Content uploaded by Emre Sezgin
Author content
All content in this area was uploaded by Emre Sezgin on Jul 23, 2020
Content may be subject to copyright.
Article
Understanding the perception
towards using mHealth applications
in practice: physicians’ perspective
Emre Sezgin, Sevgi O
¨zkan-Yildirim and Soner Yildirim
Middle East Technical University
Abstract
The objective of this study was to investigate physicians’ perceptions to use mobile health applications in
practice, and to identify influencing factors to use the technology. An mHealth technology acceptance model
was proposed (M-TAM), and a cross-sectional survey was implemented using structured questionnaire to
collect data. Online tools were used for inviting participants (physicians) and data collection from Turkey. The
data was analyzed using Confirmatory Factor Analysis (CFA) and Structural Equation Modeling (SEM). A total
of 128 physicians participated in the survey. The model explained the perception of physicians towards mHealth
application use by 51% of total variance. The influential factors were identified as Effort Expectancy, Mobile
Anxiety, Perceived Service Availability and Technical Training and Support. The study provided a new model to
the literature of health information technology. Findings of the research contributed by unveiling latent con-
structs and their influence on physicians’ perceptions towards a new healthcaretechnology: mHealth applications.
Keywords
mobile health, health information technology, technology acceptance, physicians’ perception
Submitted: 5 May, 2016; Accepted: 23 November, 2016.
Different factors affect the perception of user and non-user physicians towards mHealth
applications.
Introduction
Alignment of healthcare services with the mobile
platform is a promising technological leap. The mobi-
lity of healthcare services has increased reachability,
accessibility and has the ability to reach more indi-
viduals, especially in rural areas (Sarker, 2003; Nah
and Siau, 2005; Ve´lez et al., 2014; Anderson et al.,
2013). Mobile health (mHealth) can be defined as the
healthcare related technology providing mobile
information communication and network systems
together with services (Adibi, 2015). In that, mHealth
comprises mobile devices and other peripherals used
by healthcare providers, patients and customers in
order to gather, store and analyze data in the decision
making processes (Sezgin and O
¨zkan-Yildirim,
2014). In the market, there is a vast number of assis-
tive mobile health services for use in diagnostic stages
and health management (World Health Organisation,
2011; Istepanian et al., 2010; Organisation for Eco-
nomic Co-operation and Development, 2015). There
is a high rate of increase in mobile device use world-
wide. Gartner reported that smart phone use world-
wide has been rapidly increasing since 2007 (Gartner
Inc., 2012). In Turkey also, there is a high level of use
of mobile phones. The Turkish Statistical Institute
(TUIK in Turkish) reported that 96.8%of the Turkish
population are using mobile phones, and in 2015, it
was reported that 58.7%of individuals in Turkey have
Corresponding author:
Emre Sezgin, Middle East Technical University, School of
Informatics, 06800, C¸ankaya, Ankara, Turkey.
Email: esezgin@metu.edu.tr
Information Development
1–19
ªThe Author(s) 2016
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0266666916684180
journals.sagepub.com/home/idv
smart phone Internet access (Turkish Statistical Insti-
tute, 2015).
Considering the common trajectory in mobile
access and connectivity, the use of mobile devices
in healthcare services has inevitably gained impor-
tance (Tachakra et al., 2003; Atluri et al., 2015; Kim
et al., 2016). According to reports by Wolters Kluwer
and Deloitte, providing healthcare services has been
thriving in terms of development of new mobile tech-
nologies as well as their use by healthcare providers
(Deloitte, 2013; Wolters Kluwer, 2013). In Turkey,
the use of mobile health and the associated market for
healthcare services have not yet reached maturity.
The Ministry of Health and the Social Security
Institution have initiated projects regarding health
management and tracking; however, real life imple-
mentations have been limited to private initiatives
only (Tezcan, 2016).
Literature review
Turkish health literature with regards to mobile health
has been said to be very limited. A current research in
Turkey revealed that the numbers of potential users
are high and there is a need for developments in
mobile health considering demographic characteris-
tics (Doganyigit and Yılmaz, 2015). On the other
hand, regarding to the global developments, the liter-
ature provided a number of studies about the use of
mobile devices and applications in healthcare services
(Gagnon et al., 2016; Hampton, 2012). Some of these
studies raised concerns in that regard. Varshney
(2014) noted that the increasing use of mobile health
services may reduce the quality of the service deliv-
ered by increasing frequency of interruptions in deliv-
ery for healthcare providers, and requiring
multitasking in daily routines may lead to errors in
practice. Mobile technologies were also argued to
have clinical risks, such as data security, confidenti-
ality and electromagnetic interference (Visvanathan
et al., 2011). In that regard, assessing the impact of
mobile technologies in terms of the actual use and
adoption of the technology by healthcare providers
has been a significant concern (Kahn et al., 2010;
Okazaki et al., 2012).
From the technical perspective, healthcare technol-
ogies are the shining stars. Health information tech-
nologies provide tools and utilities to improve
medical communications as well as helping to lower
expenses through the quality of services and saving
time in providing healthcare services (Chaudhry et al.,
2006; Siau and Shen, 2006; Becker et al., 2014). How-
ever, Ward (2013) underlines that a socio-technical
approach is required while implementing new infor-
mation technologies which blend social drivers and
technological decisions. In parallel, the literature
reveals that healthcare technologies may not be as
promising as they seem, since factors like users’ pre-
vious experiences, risk levels and organizational cul-
ture can remarkably affect the attitudes of
professionals (Ludwick and Doucette, 2009; Li
et al., 2013). Thus, questions remain as to the willing-
ness and intentions of healthcare providers to use
these technologies (Ducey and Coovert, 2016; Kum-
mer et al., 2013; Hung and Jen, 2010). Consequently,
understanding the intentions of healthcare providers
to use new technologies is as important as developing
a fully functional healthcare service application. To
address this issue, the US Health Information Tech-
nology for Economic and Clinical Health (HITECH)
Act provided subsidies in 2009, allocating around
US$ 22 million to increase the adoption of informa-
tion technology in US healthcare (United States Gov-
ernment Publishing Office, 2009; Moores, 2012).
However, adoption and use rates remained below
expectations, with only 17%use rate of electronic
health records technology, even though there was a
noticeable increase in healthcare service quality due
to information technology use (Moores, 2012). To
understand the key elements in adoption, an
in-depth look at the perceptions and factors influen-
cing the use of mobile health technologies by health-
care providers is required.
Recent studies revealed that the adoption of mobile
healthcare services (i.e. management, monitoring and
communications in healthcare systems with patients
and providers) is a popular field of study (Hung and
Jen, 2010; Wu et al., 2010; Chen et al., 2010; Han
et al., 2006; Piette et al., 2011; Lin and Yang, 2009;
Lin, 2011; Iredale et al., 2011; Wu et al., 2009). Even
though there have been studies assessing the attitudes
of users towards particular health technologies (Steele
et al., 2009; Lin et al., 2012; Zailani et al., 2014;
Sezgin and O
¨zkan-Yildirim, 2016), there is relatively
limited research on healthcare providers’ acceptance
of mobile health technologies (Jersak et al., 2013).
The literature of technology perception (not usage)
studies in mobile and healthcare technologies are also
scarce. One of the studies was conducted by Hong et al.
(2006) investigating the perceptions toward using
mobile data services in e-government, and they col-
lected data from potential users in Hong Kong. The
2Information Development
study revealed direct and mediating effect of per-
ceived usefulness and ease of use. Steele et al.
(2009) conducted a focus group study to understand
elderly persons’ perceptions about wireless sensors
in healthcare assistance, revealing ‘cost’ and ‘inde-
pendency’ as the important factors. Another study
investigated patients’ perception towards a home tel-
ecare system (Rahimpour et al., 2008). Its findings
provided the common concerns about the technology
as ‘ease of use’, ‘cost’, ‘low self-efficacy’, ‘anxiety’
and ‘clinical support’.
In the literature, there is a lack of studies investi-
gating the perceptions of physician non-users (i.e.
potential users) of mobile health applications, in order
to understand their attitudes to use (Fiordelli et al.,
2013). This study contributes to the literature by
focusing on the assessment of perceptions towards
using mHealth applications by non-user physicians.
Research objectives
The objectives of the study are
(1) providing an insight about the perceptions of
physicians of mHealth applications
(2) proposing a research model to assess influen-
cing factors in mHealth application use
(3) providing implications for managers and
developers about physicians’ mHealth
perceptions.
In this context, a Mobile Health Technology
Acceptance Model (M-TAM) has been proposed to
identify and understand the influencing factors of per-
ception in using mHealth applications. It is believed
that the findings of this study would help to improve
the quality of the healthcare services.
Research questions
RQ1: ‘‘What are the factors influencing physicians’
perception to use mHealth applications?’’
In that regard, the effects on the behavioral inten-
tions were hypothesized to seek answers for this
research question. The concept of Behavioural Inten-
tion (BI) emerged as behavior predictor in TPB
(Ajzen, 1991), and it was widely used and validated
in many acceptance studies (Venkatesh and Bala,
2008; King and He, 2006; Holden and Karsh,
2010; Yousafzai et al., 2007; Or and Karsh, 2012).
RQ2: ‘‘What are the relationships among the factors
influencing the perception to use of mHealth
applications?’’
The second research question was for seeking inter-
relations among the constructs. In addition to the
direct effect of constructs to the behavioral intention,
their impact on behavioral intention would also be
observed via mediating effects over PE and EE
(Moores, 2012).
Research model
This study aims providing insight about the perceptions
of physicians towards mHealth applications and the
nature of influences towards their adoption. The study
embraced a number of behavioral studies of technology
acceptance and adoption theories in the literature in
order to establish the model: Mobile Health Technol-
ogy Acceptance Model (M-TAM) (Sezgin and O
¨zkan-
Yildirim, 2014). Unified Theory of Acceptance and
Use of Technology 1 (UTAUT) was considered as the
basis for the theoretical structure of M-TAM. The base
model was expanded by integrating with other pioneer-
ing behavioral theories—Technology Acceptance
Model (TAM) (Davis, 1989; Venkatesh and Davis,
2000), Theory of Planned Behavior (TPB) (Ajzen,
1991) and Innovation Diffusion Theory (IDT) (Rogers
and Shoemaker, 1971).
The constructs of the research model were devel-
oped utilizing the theories within the literature. In
addition, experts’ opinions were included in the study
to reach a consensus about a scalable measurement of
the constructs and relationships within the model. In
that regard, card sorting methodology was employed
to identify theories and constructs to be included in
the study. Four experts were employed for the sorting.
They were academic professionals experienced in the
field of technology acceptance and behavioural the-
ories at graduate level of knowledge. Theory and con-
struct lists were given to experts, and the concept of
the study was explained. These constructs, definitions
and references are given in Table 1.
Figure 1 illustrates the proposed model. The con-
structs, ‘computer anxiety’ and ‘computer self-effi-
cacy’, were re-defined to suit the mobile platform as
‘mobile anxiety and ‘mobile self-efficacy’. Through-
out the paper, ‘mHealth application’ refers to the soft-
ware provided in online application stores (i.e.
Google Play and iOS App Store), which were used
by physicians for professional purposes, such as in
diagnosis/decision making process or management.
Sezgin et al: Understanding the perception towards using mHealth applications in practice 3
Hypotheses of the conceptual model were for-
mulated with respect to the relationships among
the constructs. They were categorized under the
research questions. Performance expectancy,
Effort expectancy and the other constructs in rela-
tion to Behavioral intention were categorized
under RQ1. Constructs influencing Performance
Expectancy and Effort Expectancy were categor-
ized under RQ2.
RQ1: ‘‘What are the factors influencing physicians’
perception to use mHealth applications?’’
Performance Expectancy and Effort Expectancy. Perfor-
mance Expectancy (PE) investigates users’ attitudes
to identify the relationship between ‘‘job perfor-
mance’’ and ‘‘using a technology’’. Effort Expectancy
was first used in UTAUT, as the successor of per-
ceived ease of use of TAM (Davis 1989). This con-
struct was used to explain relations between user
Table 1. Constructs, definitions and references.
Constructs and
Abbreviations
Theories of
Constructs Definition References
Behavioral intention (BI) UTAUT ‘‘the degree to which a person has formulated
conscious plans to perform or not perform
some specified future behavior’’
(Venkatesh et al., 2003)
Effort expectancy (EE) UTAUT ‘‘the degree of ease associated with the use of
the system.’’
(Kijsanayotin et al., 2009;
Venkatesh et al., 2003)
Performance
expectancy (PE)
UTAUT ‘‘the degree to which an individual believes
that using the system will help him or her to
attain gains in job performance.’’
(Kijsanayotin et al., 2009;
Venkatesh et al., 2003)
Social influence (SI) TPB/UTAUT ‘‘the degree to which an individual perceives
that important others believe he or she
should use the new system’’
(Venkatesh et al., 2003; Ajzen,
1991; Kijsanayotin et al.,
2009)
Habit (HB) UTAUT2 ‘‘constitutes the level of routinization of
behavior, i.e. the frequency of its
occurrence’’
(Gagnon et al., 2003; Venkatesh
et al., 2012)
Personal innovativeness
in the domain of
IT (PI)
IDT ‘‘the willingness of an individual to try out any
new IT, plays an important role in
determining the outcomes of user
acceptance of technology’’
(Yi et al., 2006; Hung et al., 2012;
Rogers, 1995; Agarwal and
Prasad, 1998)
Result demonstrability
(RD)
TAM2 ‘‘the extent to which the tangible results of
using an innovation can be observable and
communicable’’
(Yi et al., 2006; Venkatesh and
Davis, 2000)
Compatibility (CO) IDT ‘‘the degree to which an innovation is
perceived as being consistent with the
existing practices, values, needs and
experiences of the health care professional’’
(Schaper and Pervan, 2007; Wu
et al., 2007; Chen et al., 2010;
Rogers, 1995)
Computer self- efficacy TAM3 ‘‘the degree to which an individual believe that
he or she has the ability to perform specific
task/job using computer’’
(Schaper and Pervan, 2007;
Aggelidis and Chatzoglou,
2009; Venkatesh and Bala,
2008)
Computer anxiety TAM3 ‘‘the degree of an individual’s apprehension, or
even fear, when she/he is faced with the
possibility of using computers’’
(Schaper and Pervan, 2007;
Aggelidis and Chatzoglou, 2009;
Venkatesh and Bala, 2008)
Technical support and
training (TT)
UTAUT ‘‘the technical support and the amount of
training provided by individuals of
knowledge’’
(Aggelidis and Chatzoglou, 2009;
Wu et al., 2007; Venkatesh
et al., 2003)
Perceived service
availability (PS)
UTAUT ‘‘the degree to which an innovation is
perceived as being able to support pervasive
and timely usage’’
(Wu et al., 2011; Venkatesh
et al., 2003)
4Information Development
attitudes and their perception about easiness towards a
technology (Venkatesh et al., 2003). In the literature,
PE and EE were used as major factors to explain user
behaviors (Kijsanayotin et al., 2009; Venkatesh et al.,
2003; Holden and Karsh 2010).
H1. Performance expectancy will positively affect beha-
vioral intention of health professionals.
H2. Effort expectancy will positively affect behavioral
intention of health professionals.
Constructs influencing Behavioral Intention. On the other
side, additional constructs were proposed in the liter-
ature to explain user behavior. In that regard, Social
Influence (SI) was used for investigating influence of
social environment on user perceptions ( Kijsanayotin
et al., 2009). Considering the trend in mobile health-
care, compatibility with the existing healthcare tech-
nologies could affect the intention to use. Thus,
compatibility was used for assessing if the technol-
ogy was perceived consistent with the current prac-
tices (Wu et al., 2007). Similarly, training to be
competent in technology use (Aggelidis and Chatzo-
glou, 2009) and availability of mobile services for
timely use (Wu et al., 2011) were influential factors
in intention to use.
H3. Social influence will positively affect behavioral
intention of health professionals.
H4. Compatibility will positively affect behavioral
intention of health professionals.
H5. Technical support and Training will have a direct
effect on behavioral intention of health professionals.
H6. Perceived service availability will positively affect
behavioral intention of health professionals.
In healthcare, routinization and high frequency of
repetition in routine tasks could lead to habitual beha-
viors (Gagnon et al., 2003). Thus, the current state of
mobile use motivates to investigate the relationship
between habit and behavioral intention in mobile
health applications. On the other hand, healthcare pro-
viders’ apprehension in using mobile technologies
(mobile anxiety) (Schaper and Pervan, 2007), their per-
ceived ability to perform specific tasks using mobile
technologies (mobile self-efficacy) (Aggelidis and
Chatzoglou, 2009), and their willingness to try new
mobile technologies (Personal innovativeness) (Hung
et al., 2012) would impact the behavioral intention.
H7. Habit will positively affect behavioral intention of
health professionals.
H8. Mobile anxiety will negatively affect behavioral
intention of health professionals.
Figure 1. Proposed model for M-TAM.
Sezgin et al: Understanding the perception towards using mHealth applications in practice 5
H9. Mobile self-efficacy will positively affect beha-
vioral intention of health professionals.
H10. Personal innovativeness will positively affect
behavioral intention of health professionals.
RQ2: ‘‘What are the relationships among the factors
influencing the perception to use mHealth
applications?’’
Constructs influencing Performance Expectancy and Effort
Expectancy. Literature suggested additional constructs
to investigate physicians’ intention to use healthcare
technologies. Mobile anxiety, Self-efficacy, Personal
Innovativeness, Habit, Perceived Service Availabil-
ity, Result Demonstrability, Technical Training and
Support and Compatibility were the prior constructs
included to the study. The hypotheses were formu-
lated as the following:
H11. Mobile anxiety will have a direct effect on effort
expectancy.
H12. Mobile self-efficacy will have a direct effect on
effort expectancy.
H13. Personal innovativeness will have a direct effect on
effort expectancy.
H14. Habit will have a direct effect on effort expectancy.
H15. Personal innovativeness will have a direct effect on
performance expectancy.
H16. Perceived service availability will have a direct
effect on performance expectancy.
H17. Perceived service availability will have a direct
effect on effort expectancy.
H18. Result Demonstrability will have a direct effect on
effort expectancy.
H19. Result Demonstrability will have a direct effect on
performance expectancy.
H20. Technical support and Training will have a direct
effect on performance expectancy.
H21. Technical support and Training will have a direct
effect on effort expectancy.
H22. Compatibility will have a direct effect on perfor-
mance expectancy.
H23. Compatibility will have a direct effect on effort
expectancy
Methodology
Target sample and data collection
Convenience sampling was employed in data collec-
tion. The data was collected using an online survey
tool (www.qualtrics.com). The target sample was the
physicians (i.e. general practitioners and specialist
medical practitioners), who are ‘actively working in
a health institution in Turkey and not using mobile
health applications’ (non-users). After the approval by
the ethical board of the university (METU), the sur-
vey was transferred to qualitrics.com and linked to an
official university webpage (www.metu.edu.tr). The
survey was announced online by posting on the social
network groups in Facebook, Twitter and Linkedin
and by sending e-mails to the physicians. In addition
to the online announcement, formal notifications were
delivered to the physicians informing them about the
aim and context of the study and the agreement
notice. They were also introduced to mHealth appli-
cations via the website. The survey was accessible for
six months (June 2015-November 2015). With respect
to the subscribers of the social network channels,
email lists and mail groups, the survey was distributed
to approximately 1031 participants (out of 135 thou-
sand registered physicians in Turkey).
Cross-sectional survey method was used as the data
collection design. Survey was conducted with a struc-
tured questionnaire. A 5-point Likert-type scale was
used to collect scalable responses and to reduce bias
(Allen and Seaman, 2007; Krosnick and Presser,
2010). The scale was named as ‘‘1: Strongly dis-
agree’’, ‘‘2: Disagree’’, ‘‘3: Neutral’’, ‘‘4: Agree’’ and
‘‘5: Strongly agree’’. Survey was categorized in three
sections. In the first section, the participants were
informed about the purpose and the scope of study
and confidentiality of their data. In the second section,
demographic data were collected, including gender,
age, education level, type of mobile device being
used, experience in mobile device use, competency
in mobile device use and the health institution type
(7 Questions). The third section included 33 closed-
ended survey questions (Table 2). The questionnaire
was verified for the integrity of the items and seman-
tic context via a pre-test study. As the result of pre-
test, the items RD3, HB2 and MA2 were removed
from the study due to the fact that these items pre-
sented significant correlations with other construct
items.
Data analysis procedure
At the first step of analysis, descriptive statistics were
used to observe normality of the data and Cronbach’s
Alpha values were calculated for internal consistency.
IBM SPSS 23 software was used for the descriptive
analysis and reliability analysis. At the second step,
6Information Development
Structural Equation Modeling was employed to test
linear and casual models. SEM was implemented in
two stages (1) measurement model and (2) structural
model, and tested using partial least squares (PLS)
approach (Ringle et al., 2005). PLS provided a
component-based approach.
Results
In total, 128 physicians completed the questionnaire
with a response rate of 12%. Considering the
nature of the study, response rate was found at the
‘‘acceptable’’ level (Tabachnick and Fidell, 2012;
Ullman and Bentler, 2003) due to limited number of
physicians invited to the survey. After the data col-
lection, data was refined by omitting incomplete and
repetitive data. Consequently, 122 out of 128
responses formed the sample of this study.
Demographics
As presented in Table 3, the largest group of partici-
pants were male and aged between 36 and 45. The
participants had different professional backgrounds.
Table 2 . Constructs, Items, Questions and Resources.
Constructs and Items
BI BI1 I intend to use the mHealth applications.
BI2 I predict I will use mHealth applications in the next 3 months
BI3 I plan to use mHealth applications in the next 3 months
EE EE1 My interaction with mHealth applications would be clear and understandable.
EE2 It would be easy for me to become skillful at using the mHealth applications.
EE3 I would find the mHealth applications easy to use.
PE PE1 I would find mHealth applications useful in my job
PE2 Using the mHealth applications increases my productivity
PE3 Using the mHealth applications enables me to accomplish tasks more quickly
SI SI1 People who influence my behavior think that I should use the mHealth applications.
SI2 People who are important to me think that I should use the mHealth applications.
SI3 The senior health administration has been helpful in the use of the mHealth applications.
HB HB1 I frequently use mHealth applications during my life.
HB2 I feel like I must use mHealth applications.
HB3 The use of mHealth applications has become a habit for me.
PI PI1 If I heard about a new information technology, I would look for ways to experiment with it
PI2 Among my peers, I am usually the first to try out new information technologies
PI3 I like to experiment with new information technologies
RD RD1 I have no difficulty telling others about the results of using mHealth applications.
RD2 The results of using mHealth applications are apparent to me
RD3 I would have difficulty telling others about the results of using mHealth applications
MS MS1 I could complete the job using mHealth applications if there was no one around to tell me what to do as I go
MS2 I could complete the job using mHealth applications if I had never used a system like it before
MS3 I could complete the job using mHealth applications if I had used similar system before this one to do the same
job
MA MA1 The mobile applications is somewhat intimidating and wrong to me
MA2 I hesitate to use the mHealth applications for fear of making mistakes that I cannot correct
MA3 I feel apprehensive about using the mHealth applications.
TT TT1 Specialized instruction and education concerning use of mHealth applications is available to me
TT2 A specific person (or group) is available for assistance with the difficulties using mHealth applications
TT3 Specialized programs or consultant about training are available to me
PS PS1 I would be able to use mHealth applications at any time, from anywhere.
PS2 I would find mHealth applications easily accessible and portable.
PS3 mHealth applications would be available to use whenever I need it
CO CO1 Using mHealth applications is compatible with most aspects of my work
CO2 Using mHealth applications fits well with the way I like to work
CO3 Using mHealth applications fits into my work style
Sezgin et al: Understanding the perception towards using mHealth applications in practice 7
The specialist medical practitioners (79%)mostly
consisted of participants specialized at Pulmonology,
Primary care, Ophthalmology, Anesthesia, Internal
medicine, Pediatrics, Cardiology and Gynecology.
General practitioners (21%) were physicians who do
not hold a specialist degree. (Table 3). Many respon-
dents prefer both tablet PC and smart phone but smart
phones were the primary mobile devices in use. Expe-
rience in mobile device use clustered around ‘‘1 to 5
years’’. Eighty percent of participants reported them-
selves at ‘‘good’’ and ‘‘moderate’’ levels of compe-
tency in mobile device use. Public hospitals and
Training and Research hospitals were the majority
of health institutions that participants were currently
working. Survey logs presented that participants were
mainly from 3 biggest cities of Turkey (Ankara
(37%), Istanbul (25%) and Izmir (22%).
Descriptives
Normality analysis has been completed to resolve the
model testing method (Tabachnick and Fidell, 2012).
The mean values were around 3, and the constructs had
low standard deviations (between 0.4 and 0.7). Posi-
tive Skewness and positive Kurtosis were observed in
the data. In addition to that, the data was found to be in
the tolerable interval for non-severe violation
(between þ1.5 and -1.5). Thus, in the following stages,
multivariate normality was assumed, and Structural
Equation Modeling procedures were followed (Kline,
2010). In addition to that, considering the sample size,
Shapiro-Wilk test was completed to test normality
(Ghasemi and Zahediasl, 2012). It was observed that
the data is not perfectly normally distributed (p<0.05)
(Table 4). Missing data analysis was conducted to
observe the missing data relationships, and list-wise
deletion approach was used to handle missing data.
The overall reliability of the constructs was found sig-
nificantly reliable at 0.825, and the constructs were
found reliable with Alpha values at the acceptable
level (>0.70) (Steel et al., 1997).
Measurement model
Construct validity was tested via convergent validity
and discriminant validity in order to provide evidence
that expected relations were met and no unexpected
relations have occurred. In that regard, the convergent
validity test was conducted employing Fornell and
Larcker’s (1981) procedure. The procedure involved
analyses of item reliability, composite reliability and
average variance extracted (AVE). Item reliability
was tested by extracting factor-loading values for
each valid item. The square values of item-loadings
values were expected to be above the threshold (0.4)
in order to meet minimum requirements for the item
reliability (Hair, 2009). After that, composite reliabil-
ity test was completed. According to Nunnally and
Bernstein (1994), composite reliability was expected
to be above 0.60. The constructs met the requirement
with the minimum composite reliability value of 0.77.
Finally, convergent validity was tested by extracting
AVE. AVE values were expected to be equal to or
more than 0.50 for each construct (Segars, 1997).
AVE values of each construct met the requirement
with the minimum value of 0.529. Table 5 provides
details about the convergent validity.
Next, discriminant validity was tested. Discrimi-
nant validity helps to observe divergence within con-
structs in order to provide relational indications
(Tabachnick and Fidell, 2012). The procedure was
implemented by comparing square roots of AVE val-
ues and correlation values of constructs. Literature
suggests that the discriminant validity is ensured if
the square root of AVE value is greater than all the
correlation values for each construct (Hair, 2009). As
presented in Table 6, discriminant validity has been
met for both datasets.
Construct validity was confirmed after convergent
and discriminant validity tests, and the model was
found suitable for structural model. Here, it should be
noted that Goodness-of-Fit measure for PLS-SEM was
omitted in this study due to its measure being unsuita-
ble for identifying latent impact of the models, and
measures of the model’s predictive capabilities are
found to be more profound to assess model quality
(F. Hair Jr et al., 2014; Henseler and Sarstedt, 2012)
Structural model
In this study, reflective measurement scale was used
in the PLS testing procedure due to having highly
correlated and interchangeable items for each variable
(Hair, 2009; Petter et al., 2007). After the model def-
inition, PLS algorithm was implemented with maxi-
mum iteration set to 300. For the stability of
estimation, the algorithm was expected to converge
before reaching to the maximum number of iterations
(Wong, 2013). In that regard, the data converged in 8
iterations, and the estimations remained in good scale.
In addition to that, bootstrapping was conducted with
5000 resampling in order to generate an approximate
estimation for the normality of the data. The results of
8Information Development
Table 3. Demographic characteristics.
Categories Percentages Number of responses
Gender
Female 43% 52
Male 57% 70
Age
25-35 22% 27
36-45 44% 54
46-65 34% 41
Type of physicians*
General practitioners 21% 26
Specialist medical practitioners 79% 96
Pulmonology 13% 12
Primary care 11% 11
Ophthalmology 11% 11
Anesthesia 9% 9
Internal medicine 9% 9
Pediatrics 8% 8
Cardiology 8% 8
Gynecology 8% 8
Psychology 3% 3
Emergency 3% 3
Genetics 3% 3
Pathology 3% 3
Hematology 2% 2
Radiology 2% 2
Otorhinolaryngology 2% 2
General surgery 2% 2
Mobile device preferences
Smart Phone 97% 118
Tablet PC 60% 73
Experience in mobile device use
None 1% 1
Less than 1 year 9% 11
1-5 years 73% 89
6-10 years 8% 10
More than 10 years 9% 11
Perceived competency in mobile device use
Excellent 14% 17
Good 47% 58
Moderate 33% 40
Bad 6% 7
How can you define the type of your health institution?
Public hospital 37% 45
Training and research hospital 34% 42
Health research center 8% 10
Community clinic 5% 6
Private hospital 15% 18
On-site medical services 1% 1
*Specialistmedicalpractitionerspracticeonparticulardiseasecategoriesortypesoftreatment.Generalpractitionershavescopefor
general medical practice for individuals, families, and communities.
Sezgin et al: Understanding the perception towards using mHealth applications in practice 9
Table 4. Descriptive Statistics.
N Minimum Maximum Mean Std. Deviation Skewness Kurtosis Shapiro-Wilk
Cronbach’s
alpha values
BI 122 1,33 5,00 2,9118 ,72353 ,632 ,326 0,000 ,811
EE 122 2,00 5,00 2,9620 ,45328 1,471 1,103 0,000 ,797
PE 122 1,33 5,00 3,5744 ,56091 -,550 1,443 0,000 ,814
MA 122 1,00 5,00 2,3975 ,71564 1,021 1,293 0,000 ,859
MS 122 2,33 5,00 3,1425 ,50085 1,096 1,326 0,000 ,795
PI 122 1,33 5,00 3,0334 ,78119 ,265 -,552 0,001 ,821
HB 122 1,00 5,00 1,8730 ,54379 1,003 1,277 0,000 ,804
SI 122 1,33 5,00 3,3612 ,73587 -,784 ,223 0,000 ,832
CO 122 1,00 5,00 3,0705 ,56935 ,258 ,867 0,000 ,789
TT 122 1,00 5,00 2,5743 ,66879 1,182 ,669 0,000 ,804
RD 122 2,00 5,00 3,0533 ,50127 ,684 1,143 0,000 ,804
PS 122 1,67 5,00 3,1422 ,56705 ,610 1,125 0,000 ,805
Table 5. Item reliability, composite reliability and AVE values.
Constructs Items Item loadings Item reliability Composite reliability AVE
BI BI1 0,705 0,498 0,876 0,705
BI2 0,908 0,824
BI3 0,890 0,793
CO CO1 0,828 0,686 0,893 0,736
CO2 0,871 0,759
CO3 0,874 0,764
EE EE1 0,677 0,458 0,770 0,529
EE2 0,806 0,650
EE3 0,693 0,480
HB HB1 0,897 0,805 0,859 0,753
HB3 0,837 0,700
MA MA1 0,911 0,831 0,854 0,746
MA3 0,814 0,662
MS MS1 0,763 0,582 0,806 0,584
MS2 0,857 0,735
MS3 0,659 0,434
PE PE1 0,825 0,680 0,836 0,630
PE2 0,807 0,652
PE3 0,747 0,558
PI PI1 0,849 0,720 0,858 0,669
PI2 0,769 0,592
PI3 0,833 0,694
PS PS1 0,779 0,606 0,781 0,543
PS2 0,681 0,464
PS3 0,748 0,560
RD RD1 0,806 0,649 0,776 0,634
RD2 0,787 0,619
SI SI1 0,908 0,825 0,842 0,643
SI2 0,783 0,613
SI3 0,700 0,490
TT TT1 0,818 0,670 0,868 0,686
TT2 0,824 0,679
TT3 0,842 0,708
10 Information Development
the PLS test (path coefficients) and bootstrapping
(t-statistics) are presented in Table 7 with the signifi-
cance levels. Here, the path coefficients were
expected to be larger than 1.0, and t-statistics were
expected to be larger than 1.96 at p<0.05 (Wong,
2013; F. Hair Jr et al., 2014). Table 7 additionally
presents multicollinearity results which were found
at the acceptable level with variance inflation factor
(VIF) under 5.0 (Grewal et al., 2004). As a result, 10
of the 23 hypotheses of M-TAM were approved.
Test results provided that effort expectancy
(b¼0.215, p < 0.05), mobile anxiety (b¼-0.105,
p < 0.05), perceived service availability (b¼0.409,
p<0.001)andtechnicalsupportandtraining
(b¼-0.182, p < 0.05) had a significant influence on
behavioral intention. In addition to that, compatibility
had influence on effort expectancy (b¼0.204,
p < 0.05) and performance expectancy (b¼0.504,
p<0.001).Moreover,habit(b¼0.183, p < 0.05),
mobile anxiety (b¼-0.115, p < 0.05), mobile self-
efficacy (b¼0.242, p < 0.05), and result demonstrabil-
ity (b¼0.280, p < 0.05) had significant influence on
effort expectancy. However, compatibility, habit,
mobile self-efficacy, performance expectancy, per-
ceived innovativeness and social influence were
found to have no influence behavioral intention. The
remaining 13 hypotheses were not supported due to
non-significant path coefficient values. In the bottom
line, the determinants of behavioral intention (EE,
MA, PS and TT) accounted for 51%of total variance
explained for intention to use mobile health
applications. In addition to that, determinants of effort
expectancy explained 61%of variance, and the deter-
minants of performance expectancy explained 28%of
variance (Figure 2). Mediating effects were also
investigated by observing total effects in bootstrap-
ping, and no significant effect was found.
Discussion
The study results revealed that, when compared with the
actual users, different factors are effective in the percep-
tion of non-user physicians towards mHealth applica-
tions. The influence of Effort Expectancy (EE),
Performance Expectancy (PE), Social Influence (SI)
and Personal Innovativeness (PI) on the perception of
using mobile technologies have been previously
demonstrated by Lu et al. (2005). Furthermore, the use
of TAM, TPB, UTAUT and IDT theories proved that an
integrated model would explain a substantial amount of
variance for physicians’ perception (0.51). Thus, using
the proposed model, M-TAM, this study contributes to
the literature providing an alternative explanation and
additional factors outlining the perception of use.
The results showed that the factors influencing
physicians’ perception to use mHealth applications
were explained by the constructs of EE, Perceived
Service Ability (PS), Mobile Anxiety (MA) and Tech-
nical Support and Training (TT). Among others, EE
stood out as one of the major indicators in explaining
user intentions towards technology use. Since its first
formulation by Venkatesh (2003), the impact of EE
on Behavioral Intention (BI) has been proved in many
studies (Sezgin and O
¨zkan-Yildirim, 2014; Holden
and Karsh, 2010; Hsiao et al., 2015). The expected
findings regarding to the relationship between EE and
BI were proved, and EE significantly influenced per-
ception of behavioral intention to use mHealth appli-
cations (H2). However, this study unveiled that (EE)
ease of use was not only influential on the actual
Table 6. Discriminant validity.
BI CO EE HB MA MS PE PI PS RD SI TT
BI 0,840
CO 0,493 0,858
EE 0,528 0,666 0,728
HB 0,284 0,568 0,522 0,868
MA -0,187 0,009 -0,060 0,217 0,864
MS 0,562 0,636 0,689 0,387 -0,101 0,764
PE 0,347 0,512 0,472 0,306 -0,202 0,425 0,794
PI 0,280 0,365 0,451 0,297 0,037 0,570 0,117 0,818
PS 0,616 0,474 0,475 0,284 -0,119 0,636 0,311 0,397 0,737
RD 0,367 0,554 0,635 0,394 0,117 0,596 0,283 0,277 0,403 0,796
SI 0,095 0,345 0,170 0,206 0,025 0,145 0,435 -0,056 0,213 0,051 0,802
TT 0,188 0,546 0,509 0,569 0,070 0,440 0,260 0,349 0,288 0,480 0,235 0,828
Sezgin et al: Understanding the perception towards using mHealth applications in practice 11
users, but also non-user physicians would perceive
that their intentions could be influenced by the ease
of use. Thus, for the non-user physicians, the benefits
of mobile health applications can be regarded as per-
ceivable and substantial (Kijsanayotin et al., 2009;
Chang et al., 2007), and there was a certain level of
awareness of the technology. In addition to that, PS
was another factor significantly influencing BI (H6).
Here, the study investigated if the intention to use
would be affected by perception of mobile health
applications supporting pervasive and timely use.
Findings provided that the time and location con-
straints were not further considered as limitations to
fulfil physicians’ tasks. This result was consistent
with previous arguments in UTAUT about the effect
of perceived service availability (facilitating condi-
tions in UTAUT) while explaining the use of technol-
ogy (Venkatesh et al., 2003). Due to the fact that
dissemination of mHealth technologies was higher
in patient healthcare, the importance of mobility in
patients would have been effective in perception of
physicians towards significance of service availability
(Wu et al., 2011; Hong and Tam, 2006). On the other
side, BI was negatively influenced by mobile anxiety
(H8). Thus, it was underlined that, unlike Venkatesh’s
(2003) findings, the apprehension, intimidation and
hesitation towards using mHealth applications would
be influential in perception of actual use. The expec-
tation was that anxiety would be a predictive factor in
the perception of use, since the participants were non-
user physicians, and a certain level of reluctance was
acceptable. However, it can be claimed that mobile
anxiety would be a result of lack in fulfilment of other
factors in the model, such as self-efficacy or service
availability (Schaper and Pervan, 2007; Aggelidis and
Chatzoglou, 2009). Similar to MA, TT had signifi-
cantly negative effect on BI. This was an unexpected
result considering that the physicians would need
technical support during the stage. However, the com-
mon perception about the ease of mobile application
use would be the effective input in the reverse impact
of TT (Wu et al., 2007). In addition to that, the con-
cept of technical support and training in mobile appli-
cation use has not been widely practiced in the field of
healthcare training in Turkey. Thus, conceptualiza-
tion of TT would be relatively hard for the partici-
pants. As a result, negative impact towards using
mHealth applications was understandable.
Table 7. Hypotheses testing.
Hypotheses Path Coefficients t-Statistics p values VIF Status
PS->BI 0,409** 4,058 0,000 1,769 Approved
EE -> BI 0,215* 2,191 0,048 2,591
MA -> BI -0,105* 1,973 0,049 1,195
TT->BI -0,182* 1,993 0,049 1,775
CO -> PE 0,504** 3,775 0,000 1,891
RD->EE 0,280* 2,556 0,011 1,886
MS -> EE 0,242* 2,197 0,028 3,113
CO -> EE 0,204* 2,18 0,029 2,316
HB -> EE 0,183* 2,114 0,035 1,836
MA -> EE -0,115* 1,99 0,044 1,164
CO -> BI 0,189 1,445 0,148 2,666 Not Supported
HB -> BI 0,061 0,864 0,388 1,927
MS -> BI 0,129 1,257 0,209 3,108
PE -> BI 0,025 0,39 0,697 1,776
PI->BI -0,081 1,123 0,261 1,635
SI -> BI -0,095 1,097 0,273 1,404
PI->EE 0,104 1,343 0,179 1,58
PI->PE -0,11 1,368 0,171 1,292
PS->EE -0,002 0,046 0,963 1,729
PS->PE 0,123 1,412 0,158 1,442
RD->PE -0,013 0,146 0,884 1,599
TT->EE 0,026 0,425 0,671 1,800
TT->PE -0,007 0,097 0,923 1,577
*p<0.05,**p< 0.001
12 Information Development
On the other hand, the relationships among the
factors influencing the perception to use of mHealth
applications (RQ2) were explained by the remaining
hypotheses in the study. One of the significant rela-
tions identified was between Compatibility and PE
(H22). In other words, the perceptions of physicians
about the consistency of mHealth application with the
healthcare practices, needs and experiences were
found significant to affect beliefs towards the
mHealth’s benefits at job performance (Wu et al.,
2007; Chen et al., 2010). It was argued that higher
compatibility results in higher success rate in mobile
health acceptance (Wu et al., 2007). In parallel, the
study revealed that the perception of compatibility
had a similar effect on physicians—indicating that
there exists knowledge and concepts regarding
mHealth applications and their practical use. More-
over, MA (H11), MS (H12), HB (H14), RD (H18) and
CO (H23) were found to have significant relationships
with EE, which also indirectly affects BI. Mobile
anxiety demonstrated that the perception about the
ease of use of mHealth applications would be influ-
enced by anxiety towards the technology. In the liter-
ature, there are studies that anxiety significantly
affected the use of technology (Schaper and Pervan,
2007) and others where anxiety did not affect at all
(Aggelidis and Chatzoglou, 2009). However, this
study revealed that anxiety in use of mobile health
technologies has an undeniable influence in physi-
cians’ perception. Considering the significance of
direct and indirect relation of MA and BI, the physi-
cians’ apprehension and intimidation in mHealth
application use has been found to be critical in their
perceived ease of use, and eventually, intention to use.
Following that, H12 provided another finding in
regards to MS. Physicians demonstrated that their
Figure 2. Path analysis.
Sezgin et al: Understanding the perception towards using mHealth applications in practice 13
individual beliefs about their abilities to use mHealth
applications were related to the ease of use of the
technology. As Chen et al. (2010) stated, healthcare
providers may have high level of intention to use the
technology if they think positive about their mobile
technology skills. Thus, the indirect impact of MS on
BI over EE may indicate that physicians’ perceptions
about their skills to use the technology have effect on
their perception of actual use. On the other side, habit
provided a new perspective. Gagnon et al. (2003)
argued that habit was not an effective predictor of use
considering relations among frequency of health tech-
nology use in the past and future. Unlikely, the find-
ings unveiled the influence of habit on ease of use: the
physicians’ beliefs would be significantly influenced
by their habits of using mobile applications in terms
of their perception of ease of use of mHealth applica-
tions. RD was another significant factor influencing
EE. Yi et al. (2006) stated that if a technology helps
users to reach job relevant results that contribute to
their tasks, perceptions of ease of use are significantly
affected. Similarly, physicians’ perceptions about
ease of use are affected by their degree of beliefs
about communicable and observable results of using
mHealth applications. Similar to CO and PE relation,
CO demonstrated that perception of ease of use of
mHealth is significantly affected by the perception
about consistency of mHealth application with the
healthcare practices, needs and experiences.
However, there have been thirteen other hypoth-
eses not supported in this study. Even though the liter-
ature and expert opinions were used in identifying and
testing all the factor relations, it is the fact that the
majority of hypotheses were rejected. However, they
contributed to the literature by supporting or not sup-
porting the current evidence regarding healthcare tech-
nology use. Unlike already proven relations in the
literature i.e. CO- BI (Chen et al., 2010), PI-BI (Lu
et al., 2005), PI-EE (Yi et al., 2006), PI-PE (Kummer
et al., 2013), PS-PE (Wu et al., 2011), RD- PE (Yi et al.,
2006), MS-BI (Chen et al., 2010) and PE-BI (Venkatesh
et al., 2003); findings of this study did not support these
hypotheses. However, consistent with the literature, the
relations of HB-BI (Gagnon et al., 2003), PS-EE (Wu
et al., 2011), SI-BI (Yu et al., 2009), TT-EE and TT-PE
(Wu et al., 2007) were considered insignificant.
Practical implications
From the practical point of view, lack of using current
technologies can be arguedasalossinresources.
Even though there is an increase in investment of
healthcare technologies (Manyika et al., 2013) and
mHealth developments (Aitken and Gauntlett,
2013), international reports show that the use of
mobile services in healthcare has not reached an
effective state (Deloitte, 2013; Organisation for Eco-
nomic Co-Operation and Development, 2015; PwC
Health Research Institute, 2014). Thus, encouraging
the potential users (i.e. physicians) to benefit from the
technology would enhance health services. In that
regard, the study proposes several implications.
The study findings revealed that there are a number
of issues needed to be identified in order to increase
the use of mHealth applications by the physicians.
First, it was observed that although there is an aware-
ness of mHealth applications, there are lacks in incen-
tives and assistance for physicians. The literature
suggested that awareness of technology is an impor-
tant step in technology use (Chang et al., 2007), and
technical support and training are important as a driver
of mHealth use (Wu et al., 2005). Here, these would be
considered as key indicators when developing and
disseminating the use of the mHealth applications
(Kijsanayotin et al., 2009). Additionally, anxiety was
another key indicator observed to influence the per-
ception of use. However, the lack in use of mHealth
applications can be resulted from anxiety as well as
other subtle reasons. Hale et al. (2015) suggested that
healthcare providers have trust issues towards mobile
applications. Furthermore, Ur-Rehman and Ramzy
(2004) argued that time constraints, lack of skills and
lack of awareness would be effective in technology use.
Managerial implications
The beliefs about weakened patient-doctor relations,
increase of workloads, threat of dangerous applica-
tions and challenges to use technology were reported
as obstacles in healthcare technology use (Lin et al.,
2012). In that regard, the managers and developers
should consider personal anxieties and beliefs
towards mHealth application use. On the other side,
age and experience in using mobile device, personal
competency and type of institution would be other key
mediating elements in physicians’ perceptions to use
(Venkatesh et al., 2003). Thus, personalized or
profession-specific applications and government/
institution incentives to use mHealth would be bene-
ficial to disseminate the technology.
In the bottom line, one of the suggestions may be to
promote collaborations among patients and
14 Information Development
physicians. The policy makers should focus on pro-
viding standards in mHealth applications (Becker
et al., 2014). For developing countries, infrastructure
and regulations are suggested to be developed (Varsh-
ney, 2014) as well as taking action to reduce techno-
logical and sociocultural barriers (O’Connor et al.,
2016). Furthermore, interventions to education and
communications among physicians, management sup-
port and clinical diagnosis assistance would be useful
for developing countries (Free et al., 2013). Regard-
ing the benefits of mHealth use, such as increase in
personal time, communication and monitoring
enhancements (Steinhubl et al., 2013), it is evident
that enabling physicians to use mHealth applications
would contribute to both healthcare practice and qual-
ity of services.
Limitations
There are a number of limitations in the study that the
readers should take into account while interpreting the
results and findings. First, there is a need in the liter-
ature for the studies about the perception of health
information technology use and acceptance. Thus, the
results were interpreted considering the current liter-
ature regarding technology use of healthcare provi-
ders. Moreover, the study design has some
constraints. The study employed a cross-sectional sur-
vey on a specific set of participants in Turkey, which
may affect generalization of the results due to several
factors, such as timing, cultural impact or sample
characteristics. Furthermore, participation to the
study was on a voluntary-basis, so self-selection
biases were possible. Additionally, online survey and
quantitative approach limit capturing all relevant data
due to its self-reported nature. Another argument
about the limitation of the study would be the size
of the sample. Even though the literature approves
that the sample size was in acceptable limits (Good-
hue et al., 2012), it can be argued that the study had
limited data to represent the population. Finally, the
study was able to explain behavioral intention at 51%,
and the model was unable to predict remaining factors
in explaining perceptions of using mHealth
applications.
Conclusion
This study focused on the perception of mobile
health application use by physicians who are not
using mobile health applications in practice. This
approach brought an alternative perspective to
enlighten the literature in terms of potential inten-
tions to use mHealth applications and perceptions
towards it. In that regard, the study brought not only
non-user physicians’ perspective, but also it is the
only study, to our knowledge, investigating percep-
tion of mHealth applications acceptance by non-user
physicians.
A Mobile Technology Acceptance Model
(M-TAM) was proposed to assess physicians’ per-
ception to use mobile health applications. A cross-
sectional survey was designed based on the model,
and it was conducted on 122 physicians in Turkey.
The data was analyzed employing confirmatory fac-
tor analysis (CFA) and structural equation modeling
(SEM). Significant relations were identified, which
depicted implications for mHealth application use.
Predictive factors were discussed in explaining per-
ception to use.
The study contributed to the literature in the fol-
lowing aspects: (1) a new model was proposed to
explain physicians’ perceptions, (2) a new sample
of the population was provided, and (3) a unique
model (M-TAM) and approach (i.e. survey study
on the prospective mHealth users) has been pro-
posed. M-TAM proved its potential as a model to
be employed for acceptance of mHealth applications
in healthcare. In addition to that, this paper has
reported the first scholar research conducted in Tur-
key towards assessing physicians’ acceptance of
mobile health technology.
Further studies on acceptance of mHealth applica-
tions by healthcare providers are required to provide
insight about factors influencing the use of mHealth
by different healthcare professions. In that, this study
acts as an initiator collecting data from physicians
who are using mHealth applications in practice, and
providing information depicting differences among
user and non-user physicians. Finally, a longitudinal
survey design would be beneficial to increase the pre-
dictive value of the model.
References
Adibi S (2015) Mobile Health: A Technology Road Map.
Switzerland: Springer International Publishing.
Agarwal R and Prasad J (1998) A conceptual and opera-
tional definition of personal innovativeness in the
domain of information technology. Information Systems
Research 9(2): 204–215.
Aggelidis VP and Chatzoglou PD (2009) Using a modified
technology acceptance model in hospitals. International
Journal of Medical Informatics 78: 115–126.
Sezgin et al: Understanding the perception towards using mHealth applications in practice 15
Aitken M and Gauntlett C (2013) Patient Apps for
Improved Healthcare: From novelty to mainstream,
New Jersey: Parsippany.
Ajzen I (1991) The Theory of Planned Behavior. Organi-
zational Behavior and Human Decision Processes 50:
179–211.
Allen I and Seaman C (2007) Likert scales and data anal-
yses. Quality Progress 7: 64–65.
Anderson C, Henner T and Burkey J (2013) Tablet com-
puters in support of rural and frontier clinical practice.
International Journal of Medical Informatics 82(11):
1046–1058.
Atluri V, Rao S, Rajah T, et al. (2015) Unlocking Digital
Health: Opportunities for the mobile value chain.New
York: McKinsey & Company.
Bagozzi RP and Warshaw PR (1992) Development and test
of a theory of technological learning and usage. Human
Relations 45(7): 659–686.
Becker S, Miron-Shatz T, Schumacher N, et al. (2014)
mHealth 2.0: Experiences, possibilities, and perspec-
tives. JMIR mHealth and uHealth 2(2): e24.
Chang IC, Hwang HG, Hung WF, et al. (2007) Physicians’
acceptance of pharmacokinetics-based clinical decision
support systems. Expert Systems with Applications
33(2): 296–303.
Chaudhry B, Wang J, Wu S, et al. (2006) Systematic
review: Impact of health information technology on
quality, efficiency, and costs of medical care. Annals
of Internal Medicine 144(10): 742–52.
Chen J, Park Y and Putzer GJ (2010) An examination of the
components that increase acceptance of smartphones
among healthcare professionals. Electronic Journal of
Health Informatics 5(2): 1–12.
Chin W (1998) The partial least squares approach to struc-
tural equation modeling. Modern Methods for Business
Research 295(2): 295–336.
Davis FD (1989) Perceived usefulness, perceived ease
of use, and user acceptance of information technol-
ogy. Management Information Systems 13(3):
319–340.
Deloitte (2013) Physician Adoption of Health Information
Technology: Implications for medical practice leaders
and business partners. Available at: http://www.
deloitte.com/view/en_US/us/Industries/health-care-pro
viders/index.htm.
Doganyigit SO
¨and Yılmaz E (2015) Mobile health appli-
cations user trends in Turkey. Journalism and Mass
Communication 5(1): 44–49.
Ducey AJ and Coovert MD (2016) Predicting tablet com-
puter use: An extended technology acceptance model.
Health Policy and Technology 5(3): 268–284.
Du
¨nnebeil S, Sunyaev A, Blohm I, et al. (2012) Determi-
nants of physicians’ technology acceptance for e-health
in ambulatory care. International Journal of Medical
Informatics 81(11): 746–760.
F. Hair Jr J, Sarstedt M, Hopkins L, et al. (2014) Partial
least squares structural equation modeling (PLS-SEM):
An emerging tool in business research. European
Business Review 26(2): 106–121.
Fiordelli M, Diviani N and Schulz PJ (2013) Mapping
mHealth research: A decade of evolution. Journal of
Medical Internet Research 15(5): e95.
Fornell C and Larcker D (1981) Evaluating structural equa-
tion models with unobservable variables and measure-
ment error. Journal of Marketing Research Feb 1:
39–50.
Free C, Phillips G, Watson L, et al. (2013) The effec-
tiveness of mobile-health technologies to improve
health care service deliveryprocesses:Asystematic
review and meta-analysis. PLoS Medicine 10(1):
e1001363.
Gagnon MP, Godin G, Gagne´C,etal.(2003)Anadaptation
of the theory of interpersonal behaviour to the study of
telemedicine adoption by physicians. International
Journal of Medical Informatics 71(2): 103–115.
Gagnon MP, Ngangue P, Payne-Gagnon J, et al. (2016) m-
Health adoption by healthcare professionals: A systema-
tic review. Journal of the American Medical Informatics
Association 23(1): 212–220.
Gartner Inc. (2012) Gartner: World-wide smartphone
sales. Available at: http://www.gartner.com/it/page.
jsp?id¼1924314.
Ghasemi A and Zahediasl S (2012) Normality tests for
statistical analysis: a guide for non-statisticians. Inter-
national Journal of Endocrinology and Metabolism
10(2): 486–489.
Goodhue DL, Lewis W and Thompson R (2012) Does PLS
have advantage for small sample size or non-normal
data? MIS Quarterly 36(3): 1–16.
Grewal R, Cote JA and Baumgartner H (2004) Multicolli-
nearity and measurement error in structural equation
models: Implications for theory testing. Marketing Sci-
ence 23(4): 519–529.
Hair JF (2009) Multivariate Data Analysis.NewJersey:
Prentice Hall.
Hale K, Capra S and Bauer J (2015) A framework to assist
health professionals in recommending high-quality apps
for supporting chronic disease self-management: Illus-
trative assessment of Type 2 diabetes apps. JMIR
mHealth and uHealth 3(3): e87.
Hampton T (2012) Recent advances in mobile technology
benefit global health, research, and care. American Med-
ical Association 307(19): 2013–2014.
Han S, Mustonen P, Seppanen M, et al. (2006) Physicians’
acceptance of mobile communication technology an
exploratory study. International Journal of Mobile
Communications 4(2): 210–230.
Henseler J and Sarstedt M (2012) Goodness-of-fit indices
for partial least squares path modeling. Computational
Statistics 28(2): 565–580.
16 Information Development
Holden RJ and Karsh BT (2010) The technology accep-
tance model: Its past and its future in health care. Jour-
nal of Biomedical Informatics 43(1): 159–172.
Hong SJ and Tam KY (2006) Understanding the adoption
of multipurpose information appliances: The case of
mobile data services. Information Systems Research
17(2): 162–179.
Hsiao CH, Tang KY and Liu JS (2015) Citation-based
analysis of literature: A case study of technology accep-
tance research. Scientometrics 105(2): 1091–1110.
Hung MC and Jen WY (2010) The adoption of mobile
health management services: An empirical study. Jour-
nal of Medical Systems 36(3): 1381–1388.
Hung SY, Ku YC and Chien JC (2012) Understanding
physicians’ acceptance of the Medline system for prac-
ticing evidence-based medicine: a decomposed TPB
model. International Journal of Medical Informatics
81(2): 130–42.
Iredale R, Hilgart J and Hayward J (2011) Patient per-
ceptions of a mobile cancer support unit in South
Wales. European Journal of Cancer Care 20(4):
555–560.
Istepanian RSH, Laxminarayan S and Pattichis CS (2010)
M-Health: Emerging mobile health systems.London:
Springer-Verlag.
Jersak LC, da Costa AC and Callegari DA (2013) A
Systematic Review on Mobile Health Care.Tech.
Report 073, Faculdade de Informa´tica PUCRS–
Brazil.
Kahn J, Yang J and Kahn J (2010) ‘‘Mobile’’ Health needs
and opportunities in developing countries. Health
Affairs 29(2): 252–258.
Kijsanayotin B, Pannarunothai S and Speedie SM (2009)
Factors influencing health information technology
adoption in Thailand’s community health centers:
Applying the UTAUT model. International Journal of
Medical Informatics 78(6): 404–416.
Kim S, Lee KH, Hwang H, et al. (2016) Analysis of the
factors influencing healthcare professionals’ adoption of
mobile electronic medical record (EMR) using the Uni-
fied Theory of Acceptance and Use of Technology
(UTAUT) in a tertiary hospital. BMC Medical Infor-
matics and Decision Making 16(12): 1.
King WR and He J (2006) A meta-analysis of the Technol-
ogy Acceptance Model. Information & Management
43(6): 740–755.
Kline RB (2010) Principles and Practice of Structural
Equation Modeling. New York: Guilford Press.
Krosnick JA and Presser S (2010) Question and question-
naire design. In: Marsen PV and Wright JD (eds.)
Handbook of Survey Research. Emerald, 263–313.
Kummer TF, Scha¨fer K and Todorova N (2013) Accep-
tance of hospital nurses toward sensor-based medication
systems: A questionnaire survey. International Journal
of Nursing Studies 50(4): 508–517.
Li J, Talaei-Khoei A, Seale H, et al. (2013) Health care
provider adoption of eHealth: Systematic literature
review. Interactive Journal of Medical Research, 2(1):
e7.
Lin C, Lin IC and Roan J (2012) Barriers to physicians’
adoption of healthcare information technology: An
empirical study on multiple hospitals. Journal of Medi-
cal Systems 36(3): 1965–1977.
Lin SP (2011) Determinants of adoption of mobile health-
care service. International Journal of Mobile Commu-
nications 9(3): 298–315.
Lin SP and Yang HY (2009) Exploring key factors in the
choice of e-health using an asthma care mobile service
model. Telemedicine Journal and e-health: the official
journal of the American Telemedicine Association
15(9): 884–890.
Lu J, Yao J and Yu C (2005) Personal innovativeness,
social influences and adoption of wireless Internet ser-
vices via mobile technology. The Journal of Strategic
Information Systems 14: 245–268.
Ludwick D and Doucette J (2009) Adopting electronic
medical records in primary care: Lessons learned from
health information systems implementation experience
in seven countries. International Journal of Medical
Informatics 78(1): 22–31.
Manyika J, Chui M, Bughin J, et al. (2013) Disruptive
Technologies: Advances that will transform life, busi-
ness, and the global economy-Executive summary.New
York: McKinsey & Company.
Moores TT (2012) Towards an integrated model of IT
acceptance in healthcare. Decision Support Systems
53(3): 507–516.
Nah F and Siau K (2005) The value of mobile applications:
Autilitycompanystudy.Communications of the ACM
48(2): 85–90.
Nunnally J and Bernstein I (1994) Psychometric Theory.
New York: McGraw-Hill.
O’Connor Y, O’Connor S, Heavin C, et al. (2016) Socio-
cultural and technological barriers across all phases of
implementation for mobile health in developing coun-
tries. In: Al-Jumeily D, Hussain A, Mallucci C and Oli-
ver C (eds.) Applied Computing in Medicine and Health.
Elsevier, 212–230.
Organisation for Economic Co-Operation and Develop-
ment. (2015) OECD Digital Economy Outlook 2015.
Available at: http://ec.europa.eu/eurostat/documents/
42577/3222224/Digitalþeconomyþoutlookþ2015/
dbdec3c6-ca38-432c-82f2-1e330d9d6a24.
Okazaki S, Castan
˜eda JA, Sanz S, et al. (2012) Factors
affecting mobile diabetes monitoring adoption
among physicians: Questionnaire study and path
model. Journal of Medical Internet Research 14(6):
e183.
Or CKL and Karsh BT (2012) A systematic review of
patient acceptance of consumer health information
Sezgin et al: Understanding the perception towards using mHealth applications in practice 17
technology. Journal of the American Medical Infor-
matics Association: JAMIA 16(4): 550–560.
Petter S, Straub D and Rai A (2007) Specifying formative
constructs in information systems research. MIS Quar-
terly Dec 1:623–656.
Piette JD, Blaya JA, Lange I, et al. (2011) Experiences in
mHealth for chronic disease management in 4 countries.
ACM Proceedings of the 4th International Symposium
on Applied Sciences in Biomedical and Communication
Technologies Oct 26: 170.
PwC Health Research Institute. (2014) Top health industry
issues of 2015. Available at: http://www.pwc.com/en_
US/us/health-industries/top-health-industry-issues/
download.jhtml.
Pynoo B, Devolder P, Duyck W, et al. (2012) Do hospital
physicians’ attitudes change during PACS implementa-
tion? A cross-sectional acceptance study. International
Journal of Medical Informatics 81(2): 88–97.
Rahimpour M, Lovell NH, Celler BG, et al. (2008)
Patients’ perceptions of a home telecare system.
International Journal of Medical Informatics 77(7):
486–98.
Ringle C, Wende S and Will A (2005) Smart-PLS Version
2.0 M3. University of Hamburg.
Ringle CM, Sarstedt M and Straub D (2012) A critical look
at the use of PLS-SEM in MIS Quarterly. MIS Quarterly
36(1): 3–14.
Rogers EM (1995) Diffusion of Innovations. New York:
Free Press.
Rogers EM and Shoemaker FF (1971) Communication of
Innovations: A cross-cultural approach.NewYork:
Free Press.
Sarker S (2003) Understanding mobile handheld device use
and adoption. Communications of the ACM 46(12):
35–40.
Schaper LK and Pervan GP (2007) ICT and OTs: a model
of information and communication technology accep-
tance and utilisation by occupational therapists. Interna-
tional Journal of Medical Informatics 76(1): 212–221.
Segars AH (1997) Assessing the unidimensionality of mea-
surement: a paradigm and illustration within the context
of information systems research. Omega 25(1):
107–121.
Sezgin E and O
¨zkan-Yildirim S (2016) A cross-sectional
investigation of acceptance of health information tech-
nology: A nationwide survey of community pharmacists
in Turkey. Research in Social and Administrative Phar-
macy 12: 949–965.
Sezgin E and O
¨zkan-Yildirim S (2014) A literature review
on attitudes of health professionals towards health infor-
mation systems: From e-Health to m-Health. Procedia
Technology 16: 1317–1326.
Siau K and Shen Z (2006) Mobile healthcare informatics.
Medical Informatics and the Internet in Medicine 31(2):
89–99.
Steel RGD, Torrie JH and Dickey DA (1997) Principles
and Procedures of Statistics: A biometrical approach.
New York: McGraw-Hill.
Steele R, Lo A, Secombe C, et al. (2009) Elderly persons’
perception and acceptance of using wireless sensor net-
works to assist healthcare. International Journal of
Medical Informatics 78(12): 788–801.
Steinhubl SR, Muse ED and Topol EJ (2013) Can mobile
health technologies transform health care? JAMA: the
journal of the American Medical Association 310(22):
2395–2396.
Sun H and Zhang P (2006) The role of moderating factors
in user technology acceptance. International Journal of
Human-Computer Studies 64(2): 53–78.
Tabachnick BG and Fidell LS (2012) Using Multivariate
Statistics. Boston: Pearson Education.
Tachakra S, Wang XH, Istepanian RSH, et al. (2003)
Mobile e-health: The unwired evolution of telemedi-
cine. Telemedicine Journal and E-Health: the official
journal of the American Telemedicine Association
9(3): 247–257.
Tezcan C (2016) Sag
˘lıg
˘aYenlikc¸i Bir Bakıs¸Ac¸ısı: Mobil
Sag
˘lık. Ankara: TUSIAD Publishing.
Turkish Statistical Institute (2015) Information and Com-
munication Technology (ICT) Usage Survey on House-
holds and Individuals. Available at: http://www.turkstat.
gov.tr/PreHaberBultenleri.do?id¼18660.
Ullman J and Bentler P (2003) Structural Equation Model-
ling. New York: John Wiley & Sons.
United States Government Publishing Office (2009) Health
Information Technology for Economic and Clinical
Health Act. Available at: http://www.gpo.gov/fdsys/
pkg/PLAW-111publ5/html/PLAW-111publ5.htm.
Ur-Rehman S and Ramzy V (2004) Awareness and use of
electronic information resources at the health sciences cen-
ter of Kuwait University. Library Review 53(3): 150–156.
Varshney U (2014) Mobile health: Four emerging themes
of research. Decision Support Systems 66: 20–35.
Ve´lez O, Okyere PB, Kanter AS, et al. (2014) A usability
study of a mobile health application for rural Ghanaian
midwives. Journal of Midwifery & Women’s Health
59(2): 184–191.
Venkatesh V, Morris MG, Davis GB, et al. (2003) User
acceptance of information technology: Toward a unified
view. MIS Quarterly 27(3): 425–478.
Venkatesh V and Bala H (2008) Technology Acceptance
Model 3 and a research agenda on interventions. Deci-
sion Sciences 39(2): 273–315.
Venkatesh V and Davis FD (2000) A theoretical extension
of the Technology Acceptance Model: Four longitudinal
field studies. Management Science 46(2): 186–204.
Venkatesh V, Thong JYL and Xu X (2012) Consumer
acceptance and use of information technology: Extend-
ing the Unified Theory of Acceptance and Use of Tech-
nology. MIS Quarterly 36(1): 157–178.
18 Information Development
Visvanathan A, Gibb AP and Brady RRW (2011) Increas-
ing clinical presence of mobile communication technol-
ogy: avoiding the pitfalls. Telemedicine Journal and e-
Health 17(8): 656–661.
Ward R (2013) The application of technology acceptance
and diffusion of innovation models in healthcare infor-
matics. Health Policy and Technology 2(4): 222–228.
Wolters Kluwer (2013) Wolters Kluwer Health 2013 Phy-
sician Outlook Survey. Available at: http://woltersk
luwer.com/binaries/content/assets/wk-health/pdf/com
pany/newsroom/white-papers/wolters-kluwer-health-
physician-study-executive-summary.pdf
Wong KK (2013) Partial least squares structural equation
modeling (PLS-SEM) techniques using SmartPLS. Mar-
keting Bulletin 24: 1–32.
World Health Organisation (2011) Global Atlas on Cardi-
ovascular Disease Prevention and Control. Available
at: http://www.who.int/cardiovascular_diseases/en/.
Wu IL, Li JY, Fu CY, et al. (2010) The acceptance of
wireless healthcare for individuals: An integrative view.
Proceedings of the 12th International Conference on
Enterprise Information Systems 5: 1–6.
Wu IL, Li JY and Fu CY (2011) The adoption of mobile
healthcare by hospital’s professionals: An integrative
perspective. Decision Support Systems 51(3):
587–596.
Wu JH, Wang SC and Lin LM (2005) What Drives mobile
health care? An empirical evaluation of technology
acceptance. Proceedings of 38th Annual Hawaii Inter-
national Conference on System Sciences 150a: 1–9.
Wu JH, Wang SC and Lin LM (2007) Mobile computing
acceptance factors in the healthcare industry: A struc-
tural equation model. International Journal of Medical
Informatics 76(1): 66–77.
Wu RC, Orr MS, Chignell M, et al. (2009) Usability of a
mobile electronic medical record prototype: A verbal
protocol analysis. Informatics for Health and Social
Care 33(2): 139–149.
Yi MY, Jackson JD, Park JS, et al. (2006) Understanding
information technology acceptance by individual pro-
fessionals: Toward an integrative view. Information &
Management 43: 350–363.
Yousafzai S, Foxall GR and Pallister JG (2007) Technol-
ogy acceptance: A meta-analysis of the TAM: Part 2.
Journal of Modelling in Management 2(3): 281–304.
Yu P, Li H and Gagnon MP (2009) Health IT acceptance
factors in long-term care facilities: A cross-sectional
survey. International Journal of Medical Informatics
78(4): 219–229.
Zailani S, Gilani MS, Nikbin D, et al. (2014) Determinants
of telemedicine acceptance in selected public hospitals
in Malaysia: Clinical perspective. Journal of Medical
Systems 38: 111.
About the authors
Emre Sezgin worked as a research/teaching assistant at
Graduate School of Informatics, Middle East Technical
University (METU), Ankara, Turkey. He earned his PhD
degree from the department of Information Systems,
METU. His research interests include IT management, IT
governance and technology acceptance studies in eHealth
domain. His recent studies were about mobile technologies
and developing framework for IT assessment in organiza-
tions. Contact: Middle East Technical University, School
of Informatics, 06800, C¸ankaya, Ankara, Turkey. E-mail
address: esezgin@metu.edu.tr
Dr. Sevgi Ozkan-Yildirim received her BA and her MA in
Engineering - Electrical and Information Sciences from the
University of Cambridge, UK and her MSc in Business
Information Systems from the University of London, UK.
She has a PhD in the area of Information Systems Evalua-
tion, with a particular focus on assessing success and effec-
tiveness of information systems. She is currently a full-time
faculty member as Associate Professor of the Department
of Information Systems, Middle East Technical University.
Her research interests include information systems evalua-
tion, e-learning evaluation, technology acceptance, and
information and communication technology management
in the public sector. Contact: Middle East Technical Uni-
versity, School of Informatics, 06800, C¸ankaya, Ankara,
Turkey. E-mail address: sevgiozk@metu.edu.tr
Dr. Soner Yildirim is a Professor of Instructional Tech-
nology at the Department of Computer Education and
Instructional Technology at Middle East Technical Univer-
sity. His research interests include web based training and
instructional design; instructional and performance tech-
nologies; electronic performance support systems; online
social networks and warranty theory. Contact: Middle East
Technical University, Department of Computer Education
and Instructional Technology, 06800, C¸ ankaya, Ankara,
Turkey. E-mail address: soner@metu.edu.tr
Sezgin et al: Understanding the perception towards using mHealth applications in practice 19
