Honours Project Report Self-Management of Diabetes using Social Networking

cakeexoticInternet and Web Development

Dec 13, 2013 (3 years and 5 months ago)

256 views



Honours Project Report



Self
-
Management of Diabetes using Social
Networking



Nidheesh Sharma

Supervised by:
Dr
Hanh Le

Co
-
Supervised by:
Professor
Edwin Blake


Department of Computer Science

University of Cape Town

2011


Category

Min

Max

Chosen

1


Requirement Analysis and Design

0

20

20

2


Theoretical Analysis

0

25

0

3


Experiment Design and Execution

0

20

0

4


System Development and Implementation

0

15

5

5


Results, Findings and Conclusion

10

20

20

6


Aim Formulation and Background Work

10

15

15

7


Quality of Report Writing and Presentation

10

10

8


Adherence to Project Proposal and Quality of Deliverables

10

10

9


Overall General Project Evaluation

0

10

0

Total marks

80

80

Department of Computer Science

E
-
Health System


ii


Abstract

Diabetes is a chronic disease which is taking more lives than people actu
ally notice. Currently
diabetics
have limited ways of communicating with each other and seeking quick professional
advice.
Furthermore,
it was established by means of
questio
nnaires and prototype testing

that
there is
a
n

immense

need for an affordable and easily accessible system to
help

diabetics manage
their
condition
.
By means of
a
User
-
Centred Design, t
he E
-
Health
team

aims at developing an
inexpensive
, easily accessible a
nd

a multi
-
featured self
-
managing system to
support

diabetics with the
management of diabetes
, which is implemented on Facebook, mobile phone and a website.

After
identifying

user requirements
,

a final system was developed which was piloted for a period of
five

days.
The
key findings demonstrated that
not only
has
the

Facebook E
-
Health application achieved complete
user satisfaction
,

but
it
also attracted many potential users from all around the

world who continued
to
utiliz
e

the system on a daily basis even after the pilot run

was complete
.
To achieve a
great

number

of
loyal potential users so
swiftly

evidently
exhibit
s

that the Facebook E
-
Health application has the
prospective to expand and
for
m

the core foundation of any future systems providing online support for
patients with chronic illnesses such as diabetes.


Keywords
:

Diabetes, Social Networking, Facebook
, E
-
Health




Department of Computer Science

E
-
Health System


iii


Acknowledgments

I would like to show my ap
preciation to my supervisor,
Dr.

Hanh Le
, and co
-
supervisor, Professor Edwin
Blake, whose encouragement, support and knowledge has been invaluable to me during the course of this
project.


I would also like to thank

Dr. Ian Ross
,
Emily Jane Ryan

and
Professor Mike Lambert

for

their
v
aluable
contribution

to this project
.


Special thanks to the Diabetes South Africa Support Groups and especially Carol Hendriks, Mercia
Roman and Sandra Lewis for their contribution
which enabled the team to develop a thorough

understanding of the
diabetic users
.


Finally, I would like to thank my E
-
Health team members
, Brian Sebastian and Dalton Jacobs,

whose

informal support
,

passion

and hard work
has been indispensable

to the accomplishment of this project
.


Department of Computer Science

E
-
Health System


iv


Table of Contents

Abstract

................................
................................
................................
................................
.........................

ii

Acknowledgments

................................
................................
................................
................................
........

iii

Table of Figures

................................
................................
................................
................................
...........

vi

1.

Introduction

................................
................................
................................
................................
...........

1

1.1.

Problem Outline

................................
................................
................................
............................

1

1.2.

Proposed Solution and Division of Work

................................
................................
.....................

1

1.3.

Eth
ical Consideration

................................
................................
................................
....................

1

1.4.

Report Outline

................................
................................
................................
...............................

2

2.

Background and Motivation
................................
................................
................................
..................

3

2.1.

Introduction

................................
................................
................................
................................
...

3

2.2.

Diabetes
................................
................................
................................
................................
.........

3

2.2.1.

Type 1 Diabetes

................................
................................
................................
........................

4

2.2.2.

Type 2 Diabetes

................................
................................
................................
........................

4

2.3.

Social Networking

................................
................................
................................
........................

4

2.3.1.

Social Networking and

Healthcare
................................
................................
............................

4

2.3.2.

Facebook

................................
................................
................................
................................
...

5

2.3.3.

Opportunities

................................
................................
................................
.............................

7

2.3.4.

Challenges

................................
................................
................................
................................
.

8

2.4.

Related Systems

................................
................................
................................
............................

9

2.4.1.

Diabetes Pilot Software

................................
................................
................................
.............

9

2.4.2.

CureTogether and PatientsLikeMe

................................
................................
.........................

10

2.4.3.

SparkPeople

................................
................................
................................
............................

10

2.5.

Conclusion

................................
................................
................................
................................
..

11

3.

Requirement Analysis Chapter

................................
................................
................................
...........

12

3.1.

Introduction

................................
................................
................................
................................
.

12

3.2.

Understanding the Users and their Needs

................................
................................
...................

12

3.2.1.

Support Groups

................................
................................
................................
.......................

13

3.2.2.

Questionnaires

................................
................................
................................
.........................

15

3.2.3.

Posters

................................
................................
................................
................................
.....

19

3.2.4.

Endocrinologist

................................
................................
................................
.......................

19

3.2.5.

Dietician

................................
................................
................................
................................
..

20

3.2.6.

Pedometers

................................
................................
................................
..............................

21

3.3.

Prototypes

................................
................................
................................
................................
...

21

3.4.

Evaluations

................................
................................
................................
................................
..

22

3.5.

Summary

................................
................................
................................
................................
.....

23

Department of Computer Science

E
-
Health System


v


4.

Low Fidelity and High Fidelity Design Iterations

................................
................................
..............

25

4.1.

Introduction

................................
................................
................................
................................
.

25

4.2.

Low Fidelity Paper Prototype

................................
................................
................................
.....

25

4.2.1.

Evaluation and Findings

................................
................................
................................
..........

31

4.2.2.

Summary

................................
................................
................................
................................
.

32

4.3.

High Fidelity Interface Prototy
pe

................................
................................
...............................

32

4.3.1.

Evaluation and Findings

................................
................................
................................
..........

36

4.3.2.

Exercise Expert

................................
................................
................................
.......................

37

4.3.3.

Summary

................................
................................
................................
................................
.

37

5.

Final System

................................
................................
................................
................................
........

38

5.1.

Introduction

................................
................................
................................
................................
.

38

5.2.

Final Prototype

................................
................................
................................
............................

38

5.3.

Implementation

................................
................................
................................
...........................

41

5.3.1.

System Architecture

................................
................................
................................
................

42

5.3.2.

Database

................................
................................
................................
................................
..

43

5.3.3.

Equipment Used

................................
................................
................................
......................

44

5.4.

Method and Users

................................
................................
................................
.......................

45

5.5.

Task

................................
................................
................................
................................
.............

45

5.6.

Findings
................................
................................
................................
................................
.......

45

6.

Can Facebook Users Incorporate E
-
Health into their Lives?

................................
..............................

49

7.

Conclusion

................................
................................
................................
................................
..........

52

7.1.

Future work

................................
................................
................................
................................
.

52

8.

References

................................
................................
................................
................................
...........

54

Appendix A


Online and Support Groups Questionnaires Results

................................
...........................

57

Appendix B


Final System Evaluation Form

................................
................................
............................

69




Department of Computer Science

E
-
Health System


vi


Table of
F
igures


F
IGURE
1
:

G
ROWTH IN
I
NTERNET AND
F
ACEBOOK USE
[E
LLISON ET AL
.

2008]

................................
..............................

6

F
IGURE
2
:

D
IABETES
P
ILOT
S
OFTWARE

S
CREENSH
OTS
[D
IGITAL
A
TTITUDES
2010]

................................
......................

9

F
IGURE
3
:

C
URE
T
OGETHER
I
NTERFACE FOR
U
SER

................................
................................
................................
.......

10

F
IGURE
4
:

S
PARK
P
EOPLE

I
NTERFACE FOR
U
SER

................................
................................
................................
...........

11

F
IGURE
5
:

T
HE
I
TERATIVE
D
ESIGN
P
ROCESS
[S
HARP ET AL
.

2007]
................................
................................
...............

12

F
IGURE
6
:

A
GE GROUP OF DIABETIC
S WHO FILLED IN QUES
TIONNAIRES ONLINE
(
LEFT
)

AND OF DIABETICS WHO

FILLED
IN QUESTIONNAIRES IN

SUPPORT GROUPS
(
RIGHT
)

................................
................................
...............................

16

F
IGURE
7
:

T
HE
D
IABETIC
T
YPES OF
ONLINE GROUP
(
LEFT
)

AND OF SUPPORT GROUP
S
(
RIGHT
)

................................
.....

16

F
IGURE
8
:

T
HE METHOD OF
I
NTERNET ACCESS BY ON
LINE GROUP
(
LEFT
)

AND BY SUPPORT GROUP
S
(
RIGHT
)

...............

17

F
IGURE
9
:

T
HE USE OF SOCIAL NET
WORKS BY ONLINE GROU
P
(
LEFT
)

AND BY SUPPORT GROUP
S
(
RIGHT
)

.....................

17

F
IGURE
10
:

T
HE USE OF
F
ACEBOOK BY ONLINE GR
OUP
(
LEFT
)

AND BY SUPPORT GROUP
S
(
RIGHT
)

...............................

17

F
IGURE
11
:

T
IME SPENT ON
F
ACEBOOK BY ONLINE GR
OUP
(
LEFT
)

AND BY SUPPORT GROUP
S
(
RIGHT
)

PER WEEK

.........

18

F
IGURE
12
:

S
WEATBAND
P
EDOMETER
W
ATCH GIVEN TO
U
SERS

................................
................................
..................

21

F
IGURE
13
:

P
APER
P
ROTOTYPE
P
ROFILE
I
NTERFACE

................................
................................
................................
....

27

F
IGURE
14
:

P
APER
P
ROTOTYPE
R
ECORDING OF
G
LUCOSE
I
NTERFACE

................................
................................
.........

27

F
IGURE
15
:

P
APER
P
ROTOTYPE
C
ALORIE
C
ALCULATOR
I
NTERFACES

................................
................................
..........

29

F
IGURE
16
:

P
APER
P
ROTOTYPE
R
EMINDERS
I
NTERFACE

................................
................................
..............................

31

F
IGURE
17
:

I
NTERFACE
-
B
ASED
P
ROTOTYPE
H
OME
P
AGE

................................
................................
.............................

33

F
IGURE
18
:

I
NTERFACE
-
B
ASED
P
ROTOTYPE
S
CREENSHOT OF
G
L
UCOSE
R
ECORDING WITH
T
IME OF THE
D
AY

............

34

F
IGURE
19
:

G
RAPHICAL
R
EPRESENTATION OF THE

P
AST
D
AIL
Y
A
VERAGE
G
LUCOSE
R
ECORDINGS

............................

34

F
IGURE
20
:

I
NTERFACE
-
B
ASED
P
ROTOTYPE
C
ALORIE
C
ALCULATOR

................................
................................
...........

35

F
IGURE
21
:

I
NTERFACE
-
B
ASED
P
ROTOTYPE
S
CRE
ENSHOT OF
R
EMINDERS
F
EATURE

................................
...................

36

F
IGURE
22
:

F
INAL
P
ROTOTYPE
3

T
YPES OF
G
RAPHS TO
D
ISPLAY
G
LUCOSE
H
ISTORY

................................
..................

39

F
IGURE
23
:

F
INAL
P
ROTOTYPE
F
OOD
S
ELECTION FOR
C
ALORIES
C
ONSUMPTION

................................
.........................

40

F
IGURE
24
:

F
INAL
P
ROTOTYPE
C
ONTACT
U
S
I
NTERFACE

................................
................................
.............................

40

F
IGURE
25
:

F
INAL
P
ROTOTYPE
S
NAPSHOT OF
S
UCCESSFUL
(
LEFT
)

AND
E
RROR
M
ESSAGE
(
RIGHT
)

.............................

41

F
IGURE
26
:

F
INAL
P
ROTOTYPE
S
CREENSHOT OF
T
IME AND
D
ATE
F
ORMATS

................................
................................

41

F
IGURE
27
:

O
VERVIEW OF
F
ACEBOOK
E
-
H
EALTH
A
PPLICATION
'
S
A
RCHITECTURE

................................
......................

42

F
IGURE
28
:

D
ATABASE
D
ESIGN

................................
................................
................................
................................
....

44

F
IGURE
29
:

N
UMBER OF
D
AILY
U
SER
E
RRORS DURING THE PIL
OT RUN OF THE

F
ACEBOOK
E
-
H
EALTH
A
PPLICATION
..

47

F
IGURE
30
:

N
UMBER OF
D
AILY
L
OGINS DURING THE PIL
OT RUN OF THE
F
ACEBOOK
E
-
H
EALTH
A
PPLICATION

............

47

F
IGURE
31
:

D
EMOGRAPHICS
,

G
ENDER AND
A
GE OF ALL
U
SERS WHO

INSTALLED THE
F
ACEBOOK
E
-
H
EALTH
A
PPLICATION

................................
................................
................................
................................
......................

49

F
IGURE
32
:

D
AILY
A
CTIVE
U
SERS AND
D
AILY
L
OGINS FOR THE
F
ACEBOOK
E
-
H
EALTH
A
PPLICATION

........................

50

F
IGURE
33
:

T
OTAL
N
UMBER OF
D
AILY
U
SER
E
RRORS
WHILE USING THE
F
ACEBOOK
E
-
H
EALTH
A
PPLICATION
...........

51

Department of Computer Science

E
-
Health System


1


1.

Introduction

1.1.

Problem Outline

Currently, there are no

publicly available E
-
H
ealth applications or systems that could assist in the self
-
management of diabetes in South Africa. However, there
are

a variety of commercially available
products,
such as

the Diabetes Pilot so
ftware, which have to be purchased and are not affordable by an
average person
[Digital Attitudes 2010]
. Furthermore, there
are

no publicly available online social
support structures for diabetics in South Africa and
patients prefer

remaining anonymous whe
n
interacting with other patients.


The core aspects of this investigation are diabetes self
-
management, social support and social networki
ng,
and the applicability of E
-
H
ealth to mobile technology. The system aims to assist in the self
-
management
of diabe
tes, method of record and data keeping, which is an essential part of a patient’s daily life. In
addition, it would also provide social support for diabetic patients in areas where there is scarcity of
support groups through the use of social networking.
M
oreover, the combination of social networking
with mobile applications could provide significant social support to patients and possibly transform the
diabetic healthcare system as well as the outlook of online networking.


1.2.

Proposed Solution and Division
of Work

The team

propose
s

to develop a system
that will assist in better self
-
management of diabetes for the users.
This is
executed

by means of a User
-
Centred Design process to ensure that potential users are involved to
a great extent in the research and the testing process of the diabetes management software.
In addition, the
team focuses greatly on building a system fo
r diabetics th
at is affordable,

user
-
friendly

and easily
accessible
.
The system is implemented on three different platforms, each developed and evaluated by the
different project team members. The workload is divided into three components:




Diabetes Self
-
Management Sys
tem on a Social Network


Diabetes Self
-
Management System on a Mobile Phone


Diabetes Self
-
Management System on a Website


This report is concerned with the self
-
management system for diabetics on a social network

and to
establish if potential users can manag
e their diabetes better on a social network.
Brian Sebastian will
cover the mobile phone system in his report and Dalton Jacobs will examine the website system in his
report.


1.3.

Ethical Consideration

As user testing will be conducted in this project, Ethics

Clearance was obtained from the Science Faculty
Research Ethics at the University of Cape Town to carry out user experiments.
In addition, to engage
diabetic students in the research
, permission from Student Affairs was also obtained.



With regards to legal issues, all participants
were

required to complete consent forms before interacting
with the prototypes.
Data that is captured
was

recorded in a manner such that the individual users cannot
be identified hence preserving privacy
. As d
ata such as blood glucose levels for users
was

recorded,
all

participants
were

informed that their data will not be used by the team without their consent.

Department of Computer Science

E
-
Health System


2



1.4.

Report Outline

The next chapter outlines the background work that was formerly completed related t
o the research of this
project,
motivating the need for the proposed system. The Requirement Analysis chapter
discusses the
procedures and steps taken to identify
and understand
the user requirements

which then led to the initial
system design
.
Chapter

4 p
resents the designs, testing and evaluations of the low and high fidelity
prototype iterations, where both iterations have their own key findings.
Chapter

5
examines the
implementation of the
final prototype which involved pilot
ing the system for a period
of five

days

and
also

reviews
the key findings of the project
.

Chapter

6
investigates the additional iteration of the final
prototype to explore the opportunities of a Facebook E
-
Health application in the long run. Concluding
remarks and potential future work are presented in the final
chapter
.


Department of Computer Science

E
-
Health System


3


2.

Background and Motivation

2.1.

Introd
uction

In 2004, about 3,4 million people worldwide died from elevated blood sugar levels
[World Health
Organization 2011]
. Diabetes is a deficiency where the body is unable to convert sugar or glucose into
energy
[Diabetes Research Wellness Foundation 2007
]
. The energy obtained from glucose is transferred
to blood cells in the blood who utilize the energy to perform their functions. The hormone named insulin
is required to execute this process, and a patient is diagnosed as a diabetic when the patient’s bod
y is
unable to produce adequate amounts of insulin
[Diabetes Research Wellness Foundation 2007]
.


In today’s time, people are socially and professionally interconnected, yet there is sparse communication
in the dense population of diabetic people. Network
ing has become more prominent with the sociology of
health and medicine
[Smith and Christakis 2008]
. In economically developing nations, including South
Africa, social networks are being utilized for spreading health awareness for people suffering from
chr
onic diseases such as diabetes. South Africa is a leading example in Africa that demonstrates on a
regular basis that social networks will be implemented using cellular phones in the coming years of
technology
[Horniblow 2010]
.
In 2010, a social network ca
lled

Facebook had over 400 million users
[Horniblow 2010]
. Furthermore in South Africa, social networking rated six
th

on the list of online
activities after reading the newspaper
[Witness This 2010]
.


The team

propose
s

to develop a system
, which allows
diabetics to manage their diabetes on a daily basis
on Facebook, on a mobil
e phone and on a website. With the intention of motivating the development of
such a system on Facebook, and to comprehend the functionalities the system requires, a
summary

of
diab
etes
as well as

social networking such as Facebook is given
in this
chapter

followed by

an overview
of other related
electronic health
project
s
.


2.2.

Diabetes

Diabetes is a deficiency in an individual’s body where the body is unable to convert sugar or glucos
e into
energy

[Diabetes Research Wellness Foundation 2007]
. The energy obtained from glucose is transferred
to the blood which is then utilized by the cells to perform various operations. The hormone named insulin
is required to execute this process, and a

patient is diagnosed as being diabetic when the patient’s body is
unable to produce adequate amounts of insulin
[Diabetes Research Wellness Foundation 2007]
. A simple
blood test on a frequent basis allows the glucose levels to be monitored hence confirming the diagnosis of
diabetes.


Along with exercise the diabetic people need to carefully monitor their diet and the right diet is necessary
to balance with t
he calories burnt during physical workouts
[Gakidou et al. 2011]
. The glucose levels
must be monitored at least twice a day by doing a simple blood test which can be done by self
[Diabetes
Research Wellness Foundation 2007]
. The results from these tests ar
e usually recorded and a copy is
often taken to the doctor for further analysis
when necessary. Gakidou et al [2011
] argued that recording
the changes in these measurements can further assist with monitoring of health and have patients write
these details
on paper or fill out a table manually. However, recording measurements on paper is not the
most efficient method, as paper records can easily get damaged or lost and the data is not easy to transfer
to electronic mediums. Nevertheless, close monitoring of
the various health variables allows precise
control of diabetes. There
are two key categories of diabetes, types 1 and 2 diabetes.


Department of Computer Science

E
-
Health System


4


2.2.1.

Type 1 Diabetes

Type 1 diabetes is due to lack of sufficient quantities of insulin in an individual’s body and often presen
t
in young adults and children. This type of diabetes can also occ
ur due to genetic inheritance
[Diabetes
Research Wellness Foundation 2007]
. The treatment for type
1
involves insulin therapy
where t
he patient
is injected insulin in order to restore the ab
sent insulin in the body

[Gakidou et al. 2011]
. Essentially
insulin therapy requires the patient to also have a balance between physical exercise and food
consumption so that the treatment can be much more effective. Even though insulin is reasonably price
d
and there is no dose limit, it consists of frequent injections and sudden weight gain
[Gakidou et al. 2011]
.
However, there is no known way to prevent type 1 diabetes.


2.2.2.

Type 2 Diabetes

Type 2 diabetes is due to a relative insulin absence rather than a co
mplete deficiency like in type 1
diabetes. It mainly occurs in adults and its presence is often associated with physical inactivity and
obesity
[Gakidou et al. 2011]
.
The treatment for

type
2

greatly depends on the individual’s blood glucose
levels and may

require the patient to
do exercise frequently

[Diabetes Research Wellness Foundation
2007]
. Diet and physical exercise play an important role in the treatment of type 2 diabetes and some
patients are advised to have a personal trainer or
a
dietician
. More
over, the onset of type 2 diabetes can be
delayed by controlling diet, exercising regularly and avoiding consumption of tobacco and alcohol
[World
Health Organization 2011]
.


There are a number of symptoms of diabetes such as hazy vision, unusual thirst,
urinating repeatedly and
drowsiness

[Diabetes Research Wellness Foundation 2007]
. The most important goal of any diabetes
treatment is to gain control over the glucose levels in the body which can furthermore help avoid diabetic
complications.


2.3.

Social Net
work
ing

Social networks have been around in different forms for just over a decade
[Cerado 2006]
. They allow
individuals with common attributes, such as education, beliefs and interests, to come together and share
their experiences and information online w
ith other users
[Hojilla 2010]
. It is beneficial as it allows users
to give feedback and post comments on data that is being shared. Such networks consist of users having
an online profile which contains their information. There are many online social netw
orks; the prominent
ones are Facebook, Twitter and MySpace. In addition
,

social networks are valuable as they allow people
to interact with each other in isolated ar
eas and, according to Hojilla [2010
], these networks are usually
cheaper than the tradition
al manner of social communications; i.e. excludes travelling expenses.

On the
other hand, Hojilla’s [2010
] view may not always be correct as social networking has other requirements
too. These include having access to the Internet thus the need to have a c
omputer or mobile phone, and
this increases the overall cost to utilize a social network. Nevertheless, online networks can offer a
helping hand to those people who are remote and perhaps not in contact with their family and friends.


2.3.1.

Social Networking and

Healthcare

The Internet is an ever growing collection of various resources. Immense amount of information about
health issues is also being exchanged on the Internet which has become essential for patients living with
chronic disease
s

like diabetes
[Green

et al. 2010]
.
In addition, patients use the Internet for health
information more than communicating with their doctors
[Green et al. 2010]
.
The traditional healthcare
system has restricted selection of professionals and has restricted access to suitable p
rofessionals [
Houtart
2009;

Kalpan and Haenlein 2010
].
Houtart

[2009
] said that the traditional system is also not very cost
Department of Computer Science

E
-
Health System


5


effective and any information received from a doctor has to be paid for, but most of the online
professional advice also has to be
paid for and in some cases money may be lost to deceitful users
claiming to be doctors.


Diabetes online communities overcome most of the problems of a traditional healthcare model as there
are no restrictions of time, space or mobility. Such online societ
ies contain features such as groups,
forums, blogs, news and live chats which all allow a patient to be better informed
[Houtart 2009]
.
Kriescher [2009
]

mentioned that a number of blogs represent a personal online diary for users and is a
good manner of re
cording life events, but blogs and forums would fail when patients may require a more
instant feedback. Nonetheless, forums and blogs would be appropriate when patients need to reflect on
past answered questions. People using online sites share various eff
ects of prescriptions and diet
supplements so that they could compare and associate their personal experiences which is quite
beneficial. Most of the information shared online is sensitive as well as private for a diabetic person,
which may be improbable t
o be mentioned in a doctor
-
patient communication
[Green et al. 2010]
.
Another advantage is that online networks allow quick access to a wide range of medical information as
well as patients can get immediate feedback. Social networking is a mechanism that
is not only educating
people about different diabetes issues but is also connecting people on the basis of their experiences and
eliminating the need to physically visit a doctor
[National Center for Chronic Disease Prevention and
Health Promotion 2011]
.


Greene et al [2010
] argued that social networking can eliminate depression by reducing loneliness among
individuals but online networking can also broaden emotions such as despair and pessimism and not only
happiness and optimism as ass
ured by Smith and Ch
ristakis [2008
]. The health of a person is affected by
social networks through mechanisms such as social influence, social engagement, one
-
to
-
one
communication, access to resources and the element of social support
[Smith and Christakis 2008]
.
Furthermore,

patients’ disease management and health
outcomes

have
enhanced

to a great extent through
the use of social networks
[Green et al. 2010]
.

Social networks also allow
big

multinational firms, small
and medium sized firms, and non
-
profit and government agenci
es to engage with their potential users and
communicate

more efficiently with direct end
-
consumers
[Kalpan and Haenlein 2010]
.

Users also utilized
social networks to stay in touch with their “health buddies” who they met offline and online
[Newman et
al. 2
011]
. This would result in friends posting about their health and exercise on the network which would
encourage their “health buddies” to exercise too.
The major advantage of social networks for patients is
that they provide patients emotional support as well as offers them access to experience
-
based
information about specific medical treatments and remedies, which many patients find more valuable than
pr
ofessional advice
[Newman et al. 2011]
.

Conjointly, social networks not only connect patients that are
geographically located far away, allows them to interact anonymously
and receive social support without
being judged by others
[Newman et al. 2011]
.

Newm
an et al [2011
] realized that diabetics mostly join
social networks to seek health information soon after they have been diagnosed with diabetes. Moreover,
many online patients realized that to receive health information they also had to share some informa
tion,
hence making social networks a potential contender for future health management systems

[Newman et
al. 2011]
.


2.3.2.

Facebook

Facebook is currently the leading social network with over 700 million active users in the year of 2011

from which nearly 4 milli
on active users reside in South Africa [
Digital Buzz Blog 2011;

Social Media
Blog 2011
]
.
Each day, 20 million applications are installed on Facebook and each month, there are over
250 million people who network with Facebook from outside the official websi
te across 2 million
websites
[Digital Buzz Blog 2011]
. In addition, Facebook is easily accessible via mobile phone for over
Department of Computer Science

E
-
Health System


6


200 million people.
Facebook members also disclose more information about themselves than any other
social network which makes it a
strong candidate for patients to seek and share their experiences with one
another
[Dwyer et al. 2007]
.

Dwyer et al [2007
]
also
determined through their research, that most
Facebook users are willing to share information that would reveal their true identity thus trusting the site
much more than other social networks.


Figure 1 shows the results of a research conducted by
Ellison

et al [2008
] on a sample group

from 2006 to
2007
. Th
e figure

clearly illustrate
s how the time spent by users on the Internet has increased in the past
years
,

which
has

also had a positive effect on the amount of time spent on Facebook. Users spent nearly
a
quarte
r of their time on the Internet

using Facebook in 2007.


A reason why Facebook is more popular than other social networks such as Twitter is because Facebook
is more configurable and allows users to make applications using the software development kit and then
post them on Facebook
[Krivak 2007]
. Facebook method

calls are established over the Internet using the
HTTP, GET or POST requests which are sent to the REST server, establishing a REST
-
based interface
for the Facebook Application Programming Interface (API)
[Krivak 2007]
. Additionally, one can use
profiles,

friends, photos and events information along with the API to attach a social framework to an
application.


Facebook has also encouraged a number of websites of disease
-
specific groups to become popular and
known among the community, corresponding to essen
tial resources of health information, support and
engagement for patients with similar chronic diseases
[Green et al. 2010]. As Greene et al [2010
]
illustrated in their study sample that the largest 15 Facebook groups for diabetics consisted of 690
individ
ual posts on wall pages and discussion board written by nearly 480 distinctive users. This clearly
shows that patients are already actively using Facebook to seek support and medical information. Most o
f
the posts in Greene et al’s [2010
] study expressed p
ersonal experiences of diabetics with the management
of the disease as well as consisted of sharing of very sensitive areas of diabetes management which are
doubtful to be disclosed in patient
-
doctor relations. This allowed patients to utilize Facebook int
eractions
to support their own validations of tight control.
Newman et al’s [20
11
] research demonstrated that
patients, including diabetics, joined Facebook to receive social support, inspiration, accountability and
guidance on diabetes and weight manageme
nt.
People experiencing the same struggles, offer
optimistic
and heartening comments on social networks as well as gave quick answers at any time of the day
Figure
1
: Growth in Internet and Facebook use [Ellison et al. 2008]

Department of Computer Science

E
-
Health System


7


[Newman et al. 2011]
.

Another benefit of Facebook is that Facebook encourages social accountability
-
related communications which allows users to monitor other people’s plan and how they are pursuing
their
respective
plan. This to a great extent motivates users to
achieve their

own personal goals by
following other members of a social network or even from specific role models. Such determined groups
with health conscious people already exist on Facebook and
are also

improving health
[Newman et al.
2011]
.


Kalpan and Haenlein [20
10
] argued about the irrelevance to follow friends on social networks half
-
way
around the globe when most people do not even know their neighbours. However,
Ellison et al [2007
]
suggested that online connections often resulted in offline or face
-
to
-
face me
etings thus reflecting the
importance of social networks such as Facebook in today’s world.
Moreover, Facebook users, including
diabetics, prefer “searching” for offline users to engage and share interests with rather than to “browse”
for unfamiliar people

to meet which makes Facebook one of the social networks where patients can easily
engage
[Ellison et al. 2007]
.


Users in diabetic groups on Facebook have diverse interests and medium of communication
,

and mostly
consist of patients, family members, advertisers and researchers
[Green et al. 2010]
.

Facebook not only
allows users and patients to have interpersonal and society social support via wall posts and conversation
threads, but
it

also allow
s

users

to have
a
quick and easy access to dedicated information on diabetes
management from
the community and
friend
s. In addition, Facebook gives diabetics

an optimistic and
rational self image of themselves as diabetic individuals
[Green et al. 2010]
.

Facebook

allows users to
express and understand friends from offline social networks thus giving the users the ability to retain
lightweight contact with a wide variety of acquaintances, which makes Facebook a fitting social network
for
patients

engaging with
othe
r
patients
[Ellison et al. 2008]
.

Significantly, Facebook offers both
technical and social infrastructure for social communications
[Ellison et al. 2008]
.


2.3.3.

Opportunities

Social networks have many advantages as mentioned
previously in Section
2.3

but they also offer various
opportunities to individuals and firms to explore and expand.


Advertising on Social Networks:

Social networks

can be used as an advertising tool to reach out to the specific patients and users.
Facebook groups for diabetics and other health related groups
also
consist of wall posts that advertise
products to users such as dietary supplements and natural cures for

a disease
[Green et al. 2010]
.


Career and Contact Opportunities:

Social networks also give many prospects to evolving individuals. Currently, many students around the
world have access to social networks, and this allows them to keep in touch with frien
ds and the
community after graduation,
thus
giving them healthy connections and opportunities in terms of jobs and
internships
[Ellison et al. 2007]
.


Health Related Interactions:

The introduction of proactive feedback on social networks

with people revis
ing comments to help and
support their friends along with extra features can certainly make social networks more beneficial to
health interactions
[Newman et al. 2011]
.

Department of Computer Science

E
-
Health System


8


Self Reflection:

There are also opportunities for users of social network
s

to self
reflect
[Owen et al. 2009]
.
Taking time to
reply to others and writing of discussions allows users to reflect before responding to others. Therefore,
there are many opportunities available for social networks to expand as well as to support users with or
w
ithout health issues.


2.3.4.

Challenges

As easy as it is to gain access and utilize social networks for multiple purposes,
however,
it does
have

some
major challenges

and limitations
.


Distraction to Users:

Social networks are an immense distraction for employees at their work places and in some situations;
organisations have blocked certain websites from being accessed during working hours
[Kriescher 2009]
.
Various social networks, such as Facebook, are quit
e addictive due to the applications they provide, hence
there is apprehension about children spending more time socialising rather than learning
[Hojilla 2010]
.


Security of Social Networks:

Social networking is also not protected from hacking and other ty
pes of online mistreatment.
Recently,
Facebook
CEO

Mark Zuckerb
e
rg

experienced his
F
acebook page
being hacked too

[TechCrunch 2011]
.
Therefore, s
ecurity measures must be implemented carefully as identity theft and the misuse of personal
data is a major us
er concern. Some people create fake profiles and then threaten other users in the
network. Over 70 million users have experienced cyber bullying which is a common example of social
network exploitation
[Hojilla 2010]
. Social networking is the communication

technique for the modern
era but there is still fear about deceitful users
, a challenge

that should be addressed cautiously when
designing a network.


Inaccurate Information and Product Advertising on Social Networks:

Another challenge is that patients on

social networks are unacquainted to the degree the information on
the social network
s

is clinically correct as well as
whether

they
obtain

medical information due to some
dangerous

activities on the network by other users
[Green et al. 2010]
.
Furthermore,

advertising of some
products may be implemented with a positive and honest intention on social networks; some advertising
can harm and influence users to consume incorrect products which can be considered as a major challenge
in any given social network w
here patients are involved.

The failure to authenticate the character of poster
and the use of Facebook pages for non
-
FDA (Food and Drug Administration) approved products does
pretence

an important challenge for users to trust information of social network
s
[Green et al. 2010]
.
However, from

the research of Greene et al [2010
], it was established that not a lot of risky or deceptive
activities were sustained by diabetic groups on Facebook.


User Self Image:

Some social networks have

become a place for users desiring

to convey their healthy and happy identity
to their friends
[Newman et al. 2011]
. The fear of a lower self image of individuals with health problems
does create a challenge for them to share their issues openly on social
networks; therefore giving users an
option of how and where they prefer to post discussions could encourage more patients to utilize social
networks for health related interactions.
Efforts should be put into to ensure users have access to the right
people

on the network to support their objectives and choosing the correct communication channels.

Department of Computer Science

E
-
Health System


9



Technical and Literacy Difficulties:

There are also technical and literacy challenges that a user may face while using a social network
[Owen
et al. 2009]
.
Users

may require explanation of certain technological words or acronyms as some patients
may be new to the idea of utilizing social networks

for health information.
There are also challenges in
balancing between the security of users and their data on a social

network as well as the need to make the
services easily accessible to all members
[Owen et al. 2009]
. Furthermore, all legal and ethical procedures
should be clear and understood by all members.


Technical Difficulties for Programmers:

On the technical s
ide, Facebook API, programming support and application requirements change on a
regular basis. The Facebook programming platform changes so often that it generates problems for
software and system designers and programmers to keep up with the latest Facebo
ok requirements.
Therefore, any Facebook project would require frequent updates and changes to remain compatible with
the current platform.


All these challenges should be considered with high priority in order to meet consumer satisfaction and
the success of any future related work.


2.4.

Related Systems

2.4.1.

Diabetes Pilot Software

Diabetes Pilot Software,
well

known commercial software, is available online for diabetic patients
allowing them to easily record and maintain their glucose measurements as well as assists in maintaining
a healthy diet.
In ad
dition, this software has various

versions that can easily be installed on different
desktop computers and mobile devices including mobile phone
s and iPads

[Digital Attitudes 2010]
.

It is
compatible with most of the latest technology,
however
,

Diabetes
Pilot

S
oftware is
quite
costly and
is
not
affordable by an average user
which constraints

the
overall recognition of the product in less privileged
societies and in developing nations such as South Africa
.
Figure 2 shows two screenshots from the
software o
n a
n

iPhone, where the users can set reminders and view a detailed report as well.

Figure
2
: Diabetes Pilot Software

Screensh
ots [Digital Attitudes 2010]

Department of Computer Science

E
-
Health System


10


2.4.2.

CureTogether and PatientsLikeMe

Emotional support and information sharing are the most essential and basic services of a healthcare social
network
[Swan 2009]
. CureTogether

and PatientsLikeMe are data driven social networks that are
available for people online and both networks allow patients to join and communicate for free.
A
n

advantage

of CureTogether
network is that it
allows patients to anonymously compare and track
their
conditions with one another. Nevertheless, it focuses more on information sharing and not so much on
emotional support [
Dhillion 2010
;

Neal 2010
]. Conversely, members of PatientsLikeMe network offer
one another support based on their own experience
s

and advise each other on medical issues and how to
improve day
-
to
-
day life
[Frost and Massagli 2009]
.


A
key

disadvantage

of websites such as PatientsLikeMe and CureTogether is that they are not easily
accessible from any other device other than a compute
r thus restricting communication between patients
and the networks
[Swan 2009]
. This may be the reason to why
these healthcare networks are not well
known among various communities
, especially in South Africa which is a developing country
. In
addition, the
se networks are not directed specifically towards diabetic patients, but aimed at various
patients in general.
Figure 3 shows the interface of CureTogether network where the users can enter their
measurements which is illustrated on graphs.


2.4.3.

SparkPeople

SparkPeople is an online health community service that is provided to people with any health issues

[Newman et al. 2011]
.
SparkPeople

also focuses on people who want to lose weight as well as assists
diabetics to manage their illness
therefore

this network is not for just socialising purposes and instead
also
concentrates on health management
[Newman et al. 2011]
. However, SparkPeople i
s not a free service
and for users to view most of their results comes at an expensive cost.
This is a major drawback for
patients with chronic diseases who are looking to manage their illness online. The service is expensive
and the interface is also not
user
-
friendly, which has discouraged many members and resulted in
users

leaving the SparkPeople network
[Newman et al. 2011]
.
Figure 4 illustrates a screenshot of the
SparkPeople interface.

Figure
3
: CureTogether Interface for User

Department of Computer Science

E
-
Health System


11






2.5.

Conclusion

The prospective of social network communications for health enhancement is colossal as they

have a
diversity of positive results regarding the social aspects of a community

such as improved public health,
more interacted

communities, lower crime rates and more efficient financial markets for
consumers

and
organizations
as well as positive outcomes on psychological welfare
[
Ellison et al. 2007;

Ellison et al.
2008
].


The accomplishment of social networking is self
generating, the higher the number of users on a network,
the better the opportunity for productive associations
[Kriescher 2009]
. An individual’s actions and
physique are not the only elements that contribute to a healthy lifestyle. Other factors such as the people
around the individual, influence and support are also significant for a healthy standard of living, which are
provided

by social networks
[Smith and Christakis 2008]
. Patients spend an immense amount of
unnecessary time organising their health history, filling out forms at the hospital and even wasting a lot of
money seeking valuable professional advice. Particularly with

patients who have diseases that require
immediate attention and correction, such as diabetes, time is a delicate variable. The integration of a
social network such as Facebook with a third party, such as a website, can assist with these intensifying
issue
s as Facebook allows one to create applications that can be linked with other websites [
Ko
2010
;Krivak 2007
]. Social networks
and applications can also

allow patients to store and instantly access
their recordings, interact with other patients about their
experiences as well as view visual representations
of their recent health history.
In addition, many mobile phones allow
users to access social networks and
their applications. Such
exploitation

of a social network would revolutionise the healthcare system

and it
can facilitate important associations and be the stimulant for the prospect online networking progresses.



The

plan of how to work out the procedure of identifying the user requirements, design of the
initial
interface, and which tools and strateg
ies will be used, are introduced in the
requirements analysis

chapter.

Figure
4
: SparkPeople

Interface for User

Department of Computer Science

E
-
Health System


12


3.

Requirement Analysis

Chapter

3.1.

Introduction

The
Requirement Analysis

Chapter will give an overview and explicate in detail the procedures that were
implemented to design the diabetes self
-
management system on a social network as well as the steps
taken to identify the user requirements.
As initially the user needs are not
fully understood, t
he key goal
of this project is to develop a system by means of a User
-
Centred Design process
.

This

will not only
ensure that users are involved to a great extent in every stage of the project, both
the
research and testing
process, but w
ill also allow the team to recognize the correct user requirements and develop appropriate
prototypes on the evaluation of user needs.
In addition, User
-
Centred Design w
ill allow the system to be
user
-
friendly such that the users are able to use the system

with ease and are willing to become
loyal

to
the system.
Furthermore, this
chapter

also discusses the meetings and interviews that were conducted with
various professionals

to recognize and understand potential user requirements.















As a result, the key activities of
the design process are, identifying the correct user needs, designing and
building prototypes, and finally testing and evaluating the system until a feasible, user
-
friendly final
system materializes. As illustrated in Figure 5, these stages are preformed i
te
ratively throughout the
project; hence
the
system design
and functionalities
evolve
d as

iterations were completed
.

Identifying
user requirements, the first stage of the design process, consists of interacting with diabetics through
interviews and questionnaires. The
next

stage of the design process focuses on the designing of the
features and functionalities of the sy
stem which are derived from the
initial

stage, followed by the
building of the various types of prototypes in the third stage of the design process.
The prototypes were
then tested by the potential users in the fourth stage of the design process, and evalu
ated by means of
conducting short interviews with the users which led to identifying more user requirements,
as a result
starting the design process cycle again.
This chapter examines
on

identifying and understanding the initial
user requirements

to design

the primary system interface and features
. Chapter 4 and 5 will outline the
other stages
of the design process as they discuss the low fidelity, high fidelity and the final prototypes
that were designed, tested and evaluated.


3.2.

Understanding the Users

and

their Needs

As the focus of this research is on User
-
Centred Design, it is essential to understand the potential users of
the project as well as to identify and comprehend their needs.


In order to understand diabetics and their daily needs more thoroughly, the project team engaged with a
number of professionals. The findings from the various professionals and the impact of these findings on
the project itself will be discussed in more de
tail in this section.
Furthermore, the team interacted with a
Figure
5
: The Iterative Design Process [Sharp et al. 2007]

Identifying User
Requi
rements

Designing
Prototype/System

Testing and
Evaluation

Building
Prototype/System

Final System

Department of Computer Science

E
-
Health System


13


great number

of diabetics from both inside and outside South Africa through the means of paper and
online questionnaires as well as through participating in diabetic support group

meetings

in th
e Western
Cape of South Africa.
Various methods were utilized t
o reach the diabetic users for the project

such as

interacting with
support groups from Diabetes South Africa (DSA)

and

interacting with various online
groups and communities
.

University of Cap
e Town s
tudents
with diabetes were also encouraged

to
participate in the research.

These various groups of users will allow interactions with a diverse age group
of users in order to obtain contribution from both Type 1 and Type 2 diabetic patients.


3.2.1.

Suppo
rt Groups

In the initial stages of the project, meetings were held with different DSA support group leaders located in
Mitchells Plain in the Western Cape of South Africa.
These were conducted by the team members
personally going to the different leaders’
houses in Mitchells Plain and engaging them in a detailed
conversation about the illness and the members of the support groups.

These meetings were an essential
starting point to the research as they gave the team members an insight view of
why

and
how
diabetic

support groups are conducted. In addition, the team was given various kinds of diabetic books

such as
“Diabetes and You” by DSA which explained more about the illness as well as the daily life of South
Africans with diabetes. Such books are also g
iven to the members of the support groups

to help them
understand the illness better.
Carol H
endriks,
Mercia Roman and
Sandra Lewis are the leaders of the DSA
support groups who were approached by the team. The one
-
on
-
one meetings with the leaders allowed
the
team to understand the different support groups better even before meeting the
group
members.
Furthermore, it created a trusting bond with the leaders and the team successfully was able to obtain the
leaders support. After the one
-
on
-
one meetings, the
team participated in the support group monthly
meetings

where paper questionnaires were distributed to obtain initial requirements for the system
.
The
meetings with the different groups and their respective leaders went as following:


Carol Hendriks

Suppor
t Group
:

From the one
-
on
-
one meeting, the team
was

able to get some useful DSA books which diabetics are
provided with when they join a support group. As Carol explained to the team, these books give diabetics
more information about the illness
as well as
told real experience
-
based stories which were all set in
South Africa
n

villages and small towns. In addition, Carol i
nformed the team that unfortunately the DSA
support groups have

a limited type of
people joining the groups
. Most were women above the age of 40
years
and the reason why men did not want to participate was because of their “
pride” as they
did not
want to portray themselves as weak by

join
ing

such groups,

therefore their wives attended the support
groups on thei
r behalf.
Furthermore, y
oung people did not participate in such support groups
only
to
avoid embarrassment from friends but mostly because most support groups for diabetics in South Africa
are

aimed at the Type 2 diabetics and not at the Type 1 diabetics w
hich involved teenagers and young
adults as Carol explained.
The team was also informed that most of the members

of the group

use basic
mobile phones and do not have access to the Internet.
T
he leaders of the support groups
also
provided the
members with f
ood and free blood tests during the monthly meetings from their own money whilst the
DSA paid for the venue and the staff salary. This showed the great selfless motivation the leaders had to
help the society, therefore the team members decided to contribut
e as well by assisting the leader of the
group in buying the food. This was also seen as a tool to gain the participants trust as well as to motivate
them to participate in
the

research.


During the monthly meeting of the support group which consisted of 2
9 diabetics, the team also
participated and observed the discussions
as they
evolve
d
. Initially, all the members got their blood
glucose reading taken and the team was introduced to the group. On a regular basis, the monthly group
Department of Computer Science

E
-
Health System


14


meetings involved guest s
peak
ers, such as dieticians
, health experts as well as endocrinologists, to come
and motivate the members of the group. After all the discussions and guest speakers, the team members

of
the project gave the diabetics

a brief outline of the project and answ
ered any questions

that were asked
related to the research
. Then paper questionnaires were
distributed to all the members and each
team
member assisted the
diabetics

in

fill
ing

out the questionnaires. Just from observing, it was quite obvious
that the user
s were struggling to answer the questions as some did not know how to read and write. In
such cases, the team members helped the diabetics by reading out the questions and writing the answers
on their behalf.
However, some people in the group found reading

so difficult that they were uninspired to
fill in the questionnaires and had to be encouraged by the group leader to complete the questionnaire.
About 90%

of the diabetics in the group were females and
most
were over 50 years of age.
Some
members participated in the meetings for their husbands
and

children. Moreover, some members were not
even diabetic but just participated because of the information about health being shared or for
some
common health issues
that
they shared with diabe
tics
,

such as high blood pressure.
Surprisingly,
most of

the
support group
members were very cheerful and friendly towards the team members and appreciated
the attention and efforts the students were putting in
to understand
ing

diabetes better.


Mercia

Rom
an

Support Group
:

The one
-
on
-
one meeting with Mercia revealed almost the same
facts

about the diabetics in
Mercia’s

group as
the
diabetics in
Carol
’s

group.
The team did not get an opportunity to meet the diabetics during
their monthly support group meeting
but

found out that a number of members in Mercia’s group were
Muslims. A key finding about this group was that the Muslim diabetic members did not keep Ram
adan
fasts as diabetics are not allowed to fast
so as to

avoid

their sugar levels
from dropping too low
.
In
addition, some of the members of the group were not diabetic but participated in the monthly support
group meetings with the intention of obtaining
information related to diabetes for their parents who were
old and did not have access to health information. Mercia explained that a great number of the people join
such support groups due to lack of information and support available for diabetics in Sout
h Africa,
especially for people who are above 50 years of age and have limited access to information

due to lack of
knowledge about technology
.


Sandra Lewis

Support Group
:

There was no one
-
on
-
one meeting with Sandra,
but

the team was able to
participate
in

the monthly
support group meeting
which was attended by 27 diabetics. Again most of the people in the group were
females and over the age of 40 years. There were a few females in their thirties as well. After getting their
blood glucose levels checked,
the participants engaged in a health discussion which the team members
observed quietly. Surprisingly, the group leader was conversing with the diabetics by openly revealing
their glucose reading for the day. None of the diabetics took this offensively; in
stead everyone knew each
other in the group and laughed about the high, low or normal readings.
The group also has experts who
come in to give advice on a regular basis and again, the food was provided through the group leader’s
own money. All the particip
ants gave each other advice openly and loudly, as well as joked about the
consequences of not checking their glucose levels and of not exercising.
Even though everyone was
enjoying the discussion, there were moments where Sandra gave advice with a serious
tone.


A key problem that was observed was that
the
diabetics were not recording and checking their blood
glucose levels on a regular basis which they are supposed to do. They did not mind sharing personal
experiences and information with other diabetics
within the same group as everyone knew each other
quite well. Mostly the group leader, Sandra, gave information about various exercises and kept reminding
them to exercise, eat healthy and take their medications. She encouraged everyone to remind each othe
r
about these daily activities
, which some people did on a daily basis
.
After the discussions, the team
Department of Computer Science

E
-
Health System


15


members gave an introduction to the diabetics about the research project and distributed the
questionnaires. Sandra’s group members were very enthusiast
ic and to motivate them more, the team
members gave everyone who filled a questionnaire a voucher as a
women’s

day gift.
This really made the
group leader and the group members
more

enthused thus the team members managed to have some
constructive conversat
ions with the diabetics

as they filled in the questionnaires
.
Yet again the
participants complained about the lack of support for diabetics in South Africa, and how the monthly
support group meeting was their only way to share information and provide and s
eek support.


3.2.2.

Questionnaires

The
first

iteration

of the research focused

on the quantitative assessment and
was

executed

through the
use of questionnaires and
an online survey
.
A quantitative assessment counts, logs or measures something
of interest in
the user actions, and the description generated is usually in numeric terms [
Marsden and
Jones 2006
]. Questionnaires are a sequence of questions which are used for collecting information from
users, and are a well recognized research tool. An onli
ne versio
n of the questionnaire

was also made
available to gather information from more potential participants that are
located in other regions of South
Africa

and to engage with online communities to achieve
effective

research coverage.


T
he main
reason the team

decided to utilize questionnaires

is
because
they are cost effective and the
results can be quantified
and

questionnaires can also be
made available online on existing online
communities or even
emailed to
potential
users

[
Marsden and Jones 2006
]
. The maj
or weakness of
questionnaires however, is that they are only as good as the questions asked therefore cautious
consideration
was paid to what questions were

asked [
Marsden and Jones 2006
]
.
At the end of the
questionnaires any participants keen to partake further in the research as well as in prototype testing
voluntarily gave their email

or phone numbers
so the team can contact them later on in the research
.
As
this project has a large quant
ity of potential users, it is essential to reach a wide range of users and subject
groups, and this can be achieved especially through the use of
online questionnaires
. The advantage of
sending questionnaires via emails
or online questionnaires
is that
the

potential
users are able to answer
questions in their own surroundings and space, thus are more comfortable to give a more effective
feedback

[
Marsden and Jones 2006
]
.


Questionnaires were distributed to all the support groups that were interacted with an
d in total 56
questionnaires were completed by diabetics belonging to the various Mitchells Plain support groups.
On
the other hand, 44 questionnaires were completed on the onli
ne version of the questionnaire

which
included diabetics from all over South Af
rica as well as some diabetics from Botswana, Canada, USA and
India.


Most of the diabetics in the support groups were females and all were above the age of 30 years whereas
the diabetics in the online group were in a wide range of age groups, most being b
etween the age of 20
and 30 years old as shown in Figure 6. This
evidently

illustrates that the online society consists of
younger people in comparison to
the
social support groups which consist of
elder

people.
In addition, this
clearly shows that the new

generation of diabetics are more interactive online and are comfortable using
the Internet whereas the older generation of diabetics prefer a more traditional method of interacting
which does not consist of Internet relationships.



Department of Computer Science

E
-
Health System


16


Furthermore, another discovery through both set of questionnaires was that some diabetic
s did not know
what diabetic type
they
were. Figure 7 shows that most diabetics in the support groups were Type 2
whereas the online group had an even distribution of the two types of diabetics.


From all the questionnaires completed (100) it was evaluated that the tasks that most of the diabetics
performed on a daily basis was the measuring of glucose levels

(38 diabetics)
, monitoring diet

(28
dia
betics)
, monitoring physical exercise
(29 diabetics)
and monitoring medication

(44 diabetics)
.
Therefore, the team designed a system that allowed a diabetic to record glucose levels, monitor physical
exercise, diet and medication on a daily basis.
Very few

diabetics measured blood pressure, cholesterol,
insulin, monitored weight and sleep on a daily basis.
In addition, 62 diabetics
were
already shar
ing

their
glucose level details with friends, other diabetics or acquaintances as well as 56 diabetics
were
shar
ing

their diet plans. 32 diabetics who completed the questionnaires were also sharing their medication details
with friends, other diabetics or acquaintances

and if given the opportunity 53 diabetics would like to view
the medication treatments of othe
r diabetics
.
All this emphasizes that diabetics want to share
more
information with other diabetics if given an opportunity and are already sharing
health information with
others.


A surprising finding from the questionnaires was that 80% of the di
abetics

from the online group did

not
belong to any diabetes support group which highlights more on the fact that diabetics or patients join
online communities as a result of not finding social support in their region.
Figure 8 shows
from
where
the diabetics
in

t
he support group and the online group have access to the Internet. All the diabetics on the
online communities had access to the Internet in more than one
way,

however hardly anyone from the
support groups had access to the Internet.
Therefore, the team fo
cused on designing a system that is easily
Figure
6
:
Age group of diabetics who filled in questionnaires online (left) and of diabetics who filled in questionnaires in
support groups (right)

Figure
7
:
The Diabetic Types of
online group (left) and of support groups (right)

Department of Computer Science

E
-
Health System


17


accessible from multiple platforms, and
integrating all the platforms thus
allowing the users to utilize any
of the platforms
and still

have access to their information.


From the questionnaires it was also analysed that 88% of the diabetics from the support group were not
engaged in any social network
s

whereas 89% of the diabetics from th
e online group were engaged in at
least one social network as illustrated in Figure 9.

Furthermore, from all the diabetics that were engaged in a social network, Facebook was by far the
favourite social network. In fact out of the 46 diabetics who were using at least one social
network,
all
had Facebook accounts from which
37 logged onto Facebook on a daily basis and 7 logged onto
Facebook on a weekly basis. The diabetics from the support group did not utilize any other soc
ial network
except for Facebook,

however
,

12 diabetics from the online group were also a member of Twitter of
which only 2 diabetics logged onto Twitter daily. Figure 10 demonstrates the use of Facebook by both
support and online groups.

Figure
8
:
The method of Internet access by online group (left) and by support groups (right)

Figure
10
:
The use of
Facebook by online group (left) and by support groups (right)

Figure
9
:
The use of social networks by online group (left) and by support groups (right)

Department of Computer Science

E
-
Health System


18



Figure 11 shows how much time the users with Facebook acco
unts from the support and online group
s

spend on Facebook each week. Most users spend just a few minutes each day therefore it is essential for
the system on Facebook to provide fast efficient services and be
user
-
friendly
,
hence
not occupying a lot
of the

user’s time
.

Out of the 46 diabetics that are members of Facebook, 26 diabetics were already utilizing Facebook to
seek health r
elated information by joining groups or by just sharing wall posts with friends
, which
emphasizes on the fact that social networks are being utilized for sharing health related information

by
patients
.
Thus the system design would include a forum or a wall for diabetics to engage in discussions.
Furthermore, 42 diabetics out of the 46 wanted to use social networks to learn more about diabetes, and
32 diabetics desired to receive notifications with regard
s to their appointments and medications via social
networks.
Therefore, the design would also include a tip feature, where diabetics are given various health
tips on a daily basis hence educating them more about the illness. A reminder feature would also b
e
designed to allow users to set notifications regarding appointments, medications or other health activities.
In addition, 41 of the 46 Facebook diabetics preferred to share experiences with other diabetics about the
illness and get to meet more diabetics

on a social network.
Finally, the 46 diabetics who were engaged
with social networking were asked if they would utilize social networks to self
-
manage diabetes
, 38
diabetics agreed and the
reason why the remaining
diabetics disagreed was because they were

worried
about the privacy control of their diabetic data online.
They did not want others to see their health
information, especially friends who were not diabetic. This made the team design a system that would not
display the patients details and health
history to other users and only allow diabetics to share experiences,
diet tips and provide social support by them posting comments on the main wall or forum of the system.