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.
Today, people use the Internet to satisfy health-related information and communication needs. In Malaysia, Internet use for health management has become increasingly significant due to the increase in the incidence of chronic diseases, in particular among urban women and their desire to stay healthy. Past studies adopted the Technology Acceptance Model (TAM) and Health Belief Model (HBM) independently to explain Internet use for health-related purposes. Although both the TAM and HBM have their own merits, independently they lack the ability to explain the cognition and the related mechanism in which individuals use the Internet for health purposes.
This study aimed to examine the influence of perceived health risk and health consciousness on health-related Internet use based on the HBM. Drawing on the TAM, it also tested the mediating effects of perceived usefulness of the Internet for health information and attitude toward Internet use for health purposes for the relationship between health-related factors, namely perceived health risk and health consciousness on health-related Internet use.
Data obtained for the current study were collected using purposive sampling; the sample consisted of women in Malaysia who had Internet access. The partial least squares structural equation modeling method was used to test the research hypotheses developed.
Perceived health risk (β=.135,
The integrated model proposed and tested in this study shows that the HBM, when combined with the TAM, is able to predict Internet use for health purposes. For women who subjectively evaluate their health as vulnerable to diseases and are concerned about their health, cognition beliefs in and positive affective feelings about the Internet come into play in determining the use of health-related Internet use. Furthermore, this study shows that engaging in health-related Internet use is a proactive behavior rather than a reactive behavior, suggesting that TAM dimensions have a significant mediating role in Internet health management.
Millions of people throughout the world use the Internet and much of this activity is focused on health [
The Internet not only functions as a rich source of health information, but it also provides interactivity between professionals and health seekers through an electronic or communication tool to gain and convey health information [
Such importance placed on the Internet as a health-seeking platform helps people maintain, promote, and manage their health. Past research shows that women are more likely to use the Internet for health-related purposes than men [
Although an abundance of research can be found on Internet health care information-seeking behavior, a major focus of these studies tends to concentrate on understanding the use of the Internet for health information-seeking behavior based on the Health Belief Model (HBM). The HBM was initially developed to predict the behavioral reaction of individuals with acute or chronic diseases to the treatment they receive [
Individuals with higher perceived health risk have greater motivation to change or adopt a health-oriented behavior, including adopting a preventive health behavior such as seeking information and using information and communication channels (eg, the Internet) to satisfy health-related information and communication needs [
Results of past studies found that women tend to have a higher perceived health risk than men [
As well as perceived health risk, health consciousness is another dimension that influences health-seeking behavior.
Health consciousness is a predictor of the use of communication channels for health information seeking [
Hypothesized model based on the Health Belief Model.
Other studies that contribute toward the extant literature include those that are based on the Technology Acceptance Model (TAM) [
The TAM was developed to enable understanding of the use of technology [
Using the TAM framework, studies showed that perceived usefulness, perceived ease of use and attitude, positively influence behavioral intention to use health information technologies such as the Internet and mobile phones [
Hypothesized model based on Technology Acceptance Model.
Although many past studies on Internet use for health-related purposes adopted the TAM or HBM, the use of these theories independently has not been able to explain fully Internet health-seeking behavior. The TAM has been used to predict an individual’s technology use; however, it is an inadequate model for health-related Web use because of its heavy dependence on 2 factors: perceived usefulness and perceived ease of use of technology [
By incorporating constructs of technology acceptance based on the TAM and perceived health risk and health consciousness as explained by the HBM, an integrated model of health-related Internet use behavior is proposed whereby perceived usefulness of the Internet and attitude toward the Internet for health purposes mediate the relationship between perceived health risks as well as health consciousness and health-related Internet use behavior (
This study aimed to examine the influence of perceived health risk and health consciousness on health-related Internet use based on the HBM. The model developed for the purpose of this study incorporated the TAM to provide a better understanding of the process that affects the adoption of Internet use for health purposes. Based on the integrated model, this study set out to test the mediating effect of TAM constructs, perceived usefulness of the Internet, and attitude toward Internet use on the relationship between perceived health risk and health consciousness on Internet use for health purposes.
Research hypotheses for explaining health-related Internet use drawing upon the Health Belief Model and the Technology Acceptance Model.
Research hypotheses | Path (causal effect) | Sources |
H1: Perceived health risk toward chronic diseases consisted of perceived susceptibility to chronic diseases and perceived severity of chronic diseases has a positive effect on health-related Internet use | Perceived health risk → health-related Internet use | [ |
H2: Health consciousness has a positive effect on health-related Internet use | Health consciousness → health-related Internet use | [ |
H3: The effect of perceived health risk, consisted of perceived susceptibility to chronic diseases and perceived severity of chronic diseases, on health-related Internet use is mediated by perceived usefulness of the Internet, and attitude toward Internet use for health information and health management | Perceived health risk → perceived usefulness of the Internet → attitude toward Internet use → health-related Internet use | [ |
H4: The influence of health consciousness on health-related Internet use is mediated by perceived usefulness of the Internet, and attitude toward Internet use for health information and health management | Health consciousness → perceived usefulness of the Internet → attitude toward Internet use → health-related Internet use | [ |
Integrated model based on the Health Belief Model and the Technology Acceptance Model.
The participants in this study consisted of Malaysian females living in the state of Selangor, the most urbanized state in Malaysia. Purposive sampling was used. Women who were Internet users were selected as the sample for the purpose of this study because past research found that they tend to be educated, married, and live in urban areas [
Out of 380 questionnaires distributed, 330 completed questionnaires were obtained. From the 330 sets of questionnaires returned, 293 responses were usable after excluding cases that had not used the Internet for health-related purposes and cases with incomplete information.
As shown in
Descriptive statistics of demographic characteristics of participants (N=293).
Characteristics | n (%) | |
|
|
|
|
20-29 | 110 (37.5) |
|
30-39 | 127 (43.0) |
|
40-49 | 43 (15.0) |
|
≥50 | 13 (4.5) |
|
|
|
|
Single | 92 (31.5) |
|
Married | 195 (66.5) |
|
Others | 6 (2.0) |
|
|
|
|
Primary school | 18 (6.0) |
|
Secondary school | 138 (47.5) |
|
College/university | 135 (46.5) |
|
|
|
|
1000-2999 | 114 (39.5) |
|
3000-5999 | 133 (46.2) |
|
6000-8999 | 39 (13.5) |
|
≥9000 | 2 (0.7) |
Perceived health risk contains 2 subdimensions: perceived susceptibility to chronic diseases and perceived severity of chronic diseases. Perceived susceptibility to chronic diseases was measured by 6 items adopted from Kim and Park [
Participants’ health consciousness was measured by 11 items covering most facets of health consciousness adopted from Chen [
Items that measured perceived usefulness of the Internet for health information and health management were adopted from Davis [
Perceived ease of Internet use was assessed by the 4 items developed by Davis [
Four items on attitudes toward Internet use for health information were adopted from the study by Wong et al [
Health-related Internet use had 2 subdimensions: Internet for seeking health and medical information and Internet use to communicate health-related issues. Internet use for health information seeking was measured by 11 items and Internet usage for communication on health-related issues was measured by 5 items adopted from past studies [
We used the partial least squares structural equation modeling (PLS-SEM) method and SmartPLS software 2.0 [
There are 3 different approaches to estimate parameters in models with second-order constructs: (1) the repeated indicator approach, (2) the 2-stage approach, and (3) the hybrid approach [
In order to discover the structure of reflective latent variables and to identify the underlying variance structure of a set of indicators, this research used exploratory factor analysis (EFA) [
The Kaiser-Meyer-Olkin measure of sampling adequacy (0.816) and Bartlett’s test of sphericity results (
Reflective constructs assessment.
Construct/measure | Factor loadinga | Construct reliability | Average variance extracted | Maximum shared squared variance | Average shared square variance | |
|
|
0.916 | 0.646 | 0.092 | 0.042 | |
|
I have a higher likelihood of getting chronic diseases | 0.873 |
|
|
|
|
|
There is a great chance that I will be exposed to a chronic disease | 0.808 |
|
|
|
|
|
I would say that I am the type of person who is likely to get chronic diseases | 0.891 |
|
|
|
|
|
There is a person with chronic disease among my family members | 0.759 |
|
|
|
|
|
I have a strong possibility of attack or deterioration of chronic disease due to improper daily habits (drinking, smoking, dietary habit, lack of exercise, etc) | 0.707 |
|
|
|
|
|
It is most likely that I will catch chronic diseases in my lifetime | 0.771 |
|
|
|
|
|
|
0.900 | 0.694 | 0.022 | 0.011 | |
|
I am afraid of facing attack or deterioration of chronic diseases | 0.756 |
|
|
|
|
|
If I face attack or deterioration of chronic disease, I will have difficulty with my work life (or domestic affairs) | 0.807 |
|
|
|
|
|
If I face attack or deterioration of chronic disease, it will hinder my personal relationships | 0.896 |
|
|
|
|
|
If I face attack or deterioration of chronic disease, I will be long haunted by resultant problems | 0.865 |
|
|
|
|
|
|
0.925 | 0.608 | 0.228 | 0.140 | |
|
I have the impression that I sacrifice a lot for my health | 0.791 |
|
|
|
|
|
I consider myself very health conscious | 0.837 |
|
|
|
|
|
I think that I take health into account a lot in my life | 0.876 |
|
|
|
|
|
I think it is important to know well how to stay healthy | 0.883 |
|
|
|
|
|
My health is so valuable to me that I am prepared to sacrifice many things for it | 0.766 |
|
|
|
|
|
I have the impression that other people pay more attention to their health than I do | 0.767 |
|
|
|
|
|
I do not continually ask myself whether something is good for me | 0.665 |
|
|
|
|
|
I often dwell on my health | 0.610 |
|
|
|
|
|
|
0.905 | 0.760 | 0.336 | 0.221 | |
|
My interaction with the Internet for health information is clear and understandable | 0.857 |
|
|
|
|
|
I find the Internet for health information to be flexible to interact with | 0.880 |
|
|
|
|
|
It is easy for me to become skillful at using the Internet for health information | 0.878 |
|
|
|
|
|
|
0.928 | 0.811 | 0.344 | 0.218 | |
|
Using the Internet is useful in managing my daily health | 0.873 |
|
|
|
|
|
Using the Internet for health information is advantageous in better managing my health | 0.937 |
|
|
|
|
|
Using the Internet for health information is beneficial to me | 0.890 |
|
|
|
|
|
|
0.933 | 0.777 | 0.344 | 0.303 | |
|
Using the Internet for health information and health management would be a good idea | 0.894 |
|
|
|
|
|
Using the Internet for health information and health management would be a wise idea | 0.872 |
|
|
|
|
|
I like the idea of using the Internet for health information and health management | 0.895 |
|
|
|
|
|
Using the Internet for health information and health management would be a pleasant experience | 0.865 |
|
|
|
|
aThe total variance explained by factors=63.713%. All factor loadings were more than 0.5 and significant (
Subsequently, we assessed the construct reliability, convergent validity, and discriminant validity of reflective constructs [
To establish discriminant validity, both MSV and ASV should be less than the value of AVE. As shown in
In contrast to reflective constructs, indicators of formative constructs are not interchangeable and they do not necessarily have high intercorrelation [
In order to assess formative constructs, the collinearity issue was examined by computing correlation and the variance inflation factor (VIF).
In the second stage of the 2-stage method, latent variable scores of perceived susceptibility to chronic disease and perceived severity of chronic disease as well as latent variable scores of Internet usage for health information seeking and Internet usage for communication were estimated and used to evaluate the formative second level of perceived health risks and health-related Internet use, respectively. The VIF of indicators of health-related Internet use and PHR was less than 5, which indicates an absence of collinearity issue. Moreover, the significant factor weights of perceived susceptibility to chronic disease, perceived severity of chronic disease, Internet usage for health information seeking, and Internet usage for communication show that they make a significant contribution to perceived health risks and health-related Internet use.
Formative constructs assessment.
Construct/measure | Indicator weight |
|
Indicator outer loading | Interitem correlation, mean (range) | Variance inflation factor, maximum | |
|
|
0.536 (0.312-0.774) | 30.665 | |||
|
I use the Internet to get general health information | 0.161 | 1.501 | 0.594 |
|
|
|
I use the Internet to get information on medicine/drugs | 0.450 | 3.945 | 0.836 |
|
|
|
I use the Internet to be equipped with information before/after doctor’s appointment | –0.348 | 1.877 | 0.595 |
|
|
|
I use the Internet to get descriptions of various diseases | 0.115 | 0.793 | 0.717 |
|
|
|
I use the Internet to get information on treatments/therapy/diagnosis | 0.121 | 0.883 | 0.708 |
|
|
|
I use the Internet to get information on how to care for oneself | –0.201 | 1.468 | 0.567 |
|
|
|
I use the Internet to decide about how to treat an illness | 0.444 | 3.011 | 0.803 |
|
|
|
I use the Internet to decide about whether or not to visit a doctor | 0.097 | 0.735 | 0.735 |
|
|
|
I use the Internet to understand how to deal with an illness | 0.111 | 0.610 | 0.643 |
|
|
|
I use the Internet to get information on hospitals/clinics/other health care facilities | 0.257 | 2.112 | 0.717 |
|
|
|
I use the Internet to get information on health management (exercise, abstinence from drinking, smoking, diet, nutrition, stress, mental health, etc) | –0.002 | 0.015 | 0.505 |
|
|
|
|
|
|
0.572 (0.441-0.685) | 20.779 | |
|
I use the Internet to get online medical consultation from medical professionals | 0.601 | 3.433 | 0.931 |
|
|
|
I use the Internet to interact with people with similar health conditions | 0.280 | 1.462 | 0.833 |
|
|
|
I use the Internet to use mail to communicate with a doctor or a doctor’s office | –0.021 | 0.129 | 0.655 |
|
|
|
I use the Internet to share and exchange experiences about health and diseases | 0.289 | 1.312 | 0.765 |
|
|
|
|
|
|
0.595 | 10.549 | |
|
Internet usage for medical and health information seeking | 0.853 | 10.766 | 0.984 |
|
|
|
Internet usage to communicate for health | 0.221 | 2.021 | 0.728 |
|
|
|
|
|
|
0.005 | 10.000 | |
|
Perceived susceptibility to chronic diseases | 0.946 | 14.430 | 0.948 |
|
|
|
Perceived severity of chronic diseases | 0.319 | 1.967 | 0.324 |
|
|
Next, in testing the hypotheses developed for this study, a bootstrapping resampling method with 2000 replications was performed [
Standardized path coefficients,
Direct, indirect, and total effects.a
Path |
|
Q2 | Standardized path coefficient, |
|
||
|
|
|
|
|
||
|
|
.2395 | .1531 |
|
|
|
|
|
Perceived health risk (c1) |
|
|
.135*(.036, 234) | 2.676 |
|
|
Health consciousness (c2) |
|
|
.447***(.351, .542) | 9.168 |
|
|
|
|
|
||
|
|
.1821 | .1460 |
|
|
|
|
|
Perceived health risk (a11) |
|
|
.309***(.216, .401) | 6.538 |
|
|
Health consciousness (a21) |
|
|
.269***(.165, .373) | 5.063 |
|
|
.5284 | .4074 |
|
|
|
|
|
Perceived usefulness of the Internet (d) |
|
|
.334***(.224, .443) | 5.955 |
|
|
Perceived health risk (a12) |
|
|
.063 (–.034, .160) | 1.278 |
|
|
Health consciousness (a22) |
|
|
.270***(.167, .374) | 5.118 |
|
|
Perceived ease of Internet use (e) |
|
|
.322***(.215, .429) | 5.910 |
|
|
.3827 | .2767 |
|
|
|
|
|
Attitude toward Internet use (b1) |
|
|
.284***(.175, .392) | 5.123 |
|
|
Perceived usefulness of the Internet (b2) |
|
|
.266**(.155, .377) | 4.681 |
|
|
Perceived health risk (c’1) |
|
|
.019 (–.079, .117) | .383 |
|
|
Health consciousness (c’2) |
|
|
.211***(.107, .316) | 3.958 |
|
|
|
|
|
||
|
|
.3827 | .2767 |
|
|
|
|
|
Attitude toward Internet use, perceived usefulness of the Internet, perceived health risk (a11.d.b1) |
|
|
.029**(.013, .045) | 3.609 |
|
|
Attitude toward Internet use, perceived usefulness of the Internet, health consciousness (a21.d.b1) |
|
|
.025*(.010, .041) | 3.234 |
aArrows show the influence direction in the hypotheses. For example, perceived health risk influences (→) health-related Internet use.
*
In testing hypotheses 1 and 2 on the effect of perceived health risk to chronic disease and health consciousness on health-related Internet use, the results show support for these 2 hypotheses as perceived health risk (β=.135,
Hypothesis 3 was developed to test the mediation role of perceived usefulness of the Internet and attitude in the relationship between perceived health risk and Internet use for health information seeking. Results showed that 8 of 10 direct effects described in the structural mediated effect model in
The indirect effect of perceived health risk on health-related Internet use through perceived usefulness of the Internet and attitude toward Internet use was significant at the 95% confidence level (β=.029,
For hypothesis 4, the results showed that the indirect effect of health consciousness on health-related Internet use through perceived usefulness of the Internet and attitude toward Internet use was significant at the 95% confidence level (β=.025,
The results showed support for all the hypotheses developed in the study. Further, the model explained 38.27% of the variance in health-related Internet use. To assess the predictive accuracy of endogenous variables, we used Stone-Geisser’s Q2 [
Structural research model.
Path coefficients of the structural research model.
This study showed that there is a positive influence of perceived health risk and health consciousness on health-related Internet use, supporting hypotheses 1 and 2. It was also found that the effect of perceived health risk on health-related Internet use is fully mediated by perceived usefulness of the Internet and attitude toward Internet use for health information and health management as hypothesized in hypothesis 3. The study also supported that perceived usefulness of the Internet and attitude toward Internet use for health information and health management partially mediates the influence of health consciousness on health-related Internet use as proposed in hypothesis 4.
This study showed that perceived health risk positively affects health-related Internet use, confirming that perceived health risk is significant in influencing women’s Internet use for health-related purposes. This finding is consistent with Dillard et al’s study [
The results of this study also showed that health consciousness has a significant positive effect on health-related Internet use, supporting the relevance of the HBM, which asserts that health consciousness contributes to health behavior adoption [
The findings show that perceived usefulness of the Internet for health management and attitude toward Internet use for health-related purposes become central to women who perceive their health to be at risk and have the consciousness to seek information on health and health-related issues to manage their health and to stay healthy. Therefore, Internet use for health-related purposes is a process with perceived health risk and health consciousness as antecedents, but for this psychological orientation to translate into health-related Internet use behavior, perceived usefulness of the Internet and perceived ease of Internet use as well as attitude toward Internet use for health purposes provide the mechanism that explains health-related Internet use. In other words, for those who subjectively assess their health as susceptible to diseases and are concerned about their health, cognitive beliefs and positive affective feelings about the Internet come into play in the use of the Internet for health-related purposes.
Additionally, this integrated model shows that as health-related Internet use is predicted more by health consciousness than perceived health risk, it can be said that Internet usage for health purposes is a proactive health behavior driven by consciousness rather than a reactive health behavior. This result suggests that the Internet has become a necessary part of life for women who are health conscious and who prefer to be empowered by seeking health information online. Based on the findings of this study, the implications tend toward further promotion of Internet use for health purposes by individuals, health care service providers, and public policy makers. Knowing that health-related factors (ie, perceived health risk and health consciousness), technology-related cognitive beliefs (ie, perceived usefulness and perceived ease of use), and affective feelings toward Internet usage for health information positively influence Internet usage for searching health information, health care service providers could make greater use of the Internet to disseminate health-related information. Furthermore, health care providers can promote the use of online patient support systems or online self-care for a more seamless operation of their services. Individuals, especially women, would be motivated to seek information about health care by using the Internet, acting as opinion leaders in health and health-related issues for their family members and friends. Since the governments of all countries are keen to promote a healthy lifestyle, public policy makers could make use of the Internet to promote good health behavior, through women as the gatekeepers and as opinion leaders.
The present study has several limitations. First, the sample population focused only on working women living in urban areas. The sample was not representative of the Malaysian female population. Therefore, a more comprehensive future study is suggested to include both men and women with different ethnicities, age groups, household income levels, educational attainment levels, and place of residence for a more representative study. Second, apart from perceived health risk and health consciousness examined in this study, there are other health-related factors such as health locus of control, and health informational and decisional involvement that could be included in the deliberate reasoning process of health-related Internet use as moderator or exogenous constructs. Further, this study did not examine the influence of possible predictors of perceived ease of Internet use for health such as eHealth literacy. Therefore, we suggest that future studies could be devoted to examining the influence of these suggested constructs on health-related Internet use. Finally, based on the commonly known health-related activities that are most often performed on the Internet (namely health information seeking, communicating for health-related purposes, and purchasing drugs and health products), further studies could include purchase of drugs and other health care products as variables to enable better understanding of the use of the Internet for health maintenance activities.
Although the present study supported past research that perceived health risk and health consciousness can operate as determinants of health-related Internet use as underpinned by HBM, the HBM model is insufficient to explain the mechanism for the adoption of the Internet for health purposes. By integrating HBM and TAM, results of this study provided the insight and an understanding that perceived usefulness of the Internet for health information and attitude toward Internet usage for health purposes act as mediators on the effect of health-related factors on health-related Internet use.
Questionnaire.
average shared square variance
average variance extracted
Health Belief Model
maximum shared squared variance
Technology Acceptance Model
variance inflation factor
None declared.