Understanding factors influencing the adoption of ...

94 downloads 0 Views 729KB Size Report
[8] K. Ghorai, S. Jahan, P. Ray, M. Chylinski, Mobile phone including smart phone ... [23] M.R. Sarwar, Bangladesh health service delivery: innovative NGO and ...
International Journal of Medical Informatics 101 (2017) 75–84

Contents lists available at ScienceDirect

International Journal of Medical Informatics journal homepage: www.ijmijournal.com

Understanding factors influencing the adoption of mHealth by the elderly: An extension of the UTAUT model Rakibul Hoque a,∗ , Golam Sorwar b a b

Department of Management Information Systems, University of Dhaka, Bangladesh Department of Business and Tourism, Southern Cross University, Australia

a r t i c l e

i n f o

Article history: Received 19 November 2015 Received in revised form 10 January 2017 Accepted 1 February 2017 Keywords: mHealth Elderly UTAUT Developing country

a b s t r a c t Background: mHealth (mobile health) services are becoming an increasingly important form of information and communication technology (ICT) enabled delivery for healthcare, especially in low-resource environments such as developing countries like Bangladesh. Despite widespread adoption of mobile phones and the acknowledged potential of using them to improve healthcare services, the adoption and acceptance of this technology among the elderly is significantly low. However, little research has been done to draw any systematic study of the elderly’s intention to adopt mHealth services. Objective: The aim of this study was to develop a theoretical model based on the Unified Theory of Acceptance and Use of Technology (UTAUT) and then empirically test it for determining the key factors influencing elderly users’ intention to adopt and use the mHealth services. Methods: A face-to-face structured questionnaire survey method was used to collect data from nearly 300 participants of age 60 years and above from the capital city of Bangladesh. The data were analyzed using the Partial Least Squares (PLS) method, a statistical analysis technique based upon Structural Equation Modeling (SEM). Results: The study determined that performance expectancy, effort expectancy, social influence, technology anxiety, and resistance to change (p < 0.05) had a significant impact on the users’ behavioral intention to adopt mHealth services. The study, however, revealed no significant relation between the facilitating condition and the users’ behavioral intention to use the mHealth services (p > 0.05). Conclusions: This study confirms the applicability of UTAUT model in the context of mHealth services among the elderly in developing countries like Bangladesh. It provides valuable information for mHealth service providers and policy makers in understanding the adoption challenges and the issues and also provides practical guidance for the successful implementation of mHealth services. Additionally the empirical findings identify implications related to the design and development of mHealth services that influence potential users. Furthermore, due to a generic approach, the findings of this study could be easily modified to assist other developing countries in the planning and up-take of mHealth services. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Over the last decades, healthcare sectors have experienced tremendous changes in most of the countries in the world due to a rapid advancement in information and communication technology (ICT). Recent evidence suggests that mHealth (mobile health) could perhaps be a blessing brought about by ICT and is probably one of the most prominent services with noticeable effects upon the development of the healthcare sector [1–3]. According to the World

∗ Corresponding author. E-mail addresses: [email protected] (R. Hoque), [email protected] (G. Sorwar). http://dx.doi.org/10.1016/j.ijmedinf.2017.02.002 1386-5056/© 2017 Elsevier B.V. All rights reserved.

Health Organistaion (WHO), more than a quarter of the world’s countries, inlcuding Bangladesh, have a critical healthcare workforce shortage [4]. mHealth is considered to be an easy, low cost, and affordable solution to improve access to health services, especially for those with a significant shortage of healthcare resourecs [5]. The number of mobile phone users is rapidly increasing throughout the world including Bangladesh. According to the Bangladesh Telecommunication Regulatory Commission (BTRC), the number of Mobile Phone users in Bangladesh was 131.436 million at the end of September 2015, with an annual growth of 10.54% [6]. Statics from the Bangladesh Demographics and Health surveys indicate that household ownership of mobile phones increased from 32% to 78% from 2007 to 2011 [7]. This simply implies

76

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84

that there are clear opportunities to develop mHealth initiatives and interventions in developing nations, such as Bangladesh, to improve healthcare outcomes [8]. Leveraging the prevalence of mobile phones and their well-known potentiality to improve health services, various mHealth initiatives and services are already in place or are underway in developing countries like Bangladesh [9,10]. It is evident that mHealth transforms healthcare in ageing societies such as China, Algeria, Trinidad and Tobago with the provision of continuous monitoring in self-healthcare as well as supervised and assisted healthcare [3]. In 2011, WHO reported Bangladesh as one of the 15 countries using mHealth to raise health awareness [11]. A study by Akter and Ray suggests that mHealth services can save travel time (98%) and money (91%), and experience shorter wait times (97%) for users in Bangladesh [12]. It is also noted that most mHealth users (56%) had an income below the poverty line who could instantly receive health information and counseling services from health professionals at an affordable price through the provision of mHealth [13]. In Bangladesh, mHealth services emerged in mid-1999 [14]. In July 1999, the Swinfen Charitable Trust in the UK, established a mHealth link between the Centre for the Rehabilitation of the Paralyzed (CRP) in Dhaka (the capital of Bangladesh) and the Royal Navy Hospital, Haslar, UK [15]. As a result, CRP now has access to consultants from the Royal Navy Hospital with a variety of specialists agreed to provide consultations, free of cost. In 2003, the Telemedicine Reference Centre Ltd. (TRCL) introduced a mHealth program for health system based medical services, linking 200 specialists for their expert opinions to rural doctors [16]. In 2006, Grameenphone and the Telemedicine Reference Centre Ltd. introduced a unique mHealth Service titled “HealthLine Dial 789”. Under this platform, a patient can easily access health services simply by dialing “789” from his/her mobile phone [17]. Grameenphone’s 789 services reached the level of 10,000 calls per day within a short span of three years since their inception [18]. The Government of Bangladesh also has a wide range of mHealth initiatives designed to enhance healthcare services for better health outcomes for patients. These services include telemedicine services and pregnancy care advice via SMS and other health services via Mobile Phone. In May 2009, the Bangladesh Ministry of Health first introduced mHealth (Health Service through Mobile Phone) by establishing a local call center for delivering medical advice on a 24/7 basis to citizens in district and sub-district (upazila) hospitals across the country [2]. The mHealth services have facilitated ease of access to health care for all and especially for female, the elderly and the poor who can now easily access medical advice through mobile phone [19]. Telemedicine services in eight hospitals were formally inaugurated as part of the National Digital Innovation Fair on July 6, 2011. In 2012, a further 10 telemedicine centers in different hospitals were opened. Additionally, 15 telemedicine centers will be expected to be open in other district and sub-district hospitals by the end of 2016 [20]. Through these services, patients can access medical advice from doctor(s) of specialized hospitals without a need to physically visit them [21]. In March 2010, a pregnancy care advice service via SMS has been launched by the Ministry of Health to provide pregnancy related advice [2]. Non-government organizations (NGOs) have also introduced various mHealth services to patients in different areas of Bangladesh. For example, the Bangladesh Rural Advancement Committee (BRAC), a non-governmental development organization, has introduced the “Manoshi” project aimed at empowering community health workers to be more efficient in gathering patient information in real time and prioritizing them as per their clinical urgency in cases such as high-risk pregnancies. This is achieved through the provision of simple mobile phones [22]. “Aponjon” (“the close or dear one” in Bangla) is another mHealth service that provides health information and advice to expecting and/or new

parents through mobile networks and operates under the auspices of the Mobile Alliance for Maternal Action (MAMA) [2]. “Infolady” (a young woman with a mobile phone) is another service that was launched by a local not-for-profit organization Dnet. Infolady helps marginalized citizens (specifically women) access health information including advice with pregnancy, to around 300 Bangladeshi villages [23]. Despite the potential benefits and the various mHealth initiatives in place, recent studies show that the adoption of this service remains insignificant. A study by Ahmed et al. reported that many citizens were aware of mHealth services and perceived them as potentially useful, however very few had actually used them [9]. It should be noted that this study was conducted to only a small number of residents living in a slum area in Bangladesh. The study also indicated that mHealth appeared to be expensive. Elderly people in slum areas usually go to charlatan doctors, cheaper or free traditional healers and/or homeopathic doctors as an affordable alternative due to their low incomes [24]. Another study by GRM International and SIDA found that most people saw mHealth, the phone helplines, as “just a business”, expensive and as unsatisfactory [25]. Khan et al. found that the high consultation fees, the technical problems and trusting unfamiliar physicians are the main barriers to adopt mHealth services in Bangladesh [26]. Khatun et al. conducted a study on mHealth over 4915 randomly selected respondents from a remote rural district in Bangladesh [27]. They found only 5% of participants used the services on their mobile phone, but the majority of the respondents (73%) were aware of and keen to adopt mHealth in the future. All the above studies examining barriers to mHealth acceptance yield mixed results regarding mHealth acceptance, however, these studies clearly evidence that most people in Bangladesh are aware of and interested in using this technology. Ahmed et al. indicate that there is a considerable uncertainty as to the most appropriate design of mHealth initiatives with regards to the service mode and the content of the delivery of health service [9]. The above literature review evidences that mHealth is promising on its supply side (i.e., providing infrastructure) in Bangladesh, but little has been done on the demand side (i.e., adoption). It is, therefore, imperative to investigate factors influencing the adoption of mHealth services by all citizens. Among various aged users, the elderly are more susceptible to chronic diseases, physical disabilities and mental incapacities [28]. Globally, the proportion of the older population is increasing faster than any other age group due to an increased life expectancy and a decrease in birth rates [29]. In 2013, there were about 840 million people (10% of the overall population) throughout the world aged 60 years or more [30]. By 2025 and 2050, this figure is expected to reach to around 1.2 billion and 1.9 billion respectively [31]. In developing countries, where 80% of the older population live, the proportion of the elderly is expected to increase by 12% by 2025 [32]. The ageing population and the prevalence of their chronic conditions demands for an increase in both the quality and variety of health care services being offered. Therefore, health care spending in OECD countries has increased over the past decade at a faster rate than spending on all other goods and services. In 2012, OECD countries, on average, spent 9.3% of their GDP in health care, which is projected to increase significantly over the coming decades [33]. Studies have shown that healthcare services via mobile phone appear significant for elderly patients who are most in need of healthcare services on demand [34]. mHealth can assist elderly people to seek medical advice, register for appointments, access medical test results, and get post-diagnostic treatment for active prevention at their convenience [35]. Bangladesh is one of the twenty countries in the world with the largest elderly populations [36]. By 2025, Bangladesh, China, India, Indonesia and Pakistan together, will account for about 44% of the world’s total elderly population [37]. The most recent census

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84

77

Fig. 1. Research Model.

data (population census, 2011) shows that 7.4% of the total population in Bangladesh is 60 years or above [38]. The proportion of the elderly population in Bangladesh is projected to increase by 11.9% and 17% by 2035 and 2050 [39]. A study by Razzaque et al. indicates that more than 84% of the elderly in Bangladesh experienced health problems with most of them reporting multiple health issues [40]. Another survey found that 60% of the elderly in Bangladesh suffer from several chronic diseases (e.g., diabetes, hypertension, heart disease and cancer) at any point in time [41]. According to the socio-cultural and religious tradition of Bangladesh, elderly people used to live in joint households, and the family members would care for their own elderly members. However, this tradition has broken down due to rapid socio-economic and demographic transitions, changing social and religious values and the influence of western culture [42]. Moreover, in urban areas like the capital city of Bangladesh, Dhaka, the number of the elderly live independently at their own home or residential care has increased [43]. These changes are expected to place an unprecedented demand on the healthcare system. mHealth shows a promising future in addressing this demand in the management of chronic diseases and to transform and improve healthcare delivery for underserved patients with chronic disease, especially in the urban areas of developing countries [44]. However, little research has been done to draw any systematic study of the elderly’s adoption and usage of mHealth. To date, most studies have tended to focus on mHealth adoption issues related to physical conditions and lifestyles [45]; design and user interface [46]; security and privacy [47]; resistance to change and technology anxiety [48]; and perceived value, and perceived behavior control and attitudes [35]. In addition, previous studies have focused mainly on hospital or professional views of mHealth adoption [49]. Few studies have focused on the elderly adoption of mHealth [50,45,35]. However, current studies primarily focus on ageing female and investigate factors that influence the adoption of mHealth of middle-aged and elderly persons who were relatively healthier, more sociable and active [45]. Jen and Hung highlighted factors affecting the elderly’s family members’ intention to adopt mHealth [51]. A study by Guo et al. investigated factors affecting the mHealth experience and acceptance in the Chinese elderly context [48]. This study revealed a number of factors such as the perceived ease of use, systems quality, subjective norm, perceived usefulness, computer self-efficacy, perceived credibility technology anxiety and dispositional resistance to change that impact upon the

mHealth experience and acceptance. This study mainly focused on identifying challenges and barriers in adopting mHealth services by the elderly from a human-centric perspective; socio-technical characteristics such as facilitating condition and social influence, however, have not been addressed to the adoption of mHealth services. A study by Akter et al. investigated mHealth adoption and acceptance issues in Bangladesh from a service quality perspective [52]. They found that service quality has both a direct and indirect influence upon the intention to adopt mHealth services. This study, however, places little focus upon the adoption of mHealth by the elderly. As per the above literature review, the objective of this study was to attempt to fill the above identified gaps by investigating the adoption and acceptance issues of mHealth from elderly perspectives in a developing country like Bangladesh and hence, the following specific research question was formulated in this study. • What are the key factors influencing the elderly’s acceptance and adoption of mHealth services in developing countries like Bangladesh? The Unified Theory of Acceptance and Use of Technology (UTAUT) was used, as the theoretical framework underpinning the research, to understand and empirically test the factors that influence the elderly acceptance and adoption of mHealth services in a developing country context. 2. Theoretical framework and hypotheses In order to analyze the above research questions, this study adopts a most influential user acceptance and behaviors analysis model, the Unified Theory of Acceptance and Use of Technology (UTAUT). A number of theories including the Theory of Reasoned Action; the innovation Diffusion Theory; the Theory of Planned Behavior; the Technology Acceptance Model; the combined TAMTPB; the Motivational Model; the Model of PC Utilization, and the Social Cognitive Theory, have been developed and used to explain the intention and use of new technologies. The UTAUT model derived by comprehensive examination of various models mentioned above aiming to achieve a unified view of user acceptance [53]. It is claimed that the UTAUT model can explain as much as 70%

78

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84

Table 1 Summary of construct with measurement items. Construct

Corresponding Items

Items Sources

Performance Expectancy (PE)

PE1. I find mHealth service useful in my daily life. PE2. Using mHealth service helps me accomplish things more quickly. PE3. Using mHealth service increases my productivity. EE1. Learning how to use mHealth service is easy for me. EE2. My interaction with mHealth service is clear and understandable. EE3. I find mHealth service easy to use. EE4. It is easy for me to become skillful at using mHealth service. SI1. People who are important to me think that I should use mHealth service. SI2. People who influence my behavior think that I should use mHealth service. SI3. People whose opinions that I value prefer that I use mHealth service. FC1. I have the resources necessary to use mHealth service. FC2. I have the knowledge necessary to use mHealth service. FC3. mHealth is compatible with other technologies I use. TA1. Using mHealth services would make me very nervous. TA2. Using mHealth services make me worried. TA3. Using mHealth services may make me feel uncomfortable. TA4. Using mHealth services may make me feel uneasy and confused. RC1. I don’t want the mHealth services to change the way I deal with health relevant problems. RC2. I don’t want the mHealth services to change the way I keep myself healthy. RC3. I don’t want the mHealth services to change the way I interact with other people. BI1. I intend to use mHealth service in the future. BI2. I will always try to use mHealth service in my daily life. BI3. I plan to use mHealth service frequently. UB1: mHealth service is a pleasant experience. UB2: I use mHealth service currently. UB3: I spend a lot of time on mHealth service.

[53,66]

Effort Expectancy (EE)

Social Influence (SI)

Facilitating Condition (FC) Technology Anxiety (TA)

Resistance to Change (RC) Behavioral Intention (BI) Use Behavior (UB)

[73]

[66,73]

[73]

[45]

[63]

[64]

[73]

of the variance in intention. It has been widely adopted in different areas including e-Health and mHealth [54,55]. Using a review of prior studies of UTAUT in healthcare, we decided to adopt the most commonly used constructs in this study. The conceptual research model (Fig. 1) developed in this study postulates that four basic constructs: Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), and Facilitating Condition (FC) act as determinants of users’ Behavioral Intention (BI) and Use Behavior (UB). Among those, PE, EE and SI are significantly direct determinants of user intention to adopt technology such as mHealth. FC and BI explain user’s actual UB of technology [53]. The following sections explain the hypotheses developed in this study for investigating the research question.

H2. EE has a positive impact on the elderly’s intention to use mHealth.

2.1. Performance expectancy (PE)

H3. SI has a positive impact on the elderly’s intention to use mHealth.

PE is defined as “the degree to which an individual believes that using the system will help him or her to attain gains in job performance” [53]. Venkatesh et al. have found that PE is the strongest determinant of a user’s behavioral intention (BI) to adopt technology [53]. Pai and Huang indicate that PE affects BI to use health information systems [56]. Carlsson et al. revealed that PE has a direct positive effect on intention to use mobile devices [57]. Sun et al. empirically demonstrated that the greater the PE, the more likely mobile health services would be adopted [58]. The above discussion led this study to posit the following hypothesis: H1. PE has a positive impact on the elderly’s intention to use mHealth. 2.2. Effort expectancy (EE) EE is defined as “the degree of ease associated with the use of the system” [53]. Past studies suggest that EE has a strong influence on the users’ intention to health information system adoption and acceptance. For example, EE has been identified as an important factor directly influencing users’ intention to use mobile health monitoring systems, e-Health services via a smartphone, clinical decision support systems and mobile health [58,59]. Hence, it was hypothesized that:

2.3. Social influence (SI) SI is defined as “the degree to which an individual perceives that important others believe he or she should use the new system” [53]. Lu et al. have found that SI has strong impact on users’ intention to adopt technology [60]. Sun et al. reported that SI affects the behavioral intention to use mobile health services [58]. Wills et al. identified a significant co-relationship between SI and users’ Behavioral Intention to use digital information in healthcare [61]. As per the above discussion, the following hypothesis was posited:

2.4. Facilitating conditions (FC) FC is defined as “the degree to which an individual believes that an organizational and technical infrastructure exists to support the use of the system” [53]. Yi et al. found that FC is a direct determinant of behavioral intention and use of technology [62]. Bhattacherjee and Hikmet confirmed the critical role of infrastructure support on health information systems usage [63]. A study by Boontarig et al. suggested that FC positively influences the behavioral intention and use behavior of using smartphones for health services [59]. The above discussion resulted in the following hypotheses: H4. FC has a positive impact on the elderly’s intention to use mHealth. H5. FC has a positive impact on the elderly’s actual use of mHealth. The relationship between the behavioral intention (BI) and actual use behavior (UB) is well documented in many research fields and that indicates BI is a valid predictor of actual UB [64,65]. Venkatesh et al. empirically tested that BI explains user’s actual UB of technology [53]. Kijsanayotin et al. found that BI is a predictor of actual UB of health information technology [66]. Han et al. revealed that UB influences BI of physician to use mobile health technology

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84 Table 2 Demographics of respondents. Variable

Description

Frequency

Percentage

Gender

Male Female 60–64 65–69 70–74 More than 75 1–2 3–5 SSC or below HSC Bachelor Other Yes No 1–3

181 93 38 162 52 22 197 77 41 140 68 25 238 36 58

66% 34% 14% 59% 19% 8% 72% 28% 15% 51% 25% 9% 87% 13% 21%

4–7 8–10 More than 10

173 38 5

63% 14% 2%

Age

Chronic Disease Educational Qualification

Own Mobile Phone Mobile phone usage experience

[67]. Based on the above literature, this study posited the following hypothesis: H6.

BI has a positive impact on the actual use of mHealth.

The basic UTAUT model has been expanded with additional variables [54,68,69]. Among other variables, technology anxiety (TA) and resistance to change (RC) have been identified as major concern that influences the elderly’s intention to adopt mHealth services. There is general perception that older adults have fewer new technology skills and less technological self-efficiency than younger adults [70]. Moreover, declining physical and cognitive capabilities possibly cause them to suffer a higher level of anxiety potentially further reducing their intention to use innovative technology [35]. TA is a negative emotional response, and pertains to the fear or discomfort people experience when they think of using technology [71]. On the other hand, RC reduces the adoption intention of technology [48]. It is also argued that elderly people have a high propensity for RC [72]. Therefore, this study extends the basic UTAUT with additional variables, TA and RC shown in Fig. 1, aiming to enhance the prediction of the elderly’s intention to use mHealth. Hence, the following hypotheses were proposed: H7. TA has a negative impact on the elderly’s intention to use mHealth. H8. RC has a negative impact on the elderly’s intention to use mHealth. 3. Methodology 3.1. Measurement instruments To ensure the validity of all measures, the measurement items for latent constructs within the proposed model were developed from prior studies. The detailed items of each construct and their sources are listed in Table 1. 3.2. Questionnaire design and data collection A structured questionnaire was originally developed in English, and then translated into the local language (Known as Bangla) by a professional translator, a native Bangla speaker with an excellent command of the English language and good knowledge of mHealth, ensuring that both versions were considered to be similar by a group of experts. This translation process is important for capturing the perception, which has already been reflected in the

79

works of Andaleeb [74]. The questionnaire was divided into Part A and Part B. Part A contains the demographic information. Respondents were asked information about their age, gender, educational qualifications, chronic disease and mobile phone usage experience. Part B includes questions for the different constructs presented in the research model, shown in Fig. 1, using a 5-point Likert scale ranging from (1) “strongly disagree” to (5) “strongly agree. A pilot study was conducted to test the effectiveness and appropriateness of the questionnaire, for which 15 samples from selected retired staff from the University of Dhaka and the Bangladesh University of Engineering and Technology (BUET) were recruited. They were selected due to their age bracket (> = 60), technical knowledge, and experience in mobile/internet usage. The feedback from the pilot study was used to review the questionnaire with context-specific adjustments. The study population consisted of elderly people in Dhaka City Corporation, Bangladesh. Dhaka City Corporation is the largest and most populated urban territory developed around a central place of Dhaka, Capital city of Bangladesh. It is associated with the task of running the affairs of the city of Dhaka and covers a population of about 9 million, among them 7.5% are elderly [40]. This study recruited participants (i.e., elderly people) from Dhaka City Corporation for several reasons. Firstly, the Dhaka City Corporation areas have 100% network coverage allowing access to mHealth service from anywhere. Secondly, in addition to simple communication, people including the elderly in Dhaka city frequently use mobile technology for other various online activities such as searching information, e.g., news and health, online shopping, gaming, etc. [75]. Thirdly, large numbers of elderly in Dhaka City Corporation suffer from various chronic medical conditions [41]. There is considerable variation in the opinions observed in the literature in regard to the selection/calculation of optimum sample size in different types of statistical analysis [76]. For example, statistical analysis including structural equation modeling (SEM) recommends sampling of 200 as fair and 300 as good [77]. Hair et al. also recommended a sample size of 200 to test a model using SEM. In Hoelter, 200 was suggested as a ‘critical sample size’ that can be used in any common estimation procedure for valid results [78]. Roscoe reported that in a multivariate research, such as multiple regression analysis, the sample size should be at least 10 times the number of items in the study [79]. In our study, the proposed multivariate regression model consists of 26 items including both independent and dependent. As per Roscoe and previous studies, a sample size of 300 was selected in this study for data analysis using SEM [79]. Potential participants were randomly selected from the National ID (NID) Database of Election Commission. The sample was drawn by a simple random procedure that is equivalent to a lottery system. In this study, National ID numbers of elderly individuals were randomly selected as winners, potential participants, in an unbiased manner. Simple random sampling can eliminate the bias by giving all individuals an equal chance to be chosen. In addition, it can provide an accurate picture of the population [80]. Data were collected using a questionnaire developed in the participant’s native language. Initial contact was made by the researchers with individuals and their families to provide information, in a form of information sheet, including the definition of mHealth and the purpose of the study. The questionnaires were personally distributed by the researchers and individuals willing to participate were asked to read a consent form (or had it read to them) and then sign or mark it with a thumb print. Respondents were also made aware of their rights to withdraw participation at any time during the study. In terms of survey interaction, in-home and location intercept techniques, though more time consuming and costly, provide a higher response rate than a postal mail, telephone or online survey in the context of a developing country [80].

80

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84

Table 3 The measurement model. Constructs

Items

Loadings

AVE

CR

Cronbach’s alpha

Behavioral Intention (BI)

BI1 BI2 BI3 EE1 EE2 EE3 EE4 FC1 FC2 FC3 PE1 PE2 PE3 RC1 RC2 RC3 SI1 SI2 SI3 TA1 TA2 TA3 TA4 UB1 UB2 UB3

0.8860 0.8510 0.8578 0.8858 0.9000 0.8620 0.8726 0.8823 0.8813 0.8754 0.8976 0.8982 0.8744 0.9434 0.9510 0.9652 0.8331 0.8430 0.8356 0.8118 0.8383 0.8380 0.8341 0.9144 0.9154 0.9026

0.7483

0.8992

0.8317

0.7748

0.9322

0.9037

0.7738

0.9112

0.8550

0.7923

0.9196

0.8691

0.9087

0.9676

0.9497

0.7010

0.8755

0.7879

0.6899

0.8990

0.8506

0.8296

0.9359

0.8976

Effort Expectancy (EE)

Facilitating Conditions (FC)

Performance Expectancy (PE)

Resistance to Change (RC)

Social Influence (SI)

Technology Anxiety (TA)

Use Behavior (UB)

AVE: Average Variance Extracted, CR = Composite Reliability.

This study has ethical approval from the Human Research Ethics Committee at Southern Cross University, Australia. 3.3. Data analysis The Partial Least Squares (PLS) method, a statistical analysis technique based on the Structural Equation Modeling (SEM), was used to test and validate the proposed model and the relationships among the hypothesized constructs. SEM is a widely accepted paradigm to gauge the validity of theories with empirical data [81]. SmartPLS software, one of the well-known software applications for PLS-SEM, was used to analyze the data [82]. Data from questionnaires were captured into Microsoft excel and imported into SmartPLS software for statistical analysis. 4. Results 4.1. Demographic characteristics of sample The study distributed 300 questionnaires, of which 274 were received completed for further analysis. The demographic characteristics of respondents presented in Table 2 shows that percentages were 66% males and 34% females who participated in the study. The majority of respondents (59%) were aged between 65 and 69 years. About 72% of respondents suffered from at least one chronic medical condition. In terms of access, 87% of respondents owned a mobile phone. Most of the participants (85%) had attained at least a HSC (Higher Secondary Certificate equivalent to year 12) level education, with 79% of them having more than 3 years of mobile phone usage experience.

Table 4 Correlation matrix and square root of the AVE.

BI EE FC PE RC SI TA UB

BI

EE

FC

PE

RC

SI

TA

UB

0.8650 0.3560 0.2531 0.5314 0.4761 0.3578 0.4219 0.4461

0.8802 0.1940 0.2123 0.2056 0.1204 0.1838 0.3146

0.8797 0.3501 0.2468 0.2625 0.1622 0.2028

0.8901 0.2627 0.3148 0.3266 0.2535

0.9533 0.2493 0.2680 0.3141

0.8373 0.1808 0.2450

0.8306 0.4105

0.9108

assessed by an average variance extracted (AVE) and items loading with at least 0.50 of AVE for construct validity [82]. The loadings, AVE, composite reliability and Cronbach’s alpha (˛) are presented in Table 3. It can be seen from Table 3 that the calculated Cronbach’s alpha (˛) values ranged from 0.7879 to 0.9497 and composite reliability values ranged from 0.8755 to 0.9676, which supports strong internal reliability. Table 3 also shows that the estimated constructs loading ranged from 0.8118 to 0.9652 and AVE ranged from 0.6899 to 0.9087 are greater than the recommended levels. Therefore, the conditions for convergent validity are satisfied in this study. In contrast, the discriminant validity was assessed by the square root of the AVE and cross loading matrix. The square root of the AVE of a construct should be greater than its correlation with other constructs for satisfactory discriminant validity [82]. The diagonal elements must be larger than the entries in corresponding columns and rows to satisfy discriminant validity [83]. The results shown in Table 4 reveals that all constructs in this study confirm the discriminant validity of the data. 4.3. Hypothesis testing

4.2. Measurement model The measurement model was assessed by examining the internal reliability, convergent and discriminant validity [82]. The internal reliability was evaluated considering Cronbach’s alpha (˛) and composite reliability where the level of 0.70 is an indicator for acceptable internal consistency [76]. Convergent validity was

The structural model was developed to identify the relationships among the constructs in the research model. The study tested the relationship between dependent and independent variables by path coefficient (ˇ) and t-statistics. The PLS results for the structural model are shown in Table 5. The results show that the relationships between PE and BI (t = 5.1920, ˇ = 0.3193, p < 0.05), EE and

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84 Table 5 Structural model. Hypothesis

Path

B

t- Statistics

Comments

H1 H2 H3 H4 H5 H6 H7 H8

PE −> BI EE −> BI SI −> BI FC −> BI FC −> UB BI −> UB TA −> BI RC −> BI

0.3193 0.1879 0.1441 0.0298 0.0979 0.4148 −0.1897 −0.2645

5.1920 3.6747 2.8407 0.5537 1.3698 4.6749 3.2097 4.5724

Supported Supported Supported Not Supported Not Supported Supported Supported Supported

Significant at P < 0.05.

BI (t = 3.6747, ˇ = 0.1879, p < 0.05), SI and BI (t = 2.8407, ˇ = 0.1441, p < 0.05), TA and BI (t = 3.2097, ˇ = −0.1897, p < 0.05), RC and BI (t = 4.5724, ˇ = −0.2645, p < 0.05), BI and UB (t = 4.6749, ˇ = 0.4148, p < 0.05) were significant. Thus H1, H2, H3, H6, H7 and H8 were supported. However, the relationships between FC and BI (t = 0.5537, ˇ = 0.0298, p > 0.05), FC and UB (t = 1.3698, ˇ = 0.0979, p > 0.05) were insignificant and not supporting H4 and H5 in the current study. 5. Discussion This study applied the UTAUT model to determine the users’ (i.e., the elderly in Bangladesh) behavioral intention to adopt and use the mHealth services. The empirical findings have provided insight into the UTAUT constructs such as PE, EE, SI, TA and RC influencing the adoption of mHealth. The study provides empirical support for the hypotheses proposed in relation to PE, EE, SI and BI in mHealth adoption and usage. Our findings are consistent with the results of previous studies on the application of UTAUT in e-Health including mHealth adoption. Hennington et al. found that PE and EE have a significant influence on physician adoption of electronic medical records (EMR) [84]. Phichitchaisopa and Naenna revealed that the factors such as PE and EE have a significant influence on the adoption of healthcare information technology [85]. Holtz and Krein found that PE and SI are significant factors in electronic medical record (EMR) systems adoption [86]. AlAwadhi and Morris indicated that SI becomes more significant when individuals have inadequate experience of online services [87]. Wills et al. revealed that SI plays a greater role in EMR adoption [61]. This study identified TA and RC had a significant but negative influence on BI to use mHealth. Compared to developed countries, most of the elderly in Bangladesh usually prefer face-to-face professional healthcare services rather than an innovative information technology based healthcare service like mHealth [9]. This is due to a reason that mobile phones were not commonplace in Bangladesh when the participants were born about 60 years ago [88]. This finding appears to be unaligned with the mobile phone usage profiles, for example 79% of the interviewed citizens have four or more years of mobile phone usage, as shown in Table 2. However, in Bangladesh, the majority of older people, especially older women, often depend upon their children or husband for help using their phones, even when making a phone call to contact their families. In addition, women’s mobile phone usage and their finance is highly controlled by men [9]. Due to this local cultural perspective, using mobile phones for purposes other than making a typical phone call, such as receiving health care advice is still a comparatively new concept in Bangladesh [89]. Therefore, the authors believe that the finding may be a true reflection of the present context of mHealth in Bangladesh. In addition, the digital divide, practicing traditional culture, limited technical skills, and information seeking behavior increase technology anxiety and the resistance to change in the elderly to adopt new technologies like mHealth [90]. Research has found that elderly people have less technological self-efficiency, a higher level of anxiety, resistance to change, less control over ICT

81

and fewer new technology skills, which are all likely to negatively affect acceptance of innovative technology such as mHealth [91]. Tung and Chang identified TA as the most important variable in their study that negatively affects BI [92]. Guo et al. revealed that TA can reduce the adoption intention of elderly people by reducing the perceived ease of use (EE) [48]. Some other studies have also argued that TA creates major barriers for the elderly in adopting innovations [93]. It is found that elderly people resist changing their medical habits and choose to continue their previous behaviors despite the emergence of mHealth services. However, the study did not find any significant relationship between FC and BI to the adoption and acceptance of mHealth. This finding may appear apparently surprising given that many previous studies confirm a positive relationship between FC and BI in technology adoption. For example, Phichitchaisopa and Naenna found that FC significantly affects the adoption of healthcare information technology [85]. Our finding could be reflective of the fact that most elderly people in Bangladesh are extensively dependent on their adult children for their living [94]. As a consequence, it is unlikely that they are aware of the importance of technical, monetary and/or infrastructural resource requirements, and their impact on the mHealth service adoption and usage. It is also true if both PE and EE constructs are present in a model and significant, FC becomes non significant in predicting user’s behavioral intention to the adoption and acceptance of technology [53]. This study examined the association between BI and UB and identified a positive relationship between them. Carlsson et al. indicated that the higher the elderly’s BI to adopting mHealth is, the greater their UB in actual use of the technology [57]. Ifinedo also indicated an association between user’s BI and UB in technology adoption and usage [95].

6. Implications The findings of this study offer a significant theoretical contribution to the literature. This research has applied the UTAUT model to determine elderly behavioral intention to adopt mHealth services in the context of developing countries [53]. Technology acceptance models are frequently applied for studying user acceptance of technologies, however few studies have investigated technology acceptance of mHealth care in general [50]. In addition, most of the prior studies primarily adopted the basic TAM or UTAUT model with PE and EE constructs and implemented them either in the context of a developed nation or with little focus on the elderly’s adoption of mHealth in the context of developing nations [96,97]. To the best our knowledge, this study is the first attempt extending the constructs of the original UTAUT model with additional variables, i.e., technology anxiety and resistance to change, and empirically testing them to validate their applicability in mHealth applications in the context of the elderly in a developing country. This study believes that the proposed extension of the UTAUT model makes a significant contribution to the literature in that it is one of the few to investigate the factors influencing mHealth adoption among elderly in the context of developing nations. The results also provide further support for the utility of UTAUT with additional variables in mHealth adoption. The findings of this study offer practical implications for the benefits of improving the adoption of mHealth in the context of a developing nation. A systematic review conducted by Chib et al. indicates that one of the main reasons for low uptake of mHealth in developing countries is the lack of adequate research focusing on the theoretical understanding of its adoption process [98]. Our empirical findings provide a practical guideline to the successful adoption of mHealth services in developing countries. With an increased knowledge of the user perception of mHealth services, an expected benefit for

82

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84

mHealth technology developers and providers is to understand the challenges/issues in regards to the design and implementation of successful mHealth services. Furthermore, due to a generic approach, the findings of this study could be easily modified to assist other developing countries in the planning and up-take of mHealth services.

7. Limitations and future research directions There are some limitations in the study due to (1) age-specific participants and (2) specific geographic context chosen for administering the survey. The study was conducted to the sample population of elderly living in the capital of Bangladesh. Hence, the results may raise concerns about the generalizability of the findings. Further research, therefore, would extend the current study to include additional demographic and geographic factors such as the age, income and education of citizens nationally, and other variables such as trust and personal innovativeness in IT to uncover a more generalized view of the proposed model in the context of a developing country like Bangladesh. Secondly, due to the cross-sectional design, the proposed study cannot confirm the causality and contingent effects of users’ level of experience before and after the mHealth system and service adoption. It is also known that cross-sectional data might not have been representative of the actual use of technology. Further study could use longitudinal data to unfold the causal relationship among variables over time and to more accurately reflect actual technology usage. Another extension could also adopt both quantitative and qualitative approaches to unveil users’ in-depth views and/or opinions on the issues. Although, UTAUT is generally applied to a specific technology, the literature shows the applicability of UTAUT in a general or broad category of technology, for example e-Health and e-Government [99]. This study is the reflection of the adoption of a broad category of mHealth technology similar to some previous studies, e.g., Faqih & Jaradat [97].

8. Conclusions This study attempts to extend the well-validated UTAUT framework with additional variables, i.e., technology anxiety and resistance to change and implement them in a novel context of mHealth adoption and use by elderly people in a developing country. The study found that performance expectancy, effort expectancy, social influence, technology anxiety and resistance to change significantly influence the elderly’s behavioral intention to use mHealth. The results of the study, however, suggest that there is no significant relation between facilitating conditions and user behavioral intentions to adopt the mHealth services. The results appear inconsistent with some other studies, however, it is our belief that the findings truly reflect on the socio-economic and cultural aspect of developing countries like Bangladesh. In such a context, the technology adoption decisions made by most senior citizens, who are dependent on their adult family members, are more influenced by social pressures rather than by facilitating factors, such as the availability of human assistance. The findings of the study contribute to the body of research informing mHealth design and development to maximize its adoption and up-take in developing countries. The findings provide valuable information for mHealth service providers, planners, and policy makers to develop strategy and policy for the successful implementation and acceleration of the adoption of this technology among end users, particularity the elderly, in a developing environment such as Bangladesh.

Conflict of interest There is no conflict of interest for this study. Authors contribution Both authors made a substantial contribution to this study. Rakibul Hoque conceived and designed the study, conducted data collection and analysis, interpreted findings, and drafted the manuscript. Golam Sorwar led the study in the development of the research model, reviewed the study design and interpretation of study findings, and drafted the manuscript. Both authors contributed to the revision of the manuscript and drafted the final version submitted. Summary points What was already known on the topic? • mHealth (mobile health) may be the blessing of ICT and is probably one of the most prominent services with noticeable effect on the development of healthcare sector. • Healthcare services via mobile phone appear significant for elderly patients who are most in need of healthcare. What this study added to the body of knowledge? • This study confirmed the applicability of a modified version of UTAUT model in the mHealth application in the context of developing countries like Bangladesh. • This study identified that technology anxiety and resistance to change have a significant influence on the elderly’s behavioral intention to use mHealth in a developing country. Acknowledgements This work was supported by seed research grant of School of Business and Tourism (SBaT), Southern Cross University, under grant No. (31496). The authors, therefore, gratefully acknowledge the SCBS technical and financial support. References [1] M. Sharifi, M. Ayat, M. Jahanbakhsh, N. Tavakoli, H. Mokhtari, W.K. Wan Ismail, E-health implementation challenges in Iranian medical centers: a qualitative study in Iran, Telemed. e-Health 19 (2) (2013) 122–128. [2] R. Hoque, M. Mazmum, Y. Bao, e-Health in Bangladesh: current status, challenges, and future direction, Int. Technol. Manage. Rev. 4 (2014) 87–96. [3] G. Chiarini, P. Ray, S. Akter, C. Masella, A. Ganz, mHealth technologies for chronic diseases and elders: a systematic review, IEEE J. Sel. Areas Commun. 31 (9) (2013) 6–18. [4] Global Health Workforce Alliance (GHWA), Global Health Workforce Crisis. Key messages −2013, Online (http://www.who.int/workforcealliance/media/KeyMessages 3GF.pdf), (Accessed 22.2.2016). [5] K. Källander, J.K. Tibenderana, O.J. Akpogheneta, D.L. Strachan, Z. Hill, A.H. Asbroek, S.R. Meek, Mobile health (mHealth) approaches and lessons for increased performance and retention of community health workers in low-and middle-income countries: a review, J. Med. Internet Res. 15 (2013) e17. [6] Bangladesh Telecommunication Regulatory Commission, Mobile Phone Subscribers in Bangladesh, (2015). [7] NIPORT, Mitra and Associates, and ICF International, Bangladesh Demographic and Healthsurvey 2011, Dkaka and Calverton MA: National Institute of Population Research and Training Mitra and Associates and Macro International, (2013). [8] K. Ghorai, S. Jahan, P. Ray, M. Chylinski, Mobile phone including smart phone based persuasive system design for controlling hypertension and diabetes in Bangladesh, 24th Australasian Conference on Information System (2013). [9] T. Ahmed, G. Bloom, M. Iqbal, H. Lucas, S. Rasheed, L. Waldman, A. Bhuiya, E-health and M-health in Bangladesh: Opportunities and Challenges, 2014.

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84 [10] W. Chigona, M. Nyemba, A. Metfula, A review on mHealth research in developing countries, J. Commun. Inf. 9 (2) (2012). [11] M. Kay, mHealth: New Horizons for Health Through Mobile Technologies, World Health Organization, 2011. [12] S. Akter, P. Ray, mHealth-an ultimate platform to serve the unserved, Yearb. Med. Inf. (2010) 94–100. [13] S. Motamarri, S. Akter, P. Ray, C.L. Tseng, Distinguishing ‘mHealth’from other healthcare services in a developing country: a study from the service quality perspective, Commun. AIS 34 (1) (2014) 669–692. [14] A. Nessa, M. Ameen, S. Ullah, K. Kwak, Applicability of telemedicine in Bangladesh: current status and future prospects’, Int. Arab J. Inf. Technol. (IAJIT) (2) (2008) 138–145 (abs/0911.1). [15] D.J. Vassallo, F. Hoque, M.F. Roberts, V. Patterson, P. Swinfen, R. Swinfen, An evaluation of the first year’s experience with a low-cost telemedicine link in Bangladesh, J. Telemed. Telecare 7 (3) (2001) 125–138. [16] S. Zakir, eHealth initiatives and projects of Telemedicine Reference Center Ltd. (TRCL) Dhaka, Bangladesh,S pecial contribution (2008). [17] G. Ivatury, J. Moore, A. Bloch, A doctor in your pocket: health hotlines in developing countries, Innovations 4 (1) (2009) 119–153. [18] M.S. Akter, Service Quality Dynamics of Mhealth in a Developing Country (Doctoral Dissertation, The University of New South Wales, 2012. [19] M.R. Hoque, G. Sorwar, Factors influencing mHealth acceptance among elderly people in Bangladesh, in: The 26th Australasian Conference on Information Systems (ACIS), Australia, 2015. [20] Directorate General of Health Services, Publication http://dghs.gov.bd/bn/ licts file/images/Health Bulletin/HB2012 CH/HB2012 CH17 HIS-eHealthMBT.pdf, (2013) (Accessed 25.03.2016). [21] Health Bulletin, Ministry of Health and Family Welfare, Government of the People’s Republic of Bangladesh, 2014. [22] CHMI, ICT in Health: Bangladesh is Moving Ahead, Center for Health Market Innovation, (2014). [23] M.R. Sarwar, Bangladesh health service delivery: innovative NGO and private sector partnerships, IDS Bull. 46 (3) (2015) 17–28. [24] P.K. Jiban, H.M. Muazzam, Health care practices of older people in sylhet: a study conducted in slum areas of sylhet city of Bangladesh, Int. J. Human. Soc. Sci. Educ. (IJHSSE) 3 (4) (2016) 90–100. [25] Swedish International Development Cooperation Agency (SIDA), (2010). Reality Check Bangladesh 2009 − listening to poor people’s realities about primary healthcare and primary education − Year 3. [26] N.U. Khan, S. Rasheed, T. Sharmin, T. Ahmed, S.S. Mahmood, F. Khatun, A. Bhuiya, Experience of using mHealth to link village doctors with physicians: lessons from Chakaria, Bangladesh, BMC Med. Inform. Decis. Mak. 15 (2015) 62. [27] F. Khatun, A.E. Heywood, P.K. Ray, S.M.A. Hanifi, A. Bhuiya, S.T. Liaw, Determinants of readiness to adopt mHealth in a rural community of Bangladesh, Int. J. Med. Inf. 84 (2015) 847–856. [28] Z.N. Kabir, The Emerging Elderly Population in Bangladesh: Aspects of Their Health and Social Situation, Department of Clinical Neuroscience, Occupational Therapy and Elderly Care Research (NEUROTEC), Division of Geriatric Medicine, Karolinska Institutet, 2001. [29] J.M. Ortman, V.A. Velkoff, H. Hogan, An Aging Nation: The Older Population in the United States, US Census Bureau, Washington, DC, 2014, pp. 25–1140. [30] WHO, World Health Statistics 2013: A Wealth of Information on Global Public Health, World Health Organization, 2013. [31] UN, World Population Prospects: The 2012 Revision, Department of Economic, United Nations Publications, 2012. [32] D. Hutton, Older People in Emergencies: Considerations for Action and Policy Development, World Health Organization, 2008. [33] D.L. Maisonneuve, J.O. Martins, Public Spending on Health and Long-Term Care, 2013. [34] Pew Internet, Mobile Access (2010), Available from: http://www.pewinternet. org/∼/media/Files/Reports/2010/PIP MobileAccess2010.pdf (Accessed 23.05.16). [35] Z. Deng, X. Mo, S. Liu, Comparison of the middle-aged and older users’ adoption of mobile health services in China, Int. J. Med. Inf. 83 (3) (2014) 210–224. [36] A.M. Munsur, I. Tareque, K.M. Rahman, Determinants of living arrangements, health status and abuse among elderly women: a study of rural Naogaon district, Bangladesh, J. Int. Womens Stud. 11 (4) (2013) 162–176. [37] H. Chaklader, M. Haque, M. Kabir, Socio-economic situation of urban elderly population from a microstudy, in: M. Kabir (Ed.), The Elderly Contemporary Issues, Bangladesh Association of Gerontology, Dhaka, 2003, pp. 1–13. [38] BBS Bangladesh Population Census, (2011), Bangladesh Bureau of Statistics, Ministry of Planning, Bangladesh, 2011. [39] U. Desa, World Population Prospects: the 2008 Revision, New York : Department for Economic and Social Affairs, 2009. [40] A. Razzaque, L. Nahar, M.A. Khanam, P.K. Streatfield, Socio-demographic differentials of adult health indicators in Matlab Bangladesh: self-rated health, health state, quality of life and disability level, Glob. Health Action 3 (2010) 70–77. [41] M.A. Khanam, P.K. Streatfield, Z.N. Kabir, C. Qiu, C. Cornelius, A. Wahlin, Prevalence and patterns of multimorbidity among elderly people in rural Bangladesh: a cross-sectional study, J. Health Popul. Nutr. (2011) 406–414. [42] M.N. Islam, D.C. Nath, A future journey to the elderly support in Bangladesh, J. Anthropol. (2012) [Online] (Accessed 3.2.2016) Available from: http://www. hindawi.com/journals/janthro/2012/752521/.

83

[43] S. Shahanaz, Quality of Life Among Elder People at Elderly Care Homes in Bangladesh (Doctoral Dissertation, Bangladesh Health Professions Institute (The academic institute of CRP) University of Dhaka), 2015. [44] S. Hamine, E. Gerth-Guyette, D. Faulx, B.B. Green, A.S. Ginsburg, Impact of mHealth chronic disease management on treatment adherence and patient outcomes: a systematic review, J. Med. Internet Res. 17 (2) (2015). [45] L. Xue, C.C. Yen, L. Chang, H.C. Chan, B.C. Tai, S.B. Tan, M. Choolani, An exploratory study of ageing women’s perception on access to health informatics via a mobile phone-based intervention, Int. J. Med. Inf. 81 (2012) 637–648. [46] A. Lorenz, R. Oppermann, Mobile health monitoring for the elderly: designing for diversity, Pervasive Mob. Comput. 5 (2009) 478–495. [47] L.H. Iwaya, M.A.L. Gomes, M.A. Simplício, T.C.M.B. Carvalho, C.K. Dominicini, R.R. Sakuragui, P. Håkansson, Mobile health in emerging countries: a survey of research initiatives in Brazil, Int. J. Med. Inf. 82 (5) (2013) 283–298. [48] X. Guo, Y. Sun, N. Wang, Z. Peng, Z. Yan, The dark side of elderly acceptance of preventive mobile health services in China, Electron. Mark. 23 (1) (2013) 49–61. [49] I.L. Wu, J.Y. Li, C.Y. Fu, The adoption of mobile healthcare by hospital’s professionals: an integrative perspective, Decis. Support Syst. 51 (3) (2011) 587–596. [50] M. Cocosila, N. Archer, Adoption of mobile ICT for health promotion: an empirical investigation, Electron. Mark. 20 (2010) 241–250. [51] W.Y. Jen, M.C. Hung, An empirical study of adopting mobile healthcare service: the family’s perspective on the healthcare needs of their elderly members, Telemed. e-Health 16 (2010) 41–48. [52] S. Akter, J. D’Ambra, P. Ray, Service quality of mHealth platforms: development and validation of a hierarchical model using PLS, Electron. Mark. 20 (2010) 209–227. [53] V. Venkatesh, M.G. Morris, G.B. Davis, F.D. Davis, User acceptance of information technology: toward a unified view, MIS Q. 27 (3) (2003). [54] P.A. Nuq, B. Aubert, Towards a better understanding of the intention to use eHealth services by medical professionals: the case of developing countries, Int. J. Healthcare Manage. 6 (4) (2013) 217–236. [55] C.P. Lin, B. Anol, Learning online social support: an investigation of network information technology based on UTAUT, CyberPsychol. Behav. 11 (2008) 268–272. [56] F.Y. Pai, K.I. Huang, Applying the Technology Acceptance Model to the introduction of healthcare information systems, Technol. Forecast. Soc. Change 78 (2011) 650–660. [57] B. Carlsson, Internationalization of innovation systems: a survey of the literature, Res. Policy 35 (2006) 56–67. [58] Y. Sun, N. Wang, X. Guo, Z. Peng, Understanding the acceptance of mobile health services: a comparison and integration of alternative models, J. Electron. Commerce Res. 14 (2) (2013) 183–200. [59] W. Boontarig, W. Chutimaskul, V. Chongsuphajaisiddhi, B. Papasratorn, Factors influencing the Thai elderly intention to use smartphone for e-Health services, IEEE Symposium on Humanities, Science and Engineering Research (SHUSER) (2012) 479–483. [60] J. Lu, J.E. Yao, C.S. Yu, Personal innovativeness: social influences and adoption of wireless Internet services via mobile technology, J. Strat. Inf. Syst. 14 (2005) 245–268. [61] M.J. Wills, O.F. El-Gayar, D. Bennett, Examining healthcare professionals’ acceptance of electronic medical records using UTAUT, Issues Inf. Syst. 9 (2008) 396–401. [62] M.Y. Yi, J.D. Jackson, J.S. Park, J.C. Probst, Understanding information technology acceptance by individual professionals: toward an integrative view, Inf. Manage. 43 (2006) 350–363. [63] A. Bhattacherjee, N. Hikmet, Reconceptualizing organizational support and its effect on information technology usage: evidence from the health care sector, J. Comput. Inf. Syst. 48 (2008). [64] V. Venkatesh, F.D. Davis, A theoretical extension of the technology acceptance model: four longitudinal field studies, Manage. Sci. 46 (2000) 186–204. [65] B.H. Sheppard, J. Hartwick, P.R. Warshaw, The theory of reasoned action: a meta-analysis of past research with recommendations for modifications and future research, J. Consum. Res. (1988) 325–343. [66] B. Kijsanayotin, S. Pannarunothai, S.M. Speedie, Factors influencing health information technology adoption in Thailand’s community health centers: applying the UTAUT model, Int. J. Med. Inf. 78 (2009) 404–416. [67] S. Han, P. Mustonen, M. Seppanen, M. Kallio, Physicians’ behavior intentions regarding the use of mobile technology: an exploratory study. PACIS 2004 Proceedings, 49. [68] J.C. Oh, S.J. Yoon, Predicting the use of online information services based on a modified UTAUT model, Behav. Inf. Technol. 33 (7) (2014) 716–729. [69] E.L. Slade, M. Williams, Y. Dwivedi, An extension of the UTAUT 2 in a healthcare context, Proceeding of the UK Academy for Information Systems (2013). [70] S.J. Czaja, N. Charness, A.D. Fisk, C. Hertzog, S.N. Nair, W.A. Rogers, J. Sharit, Factors predicting the use of technology: findings from the center for research and education on aging and technology enhancement (CREATE), Psychol. Aging 21 (2) (2006) 333. [71] M.L. Meuter, A.L. Ostrom, M.J. Bitner, R. Roundtree, The influence of technology anxiety on consumer use and experiences with self-service technologies, J. Bus. Res. 56 (11) (2003) 899–906.

84

R. Hoque, G. Sorwar / International Journal of Medical Informatics 101 (2017) 75–84

[72] J.A.A. Smither, C.C. Braun, Technology and older adults: factors affecting the adoption of automatic teller machines, J. Gen. Psychol. 121 (4) (1994) 381–389. [73] V. Venkatesh, J.Y. Thong, X. Xu, Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology, MIS Q. 36 (2012) 157–178. [74] S. Andaleeb, Caring for children: a model of healthcare service quality in Bangladesh, Int. J. Qual. Health Care 20 (2008) 339–345. [75] R. Hoque, An empirical study of mHealth adoption in developing countries: the moderating effect of gender concern, BMC Med. Inform. Decis. Mak. (2016). [76] J.F. Hair, R.E. Anderson, R.L. Tatham, C. William, Black, in: Multivariate Data Analysis, 1998. [77] R.B. Kline, Principles and Practice of Structural Equation Modeling, Guilford Publications, 2015. [78] J.W. Hoelter, The analysis of covariance structures goodness-of-fit indices, Sociol. Methods Res. 11 (1983) 325–344. [79] J.T. Roscoe, Fundamental Research Statistics for the Behavioral Sciences, in: John T. Roscoe (Ed.), Holt, Rinehart and Winston, New York, NY, 1975. [80] N.K. Malhotra, Marketing Research: An Applied Orientation, 5th ed., Pearson Education, India, 2008. [81] O. Götz, K. Liehr-Gobbers, M. Krafft, Evaluation of structural equation models using the partial least squares (PLS) approach, in: Handbook of Partial Least Squares, Springer, Berlin Heidelberg, 2010, pp. 691–711. [82] J.F. Hair Jr., G.T.M. Hult, C. Ringle, M. Sarstedt, A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM), SAGE Publications, Incorporated, 2013. [83] J. Henseler, C.M. Ringle, R.R. Sinkovics, The use of partial least squares path modeling in international marketing, Adv. Int. Market. 20 (1) (2009) 277–319 (2009). [84] A. Hennington, B.D. Janz, Information systems and healthcare XVI: physician adoption of electronic medical records: applying the UTAUT model in a healthcare context, Commun. Assoc. Inf. Syst. 19 (2007) 5. [85] N. Phichitchaisopa, T. Naenna, Factors affecting the adoption of healthcare information technology, EXCLI J. 12 (2013) 413–436. [86] B. Holtz, S. Krein, Understanding nurse perceptions of a newly implemented electronic medical record system, J. Technol. Hum. Serv. 29 (2011) 247–262.

[87] S. AlAwadhi, A. Morris, (2008). The Use of the UTAUT Model in the Adoption of E-government Services in Kuwait. In Hawaii International Conference on System Sciences, Proceedings of the 41 st Annual (pp. 219–219). [88] S.M. Ahmed, M.A. Hossain, M.R. Chowdhury, Informal sector providers in Bangladesh: how equipped are they to provide rational health care? Health Policy Plan. 24 (6) (2009) 467–478. [89] N. Nisha, M. Iqbal, A. Rifat, S. Idrish, Exploring the role of service quality and knowledge for mobile health services, Int. J. E-Bus. Res. (IJEBR) 12 (2) (2016) 45–64. [90] A. Islam, K. Tsuji, Bridging digital divide in Bangladesh: study on community information centers, Electron. Libr. 29 (4) (2011) 506–522. [91] R. Steele, A. Lo, C. Secombe, Y.K. Wong, Elderly persons’ perception and acceptance of using wireless sensor networks to assist healthcare, Int. J. Med. Inf. 78 (12) (2009) 788–801. [92] F.C. Tung, S.C. Chang, Nursing students’ behavioral intention to use online courses: a questionnaire survey, Int. J. Nurs. Stud. 45 (2008) 1299–1309. [93] M.H. Ryu, S. Kim, E. Lee, Understanding the factors affecting online elderly user’s participation in video UCC services, Comput. Hum. Behav. 25 (2009) 619–632. [94] N. Alam, H.R. Chowdhury, M.A. Bhuiyan & P.K. Streatfield Causes of death of adults and elderly and healthcare-seeking before death in rural Bangladesh, J. Health Popul. Nutr. 28 (5) (2010) 520. [95] P. Ifinedo Technology acceptance by health professionals in Canada: An analysis with a modified UTAUT model. In System Science (HICSS), 2012 45th Hawaii International Conference on (pp. 2937–2946) (2012). [96] E. Maillet, L. Mathieu, C. Sicotte, Modeling factors explaining the acceptance, actual use and satisfaction of nurses using an Electronic Patient Record in acute care settings: an extension of the UTAUT, Int. J. Med. Inf. 84 (1) (2015) 36–47. [97] K.M. Faqih, M.I.R.M. Jaradat, Mobile healthcare adoption among patients in a developing country environment: exploring the influence of age and gender differences, Int. Bus. Res. 8 (9) (2015) (142). [98] A. Chib, M.H. Velthoven, J. Car, mHealth adoption in low-resource environments: a review of the use of mobile healthcare in developing countries, J. Health Commun. 20 (2015) 4–34. [99] S. Dünnebeil, A. Sunyaev, I. Blohm, J.M. Leimeister, H. Krcmar, Determinants of physicians’ technology acceptance for e-health in ambulatory care, Int. J. Med. Inf. 81 (11) (2012) 746–760.