Understanding Gender Differences in m-Health Adoption: A Modified Theory of Reasoned Action Model

Abstract

Background: Mobile health (m-health) services are becoming increasingly popular in healthcare, but research on m-health adoption is rare. This study was designed to obtain a better understanding of m-health adoption intention. Materials and Methods: We conducted an empirical research of a 481-respondent sample consisting of 44.7% women and 55.3% men and developed a modified theory of reasoned action (TRA) model by incorporating the nonlinearities between attitude and subjective norms and the moderating effect of gender. Results: The results indicate that, based on the study population in China: (1) facilitating conditions, attitude, and subjective norms are significant predictors of m-health adoption intention; (2) the model including the nonlinearities enhances its explanatory ability; (3) males enjoy a higher level of m-health adoption intention compared with females; (4) the modified TRA model can predict men's behavior intention better than that of women; and (5) males have an Edgeworth–Pareto substitutability between attitude and subjective norms in predicting m-health adoption intention. Conclusions: Thus, we found gender differences in m-health adoption from the perspective of social psychology.

Introduction

With the development of information technology, its application in the health industry has become a hot research area in both China and some other countries,¹ gradually forming a new research field called e-health.² Currently, the popularity of mobile devices has greatly promoted the development of e-health. As a result, mobile health services have come into being.^3,4 M-health means using mobile communications and network technologies in healthcare systems.⁵ The usage of m-health has gained additional dimensions in recent years.

With potential benefits including more efficient healthcare, available remote medical monitoring and consultation, reduced cost, etc., there is a scant body of literature on the adoption of m-health.⁶ Istepanian and Lacal⁴ did a brief overview of the development trend of m-health services; they believed that the biggest obstacle to the development of m-health was the connectivity and compatibility between mobile devices. Jones et al.⁷ pointed out that m-health technology merges personal wireless communications equipment, medical equipment, and network equipment together so as to provide patients with daily health services. Through simulation, Liang et al.⁸ verified that the emergency call mode of m-health could reduce the responding time in emergency situations.

Research studies in behavioral aspects mainly consider m-health as a technology, studying people's adoption intention using the theoretic lens of the technology acceptance model (TAM), the motivational model, and others. Wilson and Lankton⁹ applied the TAM model to the health information technology (IT) area; they found that perceived ease of use and perceived usefulness affect the patients' intention to adopt e-health. López-Nicolás et al.¹⁰ used the TAM model as a basic model to study m-health usage intention and verified that perceived ease of use and perceived usefulness have a positive influence on patients' m-health adoption. Mohamed et al.¹¹ applied the TAM model to explore m-health users' attitude and also verified the model's basic conclusion. Cocosila¹² used the motivational model to study how prior attitude influences m-health adoption behavior and found that intrinsic motivation and extrinsic motivation have positive effects on m-health adoption. López-Navarro et al.¹³ studied the adoption of m-health from the perspective of health risk perception and trust in public.

The theoretical framework for this study is based on the theory of reasoned action (TRA) model. From the perspective of social psychology, the TRA model takes attitude and subjective norms as the key factors of behavior intention,¹⁴ which is one of the most basic and influential IT acceptance models.¹⁵ Ajzen¹⁶ pointed out that in some specific areas, the explanatory ability of the model could be enhanced by modifying the model's theory mechanism. The model enhances our comprehension of how attitude and subjective norms (SNs) affect technology adoption, but it ignores the nonlinear relationship between two key constructs.¹⁷

Although we take the user's perceived ease of use, perceived usefulness, age, motivation, and other factors into consideration while studying m-health adoption, there are also some consumer attributes not incorporated.¹⁸ Although numerous studies in various settings have extensively involved the impact of gender, such as on buying behaviors¹⁹ and on blog switching behavior,²⁰ the role of gender in m-health adoption has rarely been studied.

Thus, this study aims to fill the research gap by applying a modified TRA model to m-health research. The purpose of this research is to determine whether by incorporating the nonlinearities of independent variables and the mediator gender into the TRA model we can predict users' m-health adoption intention and to study the gender differences in the intention, especially the gender differences in the nonlinearities. The remainder of the article is organized as follows: in the next section, we elaborate on the theoretical foundations and hypotheses of this research, and then we provide an overview of the proposed methodology, followed by a discussion of the results and a conclusion to our research work.

Materials and Methods

M-Health

As a type of e-health services, m-health mainly provides health services through mobile devices. The major functions of m-health consist of health consulting, hospital registering, and location-based services.²¹ M-health has more advantages over e-health in portability, mobility, and ubiquity,²² so it features personalization, interactivity, and generalization.

In terms of the functions, m-health services can be classified into five types:

1. Health information retrieval. Through m-health services, users can query health information, emergency measures of all kinds of common diseases, and visiting time.²³

2. Remote reservation. Users can register through intelligent mobile devices and visit doctors according to the appointed time.²⁴

3. Remote diagnosis. Users can record physical indicators data using digital detection equipment and upload the data to service platforms through mobile devices, achieving real-time detection, diagnosis, and monitoring of user health status.²⁵

4. Electronic medical records access. Users can log-on to m-health service platforms, refer to their electronic medical records, and browse the results or reports as well as doctors' advice information.²⁶

5. Health consultation. M-health service platforms also periodically send health guidance, disease prevention and controls, appointment notices, doctors' advice statements, and some specific notices or reminding to users.²⁷

As m-health is a new, emerging phenomenon, empirical studies on these issues are still rare. Mohamed et al.²⁸ tested whether technology design affected acceptance of m-health informatics. Hung and Jen²⁹ discovered that the perceived usefulness and perceived ease of use are different for different m-health adopters through the TAM model and that the impact of age factor is also significant at the same time. Akter et al.³⁰ studied the degree of acceptance of m-health information services and pointed out trustworthiness as a secondary variable has direct and indirect effects on continuous intention, but there is no mediating role. Wu et al.³¹ studied acceptance of m-healthcare systems with the TAM model and the diffusion of innovations theory and found that compatibility and self-efficacy affect both perceived usefulness and perceived ease of use, whereas technical support and training have no significant effects. Yang and Wang³² studied acceptance behavior for asthma care mobile service and added SN and attitude toward use to the TAM model, finding that SN has direct and indirect impact on behavior intention and that the perceived ease of use and attitude toward use can mediate the perceived useful effects on the dependent variable. Thus, in m-health contexts, the TRA model is rarely used.

TRA Model

The TRA model explains or predicts individual specific behaviors in many areas with volitional control and achieved good effects.^14,33 The premise hypothesis of the model is that people's behavior intention influence their behaviors, and, in turn, behavior intentions are influenced by attitude and SN. “The attitudinal component refers to the person's attitude toward performing the behavior under consideration,”¹⁴ whereas the normative component refers to “The person's subjective norm, that is, his perception that most people who are important to him think he should or should not perform the behavior in question.”¹⁴ So the basic framework of the TRA model is that behavior is affected by behavior intention, and behavior intention is affected by the inner factors attitude toward the behavior (ATTB) and SNs.

In addition to internal factors, behavior intention is also affected by external factors such as facilitating conditions (FC).³⁴ The term FC means resources needed to achieve the behavior such as time, knowledge, and other specialized resources.³⁵ Much research has introduced FC into the TRA model and tested their impacts on behavior intention.^17,36,37 Thus, the common modified TRA model is formed with attitude, SNs, and FC as independent variables.^17,36,37 Thus, we suggest these hypotheses:

H1. FC have a positive influence on m-health adoption intention.

H2. ATTB has a positive influence on m-health adoption intention.

H3. SNs have a positive influence on m-health adoption intention

Nonlinearities between ATTB and SN

Although attitude and SNs are two key constructs of the TRA model,^14,16 the nonlinearities between them are often overlooked. The hypothesis based on the theoretical independence between them is likely to underestimate or overestimate the explanatory ability of their interpretation of behavior intention,³⁸ and their nonlinearities are seldom incorporated in the model on IT contexts.¹⁷ Rabow et al.³⁹ discovered that ATTB * SN has a strong influence on behavior intention in the study of adult alcohol consumption. Terry et al.⁴⁰ verified that ATTB * SN can well predict behaviors. Titah and Barki¹⁷ first introduced ATTB * SN as a fourth variable to the TRA model to explain IT acceptance and found that the explanatory ability of the model was improved.

Current articles have proven an Edgeworth–Pareto substitutability between ATTB and SNs.^17,40,41 The Edgeworth–Pareto substitutability declares that if the combined influence of the two factors is weaker than the sum of two variables' independent influence, there is a negative synergy.¹⁷ So when the influence of a factor increases, the marginal effect of another factor will be reduced. Instead, complementarity means that when the influence of a factor increases, the influence of another factor will be increased. The relationship above can be illustrated by the following formulas:

Assume that X ₁ and X ₂ are impact factors of Y, and considering their nonlinear relationship, the function can be written as \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}Y = \beta_0 + \beta_1X_1 + \beta_2X_2 + \beta_3X_1 * X_2 + \xi\end{align*} \end{document}

When β ₃>0, X ₁ and X ₂ are complements; when β ₃<0, X ₁ and X ₂ are substitutes; and when β ₃=0, X ₁ and X ₂ are independent.

M-health is accepted by users voluntarily. Users can decide on the application degree and extent according to their requirements. But when one is influenced by strong SNs (relatives, influential people) and then has m-health adoption intention, changes in his or her attitude will only slightly influence his or her adoption intention. In turn, when one has a very positive attitude toward m-health, thinking that m-health is particularly useful, changes in SNs will only slightly influence the adoption intention.⁴² So the Edgeworth–Pareto substitutability between attitude and SNs in m-health adoption intention seems plausible. Thus, we suggest:

H4. The nonlinearities between ATTB and SN have a negative influence on m-health adoption intention, indicating substitutability.

Gender Differences

The influences of personal characteristics on behavior intention has attracted considerable research attention.^19,33 In IT acceptance contexts, Gefen and Straub⁴³ added the factor of gender into the TAM model as a part of social and cultural factors to IT context, drawing the conclusion that the influences of perceived ease of use and perceived usefulness in e-mail technology diffusion differ by gender. Venkatesh et al.² put gender and other factors (age, experience, etc.) as moderator variables into the unified theory of acceptance and use of technology model. Okazaki and Mendez⁴⁴ took gender as a moderator in mobile commerce acceptance and found that there is a gender difference. In TRA model applications, Al-maghrabi and Dennis⁴⁵ used the TRA model to study the factors of continuance online shopping and found the gender differences cannot be ignored. Ryu and Han³³ introduced gender into the TRA model to study tourists' behavioral intention and found that gender had a significant influence on the relationship between attitude and behavior intention but no significant influence on the relationship between SNs and behavioral intention.

According to social role theory, women and men play different roles and behave differently in society because they are socialized differently.⁴⁶ Women in general are relatively passive and do not like to adventure. Instead, men are more active and adventurous, so men are more positive about exploring new things.⁴⁷ “Many researchers have proposed that women are characterized as more communal while men as more agentic.”²⁰ In particular, women are likely to be more anxious about mobile technology.⁴⁸ On the other hand, men are expected to be more task-oriented than women. If they or their friends think something is useful and important, they may have a higher level of usage intention.⁴⁹ Lin et al.⁵⁰ found that the positive influence of perceived usefulness on Facebook continuance intention will be stronger for men than for women. Venkatesh et al.⁵¹ suggested that men tend to rely less on FC when considering use of a new technology, whereas women tend to place greater emphasis on external supporting factors.

So the gender factor may affect m-health services adoption intention. We incorporated gender as a moderator variable into the model, exploring the gender differences in the relationship between dependent and independent variables. Thus, we suggest:

H5. Gender has a significant moderating role in the relationship between FC and m-health adoption intention.

H6. Gender has a significant moderating role in the relationship between ATTB and m-health adoption intention.

H7. Gender has a significant moderating role in the relationship between SNs and m-health adoption intention.

H8. Gender has a significant moderating role in the relationship between ATTB * SN and m-health adoption intention.

In summary, this article was designed (a) to examine the construct validity of a modified TRA model in explaining people's m-health adoption intention, (b) to study the effects of FC, attitude, and SN on m-health adoption intention, (c) to test the effect of the nonlinearities relationship between ATTB and SN on m-health adoption intention, (d) to discover the effects of gender as a moderator variable in a modified TRA model, and (e) to discover the gender differences of the nonlinearities relationship between ATTB and SN. Figure 1 presents the proposed model of our research.

Fig. 1.

Proposed model. The dotted lines signify the moderating effect. AI, adoption intention; ATTB, attitude toward the behavior; FC, facilitating conditions; SN, subjective norm.

Results

Data Collection and Measures

To test the study hypothesis, we used a questionnaire to collect data in a field survey of customers of a large company providing m-health services in Harbin, China. Construct measures were adapted from previous studies with all items assessed on 5-point Likert-type scales. To ensure our data were robust, we distributed our questionnaires in different industries and covered people of different ages. We sent out 510 questionnaires, from which 481 valid questionnaires were collected after removing the invalid returns. Table 1 shows the demographic information of our respondents.

Table 1.

Respondent Demographics

	FREQUENCY	PERCENTAGE
Gender
Female	215	44.7%
Male	266	55.3%
Age (years)
≤30	299	62.2%
30–60	115	23.9%
≥60	67	13.9%
Mobile device usage experience (years)
<4	186	38.7%
4–8	174	36.2%
>8	121	25.1%

As shown in Table 2, a preliminary assessment of the survey instrument was conducted by partial least squares (PLS). It shows that the composite reliabilities are greater than 0.85 and that the average variances extracted are greater than 0.65, highly above the suggested cutoff values of 0.70 and 0.50, respectively.^52,53 Factor loadings are found to be significant and higher than 0.70. All these suggest good construct reliability.

Table 2.

Measures for Constructs

CONSTRUCT	CR	AVE	LOADINGS	MEAN (STANDARD DEVIATION)
FC²	0.854/0.840/0.888	0.662/0.640/0.726		3.520 (1.097)/3.501 (1.290)/3.543 (0.858)
• I can use the services easily.			0.876/0.883/0.868
• I have the human and technological resources necessary to use the services.			0.704/0.650/0.825
• I have the knowledge necessary to use the services.			0.850/0.848/0.862
ATTB³⁵	0.943/0.943/0.942	0.806/0.805/0.804		3.532 (1.181)/3.423 (1.198)/3.659 (1.131)
• Using m-health is a good idea.			0.875/0.869/0.883
• Using m-health is a wise idea.			0.904/0.892/0.919
• I like using m-health.			0.923/0.920/0.926
• Using m-health is interesting.			0.890/0.909/0.857
SN³⁵	0.922/0.925/0.918	0.856/0.860/0.849		3.513 (1.126)/3.436 (1.146)/3.605 (1.081)
• People who influence me think that I should use m-health services.			0.948/0.918/0.918
• People who are important to me think that I should use m-health services.			0.931/0.936/0.925
AI⁵⁴	0.929/0.930/0.926	0.814/0.816/0.807		3.417 (1.161)/3.335 (1.252)/3.526 (1.035)
• I will use the services in the next 3 months.			0.889/0.897/0.878
• I predict I will use the services in the next 3 months.			0.905/0.898/0.912
• I plan to use the services in the next 3 months.			0.913/0.916/0.905

Data are for total population/males/females.

AI, adoption intention; ATTB, attitude toward the behavior; AVE, average variances extracted; CR, composite reliability; FC, facilitating conditions; m-health, mobile health; SN, subjective norm.

The discriminant validity can be assessed by comparing the square root of average variances extracted and the correlations.⁵⁵ As shown in Table 3, from the item-level correlation matrix we can observe that all the correlations are lower than the root of average variances extracted. Therefore, the discriminant validity is acceptable. Thus, our measurement model has been tested as reliable.

Table 3.

Discriminant Validity for Constructs

	FC	ATTB	SN	AI
FC	0.814/0.800/0.852
ATTB	0.579/0.593/0.568	0.898/0.897/0.897
SN	0.568/0.620/0.504	0.687/0.687/0.679	0.925/0.927/0.921
AI	0.524/0.597/0.422	0.539/0.595/0.451	0.572/0.679/0.412	0.902/0.903/0.898

Data are for total population/males/females.

AI, adoption intention; ATTB, attitude toward the behavior; FC, facilitating conditions; SN, subjective norm.

Structural Model

The model analysis is conducted in two stages. In the first stage, to test our model and the role of the nonlinearities between ATTB and SN, we built model M1 without the nonlinearities and model M2 with the nonlinearities. We used PLS to test our models with all of the data. The results are shown in Table 4.

Table 4.

Partial Least Squares Results in Stage 1

	MODEL 1			MODEL 2
PATH	β	T STATISTICS	COEFFICIENT	β	T STATISTICS	COEFFICIENT
FC→AI	0.240	5.813	p<0.05	0.241	6.664	p<0.05
ATTB→AI	0.192	4.117	p<0.05	0.386	3.896	p<0.05
SN→AI	0.304	7.474	p<0.05	0.501	4.892	p<0.05
ATTB ^* SN→AI				–0.367	2.098	p<0.10
R ²	0.336			0.407

AI, adoption intention; ATTB, attitude toward the behavior; FC, facilitating conditions; SN, subjective norm.

Second, to discover the role of gender moderator, we divided our data into two parts according to gender and tested our modified model separately. To perform the subgroup analysis, we used the formula suggested by Ahuja and Thatcher⁵⁶ to calculate the discrepancy of the designated paths by the standard error and the corresponding path coefficient. The results are shown in Table 5.

Table 5.

Partial Least Squares Results in Stage 2

	MODEL 3 (MALES)			MODEL 4 (FEMALES)			COMPARISON OF PATHS
PATH	β	T STATISTICS	COEFFICIENT	β	T STATISTICS	COEFFICIENT	T STATISTICS	COEFFICIENT
FC→AI	0.242	7.161	p<0.05	0.222	5.210	p<0.05	5.605	p<0.05
ATTB→AI	0.331	4.709	p<0.05	0.284	2.605	p<0.05	5.430	p<0.05
SN→AI	0.570	7.474	p<0.05	0.212	1.688	p<0.10	39.44	p<0.05
ATTB ^* SN→AI	–0.300	8.082	p<0.05	–0.115	0.569		12.39	p<0.05
R ²	0.529			0.256

AI, adoption intention; ATTB, attitude toward the behavior; FC, facilitating conditions; SN, subjective norm.

Discussion

Our research aims to inspect one modified TRA model and the mediating role of gender in explaining people's m-health adoption intention. Furthermore, we examine the Edgeworth–Pareto substitutability between attitude and SNs and its gender differences.

Findings

First, the effects of FC, attitude, and SNs on behavioral intention are tested. The regression path from FC to adoption intention is significant, thus supporting Hypothesis 1. FC have a positive influence on m-health adoption intention. It means that if someone possesses or he or she thinks he or she possesses the right thing necessary for m-health usage, he or she would have a high stage of adoption intention. The regression path from attitude to adoption intention is significant, thus supporting Hypothesis 2. It means that if one has a positive attitude toward m-health, he or she would expect to use m-health services. The regression path from SNs to adoption intention is also significant, thereby supporting Hypothesis 3. It demonstrates that one would have a higher probability to adopt m-health services if other people who are important to him or her or can influence him or her to think he or she should. All the factors are significant, and the model explains a certain proportion (33.6%) of the variance in adoption intention, indicating that the TRA model can be used in m-health contexts.

Second, from the comparison of Model 1 and Model 2 we can come to a conclusion that when incorporating the nonlinearities between ATTB and SN, the nonlinear model explains a significantly greater proportion of the variance than the linear model (33.6% versus 40.7%), indicating that the nonlinear model predicts behavior better. What is more, the effect of the nonlinearities between ATTB and SN is significant and negative (β=–0.376, t=2.098, p<0.01); thus the Edgeworth–Pareto substitutability between attitude and SNs is proved in m-health adoption.

Third, as shown in Table 5, the two groups had a significant difference in m-health adoption intention. From whichever aspect, FC (0.242 versus 0.222, t=5.605, p<0.01), attitude (0.331 versus 0.284, t=5.430, p<0.01), or SNs (0.570 versus 0.212, t=39.44, p<0.01), males always have a higher level of m-health adoption intention compared with females. Thus, our Hypotheses 5–7 are all supported, and we can conclude that men find it easy to adopt m-health. In particular, the estimates for path coefficients from SNs to adoption intention across groups have a significant difference, indicating that men find it easier to adopt m-health when influenced by important or influential people.

Finally, we come to the gender differences on the nonlinearities between ATTB and SN. From Table 5, based on the data from the study population in China, we also find that the path coefficient from the interaction to adoption intention is significant in the male group (β=–0.300, t=8.082, p<0.01) but not significant in the female group (β=–0.115, t=0.569, p>0.1). So for men, when attitude is high, increases in SNs have a decreasing marginal impact on adoption intention. Alternatively, when SNs are high, increases in attitude have a decreasing marginal impact on adoption intention. But, for women the phenomenon is not significant. So men have an Edgeworth–Pareto substitutability between attitude and SNs, whereas women do not. What is more, for men the modified TRA model explains 52.9% of the variability of adoption intention, but for women it is only 25.6%. So other factors may exist affecting women adopting m-health besides social psychological factors. So the modified TRA model fits males better.

Limitations and Future Research

Some limitations still exist in the present study. This article only involves one nonlinearity; the others such as FC*ATTB or FC*SN are not incorporated, which have been studied by Bansal and Taylor⁵⁷ but have not been introduced to m-health adoption research. Second, the modified TRA model fits the male gender well, but for the female gender the model fails to explain the variability by more than 30%; in other words, our model does not incorporate the main factors of female. Finally, although we have found that women do not have an Edgeworth–Pareto substitutability between attitude and SNs, we did not find out the inherent reasons. Our future research aims to develop a suitable model to explain the m-health adoption behavior and find out the reason why females and males perform differently according to the nonlinearities between ATTB and SN.

Contributions

In spite of these limitations above, the study makes several theoretical contributions. First, the study introduces a modified TRA model with the nonlinearities between attitude and SNs to explain or predict people's m-health adoption intention for the first time. Second, we use gender as a moderator into the modified TRA model and discover the gender differences of the relationship between social psychological factors and m-health adoption intention. Third, we find that the modified TRA model based on social psychological factors fits men better than women for the reason that they are differently socialized.⁴⁶ Finally, for the first time we find the gender differences of the Edgeworth–Pareto substitutability between attitude and SNs in m-health adoption.

On practical grounds, the findings of the present study have value for m-health services providers. The Edgeworth–Pareto substitutability between attitude and SNs means that the high level of one factor can compensate for the other factor in low level. Providers can use this mechanism to gain deeper insights into the interaction of attitude and SNs. To develop male users, they can take advantage of the substitution effect to make informed decisions on how much effort to invest to enhance potential users' positive attitude or strong SNs.

Conclusions

Although m-health services have drawn considerable attention in the research field of the world today, studies on the relationship between social psychological factors and m-health adoption are rarely conducted. The proposed research model reveals the effect of the nonlinearities between attitude and SNs and the role of the moderator gender. Our findings are as follows: (1) when the TRA model includes the nonlinearities, it may enhance the explaining ability of the model; (2) males have a higher-level adopt intention of m-health compared with females; (3) the modified TRA model predicts males' behavior intention better than that of females; and (4) Edgeworth–Pareto substitutability exists between attitude and SNs in explaining m-health adoption intention, but this phenomenon exists only when it comes to males. Our study provides a new understanding on both gender differences in m-health adoption and the TRA model usage.

Footnotes

Acknowledgments

This study was partially funded by the NSFC 71101037, NCET-12-0146, and the Hong Kong Scholars Program. We would like to thank the editors and two anonymous reviewers for their insightful comments and suggestions.

Disclosure Statement

No competing financial interests exist.

References

Bhattacherjee

, Hikmet

. Physicians' resistance toward healthcare information technologies: A dual-factor model. 40th International Conference on System Sciences, HICSS 2007. New York: IEEE, 2007; 141–151.

Venkatesh

, Morris

, Davis

. User acceptance of information technology: Toward a unified view. MIS Q, 2003; 27:425–478.

Liu

, Zhu

, Holroyd

, Seng

. Status and trends of mobile-health applications for iOS devices: A developer's perspective. J Syst Software, 2011; 84:2022–2033.

Istepanian

, Lacal

. Emerging mobile communication technologies for health: Some imperative notes on m-health. Engineering in Medicine and Biology Society. Proceedings of the 25th Annual International Conference of the IEEE. New York: IEEE, 2003; 2:1414–1416.

Istepanian

, Laxminarayan

, Pattichis

. M-health: Emerging mobile health systems. In: Istepanian

, Laxminarayan

, Pattichis

, eds. M-health: Emerging mobile health systems. Berlin: Springer, 2006:182–185.

Guo

, Sun

, Yuan

, Yan

. Privacy-personalization paradox in adoption of mobile health service: The mediating role of trust. Proceedings of International Conference on Management Science and Engineering. 2012. Available at http://aisel.aisnet.org/pacis2012/27 (last accessed February 15, 2012 ).

Jones

, van Halteren

, Dokovsky

, Koprinkov

, Peuscher

, Bults

, Konstantas

, Widya

, Herzog

. Mobihealth: Mobile services for health professionals. In: Istepanian

, Laxminarayan

, Pattichis

, eds. M-Health: Emerging mobile health systems. Berlin: Springer, 2006:237–246.

Liang

, Lu

, Chen

, Lin

, Shen

. PEC: A privacy-preserving emergency call scheme for mobile healthcare social networks. J Commun Netw, 2011; 13:102–112.

Wilson

, Lankton

. Modeling patients' acceptance of provider-delivered e-health. J Am Med Inform Assoc, 2004; 11:241–248.

10.

López-Nicolás

, Molina-Castillo

, Bouwman

. An assessment of advanced mobile services acceptance: Contributions from TAM and diffusion theory models. Inform Manage, 2008; 45:359–364.

11.

Mohamed

AHH

, Tawfik

, Al-Jumeily

, Norton

. MoHTAM: A technology acceptance model for mobile health applications. In: 2011 Developments in E-systems Engineering. New York: IEEE, 2011:13–18.

12.

Cocosila

. Role of user a priori attitude in the acceptance of mobile health: an empirical investigation. Electron Markets, 2013; 23:15–27.

13.

López-Navarro

MÁ

, Llorens-Monzonís

, Tortosa-Edo

. The effect of social trust on citizens' health risk perception in the context of a petrochemical industrial complex. Int J Environ Res Public Health, 2013; 10:399–416.

14.

Ajzen

, Fishbein

. Understanding attitudes and predicting social behaviour. Englewood Cliffs, NJ: Prentice Hall, 1980.

15.

Omar

, Owusu-Frimpong

. Life insurance in Nigeria: An application of the theory of reasoned action to consumers' attitudes and purchase intention. Service Ind J, 2007; 27:963–976.

16.

Ajzen

. The theory of planned behavior. Organizational Behav Hum Decis Processes, 1991; 50:179–211.

17.

Titah

, Barki

. Nonlinearities between attitude and subjective norms in information technology acceptance: A negative synergy?. MIS Q, 2009; 33:827–844.

18.

Chellappa

, Sin

. Personalization versus privacy: An empirical examination of the online consumer's dilemma. Inform Technol Manage, 2005; 6:181–202.

19.

Homburg

, Giering

. Personal characteristics as moderators of the relationship between customer satisfaction and loyalty—An empirical analysis. Psychol Market, 2001; 18:43–66.

20.

Zhang

, Lee

, Cheung

, Chen

. Understanding the role of gender in bloggers' switching behavior. Dec Support Syst, 2009; 47:540–546.

21.

Ivatury

, Moore

, Bloch

. A doctor in your pocket: Health hotlines in developing countries. Innov Technol Governance Globalization, 2009; 4:119–153.

22.

Akter

, D'Ambra

, Ray

. Service quality of mhealth platforms: Development and validation of a hierarchical model using PLS. Electron Markets, 2010; 20:209–227.

23.

Germanakos

, Mourlas

, Samaras

. A mobile agent approach for ubiquitous and personalized eHealth information systems. Proceedings of the Workshop on 'Personalization for e-Health' of the 10th International Conference on User Modeling (UM'05). Edinburgh: 2005:67–70. Available at http://cgi.csc.liv.ac.uk/∼floriana/UM05-eHealth/UM05-W6.pdf (last accessed February 15, 2012 ).

24.

Baarda

, Jones

, Stemerding

. BAN-based m-health services: Experiences and prospects. Technical report TR-CTIT-05-67. Enschede, The Netherlands: Centre for Telematics and Information Technology, University of Twente, 2005.

25.

Istepanian

, Jovanov

, Zhang

. Guest editorial introduction to the special section on m-health: Beyond seamless mobility and global wireless health-care connectivity. IEEE Trans Inform Technol Biomed, 2004; 8:405–414.

26.

Jovanov

. Wireless technology and system integration in body area networks for m-health applications. 27th Annual International Conference of the Engineering in Medicine and Biology Society. New York: IEEE, 2006:7158–7160.

27.

van Velthoven

, Car

, Atun

. Telephone consultation for improving health of people living with or at risk of HIV: A systematic review. PloS One, 2012; 7:e36105.

28.

Mohamed

AHH

, Tawfik

, Norton

, Al-Jumeily

. Does e-health technology design affect m-health informatics acceptance? A case study. Proceedings of IEEE-EMBS International Conference on Biomedical and Health Informatics. New York: IEEE, 2012; 968–971.

29.

Hung

M-C

, Jen

W-Y

. The adoption of mobile health management services: An empirical study. J Med Syst, 2012; 36:1381–1388.

30.

Akter

, D'Ambra

, Ray

. Trustworthiness in mhealth information services: An assessment of a hierarchical model with mediating and moderating effects using partial least squares (PLS). J Am Soc Inform Sci Technol, 2011; 62:100–116.

31.

J-H

, Wang

S-C

, Lin

L-M

. Mobile computing acceptance factors in the healthcare industry: A structural equation model. Int J Med Inform, 2007; 76:66–77.

32.

Yang

H-Y

, Wang

M-J

. What factors affect physicians' decisions to use an e-health care system?. Health, 2012; 4:1023–1028.

33.

Ryu

, Han

. Predicting tourists' intention to try local cuisine using a modified theory of reasoned action: The case of New Orleans. J Travel Tourism Market, 2010; 27:491–506.

34.

Triandis

. Values, attitudes, and interpersonal behavior. Nebraska Symp Motivation, 1980; 27:195–259.

35.

Taylor

, Todd

. Understanding information technology usage: A test of competing models. Inform Syst Res, 1995; 6:144–176.

36.

Teo

, Al-Fadhli

, Chou

P-N

, Chen

W-F

, Dwyer

, Loannou

. Influence of user characteristics on teachers' intention to use technology: Some research evidence. Int J Instruct Media, 2011; 38:7–16.

37.

Püschel

, Mazzon

, Hernandez

JMC

. Mobile banking: Proposition of an integrated adoption intention framework. Int J Bank Marketing, 2010; 28:389–409.

38.

Liska

. A critical examination of the causal structure of the Fishbein/Ajzen attitude-behavior model. Soc Psychol Q, 1984; 47:61–74.

39.

Rabow

, Neuman

, Hernandez

ACR

. A configurational approach to contingent contingency in attitude-behavior relationship. Soc Psychol Q, 1987; 50:50–63.

40.

Terry

, Hogg

, McKimmie

. Attitude‐behaviour relations: The role of in‐group norms and mode of behavioural decision‐making. Br J Soc Psychol, 2000; 39:337–361.

41.

Shih

Y-Y

, Fang

. The use of a decomposed theory of planned behavior to study Internet banking in Taiwan. Internet Res, 2004; 14:213–223.

42.

Weber

. Edgeworth on complementarity, or Edgeworth, Auspitz-Lieben, and Pareto de-homogenized. History Political Econ, 2005; 37:293–307.

43.

Gefen

, Straub D

. Gender differences in the perception and use of e-mail: An extension to the technology acceptance model. MIS Q, 1997; 21:389–400.

44.

Okazaki

, Mendez

. Exploring convenience in mobile commerce: Moderating effects of gender. Comput Hum Behav, 2013; 29:1234–1242.

45.

Al-maghrabi

, Dennis

. The driving factors of continuance online shopping: Gender differences in behaviour among students—The case of Saudi Arabia. Int J Business Inform Syst, 2012; 9:360–384.

46.

Saad

, Gill

. Applications of evolutionary psychology in marketing. Psychol Market, 2000; 17:1005–1034.

47.

Powell

, Ansic

. Gender differences in risk behaviour in financial decision-making: An experimental analysis. J Econ Psychol, 1997; 18:605–628.

48.

Gilbert

, Lee-Kelley

, Barton

. Technophobia, gender influences and consumer decision-making for technology-related products. Eur J Innov Manage, 2003; 6:253–263.

49.

Eagly

. Sex differences in social behavior: A social-role interpretation. Mahwah, NJ: Lawrence Erlbaum Associates, 1987.

50.

Lin

, Li

, Califf

, Featherman

. Can social role theory explain gender differences in Facebook usage. 46th Hawaii International Conference on System Sciences. New York: IEEE, 2013; 690–699.

51.

Venkatesh

, Thong

JYL

, Xu

. Consumer acceptance and use of information technology: Extending the unified theory of acceptance and use of technology. MIS Q, 2012; 36:157–178.

52.

Chin

. Commentary: Issues and opinion on structural equation modeling. MIS Q, 1998; 22:vii–xvi.

53.

Wetzels

, Odekerken-Schroder

, Van Oppen

. Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. MIS Q, 2009; 33:177–190.

54.

Johnston

, Warkentin

. Fear appeals and information security behaviors: An empirical study. MIS Q, 2010; 34:549–566.

55.

Fornell

, Larcker

. Evaluating structural equation models with unobservable variables and measurement error. J Market Res, 1981; 18:39–50.

56.

Ahuja

, Thatcher

. Moving beyond intentions and toward the theory of trying: Effects of work environment and gender on post-adoption information technology use. MIS Q, 2005; 29:427–459.

57.

Bansal

, Taylor

. Investigating interactive effects in the theory of planned behavior in a service-provider switching context. Psychol Market, 2002; 19:407–425.