ActiThings Toolkit

Towards Supporting Older Adults’ Adherence to Home-based Physical Exercise

Programs by Providing Notifications in Opportune Moments

Elke Beck, Kai von Holdt and Jochen Meyer

OFFIS Institute for Information Technology, Escherweg 2, 26121 Oldenburg, Germany

Keywords: Health, Primary Prevention, Older Adults, Physical Activity, Ambient Reminders, Opportune Moments,

User Study.

Abstract: Home-based physical exercise programs can delay or even prevent age-related frailty among older adults,

but insufficient adherence to these programs, especially long-term, is a reoccurring problem. We suggest

aiding the integration of simple exercises into daily routine by using innovative interactive tools which

provide older adults with environmental prompts for exercises in opportune moments. In this paper, we

report on our lessons learned with exploring the usefulness of our ActiThings toolkit, which includes a

broad range of unobtrusive, portable, ubiquitous sensing and feedback technologies.

1 INTRODUCTION

Lifelong physical activity is one of the key

determinants of healthy aging. According to the

World Health Organization, at least 150 min of

moderate exercise per week has considerable

benefits for health among older adults and could

delay or even prevent age-related frailty. In

particular for fall prevention, older adults are

recommended to do muscle-strengthening activities

and physical activity that enhances balance.

However, globally, 1 out of 4 adults (18+ years) is

insufficiently active. Embedding physical activity

into the daily life requires changes of behaviours and

habits, which is a considerable challenge for many,

particularly elderly persons. There is a need for

opportunities to conduct exercises during daily life,

in an unobtrusive manner, and with little extra effort.

Strength, coordination, and balance home exercises

programs, such as HOPE (Clegg, Barber, Young,

Iliffe, and Forster, 2014), LIFE, or OTAGO address

these needs and are proven to be effective. Still,

insufficient adherence to exercise programs,

especially long-term, is a reoccurring problem.

Approximately 50% of people who start a physical

exercise programme will stop within the first 6

months (Robison and Rogers, 1994). Research on

the factors for adherence to exercise programs has

shown that besides demographic, physical and

psychological factors, supervision and motivational

support of the participating people provide

encouragement to exercise and thus contribute to a

better program adherence (Picorelli, Pereira, Pereira,

Felício, and Sherrington, 2014; Chao, Foy, and

Farmer, 2000; King, Taylor, Haskell, and Debusk,

1988).

2 RELATED WORK

A lot of research investigated the role of technology

to support healthy sedentary elderly in doing

physical activity in general (e.g., Fan, Forlizzi, and

Dey, 2012). Within this paper we will focus on

exploring how technological solutions can support

the adherence of healthy older adults to home-based

physical exercise programs.

Behaviour change theories offer numerous

techniques to induce physical exercise habits such as

providing information on consequences of

behaviour, goal setting, or providing rewards for

successful behaviour (Michie, Ashford, Sniehotta,

Dombrowski, Bishop, and French, 2011). For

instance, van het Reve, Silveira, Daniel, Casati, and

de Bruin (2014) investigated a physical exercise

intervention program for older adults, which

comprised individual and social motivation

strategies provided via a tablet-PC app. They found

that these technology-supported techniques can lead

Beck, E., Holdt, K. and Meyer, J.

ActiThings Toolkit - Towards Supporting Older Adults’ Adherence to Home-based Physical Exercise Programs by Providing Notiﬁcations in Opportune Moments .

DOI: 10.5220/0006647705190526

In Proceedings of the 11th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2018) - Volume 5: HEALTHINF, pages 519-526

ISBN: 978-989-758-281-3

519

to higher adherence rates compared to traditional

exercise plans (van het Reve et al., 2014).

Such intervention programs use of a multitude of

motivation strategies in parallel (e.g. Irvine, Gelatt,

Seeley, Macfarlane, and Gau, 2013; van het Reve et

al., 2014): Educational approaches aim improving

(health) knowledge among intervention participants.

Plan related approaches help participants to do

detailed planning of when and where to do physical

exercises. Participants may be rewarded for

attempting to achieve or actually achieving an

exercise goal. Social connections and social

awareness may be established between participants

and coaches via technological means. Technology,

such as a virtual gym application, can also create a

group experience while exercising (Baez, Far,

Ibarra, Ferron, Didino, and Casati, 2017), enabling

also less fit individuals to train as much as fitter

individuals. To support the (self-) monitoring of a

person’s physical exercise habits and performance,

exercise program websites can prompt older adults

to self-report activities and frequency of exercise

(Irvine et al., 2013), and, activity-tracking

technology can help older adults to get aware of

their physical activity level through monitoring their

physical activity and providing them with these data.

Furthermore, motion-tracking technology can

provide feedback on the correctness of physical

exercise conduction. Increasing positive emotions

may help the program participant with performing

the physical exercise, e.g., by making exercise fun.

This type is related to gamification of physical

activity or so-called exergames (games that involve

physical exercises).

To conclude, there are several different strategies

to support older adults’ adherence and motivation to

physical exercise with technology, but there is one

reoccurring effective characteristic among many of

the discussed studies: taking individual needs and

preferences of older adults towards physical

exercises into account. Such an individualized

tailoring of programs considers the personal level of

fitness (Baez et al., 2017), level of motivation

(Marcus, Bock, Pinto, Forsyth, Roberts, and

Traficante, 1998), and recognizes unique barriers in

older adults (e.g., intermittent illness or caregiving)

(Brawley, Rejeski, and King, 2003).

To achieve long-term adherence to physical

exercise beyond intervention participation, the

formation of habits and routines for physical

exercises in daily life is of utmost importance.

Research studies (e.g., Opdenacker, Boen,

Coorevits, and Delecluse, 2008) recommend

following a lifestyle approach to promoting physical

activity as an alternative to structured exercise

programs. This approach encourages older adults to

increase and maintain their physical activity level by

embedding balance, strength, and flexibility

exercises in their daily life through activities like, for

instance, walking, cycling, or swimming. Still,

identifying opportunities for physical exercise in

daily life can be a challenge. A suitable behavior

change strategy to overcome this issue is to use

environmental prompts to cue older adults about

performing physical exercises (Michie et al., 2011).

Some technology-supported home-based physical

exercise programs remind their participants via an

alarm clock or display invitations to start due

exercises (van het Reve et al., 2014), but these

technology-delivered prompts hardly consider the

suitability of this particular moment for an exercise.

Health intervention studies provide evidence that

health-related messages shown to people in suitable

points in time and place are effective in changing

people’s behaviour, e.g. in preferring to take the

stairs instead of an elevator after a point-of-decision

prompt. Intille (2004) explored the concept of non-

intrusive, “just-in-time” messaging for behaviour

change by using ubiquitous computing technology.

We assume this concept to be of value also for

triggering the conduction of physical exercises from

a home-based intervention program during the daily

lives of seniors.

Within our research about how technology can

support the integration and long-term maintenance

of a physical exercise program in the daily lives of

elderly, we build on top by exploring the usefulness

of innovative interactive tools which provide older

adults with environmental prompts for exercises in

opportune moments.

3 OUR APPROACH: ACTI-

THINGS TOOLKIT

Our research goal is supporting older adults’

adherence to home-based physical exercise

programs by reminding them about physical

exercises in possibly opportune moments. We

understand opportune moments as points in time

when a person has the capability and willingness to

get reminded about a physical exercise and to

conduct it. Such opportune moments are profoundly

dependent on individual preferences and may

happen any time throughout the day. They are

influenced by, e.g., having a sufficient amount of

time (a couple of minutes), being in a suitable spot at

HEALTHINF 2018 - 11th International Conference on Health Informatics

520

home, when interrupting the prevalent activity is not

a problem, and the like. Thus, the identification of

opportune moments (in general, as well as for a

single person) is a huge challenge. In our approach,

we deal with this complexity by identifying a

person’s routine activities at home which may fulfill

the conditions for doing a short physical exercise.

For instance, a person may be engaged in the routine

activity of making coffee, where she sits on a

kitchen chair, waiting for a couple of minutes until

the coffee machine is done. Such a moment could be

appropriate for making an environmental prompt to

the person, suggesting an instantaneous completion

of a physical exercise. Thus, we aim to integrate

physical exercises into the daily life of older adults

by intertwining the exercises with already

established activities at home, which should also

ease the formation of physical exercise habits.

Within our research, we are interested in examining

sensor-based technologies for detecting such routine

activities and interactive technologies for providing

visual and/or acoustic feedback (i.e., environmental

prompts to do a simple physcial exercise) to people.

Regarding the detection of opportune moments,

we chose to explore a broad range of unobtrusive,

portable and ubiquitous sensing technologies, which

identify the traces a person leaves in the

environment, when engaging in activities of daily

living. While daily activities like lying, sitting,

walking, running, or cycling can be recognized with

wearable accelerometry-based motion detectors

(e.g., Pärkkä, Ermes, Korpipaa, Mantyjarvi, Peltola,

and Korhonen, 2006), we need to go beyond by

capturing further data, such as a person’s location at

home, usage of objects (Tapia, Intille, and Larson,

2004), and interaction with devices. Next to

detecting opportune moments for notification and

physical exercise, suitable interactive technology

needs to notify an older adult about this physical

exercise opportunity. HCI research on reminders for

older adults has shown that, in order to be effective,

reminder systems need to be highly personalizable,

enabling the person to choose between diverse

modalities (vision, speech, sound, touch) (McGee-

Lennon, Wolters, and Brewster, 2011). Older adults

are a heterogenic group with different daily routines

and habits, varying capabilities, and also

considerable differences with regard to novel

technology use and acceptance. When selecting and

developing sensing and feedback technology for

supporting older adults, we must take these user-

group specific and personal differences into account.

Therefore, our focus is on developing a modular

toolkit, called ActiThings, with high- and low-

fidelity, small, easy to use, ambient, sensor-based

and interactive technologies.

We followed an iterative, human-centred

interaction design approach for achieving our

research goal. In a multi-stage process, we first

developed and evaluated conceptual designs of the

ActiThings toolkit. In a second stage, we iterated our

concepts based on the evaluation findings and built

functional prototypes, which we evaluated again.

For each stage, we investigated the usefulness of the

ActiThings toolkit by involving older adults in our

evaluation activities.

4 FIRST STAGE: CONCEPTUAL

DESIGNS

Within our first design stage, we developed three

conceptual designs. Thereby, we explored different

ways when – in which situations or activities – and

how – via which interactive devices and with which

information contents – to notify a person.

4.1 Three Prototypes

The first prototype, ActiMove, suggests to continue

with some physical exercise after the completion of

a routine (household) activity at home (e.g., opening

drawers, watering flowers). It consists of a device

with an accelerometer, which can be attached to

many household items (e.g. a watering can) to

identify their usage (see Figure 1). When the device

detects the termination of the activity (the household

item is inactive), it notifies the person with an

acoustic alarm and a red light about a possibly

opportune moment for an exercise. Moreover, the

prototype indicates the number of planned and

completed exercises via blue and green lights.

Figure 1: Conceptual design illustrations of the ActiMove

prototype.

The second prototype, ActiWait, aims to identify

potential waiting times of a person during an activity

of an electric-powered device at home (e.g., waiting

for the kettle to boil water). A power sensing device

identifies e.g. the use of a kettle (see Figure 2). A

ActiThings Toolkit - Towards Supporting Older Adults’ Adherence to Home-based Physical Exercise Programs by Providing Notiﬁcations

in Opportune Moments

521

nearby placed tablet-based PC then suggests

specific, simple physical exercises such as toe-heels-

stand, repetitive chair rises or balance exercises from

established programs such as HOPE (Clegg et al.,

2014).

Figure 2: Conceptual design illustrations: ActiWait

prototype (pic. 1-4); ActiConverge prototype (pic. 4-6).

The third prototype, ActiConverge, is a variation

of the ActiWait prototype. Its aim is to make

suggestions of the aforementioned simple physical

exercises when a person arives and stays at a

particular place at home (e.g., when entering the

kitchen). Thus, it senses the proximity of a person to

the tablet PC via near-body-worn Bluetooth beacons

(Estimote or KST Particle) (see Figure 2).

4.2 Evaluation Process

To assess the usefulness of our conceptual designs,

we conducted two focus groups with (in sum) five

older adults who were actively participating in a p.a.

intervention study (two females, 3 males, aged

between 70 and 76 years). Each focus group lasted

about 30-40 minutes and consisted of a moderated

discussion. Participants discussed whether they

wanted to get a reminder about doing physical

exercises and, if so, how they would like to get

reminded. They received handouts, which described

and illustrated the conceptual designs of each

prototype (see Figure 1 and 2 for the illustrations).

These conceptual design handouts were used to

stimulate a discussion about the pros and cons of

each prototype, and to identify basic objections, as

well as further ideas and wishes.

We noted the verbal expressions of the

participants during the focus group in a protocol. We

then categorized and summarized the information

from these protocols in an explorative manner.

4.3 Evaluation Results

The focus group participants expected the prototypes

to do more than only providing a generic reminder

for physical exercises. They wanted the prototypes

to provide rewards for conducting the exercises,

make the benefit of the physical exercise clear to the

user, monitor the completion of exercises, compare

the number of conducted exercises with a training

schedule and inform the user about deviations from

this schedule. Some study participants appreciated

the use of a tablet PC for the ActiConverge and

ActiWait prototype, because it could have additional

features, such as a possibility to document

conducted or dismissed exercises.

At the same time, the study participants regarded

the ease of use and unobtrusiveness of the

prototypes as highly important. One participant

criticized the ActiMove prototype for the smallness

of its buttons and considered the interaction with the

device as cumbersome. Moreover, ActiMove cannot

be placed on every household item, because the item

gets unhandy with the device on it or it is immersed

in water (e.g., one participant puts the watering can

in water when using it).

Albeit there are many opportune moments for

physical exercises in daily life, there are situations

when being notified by the prototypes is not

appreciated. For instance, one participant does not

want to get a reminder for a physical exercise when

she has visitors around at home. Also our idea to

sense power use in order to identify possible waiting

times with physical inactivity is not always

applicable. Participants explained that they did

additional activities while they were waiting for an

electrical device to complete a task (such as boiling

water with a kettle). Due to the heterogeneity of

people’s daily activities and habits, identifying

suitable opportune moments for reminding about

physical exercises is a highly complex task.

4.4 Lessons Learned

Our focus group discussions showed that our

prototypes should provide additional, motivational

features, not only a notification to physical activity,

to be of value for adhering to p.a. interventions.

Further, the prototypes should not require the users

to perform many interactions with the interface

(pressing buttons, adjusting settings) during their

daily life use. The focus groups also showed that the

prototypes should tailor the notification much more

towards individual needs of users. Based on the

focus group discussion, we thus extended our list of

HEALTHINF 2018 - 11th International Conference on Health Informatics

522

possible opportune moments for a short physical

exercise, i.e.:

▪ After sitting or lying for one or more hours

on a chair / couch (because of watching tv,

reading a book or newspaper)

▪ When entering a room, e.g., the kitchen or

the bed room

▪ While doing household chores, e.g.,

washing dishes, brushing teeth, cooking

▪ During waiting times, e.g., while the kettle

is boiling water, the coffee machine is

making coffee

▪ In the morning, after wake-up, when still

lying in bed

5 SECOND STAGE:

FUNCTIONAL PROTOTYPES

Based on our findings from user evaluation, we

refined, as well as altered our conceptual designs

and developed functional prototypes. Due to several

critical issues with usability and use constraints

which came up during the focus groups, we did not

further pursue the concepts of sticking a device on

household items (ActiMove prototype) and of using

power sensing to identify waiting times (ActiWait

prototype). While these concepts may be useful for

some older adults in p.a. interventions, we aimed at

identifying concepts which work for various

situations at home, taking the heterogeneity of

people’s daily activities and habits into account. We

therefore focused on extending the ActiConverge

prototype with further motivational features and user

input devices. Moreover, we explored another

concept which focuses on providing unobtrusive

reminders during sedentary behaviour.

5.1 Two Prototypes

We developed two functional prototypes for the

ActiThings toolkit. A new prototype, called ActiSit,

addresses prolonged sedentary behaviour. It collects

sensor data on sitting activity to detect possibly

opportune moments for physical activity. The

relocatable pillow is equipped with a pressure sensor

that identifies a sitting person and measures the time

of sitting (see Figure 3).

Figure 3: ActiSit prototype (pillow, table lamp).

Based on the time passed, a table lamp, which

serves as an ambient light display, is used to

motivate some physical activity (i.e. doing a little

exercise, or at least, stand up for a moment before

sitting down again). The lamp gradually increases its

light intensity and changes its colour from a neutral

white to blue, until the maximum of a predefined

sitting time is reached. The sitting time counter is

resent by simply decreasing the weight on the pillow

(e.g., standing up).

The ActiConverge prototype was refined with

additional features for documenting the adherence to

the physical exercise intervention. The tablet-based

PC then presents exercises, gives guidance on how

to conduct the exercises, allows writing an exercise

diary, and presents badges and achievements.

Moreover, we added two Bluetooth buttons for a

quick logging of (non-)completion of physical

exercises, which can be placed anywhere according

to user preferences (see Figure 4).

Figure 4: ActiConverge prototype (tablet-based PC with

exercise program, Bluetooth beacon, Bluetooth buttons).

5.2 Evaluation Process

In the second stage, we conducted a user evaluation

with five older adults (3 males, 2 females, aged

between 71 and 80 years), who had indicated their

interest in participating in a particular p.a.

intervention (people on a physical activity

intervention waiting list). The evaluation took place

in an apartment-like lab, which allows testing the

prototypes under more naturalistic conditions than in

an artificial lab environment.

ActiThings Toolkit - Towards Supporting Older Adults’ Adherence to Home-based Physical Exercise Programs by Providing Notiﬁcations

in Opportune Moments

523

The evaluation session with each participant

lasted approximately one hour and consisted of three

parts. First, a pre-interview (5 to 10 minutes) about

the participants’ living situation (e.g., time spent at

home/outside), activity habits in daily life,

previously used reminder tools, and daily ICT use

was carried out. In the second part, the participants

were introduced to the prototypes, their purpose and

functionality. We asked them to try out the

prototypes and verbalize their experiences and first

impressions with the prototypes. As a starting point,

we also offered several examples for possible

opportune moments for a short physical exercise

(see section 4.4) and asked the participants to reflect

on their appropriateness for their own daily life

and/or to come up with further descriptions of

opportune moments at home for a short physical

exercise.

After this hands-on demo part, we conducted a

short, semi-structured, focused interview with the

participants, which aimed at discussing the

experiences and perceptions of the participants in

great detail. The interview was structured around

topics. These topics focused the interview towards

the research interests, and they acted as a guiding

trigger for the study participants. During the

interview the prototypes were also available for use.

We discussed the prototypes with regard to overall /

first impressions, positive and negative issues,

usefulness, missing or undesired features, kind of

information provided to user, design, and possible

usage problems in social situations at home.

We noted all verbal expressions of the

participant during the evaluation session in a

protocol. These protocols were then used for

categorizing and summarizing the information in an

explorative manner.

5.3 Evaluation Results

All participants saw several opportunities for a

reminder for physical exercise within or after

prolonged periods of physical inactivity (sitting or

lying when watching TV, listening to the radio,

using the PC/laptop). Three participants also spotted

opportune moments right after periods of sleeping in

order to get some stimulation. In some cases, an

interruption needs to be postponed after finishing the

primary activity, such as not being interrupted

during the after-lunch nap but after, and not when

visitors are around. Two participants would also like

to get reminded during times of physical activity,

such as doing household chores.

Overall, participants experienced both prototypes

(ActiSit and ActiConverge) as useful. While

ActiConverge devices like the tablet PC were

considered as easy to use and understand by all

participants, three participants mentioned difficulties

to grasp the purpose and functionality of novel

devices like the beacon and the smart button.

Regarding the reminders for physical exercise by

the ActiSit and the ActiConverge prototype,

participants requested an additional acoustic

modality for getting notified. Moreover, three

participants would like to get reminders on their

own, personal devices, such as smartphones or e-

readers. First off, because they are already

accustomed to their mobile devices. Secondly, a

device used along with sedentary activities could

itself be used for receiving a reminder, for instance,

getting a notification on the e-reader after reading

for a very long time.

Two participants stressed the need for a temporal

and spatial flexibility of the provision of reminders.

They want to be reminded on different places within

the home and, moreover, the reminder system should

adapt to the personal daily routine. As a solution,

participants suggested to have a multitude of devices

of the same kind placed at different spots in the

home or to use a wearable device for feedback,

because they would not carry the tablet or beacon

around at home. One participant would prefer to

have every physical activity measured by a wrist-

worn device (like a standard fitness tracker).

Moreover, one participant’s preference for a

reminder device depends on environmental

characteristics of a situation, such as getting

reminded by the light of the lamp in evenings when

sitting in a comfy chair (fit to situation). Two study

participants would put devices for reminding (like

the tablet PC of ActiConverge) in a central or highly

frequented place in the home, e.g., on a side table in

the corridor or mounted on the wall. The device then

reminds the person when passing by.

Three participants expressed their preference for

the ActiConverge prototype, especially because they

appreciated the suggestion of particular exercises on

the tablet PC when getting a reminder. Two

participants missed a social communication feature

of the prototypes, in order to get animated by friends

to do some physical exercise.

5.4 Lessons Learned

The user evaluation of the two functional prototypes

has shown that older adults see opportunities for a

reminder to physical exercise also during daily life

HEALTHINF 2018 - 11th International Conference on Health Informatics

524

situations of sedentary behaviour. Moreover, this

evaluation indicated again the importance of

tailoring the provision of reminders and the

collection of sensing and feedback technologies to

the individual needs and preferences of older adults.

The devices of the ActiThings toolkit need to be able

to both sense behaviour and provide reminders at

many different locations at home. Thereby,

particular sensing and notification devices seem to

fit particular situations at home. As McGee-Lennon

et al. (2011) note, such a contextualisation (right

location, time and device) is crucial for successful

reminder strategies. Moreover, our prototypes

should provide additional feedback modalities, e.g.,

acoustic notifications.

Our study results indicate that the toolkit should

be extended with accustomed, personal smart

devices (e.g., desktop PC, e-reader) for tracking their

time of use (and consequently tracking sedentary

behaviour related to device use) and for providing

prompts to do a physical exercise. Familiar and easy

to use devices like tablet PCs and smartphones are

useful as basic interaction tools for providing the

exercise program and deploying further motivational

strategies (e.g., social interactions among people).

This is in line with a study by Young, Willis,

Cameron, and Geana (2013) on adoption barriers of

home-based health information technologies among

older adults. They found that people with a rather

sceptical attitude prefer devices they are already

familiar with and thus new health information

systems should build on these preferences.

Consequently, the ActiThings toolkit should also

include “low-threshold” devices to mitigate possible

adoption barriers within our addressed age group.

Moreover, tablet PCs work well as ambient

reminder displays, when placed at central, highly

frequented places at home. Bluetooth buttons should

be offered as additional interaction possibilities if

desired.

Based on our user evaluation, we need to rethink

the use of Estimote or KST Particle Bluetooth

beacons as portable devices. Our study findings

suggest the use of wearable devices for tracking

active and sedentary times, as well as the person’s

proximity to an ambient display during the day.

Wearable devices can as well be used for providing

prompts to do a physical exercise. As shown by

research on wearable devices for older adults, smart

watches offer useful notification services and

activity tracker features, but these devices have the

limitation of some people not wearing it every day

or in every situation (Fernández-Ardèvol and

Rosales, 2017). The same applies to head-mounted

displays, which were found to be useful to provide

reminders to older adults (Kunze, Henze, and Kise,

2014). However, there might be use constraints for

such devices during some of our identified

notification opportunities, such as periods of

sleeping.

6 CONCLUSIONS

The effectiveness of home-based physical exercise

programs to prevent frailty among older adults is

very much dependent on people’s long-term

commitment to physical exercise. Within our

research we aim at supporting the adherence to

exercise programs by using ubiquitous, sensor-

based, and interactive technologies for prompting

older adults in opportune moments in daily life to do

physical exercises. Based on our research findings

so far, we see a high potential for our approach to

integrate physical activity in daily life. The

ActiThings toolkit prototypes enable environmental

prompts to physical exercise in a variety of

situations at home, taking individual needs and

preferences of older adults into account. As for

future work, we plan to conduct a field study to

assess the effectiveness of the ActiThings toolkit to

enhance older adults’ adherence to a physical

exercise program.

ACKNOWLEDGEMENTS

This research is funded by the German Federal

Ministry of Education and Research (AEQUIPA,

Project No. 01EL1422C).

REFERENCES

Baez, M., Far, I. K., Ibarra, F., Ferron, M., Didino, D., and

Casati, F. (2017). Effects of online group exercises for

older adults on physical, psychological and social

wellbeing: a randomized pilot trial. PeerJ, 5, e3150.

Brawley, L. R., Rejeski, W. J., and King, A. C. (2003).

Promoting physical activity for older adults: the

challenges for changing behavior. American journal of

preventive medicine, 25(3), 172-183.

Chao, D., Foy, C. G., and Farmer, D. (2000). Exercise

adherence among older adults: challenges and

strategies. Controlled clinical trials, 21(5), 212-217.

Clegg, A., Barber, S., Young, J., Iliffe, S., and Forster, A.

(2014). The Home-based Older People's Exercise

(HOPE) trial: a pilot randomised controlled trial of a

ActiThings Toolkit - Towards Supporting Older Adults’ Adherence to Home-based Physical Exercise Programs by Providing Notiﬁcations

in Opportune Moments

525

home-based exercise intervention for older people

with frailty. Age and ageing, 43(5), 687-695.

Fan, C., Forlizzi, J., and Dey, A. (2012). Considerations

for technology that support physical activity by older

adults. In Proceedings of the 14th international ACM

SIGACCESS conference on Computers and

accessibility, 33-40.

Fernández-Ardèvol, M., and Rosales, A. (2017). My

interests, my activities: Learning from an

intergenerational comparison of smartwatch use. In

International Conference on Human Aspects of IT for

the Aged Population, 114-129.

Intille, S. S. (2004). Ubiquitous computing technology for

just-in-time motivation of behavior change. Medinfo,

107, 1434-1437.

Irvine, A. B., Gelatt, V. A., Seeley, J. R., Macfarlane, P.,

and Gau, J. M. (2013). Web-based intervention to

promote physical activity by sedentary older adults:

randomized controlled trial. Journal of Medical

Internet Research, 15(2).

King, A. C., Taylor, C. B., Haskell, W. L., and Debusk, R.

F. (1988). Strategies for increasing early adherence to

and long-term maintenance of home-based exercise

training in healthy middle-aged men and women. The

American journal of cardiology, 61(8), 628-632.

Kunze, K., Henze, N., and Kise, K. (2014). Wearable

computing for older adults: initial insights into head-

mounted display usage. In Proceedings of the 2014

ACM International Joint Conference on Pervasive and

Ubiquitous Computing: Adjunct Publication, 83-86.

Marcus, B. H., Bock, B. C., Pinto, B. M., Forsyth, L. A.

H., Roberts, M. B., and Traficante, R. M. (1998).

Efficacy of an individualized, motivationally-tailored

physical activity intervention. Annals of behavioral

medicine, 20(3), 174-180.

McGee-Lennon, M. R., Wolters, M. K., and Brewster, S.

(2011). User-centred multimodal reminders for

assistive living. In Proc. of the SIGCHI Conference on

Human Factors in Computing Systems, 2105-2114.

Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S.

U., Bishop, A., and French, D. P. (2011). A refined

taxonomy of behaviour change techniques to help

people change their physical activity and healthy

eating behaviours: the CALO-RE taxonomy.

Psychology & health, 26(11), 1479-1498.

Opdenacker, J., Boen, F., Coorevits, N., and Delecluse, C.

(2008). Effectiveness of a lifestyle intervention and a

structured exercise intervention in older adults.

Preventive medicine, 46(6), 518-524.

Pärkkä, J., Ermes, M., Korpipaa, P., Mantyjarvi, J.,

Peltola, J., and Korhonen, I. (2006). Activity

classification using realistic data from wearable

sensors. IEEE Transactions on information technology

in biomedicine, 10(1), 119-128.

Picorelli, A. M. A., Pereira, L. S. M., Pereira, D. S.,

Felício, D., and Sherrington, C. (2014). Adherence to

exercise programs for older people is influenced by

program characteristics and personal factors: a

systematic review. Journal of physiotherapy, 60(3),

151-156.

Robison, J. and Rogers, M. A. (1994). Adherence to

exercise programmes. Sports Medicine, 17(1), 39-52.

Tapia, E. M., Intille, S. S., and Larson, K. (2004). Activity

recognition in the home using simple and ubiquitous

sensors. Pervasive, Vol. 4, 158-175.

van het Reve, E., Silveira, P., Daniel, F., Casati, F., and de

Bruin, E. D. (2014). Tablet-based strength-balance

training to motivate and improve adherence to exercise

in independently living older people: part 2 of a phase

II preclinical exploratory trial. Journal of medical

Internet research, 16(6).

Young, R., Willis, E., Cameron, G., and Geana, M.

(2014). “Willing but Unwilling”: Attitudinal barriers

to adoption of home-based health information

technology among older adults. Health informatics

journal, 20(2), 127-135.

HEALTHINF 2018 - 11th International Conference on Health Informatics

526