ArticlePDF AvailableLiterature Review

Abstract and Figures

Wearable activity trackers (WAT) are electronic monitoring devices that enable users to track and monitor their health-related physical fitness metrics including steps taken, level of activity, walking distance, heart rate, and sleep patterns. Despite the proliferation of these devices in various contexts of use and rising research interests, there is limited understanding of the broad research landscape. The purpose of this systematic review is therefore to synthesize the existing wealth of research on WAT, and to provide a comprehensive summary based on common themes and approaches. This article includes academic work published between 2013 and 2017 in PubMed, Embase, Scopus, Web of Science, ACM Digital Library, and Google Scholar. A final list of 463 articles was analyzed for this review. Topic modeling methods were used to identify six key themes (topics) of WAT research, namely: 1) Technology Focus, 2) Patient Treatment and Medical Settings, 3) Behavior Change, 4) Acceptance and Adoption (Abandonment), 5) Self-monitoring Data Centered, and 6) Privacy. We take an interdisciplinary approach to wearable activity trackers to propose several new research questions. The most important research gap we identify is to attempt to understand the rich human-information interaction that is enabled by WAT adoption.
Content may be subject to copyright.
Wearable Activity Trackers, Accuracy, Adoption,
Acceptance and Health Impact: A Systematic Literature
Review
Grace Shin, Mohammad Hossein Jarrahi; Fei Yu; Amir Karami; Nicci Gafinowitz; Ahjung
Byun; and Xiaopeng Lu
Abstract
Wearable activity trackers (WAT) are electronic monitoring devices that enable users to track and
monitor their health-related physical fitness metrics including steps taken, level of activity, walking
distance, heart rate, and sleep patterns. Despite the proliferation of these devices in various
contexts of use and rising research interests, there is limited understanding of the broad research
landscape. The purpose of this systematic review is therefore to synthesize the existing wealth of
research on WAT, and to provide a comprehensive summary based on common themes and
approaches. This article includes academic work published between 2013 and 2017 in PubMed,
Embase, Scopus, Web of Science, ACM Digital Library, and Google Scholar. A final list of 463
articles was analyzed for this review. Topic modeling methods were used to identify six key
themes (topics) of WAT research, namely: 1) Technology Focus, 2) Patient Treatment and
Medical Settings, 3) Behavior Change, 4) Acceptance and Adoption (Abandonment), 5) Self-
monitoring Data Centered, and 6) Privacy. We take an interdisciplinary approach to wearable
activity trackers to propose several new research questions. The most important research gap we
identify is to attempt to understand the rich human-information interaction that is enabled by WAT
adoption.
Keywords: Wearable activity trackers, personal informatics, ubiquitous computing, quantified
selfers, physical activity, acceptance and adoption (abandonment), medical informatics
Citation: Shin et al (2019) Wearable Activity Trackers, Accuracy, Adoption, Acceptance
and Health Impact: A Systematic Literature Review, Journal of Biomedical Informatics, 29
(May 2019), 103153.
1.Introduction
A growing number of wearable activity trackers (WAT) (e.g., Fitbit, Xiaomi, Garmin, Samsung
Gear Fit) provide an opportunity for self-monitoring and the potential to change personal behavior
towards a healthier lifestyle. According to the International Data Corporation (IDC) Worldwide
Quarterly Wearable Device, the overall market for wearables is expected to grow from 113.2
million units sold in 2017, to 222.3 million in 2021 (IDC, 2016.). Out of approximately 26.3 million
WAT units sold in the third quarter of 2017, the most popular devices were Fitbit (13.7%) and
Xiaomi (13.7%), and the Apple Watch (10.3%) (IDC, 2016).
These WAT are typically attached to the body, mainly the wrist, to help enhance or complement
human health management capabilities. They make it easier for users to generate health-related
data and to track their daily activities such as step count, calories burned, heart rate and even
sleep pattern. WAT also quantify user behavior and state of health, and automatically upload their
data to mobile apps and linked websites. Furthermore, the data generated from WAT have shown
significant value in patient care and self-health management in clinical settings (Bellicha et al.,
2017; Thomas et al., 2017; Mercer et al., 2016), particularly for patients with lifestyle diseases
such as obesity and diabetes.
Research on WAT has been carried out by a variety of intellectual communities, often with
divergent disciplinary interests. These include psychology (e.g., Karapanos et al., 2016; Butryn et
al., 2016), health sciences (e.g., Yavelberg et al., 2018; Gillinov, et al., 2017), medical
engineering (e.g., Adam Noah et al., 2013; Dontje et al., 2015), human-computer interaction (e.g.,
Munson et al., 2015; Tang et al., 2016), behavior science (e.g., Washington et al., 2014; Valbuena
et al., 2015), computer science (e.g., Randriambelonoro et al., 2015; Epstein et al., 2016), and
sports science (e.g., Smith, 2016; Stahl et al., 2016; Sears et al., 2017). Few have attempted to
summarize the body of WAT-related research to date, or to examine how research trends have
evolved over time. Of those, most are limited to sub-topics or specific populations of research
subjects. For example, Evenson et al. (2015) presented a systematic review of 22 laboratory-
based studies focused only on the validity and reliability of WAT that register step, distance,
physical activity (PA), and energy expenditure. Furthermore, only studies of popular WAT, namely
Fitbit and Jawbone Up, were examined. Coughlin and Stewart (2016) used bibliographic searches
in PubMed and behavioral sciences collections in their review of research that considers the
effectiveness of WAT in promoting PA and weight loss. Lewis et al. (2015) systematically
reviewed eleven empirical studies which centered on PA interventions, and synthesized the
efficacy and feasibility of data collection mechanisms.
Despite the proliferation of WAT since their commercial introduction in 2013, and growing
interest in the use and impact of these devices in varying contexts, research is fragmented
across disciplinary lines. There is a clear need for a systematic review of the broader landscape
of relevant literature, which provides a coordinated view and agenda for future research. The
purpose of this article is to present a comprehensive overview and synthesis of the state of
research on WAT performed between 2013 and 2017. We use topic modeling methods to
identify common topics being investigated, and analyze the research methods used and
characteristics of study participants. In addition, this review identifies important gaps in the
current body of published literature, outlining directions for future research. The primary
contributions of this paper are twofold: to understand key research areas, and trends and to
create cross-awareness among different professional disciplines interested in the use and
impact of WAT by providing a holistic perspective.
2. Research Methods
Extraction of articles
A literature search was undertaken of recent English language WAT-focused research.
Databases consulted included PubMed, Embase, Scopus, Web of Science, ACM Digital Library,
and Google Scholar. It is important to note that Google Scholar is increasingly considered a viable
source for literature search and review. Even though it may include non-peer reviewed articles,
recent sociometric research delineates its utility in offering a broader coverage of research for
most scholarly traditions (Harzing and Alakangas, 2016; de Winter et al., 2014).
The search encompassed articles published after 2013, reflecting the release dates of the most
commercially popular WAT: Jawbone Up in November 2012 and Fitbit in May 2013. We used the
following keywords: “wearable activity trackers”; “wearable fitness trackers”; “wearable activity
monitor”; “wearable activity tracking devices”; “consumer-activity trackers”; “consumer fitness
tracking device” and “consumer-wearable activity trackers”.
Inclusion and exclusion criteria
We excluded articles which: 1) primarily involved traditional pedometers (which only register
information through mechanical means); 2) experiential research on WAT or prototypes
developed in labs (e.g., UbiFit (Munson and Consolvo, 2012)); 3) studies of mobile tracking
applications (e.g., Apple ihealth app (Goyal et al., 2016)); 4) studies of the hardware architecture
of WAT (e.g., Ahn et al., 2016).
1247 articles were initially retrieved, and refined to 463 for this literature review. We included
extended abstracts for conferences (e.g., Ayobi, 2016; Tang; 2016) and excluded sources in
languages other than English, non-peer reviewed articles, letters to the editors, magazine articles,
and patents. 525 sources did not meet the inclusion criteria and 259 duplicate items were removed
from the final corpus. Three authors independently examined retrieved articles, and any
disagreements about inclusion were resolved through conversation. Articles’ data were extracted
manually and included: author(s), title, journal name, publication year, abstract, keywords,
conclusion.
Data analysis
Data analysis builds on topic modeling techniques, which help the discovery of meaningful
semantic patterns (topics) and the clustering of large volumes of otherwise unlabeled or
unstructured documents (Karami et al., 2015a; Karami et al., 2015b; Aggarwal and Zhai, 2012).
This review aims to provide a holistic perspective that represents multiple sides and approaches
(Cooper, 1988). Topic modeling enables a more objective synthesis of articles and key themes,
by applying computational techniques to minimize bias associated with relying exclusively on a
subjective interpretation of research data (Karami, 2015). Computational approaches to literature
analysis may provide greater validity, thus offering a more objective approach to identifying
relevance and connection between articles in literature reviews (Mortenson and Vidgen, 2016).
One of the most widely-used and robust topic modeling approaches (Lee et al., 2010), latent
Dirichlet allocation (LDA), has been applied in a wide range of domains such as analysis of health
and medical corpi (e.g., Webb et al., 2018; Karami et al., 2018a; Karami et al., 2018b; Shaw and
Karami, 2017; Ghassemi et al., 2014), election data (e.g., Karami et al., 2018), and scientific and
historical topics (e.g., Griffiths and Steyvers, 2004). The methodology assumes that each topic is
a distribution over words and each document is a mixture of the topics in a corpus (Karami et al,
2018c; Blei et al., 2003).
The MALLET implementation of LDA based on a Java platform using 1000 iterations (McCallum,
2002) was used in this study to identify key topics and keywords in the abstract and conclusions
of the collected articles. LDA was subsequently re-applied using a different number of topics,
ranging from 5 to 30, and the most precise number of topics (6) was estimated, based on the trial
and error strategy recommended by Chang et al., (2009). LDA enabled us to define 6 distinct
topics (see Table 1) by grouping together semantically related terms (e.g., diabetes, control,
condition, risk, and mHealth). It also automatically assigned articles to different topics. Finally, we
identified the core themes that define the 6 topics (e.g., privacy or behavior change) based on the
keywords assigned to each topic by LDA and the articles it most strongly associated with them.
3. Results
Analysis of the 463 wearable activity tracker (WAT) studies shows a significant growth rate in
annual publication between 2013 and 2017 (see Figure 1). Around 200 articles were published in
2017, more than twenty times the number that had appeared just four years earlier.
Figure 1. Number of WAT studies per annum between 2013 and 2017
Topic modelling analysis enabled researchers to identify six thematic topics central to these
studies: 1) Technological Focus; 2) Patient Treatment and Medical Settings; 3) Behavior Change;
4) Acceptance and Adoption (Abandonment); 5) Self-monitoring Data Centered and 6) Privacy.
As noted, analysis of the terms and keywords connected to each topic, as well as the articles
most strongly associated with each of the topic categories, helped us identify/define each topic.
Table 1 displays, for example, that the terms ‘usability’, ‘acceptance’, ‘users’, and ‘interviews’
appear frequently in Acceptance and Adoption (Abandonment) themed studies. These words
appear less frequently in other categories such as Behavior Change or Privacy. Similarly, the
words ‘accuracy’, ‘validity’, ‘reliability’, and ‘accurate’ appear frequently in studies that we infer to
have had a Technology Focus theme.
Topic/theme
Frequently used terms in each topic topic
Technology Focus
activity; physical; system; accuracy; accurate; validity; algorithms; reliability; sensors;
daily; steps; active; significant; week; monitor; increase; data; methods; assess;
body; analysis; walking expenditure; energy; actigraph; compare
Patient Treatment and
Medical Settings
diabetes; behavior; messages; control; condition; cognitive; risk; mPFC; model;
states; type; health technologies; care; disease; mhealth; chronic patients; support;
improve; medical; adherence; clinical; duration; cancer; treatment; metrics; measure
Behavior Change
children; students; learning; change behavior; fitness; interventions; school;
adolescents; enhanced; sedentary; trial; self-quantification; healthy; social; support;
group; time; sitting; workers; control; strategies; feasibility; incentives; effectiveness;
increasing; levels; potential
Acceptance, Adoption and
Abandonment
adults; older; wearable trackers; participants; usability; acceptance; context; users;
design; engagement; understanding; interview; activity; data; age; technology;
people; motivation; social; life
Self-monitoring Data
Centered
data; health; information; sensors; monitoring; collection; individual; personal;
consumer; future; systems; applications; based; important; provide; approach;
medical; services
Privacy
data; privacy; healthcare; information; smart; challenges; collection; internet
solutions; system; security; services; mobile; things; home; video
Table 1. Key themes and associated terms
Figure 2 outlines the distribution of these research topics/themes in material published since
2013. Publications with a ‘Technology Focus’ make up the largest portion (26%), closely followed
by papers featuring ‘Patient Treatment and Medical Settings’ (23%). ‘Behavior Change’,
‘Acceptance and Adoption (Abandonment)’, ‘Privacy concerns’, and ‘Self-monitoring Data
Centered’ themed works make up the balance.
Figure 2. Distribution of main topics in wearable activity tracker (WAT) studies,
published between 2013 and 2017
Figure 3 shows the change over time (2013 - 2017) in the annual proportions of publications’
study themes. Technology Focused research (average 38.47%) is consistently of greatest
interest, whilestudies of WAT for Patient Treatment and Medical Settings have increased over
over the study period. This may reflect the fact that research centered on improving the
technological functionality of these WAT and their medical implications has gained more traction
over the past few years. Interestingly, research with a technology focus dropped significantly from
38.47% in 2013 to 19.89% in 2014. In contrast, research on Acceptance and Adoption
(Abandonment) increased significantly in 2014, from 10.92% to 28.36%. An important report
published in 2014 may provide an explanation. Ledger and McCaffrey (2014) surveyed U.S.
consumers who owned WAT and revealed that half had ceased using the device. Such reports
raised subsequent concerns over usability issues that could have hampered long-term
engagements with WAT.
Figure 3. The evolution of research themes from 2013 to 2017
The remainder of the Results section provides more details and examples in relation to these six
key research themes.
Theme 1: Technology Focus (reliability, accuracy and validity) (121 articles)
These studies primarily assess the technological functioning of WAT (see Table 2), and seek to
provide design implications for improving features and performance, particularly in relation to what
Codella et al. (2018) term the ‘data quality’ provided by activity trackers. Research in this field is
typically focused on accuracy, validity, and important features related to data collection and
analysis (e.g. algorithms for registering and analyzing energy expenditures associated with
different levels of PA). WAT research with a technology focus examines the quality of data
collection for both healthy adults and different populations of patients (e.g., Thorup et al., 2017;
Wong et al., 2018; Takacs et al., 2014; Singh et al., 2016; Husted and Llewellyn, 2017; Beevi et
al., 2016; Leth et al., 2017).
A common approach used here is the comparison of multiple devices, or across devices or mobile
applications, in terms of the accuracy and reliability of data collection and representation. For
example, Fitbit and Nike+ Fuelband were used to test accuracy in detecting steps for individuals
with stroke and brain injury in Fulk et al., (2014). Similarly, other studies evaluated and compared
the accuracy of step count tracking by various WAT devices such as Fitbit, Nike+ Fuelband,
Jawbone Up, Microsoft Band 2 (MB), Omron HJ-303, Flyfit, Sportline Fitness Monitors, and A&D
Activity Monitor (e.g., Case et al., 2015; Husted and Llewellyn, 2017; Fokkema et al., 2017).
Moststudies in this theme drew on experimental research (e.g., Alinia et al., 2017; Ferrara et al.,
2017; Case et al., 2015), usually conducted in laboratory settings, focusing on PA such as walking
on a treadmill or normal jogging, rather than investigating the technology in natural settings. Only
a few studies collected data outside the laboratory, but over short periods of time: Chu et al.
(2017) required participants to wear their WAT for 7 days; and Kooiman et al. (2015) required
participants to wear WATs for 30 mins in a laboratory setting and one day in free-living conditions,
to test device reliability.
Experimental studies or short-term perspectives are well-suited to study the technological
performance of WAT (e.g., Diaz et al., 2016). However, these studies tended to be more
concerned with technological performance, and therefore provide a narrower perspective on
users’ interactions with the system (e.g., Case et al., 2015). The experimental research method
may not discover the social realities and user behaviors that may be formed over an extended
period of time in a natural environment (e.g., Jarrahi et al., 2017).
The development of WAT has progressed rapidly, and research on their accuracy has not kept
pace with the diversity of devices in use. As mentioned in the Introduction, the most commercially
popular wearables in this study period were Fitbit (13.7%) and Xiaomi (13.7%), based on sales in
the third quarter of 2017. This is despite the fact that the Xiaomi device entered the commercial
market in 2017, and little research has been performed on its accuracy or reliability. Further
examples reflecting the difficulty of reporting on current developments are the studies that
employed the Nike+ Fuelband device to test WAT accuracy and reliability (e.g., Kooiman et al.,
2015; Case et al., 2015; Fulk et al., 2014; Guo et al., 2013; Lee et al., 2014). Yet Nike had
discontinued distribution of FuelBands in April 2014 and their newest product, the Apple Watch
Nike+, features different technological properties (Rita Guerra, 2015).
Author (year)
Research
Focus
Sample
Background
Research
Method
Device Usage for data
WAT model
Adam Noah et
al. (2013)
Validity
Participants with no
cardiovascular,
neuromuscular or
orthopaedic
conditions
Experiment
6-min sets of treadmill
walking, jogging and stair
stepping.
Fitbit
Actical
Fulk et
al.(2014)
Comparison
Participants with
chronic stroke and
traumatic brain injury
Experiment
Two minute walk test
(2MWT)
Fitbit
Nike+ Fuelband
Lee et al.
(2014)
Validity
Healthy adults
Experiment
Different devices worn
simultaneously while
completing a 69-min
protocol.
Fitbit One, Zip
Jawbone Up
ActiGraph
Basis
DirectLife
BodyMedia FIT
Case et al.
(2015)
Comparison
Healthy adults
Experiment
Each participant were
asked to wear all the
devices: Twice on a
treadmill set at 3.0 mph for
500 and 1500 steps at the
same time
Fitbit
Nike+ Fuelband
Jawbone Up
Pedometer
Mobile apps
Kooiman et al.
(2015)
Validity and
Reliability
University staff
working in the office
Experiment:
laboratory &
free-living
conditions
Participants were asked to
wear 10 different trackers:
Walk for 30 minutes on a
treadmill at 4.8 km / h
walking speed and one
working day in free-living
conditions
Lumoback
Fitbit
Nike+ Fuelband
Pedometer
iPhone Moves
Misfit Shine
Withings Pulse, etc.
Diaz et al.
(2016)
Validity
Adult females
Experiment
Four-phase treadmill
exercise timed session
Fitbit One
Fitbit Flex
Kaewkannate
et al. (2016)
Comparison
Real users of
wearable fitness
trackers
Experiment
Subjects wore all four
devices for 4 weeks
Withings Pulse
Misfit Shine
Jawbone Up
Fitbit Flex
Chu et al.
(2017)
Comparison
Adult participants (35
male; 69 female)
Experiment
Fitbit Flex and ActiGraph
devices worn at the same
time for 7 days
Fitbit Flex
ActiGraph
Table 2. Examples of WAT studies with a Technological Focus
Theme 2: Patient Treatment and Medical Settings (104 articles)
Figure 2 suggests that research on the subject of WAT device use for Patient Treatment and
Medical Settings is the second most common research theme, and empirical interest in this
domain has increased steadily over time (see Figure 3).
Patients’ use of WAT overcomes barriers of conventional health measurements and allows
collection of new types of patient monitoring information (e.g., steps (Cook et al., 2013), heart rate
(Kroll et al., 2016), and sleep (Gruweza et al., 2017)). This in turn creates opportunities to better
manage patients’ functional status assessment and support beyond the clinical setting (Chum et
al., 2017). For example, conventional health measurements may involve medical tests performed
either in medical offices (e.g., weight and blood pressure) or may be built on clinical laboratory
tests (e.g., blood tests measuring glucose and cholesterol levels). Although such tests reveal
significant information about a patient’s health status, they do not explicitly account for, for
example, nutritional habits (e.g., food consumption, nutritional content, number of calories) and
physical fitness activities over an extended period of time. Furthermore, there are several barriers
to acquiring and reporting functional assessments of mobility during hospitalization, such as
reliable measurement reporting, consistent reporting, and data retrieval capabilities (Cook et al,.
2013).
Studies focused on this theme address some of these issues, and examine the feasibility of using
WAT for monitoring and rehabilitating patients (see Table 3). As shown in Table 1, keywords in
this group of publications reveal a frequent focus on chronic diseases, such as ‘diabetes’, and
‘cancer,’ which require lifetime monitoring of PA and dietary habits (Asif, 2014; Kushi et al., 2012).
These are among the most studied patient subpopulations who may benefit from ‘clinical’ and
‘treatment’ designs based on WAT. For example, Fitbit devices have been used to measure
mobility recovery during hospitalization after cardiac surgery (Cook et al., 2013); examine activity
levels of participants before and after knee arthroplasty (Roe, et al., 2016), and to remotely
monitor patient postoperative mobility (Appelboom et al., 2015).
Overall, studies focused on the Patient Treatment and Medical Settings topic illustrate the great
potential of incorporating WAT into medical settings. Between 2013 and 2016, studies were more
focused on the usability of WAT devices, and their feasibility as measurement tools of patients’
recovery. From 2017, studies reflect a greater focus on the data generated by WAT, and their
potential use in monitoring patient-generated data in hospital settings, or providing more timely
feedback that enables faster planning and intervention (Shinde et al., 2017; Gresham et al., 2017).
A systematic study by Van Remoortel et al., (2012), conducted before Fitbit was available,
revealed that activity monitors had been appropriately validated for use in assessing physical
activity in chronic patient populations. The authors concluded from their analysis that, due to low
accuracy in recording slow walking speeds and lack of information about validated activity
monitors in chronic patients, the need remained for proper validation studies for the population.
More recently, the validity of Fitbit was tested among groups of patients with COPD (Chronic
Obstructive Pulmonary Disease) (De Sousa Sena et al. 2015); results suggest that Fitbit One is
a valid device for counting steps in patients with COPD and feasible for long-term usability.
Similarly, Treacy et al. (2017) studied the validity of WAT for measuring step counts in
rehabilitation inpatients; results support that Fitbit One and StepWatch were accurate in counting
step counts of slow-walking individuals.
In most studies, number of steps were used as the primary parameter to measure or monitor
patients for functional recovery after surgery (Appelboom, 2015), mobility recovery after major
surgery in elderly populations (Cook, 2013), major depressive disorder (Chum et al. 2017), cancer
patients (Shinde et al. 2017), children with lymphoblastic leukemia (Hooke et al. 2016), etc. In
addition, in some studies, heart rate was used to monitor hospitalized patients for clinical
deterioration (Kroll et al. 2016).
Researchers have also examined the possibility of augmenting patient records during
hospitalization with data from WAT (Shinde et al., 2017; Sprint et al., 2017). Such data could
create new possibilities for clinicians, who would no longer be limited to relying on information
from patient-reported outcome (PRO) questionnaires, which can be burdensome to patients,
particularly when performed in serial (Shinde et al., 2017).
Author (year)
Research
Focus
Sample Size
Sample Background
Research Method
WAT Usage for data
WAT model
Cook et al.
(2013)
WAT use in
recovery
128
Postoperative,
cardiac surgical
population
Quantitative
(collect steps from
Fitbit device)
5-7 days
Fitbit
Appelboom et
al. (2015)
WAT use in
recovery
27
In-patients
Experiments
Patients asked to wear a
FitBit Zip device for 10 to
15 minutes of all
sessions
Fitbit Zip
Roe et al.
(2016)
WAT use in
recovery
54
Postoperative, knee
arthroplasty surgery
Observational
Series
5-7 days at 4 time
periods: before surgery,
day 1-3, 6 weeks and 12
month after surgery
Fitbit
Gresham et al.
(2017)
Data use in
hospital setting
35
Patients, most with
gastrointestinal
cancers
Experiment
Device worn for 3
consecutive clinic visits.
Fitbit Charge HR
Shinde et al.
(2017)
Data use in
hospital setting
35
Cancer patients
Experiment
Device worn
continuously through 3
consecutive clinic visits
Fitbit Charge HR
Schrager et al.
(2017)
WAT use for
physicians
30
Emergency
medicine residents
Cohort study
6 months
Fitbit
Table 3. Examples of WAT studies focused on Patient Treatment and Medical Settings
It is important to note that the methodology with which we initially designated papers to ‘Patient
Treatment and Medical Settings’ is based on their predominant study focus, and whether they
were primarily directed at the medical implications of WAT, particularly for different patient
populations. As discussed in the Methods section, our classifications were additionally guided by
keyword pools.
A notable feature of our review of the WAT research corpus, was the number of studies that
connected secondarily with medical contexts in addition to their primary focus. For example,
Kumari and Hook (2017) examined patients’ privacy concerns while using WAT in medical
settings. To evaluate the number of articles with other categories of focus (Privacy, Data,
Technology Focus, Adoption and Behavior Change) that are also related to medical settings, we
performed an additional bibliometric analysis (see Table 4).
The steps for the bibliometric analysis are as follows:
Isolated 104 papers with the theme "patient treatment and medical settings" from our
total 463-record list
Created a list of terms we would like to search in the remaining 359 citation records
which included, “doctor,” “physician,” “nurse,” “clinician” “disease,” “hospital,” “patient,”
“diabetes,” “obesity,” “overweight,” “caregiver,” “medical,” “medicine,” and “clinical.”
Created a rule to use the above terms to search the title, abstract, and conclusion fields
of those records. A match in any of these fields was used to create a new sub-list of
citations.
Grouped new list of records that related to medical settings and manually checked the
relevance of each. Some irrelevant records extracted by the rule were excluded.
Theme
Number of Papers related to medical settings
(%)
Technology Focus
27 (31.76%)
Behavior Change
18 (21.17%)
Acceptance, Adoption and Abandonment
17 (20%)
Self-monitoring Data Centered
16 (18.82%)
Privacy
7 (8.25%)
Total
85 (100%)
Table 4. Number of WAT research articles focused on other study categories, which also
included Medical Settings
Theme 3: Behavior Change (83 articles)
Investigations under this theme focus on whether the use of WAT can positively affect PA, or
other health-related behaviors. This research explores whether wearing WAT enables users not
only to easily track their patterns of health and well-being, but also to achieve meaningful
behavioural change towards self-improvement (see Table 5). As shown in Table 1, among
frequently used keywords in this body of research were ‘sedentary’, ‘sitting’, ‘intervention’, and
‘self-quantification’.
Evidence suggests that sedentary behavior has adverse effects on health in both the short and
long term (Healy et al., 2011; Matthews et al., 2012; Pina et al., 2012). Growing evidence indicates
that breaking up participants' prolonged sedentary time for even a short period of PA can have a
positive impact on health (Healy et al., 2011; Matthews et al,. 2012). Use of WAT has been
considered to have created opportunities for users to modify their health behavior, through self-
monitoring, goal setting, and reinforcement (Evenson et al., 2015; Mercer et al., 2016; Miyamoto
et al., 2016). WAT use also enhances user self-awareness, and self-knowledge, and the ability to
become more engaged in self-monitoring than ever before (Ananthanarayan et al., 2014; Chung
et al., 2016). Ploderer et al. (2014) emphasize the role of social interaction and self-reflection in
how WAT systems may reinforce positive behavior change.
Many articles within this theme specifically examine the use of WAT as tools within an intervention
or feasibility study, and consequently measure PA as an outcome variable (often drawing on such
methods as randomized control trials (RCTs) (Cadmus-Bertram et al., 2015a; 2015b; Wang et al.,
2015, 2016). For example, Hooke et al. (2016) assessed the feasibility of using intervention
methods, namely daily emails with feedback (coaching email), and screenshots of participants’
previous day’s steps with a Fitbit device, as a means to increase daily steps. Cadmus-Bertram et
al. (2015a, 2015b) and Wang et al. (2015, 2016) investigated the use of WAT-centered self-
monitoring interventions to assess changes in the PA of obese or overweight adults, mostly
women, employing RCT methodologies. Short text messages were used as an intervention
method to monitor and improve PA. In these cases, results indicated that providing more
personalized, adaptive short messages could prove more effective than WAT devices combined
with simple one-size-fits-all reminders (Cadmus-Bertram et al., 2015a, 2015b; Wang et al., 2015,
2016).
Most studies used number of steps and active minutes generated by WAT devices as parameters
to examine changes in PA. For example, Hayes and Van Camp (2015) conducted a study of
Fitbit-based physical activity intervention for children. The number of steps per day taken during
the intervention period (mean=1,956 steps) was 47% higher than during baseline activity
(mean=1,326 steps).
Similarly, Cadmus-Bertram et al. (2015a) conducted a 16-week Fitbit based intervention study
with postmenopausal women. The intervention group (n=25) employed web-based, self
monitoring Fitbit devices and the control group (n=26) used standard pedometers. Results
indicate that the intervention group increased their moderate-to-vigorous physical activity minutes
per week by 62 minutes (baseline: 172 min/week, after intervention: 234 min/week.) In addition,
they recorded an increase of 786 average steps per day (baseline: 5906/day, after intervention:
6695/day). The control group with pedometers experienced non-significant increase in physical
activity. Based on this research, active behavior was increased in association with the use of
when people wore a WAT device such as Fitbit.
A few studies employed qualitative methods to measure behavior change from WAT use. For
instance, ethnographic study methods (one-on-one interviews at the beginning and the end of
WAT-based interventions) were used to investigate the potential role of WAT in changing the
behavior of obese and diabetic patients (Randriambelonoro et al., 2015). Most of these
participants reported changes in their lifestyle having used a Fitbit tracker for one month.
Whereas these findings suggest that WAT may have potential as an intervention tool to increase
activity levels through self-monitoring in the short term, there is still a need to evaluate the long-
term impacts of these devices (Fritz et al., 2014; Clawson et al., 2015; Jarrahi et al., 2017; Shin
et al., 2018).
A few long term studies (6 months based on TTM theory) indicate that WAT use helps manage
weight. Hartman et al. (2016) conducted a 6 month intervention study including the use of Fitbit
to monitor steps, smartphone app and coaching calls. Results indicate that the intervention group
had lost much more weight (4.4 kg vs 0.8 kg, p=0.004) than the control group. Similarly, Ashe et
al. (2015) performed a 6-month physical activity intervention measured with Fitbit. In this period,
intervention participants recorded significant increases in steps per day as well as weight loss
compared to the control group.
Research focused on child and adolescent user populations is another emerging field of interest
in WAT-enabled behavior change. Thematic textual analysis indicates that some frequently
occurring keywords in this body of research (see Table 1) were ‘children’, ‘students’, ‘youth’, and
‘adolescents’. This reflects growing recognition of the importance of PA on child and adolescent
health, particularly in the prevention of childhood metabolic and cardiovascular diseases. For
example, Hooke et al. (2016) reported the effectiveness of WAT device use as an intervention for
children with acute lymphoblastic leukemia, by delivering reinforcement, self-monitoring, goal
setting, gamification and feedback, as well as PA measurement (Hooke et al., 2016).
In terms of the direct impacts and interactions of WAT on children, we found two somewhat
contrasting research findings. On the one hand, a group of empirical studies document that WAT
use is likely to increase the activity levels of research participants compared to baseline or control
groups (Hayes and Van Camp, 2015; Chung et al., 2016; Hooke et al., 2016). This research also
suggests that children and adolescents engage positively with WAT. On the other hand, Schaefer
et al. (2016) reported that adolescents’ engagement with WAT declined over time, due to their
limited access to technology and weak motivation to participate (Schaefer et al., 2016). This work
further suggests that it is difficult to induce sustainable change in under-resourced urban youth
populations through using WAT alone, without adequate researcher training, programmatic
support, technology access, and some form of prior motivation to sustain ongoing use and
engagement.
Author (year)
Research focus
Sample
Size
Sample Background
Research Method
WAT Usage for data
WAT model
Schaefer et al.
(2014)
Promote physical
activity with urban
youth
25
Children
aged 7-10
Qualitative study
4 weeks; use of
different WAT each
week for 7
consecutive days
Actical,
SenseWear,
Polar Active,
etc.
Washington et al.
(2014)
Physical activity
intervention
11
Healthy adults
Intervention
3 weeks
Fitbit
Cadmus-Bertram
et al. (2015a)
Physical activity
intervention
25
Obese/overweight,
postmenopausal
women
Intervention
16 weeks
Fitbit
Cadmus-Bertram
et al. (2015b)
Physical activity
intervention
51
Obese/overweight
postmenopausal
women
Randomized Trial
4 weeks
Fitbit
Hayes and Van
Camp (2015)
Promote physical
activity during school
recess
6
Elementary students
Intervention
20 mins, 1 to 4 days
per week in which 22
sessions occurred
Fitbit
Randriambelonor
o et al. (2015)
Promote physical
activity
18
Diabetic and obese
patients
One-month in-situ
study
Interviews
1 month
Fitbit
Wang et al.
(2015)
Physical activity
intervention
67
Overweight/obese
adults
Randomized Trial
6 weeks
Fitbit One
Wang et al.
(2016)
Promote physical
activity
67
Overweight/obese
adults
Calculated mean
score of the Likert-
type responses that
assessed usability
and level of
engagement
6 weeks
Fitbit One
Chung et al.
(2016)
Promote physical
activity in young
adults
12
Adolescents
aged 19-20
Randomized
Controlled Trial
2 months
Fitbit
Hooke et al.
(2016)
Promote physical
activity
17
Children with cancer
aged 6-15
Intervention
3 days for baseline
2 weeks for
intervention
Fitbit
Table 5. Examples of WAT studies focused on Behavior Change
Theme 4: Acceptance, Adoption, and Abandonment (78 articles)
As WAT use has become ubiquitous, an important body of research has developed that considers
the acceptance and adoption implications of these devices. Some prominent keywords in this
topic (see Table 1) are ‘acceptance’, ‘usability’, ‘interviews’, and ‘design’. Yet in the years since
WAT introduction in 2012, designers have continued to encounter difficulties, as evidenced by
reports that many devices have failed to achieve sustained user engagement (Dibia, 2015; Meyer
et al., 2016). This design challenge is not unique to WAT, and many technologies aimed at helping
users change their everyday behavior suffer from the same issue of user sustainability (Consolvo
et al., 2009; Zhao et al., 2017; Gulotta et al., 2016). Acceptance and adoption (abandonment)
research, building on a context-aware approach clearly indicates that a WAT device in itself, does
not determine the desired ends (e.g., increased PA (Jarrahi et al., 2017)), and that WAT influence
is mediated by many social, cognitive and psychological factors (Mackintosh, 2016; Arigo, 2015;
Karapanos et al., 2016; Butryn et al. 2016).
The acceptance and adoption (abandonment) research reviewed here provides design
implications for various use contexts. Several field studies that explored how WAT are used and
experienced in the daily life of novice users, identified critical acceptance and adoption
(abandonment)-related issues that influence long-term daily use. Features included wearability,
device appearance, display and interaction, as well as the modeling and technical aspects of data
measurement and presentation (Meyer et al., 2015; Kim et al., 2016). Kim et al. also emphasized
that users expect a more sophisticated, personalized interaction system and data analysis over
time. A tangible outcome of this stream of research is a set of guidelines for WAT design, such
as device appearance, intensity of interaction, wearability and technical aspects of measurement
and data validity. More specifically, these studies are centered around two closely related
subtopics: WAT acceptance and WAT abandonment.
WAT acceptance
A clear focus of acceptance and adoption (abandonment) research has been on understanding
users' rationale behind retaining or abandoning their devices. This body of literature addresses
why individuals engage with or accept WAT and suggests directions for future research and
design (Mercer et al., 2016; Darvallet al., 2016; Fritz et al., 2014). In particular, research on this
subtopic has explored the use, adoption and acceptance of WAT by different user populations,
and the ways in which various contextual factors may shape their interactions with these devices
(see Table 6).
Some of the most studied users in this subtopic, as indicated in Table 1, are older adults.
Research has centered on this population’s acceptance of WAT, and evaluated their use of and
attitudes toward WAT. In turn, these observations can reveal potential usability barriers to
acceptance (Preusse et al., 2016; Fausset et al., 2013; McMahon et al., 2016). Findings suggest
that older participants may not always accept and use WAT, even if they have initially accepted
the technology (Fausset et al., 2013; McMahon et al., 2016)). This observation is consistent with
the description by Blanson et al. (2011) of four groups of technology users: “first glimpsers,” “early
dropouts,” “late dropouts,” and “maintainers” (Blanson et al., 2011). Preusse et al. (2016)
suggested that older adults’ WAT interaction may be improved by addressing ‘acceptance-
barriers’ during deployment, such as by providing tutorials on challenging features and
communicating the device’s usefulness (Preusse et al., 2016).
Several studies have investigated the adoption of WAT acceptance by users already highly
motivated by their own acute health needs. For example, the usefulness of WAT with participants
over age 50, who had been diagnosed with a chronic illness (e.g., vascular disease, diabetes,
arthritis, or osteoporosis) was examined by Mercer et al. (2016). They concluded that while this
group of participants may recognize the use and ease of use of WAT, elder participants may
benefit from cheaper and more compatible devices, and more comprehensive set-up assistance.
Naslund et al. (2015) used qualitative follow-up interviews to evaluate the acceptability of WAT to
overweight and obese individuals with serious mental illness (SMI); they demonstrated that WAT
were useful for motivating patient PA through personal goal setting, self-monitoring and the
devices’ social connectivity enhancements (Naslund et al., 2015).
Other studies have focused on healthy users who have already integrated WAT monitoring into
their lives at the time of study. For instance, the participants in the study by Fritz et al., (2014) had
already adopted their own WAT for between 3 and 54 months (rather than using research project-
allocated devices). Investigations into the long-term influence of WAT on wearers’ activities and
attitudes, showed that despite changing goals and practices over time, many wearers continued
to gain value and motivation from the technology. Similarly, Jarrahi et al. (2017) conducted a
qualitative study to explore how various types of existing motivations shaped the perception and
adoption of Fitbit trackers. They suggested that unique life priorities, personal circumstances, and
personalities could interact with the affordances of WAT, resulting in different outcomes, ranging
from abandonment to strong acceptance.
Overall, one of the key conclusions of acceptance and adoption (abandonment) research in the
current context is that the successful design of an all-purpose, universal WAT is unreasonable.
Rather, a variety of design concepts and data models should continue to emerge that align with
the personal preferences of various groups of users. Features such as self-monitoring, goal-
based gamification, continuous feedback, and social support seem to encourage increased
activity levels and healthy behaviors for many users. But different user groups present different
sets of needs, different design personas and consequently different WAT device interactions. As
Clawson et al argue (2015, p.655): “Health-tracking is not limited to a specific subset of users and
our findings highlight the need to design solutions that embrace a diversity of users and have the
ability to support these users over time.”
What is considered usable for one group may not cater to the needs of another. Design criteria
therefore vary across different user populations. For example, the reviewed literature suggests
that WAT targeting patients with chronic illness should focus on mHealth, specifically integrating
programs that designed for specific patients populations and recommend a customized regimen
and specific levels of physical activities (Mercer et al. 2016). In contrast, WAT design targeting a
healthy office workers with sedentary behavior may follow and reinforce a rather different adoption
pathway, for example, by repeatedly reminding the user to engage in more physical activities
(Sloan et al. 2018).
Author (year)
Research focus
Sampl
e Size
Sample Background
Data collection method
WAT usage for data
collection
WAT model
Fausset et
al. (2013)
Older adults
acceptance of WAT
8
Older adults
Experiment + Interview
2 weeks
Striiv
Fitbit
Nike+ Fuelband
MyfitnessPal
Fritz et al.
(2014)
Long term use of
activity tracker
30
Users who had been
using such a WAT
for at least three
months
Interviews
Between 3 and 54
months
Nike+ Fuelband
Fitbit Ultra
Darvall et al.
(2016)
Acceptability
10
Aged over 18 years
Descriptive 22-item
questionnaire
Ten-week pilot trial
Fitbit Zip
McMahon et
al. (2016)
Older adults
acceptance of WAT
95
Older adults
Randomized controlled
trial :- survey 10 weeks
and 8 months after
start date
Fitbit One device for use
throughout 8-month
study
Fitbit One
Mercer et al.
(2016)
WAT acceptance of
older adults with
chronic illness
32
Aged over 50
previously diagnosed
with a chronic illness
Questionnaire
Each device worn for at
least 3 days
Fitbit Zip
Misfit Shine
Jawbone Up
Withings
Preusse et
al.
(2016)
Older adults’
acceptance
16
Older adults
Questionnaires +
interviews
Heuristic evaluation
28-day field study
Fitbit
Jarrahi et al.
(2017)
Role of motivation in
WAT use
29
Worked in the same
academic institution
Interviews
Between 2 weeks and
35 months
Fitbit
Naslund et
al.
(2015)
WAT acceptance of
participants with
serious mental
illness
10
Participants with
serious mental
illness
Exploratory study +
interviews
80 and 133 days
Fitbit
Nike
Apple smartphone
app
Table 6. Examples of Acceptance and Adoption (Abandonment) studies that focus on Acceptance
WAT abandonment
In spite of the growing perception of the value of WAT, recent studies report that some users
abandon their WAT after short term use, which is unlikely to reflect the devices’ potential to help
achieve a sustainably active lifestyle (Ledger and McCaffrey, 2014; Shih et al., 2015; H. Lee and
Y. Lee, 2017) (see Table 7). The popularity of purchasing and wearing WAT obscures the fact
that full understanding of user acceptance, adoption, and sustained engagement remains elusive
(Motti and Caine, 2014). Increasing concern that WAT frequently fail to inspire long-term use and
acceptance (Fritz et al., 2014), has led researchers to investigate why some users are unwilling
or unable to integrate WAT into their daily activities.
Few studies of device adoption or abandonment embraced multiple brands of WAT in their
sampling approach (as indicated in Table 6), and we did not find one that specifically examined
and compared abandonment across various devices. However Clawson et al. (2015) explored
why users abandoned their WAT devices, by analyzing advertisements for secondary sales of
personal health-tracking technologies on Craigslist (https://craigslist.org/), and identified a
complex and dynamic range of rationales behind users’ desire to sell their WAT. These included
lifestyle changes, disappointment (mismatched hopes with device capabilities) and decisions to
upgrade. The authors suggested that device design could better accommodate the unpredictable
trajectories of everyday life that users experience (Clawson et al., 2015). Harrison et al. (2015)
also adopted an indirect approach to the issue by exploring users’ ‘barriers to engagement’ with
WAT. From surveys and contextual interviews, they highlighted two important design factors that
can facilitate long-term use, namely customizing the user experience of both the device’s “tracking
and social functionalities”, and of its aesthetics and physical design (Harrison et al., 2015, p4).
Lazar et al. (2015) and Epstein et al., (2016a, 2016b) identified further reasons why some users
may abandon WAT. These included the ineffectiveness of collected data in motivating use, and
devices’ high maintenance requirements. Surveys by Epstein et al. (2016a) identified that users
may have mixed feelings after having disengaged with their device, ranging from relief and
freedom to feelings of frustration and guilt about having abandoned it. The authors suggest that
lapsed users may show more interest in alternative forms of data visualization and textual cues
and that design could take their past history into account (Epstein et al., 2016b). Other design
recommendations included the personalization of features and user responsiveness, and de-
emphasis of the importance of long-term user commitment (Lazar et al., 2015).
Studies focusing on the use of WAT for medical interventions reported much lower rates of device
abandonment (e.g., Naslund et al. 2015; Mercer et al. 2016) as opposed to those examining
adoption of WAT outside of medical contexts. One possible explanation lies in the typically short-
term timeframes (often a few days) used in medical intervention studies, as opposed to studies
of WAT uses in users’ daily lives of users and over long terms (more than a few months) (e.g.,
Jarrahi et al. 2017; Fritz et al. 2014). User engagement and disengagement with these devices
commonly take place over an extended period of time, and may not be effectively observed in
short term studies.
Author
(year)
Research Focus
Sample
Size
Sample Background
Data collection method
WAT usage for data
collection
WAT Model
Clawson
et al.
(2015)
Health motivations and
reasoning for
abandonment
462
Advertisers selling
devices on Craigslist
(US) during 1 month
Inductive and deductive
method
n/a
Fitbit
Harrison et
al.
(2015)
Abandonment and
barriers to engagement
24
Current and previous
tracker users
Survey and contextual
interviews
Various, between
2 weeks - 3 years
Fitbit
Jawbone Up
Misfit
Lazar et
al. (2015)
Abandonment
17
Employees in technology
company
Survey + interview
Approximately 2 months
Lumoback
Fitbit
Nike+Fuelband
Misfit Shine
Samsung Gear
Etc.
Epstein et
al. (2016a)
Abandonment
141
Lapsed Fitbit users on
AmazonTurk + snowball
sampling
Survey
Ranging from less than a
week to over two years
Fitbit
Epstein et
al. (2016b)
Abandonment
141
Lapsed Fitbit users on
AmazonTurk + snowball
sampling
Survey
Participants were
required to wear Fitbit at
least three days
Fitbit
Table 7. Examples of Acceptance and Adoption (Abandonment) studies that focus on WAT
Abandonment
Overall, one of the key conclusions of WAT acceptance and adoption (abandonment) research in
the current context is that the successful design of an all-purpose, universal device is
unreasonable. Rather, a variety of design concepts and data models should continue to emerge
that align with the personal preferences of various groups of users. Features such as self-
monitoring, goal-based gamification, continuous feedback, and social support seem to encourage
increased activity levels and healthy behaviors for many users. But different user groups present
different sets of needs, different design personas and consequently different WAT device
interactions. As Clawson et al argue (2015, p.655): “Health-tracking is not limited to a specific
subset of users and our findings highlight the need to design solutions that embrace a diversity of
users and have the ability to support these users over time.”
What is considered usable for one group may not cater to the needs of another . Design criteria
therefore vary across different user populations. For example, the reviewed literature suggests
that WAT targeting patients with chronic illness should focus on mHealth, specifically integrating
programs that designed for specific patients populations and recommend a customized regimen
and specific levels of physical activities (Mercer et al. 2016). In contrast, WAT design targeting a
healthy office workers with sedentary behavior may follow and reinforce a rather different adoption
pathway, for example, by repeatedly reminding the user to engage in more physical activities
(Sloan et al. 2018).
Theme 5: Self-monitoring data centered (48 articles)
Research studies in this theme are more focused on WAT affordances for collection and analysis
of personal data, rather than on their technological features. As shown in Table 1, prominent
keywords in this topic were ‘data’, ‘personal’, ‘information’, and ‘collection’. For example, van
Berkel et al. (2015) attempted to understand how people used WAT-generated data, and how
they interpreted their records on activity steps, activity level, sleep, heart rate, etc. (van Berkel et
al., 2015).
Personal informatics research and studies of the characteristics and behaviors of “quantified
selfers” are two important sub-themes (see Table 8). Personal Informatics (PI) systems are
defined as “those that help people collect personally relevant information for the purpose of self-
reflection and gaining self-knowledge” (Li et al. 2013). In general, PI, as an emerging research
space, concerns activities that contribute to users’ understanding and knowledge of their own
behavior, through more effective collection of personal data and reflection of “self” (Rapp and
Tirassa, 2017). Epstein et al. (2015) adopted the concept of PI and more specifically 'lived
informatics' to examine characteristics across the spectrum of self-trackers' habits, the decision
to track, their tool choice, information collection, and lifestyle integration.
People have used almost any tool in the past to collect personal information. These digital and
non-digital tools have varied in physical form (paper, websites, devices, etc.) and level of support
(manual vs. automated). Since their introduction in 2012, modern WAT have been argued to open
up wholly new opportunities to assist users in recording and monitoring their broader PI, namely
the ability to track, manage, depict, reflect and use their data (Rapp and Cena, 2016). Collections
of personal informatics, enabled by the use of WAT and other personal ubiquitous devices, have
been promoted as helping users improve self-knowledge, by providing a personal history and
means to review or analyze personal data (Rapp et al., 2018; Elsden et al., 2016).
Research on PI and quantified selfers is closely related (Rapp et al., 2018). Quantified selfers (Q-
selfers), a somewhat exceptional (and widely studied) user population in PI studies, are
individuals who proactively collect and act upon their own personal data, in order to optimize self-
approbation of their lives (Shin et al., 2015; Choe et al., 2014; Elsden and Kirk, 2014). An
important challenge for Q-selfers is to maintain consistent data collection over a long period of
time (van Berkel et al., 2015). Furthemore, tracking data seems to be used in the short term, often
on the day that the data was generated (Rooksby et al., 2014).
Q-selfers’ primary motivation to collect personal data is in order to answer questions about
themselves. Once they have an answer to their query, they move on to another or stop self-
tracking behavior. Future WAT tool design could improve response to users such as quantity
selfers’ real-time, dynamic information needs, the better to engage their interest in more sustained
data collection (van Berkel et al., 2015).
Author
(year)
Research Focus
Sample
Size
Sample Background
Data collection method
WAT usage for data
collection
WAT Model
Choe et
al. (2014)
Quantified
Selfers
52
Quantified Selfers
Transcribed this entire 52
videos
Video uploaded since
January 2012
Fitbit, ZEO,
WIFI-scale,
etc.
Shin et
al.
(2015)
Quantified
Selfers
15
participants
30 posts
Fitbit users and
Quantified Selfers
Qualitative interview
Explored the popular posts
on the Quantified Self forum
30 most popular posts
in 2015
Fitbit
Epstein
et al.
(2015)
Personal
Informatics
281 for
survey
22 for
interviews
281 past and
present trackers
22 trackers
Survey+Interviews
Not specified
Fitbit
Nike
MyFitnessPal,
etc.
Ayobi et
al. (2016)
Personal
Informatics
20
Not specified
Grounded Theory Literature
Review
Keyword searched
from 2010 to 2015
Not specified
Table 8. Examples of studies that feature a Data Centric perspective of WAT
Theme 6: Privacy (29 articles)
The rising popularity of WAT raises a number of concerns related to user privacy. In particular,
there is increasing concern about how providers of these devices use the data they record, and
the personal privacy protections that are afforded to users (Paul and Irvine, 2014; Li, 2016). As
shown in Table 1, prominent keywords in this topic are ‘privacy’, ‘information’, and ‘security’.
Paul and Irvine (2014) examined privacy policies and the type of data collected by four WAT
services (Fitbit, Jawbone Up, Basis, and Nike+) to identify privacy risks for users. WAT vendors
gain the right to commercially use or share users’ data with partners. Furthermore, these devices
are constantly connected to the internet, and do not allow offline use. Users may not use the
device without agreeing to such service privacy policies. Paul and Irvine (2014) pointed to the
rights to data ownership as being a privacy issue.
In a similar vein, Cyr et al. (2014) studied the security and privacy properties of the Fitbit device
by analyzing the security properties of the network traffic between the device, smartphone or
computer app, and the Fitbit web service. This work also highlighted privacy concerns: they
identified that Fitbit devices collected unrelated information about users while failing to provide all
of the data collected (such as per-minute activity data) to the device owner (Cyr et al., 2014).
Personality, personal trust and the device’s usability could affect users’ privacy perception of WAT
(Lamb et al. 2016). For example, more neurotic users tend to be more conscious of data privacy,
whereas users with more personal trust were less concerned with the privacy implications of WAT.
Based on a greater understanding of user personalities, Lamb et al. (2016) recommended that
designers and developers customize the default settings of WAT for different user groups with
varying privacy-related interests.
The primary question that these studies are concerned with is whether the privacy, security and
discoverability risk of WAT is greater than the benefits that these wearable technologies can
provide (see Table 9). The common conclusion of this body of research underscores the need to
balance privacy and security issues with the potential of improvements that may arise in relation
to consumers, the health care system, and the health and wellness of society (Kumari and Hook,
2017).
Even though studies addressed users’ concerns about the privacy of data collected by WAT, this
theme has received the least attention (see Figures 2 ), and the number of studies has decreased
over time between 2013 (11.82%) and 2017 (6.35%) (see Figure 3). Only a few empirical studies
have examined user privacy in relation to WAT adoption (Lamb et al., 2016).
Author (year)
Research Focus
Sample
Size
Sample Background
Data collection method
WAT usage for
data collection
WAT
Model/Data
from WAT
Paul and
Irvine
(2014)
Compare privacy
policies
4 services
Privacy policies in place for
Fitbit, Jawbone Up,
Nike+,BASIS
Compare privacy
policies for Fitbit,
Jawbone Up,
Nike+,BASIS
N/A
Fitbit
Jawbone Up
Nike+
BASIS
Lamb et al.
(2016)
Personality
differences on privacy
perception
39
Young adults
Pre-interview
A week-long wearable
device usability test
Post-interview
For a week
Fitbit Flex,
Jawbone Up,
Sony SWR10
Pebble,
Samsung Gear
Kumari and
Hook (2017)
Review federal and
state legislation about
protection for data
N/A
N/A
Review
N/A
Fitbit
Jawbone Up
Table 9. Examples of studies focused on WAT Privacy implications
4. Discussion
Figure 4 illustrates the themes involved in WAT research based on three key elements that can
define studies of information technology (Sawyer and Huang, 2007). The relationship between
Technology’ and ‘Information’ components (axis 1, Figure 4), primarily reflects ‘technology’
focused studies, which have been relatively well investigated in the current literature (see Figure
3). The WAT research situated on this axis is not particularly concerned with human interaction
and instead focuses more on precision, reliability, accuracy, and the ways that WAT technological
features and capabilities can be improved to better capture and represent data. The relationship
between ‘Technology’ and the 'User' (axis 2, Figure 4), reflects the plethora of other WAT research
fields which we have classified 'Acceptance and Adoption (Abandonment', 'Behavior change',
Medical Contexts' and 'Privacy' (see Figure 3). The research situated on this axis concerns the
interaction of technology, its design and capacities, with the user's needs, perceptions and
expectations. We propose that axis 3, the more existential interrelationship between people and
their information needs, represents an integral research gap in the understanding, research and
design of WAT. A fuller explication and discussion follow.
Figure 4. WAT research themes based on the three elements of Information, Technology, and People
As opposed to acceptance and adoption (abandonment) research in this field, which has begun
to extend WAT user interaction studies outside of lab settings, research with a technology focus
has remained predominantly lab-based and short term (see Evenson et al., 2015). This growing
group of experimental studies therefore lacks longitudinal perspective, and may fail to embrace
the complicated ways in which devices’ technological functionality effectively records and reflects
users’ nuanced activity in natural settings, and over time. Users may embark on a broader range
of fitness activities in their daily lives (e.g., energy expenditure) than can be adequately captured
and assessed through lab-based experiments alone. Further, not only might user activity be more
complex in “the wild,” but the same WAT can undergo changes in technological capabilities over
time. For example, many WAT provide automatic and/or manual updates to their firmware, which
can change the measurement properties of the instrument over time, and consequently change
the reliability and validity of its data collection and analysis mechanisms. Future technology-
focused research would clearly benefit from the use of longitudinal studies of WAT in natural
settings, with greater recognition of how their technological features may evolve over time.
Another limitation of studies focusing on WAT accuracy is their lack of attention towards users’
perception of accuracy vis-à-vis the actual accuracy of these devices’ capacity for data collection
and analysis. More specifically, as these works suggest, WAT devices may not satisfactorily
reflect certain user activities (for more information see (Evenson et al. 2015)).
The current collection of research with a technological focus is concerned with objective measures
of assessing the validity and accuracy of WAT devices. Therefore, a fruitful direction for future
research could lie at the intersection of technological focus and adoption studies. Important
questions to be addressed concern 1) how users may go about practically figuring out the
accuracy and validity of data collection without access to scientific tools and methods; 2) the
extent to which certain users care about accuracy, and how trust is established when they become
aware that WAT devices may function inaccurately; and 3) how their perception of accuracy may
shape users’ decision to abandon or continue using WAT devices.
Many acceptance and adoption (abandonment) aspects of WAT adoption (axis 2, Figure 4) have
also been covered relatively well in the current literature. Existing research has examined topics
such as the interaction of various user populations with WAT (e.g., older adults or children), device
usability, and why users may adopt or abandon these devices. Because of the health-related
implications of these devices, existing research has also examined the use and effect of WAT for
health, medical and clinical purposes. This body of work has focused on the ways these devices
may impact physical activity (PA) or other health-related behaviors of average users or patients
in or outside of medical settings. Research has also directed attention to the privacy dimension
of WAT adoption, how it may interact with the security of these devices, and how users’ privacy
is handled (or mishandled) by these technologies.
Finally, this review has identified that what is often missing from this body of work, is a focus on
the interaction between the user and multiple WAT devices. The vast majority of the WAT studies
under review consider these devices in isolation from other devices and technological options
(e.g., social, navigational, informational etc.). In practice, given the plethora of available consumer
technologies, many users may also engage with multiple devices and technologies, to monitor
their fitness/health activities. They may therefore use WAT in tandem with smart watches or
mobile applications, to help track similar or different activities. In order to address this gap in
research, future work could build from an ecological perspective (Gibson, 1977; Nardi and O’Day,
1999). This would help focus our understanding of how the use of each device may fit into the
user’s broader lifestyle, how or why users may pick and choose between them, and how devices
may interact with one another within the user’s behavioral ecology (Bødker and Klokmose, 2012;
Coughlan et al., 2012; Jarrahi et al., 2017).
The development of WAT is creating new ways of understanding human-information interaction
(axis 3, Figure 4). Thematic analysis in this review makes it clear that this dimension has not been
addressed as adequately as have the other two dimensions of technology focus and acceptance
and adoption (abandonment) (axes 1 and 2, Figure 4). Adopting a personal informatics approach,
current data-centered research (with the exception of Rooksby et al. (2014) and Rapp and Cena
(2016)) is focused more on human-data interaction (Mortier et al., 2014), and therefore does not
typically go beyond a narrow preoccupation with data collection or self-tracking through WAT. Nor
does it address people’s use of these data for lifelogging or optimizing certain behaviors such as
physical activity or dietary habits (Hoy, 2016). As such, this research is heavily weighted toward
the study of Q-selfers, who are an eccentric subpopulation of WAT users (Choe et al., 2014). Liu
et al. (2014) correctly argues that the relationship between information seeking contexts and use
of self-tracking technology is still unclear.
To address this gap in the understanding of the rich human-information interaction enabled by
WAT adoption, future research could draw on the theoretical foundations of information behavior
(Wilson, 1999) or information practice (Savolainen, 2007). These frameworks provide insight into
the unique ways people encounter, seek, use, and make sense of information with an emphasis
on the information context within which they seek information. An Information Behavior approach
helps identify the cognitive, social and ecological dimensions of human-information interaction:
the impact of individual characteristics, the meanings and values that users assign to social
aspects of WAT-enabled information seeking, and finally, the environmental factors that shape
the use of information artifacts (Fidel, 2012; Pettigrew et al., 2001). A practice-centric approach
specifically helps explicate the dynamic relationships between the user, WAT, the information
artifacts embodied in WAT, and the forces of their social context (Feldman and Orlikowski, 2011).
From this perspective, information activities intertwined with the adoption of WAT are intrinsically
social and can only be understood in relation to the broader context of everyday life information
seeking (Savolainen, 1995), and the information activities and inspiration of other actors (Cox,
2012).
Limitations
Mobile tracking applications were excluded from this review since the focus of this work was on
commercially available wearable activity trackers (WAT) as a recent technological phenomenon.
WAT devices share some similar features with mobile applications but also have distinct
consequences by offering a physical manifestation (i.e., wearability). For more information on the
differing consequences of digital and physical dimensions of WAT see Jarrahi, 2015. This noted,
we recognize that a major limitation of this review is having included only commercially available
activity tracking devices and not all types of digital technologies, applications and platforms that
may provide similar functionalities.
5. Conclusion
This review summarizes evidence from 463 studies of wearable activity tracking (WAT) devices
published between 2013 and 2017. Using topic modeling techniques, we identified six common
themes that capture the existing state of research on WAT: 1) Technological Focus; 2) Patient
Treatment and Medical Settings; 3) Behavior Change; 4) Acceptance and Adoption
(Abandonment); 5) Self-monitoring Data Centered, and 6) Privacy. We describe these topics, the
subjects being studied, and the research methods used. We also shed light on research trends
between 2013 and 2017, and provide direction for future research. An important research gap
identified in the reviewed literature lies in the understanding of the rich human-information
interaction that is enabled by WAT adoption.
The primary contributions of this article are firstly, our accumulation of the highly dispersed
research on WAT, as prominent instances of ubiquitous computing; and secondly, our description
and analysis of a diverse set of disciplinary interests in WAT; that has implications for personal
health and fitness data management.
In conclusion, this article suggests that WAT devices are multi-dimensional technologies with
complex impacts. Understanding WAT and their technological and non-technological aspects
requires various research perspectives. This work raises interdisciplinary awareness about the
current landscape of WAT use and the related diversity of interesting research opportunities and
challenges.
6. References
Adam Noah, J., David K. Spierer, Jialu Gu, and Shaw Bronner. 2013. “Comparison of Steps and
Energy Expenditure Assessment in Adults of Fitbit Tracker and Ultra to the Actical and
Indirect Calorimetry.” Journal of Medical Engineering & Technology 37 (7): 45662.
Aggarwal, Charu C., and Chengxiang Zhai. 2012. Mining Text Data. Springer Science &
Business Media.
Alinia, Parastoo, Chris Cain, Ramin Fallahzadeh, Armin Shahrokni, Diane Cook, and Hassan
Ghasemzadeh. 2017. “How Accurate Is Your Activity Tracker? A Comparative Study of
Step Counts in Low-Intensity Physical Activities.” JMIR mHealth and uHealth 5 (8): e106.
Ananthanarayan, Swamy, Nathan Lapinski, Katie Siek, and Michael Eisenberg. 2014. “Towards
the Crafting of Personal Health Technologies.” In Proceedings of the 2014 Conference on
Designing Interactive Systems, 587–96. DIS ’14. New York, NY, USA: ACM.
Appelboom, Geoff, Blake E. Taylor, Eliza Bruce, Clare C. Bassile, Corinna Malakidis, Annie
Yang, Brett Youngerman, et al. 2015. “Mobile Phone-Connected Wearable Motion Sensors
to Assess Postoperative Mobilization.” JMIR mHealth and uHealth 3 (3): e78.
Arigo, Danielle. 2015. “Promoting Physical Activity among Women Using Wearable Technology
and Online Social Connectivity: A Feasibility Study.” Health Psychology and Behavioral
Medicine 3 (1): 391409.
Ashe, Maureen C., Meghan Winters, Christiane A. Hoppmann, Martin G. Dawes, Paul A.
Gardiner, Lora M. Giangregorio, Kenneth M. Madden, et al. 2015. “‘Not Just Another
Walking Program’: Everyday Activity Supports You (EASY) Modela Randomized Pilot
Study for a Parallel Randomized Controlled Trial.” Pilot and Feasibility Studies 1 (1): 4.
Beevi, Femina H. A., Jorge Miranda, Christian F. Pedersen, and Stefan Wagner. 2016. “An
Evaluation of Commercial Pedometers for Monitoring Slow Walking Speed Populations.”
Telemedicine Journal and e-Health: The Official Journal of the American Telemedicine
Association 22 (5): 44149.
Bellicha, Alice, Sandrine Macé, and Jean-Michel Oppert. 2017. “Prescribing of Electronic
Activity Monitors in Cardiometabolic Diseases: Qualitative Interview-Based Study.” Journal
of Medical Internet Research 19 (9): e328.
Berkel, Niels van, Chu Luo, Denzil Ferreira, Jorge Goncalves, and Vassilis Kostakos. 2015.
“The Curse of Quantified-Self: An Endless Quest for Answers.” In Adjunct Proceedings of
the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing and
Proceedings of the 2015 ACM International Symposium on Wearable Computers, 97378.
UbiComp/ISWC’15 Adjunct. New York, NY, USA: ACM.
Blanson Henkemans, O. A., W. A. Rogers, and A.M.C. Dumay. 2011. “Personal Characteristics
and the Law of Attrition in Randomized Controlled Trials of eHealth Services for Self-Care.”
Accessed March 15, 2019.
https://www.narcis.nl/publication/RecordID/oai:tudelft.nl:uuid%3A2f26db58-8ffc-46f9-9a26-
804f87d800ae.
Blei, David M., Andrew Y. Ng, and Michael I. Jordan. 2003. “Latent Dirichlet Allocation.” Journal
of Machine Learning Research: JMLR 3 (Jan): 9931022.
Bødker, Susanne, and Clemens Nylandsted Klokmose. 2012. “Dynamics in Artifact Ecologies.”
In Proceedings of the 7th Nordic Conference on Human-Computer Interaction: Making
Sense Through Design, 448–57. NordiCHI ’12. New York, NY, USA: ACM.
Butryn, Meghan L., Danielle Arigo, Greer A. Raggio, Marie Colasanti, and Evan M. Forman.
2016. “Enhancing Physical Activity Promotion in Midlife Women with Technology-Based
Self-Monitoring and Social Connectivity: A Pilot Study.” Journal of Health Psychology 21
(8): 154855.
Cadmus-Bertram, Lisa A., Bess H. Marcus, Ruth E. Patterson, Barbara A. Parker, and Brittany
L. Morey. 2015a. “Randomized Trial of a Fitbit-Based Physical Activity Intervention for
Women.” American Journal of Preventive Medicine 49 (3): 41418.
Cadmus-Bertram, Lisa, Bess H. Marcus, Ruth E. Patterson, Barbara A. Parker, and Brittany L.
Morey. 2015b. “Use of the Fitbit to Measure Adherence to a Physical Activity Intervention
Among Overweight or Obese, Postmenopausal Women: Self-Monitoring Trajectory During
16 Weeks.” JMIR mHealth and uHealth 3 (4): e96.
Case, Meredith A., Holland A. Burwick, Kevin G. Volpp, and Mitesh S. Patel. 2015. “Accuracy of
Smartphone Applications and Wearable Devices for Tracking Physical Activity Data.”
JAMA: The Journal of the American Medical Association 313 (6): 62526.
Chang, Jonathan, Sean Gerrish, Chong Wang, Jordan L. Boyd-graber, and David M. Blei. 2009.
“Reading Tea Leaves: How Humans Interpret Topic Models.” In Advances in Neural
Information Processing Systems 22, edited by Y. Bengio, D. Schuurmans, J. D. Lafferty, C.
K. I. Williams, and A. Culotta, 28896. Curran Associates, Inc.
Choe, Eun Kyoung, Nicole B. Lee, Bongshin Lee, Wanda Pratt, and Julie A. Kientz. 2014.
“Understanding Quantified-Selfers’ Practices in Collecting and Exploring Personal Data.” In
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 1143
52. CHI ’14. New York, NY, USA: ACM.
Chu, Anne H. Y., Sheryl H. X. Ng, Mahsa Paknezhad, Alvaro Gauterin, David Koh, Michael S.
Brown, and Falk Müller-Riemenschneider. 2017. “Comparison of Wrist-Worn Fitbit Flex and
Waist-Worn ActiGraph for Measuring Steps in Free-Living Adults.” PloS One 12 (2):
e0172535.
Chum, Jenny, Min Suk Kim, Laura Zielinski, Meha Bhatt, Douglas Chung, Sharon Yeung,
Kathryn Litke, et al. 2017. “Acceptability of the Fitbit in Behavioural Activation Therapy for
Depression: A Qualitative Study.” Evidence-Based Mental Health 20 (4): 12833.
Chung, Arlene E., Asheley C. Skinner, Stephanie E. Hasty, and Eliana M. Perrin. 2016.
“Tweeting to Health: A Novel mHealth Intervention Using Fitbits and Twitter to Foster
Healthy Lifestyles.” Clinical Pediatrics, June. https://doi.org/10.1177/0009922816653385.
Clawson, James, Jessica A. Pater, Andrew D. Miller, Elizabeth D. Mynatt, and Lena Mamykina.
2015. “No Longer Wearing: Investigating the Abandonment of Personal Health-Tracking
Technologies on Craigslist.” In Proceedings of the 2015 ACM International Joint
Conference on Pervasive and Ubiquitous Computing, 647–58. UbiComp ’15. New York,
NY, USA: ACM.
Codella, J., C. Partovian, H. Y. Chang, and C. H. Chen. 2018. “Data Quality Challenges for
Person-Generated Health and Wellness Data.” IBM Journal of Research and Development
62 (1): 3:13:8.
Consolvo, Sunny, David W. McDonald, and James A. Landay. 2009. “Theory-Driven Design
Strategies for Technologies That Support Behavior Change in Everyday Life.” In
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 405
14. CHI ’09. New York, NY, USA: ACM.
Cook, David J., Jeffrey E. Thompson, Sharon K. Prinsen, Joseph A. Dearani, and Claude
Deschamps. 2013. “Functional Recovery in the Elderly after Major Surgery: Assessment of
Mobility Recovery Using Wireless Technology.” The Annals of Thoracic Surgery 96 (3):
105761.
Cooper, Harris M. 1988. “Organizing Knowledge Syntheses: A Taxonomy of Literature
Reviews.” Knowledge in Society 1 (1): 104.
Coughlan, Tim, Trevor D. Collins, Anne Adams, Yvonne Rogers, Pablo A. Haya, and Estefanía
Martín. 2012. “The Conceptual Framing, Design and Evaluation of Device Ecologies for
Collaborative Activities.” International Journal of Human-Computer Studies 70 (10): 765
79.
Coughlin, Steven S., and Jessica Stewart. 2016. “Use of Consumer Wearable Devices to
Promote Physical Activity: A Review of Health Intervention Studies.” Journal of
Environment and Health Sciences 2 (6). https://doi.org/10.15436/2378-6841.16.1123.
Cox, Andrew M. 2012. “An Exploration of the Practice Approach and Its Place in Information
Science.” Journal of Information Science and Engineering 38 (2): 17688.
Cyr, Britt, Webb Horn, Daniela Miao, and Michael Specter. 2014. “Security Analysis of Wearable
Fitness Devices (fitbit).” Massachusetts Institute of Technology, 1.
Darvall, J. N., A. Parker, and D. A. Story. 2016. “Feasibility and Acceptability of Remotely
Monitored Pedometer-Guided Physical Activity.” Anaesthesia and Intensive Care 44 (4):
5016.
De Sousa Sena, Riany, Jean Bourbeau, Pei Zhi Li, and Sara Ahmed. 2015. “Validity and
Usability Testing of the Fitbit Pedometer in Patients with COPD.” The European
Respiratory Journal: Official Journal of the European Society for Clinical Respiratory
Physiology 46 (suppl 59): PA2067.
Dibia, Victor. 2015. “An Affective, Normative and Functional Approach to Designing User
Experiences for Wearables,” July. Accessed January 30, 2018.
https://doi.org/10.2139/ssrn.2630715.
Dontje, Manon L., Martijn de Groot, Remko R. Lengton, Cees P. van der Schans, and Wim P.
Krijnen. 2015. “Measuring Steps with the Fitbit Activity Tracker: An Inter-Device Reliability
Study.” Journal of Medical Engineering & Technology 39 (5): 28690.
Elsden, Chris, and David S. Kirk. 2014. “A Quantified Past: Remembering with Personal
Informatics.” In Proceedings of the 2014 Companion Publication on Designing Interactive
Systems, 45–48. DIS Companion ’14. New York, NY, USA: ACM.
Elsden, Chris, David S. Kirk, and Abigail C. Durrant. 2016. “A Quantified Past: Toward Design
for Remembering With Personal Informatics.” HumanComputer Interaction 31 (6): 518
57.
Epstein, Daniel A., An Ping, James Fogarty, and Sean A. Munson. 2015. “A Lived Informatics
Model of Personal Informatics.” In Proceedings of the 2015 ACM International Joint
Conference on Pervasive and Ubiquitous Computing, 731–42. UbiComp ’15. New York,
NY, USA: ACM.
Epstein, Daniel A., Jennifer H. Kang, Laura R. Pina, James Fogarty, and Sean A. Munson.
2016. “Reconsidering the Device in the Drawer: Lapses as a Design Opportunity in
Personal Informatics.” Proceedings of the ... ACM International Conference on Ubiquitous
Computing . UbiComp (Conference) 2016 (September): 82940.
Evenson, Kelly R., Michelle M. Goto, and Robert D. Furberg. 2015. “Systematic Review of the
Validity and Reliability of Consumer-Wearable Activity Trackers.” The International Journal
of Behavioral Nutrition and Physical Activity 12 (December): 159.
Fausset, Cara Bailey, Tracy L. Mitzner, Chandler E. Price, Brian D. Jones, Brad W. Fain, and
Wendy A. Rogers. 2013. “Older Adults’ Use of and Attitudes toward Activity Monitoring
Technologies.” Proceedings of the Human Factors and Ergonomics Society ... Annual
Meeting Human Factors and Ergonomics Society. Meeting 57 (1): 168387.
Feldman, Martha S., and Wanda J. Orlikowski. 2011. “Theorizing Practice and Practicing
Theory.” Organization Science 22 (5): 124053.
Ferrara, Cynthia M., Shelby Smyth, Erin Mullan, and Christopher Burke. 2017. “Caloric
Expenditure Using Indirect Calorimetry, Apple Watch Sport, and Fitbit Zip.” Journal of
Exercise Physiology Online / American Society of Exercise Physiologists 20 (3). Accessed
January 30, 2018. http://www.asep.org/asep/asep/JEPonlineJUNE2017_Ferrara.pdf.
Fidel, Raya. 2012. Human Information Interaction: An Ecological Approach to Information
Behavior. MIT Press.
Fokkema, Tryntsje, Thea J. M. Kooiman, Wim P. Krijnen, Cees P. VAN DER Schans, and
Martijn DE Groot. 2017. “Reliability and Validity of Ten Consumer Activity Trackers Depend
on Walking Speed.” Medicine and Science in Sports and Exercise 49 (4): 793800.
Fritz, Thomas, Elaine M. Huang, Gail C. Murphy, and Thomas Zimmermann. 2014. “Persuasive
Technology in the Real World: A Study of Long-Term Use of Activity Sensing Devices for
Fitness.” In Proceedings of the SIGCHI Conference on Human Factors in Computing
Systems, 487–96. CHI ’14. New York, NY, USA: ACM.
Fulk, George D., Stephanie A. Combs, Kelly A. Danks, Coby D. Nirider, Bhavana Raja, and
Darcy S. Reisman. 2014. “Accuracy of 2 Activity Monitors in Detecting Steps in People with
Stroke and Traumatic Brain Injury.” Physical Therapy 94 (2): 22229.
Ghassemi, Marzyeh, Tristan Naumann, Finale Doshi-Velez, Nicole Brimmer, Rohit Joshi, Anna
Rumshisky, and Peter Szolovits. 2014. “Unfolding Physiological State: Mortality Modelling
in Intensive Care Units.” KDD: Proceedings / International Conference on Knowledge
Discovery & Data Mining. International Conference on Knowledge Discovery & Data Mining
2014 (August): 7584.
Gibson, James J. 1977. “The Theory of Affordances, In” Perceiving, Acting and Knowing”, Eds.
RE Shaw and J. Bransford.” Erlbaum.
Gillinov, Stephen, Muhammad Etiwy, Robert Wang, Gordon Blackburn, Dermot Phelan, A. Marc
Gillinov, Penny Houghtaling, Hoda Javadikasgari, and Milind Y. Desai. 2017. “Variable
Accuracy of Wearable Heart Rate Monitors during Aerobic Exercise.” Medicine and
Science in Sports and Exercise 49 (8): 16971703.
Gouveia, Rúben, Evangelos Karapanos, and Marc Hassenzahl. 2015. “How Do We Engage
with Activity Trackers?: A Longitudinal Study of Habito.” In Proceedings of the 2015 ACM
International Joint Conference on Pervasive and Ubiquitous Computing, 130516.
UbiComp ’15. New York, NY, USA: ACM.
Gresham, Gillian K., Elad Neeman, Andrew Eugene Hendifar, Quanlin Li, Richard Tuli, Bj Rimel,
Christine S. Walsh, et al. 2017. “Assessing Performance Status and Clinical Outcomes with
Wearable Activity Monitors.” Journal of Clinical Orthodontics: JCO 35 (15_suppl): 6571
6571.
Griffiths, Thomas L., and Mark Steyvers. 2004. “Finding Scientific Topics.” Proceedings of the
National Academy of Sciences of the United States of America 101 Suppl 1 (April): 5228
35.
Gulotta, Rebecca, Jodi Forlizzi, Rayoung Yang, and Mark Wah Newman. 2016. “Fostering
Engagement with Personal Informatics Systems.” In Proceedings of the 2016 ACM
Conference on Designing Interactive Systems, 286300.
Guo, Fangfang, Yu Li, Mohan S. Kankanhalli, and Michael S. Brown. 2013. “An Evaluation of
Wearable Activity Monitoring Devices.” In Proceedings of the 1st ACM International
Workshop on Personal Data Meets Distributed Multimedia, 31–34. PDM ’13. New York,
NY, USA: ACM.
Harrison, Daniel, Paul Marshall, Nadia Bianchi-Berthouze, and Jon Bird. 2015. “Activity
Tracking: Barriers, Workarounds and Customisation.” In Proceedings of the 2015 ACM
International Joint Conference on Pervasive and Ubiquitous Computing, 61721. UbiComp
’15. New York, NY, USA: ACM.
Hartman, Sheri J., Sandahl H. Nelson, Lisa A. Cadmus-Bertram, Ruth E. Patterson, Barbara A.
Parker, and John P. Pierce. 2016. “Technology- and Phone-Based Weight Loss
Intervention: Pilot RCT in Women at Elevated Breast Cancer Risk.” American Journal of
Preventive Medicine 51 (5): 71421.
Harzing, Anne-Wil, and Satu Alakangas. 2016. “Google Scholar, Scopus and the Web of
Science: A Longitudinal and Cross-Disciplinary Comparison.” Scientometrics 106 (2): 787
804.
Hayes, Lynda B., and Carole M. Van Camp. 2015. “Increasing Physical Activity of Children
during School Recess.” Journal of Applied Behavior Analysis 48 (3): 69095.
Healy, Genevieve N., Charles E. Matthews, David W. Dunstan, Elisabeth A. H. Winkler, and
Neville Owen. 2011. “Sedentary Time and Cardio-Metabolic Biomarkers in US Adults:
NHANES 200306.” European Heart Journal 32 (5): 59097.
Hooke, Mary C., Laura Gilchrist, Lynn Tanner, Nicole Hart, and Janice S. Withycombe. 2016.
“Use of a Fitness Tracker to Promote Physical Activity in Children With Acute
Lymphoblastic Leukemia.” Pediatric Blood & Cancer 63 (4): 68489.
Hoy, Matthew B. 2016. “Personal Activity Trackers and the Quantified Self.” Medical Reference
Services Quarterly 35 (1): 94100.
Husted, Hannah M., and Tamra L. Llewellyn. 2017. “The Accuracy of Pedometers in Measuring
Walking Steps on a Treadmill in College Students.” International Journal of Exercise
Science 10 (1): 14653.
Jarrahi, Mohammad Hossein. 2015. “Digital and Physical Materiality of Information
Technologies: The Case of Fitbit Activity Tracking Devices.” In 2015 48th Hawaii
International Conference on System Sciences (HICSS), 176877. IEEE Computer Society.
Jarrahi, Mohammad Hossein, Nicci Gafinowitz, and Grace Shin. 2017. “Activity Trackers, Prior
Motivation, and Perceived Informational and Motivational Affordances.” Personal and
Ubiquitous Computing, November, 116.
Jarrahi, Mohammad Hossein, Sarah Beth Nelson, and Leslie Thomson. 2017. “Personal Artifact
Ecologies in the Context of Mobile Knowledge Workers.” Computers in Human Behavior 75
(October): 46983.
Karami, Amir. 2015. University of Maryland, Baltimore County, ProQuest Dissertations
Publishing. 3721828.
Karami A, Gangopadhyay A, Zhou B, Kharrazi H. 2015a. A fuzzy approach model for
uncovering hidden latent semantic structure in medical text collections. Proceedings of the
iConference. Newport Beach, California.
Karami A, Gangopadhyay A, Zhou B, Karrazi H. 2015b. Flatm: A fuzzy logic approach topic
model for medical documents. In the 2015 Annual Conference of the North American Fuzzy
Information Processing Society (NAFIPS) held jointly with 2015 5th World Conference on
Soft Computing (WConSC), Redmond, WA. IEEE.
Karami, Amir, London S. Bennett, and Xiaoyun He. 2018. “Mining Public Opinion about
Economic Issues: Twitter and the U.S. Presidential Election.” arXiv [cs.SI]. arXiv. Accessed
January 30, 2018. http://arxiv.org/abs/1802.01786.
Karami A, Dahl AA, Turner-McGrievy G, Kharrazi H, Shaw G. 2018a. Characterizing
diabetes, diet, exercise, and obesity comments on Twitter. International Journal of
Information Management. 38(1):1-6.
Karami A, Webb F, Kitzie VL. 2018b. Characterizing transgender health issues in twitter.
Proceedings of the Association for Information Science and Technology. Vancouver,
Canada.
Karami, Amir, Aryya Gangopadhyay, Bin Zhou, and Hadi Kharrazi. 2018c. "Fuzzy
approach topic discovery in health and medical corpora." International Journal of Fuzzy
Systems 20, no. 4 (2018): 1334-1345.
Karapanos, Evangelos, Rúben Gouveia, Marc Hassenzahl, and Jodi Forlizzi. 2016. “Wellbeing
in the Making: Peoples’ Experiences with Wearable Activity Trackers.” Psychology of Well-
Being 6 (June): 4.
Kim, Da-Jung, Yeoreum Lee, Saeyoung Rho, and Youn-Kyung Lim. 2016. “Design
Opportunities in Three Stages of Relationship Development Between Users and Self-
Tracking Devices.” In Proceedings of the 2016 CHI Conference on Human Factors in
Computing Systems, 699–703. CHI ’16. New York, NY, USA: ACM.
Kooiman, Thea J. M., Manon L. Dontje, Siska R. Sprenger, Wim P. Krijnen, Cees P. van der
Schans, and Martijn de Groot. 2015. “Reliability and Validity of Ten Consumer Activity
Trackers.” BMC Sports Science, Medicine and Rehabilitation 7 (October): 24.
Kroll, Ryan R., J. Gordon Boyd, and David M. Maslove. 2016. “Accuracy of a Wrist-Worn
Wearable Device for Monitoring Heart Rates in Hospital Inpatients: A Prospective
Observational Study.” Journal of Medical Internet Research 18 (9): e253.
Kumari, Vishakha, and Sara Anne Hook. 2017. “The Privacy, Security and Discoverability of
Data on Wearable Health Devices: Fitness or Folly?” In Universal Access in Human
Computer Interaction. Human and Technological Environments, edited by Margherita
Antona and Constantine Stephanidis, 10279:5064. Lecture Notes in Computer Science.
Cham: Springer International Publishing.
Lamb, Karen, Hsiao-Ying Huang, Andrew Marturano, and Masooda Bashir. 2016. “Users’
Privacy Perceptions About Wearable Technology: Examining Influence of Personality,
Trust, and Usability.” In Advances in Human Factors in Cybersecurity, 5568. Advances in
Intelligent Systems and Computing. Springer, Cham.
Ledger, Dan, and Daniel McCaffrey. 2014. “Inside Wearables: How the Science of Human
Behavior Change Offers the Secret to Long-Term Engagement.” Endeavour Partners 200
(93): 1.
Lee, H., and Y. Lee. 2017. “A Look at Wearable Abandonment.” In 2017 18th IEEE International
Conference on Mobile Data Management (MDM), 39293.
Lee, Jung-Min, Youngwon Kim, and Gregory J. Welk. 2014. “Validity of Consumer-Based
Physical Activity Monitors.” Medicine and Science in Sports and Exercise 46 (9): 184048.
Lee, S., J. Baker, J. Song, and J. C. Wetherbe. 2010. “An Empirical Comparison of Four Text
Mining Methods.” In 2010 43rd Hawaii International Conference on System Sciences, 1
10.
Leth, Soren, John Hansen, Olav W. Nielsen, and Birthe Dinesen. 2017. “Evaluation of
Commercial Self-Monitoring Devices for Clinical Purposes: Results from the Future Patient
Trial, Phase I.” Sensors 17 (1). Accessed January 30, 2018.
https://doi.org/10.3390/s17010211.
Lewis, Zakkoyya H., Elizabeth J. Lyons, Jessica M. Jarvis, and Jacques Baillargeon. 2015.
“Using an Electronic Activity Monitor System as an Intervention Modality: A Systematic
Review.” BMC Public Health 15 (June): 585.
Li, Ian, Jon Froehlich, Jakob E. Larsen, Catherine Grevet, and Ernesto Ramirez. 2013.
“Personal Informatics in the Wild: Hacking Habits for Health & Happiness.” In CHI ’13
Extended Abstracts on Human Factors in Computing Systems, 3179–82. CHI EA ’13. New
York, NY, USA: ACM.
Liu, Ying-Hsang, Paul Scifleet, Lisa M. Given, and Others. 2014. “Contexts of Information
Seeking in Self-Tracking and the Design of Lifelogging Systems.” In MindTheGap@
iConference, 4450.
Mackintosh, Kelly. 2016. “Mission Possible: Using Ubiquitous Social Goal Sharing Technology
to Promote Physical Activity in Children.” Movement, Health & Exercise 5 (2): 1.
Matthews, Charles E., Stephanie M. George, Steven C. Moore, Heather R. Bowles, Aaron Blair,
Yikyung Park, Richard P. Troiano, Albert Hollenbeck, and Arthur Schatzkin. 2012. “Amount
of Time Spent in Sedentary Behaviors and Cause-Specific Mortality in US Adults.” The
American Journal of Clinical Nutrition 95 (2): 43745.
McCallum, A. K. 2002. “MALLET: A Machine Learning for Language Toolkit.” Accessed January
30, 2018. http://www.citeulike.org/group/3030/article/1062263.
McMahon, Siobhan K., Beth Lewis, Michael Oakes, Weihua Guan, Jean F. Wyman, and
Alexander J. Rothman. 2016. “Older Adults’ Experiences Using a Commercially Available
Monitor to Self-Track Their Physical Activity.” JMIR mHealth and uHealth 4 (2): e35.
Mercer, Kathryn, Lora Giangregorio, Eric Schneider, Parmit Chilana, Melissa Li, and Kelly
Grindrod. 2016. “Acceptance of Commercially Available Wearable Activity Trackers Among
Adults Aged Over 50 and With Chronic Illness: A Mixed-Methods Evaluation.” JMIR
mHealth and uHealth 4 (1): e7.
Mercer, Kathryn, Melissa Li, Lora Giangregorio, Catherine Burns, and Kelly Grindrod. 2016.
“Behavior Change Techniques Present in Wearable Activity Trackers: A Critical Analysis.”
JMIR mHealth and uHealth 4 (2): e40.
Meyer, Jochen, Jutta Fortmann, Merlin Wasmann, and Wilko Heuten. 2015. “Making Lifelogging
Usable: Design Guidelines for Activity Trackers.” In MultiMedia Modeling, 32334. Lecture
Notes in Computer Science. Springer, Cham.
Meyer, J., J. Schnauber, W. Heuten, H. Wienbergen, R. Hambrecht, H. J. Appelrath, and S.
Boll. 2016. “Exploring Longitudinal Use of Activity Trackers.” In 2016 IEEE International
Conference on Healthcare Informatics (ICHI), 198206.
Miyamoto, Sheridan W., Stuart Henderson, Heather M. Young, Amit Pande, and Jay J. Han.
2016. “Tracking Health Data Is Not Enough: A Qualitative Exploration of the Role of
Healthcare Partnerships and mHealth Technology to Promote Physical Activity and to
Sustain Behavior Change.” JMIR mHealth and uHealth 4 (1): e5.
Mortenson, Michael J., and Richard Vidgen. 2016. “A Computational Literature Review of the
Technology Acceptance Model.” International Journal of Information Management 36 (6,
Part B): 124859.
Mortier, Richard, Hamed Haddadi, Tristan Henderson, Derek McAuley, and Jon Crowcroft.
2014. “Human-Data Interaction: The Human Face of the Data-Driven Society,” October.
Accessed January 30, 2018. https://doi.org/10.2139/ssrn.2508051. Available at
https://ssrn.com/abstract=2508051 or http://dx.doi.org/10.2139/ssrn.2508051
Munson, Sean A., Erin Krupka, Caroline Richardson, and Paul Resnick. 2015. “Effects of Public
Commitments and Accountability in a Technology-Supported Physical Activity
Intervention.” In Proceedings of the 33rd Annual ACM Conference on Human Factors in
Computing Systems, 1135–44. CHI ’15. New York, NY, USA: ACM.
Nardi, Bonnie A., and Vicki O’Day. 1999. Information Ecologies: Using Technology with Heart.
MIT Press.
Naslund, John A., Kelly A. Aschbrenner, Laura K. Barre, and Stephen J. Bartels. 2015.
“Feasibility of Popular m-Health Technologies for Activity Tracking among Individuals with
Serious Mental Illness.” Telemedicine Journal and E-Health: The Official Journal of the
American Telemedicine Association 21 (3): 21316.
Ng, Ada, Madhu Reddy, Alyson K. Zalta, and Stephen M. Schueller. 2018. “Veterans’
Perspectives on Fitbit Use in Treatment for Post-Traumatic Stress Disorder: An Interview
Study.” JMIR Mental Health 5 (2): e10415.
Paul, Greig, and James Irvine. 2014. “Privacy Implications of Wearable Health Devices.” In
Proceedings of the 7th International Conference on Security of Information and Networks,
117:117–117:121. SIN ’14. New York, NY, USA: ACM.
Pettigrew, Karen E., Raya Fidel, and Harry Bruce. 2001. “Conceptual Frameworks in
Information Behavior.” Annual Review of Information Science and Technology (ARIST) 35
(43-78). http://faculty.washington.edu/fidelr/RayaPubs/ConceptualFrameworks.pdf.
Pina, L. R., E. Ramirez, and W. G. Griswold. 2012. “Fitbit+: A Behavior-Based Intervention
System to Reduce Sedentary Behavior.” In 2012 6th International Conference on
Pervasive Computing Technologies for Healthcare (PervasiveHealth) and Workshops,
17578.
Ploderer, Bernd, Wolfgang Reitberger, Harri Oinas-Kukkonen, and Julia van Gemert-Pijnen.
2014. “Social Interaction and Reflection for Behaviour Change.” Personal and Ubiquitous
Computing 18 (7): 166776.
Preusse, Kimberly C., Tracy L. Mitzner, Cara Bailey Fausset, and Wendy A. Rogers. 2016.
“Older Adults’ Acceptance of Activity Trackers.” Journal of Applied Gerontology: The
Official Journal of the Southern Gerontological Society 36 (2): 12755.
Randriambelonoro, Mirana, Yu Chen, Antoine Geissbuhler, and Pearl Pu. 2015. “Exploring
Physical Activity Monitoring Devices for Diabetic and Obese Patients.” In Adjunct
Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous
Computing and Proceedings of the 2015 ACM International Symposium on Wearable
Computers, 1003–8. UbiComp/ISWC’15 Adjunct. New York, NY, USA: ACM.
Rapp, Amon, and Federica Cena. 2016. “Personal Informatics for Everyday Life: How Users
without Prior Self-Tracking Experience Engage with Personal Data.” International Journal
of Human-Computer Studies 94 (October): 117.
Rapp, Amon, Federica Cena, and Alessandro Marcengo. 2018. “Editorial of the Special Issue
on Quantified Self and Personal Informatics.” Computers 7 (1): 14.
Rapp, Amon, Alessandro Marcengo, Luca Buriano, Giancarlo Ruffo, Mirko Lai, and Federica
Cena. 2018. “Designing a Personal Informatics System for Users without Experience in
Self-Tracking: A Case Study.” Behaviour & Information Technology, February, 132.
Rapp, Amon, and Maurizio Tirassa. 2017. “Know Thyself: A Theory of the Self for Personal
Informatics.” HumanComputer Interaction 32 (5-6): 33580.
Rita Guerra, A. 2015. “Nike Finally Explains Why the FuelBand Was Cancelled.” Blasting News.
May 6, 2015. Accessed January 30, 2018. https://uk.blastingnews.com/tech/2015/05/nike-
finally-explains-why-the-fuelband-was-cancelled-00381991.html.
Roe, Justin, Lucy Salmon, and Joshua Twiggs. 2016. “Objective Measure of Activity Level After
Total Knee Arthroplasty with The Use of the ‘Fitbit’ Device.” Orthopaedic Journal of Sports
Medicine 4 (2_suppl): 2325967116S00012.
Rooksby, John, Mattias Rost, Alistair Morrison, and Matthew Chalmers Chalmers. 2014.
“Personal Tracking As Lived Informatics.” In Proceedings of the 32Nd Annual ACM
Conference on Human Factors in Computing Systems, 1163–72. CHI ’14. New York, NY,
USA: ACM.
Savolainen, Reijo. 1995. “Everyday Life Information Seeking: Approaching Information Seeking
in the Context of ‘way of Life.’” Library & Information Science Research 17 (3): 25994.
Savolainen, Reijo. 2007. “Information Behavior and Information Practice: Reviewing the
‘umbrella Concepts’ of Information-Seeking Studies.” The Library Quarterly 77 (2): 10932.
Sawyer, Steve, and Haiyan Huang. 2007. “Conceptualizing Information, Technology, and
People: Comparing Information Science and Information Systems Literatures.” Journal of
the American Society for Information Science. American Society for Information Science 58
(10): 143647.
Schaefer, Sara E., Cynthia Carter Ching, Heather Breen, and J. Bruce German. 2016.
“Wearing, Thinking, and Moving: Testing the Feasibility of Fitness Tracking with Urban
Youth.” American Journal of Health Education / American Alliance for Health, Physical
Education, Recreation, and Dance 47 (1): 816.
Sears, Tiffany, Elmer Alvalos, Samantha Lawson, Ian McAlister, L. Chris Eschbach, and
Jennifer Bunn. 2017. “Wrist-Worn Physical Activity Trackers Tend to Underestimate Steps
During Walking.” International Journal of Exercise Science 10 (5): 76473.
Shaw, George, and Amir Karami. 2017. “Computational Content Analysis of Negative Tweets
for Obesity, Diet, Diabetes, and Exercise.” Proceedings of the Association for Information
Science and Technology 54 (1): 35765.
Shih, Patrick C., Kyungsik Han, Erika Shehan Poole, Mary Beth Rosson, and John M. Carroll.
2015. “Use and Adoption Challenges of Wearable Activity Trackers.” iConference 2015
Proceedings. Available at http://hdl.handle.net/2142/73649
Shinde, Arvind Manohar, Gillian K. Gresham, Andrew Eugene Hendifar, Quanlin Li, Brennan
Spiegel, Bj Rimel, Christine S. Walsh, Richard Tuli, Steven Piantadosi, and Robert A.
Figlin. 2017. “Correlating Wearable Activity Monitor Data with PROMIS Detected Distress
and Physical Functioning in Advanced Cancer Patients.” Journal of Clinical Orthodontics:
JCO 35 (15_suppl): e21689e21689.
Shin, Grace, Cheon, Eun Jeong, and Jarrahi, Mohammad Hossein. 2015. “Understanding
Quantified-Selfers’ Interplay between Intrinsic and Extrinsic Motivation in the Use of
Activity-Tracking Devices.” iConference 2015 Proceedings. Available at
https://www.ideals.illinois.edu/handle/2142/73740.
Shin, Grace, Feng, Yuanyuan Feng, Jarrahi, Mohammad Hossein, and Gafinowitz, Nicci. 2018.
“Beyond novelty effect: a mixed-methods exploration into the motivation for long-term
activity tracker use.” JAMIA Open. Doi:10.1093/jamiaopen/ooy048.
Singh, Abhishek K., Craig Farmer, Maayken L. E. Van Den Berg, Maggie Killington, and
Christopher J. Barr. 2016. “Accuracy of the FitBit at Walking Speeds and Cadences
Relevant to Clinical Rehabilitation Populations.” Disability and Health Journal 9 (2): 320
23.
Sloan, Robert A., Youngdeok Kim, Aarti Sahasranaman, Falk Müller-Riemenschneider, Stuart J.
H. Biddle, and Eric A. Finkelstein. 2018. “The Influence of a Consumer-Wearable Activity
Tracker on Sedentary Time and Prolonged Sedentary Bouts: Secondary Analysis of a
Randomized Controlled Trial.” BMC Research Notes 11 (1): 189.
Smith, John D. 2016. “Accuracy Of Wrist-Worn Activity Monitors At Three Walking Speeds On
The Treadmill.” Medicine & Science in Sports & Exercise 48: 783.
Sprint, Gina, Diane Cook, Douglas Weeks, Jordana Dahmen, and Alyssa La Fleur. 2017.
“Analyzing Sensor-Based Time Series Data to Track Changes in Physical Activity during
Inpatient Rehabilitation.” Sensors 17 (10). https://doi.org/10.3390/s17102219.
Stahl, Sarah E., Hyun-Sung An, Danae M. Dinkel, John M. Noble, and Jung-Min Lee. 2016.
“How Accurate Are the Wrist-Based Heart Rate Monitors during Walking and Running
Activities? Are They Accurate Enough?” BMJ Open Sport & Exercise Medicine 2 (1):
e000106.
Takacs, Judit, Courtney L. Pollock, Jerrad R. Guenther, Mohammadreza Bahar, Christopher
Napier, and Michael A. Hunt. 2014. “Validation of the Fitbit One Activity Monitor Device
during Treadmill Walking.” Journal of Science and Medicine in Sport / Sports Medicine
Australia 17 (5): 496500.
Tang, Lie Ming, Margot Day, Lina Engelen, Philip Poronnik, Adrian Bauman, and Judy Kay.
2016. “Daily & Hourly Adherence: Towards Understanding Activity Tracker Accuracy.” In
Proceedings of the 2016 CHI Conference Extended Abstracts on Human Factors in
Computing Systems, 3211–18. CHI EA ’16. New York, NY, USA: ACM.
Thomas, J. Graham, Hollie A. Raynor, Dale S. Bond, Anna K. Luke, Chelsi C. Cardoso, Gary D.
Foster, and Rena R. Wing. 2017. “Weight Loss in Weight Watchers Online with and without
an Activity Tracking Device Compared to Control: A Randomized Trial.” Obesity 25 (6):
101421.
Thorup, Charlotte Brun, Jan Jesper Andreasen, Erik Elgaard Sørensen, Mette Grønkjær, Birthe
Irene Dinesen, and John Hansen. 2017. “Accuracy of a Step Counter during Treadmill and
Daily Life Walking by Healthy Adults and Patients with Cardiac Disease.” BMJ Open 7 (3):
e011742.
Tran, Tiffany, and Dan Nathan-Roberts. 2018. “Design Considerations in Wearable Technology
for Patients with Bipolar Disorder.” Proceedings of the Human Factors and Ergonomics
Society ... Annual Meeting Human Factors and Ergonomics Society. Meeting 62 (1): 1187
91.
Treacy, Daniel, Leanne Hassett, Karl Schurr, Sakina Chagpar, Serene S. Paul, and Catherine
Sherrington. 2017. “Validity of Different Activity Monitors to Count Steps in an Inpatient
Rehabilitation Setting.” Physical Therapy 97 (5): 58188.
Valbuena, Diego, Raymond Miltenberger, and Elizabeth Solley. 2015. “Evaluating an Internet-
Based Program and a Behavioral Coach for Increasing Physical Activity.” Behavior
Analysis: Research and Practice 15 (2): 122.
Van Remoortel, Hans, Santiago Giavedoni, Yogini Raste, Chris Burtin, Zafeiris Louvaris, Elena
Gimeno-Santos, Daniel Langer, et al. 2012. “Validity of Activity Monitors in Health and
Chronic Disease: A Systematic Review.” The International Journal of Behavioral Nutrition
and Physical Activity 9 (July): 84.
Wang, Julie B., Lisa A. Cadmus-Bertram, Loki Natarajan, Martha M. White, Hala Madanat,
Jeanne F. Nichols, Guadalupe X. Ayala, and John P. Pierce. 2015. “Wearable
Sensor/Device (Fitbit One) and SMS Text-Messaging Prompts to Increase Physical Activity
in Overweight and Obese Adults: A Randomized Controlled Trial.” Telemedicine Journal
and E-Health: The Official Journal of the American Telemedicine Association 21 (10): 782
92.
Wang, Julie B., Janine K. Cataldo, Guadalupe X. Ayala, Loki Natarajan, Lisa A. Cadmus-
Bertram, Martha M. White, Hala Madanat, Jeanne F. Nichols, and John P. Pierce. 2016.
“Mobile and Wearable Device Features That Matter in Promoting Physical Activity.” Journal
of Mobile Technology in Medicine 5 (2): 211.
Washington, Wendy Donlin, Kelly M. Banna, and Amanda L. Gibson. 2014. “Preliminary
Efficacy of Prize-Based Contingency Management to Increase Activity Levels in Healthy
Adults.” Journal of Applied Behavior Analysis 47 (2): 23145.
Webb F, Karami A, Kitzie V. 2018. Characterizing Diseases and disorders in Gay Users' tweets.
Proceedings of the Southern Association for Information Systems (SAIS). Atlanta, Georgia.
Wilson, T. D. 1999. “Models in Information Behaviour Research.” Journal of Documentation 55
(3): 24970.
Winter, Joost C. F. de, Amir A. Zadpoor, and Dimitra Dodou. 2014. “The Expansion of Google
Scholar versus Web of Science: A Longitudinal Study.” Scientometrics 98 (2): 154765.
Wong, Christopher K., Helena M. Mentis, and Ravi Kuber. 2018. “The Bit Doesn’t Fit: Evaluation
of a Commercial Activity-Tracker at Slower Walking Speeds.” Gait & Posture 59 (January):
17781.
“Worldwide Wearables Market Grows 7.3% in Q3 2017 as Smart Wearables Rise and Basic
Wearables Decline, Says IDC.” n.d. Www.idc.com. Accessed January 30, 2018.
https://www.idc.com/getdoc.jsp?containerId=prUS43260217.
Yavelberg, Loren, Dessi Zaharieva, Ali Cinar, Michael C. Riddell, and Veronica Jamnik. 2018.
“A Pilot Study Validating Select Research-Grade and Consumer-Based Wearables
Throughout a Range of Dynamic Exercise Intensities in Persons With and Without Type 1
Diabetes: A Novel Approach.” Journal of Diabetes Science and Technology 12(3):56976.
Zhao, Zhao, Ali Arya, Anthony Whitehead, Gerry Chan, and S. Ali Etemad. 2017. “Keeping
Users Engaged Through Feature Updates: A Long-Term Study of Using Wearable-Based
Exergames.” In Proceedings of the 2017 CHI Conference on Human Factors in Computing
Systems, 1053–64. CHI ’17. New York, NY, USA: ACM.
... Mounting evidence suggests the benefits of wearable devices for tracking and monitoring health as safe and cost-effective tools to promote health behavior change such as enhancing PA [22]. Further, the data generating capabilities of wearables provide substantial value to the users in their health management [23]. In medical settings and patient treatment, incorporating and use of wearables data in the healthcare decision-making has shown great potential in patient monitoring and enhancing planning and intervention by providing a timelier feedback [23]. ...
... Further, the data generating capabilities of wearables provide substantial value to the users in their health management [23]. In medical settings and patient treatment, incorporating and use of wearables data in the healthcare decision-making has shown great potential in patient monitoring and enhancing planning and intervention by providing a timelier feedback [23]. ...
... There are several acceptance and abandonment related issues that substantially influence long-term wearable use. Some of these factors include device appearance, display and interaction, wearability, perceived usefulness or risks, and other technical issues such as data measurement and presentation [23,50]. Other factors that influence wearable use include development of more personalized wearables for various groups of populations who have different needs and preferences. ...
Preprint
Full-text available
Background: With an increase in aging population and chronic medical conditions in the United States (US), the role of informal caregivers has become paramount as they engage in the care of their loved ones. Mounting evidence suggests that such responsibilities place significant burdens on informal caregivers and can negatively impact their own health. New wearable health and activity trackers (wearables) are increasingly being used to facilitate and monitor healthy behaviors and to improve health outcomes. Although prior studies have examined the efficacy of wearables in improving health and wellbeing in the general population, little is known about their benefits among informal caregivers. Objective: This study examines the association between use of wearables and levels of physical activity among informal caregivers in the US. Methods: We utilized data from the National Cancer Institute’s Health Information National Trends Survey (HINTS 5-Cycle 3 [2019] & Cycle 4 [2020]) for a nationally-representative sample of 1,273 community-dwelling informal caregivers (aged ≥18 years; females = 60%; some college or more in education = 75.7%; caring for patients with ≥one chronic medical condition = 67.3%) in the US. Using jackknife replicate weights, a multivariable logistic regression was fit to assess an independent association between the use of wearables and a binary outcome: meeting or not meeting the current World Health Organization’s recommendation on physical activity for adults (≥150 minutes of at least moderate-intensity physical activity per week). Results: Over one-third (37.8%; n=466) of the informal caregivers met the recommendations for adult physical activity. However, those who reported using wearables (n=390) had slightly higher odds of meeting physical activity recommendations (Adjusted OR= 1.10; 95% Confidence Interval: 1.04, 1.77; P=.042) compared to those who did not use wearables. Conclusions: The results demonstrate a positive association between the use of wearables and levels of physical activity among informal caregivers in the US. Therefore, efforts to incorporate wearable technology into the development of health-promoting programs or interventions for informal caregivers could potentially improve their health and wellbeing. Future longitudinal studies are required to further support the current findings.
... Research on lifestyle is largely dependent on individuals' subjective reports and memories, limiting the reliability of those reports. Therefore, objective measures of lifestyle elements are an important issue in this field [3,4]. ...
... A wearable activity tracker is an electronic device attached to the body surface, which tracks and monitors health-related behaviors, including step counts, heart rates (HRs), and sleep patterns [4]. It continuously generates health-related data and provides significant value in patient care [5]. ...
... In addition, the large amount of information obtained through this device could be utilized for research purposes [11]. Researchers may also incorporate lifestyle information into studies of behavioral intervention and monitoring programs to improve the management of various types of disease [4]. Our study also suggested that parameters collected by an activity tracker could be used to improve clinicians' ability to identify healthy and unhealthy lifestyles beyond self-rewww.e-enm.org ...
Article
Lifestyle is a critical aspect of diabetes management. We aimed to define a healthy lifestyle using objectively measured parameters obtained from a wearable activity tracker (Fitbit) in patients with type 2 diabetes. This prospective observational study included 24 patients (mean age, 46.8 years) with type 2 diabetes. Expectation-maximization clustering analysis produced two groups: A (n=9) and B (n=15). Group A had a higher daily step count, lower resting heart rate, longer sleep duration, and lower mean time differences in going to sleep and waking up than group B. A Shapley additive explanation summary analysis indicated that sleep-related factors were key elements for clustering. The mean hemoglobin A1c level was 0.3 percentage points lower at the end of follow-up in group A than in group B. Factors related to regular sleep patterns could be possible determinants of lifestyle clustering in patients with type 2 diabetes.
... Although no participants in this study directly questioned the accuracy of wearable activity trackers, the algorithms embedded within devices are typically developed using the movement patterns of healthy people, which may lead to higher data inaccuracies when applied to people with HD, who often present with motor symptoms involving their upper limbs and pathological gait characteristics. A recent review of wearable activity tracker research [25] found a substantial increase in the literature focusing on wearable activity trackers in recent years, with the most common research theme focusing on concerns about the reliability, accuracy, and validity of the technology itself. ...
... Concerns regarding data privacy were also raised in this study. These concerns are echoed in the literature [24,25,33]. Although patients may be open to sharing their personal information for the purpose of clinical benefit, there is a risk of data misuse. ...
Article
Background: There are early indications that lifestyle behaviors, specifically physical activity and sleep, may be associated with the onset and progression of Huntington disease (HD). Wearable activity trackers offer an exciting opportunity to collect long-term activity data to further investigate the role of lifestyle, physical activity, and sleep in disease modification. Given how wearable devices rely on user acceptance and long-term adoption, it is important to understand users' perspectives on how acceptable any device might be and how users might engage over the longer term. Objective: This study aimed to explore the perceptions, motivators, and potential barriers relating to the adoption of wearable activity trackers by people with HD for monitoring and managing their lifestyle and sleep. This information intended to guide the selection of wearable activity trackers for use in a longitudinal observational clinical study. Methods: We conducted a mixed methods study; this allowed us to draw on the potential strengths of both quantitative and qualitative methods. Opportunistic participant recruitment occurred at 4 Huntington's Disease Association meetings, including 1 international meeting and 3 United Kingdom-based regional meetings. Individuals with HD, their family members, and carers were invited to complete a user acceptance questionnaire and participate in a focus group discussion. The questionnaire consisted of 35 items across 8 domains using a 0 to 4 Likert scale, along with some additional demographic questions. Average questionnaire responses were recorded as positive (score>2.5), negative (score<1.5), or neutral (score between 1.5 and 2.5) opinions for each domain. Differences owing to demographics were explored using the Kruskal-Wallis and Wilcoxon rank sum tests. Focus group discussions (conducted in English) were driven by a topic guide, a vignette scenario, and an item ranking exercise. The discussions were audio recorded and then analyzed using thematic analysis. Results: A total of 105 completed questionnaires were analyzed (47 people with HD and 58 family members or carers). All sections of the questionnaire produced median scores >2.5, indicating a tendency toward positive opinions on wearable activity trackers, such as the devices being advantageous, easy and enjoyable to use, and compatible with lifestyle and users being able to understand the information from trackers and willing to wear them. People with HD reported a more positive attitude toward wearable activity trackers than their family members or caregivers (P=.02). A total of 15 participants participated in 3 focus groups. Device compatibility and accuracy, data security, impact on relationships, and the ability to monitor and self-manage lifestyle behaviors have emerged as important considerations in device use and user preferences. Conclusions: Although wearable activity trackers were broadly recognized as acceptable for both monitoring and management, various aspects of device design and functionality must be considered to promote acceptance in this clinical cohort.
... Research on personal informatics and everyday self-tracking behavior-including activity trackers-often distinguishes between different stages in interaction between humans and tracking devices, showing how people transition between preparation, collection, integration, reflection, and action (i.e., the five-stage model by Li et al., 2010). Studies in this field have informed designers about different user applications and improved the usability of activity trackers (Shin et al., 2019). Yet, there is far less attention for the skills that are required to use activity trackers for personal benefits. ...
... It has been demonstrated that more women than men have participated in research studies investigating the effectiveness of wearable activity devices when incorporated into all-encompassing weight loss programs [31]. It is also essential to keep in mind that the functionality of wearable activity tracker devices is not guaranteed to be accurate [32]. The need to adapt the type of device to the characteristics of the users has also been discussed, and it has been seen that it is important for long-term use to facilitate the user experience in terms of functionality, aesthetics, and physical design [33]. ...
Article
Full-text available
Regular physical activity has a positive impact on our physical and mental health. Adhering to a fixed physical activity regimen is essential for good health and mental wellbeing. Today, fitness trackers and smartphone applications are used to promote physical activity. These applications use step counts recorded by accelerometers to estimate physical activity. In this research, we performed a two-level clustering on a dataset based on individuals’ physical and physiological features, as well as past daily activity patterns. The proposed model exploits the user data with partial or complete features. To include the user with partial features, we trained the proposed model with the data of users who possess exclusive features. Additionally, we classified the users into several clusters to produce more accurate results for the users. This enables the proposed system to provide data-driven and personalized activity planning recommendations every day. A personalized physical activity plan is generated on the basis of hourly patterns for users according to their adherence and past recommended activity plans. Customization of activity plans can be achieved according to the user’s historical activity habits and current activity objective, as well as the likelihood of sticking to the plan. The proposed physical activity recommendation system was evaluated in real time, and the results demonstrated the improved performance over existing baselines.
... Many wearable devices have been released since. In addition to being used for personal physical activity tracking, these wearables are also increasingly being used in research as tools to collect health data [2][3][4]. These wearables are especially designed for long-term usage and thus facilitates longterm recording with low participant burden. ...