Computers in Industry 69 (2015) 22–29

Contents lists available at ScienceDirect

Computers in Industry
journal homepage: www.elsevier.com/locate/compind

Impact of Mobile Diabetes Self-Care System on patients’ knowledge,

behavior and efficacy
Sophie Huey-Ming Guo a, Her-Kun Chang b, Chun-Yi Lin b,*
a
Department of Nursing, Mackay Medical College, New Taipei, Taiwan
b
Department of Information Management, Chang Gung University, Taoyuan, Taiwan

A R T I C L E I N F O A B S T R A C T

Article history: Evaluating the impact of emerging mobile technology on chronic disease management such as diabetes
Received 30 January 2014 mellitus has been a persistent concern among healthcare professionals. However, most of the previous
Received in revised form 2 November 2014 studies have focused on assessing metabolic outcomes, rather than changes in patients’ self-care ability
Accepted 4 November 2014
and practices, which have been found critical in properly managing the disease. The research team
Available online 12 December 2014
developed a system, called Mobile Diabetes Self-Care System, for people with diabetes, which facilitates
the patients to enhance their self-care ability and practices with the flexibility of timing, location and
Keywords:
choices. The study evaluates the system’s effectiveness in patients’ self-care knowledge, behavior and
Diabetes
efficacy. Twenty-eight patients with type 2 diabetes participated in the six-week intervention.
Self-care
Mobile interactive system Questionnaires were used to measure their changes before and after the intervention. The results
indicate that the mobile system enhanced the patient’s self-care knowledge and behavior by 17% and
22%, respectively, with statistical significance, yet only marginally increased their self-efficacy. The
majority of the participants have enjoyed using the system. Remarks on the biggest advantages of using
the system include portability, convenience in maintaining and accessing personal records, and
flexibility in learning necessary information. Integrating mobile technology with patient education and
support services bears great potential in combating the global burden of chronic diseases.
ß 2014 Elsevier B.V. All rights reserved.

1. Introduction global concern [6–9]. In addition to medical interventions,

government policies and healthcare programs have recommended
The prevalence of diabetes is increasing in most countries of educating patients to engage in self-care actively.
the world. The number of global diabetes population changes Explicit principles and appropriate strategies which enable
dramatically from 171 million in 2000 to 366 million in 2011, the patients to better manage their diabetes would be required.
according to the report of International Diabetes Federation [1]. The guideline on diabetes self-care education, AADE7TM, an-
The diabetes epidemic has largely exceeded previous prediction. nounced by American Association of Diabetes Education (AADE)
WHO had estimated the population to reach 366 million in 2030 and endorsed by the American Diabetic Association, contains
[2], yet a renewed estimation by International Diabetes Federation essential information for both diabetes educators and patients
pronounced that the population will reach 552 million by 2030 [10–12]. This guideline includes a set of daily activities such
[1]. In Asia, approximately one in ten adults suffers from such a as healthy eating, active exercising, monitoring health conditions,
disease [3]. Additionally, conditions accompanying to diabetes medication taking, problem-solving and risks reduction [11,13].
(e.g. kidney damage caused by diabetes) are frequent causes of Living up to the guideline to modify lifestyle requires many efforts,
poor quality of life and higher healthcare uses [4,5]. With the such as learning diabetes food exchange knowledge. Studies
expanded number of patients, improving diabetes care to relieve reported that patients did not adhere to recommendations rigidly
burdens caused by the disease is widely viewed as a significant [14,15]. Limited involvement in diabetes self-care is often caused
by lack of knowledge, low confidence and feelings of helplessness
[16]. Delays or failed access to needed care is another reason
Abbreviations: WHO, World Health Organization; AADE, American Association of [17,18]. Clinical settings play a central role in offering diabetic
Diabetes Education. services, but many of them over-emphasize on stabilizing
* Corresponding author at: 259 Wen-Hwa 1st Road, Kwei-Shan, Taoyuan, metabolic status through medical interventions rather than
Taiwan. Tel.: +886 3 2118800x5828; fax: +886 3 2118020.
patients’ self-care ability enhancement [7,19]. An effective tool
E-mail address: [email protected] (C.-Y. Lin).

http://dx.doi.org/10.1016/j.compind.2014.11.001
0166-3615/ß 2014 Elsevier B.V. All rights reserved.
 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29 23

or method to enhance patient’s self-care ability is required. Use of complication risk management, etc. The system also provides
information technologies would be helpful in reaching the quizzes that patients can self-test.
objective [20,21].  An interactive system which facilitates the patients to manage
Mobile technologies have matured to the point where their behavior
healthcare services for chronic disease could be provided beyond The system is designed to facilitate the patients in managing
hospital borders [22,23]. The usage of mobile-health (m-health) their diabetic relevant conditions in a less intrusive approach,
technologies has the potential to enhance a patient’s self-care where the patients can acquire relevant knowledge or actionable
ability, thereby modify their lifestyles and improve metabolic advises without time and geographical constraints. The users
conditions [23–28]. Despite of the fact that the benefits of could choose the path and set the speed of managing their own
m-health interventions for diabetic patients are well established, health with multiple options facilitated by the software system.
the impacts on diabetes self-care practices have not been The system acts as coach in facilitating the patients to reach their
extensively validated. Most m-health studies have emphasized own goals. It actively monitors individual health status, and sends
on assessing clinical metabolic outcome, such as decreased level of out reminders or appropriate advises timely. For example, as a
HbA1C [29], rather than validating self-care processes from the patient logs his/her blood glucose level using the mobile device
patient’s perspectives. Little is known about the effect of m-health routinely, the system helps monitoring the trend and changes in
on self-care processes [20,29–31]. Outcomes of self-care ability blood glucose level. When a patient’s blood glucose exceeds an
enhancement can be categorized as immediate (knowledge alarming threshold, the system automatically creates a reminder
and skills acquisition), intermediate (behavioral change) and long message and recommends plans for the patient to act on.
term manifestation (improved health status) [7,11,12]. Studies The system interacts with the users in a dynamic way.
have suggested adding focuses on identifying immediate and Options of exercise regimes are suggested pursuant to the
intermediate outcomes [7,29]. dietary intakes recorded by the patient, and an interface is
The purpose of this study is to assess the effects of the provided for logging exercise time for each chosen activity.
mobile-based interactive system, called Mobile Diabetes Self-Care  An active and responsive consultation system
System, developed by the research team. The effects are evaluated
Consultation services are provided by care managers at the
through three dimensions: (1) knowledge acquisition; (2) self-care
service center upon patients’ request. The database contains
behavior change, and (3) self-care efficacy.
longitudinal and comprehensive records that provide abundant
information for care managers to suggest behavioral adjust-
ments for patients. The patients may call or send messages to
2. The mobile support system
the service center for any questions, concerns, or distresses
encountered in self-care. The care managers provide not only
2.1. Mobile Diabetes Self-Care System design and development
instructions on knowledge and skills solicited by the patients,
but also psychosocial supports for patients’ reconciling
The research team developed the Mobile Diabetes Self-Care
diabetes conditions with daily lives. The care manager may
System in supporting diabetes patients for their daily care. The
be instructive in leading the patients navigating through the
system consists of a service center and a support application
possibly rocky journey of self-care management, but not in a
embedded in a mobile device. The processes of developing the
directive manner.
Mobile Diabetes Self-Care System included (1) analyzing require-
ments, (2) designing a service framework, and (3) developing
the Mobile Self-Care Support application. 2.2. Framework of the Mobile Diabetes Self-Care System
The multidisciplinary research team consists of endocrino-
logists, a senior nurse, a diabetes educator, and informational The three major features are accomplished through integrating
technology developers. The research team firstly gathered patient activities and care manager activities facilitated by the
information through extensive interviews with patients in a local mobile Diabetes Self-Care System. Fig. 1 shows the overall frame-
community and a tertiary hospital with the questions of ‘‘how do work of the system for individuals with diabetes.
you manage your diabetic conditions currently?’’, ‘‘what are the The tasks that the patients and care-managers perform are
barriers to managing the conditions as you would expect?’’, ‘‘what described as follows.
may help you better manage the disease?’’ Though not necessarily Patient-side tasks are facilitated through a Mobile Self-Care
a complete user story is collected for each user, the transcripts of Support application.
the interviews were carefully analyzed and discussed among team
members in developing the required functionality of the support  Acquiring self-care knowledge and skills related to diabetes care.
care system. Analysis of the interview transcripts revealed that  Recording personal activities relevant to diabetes care, such as,
the patients encountered the following issues in self-care: (1) dietary intake (types and amount of food or beverages), exercises
knowledge deficits or inaccurate information on diabetes care; (2) taken (type, strength, and duration of exercises), medicine intake
lack of easily accessible venues for disease management informa- (type, dosage and frequency of medications), foot care, and any
tion; (3) perceived difficulty in controlling diabetes. For example, relevant complications, on chosen devices (Internet connected
one female interviewee said, ‘‘I do not know how to control diabetes. smart-phone, tablet or desk-top computer).
I cannot eat any food with sugar in it. Nurses have given me diabetes  Relaying the above records through Internet to the service
information sheets, but I lose these information sheets.’’ Based on the center.
interview results, the research team developed a system that
characterizes three major features described below. Care manager-side tasks are facilitated by a service center.

 A diabetes self-care knowledge base  Monitoring metabolic variations as well as health conditions of
This knowledge base contains information as well as the patient and alarming the patient if necessary.
instructions on developing self-care skills. For example, indica-  Sending alarm messages as well as advises to patients whose
tion of blood glucose levels and modification factors, dietary conditions have crossed the individually set threshold.
information and intake advises, foot care importance and  Providing self-care consultation as well as psychosocial supports.
24 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29

Fig. 1. Framework for the Mobile Diabetes Self-Care System.

The knowledge base and behavioral change features of collected prior to the intervention and again post the six-week
the system are segmented into seven self-care functions. The intervention.
contents of each function were based on the AADE7TM self-care
behavior, the relevant literature [11] and guidelines by Taiwan 3.1. Subjects
Association of Diabetes Educators (www.tade.org.tw). These
functions are (1) Physical information management, (2) Diet Thirty patients with type 2 diabetes were recruited from the
management and Calorie calculation practice, (3) Exercise outpatient department of endocrinology and metabolism at three
management, (4) Medication management, (5) Foot management, tertiary-care hospitals in Taiwan. The size of the test group is
(6) Risk management, and (7) Emergence support. The diabetes limited due to available mobile devices embedded with the
educator and the nurse in the research team analyzed and developed system, blood pressure meters, and glucometers. Ten
organized results from the system requirement analyses through patients were recruited from each outpatient department. Poten-
use cases, flowchart and interface blueprints for technical tial subjects referred by endocrinologists were screened by the
developers to embed these functions in the application software researchers to determine their eligibility. The inclusion criteria are
for a mobile device. (1) 20 years of age or older; (2) reading and writing ability; and (3)
Use case is a highly effective method for discovering user mobile device familiarity. Subjects were excluded if they had
requirements so that the system response can be outlined for a severe vision or dexterity problems or if they had an alcohol or
precise sequence of user actions. Fig. 2 shows an example of the drug abuse problems. Eligible subjects were then asked if they
designed use cases where one patient learned diet management were willing to participate in the project, and agree to undergo
while interacting with the Diet Management function. Table 1 blood glucose tests and use the Mobile Self-Care Support
presents annotations from use case diagrams in Fig. 2. The diet application. A researcher then called the recruited patients to
plans for diabetic persons are frequently characterized by more schedule a research orientation visit, which includes explanations
frequent meals in lesser portions. Sometimes, light meals or of the study objectives, process and operation of the mobile
‘‘snacks’’ are even recommended as necessary. The value of the Diabetes Self-Care System. Written consent was obtained from
system, for this particular example, is to provide handy informa- the subjects before enrollment in the study. Of the 30 diabetic
tion on assortments of food exchange for users’ reference when patients recruited as subjects, 28 completed the study (93.3%).
hunger or desire strikes, and help logging/calculating nutrients One patient dropped out for personal reasons, and one patient
and calories intake daily over time. The goal is to help the user failed to complete the questionnaires.
being consciously aware of his/her own health status and available
choices while still enjoy the quality of life. The prototype of the 3.2. Procedure
Mobile Self-Care Support application was revised ten times based
on the comments given by the diabetes educator and the senior The subjects were required to complete a questionnaire on (1)
nurse, and then tested by three patients with diabetes before basic demographic and diabetes characteristics and (2) knowledge,
the study was conducted. behavior and efficacy on diabetes self-care at the beginning. All
subjects received at least 30 min face-to-face orientation in using
3. Evaluation method the mobile device embedded with the Mobile Diabetes Self-Care
System and a set of instruction manuals for using the Mobile Self-
A pre- and post-test design was used to evaluate the Care Support application, blood pressure meters, and glucometers.
effectiveness of the Mobile Diabetes Self-Care System in improving All subjects were free to use the self-care support system and
patient knowledge, behavior and efficacy. Baseline data were equipment during the study period but were required to return the
 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29 25

Fig. 2. Use case diagram of the diet management function.

equipment upon completion of the project. Subjects’ skills in using confidentiality of the data, password is required to access the
the self-care support system were verified through their return Mobile Self-Care Support Application.
demonstration after completing the training program. One week
after the training program, care managers followed up with each 3.3. Measurements
subject by phone to confirm their proper utilization.
After the six-week intervention, the subjects met with the Questionnaires of diabetes self-care knowledge [32], behavior
investigators and completed the post-test questionnaires, returned [33,34] and efficacy [33,34] are drawn from indexes of AADE
the devices, and received complementary diabetes self-care outcome system [11,13]. All three questionnaires have been
handbooks. To protect privacy of the subjects and to maintain validated in previous researches [32–34]. Each of the three

Table 1
An example of user interacting with the diet management function.

User task Learn diet management

User task Learn the skill and knowledge of diet management. Understand a user’s BMI (Body Mass Index) and ideal daily calorie intake as
description recommended by the mobile self-care support application (Mobile SC Support App)

Actors 1. The user

2. Mobile SC Support App
3. Service Center

Pre-conditions 1. The user may have stored his/her body height (BH), body weight (BW) and daily activity (DA) type data, or input the data later
2. Mobile device is in operation

Process 1. The user opens the main page of Mobile SC Support App
2. Mobile SC Support App shows function menu in the screen
3. The user selects ‘‘Diet Management’’ function from the menu
4. The user selects ‘‘Ideal Daily Calories Need’’ function from the menu
5. Mobile SC Support App shows his/her BH, BW, and DA type data
6. The user confirms/inputs the BH, BW, and DA type data. If the user’s BMI went beyond ideal range, a message will be sent to the
Service Center
7. Mobile SC Support App displays the user’s current BMI and ideal BW range. Mobile SC Support App displays the user’s ideal calorie intake
amount and Diet Plans options for different daily calorie intake. If the user’s BMI went beyond ideal range, a dialog will display a reminder
and a link to the knowledge database on the importance of weight control and diabetes
8. The user selects ‘‘Diet Plan’’ based on his/her chosen calorie intake

Post-conditions If the user modifies any data of BH, BW, DA type, and Diet Plan, then the Mobile SC Support App updates the data at both mobile device
and Service Center

Non-function 1. Security: The user must input correct password in order to access and activate all functions
2. Fault Tolerance: Mobile SC Support App will respond error message when the user inputs invalid data
26 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29

questionnaires contains six domains (1) generic health, (2) blood Table 2
Demographic characteristics of the subjects.
glucose monitoring, (3) healthy eating, (4) physical activities, (5)
foot care and complication risks, and (6) medication taking, which Characteristics Frequency Percentage
are organized around the AADE7TM standard [12,35]. (n = 28) (%)

Gender
 Knowledge: There are 33 items assessing diabetes care Male 17 60.7
knowledge [32]. For example, ‘‘Diabetes related symptoms such Female 11 39.3
Age (mean  SD) (45.4  12.4)
as sweating is a sign of low blood glucose,’’ with possible
<40 years 11 39.3
selections of: (1) Yes, (2) No, or (3) I don’t know. Each correct >=40 years 17 60.7
answer is scored as 1 point, and each incorrect or unknown Job
answer is scored as zero. The correct answers are summed to Employed 24 85.7
Not employed 4 14.3
obtain a total score ranging from 0 to 33, where a higher score
Education level
indicates better knowledge of diabetes care. Elementary school 7 25.0
 Behavior: There are 36 items assessing the degree to which the Senior high school 5 17.9
patient follows the recommended self-care activities [33,34]. For College or university 9 32.1
example, subjects are asked how frequently they comply with Master or PhD 7 25.0
Duration of type 2 diabetes (mean  SD) (6.77  7.53)
the recommended daily diet in a typical week. Behavior is
<2 year 7 25.0
measured on a five-point ordinal scale with ‘‘0 = never’’, ‘‘1 = 1–3 2–9years 15 53.6
times per week’’, ‘‘2 = 4–5 times per week’’, ‘‘3 = more than 5 >=10 years 6 21.4
times per week’’, to ‘‘4 = always’’. A higher score indicates more BMI
Normal weight = 18.5–23.9 9 32.2
frequent self-care behavior.
Overweight = 24–26.9 6 21.4
 Efficacy: Subjects are asked to evaluate how confident they are in Obesity = 27 or greater 13 46.4
performing the 36 items listed under the self-care behavioral
activities [33,34]. The responses are given on a five-point Likert
scale from ‘‘0 = not at all’’ to ‘‘4 = very confident’’. For example,
subjects are asked to rate their confidence in taking care of their difference between pre-test and post-test score is not statistically
complications. A higher score indicates better efficacy in self- significant, however.
care practice. The study further analyzes the six domains of the measures by
comparing changes in median scores from pre-to-post interven-
The content validity of these scales is established by the tion (see Table 4). The six domains are (1) generic health, (2) blood
expert review process. The reliability of the scales is measured glucose (BG) monitoring, (3) healthy eating, (4) physical activities,
by (1) The Kuder-Richardson test score for knowledge, which is (5) medication taking, and (6) foot care with complication risks
0.82, and (2) Cronbach’s a for the behavior and efficacy scales, care. In the knowledge measure, the percentages of improvement
which is 0.94, and 0.95, respectively. This means the scales are in these 6 domains range from 0% to 50%. Medication-taking
of good reliability and measure knowledge level, behavior, domain exhibits the highest percentage of improvement (50%),
efficacy consistently. followed by physical activities domain (33%) and healthy eating
domain (30%). But, the results indicate that there are no
3.4. Data analysis improvements in the generic health domain and foot care domain.
Overall, the improvement in the knowledge domains is 16%. In the
Given the small number of subjects for this study, the research behavior measure, healthy eating domain exhibits the highest
team still utilizes statistical analyses in the intention to objectively percentage of improvement (41%), while generic health domain
interpret the observed increases in participants’ measured and foot care domain exhibit the least improvement (8%). Overall,
knowledge, behavior and self-efficacy scores. Demographic data the improvement in the behavior domains is 26%. The improve-
are presented as proportions or as means  standard deviations. A ments observed in the six self-care domains from pre- to post-
paired t-test is used to compare the pre- and post-test of knowledge, intervention in knowledge and behavior measures are graphically
behavior and efficacy scores. All tests are two-tailed tests. Statistical presented by Radar-Chart in Figs. 3 and 4, respectively.
analyses are performed using SPSS version 17 for Windows.
5. Discussion
4. Results
The proposed Mobile Diabetes Self-Care System constitutes a
Table 2 presents the demographic profile of the 28 subjects. The service center (care manager-side) and the Mobile Self-Care
mean age is 45.4 years old, and 60.7% of them are male. Support application (patient-side). The results indicate, that it is
Approximately 67.8% of the subjects were either overweight or feasible to implement the Mobile Diabetes Self-Care System with
obese as indicated by their Body Mass Index (BMI). Mean duration positive effects on self-care knowledge and behavior. The efforts of
of their disease history with type 2 diabetes is 6.77 years. the requirements analysis aim to facilitate developers in building a
Table 3 presents the diabetes self-care knowledge, behavior
and efficacy scores before and after the intervention, as well
Table 3
changes by the participants. The average score of self-care Comparative change in self-care knowledge, behavior, and efficacy.
knowledge increases from 21.2 at pre-intervention to 24.7 at
post-intervention, an increase of 17%, with statistical significance. Variable Pre-intervention Post-intervention Change p value

The second measure is self-reported change in self-care behavior. Mean SD Median Mean SD Median Mean SD
The average score is 71.0 at pre-intervention, and increases to 86.7 Knowledge 21.2 5.2 21.5 24.7 3.1 25.0 3.5 3.5 <.001
at post-intervention, an increase of 22%, with statistical signifi- Behavior 71.0 23.9 68.5 86.7 26.8 86.5 15.8 33.9 .015
cance. The third measure is efficacy in performing the daily Efficacy 94.0 24.5 97.0 96.6 27.2 101.5 2.6 23.0 .067
self-care activities. The mean score at pre-test is 94.0 (SD = 24.5); 1. p value less than 0.05 indicates significant difference.
while the mean post-test score increases to 96.6 (SD = 27.2). The 2. Comparison between columns values for mean match (paired t-test).
 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29 27

Table 4
Improvement in six domains of self-care knowledge and behavior measure.

Domains Knowledge Behavior

Pre-test Post-test Improvement (%) Pre-test Post-test Improvement (%)

Generic health 6.0 6.0 0 6.5 7.0 8

BG monitoring 3.0 3.5 17 15.5 19.0 23
Healthy eating 5.0 6.5 30 13.5 19.0 41
Physical activities 3.0 4.0 33 16.5 20.0 21
Medication taking 2.0 3.0 50 10.0 11.0 10
Foot care 3.0 3.0 0 12.0 13.0 8

system that could better integrate recommended diabetes self- has pointed out that patients who have been diagnosed with the
care knowledge and skills with daily activities. The objectives of disease long time prior to participation in an intervention program
the system are to reduce the gaps in self-care practices between may have acquired sufficient knowledge or developed strategies to
ideal (professional expectation) and reality (patient activities). At manage their conditions and benefit less from such professional
the end of study, the researchers received positive feedbacks from supports than does a novice patient [30]. Future studies may explore
many participants to an open-ended question: ‘‘what is the biggest the impacts of such interventions on patients with different
advantage of using the system?’’ The participants find the characteristics, such as history of the disease, baseline efficacy,
following characteristics attractive: (1) It is easy to carry and etc. Another issue concerns the sustainability of the observed
has access to personal health data (8 persons), (2) It is convenient changes, and potential techniques of making the changes viable.
for recording and querying personal heath data (5 persons), (3) It is Comparison of the six domains indicates that the highest
flexible to learn and increase self-care knowledge (2 persons), (4) It percentage improvement for the knowledge and the behaviors
is helpful in seeking medical services (1 person), and (5) the system measures is healthy-eating and medication-taking domain respec-
is good in general (2 persons). Whereas, some of the subjects tively. Other than medication, diet management is critical in self-
suggest that complementary features such as bright graphic care for diabetes patients. But it may seem extremely perplexing for
illustrations and speech input functions are helpful. User com- patients to change their food choices and follow the most suitable
ments are valuable to direct further design works. calorie meal plan [38,39]. According to the principle of AADE7TM
The study finds that the system effectively improves patients’ [12], a patient with diabetes is expected to count calories regularly
self-care ability in knowledge and behavior, but only marginally in and acquire sufficient knowledge of using ‘‘food exchange list’’ (a list
efficacy. As observed in other diabetes intervention program developed by American Diabetes Association for diet substitution)
studies [27,36,37], the proposed intervention reinforces the self- in choosing the right foods with the right amounts. Increasing
care knowledge and behavior of the participants. The proposed access to diabetes care information through mobile health
self-care system includes education modules to enable mobile applications can decrease barriers to managing diabetes.
learning of appropriate self-care knowledge. The self-care knowl- In building sustainable self-care practices, positive feedbacks
edge can be transformed into self-care behavior, which is critical in from behavioral changes and knowledge acquisition are essential
effective diabetes control [7,11,12]. The results indicate that in encouraging the patients to further engage in appropriate
patient efficacy did not significantly differ after the intervention. behaviors toward healthy lifestyle [11]. Setting attainable
The research subjects of the study are characterized with an objectives of self-care practices, even if the objectives are far from
average of 6.77-year experience with diabetes. During the period of ideal metabolic control, fosters the patients with a sense of success,
time, the patients may have gradually developed an attitude competence, and engagement that drive them acting toward better
toward their efficacy of self-management. In contrast, the impacts metabolic control as the objectives are advanced gradually [7]. The
of a short-term, experimental mobile system intervention may not Mobile Diabetes Self-Care System offers quizzes for the user to test
encompass such a strong impact as to change their efficacy. A if their concept on diabetes self-care is correct, and allows the
stronger effect may be obtained from novice patients or studies patients to review his/her records and trends of self-care activity
with a longer intervention period. Furthermore, a literature review as well as physiological outcomes (blood sugar, blood pressure,

Fig. 3. Improvements in six domains of knowledge measure. Fig. 4. Improvements in six domains of behavior measure.
28 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29

body mass index, etc.). The feedback offered by the Mobile References
Diabetes Self-Care System motivates the user to sustain current
[1] D.R. Whiting, et al., IDF diabetes atlas: global estimates of the prevalence of
success, or keep working toward higher objectives. diabetes for 2011 and 2030, Diabetes Research and Clinical Practice 94 (3) (2011)
Comprehensive approaches to diabetes management should be 311–321.
presented to the patient as a means of increasing the freedom in [2] S. Wild, et al., Global prevalence of diabetes estimates for the year 2000 and
projections for 2030, Diabetes Care 27 (5) (2004) 1047–1053.
their lives rather than simply intensifying clinical control (e.g., [3] Y.-M. Mu, et al., Managing diabetes in Asia: overcoming obstacles and the role of
metabolic index) [7]. The mobile health technologies offer flexible DPP-IV inhibitors, Diabetes Research and Clinical Practice 95 (2) (2012) 179–188.
venues to enable the patients in managing their disease conditions [4] K.A. McBrien, et al., Health care costs in people with diabetes and their association
with glycemic control and kidney function, Diabetes Care 36 (5) (2013)
and rectify knowledge deficiencies related to diabetes care [20]. The 1172–1180.
Mobile Diabetes Self-Care System offers multiple approaches with [5] S.W. Klarenbach, P. Jacobs, International comparison of health resource utilization
abundant options, such as diet plans, exercises regimes, for users to in subjects with diabetes an analysis of Canadian and American National Health
Surveys, Diabetes Care 26 (4) (2003) 1116–1122.
adopt in managing their diabetes conditions. In the future, the
[6] W.H. Herman, et al., The 10-year cost-effectiveness of lifestyle intervention or
development of a mobile system for people with chronic diseases metformin for diabetes prevention: an intent-to-treat analysis of the DPP/DPPOS,
must evolve into user-oriented thinking as patients and consumers Diabetes Care 35 (4) (2012) 723–730.
increase their health and technology literacy [28,40,41]. [7] H.A. Wolpert, B.J. Anderson, Management of diabetes: are doctors framing the
benefits from the wrong perspective? British Medical Journal 323 (7319) (2001)
Healthcare Information technologies are still evolving (e.g., 994–996.
Internet, smart phones, and mobile devices) and generating value- [8] A. Morden, C. Jinks, B.N. Ong, Rethinking ‘risk’ and self-management for chronic
added services that make complex tasks manageable, which make illness, Social Theory & Health 10 (2012) 78–99.
[9] A. Kennedy, A. Rogers, P. Bower, Support for self care for patients with chronic
lifestyle or behavior interventions for chronic disease care even disease, British Medical Journal 335 (7627) (2007) 968–970.
more effective. The service agents and health care providers are [10] American Medical Association Position Statement, Standards of medical care in
seeking optimal solutions for managing targeted health conditions, diabetes-2010, Diabetes Care 33 (Suppl. 1) (2010) S11–S61.
[11] L. Haas, et al., National standards for diabetes self-management education and
i.e., identifying technologies that can assist individuals with support, Diabetes Care 35 (11) (2012) 2393–2401.
specific disease in promoting their own health, or motivate people [12] American Association of Diabetes Educators, AADE Guidelines for the Practice of
with the capability to perform certain health tasks. Diabetes Self-Management Education and Training (DSME/T), American Associ-
ation of Diabetes Educators, Chicago, IL, 2009.
[13] D. Kent, et al., Healthy coping: issues and implications in diabetes education and
care, Population Health Management 13 (5) (2010) 227–233.
6. Limitations [14] T. Alcorn, Y. Ouyang, Diabetes saps health and wealth from China’s rise, Lancet
379 (9833) (2012) 2227–2228.
Despite its contributions, this study has certain limitations. [15] M.M. Funnell, R.M. Anderson, The problem with compliance in diabetes, Journal
of the American Medical Association 284 (13) (2000) 1709.
Firstly, the subjects of the study constitutes a small group of [16] D.F. Blackburn, J. Swidrovich, M. Lemstra, Non-adherence in type 2 diabetes:
patients with type 2 diabetes without a control group, which made practical considerations for interpreting the literature, Patient Preference and
it impossible to analyze or rule out other impacts on the observed Adherence 7 (2013) 183.
[17] Diabetes UK, Diabetes UK survey of people with diabetes and access to healthcare
effects. Secondly, self-reported behaviors may not represent actual
services 2009, vol. 1, 2009, pp. 1–16, http://www.diabetes.org.uk/Documents/
behavior. However, the progress of patients in meeting behavioral Reports/Members%27_survey_report_22.
goals in this study is consistent with previous studies [29]. Thirdly, [18] A. Ghorob, T. Bodenheimer, Sharing the care to improve access to primary care,
the subjects of this study have higher education level than that of New England Journal of Medicine 366 (21) (2012) 1955–1957.
[19] R. Jansink, et al., Primary care nurses struggle with lifestyle counseling in diabetes
people of similar ages in Taiwan. The impacts of a mobile system on care: a qualitative analysis, BMC Family Practice 11 (1) (2010) 41.
ordinary educational level patients may need to be further verified. [20] S. Tani, et al., Development of a health management support system for patients
with diabetes mellitus at home, Journal of Medical Systems 34 (3) (2010)
223–228.
7. Conclusions [21] D.K. King, et al., What patients want: relevant health information technology for
diabetes self-management, Health and Technology 2 (3) (2012) 147–157.
[22] S.J. Lee, J. Kim, M. Lee, The design of the m-health service application using a
This study demonstrates that the Mobile Diabetes Self-Care Nintendo DS game console, Telemedicine and e-Health 17 (2) (2011) 124–130.
System is effective in improving diabetes patients’ self-care [23] Y. Lee, et al., Mobile personal health care system for patients with diabetes, in:
knowledge and facilitating their behavior changes. The study also Aging Friendly Technology for Health and Independence, Springer, Berlin/Heidel-
berg, 2010, pp. 94–101.
demonstrates that education and support to diabetes patients for [24] J. Baron, S. Newman, A systematic review of the effectiveness of mobile health
self-care through mobile devices, with feedbacks and freedom of interventions for the management of diabetes, in: 1st International Congress on
choices, would be a feasible and acceptable approach for patients Telehealth and Telecare, London, UK, International Journal of Integrated Care 11
(6) 2011.
with limited access to professional venues of care. [25] T. Norris, R. Stockdale, S. Sharma, Mobile health: strategy and sustainability,
Interests in mobile-based healthcare research for diabetes care Journal of Information Technology in Healthcare 6 (5) (2008) 326–333.
have lasted for years; many lessons are still to be learned. For [26] H.J. Lee, et al., Ubiquitous healthcare service using Zigbee and mobile phone for
elderly patients, International Journal of Medical Informatics 78 (3) (2009) 193–198.
example, future work is needed to evaluate long-term effect of self-
[27] T. Chomutare, et al., Features of mobile diabetes applications: review of the
care systems on different dimensions of diabetes care, to follow up literature and analysis of current applications compared against evidence-based
participants more completely, and to analyze patients with guidelines, Journal of Medical Internet Research 13 (2) (2011) e65.
different characteristics, such as those who have poor control [28] M. Gupta, et al., Health@ home – an e-service model for disease prevention and
healthcare in the home, Electronic Healthcare 1 (2009) 17–24.
over their disease. Furthermore, researchers could consider [29] M. Peeples, in: K. Weinger, C.A. Carver (Eds.), Educating Your Patient with
integrating health technologies with professional support in Diabetes, Contemporary Diabetes, ‘‘Measuring Diabetes Outcomes and Measuring
intervention development and exploring the possibility of Behavior Educating Your Patient with Diabetes’’, Humana Press, New York, 2009,
pp. 159–175.
decreasing health service utilizations. [30] B.M. Costa, et al., Effectiveness of IT-based diabetes management interventions: a
review of the literature, BMC Family Practice 10 (1) (2009) 72.
Acknowledgments [31] S.I. Kim, H.S. Kim, Effectiveness of mobile and Internet intervention in patients
with obese type 2 diabetes, International Journal of Medical Informatics 77 (6)
(2008) 399–404.
The authors would like to thank the anonymous reviewers for [32] Y.F. Chong, H.D. Lin, The effects of health education on knowledge attitude and
their valuable comments and suggestions to improve the quality of practice in seventy-four type 2 diabetic patients, Endocrinology & Diabetology 13
(3) (2000) 1–17.
the paper. This work was supported by research grants from
[33] M.L. Parchman, et al., Glucose control self-care behaviors, and the presence of
Chang Gung University (CGURP UARPD3A0151) and Chang Gung the chronic care model in primary care clinics, Diabetes Care 30 (11) (2007)
Memorial Hospital (CMRPD-390011). 2849–2854.
 S.H.-M. Guo et al. / Computers in Industry 69 (2015) 22–29 29

[34] A.C. Hurley, C.A. Shea, Self-efficacy: strategy for enhancing diabetes self-care, in nursing practice including nursing care and clinical
Diabetes Educator 18 (2) (1992) 146. instructor.
[35] M.M. Funnell, et al., National standards for diabetes self-management education,
Diabetes Care 32 (Suppl. 1) (2009) S87.
[36] S.C. Wangberg, An Internet-based diabetes self-care intervention tailored to self- Her-Kun Chang is a joint professor with Graduate
efficacy, Health Education Research 23 (1) (2008) 170. Institute of Business and Management, and Department
[37] K. Knight, et al., A systematic review of diabetes disease management programs, of Information Management at Chang Gung University,
American Journal of Managed Care 11 (4) (2005) 242–250. Taiwan. He received his BS and PhD degrees in Computer
[38] E. West, et al., Focus groups show need for diabetes awareness education among and Information Science from National Chiao Tung
African Americans, California Agriculture 56 (4) (2002) 139–143. University, Taiwan, in 1989 and 1994, respectively.
[39] E. Lee, et al., Review of mobile terminal-based tools for diabetes diet management, His current research interests include health information
Studies in Health Technology and Informatics 169 (2011) 23. management, distributed information processing, data
[40] C. Pagliari, Design and evaluation in eHealth: challenges and implications for an mining, and mobile technology enabled services.
interdisciplinary field, Journal of Medical Internet Research 9 (2) (2007) e9.
[41] J.E. Van Gemert-Pijnen, et al., A holistic framework to improve the uptake and
impact of eHealth Technologies, Journal of Medical Internet Research 13 (4)
(2011) e111.

Sophie Huey-Ming Guo is an assistant professor in Chun-Yi Lin is an assistant professor in Dept. of
Department of Nursing, Mackay Medical College in Information Management at Chang Gung University
Taiwan; she teaches courses in the area of nursing in Taiwan. He receives his PhD in Industrial Engineering
related and nursing informatics in the college. Her and Operations Research from University of California
specialties are health informatics and nursing (diabetes at Berkeley in the United States. His current research
care, pediatric nursing). Sophie holds a M.S. degree in interests include health information management,
Nursing Informatics and a PhD degree in Management quality management, production management and
from Chung Gung University. She also has experience data mining.