MyHealthFrame

Design and Evaluation of a Minimally Invasive Communication Platform for

Telemedicine Services Aimed at Older Adults

Mohammad Hossein Nassabi

, Harm op den Akker

1,2

, Marian Bittner

, Coen Kemerink

Bert-Jan van Beijnum

, Hermie Hermens

1,2

and Miriam Vollenbroek

1,2

Telemedicine Group, University of Twente, Enschede, The Netherlands

Telemedicine Group, Roessingh Research and Development, Enschede, The Netherlands

Keywords:

Telemedicine, Persuasive Technology, Adherence, Physical Activity.

Abstract:

MyHealthFrame is a communication platform that telemedicine (and well-being) services can leverage to

deliver motivational messages and notiﬁcations to their end-users. Instead of being a telemedicine service in

itself, MyHealthFrame is a channel through which external services can reach their users to provide reminders

or deliver simple information such as number of steps. To its end-users, MyHealthFrame is a tablet device

which is designed to be perceived as a photoframe and can be immersed in the users’ living environment. In

this paper, we describe the design and the preliminary assessment of the platform. The results of the feasibility

study with ﬁve older adults (65+) are promising.

1 INTRODUCTION

Patient participation is essential to successfully move

care from hospitals to homes. However, the uptake

and adherence to telemedicine systems targeting the

home environment remains poor (McGee-Lennon and

Brewster, 2011). To address this gap, special attention

must be paid to the characteristics of the end-users.

Especially in the case of older adults, not only should

researchers consider the cognitive and physical im-

pairments, but technological literacy and ease of use

also play signiﬁcant roles in accepting a home-based

system.

Some telemedicine systems support the elderly

in establishing a healthy behavior in order to re-

duce future health complications. Researchers have

been using persuasive technology to promote such

changes to daily living. Virtual activity coaches are

the prime example of systems in which techniques

such as goal-setting, increasing self-awareness and re-

inforcing proper attitudes are used to increase levels

of physical activity. Although the older adults’ atti-

tude towards technology is not uniform, it has been re-

ported that they have greater fear and anxiety of using

computers comparing to other age groups (Barnard

et al., 2013). This can lead to the elderly not being

keen on using technology. In addition, some barriers

can further demotivate the older adult end-user from

accessing the health service. Consider the example of

a web-based rehabilitation service: The elderly has to

turn on his computer, log into the Operating System

(OS), open up the browser and log into the website to

access the rehabilitation service. In a more dramatic

scenario, changes to the elderly’s computer setting

(e.g., software updates) can result in the user becom-

ing discouraged and not using the service at all. This

example can become more complicated if the elderly

has to use more telemedicine systems (and well-being

services) to address the health needs in different do-

mains. Each of these services may also use persuasive

technology to motivate their end-users. However, the

end-users will never receive the persuasive elements

if they stop referring to the corresponding websites or

devices.

The MyHealthFrame platform is designed to facil-

itate long-term communication between telemedicine

systems and end-users who will beneﬁt from the ser-

vices offered by such systems. Therefore, MyHealth-

Frame in itself is not a telemedicine system. How-

ever, it is an always-available communication chan-

nel through which a simpliﬁed subset of function-

ality from a telemedicine system is delivered in the

home environment. As such, it can be viewed as a

proxy to the older adults’ daily living. Moreover,

322

Nassabi, M., Akker, H., Bittner, M., Kemerink, C., Beijnum, B-J., Hermens, H. and Vollenbroek, M.

MyHealthFrame - Design and Evaluation of a Minimally Invasive Communication Platform for Telemedicine Services Aimed at Older Adults.

DOI: 10.5220/0005705703220329

In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2016) - Volume 5: HEALTHINF, pages 322-329

ISBN: 978-989-758-170-0

the platform can be employed as a hub for collect-

ing sensor data and passing this data to suitable ex-

ternal services. MyHealthFrame, with its ubiquitous

front-end, attempts to overcome barriers to initiating

service use by: (1) Reminding the user about the ex-

istence of the external services by delivering the per-

suasive elements generated by them and (2) Present-

ing simple information originating from the external

services. Consequently, MyHealthFrame has the po-

tential to yield higher user engagement and possibly

long-term adherence to the offered services.

In this paper, we discuss the design, develop-

ment and preliminary assessment of MyHealthFrame.

The assessment will be done for services targeting

the physical domain with which the older adults can

achieve a physically active lifestyle.

2 BACKGROUND

To the best knowledge of the authors, no ICT sys-

tem similar to MyHealthFrame—as a channel to de-

liver persuasive elements from external health ser-

vices to older adult end-users—exists. The type of in-

formation delivered via MyHealthFrame is mainly re-

lated to health persuasion. Therefore, in this section,

we brieﬂy describe persuasive technology and sub-

sequently present several systems that promote self-

management and harness persuasive elements to es-

tablish healthy behavior change. These systems have

been evaluated by either patients or healthy subjects.

Our main interest here is to explore the opportunities

for the delivery of persuasive content to older-adult

end-users.

2.1 Persuasive Technology

Persuasive technology is deﬁned as the study of inter-

active systems designed to deliver attitude or behav-

ior change. Studies regarding the technology have al-

ready been performed in various domains such as ed-

ucation, marketing, safety, entertainment and health,

where the latter has attracted most contributions.

(Oinas-Kukkonen and Harjumaa, 2009) present

a Persuasive System Design (PSD) model in which

persuasive techniques are classiﬁed into four cate-

gories: (1) Primary Task Support, which are tech-

niques that help users perform the main activity and

achieve their goals; examples include personaliza-

tion and self-monitoring; (2) Dialogue Support, ad-

dressing computer-human interaction bearing in mind

similarities to human interactions such as reminders,

praise, feedback and rewards; (3) Credibility Support

to promote the user’s trust in the system for exam-

ple, veriﬁability and trustworthiness; and ﬁnally (4)

Social Support Techniques, leveraging the social fac-

tors to increase system’s persuasiveness by compe-

tition, recognition and normative inﬂuence (namely,

peer pressure).

2.2 Self-management Promoting

Systems

A well-known category of persuasive systems are vir-

tual activity coaches in which a pedometer or an ac-

celerometer sensor is used to measure physical ac-

tivity so that the intervention can be adjusted ac-

cordingly. For example, in Fish’n’steps (Lin et al.,

2006), a virtual pet is presented on a ﬁxed LCD dis-

play whose life state is dependent on the user’s ac-

tivity. In Ubiﬁt Garden (Consolvo et al., 2008), the

authors use a PDA mobile device to depict a garden

metaphor in which the number of ﬂowers increases

as a result of achieving activity goals. In Flowie (Al-

baina et al., 2009), a ﬂower metaphor representing the

end-user’s physical performance is provided through

a photoframe device. Extra user interfaces have also

been introduced for goal-setting. The authors evalu-

ated the prototype in a panel consisting of two older

adults (aged 65+). The results showed that the panel

had a positive attitude toward adopting the system.

Some systems have used the TV as their ser-

vice delivery medium. In (Giordano et al., 2009), the

Philips Motiva system was connected to patients’ TV

through which the educational videos were shown.

The authors in (Stojmenova et al., 2013) used an in-

teractive TV to increase adherence to medication by

sending reminders. However, an intrinsic problem

with systems employing TV as their medium is the

communication loss when the TV is off.

Beside the more conventional modalities, re-

searchers have also used robots to provide health ser-

vices. For example, in (Johnson et al., 2014), a so-

cially assistive robot was integrated into a smart home

to support the elderly’s independent living. However,

the long-term acceptance of such robots has scarcely

been studied (de Graaf et al., 2015).

The above-mentioned studies show the range of

different options that can be chosen as delivery de-

vice in telemedicine systems. We can consider multi-

ple criteria to compare various options and choose the

most suitable device for a given application. Portabil-

ity, Affordability, Availability, Information Richness

(op den Akker et al., 2014), Computation Power, Re-

quired Infrastructure, Learning Curve and Interaction

Type can be named as some criteria for consideration.

MyHealthFrame - Design and Evaluation of a Minimally Invasive Communication Platform for Telemedicine Services Aimed at Older

Adults

323

3 CONCEPT OVERVIEW

Our aim in this research was to design an afford-

able interaction device that could be immersed in the

home. This led on from previous research indicating

that the homes of the elderly can be an ideal loca-

tion for providing persuasive elements (Cabrita et al.,

2015). We ruled out mobile devices as we could not

ﬁnd strong evidence of elderly preference for such de-

vices in home-based systems (McGee-Lennon et al.,

2012). Moreover, we selected a photoframe represen-

tation for the end-user device as it seems to easily be-

come a natural part of the existing living environment.

This kind of end-user device has already been inves-

tigated with older adults (or with their caregivers) in

(Mynatt et al., 2001), (Dadlani et al., 2010) and (Al-

baina et al., 2009).

MyHealthFrame provides a single point of refer-

ence for the older adult end-users via which motiva-

tional messages, reminders and feedback are commu-

nicated.

Fitbit server

MyHealthFrame

Server

Reminders and

motivational

content

Reminders and

motivational

content

Physical

activity data

GNPT

cognitive trainer

Reminders,

feedback,

images,

notifications

MyHealthFrame

Tablet

Pre-frail elderly

CoCo

physical trainer

Figure 1: MyHealthFrame presents the motivational mes-

sages to the older adult end-user.

To its end-users, MyHealthFrame is a tablet de-

vice, designed to be perceived as a digital photoframe.

The tablet requires minimal explicit user interaction

and is placed at a location that is easily accessible

by the elderly end-user. When idle, it can show im-

ages of family members or delightful scenery to im-

prove the the elderly’s mood and mental state. Exter-

nal telemedicine systems can trigger the tablet to con-

vey persuasive elements to the elderly. In addition, the

tablet can transmit sensor data to the relevant external

systems.

The ﬁrst major exploitation of the platform will

be in the PERSSILAA project in which multiple

electronic health services in the domains of phys-

ical, cognitive and healthy nutrition are offered to

pre-frail community dwelling older adults. Partici-

pants in PERSSILAA use their personal computers

to access electronic health services (some of which

are shown in Figure 1) to train and thereby improve

their health. For example, PERSSILAA users follow a

12-week physical exercise program on the Condition

Coach (CoCo) system: a web-based physical trainer

allowing the elderly to perform exercises in their

home setting (Tabak et al., 2014). Moreover, cognitive

exercises are provided through the Guttmann Neurop-

ersonal Trainer (GNPT) which is a Java-based desk-

top application (Solana et al., 2015). The users are

also provided with Fitbit sensors to track their physi-

cal activity.

4 REQUIREMENT ELICITATION

The main functional requirements considered for the

MyHealthFrame platform are listed below. Not all

the functional requirements have been implemented

in the ﬁrst prototype.

1. An external telemedicine (or well-being) service

can register itself with the platform.

2. The end-users can register themselves with the

platform by providing the credentials for the ex-

ternal services.

3. An external service can send motivational items

via the platform to registered end-users.

4. The platform queues messages based on their pri-

ority.

5. The end-user tablet ﬁlters the received messages

based on their validity period.

6. The platform collects sensor data and sends it to

the corresponding services.

7. The platform can update parts of the user interface

in the tablet to present simple information such as

step count originating from the external services.

We took a user-centered approach while design-

ing the ﬁrst prototype of the system for requirements

related to end-users. Initially, we presented the con-

cept and ideas regarding to MyHealthFrame to a focus

group consisting of four older adults aged between

65 and 75. Subsequently, the participants were asked

several questions about the concept and the features

they would like to see included in the prototype. All

participants mentioned that they would place such a

solution in the living room for example, next to the

TV, and some mentioned that they would favor the in-

clusion of a clock and an overview of their physical

activity. The focus group was also asked to give their

preferences for the activity sensor and all stated that

they would like to use it to record their daily move-

ment level.

HEALTHINF 2016 - 9th International Conference on Health Informatics

324

5 PROTOTYPE DESIGN AND

IMPLEMENTATION

The ﬁrst prototype was built to assess the feasibility

and potential uptake of the solution by older adults.

5.1 System Architecture

The architecture used in the prototype is shown in

Figure 2. The user can interact with external ser-

vices via their proprietary browser-based applica-

tions. The Java programming language is used to de-

velop the required JSON-RPC 2.0 web-services ex-

posed by MyHealthFrame server. Since the CoCo

physical trainer’s functionality to generate motiva-

tional messages is under development, we created a

back-end control panel (i.e., caregiver user interface)

that allows an experiment conductor to inject motiva-

tional items into the system.

Figure 2: The push mechanism is used to communicate per-

suasive elements to the MyHealthFrame tablet.

In the prototype implementation, we have used

the Google Cloud Messaging (GCM) service to push

items to the tablet. The decision was made to accel-

erate the development process. Moreover, the tablet

is set to pull sensor data (i.e., step count) every 15

minutes using the Bluetooth protocol. However, the

successful synchronization of the accelerometer data

depends on: (1) the proximity of the sensor and the

tablet; and (2) the sensor not being idle. The classes

in the back-end that are deﬁned for communicating

with the tablet are shown in Figure 3. Despite consid-

ering a single device per user for our ﬁrst prototype,

the software architecture allows for multiple user de-

vices.

Each message sent to the tablet is deﬁned as an

ActionMessage object containing a pair of action type

(e.g., image) and content (i.e.,parameters). After ver-

+getAge()

+getEmail()

+getGender()

+setAge()

+setEmail()

+setGender()

-age

-gender

-email

Person

+getReceivers()

+sendAction()

-receivers : Receiver

Sender

+executeActions()

+setDevices()

+addAction()

-devices : IDevice

-actionMessages : ActionMessage

Receiver

+handle()

-externService : IExternService

-successor : ActionMessage

-targetDevice : IDevice

-requestedAction : string

-requestedContent : string

ActionMessage

DialogActionHandlerTextActionHandler MarqueeActionHandler

+addContent()

+sendAllActions()

IExternService

-successor

0..1

-sender1

+isActionAllowed() : bool

+getId()

+setId()

-Id

IDevice

AndroidDevice

GoogleCloud

0..*

+sendAction()

HFService

AlertActionHandlerImageActionHandler

Figure 3: Extensibility and multiple end-user devices were

considered when designing the system.

Table 1: The various types of messages that can be sent to

the MyHealthFrame tablet.

Type Parameters

Image imageAddress, playMusic?

Text messageText, position

fontSize, fontType

fore/background colour

MarqueeText Same as a text message plus

numberOfRepititions

Dialog messageText, dialogDuration

Voice messageText, voiceGender

voiceSpeed

GetDisplayedData No Parameters

ResetFrame No Parameters

ifying that the end-user’s device is capable of exe-

cuting the requested ActionMessage, the message is

transmitted via the GCM service.

5.2 Communicated Items

A communicated item can be an image, a textual item,

a marquee text, a voice message and a dialog item. In

case of the later, a user dialog is shown on the tablet

that requires the user to press a button.

Each item can be parametrized according to its

type, thereby allowing for some levels of customiza-

tion. For example, if a text message is sent, beside

the content, the text location, size and color can be

adjusted. As another example, each image can be ac-

companied by a random music sample 10 seconds

long.

Table 1 lists the message types implemented in the

ﬁrst prototype. The last two types were added to help

the experiment conductor to verify that the tablet is

always in a consistent state.

MyHealthFrame - Design and Evaluation of a Minimally Invasive Communication Platform for Telemedicine Services Aimed at Older

Adults

325

Figure 4: Message parameters can be set via the caregiver

user interface.

5.3 User Interfaces

Two user interfaces (UI) were designed for the ﬁrst

prototype. The caregiver UI shown in Figure 4 is a

secure website that the authenticated study conductor

can use to send items to the elderly’s tablet. Following

the successful transmission, a notiﬁcation is provided.

The care receiver UI is the interface designed for

older adults. This interface is integrated in an Android

application. The application cannot be closed unless

the tablet is out of power.

Figure 5: The implemented Android app on MyHealth-

Frame tablet is always running and cannot be closed.

Textual messages can be rendered in either of the

four boxes shown in Figure 5. The strong contrast

between the white number and the black background

makes it easy to read the time. The step counter in the

lower right corner shows the latest received number

of steps for the Fitbit sensor. The shoe color changes

from yellow to green if the step count is higher than a

pre-deﬁned threshold.

6 EVALUATION

We investigated the feasibility of MyHealthFrame

in a small exploratory study. The only external

telemedicine system based on which motivational

content was communicated was the browser-based

CoCo physical exercise trainer. This system provides

a set of exercises in the form of narrated videos ac-

companied by textual descriptions. Each exercise set

consists of a warm-up, main phase and cool-down

stage containing four, nine and four exercises respec-

tively. The participants had to use their own comput-

ers to access the physical trainer. Step counts were

collected using the Fitbit Zip sensor given to each par-

ticipant.

CoCo’s motivational message-generator service is

currently under development. Therefore, this study

was done as a Wizard of Oz experiment: the partic-

ipants were told that an intelligent system communi-

cates with them through the tablet, although in real-

ity, a nursing student was sending messages based on

participant’s activity levels and their progress in the

physical trainer.

6.1 Study Protocol

The goal of the study was to evaluate the feasibility

of MyHealthframe with a group of older adults and to

collect their feedback about the system. The study du-

ration was three weeks and consisted of two phases.

The data collected during the initial phase (ﬁrst two-

weeks) was used as a baseline for sending persuasive

elements in the the second phase (last week) of the

study. During the registration meeting, the research

conductor explained the goal of the research and col-

lected the signed informed consent form. Moreover,

the participants ﬁlled in a short questionnaire about

their physical activity levels and were given access

to the CoCo physical trainer. The participants were

asked to perform the exercises on the trainer three

times per week using their notebook computers. The

Fitbit sensors were given to participants in the same

day. In the third week, the participants were provided

with the tablets for a one week period.

At the end of the study, the nursing student held

an interview with the participants asking them to

ﬁll in the Computer System Usability Questionnaire

(CSUQ) (Lewis, 1995) containing 19 items on over-

all system usability (OVERALL), system usefulness

(SYSUSE), information quality (INFOQUAL) and

interface equality (INTERQUAL). The items can be

answered using 7-point Likert scales, ranging from

”Strongly agree” with one point to ”Strongly dis-

agree” with seven points, and an additional Not Ap-

plicable (N/A) point. Moreover, to collect further in-

formation regarding user acceptance, the six extra

questions listed below were asked during the inter-

view of which some focused on the feedback given

and future potential improvements to the system.

1. How was your experience with the system and

HEALTHINF 2016 - 9th International Conference on Health Informatics

326

the provided feedback?

2. What did you learn from the given feedback?

3. What could be improved when giving feedback?

4. Have you beneﬁted from the feedback or

reminders? Please elaborate on your answer.

5. What would you like to change in the system?

6. Would you recommend MyHealthFrame to a

friend? Please elaborate on your answer.

6.2 Participants

Five participants aged between 65-68 were recruited

for the pilot consisting of two couples (Subjects 1 & 2

and Subjects 3 & 4) and a single male subject (Sub-

ject 5). Only four tablets were distributed amongst

the participants resulting in Subjects 1 & 2 receiving

a single tablet.

All of the participants mentioned that they cycled

every day. Three participants mentioned exercising at

a gym or a sport center for a minimum of once per

week e.g., playing tennis. All participants were able

to do daily activities such as shopping and performing

household tasks.

6.3 Communication Timing and

Content

Motivational messages were sent to the tablet both at

ﬁxed and time-varying moments. Based on the time

of the day (morning, noon, evening), three generic im-

ages accompanied by a text were communicated to the

participants e.g., the “Good Morning” message was

sent at 9:00 am. In addition, after performing six (out

of nine) exercises on the CoCo system, positive feed-

back in the form of an image would be sent. More-

over, a motivational message was sent if the number

of steps was more than the average step count for a

speciﬁc person (based on two weeks collected data).

If the participant had not been exercising on the

physical trainer and the step count was lower than

5000 steps on a given day, a reminder was sent to the

user to perform physical activity e.g., a textual mes-

sage containing “It is a great day to exercise”.

6.4 Results

The number of steps measured during the study is

shown in Figure 6. The average number of steps in

some cases (e.g., subject 1) is signiﬁcantly higher than

the recommended 7000-8000 steps for the 65+ age

group. The step counts supports the ﬁnding that the

studied participants, other than subject 5, are more ac-

tive than the general population.

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

Subject1 Subject2 Subject3 Subject4 Subject5

Average number of daily steps

With

System

Without System

Figure 6: The average number of daily steps collected from

participants in during the study.

Table 2: Four scores are calculated based on the answers to

the CSUQ usability questionnaire.

Based Average Missing

Score on questions responses value ratio

OVERALL 1 to 19 3.04 0.31

SYSUSE 1 to 8 2.74 0.11

INFOQUAL 9 to 15 3.5 0.56

INTERQUAL 16 to 18 4 0.4

Only one couple (Subject 1 and Subject 2) regu-

larly exercised using the CoCo physical trainer sys-

tem. The other users stated that they were already

physically active which was conﬁrmed by sensor data.

The participants could not provide answers to all

19 questions of the CSUQ questionnaire. Following

the guidelines of (Lewis, 1995), the questionnaire can

be summarized into the four scores listed in Table 2.

This table also contains the ratio for missing values

per score deﬁned using Equation 1:

Missing value ratio=

∑

N/A Answers

#Questions ×#Participants

(1)

Unfortunately questions regarding information

and interface quality seemed ambiguous or not appli-

cable to most participants. However, the SYSUSE

score suggested that the participants were satisﬁed

with the system.

All participants were satisﬁed with using the Fit-

bit Zip sensor; indeed, Subject 5 mentioned that he

became more active because of the sensor. Subjects

3 and 4 also stated that presenting the step counts on

the tablet motivated them to compete with each other.

Subjects 1, 2 and 3 mentioned that they liked the

received feedback shown on the tablet. The most ac-

tive participant, Subject 4, did not notice any feed-

back as she was mainly outside doing voluntary work.

Subject 5, the least active participant, could not under-

stand how MyHealthFrame could help him.

MyHealthFrame - Design and Evaluation of a Minimally Invasive Communication Platform for Telemedicine Services Aimed at Older

Adults

327

7 DISCUSSION AND FUTURE

WORK

The acceptance of a computer system is inﬂuenced

by multiple factors including user satisfaction, system

usability, perceived ease of use and perceived useful-

ness (Acton et al., 2004). In this feasibility study, we

attempted to collect subjective measurements related

to the usability and usefulness of the MyHealthFrame

platform with a limited number of participants. Even

though the qualitative results regarding system useful-

ness are limited, still they indicate a positive attitude

towards accepting the system.

The low value for information and interface qual-

ity and the corresponding high missing value ratios

in Table 2 suggest that better explanations should be

given about the system and its intended role. Al-

though the participants were given a one-page de-

scription of the overall system at the beginning of the

study, their answers to the extra questions about the

system suggested some levels of confusion. For ex-

ample, one participant suggested improvements to the

CoCo physical trainer despite the fact that the ques-

tion concerned MyHealthFrame. Another reason for

such confusion can be the introduction of CoCo and

Fitbit without taking into account the differences in

the learning abilities of the participants. Therefore,

one inclusion criteria for future studies can be the par-

ticipant’s prior acquaintance with web-based systems

and his or her computer literacy.

The most important consideration for future stud-

ies is the inclusion of participants who beneﬁt most

from the platform. All participants in this study were

moderately to highly active in comparison with peo-

ple in the same age group. Consequently, there was

no need for them to follow the online physical exer-

cises. This inevitably led to fewer feedback moments

related to their progress.

An interesting extension to the platform would

be the addition of competition features for older

adult couples. As stated by two participants, such

functionality could have motivated them to become

more physically active. However, the competition fea-

tures would not need to be limited to physical ac-

tivity and could be generalized to make use of any

health/service information. For example, the user

progress in GNPT cognitive trainer can also be con-

sidered as the competition information source. In this

case, a total score could be aggregated based on scores

in each health domain. Inclusion of such features in a

future prototype would require implementing a light

computation engine in the back-end. Moreover, the

user interface would need to be modiﬁed to present

the couple progress in a single tablet.

The calculated total scores can also be used to

implement other game mechanics than (intra-couple)

competition such as leaderboards, challenges, social

networking and awarding points amongst all partici-

pants. For example, a reward scheme can be deﬁned

for end-users motivating them to become more en-

gaged with the external telemedicine systems. The

scheme can reward users in terms of prizes in the form

of electronic badges as done in Fitocracy (Hamari

and Koivisto, 2013) to promote individuals’ achieve-

ments.

Our main focus for next prototypes will be on im-

proving the user interface for participants belonging

to the target group (pre-frail end-users). Speciﬁcally,

we plan to perform a task-oriented user study. Par-

ticipants will be asked to perform a pre-deﬁned set of

tasks using MyHealthFrame tablet and will be subse-

quently interviewed and asked to ﬁll in the User Expe-

rience Questionnaire (Laugwitz et al., 2008). Setting

up the study in this way will allow us to compare and

improve various design elements in the system which

can eventually lead to a more effective solution.

The architecture of the platform allows users to

use multiple Android devices. Consequently, it is

possible not only to send messages to the devices at

home but also to communicate with the mobile de-

vices that the user carries. Therefore, as a future ex-

tension, we would like to assess the effectiveness of

cross-platform electronic health content delivery.

After analyzing the log ﬁles from the study, some

end-user mistakes when sending messages based on

time of the day were detected (made by the experi-

ment conductor). Therefore, to minimize the human

error in future studies, a scheduler component will be

included in the platform that can be used for ﬁxed

time/rule-based messages.

8 CONCLUSION

In this paper, we have presented MyHealthFrame,

a communication platform that third-party electronic

health service providers can exploit to deliver persua-

sive elements such as motivational messages and re-

minders. The platform can also gather sensor data and

send it to the relevant external services.

The preliminary results from the feasibility study

yielded small yet positive indications. In addition, the

study highlighted essential improvement points for

future evaluations such as a more strict inclusion cri-

teria for participants. The focus of the next study will

be on improvements to usability and collecting more

credible evidence about the platform’s effectiveness.

HEALTHINF 2016 - 9th International Conference on Health Informatics

328

ACKNOWLEDGEMENTS

The authors would like to thank the European Com-

mission for providing the funding of this research

through the PERSSILAA FP7 project. Moreover, the

authors thank Sanne Frazer for her contributions to

the study.

REFERENCES

Acton, T., Golden, W., Gudea, S., and Scott, M. (2004).

Usability and acceptance in small-screen information

systems. In Proceedings of 9th European Collabora-

tive Electronic Commerce Technology and Research

Conference.

Albaina, I. M., Visser, T., van der Mast, C. A., and Vas-

tenburg, M. H. (2009). Flowie: A persuasive virtual

coach to motivate elderly individuals to walk. In Per-

vasive Computing Technologies for Healthcare, 2009.

PervasiveHealth 2009. 3rd International Conference

on, pages 1–7. IEEE.

Barnard, Y., Bradley, M. D., Hodgson, F., and Lloyd,

A. D. (2013). Learning to use new technologies by

older adults: Perceived difﬁculties, experimentation

behaviour and usability. Computers in Human Behav-

ior, 29(4):1715–1724.

Cabrita, M., Nassabi, M. H., op den Akker, H., Tabak, M.,

Hermens, H., and Vollenbroek, M. (2015). An unob-

trusive system to monitor physical functioning of the

older adults: Results of a pilot study. In International

Workshop on Personalisation and Adaptation in Tech-

nology for Health.

Consolvo, S., McDonald, D. W., Toscos, T., Chen, M. Y.,

Froehlich, J., Harrison, B., Klasnja, P., LaMarca, A.,

LeGrand, L., Libby, R., et al. (2008). Activity sensing

in the wild: a ﬁeld trial of ubiﬁt garden. In Proceed-

ings of the SIGCHI Conference on Human Factors in

Computing Systems, pages 1797–1806. ACM.

Dadlani, P., Sinitsyn, A., Fontijn, W., and Markopoulos, P.

(2010). Aurama: caregiver awareness for living inde-

pendently with an augmented picture frame display.

Ai & Society, 25(2):233–245.

de Graaf, M. M., Allouch, S. B., and Klamer, T. (2015).

Sharing a life with harvey: Exploring the acceptance

of and relationship-building with a social robot. Com-

puters in human behavior, 43:1–14.

Giordano, A., Scalvini, S., Zanelli, E., Corr

a, U., Longob-

ardi, G., Ricci, V., Baiardi, P., and Glisenti, F. (2009).

Multicenter randomised trial on home-based teleman-

agement to prevent hospital readmission of patients

with chronic heart failure. International journal of

cardiology, 131(2):192–199.

Hamari, J. and Koivisto, J. (2013). Social motivations to use

gamiﬁcation: An empirical study of gamifying exer-

cise. In ECIS, page 105.

Johnson, D. O., Cuijpers, R. H., Juola, J. F., Torta, E., Si-

monov, M., Frisiello, A., Bazzani, M., Yan, W., We-

ber, C., Wermter, S., et al. (2014). Socially assis-

tive robots: A comprehensive approach to extending

independent living. International Journal of Social

Robotics, 6(2):195–211.

Laugwitz, B., Held, T., and Schrepp, M. (2008). Con-

struction and evaluation of a user experience ques-

tionnaire. Springer.

Lewis, J. R. (1995). Ibm computer usability satisfac-

tion questionnaires: psychometric evaluation and in-

structions for use. International Journal of Human-

Computer Interaction, 7(1):57–78.

Lin, J. J., Mamykina, L., Lindtner, S., Delajoux, G., and

Strub, H. B. (2006). Fish’n’steps: Encouraging phys-

ical activity with an interactive computer game. In

Proceedings of the 8th International Conference on

Ubiquitous Computing, UbiComp’06, pages 261–278,

Berlin, Heidelberg. Springer-Verlag.

McGee-Lennon, M., Smeaton, A., and Brewster, S. (2012).

Designing home care reminder systems: lessons

learned through co-design with older users. In Per-

vasive Computing Technologies for Healthcare (Per-

vasiveHealth), 2012 6th International Conference on,

pages 49–56. IEEE.

McGee-Lennon, M. R. and Brewster, S. (2011). Reminders

that make sense: Designing multimodal notiﬁcations

for the home. In Pervasive Computing Technologies

for Healthcare (PervasiveHealth), 2011 5th Interna-

tional Conference on, pages 495–501. IEEE.

Mynatt, E. D., Rowan, J., Craighill, S., and Jacobs, A.

(2001). Digital family portraits: supporting peace of

mind for extended family members. In Proceedings of

the SIGCHI conference on Human factors in comput-

ing systems, pages 333–340. ACM.

Oinas-Kukkonen, H. and Harjumaa, M. (2009). Persuasive

systems design: Key issues, process model, and sys-

tem features. Communications of the Association for

Information Systems, 24(1):28.

op den Akker, H., Jones, V. M., and Hermens, H. J.

(2014). Tailoring real-time physical activity coaching

systems: a literature survey and model. User model-

ing and user-adapted interaction, 24(5):351–392.

Solana, J., C

aceres, C., Garc

ıa-Molina, A., Opisso, E.,

Roig, T., Tormos, J. M., and Gomez, E. J. (2015). Im-

proving brain injury cognitive rehabilitation by per-

sonalized telerehabilitation services: Guttmann neu-

ropersonal trainer. Biomedical and Health Informat-

ics, IEEE Journal of, 19(1):124–131.

Stojmenova, E., Debevc, M., Zebec, L., and Imperl, B.

(2013). Assisted living solutions for the elderly

through interactive tv. Multimedia tools and appli-

cations, 66(1):115–129.

Tabak, M., Brusse-Keizer, M., van der Valk, P., Hermens,

H., and Vollenbroek-Hutten, M. (2014). A telehealth

program for self-management of copd exacerbations

and promotion of an active lifestyle: a pilot random-

ized controlled trial. International journal of chronic

obstructive pulmonary disease, 9:935.

MyHealthFrame - Design and Evaluation of a Minimally Invasive Communication Platform for Telemedicine Services Aimed at Older

Adults

329