients (N = 728 for all variables).
Variable | Minimum | Maximum | Mean | SD | Kurtosis | Skewness | Cronbach |
No. of |
||
Health belief and concerns | 5 | 35 | 27.04 | 4.930 | –1.028 | 0.091 | 2.405 | 0.181 | 0.867 | 5 |
Subjective norm | 5 | 35 | 23.68 | 4.978 | –0.318 | 0.091 | 0.289 | 0.181 | 0.826 | 5 |
Perceived susceptibility | 3 | 21 | 12.94 | 4.157 | –0.291 | 0.091 | –0.448 | 0.181 | 0.751 | 3 |
Perceived seriousness | 4 | 28 | 20.91 | 4.801 | –0.844 | 0.091 | 0.811 | 0.181 | 0.907 | 4 |
HITa self-efficacy | 6 | 42 | 31.25 | 5.841 | –0.412 | 0.091 | 0.693 | 0.181 | 0.888 | 6 |
HIT reliability | 5 | 35 | 25.55 | 4.795 | –0.485 | 0.091 | 0.521 | 0.181 | 0.934 | 5 |
Perceived ease of use | 5 | 35 | 27.74 | 4.847 | –0.716 | 0.091 | 0.734 | 0.181 | 0.925 | 5 |
Perceived usefulness | 5 | 35 | 25.99 | 4.710 | –0.466 | 0.091 | 0.364 | 0.181 | 0.826 | 5 |
Attitude | 3 | 21 | 16.76 | 3.024 | –0.872 | 0.091 | 1.088 | 0.181 | 0.93 | 3 |
Behavioral intention | 3 | 21 | 17.07 | 3.146 | –0.921 | 0.091 | 0.956 | 0.181 | 0.919 | 3 |
a Health information technology.
We found that each measured variable satisfied the assumption of the univariate normality.
Correlation coefficients between measured variables.
Age | Diseases | Health |
Subjective |
Perceived |
Perceived |
HITa
|
HIT |
Perceived |
Perceived |
Attitude | Behavioral |
|
Age | 1 | |||||||||||
Diseases | –.117** | 1 | ||||||||||
Health belief and concerns | .144** | –.023 | 1 | |||||||||
Subjective norm | .049 | .025 | .644** | 1 | ||||||||
Perceived susceptibility | .032 | –.411** | .119** | .151** | 1 | |||||||
Perceived seriousness | .057 | –.156** | .340** | .207** | .447** | 1 | ||||||
HIT self-efficacy | –.053 | .056 | .517** | .529** | .155** | .288** | 1 | |||||
HIT reliability | .048 | .031 | .449** | .405** | .143** | .316** | .574** | 1 | ||||
Perceived ease of use | .002 | .024 | .488** | .403** | .139** | .350** | .733** | .655** | 1 | |||
Perceived usefulness | –.009 | .026 | .476** | .479** | .175** | .370** | .664** | .672** | .740** | 1 | ||
Attitude | .097** | .001 | .520** | .387** | .126** | .358** | .640** | .694** | .740** | .737** | 1 | |
Behavioral intention | .145** | .014 | .496** | .361** | .110** | .350** | .635** | .604** | .705** | .686** | .817** | 1 |
a Health information technology.
*
Standardized estimates of the Health Information Technology Acceptance Model (HITAM).
Endogenous variable | Exogenous variable | Regression |
Standardized |
CRa
|
|
SMCb |
Perceived threat | Health status | 1.167 (0.379) | .413 | 3.078 | .002 | .184 |
Health belief and concerns | 0.100 (0.029) | .117 | 3.415 | <.001 | ||
Perceived ease of use | HITc reliability | 0.372 (0.030) | .367 | 12.305 | <.001 | .665 |
HIT self-efficacy | 0.446 (0.024) | .542 | 18.275 | <.001 | ||
Perceived usefulness | Subjective norm | 0.118 (0.026) | .126 | 4.475 | <.001 | .677 |
HIT reliability | 0.276 (0.033) | .281 | 8.381 | <.001 | ||
Perceived ease of use | 0.409 (0.041) | .422 | 9.996 | <.001 | ||
Perceived threat | 0.045 (0.025) | .040 | 1.777 | .08 | ||
HIT self-efficacy | 0.102 (0.032) | .127 | 3.218 | .001 | ||
Attitude | Perceived usefulness | 0.268 (0.024) | .432 | 11.363 | <.001 | .734 |
Perceived ease of use | 0.288 (0.023) | .479 | 12.588 | <.001 | ||
Behavioral intention | Attitude | 0.955 (0.025) | .912 | 38.669 | <.001 | .831 |
a Critical ratio.
b Squared multiple correlation.
c Health information technology.
Effects of exogenous variables on endogenous variables in the Health Information Technology Acceptance Model (HITAM).
Endogenous variable | Exogenous variable | Standardized |
|
Standardized |
|
Standardized |
|
Perceived threat | Health status | .413 | .01 | 0 | .413 | .01 | |
Health belief and concerns | .117 | .006 | 0 | .117 | .006 | ||
Perceived ease of use | HITa self-efficacy | .542 | .007 | 0 | .542 | .007 | |
HIT reliability | .367 | .01 | 0 | .367 | .01 | ||
Perceived usefulness | HIT self-efficacy | .127 | .009 | .228 | .02 | .356 | .009 |
HIT reliability | .281 | .009 | .155 | .009 | .436 | .008 | |
Subjective norm | .126 | .01 | 0 | .126 | .01 | ||
Perceived threat | .040 | .06 | 0 | .040 | .06 | ||
Perceive ease of use | .422 | .008 | 0 | .422 | .008 | ||
Health belief and concerns | .005 | .04 | .005 | .04 | |||
Health status | .017 | .05 | .017 | .05 | |||
Attitude | Perceived ease of use | .479 | .02 | .182 | .005 | .661 | .02 |
Perceived usefulness | .432 | .006 | 0 | .432 | .006 | ||
HIT reliability | .364 | .03 | .364 | .03 | |||
HIT self-efficacy | .413 | .01 | .413 | .01 | |||
Subjective norm | .054 | .006 | .054 | .006 | |||
Health belief and concerns | .002 | .03 | .002 | .03 | |||
Health status | .007 | .046 | .007 | .046 | |||
Perceived threat | .017 | .05 | .017 | .05 | |||
Behavioral intention | Attitude | .912 | .01 | 0 | .912 | .01 | |
HIT reliability | .332 | .03 | .332 | .03 | |||
HIT self-efficacy | .377 | .01 | .377 | .01 | |||
Subjective norm | .049 | .006 | .049 | .006 | |||
Health belief and concerns | .002 | .04 | .002 | .04 | |||
Health status | .007 | .04 | .007 | .04 | |||
Perceived threat | .016 | .05 | .016 | .05 | |||
Perceived ease of use | .603 | .01 | .603 | .01 | |||
Perceived usefulness | .394 | .005 | .394 | .005 |
a Health information technology.
We categorized the influential factors affecting the behavioral intention to measure, store, and manage health-related data into three domains called the health zone, information zone, and technology zone. In each zone, the factors follow different mediating processes that lead to behavioral intention in health customers. In the first domain, the health zone, there is a cascade of effects starting from health status (age, disease, etc), to perceived threat, to perceived usefulness, to attitude and, finally, to behavioral intention. This further verifies the description of behavioral intention in the HBM, where an intermediate variable, attitude, leads to behavioral intention [
Consumers’ perceived threat, which is measured by perceived susceptibility and perceived seriousness, had a somewhat smaller impact on perceived usefulness than in previous studies. However, perceived threat had a statistically significant indirect effect on attitude through perceived usefulness, subsequently increasing the behavioral intention to use HIT. This indirect relationship is exemplified by the tendency of individuals to actively use HIT when they perceive a potential threat to their health. This result is consistent with the model reported by Yun [
In the second domain, the information zone, the factors have a similar cascade effect to that in the health zone. Perceived usefulness is significantly sensitive to subjective norms, such as social pressure or community competition, resulting in consumers forming positive attitude. Such a formed attitude has a consequence on behavioral intention. According to the Ajzen’ TPB [
The last domain, the technology zone, has factors with the following characteristics. HIT use forms specific HIT reliability, such as output quality and result demonstrability, and these affect perceived usefulness. More interestingly, HIT reliability also affects perceived ease of use, also affecting perceived usefulness. Similarly, HIT self-efficacy, such as HIT anxiety, playfulness, perceived enjoyment, and objective usability, also affect perceived usefulness and perceived ease of use. These two factors affect attitude and, finally, form behavioral intention. Our study revealed that HIT self-efficacy is a significant factor influencing HIT use. This suggests that consumers enjoying the use of HIT and gaining confidence in their ability to use HIT develop increased tendencies to use HIT. Specifically, the greater the self-efficacy, the greater the perceived ease of use, which is consistent with the finding reported by Yun [
Ajzen developed the TPB model that explains various human actions by integrating behavioral belief, normative belief, efficacy belief, attitude, subjective norm, and perceived behavioral control [
The key factors identified within the three zones—health status and health belief and concerns in the health zone; subjective norms and HIT reliability in the information zone; and HIT self-efficacy in the technology zone—are the predicting factors that form the HITAM with varying ranges of significance and directional relationships. By identifying the core factors that have the largest impact, HITAM is a succinct and powerful model that reevaluates and reorganizes previous findings in the field.
The survey revealed that, although TAM has expanded its utility into various areas and has been successfully implemented, its implementation in the HIT arena has been limited. Furthermore, even in the rapid advancement of information technology and its subsequent impact on health management, a model that captures and predicts various aspects of consumer acceptance is lacking. To address this gap, our research has tried to provide a robust foundation for future HIT research. Yun’s similar effort in integrating health information-seeking activities with TAM was limited to the use of the Internet and did not capture the real-life application of HIT. Therefore, by considering the latest developments in HIT, now used more ubiquitously, the model in this work possesses increased real-life applicability and predictive capability.
Using this model, many aspects of health behaviors using HIT can be explained. Thus, in the wake of the exponentially increasing presence of HIT such as the Internet and smartphone apps, the HITAM provides a valuable model of how different interactions with HIT form behavioral intention in health consumers.
The Health Information Technology Acceptance Model (HITAM). HIT = health information technology.
Health Belief Model
health information technology
Health Information Technology Acceptance Model
Technology Acceptance Model
theory of planned behavior
This work was supported by a National Research Foundation of Korea (NRF) grant, funded by the Korea government (MEST) (No.2012-0000998).
None declared.