Keywords

1 Introduction

The rapid expanding of mobile devices, including smart phones and tablet computers, has significantly changed our lifestyle in recent years; the ways in which we can access, acquire, and share information have become increasingly mobile. More importantly, mobile devices have further evolved from portable to wearable technology such as smart watches (Samsung Galaxy Gear, Sony Smartwatch) and smart glasses (Google Glass). Wearable devices are especially effective in delivering health-related information because they are literally attached to users’ body and able to monitor users’ physical conditions (heart rate), which subsequently increases user awareness and acceptance of preventive health information [1]. In other words, it is to stay connected more closely to users’ body unlike smartphone.

There are increasing needs of health information largely. First of all, users prefer to be notified of the accurate environmental conditions. For example, Seoul Air Quality Information Center provides the service of air quality massage from October 2013. The service allows users to customize the each information following kinds of air pollution (yellow dust, fine dust and ozone) and level of notification (forecast, watch, warning and emergency warning). Second emphasis is being placed on providing not only actual care but health information that functions as preventive measures [2]. As new mobile devices and services offer more and more tailoring interface, it is important for researchers to understand these trends and effects in terms of human-computer interaction (Table 1).

Table 1. Participant Characteristics
Full size table

On the contrary, concerns for privacy are concurrently increasing according to monitor and collect personal health information and data via their smart watches using embedded sensors. Information tailoring always has generated information privacy issue inevitably, because it entails sacrificing some private information. Recently data privacy issue beyond information privacy has presented according to treat vast amount of data, namely Big Data, and work of human data interaction [3]. More seriously, users’ biometric data tend to be collected by oblivious and passive ways, because individuals are not be able to aware of every signals emitted as various health data. In this light of importance, we need to understand not only providing of preventive measures using wearable devices, but also many complex and subtle privacy concerns surrounding use of personal health data.

Therefore, it is both theoretically and practically critical to explore users’ perceptions and attitudes on smart watches as preventive health tools with emphasis on 2 sides: health information tailoring and data privacy. We aimed to obtain both quantitative and qualitative data to generate 3 research questions on how users aware these issues. As such a purpose, this qualitative study intends: (1) to explicate distinct difference of awareness and attitude on the health information tailoring, namely personalization and customization, via their smart watches by classifying power users and non-power users (2) to identify significant difference of awareness of data privacy concern according to power users and non-power users in delivering health contents via a wearable device, especially the smart watch.

2 Literature Review

2.1 Emergence of a Smart Watch as Preventive Health Tools

A smart watch is a wearable computer on the wrist (a wrist watch computer) with functionality of personal information access and notification [4]. Several smart watches have been developed starting from 2000 such as the IBM Linux Watch with no sensors like PDA (Personal Digital Assistant) on the wrist. Microsoft also released their smart watch (SPOT, Smart Personal Object Technology) from 2004 offering personalized information such as weather and news [5]. Although these devices remain a very small portion of information technology industry until now, the smart watches have been getting lots of attention from 2013, as giant companies like Samsung and Sony have released their products using touch screen linked to their smartphone (Galaxy Gear, SmartWatch). According to the research of Consumer Electronics Association (CEA), global smart watch market will grow to 1.5 million shipments during 2014, and sales of smart watches are predicted to grow exponentially, selling 44 million units globally by 2016.

Smart watches have been introduced to users as a preventive health tool recently. Many major ICT companies such as Samsung, Microsoft and Apple have all announced their watches focusing fitness related functions at 2014. Samsung released their new smart watch line (Gear Fit) including personalized fitness manager application providing real time information and results of user’s workout. Microsoft created a wrist worn fitness watch (Band) connected to their clouding service (Microsoft Health) and analyzes the tracking data for personalized fitness goals to text messages [6]. Apple also introduced their smart watch (iWatch) at September 2014 with emphasizing the device’s capability as a health monitoring watch. Apple has been partnered with many health care industries to get ready for their first wearable device [7].

It has strong emphasis on health related functions because it is attached and fitted to users’ wrist like a traditional wrist watch. In other words, it is to stay connected more closely to users’ body than any other devices like smartphone and tablet pc in terms of physical side. Furthermore, it is able to track and collect users’ physical data such as movement, location, skin temperature, and even heart rate using embedded sensors beyond the functionality of general smartphone. Taking advantage of the fact that this technology is constantly worn on the wrist, most of smart watches (Pebble, Basis band) which were released from small companies offer personalized health information like tracking data of heart rate, number of steps, calories burned, sleep and perspiration [8].

However, little is known about how users access and acquire their health information on the wearable devices and potential users aware these functions. As such a thing, it is valuable to explore qualitatively the perceptions pertaining to engaging with health contents in context of wearable devices, especially smart watches, which can be described by addressing the following research question.

  • Research Question 1: For smart watch users and potential users, what are their perspectives and views of wearable technology, especially the smart watches as preventive health tools?

2.2 Health Information Tailoring

Much of the literature on health communication has conducted researches on health information tailoring. ‘Tailoring’ defined as some of methods for creating communications in which individualized information is used for their receivers, with the expectation that this individualization will enhance effects of these communication and thereby promote desired changes [9]. A meta-analytic review of 57 studies on health communication found that tailored health information produce positive effect on health behavior changes in response to their messages [10].

Information can be tailored in two ways through digital media: personalization and customization. Personalization is system-initiated, such that the system (computer, website, application) tracks and identifies individual users’ needs, interests, knowledge, and goals [11]. Personalization collects data either by directly asking users for their name, gender, date of birth and geographical location or by indirectly observing and tracking user behaviors (web history, log data) [12]. Most of these services are automatic and hence require no direct user involvement. In sum, personalization is a system-initiated ways of tailoring. In contrast, customization is a highly user-driven process of information tailoring [13]. It emphasizes user control and involvement, placing users in the driver seat and allowing them to initiate their interaction with the system [14]. In sum, users have a relatively passive role in personalization, whereas in customization, they serve as active organizers of the information [15]. By extension, these different types of information tailoring are likely to have significant psychological effects on user awareness and attitude toward the conveyed health information, we are also able to classify according to users’ tendency and pattern of technology usage.

Users who prefer to customization have been classified as “power users or early adopters.” Power users spend much time on using various gadgets offered by various technologies [16] and tend to have a precise control over their interaction with digital functions including new features. They also prefer to control their interaction with technology as much as possible [17]. Therefore, power users tend to have strong self-efficacy and clear outcome expectation [12]. In contrast, non-power users lack the interest in adopting new technologies. They tend to avoid taking a charge in controlling the device due to the lack of self-efficacy. They would rather not select among options and spend their time controlling interface.

Some studies found that customization is more effective to promote positive effect on healthy behavior than personalization in a digital world. For example, participants who customize their own avatars show a higher tendency for self-preservation [18]. In addition, customization provides users with a strong sense of user and personal agency, according to the agency model [14]. It allows them to highly personal choices and active participation in the customization service more than personalization, and trigger sense of control and involvement. This self-schema enhances their self-determination, self-efficacy and autonomy in an online environment, especially customizable interfaces [19].

By extension, users who acquire customized health information via their smart watch are likely to have a more positive attitude toward the information than those who receive personalized health information. Therefore, these different types of users are likely to have distinctive awareness and attitude on tailored health information, which can be investigated by addressing the following research questions.

  • Research Question 2: For both power users and non-power users, is there distinct difference of awareness and attitude on the health information tailoring (personalization vs. customization) via their smart watches?

2.3 Data Privacy

Many empirical privacy researches in the field of social science have focused the effect of privacy concerns and dealt with the construct of privacy concerns related behavior and attitude variables. Some previous studies treated willingness to share personal information from personalized service [20], and used level of information disclosure attitude as their privacy related variables [21].

Privacy is one of the most sensitive and complex topic because of the inconsistencies in defining and measuring privacy and also because the important relationships depend more on perceptions than on rational assessments [22]. Studies in the field of Information Systems (IS) have considered to information privacy as one of the most salient issue in information and technology based services like online shopping and transaction [23]. Privacy concerns were defined as concerns reflecting individuals’ general worries about disclosure of personal information in many studies of IS [24]. But the concept of privacy concern recently is much more understandable in specific situation and context compared to general worries and concerns for privacy [25]. In other words, contextual privacy concerns need to distinct to general concerns about personal information and must be studied in each domains.

Information tailoring service like both personalization and customization are sacrificed some of the personal information necessarily, because the vendors should collect the personal data to provide their service and users have to put out their information and data. It is impossible to get without loss of privacy since users need to offer their preferences when they use the internet personalization services [20]. Following tailoring technology gradually invades to our digital life, some users certainly have much more worried about the amount of personal information that the system would collect and utilize automatically. These privacy concerns affected negatively to online consumers in transaction although the consumers were interested in e-commerce services. The most important problem is that way of tracking is performed in oblivion ways without the awareness and control of the users [26]. This effect is more unobtrusive to non-power users who lack the expertise and interest in adopting new technologies than power users who have a strong self-efficacy and motivation using features [12]. In other words, users who prefer customization service are shown to better protection and aware of their privacy than personalization, whereas non-power users may be oblivious to their disclosure problem. As such, privacy concern related to online users’ ability to control their conditions by acquiring their personal information and data [27].

Given the concerns for information privacy documented in the literature, users who participate to customize their health information and data are more likely to have a higher privacy protection and positive evaluation of health contents than non-power users. Therefore, these different types of users are likely to have distinctive perception and attitude on data privacy protection and concern, which can be examined by the following research question.

  • Research Question 3: For both power users and non-power users, what are the significant difference of perception and attitude on concern for health data privacy related to their smart watches?

3 Methods

This study was conducted multiple methods of quantitative and qualitative data collection combining pre online survey, focus groups interviews and post in-depth interviews. The researchers analyzed the data of survey and transcripts of interviews by using QSR International (NVivo 10) with simple counting method [28] to conduct content and discourse analysis.

The participants were recruited from online communities for the smart watch users in South Korea and a large private university in Seoul, Korea. Preliminary online survey was used to make sure the participants’ basic attributes related to technology, health and privacy issues during the recruiting process. In total, 12 participants (6 males, 6 females) were interviewed and ranged in age from 21 to 41 years, with a mean age of 31 years (SD = 5.74). Each participant gave written informed consent to participate before the interviews.

3.1 Pre online Survey

This study was used by pre online survey to make sure index of power user, e-health literacy and concern for privacy before the main interviews. The participants were asked to respond to the questionnaire by marking on a 7 point Likert scale (1 = “strongly disagree”, 7 = “not at all”). In order to measure power user scale (α = 0.87), 6 items are adopted from the previous study [17]. The items are included such as “I love exploring the features that any technological gadget has to offer,” “Using any technological device comes easy to me,” and “Using information technology gives me greater control over my work environment.”

The aim of the pre online survey was to make adequate interview questions regarding health information tailoring and data privacy issue via smart watches and classify the users according to their self-reported index. All collected data was stored to Microsoft Excel file format and analysed by PASW Statistics 18.

3.2 Focus Groups Interviews

This study was conducted the 3 sessions of focus groups interviews with 4 participants each in February 2015. Each session was offered in an atmosphere of freedom by the researchers and a moderator. Each interview was lasted about 90–100 min and all procedures were audiotaped and transcribed by the researcher. All questions and answered were totally opened and the participants were asked to talk about their views and personal experiences related to using smart watch. All participants including power users asked to wear the smart watch (Sony, SmartWatch2) during the interview and freely browse the health contents via there smart watch for 5 min.

The main goal of the focus groups interview was to identify the perspectives, awareness, attitudes, perceptions and intrinsic motivation of users with respect to using a smart watch as a preventive health tool, with emphasis on health information tailoring and concern for data privacy.

3.3 Post In-depth Interviews

Our researchers undertook additional in-depth interviews personally through online chatting service (Kakao Talk) lasted about 15-20 min. We prepared new questions and asked to answer additional personal experience for discussing changed views and needs after online survey and focus groups interviews in depth.

4 Results

The researchers analyzed the results of online survey and the transcripts of 2 interviews for emerging themes regarding the smart watch as preventive health tools with health information tailoring and data privacy concern. Three main themes emerged: (1) useful high-tech toy, (2) needs of hybrid tailoring service and (3) unnecessary anxiety vs. vague fear. The characteristics of participants are as below.

4.1 Useful High-Tech Toy

There were significant two sides of perspective to using the smart watch as a preventive health tool: useful health kit but a toy of a wrist watch. Most participants (10/12) reported the potential value of the smart watches as for managing physical conditions daily. All of them mentioned usefulness and convenience of these health related functions to check their activity and conditions. As one participant said it:

  • It is surprisingly tiny and also wearable easily unlike other digital devices… It may be my personal assistant who manages my health conditions daily and automatically… It has a potential value to access, collect and manage my physical data conveniently (participant 12).

However, most participants (11/12) also expressed their worrying about health related functions. They raise various questions for inaccuracy, untrustworthiness and possibility of generalization. Some participants pointed these:

  • They (smart watches) are just funny and attractive accessories yet… I still do not believe their accuracy of digital sensors. It has a many errors collecting physical data because of their technical problems (participant 8).

  • It will be attractive but not main device like smart phone… I have a doubt that the device can collect real data and show complex statistical health information… their small screen often makes me face (participant 9).

In sum, most participants mentioned that they have a positive view of using smart watches as preventive health tools, but at the same time they expressed that the smart watch is just wearable high tech toy not a medical device for actual health care.

4.2 Needs of Hybrid Tailoring Service

According to our analysis of survey and interviews, there was no significant difference between power users (6/12) who showed power user index over 6 and non-power users (4/12) who recorded the index under 3 in terms of information tailoring. Half of users (3/6) who showed high index of power users prefer to use personalization service than customization. Likewise, half of non-power users (2/4) reported to use customization not personalization on the internet. In addition, some users (3/12) identified that they prefer both ways of tailoring depending on the service. So we reclassified users with their self-reported preference of using the information tailoring service as three types: personalization, customization and both.

Some users (5/12) who only prefer to personalization service said that they would still use personalization in health information tailoring via their smart watch. They expressed ease of use, annoying from a lot of information, fatigue of digital information and reducing the time. Some of them (4/5) noted that they have a plan for using customization in their smart watch if the device could collect sensitive biometric data such as iris and fingerprint beyond their activity.

  • I have used movie recommendation service and I feel much great tailoring… Sometimes I am very tired of much useless visual information and it makes me really annoying, stressful… I prefer to get my every information automatically because I have no much time to get all information (participant 11).

  • I will participate to organize and put my information actively although it is annoying to me… I will have to manage directly my valuable private data like health record… I think the information provider need to consider way of tailoring according to characteristics of data not economic interests (participant 2).

On the contrary, some users (4/12) who select customization reported that they will definitely use this way of health information tailoring through their smart watch. They expressed unsatisfactory experience and untrustworthiness of system, strong needs of accurate information and enjoyment of engaging with process. Two of these participants pointed out it:

  • It (personalization) was not enough to meet sensitive consumers… We need to provide our detail preference to get high-quality information… Automatic tailoring system cannot satisfy detailed demands, it just provide superficial contents, not personal (participant 10).

  • I enjoy the participating to match my preference from A to Z… I feel better to get accurate information although it spends some time to use (participant 9).

Half of them (2/4) also expressed the needs of mixed tailoring service in the second interview. They wanted to get both insightful information and convenience. They also reported that they have used many personalization services actually without awareness. A few participants (3/12) who enjoy both services mentioned that there is no significant difference between personalization and customization. Two participants argued that hybrid way of tailoring is the advanced service and will be the adequate way in context of health information via smart watch. These opinions are as below:

  • We are exposed to a lot of personalized services as mobile devices… Actually I have no choice but to use some services because we have to login… The system has provided me with some information automatically and maybe health data will be collected in the same way if users would not aware their data (participant 5).

  • I do not think two way of tailoring is different. This is a single procedure to provide personal information… Mixed way is a much superior service… I am willing to provide any data to get deep info if there is a special family history… We need also compulsory ways to prevent unexpected health problems (participant 10).

4.3 Unnecessary Anxiety Vs. Vague Fear

The researchers classified as power users (6/12) who recorded power user index over 6 scales and non-power users (4/12) who showed same index under 3. There is a distinct difference of perception about concern for health data privacy via there smart watches between power users and non-power users. However, this analysis showed different results instead of the previous studies and findings [14, 18, 19]. Power users reported various level of their concern for data privacy, but non-power users expressed rather deep concern and strong interest about their health data privacy.

Half of power users (3/6) mentioned that concern for health data privacy is an unnecessary anxiety. They focused on positive health related functions and chances to use physical data personally. Some participants said it:

  • Bio data without process and analysis is just digital signals… worry about health data is useless concern like fear… The most important thing is to construct safe firewall and develop advanced security system not worry (participant 10).

  • Any digital records and data have a bad side to exposure… but the advantages are much bigger than this negative side… new wearable technology including the smart watch will give us opportunities from check our conditions to prevent disease easily… privacy issue will be covered by these (participant 1).

  • Privacy problems of course will be occurred… I have felt much dissatisfaction at managing my health record. Even now I cannot access it easily although it is mine, my basic right… It can be innovative turning point by having the right of managing my health data on my own as smart watch collects the data (participant 12).

But to the contrary, all of non-power users (4/12) said that they have strong concern for exposing their personal information including health data. The results are totally opposite effects compared to our literature review which founded non-power users have lower protection of online privacy than power users [27]. The 4 participants expressed their some motivation and experience of general privacy issue, untrustworthiness of security system, importance of life and body.

  • Health data are related to our life and body directly… I cannot imagine that my health data will be used for other purpose in spite of myself… I strongly give my concern about collecting my any health data automatically even my devices (participant 2).

  • I heard a lot of problems, victims of online hacking like their bank account, ID from some media… I felt much fear to use internet services and cannot trust the security anymore increasingly whenever I heard these… Smart watch also collects data and submits to others to provide information, so it is not area of my wrist but others I could not know (participant 9).

  • I am concerned that my personal information is open all the time… but I have no special solution to keep my information in online. This worry is more serious in mobile environment… electronic health record system arouse my attention especially health application of wearable device (participant 3).

5 Discussion

Our qualitative research indents to explore the user perceptions of a smart watch as a preventive health tool with focus on 2 subjects: health information tailoring and data privacy. The results of this study showed that there are 2 sides of perspective of a smart watch as a preventive health tool: useful health kit but a toy of a wrist watch. In addition, the study indicates that there is no significant breakpoint between power users and non-power users according to their preference of 2 ways of information tailoring. Finally, we found that concern for health data privacy was divided into useless worry and vague fear according to the difference between power users and non-power users. Our findings will offer theoretical and practical insights for information providers, UX designers, and HCI scholars who are interested in the role of smart watches as emerging preventive health tools.

The current study has several limitations. Firstly, we relied on survey and interview accounts instead of observation of actual using. Naturalistic observations which are able to check users’ actual behavior would be necessary for further verification of power users’ index. Secondly, our interviews were conducted to a small number of participants. We need to conduct additional interviews to more random sample of participants. Finally we have concerns for generalizability. It is hard to generalize our findings to other contexts because the data were only collected from the limited demographic participants in Seoul, South Korea.

In the future, we will focus on improving the methodological tasks to identify users’ intrinsic motivation and actual behavior in context of the smart watch. In doing so, the future study would be used observational or experimental method with a larger sample and wider population to obtain more generalizable founding. Additional qualitative researches are needed to design innovative information tailoring services for promoting users to engage with health contents on the wearable technology.