Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

Accessibility of mHealth Self-Care Apps

for Individuals with Spina Bifida

by Daihua X. Yu, PhD; Bambang Parmanto, PhD; Brad E. Dicianno, MD;

and Gede Pramana, MS

Abstract
As the smartphone becomes ubiquitous, mobile health is becoming a viable technology to empower
individuals to engage in preventive self-care. An innovative mobile health system called iMHere (Internet
Mobile Health and Rehabilitation) has been developed at the University of Pittsburgh to support self-care
and adherence to self-care regimens for individuals with spina bifida and other complex conditions who
are vulnerable to secondary complications. The goal of this study was to explore the accessibility of
iMHere apps for individuals with spina bifida. Six participants were asked to perform tasks in a lab
environment. Though all of the participants were satisfied with the iMHere apps and would use them
again in the future, their needs and preferences to access and use iMHere apps differed. Personalization
that provides the ability for a participant to modify the appearance of content, such as the size of the icons
and the color of text, could be an ideal solution to address potential issues and barriers to accessibility.
The importance of personalization—and potential strategies—for accessibility are discussed.
Keywords: mobile health, accessibility, smartphone apps, wellness, self-care

Introduction
Mobile health (mHealth) initiatives are those that use mobile communications for health services and
information.1 They involve the use of mobile devices to wirelessly connect remote and highly itinerant
populations directly with healthcare systems.2 The mHealth approach has been described as a patient-
centered approach to care3 and has been popularly utilized to deliver medical reminders, provide
treatment support, and collect data4 for healthcare delivery and research.5, 6
More than 50 percent of all Americans have at least one chronic illness,7 and about one-fourth of
people with chronic conditions have a disability that limits one or more daily activities.8 Strong evidence
supports the importance of self-management skills for improved health outcomes and independence in
daily living activities for persons with disabilities.9, 10 One such population is individuals with spina bifida
(SB). SB is the most common permanently disabling birth defect in the United States.11
Individuals with SB are vulnerable to secondary complications (e.g., urinary tract infections [UTIs]
and skin breakdown)12 that can in part be prevented with self-care activities. A wellness pilot project was
developed at the Spina Bifida Association of Western Pennsylvania (SBAWP) in which two clinicians—
wellness coordinators (WCs)—supervised the care of 35 individuals with complex medical needs. The
role of the WCs as a liaison and director of care empowered the individuals to be responsible for their
own treatment. Results show that the individuals in the wellness program had shorter lengths of hospital
stays, with admission rates of only 12.9 percent compared to the national rate of 26.9 percent.13, 14 The
2 Perspectives in Health Information Management, Spring 2015

participants also had lower rates of skin breakdown (9.7 percent) and UTIs (16.1 percent) compared to
those with SB in the general population (35.5 percent and 35.2 percent respectively).
The success of the in-person SB pilot project15 in promoting wellness as well as achievements in other
wellness programs16–20 provides evidence that the goal of improving health and function through self-
management is attainable with appropriate support. However, this in-person wellness program also had
significant limitations in the number of persons that could be served due to geographic and clinical-
resource constraints. The use of mobile technology in healthcare, as with mHealth, is one way to reduce
these constraints. An mHealth system would allow WCs to serve a larger number of individuals, making
the wellness program both cost-effective and scalable. Using mHealth technologies might also improve
health outcomes of individuals by reducing secondary complications and help to reduce the cost of care
for people with chronic conditions, which account for three-quarters of healthcare expenditures in the
United States.21–23
An innovative system called iMHere (Internet Mobile Health and Rehabilitation) was developed at
the University of Pittsburgh to enable self-care for individuals with SB. As shown in Figure 1, the iMHere
system consists of smartphone apps, a clinician portal, and two-way communication connecting the two
to support self-management and service delivery.24 A web-based portal provides clinicians with the ability
to monitor individuals’ conditions and send treatment plans to individuals via smartphones. The iMHere
apps on smartphones allow individuals to set up reminders according to their preferences, respond to
alerts, and report symptoms. (See Figure 1.)
A previous study25 focused on the usability of iMHere. Three phases of usability testing were
conducted to evaluate the self-care workflow, general issues with navigation and the user interface, and
the reliability of communication between the apps and the portal.
Accessibility is a subset of usability and is the degree to which a person can use a product regardless
of ability or disability. As applied to mHealth, accessibility refers to the extent to which participants have
access to the on-screen information presented from a sensory, motor, or cognitive perspective. The aim of
the present study was to explore the accessibility of the iMHere apps, focusing on the user interface and
navigation, which might affect users’ performance and satisfaction with the system.

Methods
Six participants were recruited at the University of Pittsburgh and provided informed consent for this
study under an Institutional Review Board–approved protocol. The choice of sample size was based on
previous studies that found that 80 percent of usability problems can be revealed with only five
participants,26–28 with almost all high-severity usability problems being uncovered with only three
participants.29 In general, the estimated required sample size for a usability test depends on the problem
space.30 Since the issues related to overall usability were addressed in a different study,31 the sample size
of six participants in this descriptive study can be considered sufficient for discovering accessibility
problems of the user interface and navigation.
Inclusion criteria were (1) age 18 to 40 years, (2) diagnosis of SB and hydrocephalus, and (3) ability
to use a smartphone. Exclusion criteria were severe intellectual disability and any problem in vision,
hearing, speech, or finger movement that precluded use of the phone. Both experienced users of iMHere
who participated in the previous study and inexperienced users were included. Experienced users had
stopped using iMHere apps for more than five weeks before participating in this study to minimize the
learning effects that might carry over from the previous experiences.
Five apps that constitute the iMHere gallery were released to support preventive self-care for
managing medications (MyMeds), neurogenic bladder (TeleCath) and bowel (BMQs), mood (Mood), and
skin breakdown (SkinCare). Self-created reminders with customized alarm tones and messages prompted
individuals to perform tasks related to self-care at home. The following tasks were selected on the basis of
the complexity and types of daily activities for self-care and were randomly assigned to each participant:
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

1. Schedule a new medication: participants had to locate a correct medication, add more information
about this regimen such as their reason for taking the medication, and set up a reminder.
2. Report a skin breakdown issue: participants were required to respond to reminders, take a picture,
and fill out a form describing the affected skin including location, color, size, depth, and tissue
condition.
3. Set a schedule to monitor mood: participants were required to set a repeating alert time, such as
once per week or biweekly, and enter a valid start date for the alert.
4. Respond to a mood alert: participants had to answer 10 binary questions (yes or no) to express
their feelings.
5. Respond to a TeleCath once-per-day alert: participants had to report their number of self-
catheterizations performed.
6. Respond to a BMQ alert: participants had to report their adherence to a bowel program and report
any symptoms.

Each of the six tasks was repeated three times within two hours for a total number of data points (n)
of 108 (six participants, six tasks, repeated three times). Android devices with keyboards that participants
could physically slide out of the smartphone were used in this study. Researchers manually recorded the
time and steps to complete each task and the number of mistakes made by each participant for statistical
analysis.
Researchers also used step-by-step observation notes to record the verbal and nonverbal behaviors
and frustrations of participants in the lab tests. Weighted scores were added to all mistakes to describe the
difficulty-on-performance (DP) for participants to solve problems: a score of 1 was assigned if the
participant could solve the problem without any help; 2, if the participant needed help consisting of one
sentence; 3, if the participant needed help consisting of two to four sentences; and 4, if the participant was
unable to solve the problem. The DP score was calculated as the sum of weighted scores divided by the
total number of steps to complete a task.
After the tasks, participants were asked to complete the modified Telehealth Usability Questionnaire
(TUQ) to reveal their levels of satisfaction with the iMHere apps. TUQ is a comprehensive usability
questionnaire that covers six usability factors including usefulness, ease of use and learnability, interface
quality, interaction quality, reliability, and satisfaction and future use.32 The TUQ utilizes a seven-point
Likert scale to measure usability, with the value of 1 for least usable and 7 for most usable.
Standard deviation was calculated to reveal the dispersion patterns of the previously mentioned
variables. A paired t-test was utilized to explore the difference in the time for completing all tasks
between different testing scenarios (tasks 1 and 2, 2 and 3, or 1 and 3). Person’s correlation coefficient
and Spearman’s rank correlation coefficient were used to measure an association between the previously
mentioned variables.

Results
Six participants (five males and one female) completed this study. Their ages ranged from 23 to 36
years with an average of 29 years (SD = 5.09). No participants were excluded on the basis of the
exclusion criteria. Except for participant 4, all others were cell phone users prior to participating in the
study. Participants 4 and 5 were new to iMHere. Participants 1, 2, and 3 had tested the MyMeds app in
the previous study more than five weeks earlier. These participants remembered about 25 percent of the
processes in the MyMeds app. Participant 6 had tested the MyMeds and SkinCare apps about six months
before, but he had completely forgotten how to use these apps.
Different methods were utilized to evaluate the previously mentioned data instruments. The results
were analyzed from the perspectives of usability that include effectiveness, efficiency, and satisfaction.33

Effectiveness of Task Completion in Three Tests

In total, 108 tasks were executed by the six participants. Figure 2 illustrates the average time for all
participants to complete the tasks. Generally, they spent the most time on scheduling a medication (127
4 Perspectives in Health Information Management, Spring 2015

seconds, about 38 percent of the total time) and responding to a skincare alert (96 seconds, about 29
percent). About 33 percent of the total time was used to complete the other four tasks: responding to a
BMQ alert (36 seconds, about 11 percent), responding to a TeleCath once-per-day alert (6 seconds, about
2 percent), scheduling a mood alert (44 seconds, about 13 percent), and responding to a mood alert (26
seconds, about 8 percent). (See Figure 2.)
According to Pearson’s correlation coefficient, a slightly negative correlation was revealed between
the order of tests (tests 1, 2, and 3) and the completion time, r = −0.165, n = 108, p = 0.04. A significant
positive correlation was found between time and the steps to complete tasks, r = 0.635, n = 108, p <
0.001. A paired t-test with a 95 percent confidence interval (n = 108) revealed a significant difference in
the time for completing all tasks in test 1 (M = 71.03, SD = 75.17) and test 2 (M = 49.14, SD = 46.76)
conditions, t(35) = 2.88, p = 0.007, as well as for test 1 (M = 71.03, SD = 75.17) and test 3 (M = 47.64,
SD = 46.78) conditions, t(35) = 3.715, p = 0.001. However, no significant difference was identified for
test 2 (M = 49.14, SD = 46.76) and test 3 (M = 47.64, SD = 46.78) conditions, t(35) = 0.34, p = 0.74. (See
Table 1.)
As mentioned previously, both experienced and inexperienced users were included in this study. The
overall average time in seconds for each group of participants to complete tasks are shown in Table 1.
According to a paired t-test, no significant difference was identified between the experienced (M = 57.67,
SD = 46.86) and inexperienced participants (M = 47.5, SD = 41.27) at the p < 0.05 level, t(5) = 2.33, p =
0.067. This finding suggests that the experienced participants might not benefit in terms of time efficiency
from their prior experience with iMHere after a five-week washout period.

Efficiency of Performance
As shown in Table 2, 60 steps on average were required for a participant to complete six tasks in each
test. According to Pearson’s correlation coefficient, significant positive correlations were identified
between the following variables at the p < 0.05 levels:

1. The steps to complete tasks and the mistakes encountered by participants, r = 0.40, n =
108, p < 0.001;
2. The steps and the time to complete a task, r = 0.64, n = 108, p < 0.001; and
3. The time to complete a task and the mistakes encountered by participants, r = 0.715, n =
108, p < 0.001.

(See Table 2.)

Because of the tasks’ complexity, more steps on average were required for participants to complete
four tasks. First, adding a new medication required participants to locate a correct medication, add more
information about this regimen such as their reason for taking the medication, and set up a reminder (17
steps). Second, responding to a skincare alert required participants to respond to a reminder, take a
picture, and fill out a form describing the affected skin (11 steps). Third, when scheduling a mood alert,
participants were required to set an alert time (12 steps), such as once per week or biweekly, and enter a
valid start date for the alert. Finally, when responding to a mood alert, participants also had a relatively
high average number of 12 steps to complete a binary mood survey with answers of “Yes” or “No.”
Twenty-eight mistakes were identified throughout the participants’ tests. Sixteen mistakes (57
percent) were found when participants tried to schedule a medication. Ten mistakes (36 percent) were
associated with the task of recording a skin issue. The remaining two mistakes (7 percent) were related to
scheduling a mood alert. The overall error rate was 2.7 percent, which signifies five mistakes in the 181
steps that the participants averaged to complete all six tests. Higher error rate were identified in the tasks
of scheduling a medication and recording a skin issue.
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

As shown in Table 3, 14 mistakes (50 percent of 28) were found in test 1. In test 2, there were 12
mistakes, and only 2 mistakes occurred in test 3 (respectively, 43 percent and 7 percent). The mistakes
could be categorized into the following three groups:

• Type A: Four mistakes (14 percent) were associated with participants’ understanding of
words. For instance, participants 1, 2, and 4 had difficulty understanding the meanings of
the words “alias” and “once per week” on the MyMeds app and/or the meaning of “start
date” on the Mood app.
• Type B: Twenty-one mistakes (75 percent) were associated with participants’ familiarity
with the activity flows of the apps. Problems were identified in the following activities:
o Participants experienced problems in locating the plus symbol for adding a new
medication schedule (participant 3) or adding a new skincare report (participants 2,
3, 4, 5, and 6).
o Participants 1 and 6 saved a medication without scheduling an alert.
o Participants 3 and 6 encountered a problem in adding a new skin record.
o Participant 1 saved a skin problem without completing the survey.
o Participants 1, 2, 5, and 6 forgot to save the alias and notes for completing the task
to schedule a new medication.
• Type C: The remaining three mistakes (11 percent) were associated with participants’
interactions with the apps:
o Participant 3 did not notice that he had to click on a medication to select or expand
for a detailed list.
o Participant 1 clicked the save button twice, which caused a run-time system error
displaying the message “iMHere application is not responding” in the task for
scheduling a Mood alert.

As previously mentioned, weighted scores were added to all mistakes to describe the difficulty-on-
performance (DP) preceding task completion. As shown in Table 3, with the exception of participant 1,
who received a two-sentence reminder to save the alias and notes (DP = 3), all other participants were
able to self-correct the mistakes without any help (DP = 1) or with one-sentence assistance (DP = 2).
Additionally, a decrease in the scores for DP was observed in test 1 (total DP = 27) to test 2 (total DP =
19) and test 3 (total DP = 2). Scheduling a new medication was the most complicated task, with an
overall DP score of 31 (the sum of DP scores from test 1, test 2, and test 3) (see Table 3). The task of
responding to a skincare alert to record a skin problem was the second most complicated task, with an
overall DP score of 14. The tasks of scheduling a mood alert (total DP = 3), responding to a mood alert
(total DP = 0), responding to a TeleCath once-per-day alert (total DP = 0), and responding to a BMQ
alert (total DP = 0) were easier for participants compared to the previously mentioned tasks.
About 93 percent (n = 26) of mistakes were prevented after test 2. In test 3, participants 2, 4, 5, and 6
were able to conduct all tasks without mistakes, and participants 1 and 3 were able to self-correct the
mistakes without any help (DP = 1). Twenty mistakes (about 70 percent) were able to be corrected by
participants after a one-sentence reminder from the researcher (DP = 2) in tests 1 and 2. Two mistakes (7
percent) required a two-sentence reminder for participants, and that occurred in test 1 only.
6 Perspectives in Health Information Management, Spring 2015

Users’ Satisfaction with iMHere Apps

The iMHere apps received a high average TUQ score of 6.52 out of 7 points (93.1 percent, SD =
0.58). As shown in Figure 3, the factors for reliability and usefulness obtained lower average scores at
6.08 and 6.22 (87 percent and 89 percent; see Figure 3). Two participants gave low scores of 3 on the
questions under usefulness because they still had concerns as to how effectively iMHere improves access
to healthcare services. Since all participants encountered at least two mistakes during the tests, three of
them gave relatively lower scores of 4 for the reliability of the system in giving error messages. (See
Figure 3.)
Table 4 shows the TUQ score obtained from each participant, his or her average time to complete the
tasks, and the number of mistakes preceding task completion. According to Spearman’s correlation, a
slightly negative relationship was identified between participants’ TUQ scores and the average time, with
r = −0.257 and n = 6, but this correlation was not significant, p > 0.05. An insignificant negative
correlation was also identified for TUQ scores and mistakes, r = −0.203, n = 6, p > 0.05. Additionally, a
positive correlation was identified for the average time to complete tasks and the mistakes encountered by
participants, r = 0.464, n = 6, p = 0.354, but this correlation was not significant at the p > 0.05 level. See
Table 5.
The five iMHere apps in this study were developed to encourage important self-care activities for
individuals with SB. Depending on participants’ individual medical and psychological needs for self-care,
their preferences for using iMHere apps might differ. As shown in Table 5, all participants were interested
in using the Skincare app because they all used wheelchairs and could easily develop pressure wounds
due to insensate areas of skin. Four out of six participants thought the MyMeds app for medication
management was important because they had up to 10 routine medications.
Participants 1 and 2 said staff members would always remind them about managing their bowel
movements, so the BMQ app might not be necessary for them. The mood survey for preventing
depression was commonly recommended on a biweekly or monthly basis; participants 3, 4, and 6 thought
the Mood app might not be as important as the other apps. Participants 4, 5, and 6 did not think the
TeleCath app would be very helpful considering that they do not have problems with neurogenic bladder.
Besides their preference of apps, the following quotes highlight participants’ feedback on the iMHere
apps:

Participant 1: “Like in skincare that uses color to separate body parts, if we have different colors
to separate each application, that will help me to know which app I am using now.”
Participant 2: “Light text color is a little bit hard for me to read.”
Participant 3: “That will be great if I can change the text color and/or size.”
Participant 5: “Some buttons are kind of small, a little bit bigger should be better.”
Participant 6: “Buttons might be small, I have a big finger.”

Conclusion
In general, the smartphone is an ideal tool for implementing wellness programs,34 but it poses
accessibility challenges because of the lack of screen space,35 small form factors,36, 37 and the high number
of steps to accomplish a task in an app.38 Previous studies have suggested that auditory feedback could be
used to enhance the accessibility of mobile phones.39–43 Sound feedback can be used to improve the
usability of buttons.44 Universally designed models featuring large font sizes can help visually impaired
users to have access to e-mail messages and mobile Internet sites.45
This study explored the accessibility of iMHere apps, focusing on the user interface and navigation.
Scores from the TUQ indicated that the iMHere apps were viewed positively (6.52 out of 7 points, 93
percent). All of the participants were satisfied with the iMHere apps and would use them again in the
future. Neither the longer average time to complete tasks nor the number of mistakes significantly
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

affected participants’ perception of iMHere usability (TUQ score). Participants’ actual experiences with
the apps might play a more important role in the overall usability and satisfaction. Since the lowest score
was received under the usability factor of reliability on the TUQ, the ease of noticing and recovering from
mistakes might have a negative impact on satisfaction levels. Shorter times to complete tasks and reduced
error rates were seen over repeated trials. Learning effects, specifically, might play an important role in
the continuity of training and the effective use of iMHere apps because participants’ experiences from
prior tests appeared to have carried over to the next test.
Several important findings from this study reveal ways to improve the accessibility of smartphone
apps:

1. Appropriate use of words: Although the iMHere apps were designed by clinicians with expertise
in the care of individuals with SB, 14 percent of mistakes were still associated with participants’
perception of words. Using simple and common words such as “the reason to take medication” to
replace the word “Alias” in the MyMeds app might be more effective to ensure the readability
and understandability of the text for participants, particularly those with cognitive impairments or
problems with reading comprehension.
2. Appropriate use of text style: Using a light text color such as white or yellow on a light
background (e.g., gray) is not recommended. Using contrasting colors between the text and
background and adding shadows to text may enhance the contrast and improve readability.
3. Use of in-app directional notes: Seventy-five percent of mistakes that were encountered by
participants were related to task procedures. For instance, participants forgot to click the plus sign
to add a new schedule, forgot to save data, or saved data without completing a survey. A short,
one-sentence reminder providing directional guidance might be useful to prevent these types of
mistakes.
4. Use of large target size: Small target size of icons or buttons is not only a problem for users with
dexterity impairments, but also an accessibility issue for people with large fingers or for those
who prefer to have larger icons. Using larger icons or buttons would improve physical access to
icons and buttons.
5. Use of thematic colors: Participants highlighted the usefulness of colors to indicate the status of
whether or not a medication is scheduled (green vs. red). The use of color to separate body parts
also helped participants to correctly specify the location of problem skin areas. Using color to
separate the apps would easily let users know which app they are using.
6. Use of personalized app lists: Not all five apps are useful for all participants. Providing the ability
for users to choose which apps they want to use might help increase user satisfaction.

One limitation of this study is the small sample size. Individuals with spina bifida have a broad range
of abilities. Because all the users were able to use a smartphone, and because prior work has shown that
three to five participants are sufficient to uncover most of the basic accessibility issues of a system, our
sample was likely sufficient to assess the accessibility challenges of a larger population. In fact, our study
shows that some common accessibility challenges can be solved by a few simple design changes such as
providing the ability to change the size of text and icons or providing in-app directional notes to prevent
errors, and we expect these changes to have applicability for a broad number of users with various
impairments. More work is planned to evaluate individuals who have more complex cognitive, sensory,
or motor impairments that make use of a smartphone difficult.
Moreover, personalization provides the ability for a participant to modify the appearance of screen
contents including size, color, and a list of apps based on his or her needs and to address concerns related
to accessibility. Because users’ needs for iMHere apps and their interactional preferences, including text
color and button size, vary by individual, personalization may enhance accessibility for people with and
without disabilities. For instance, a larger display text could benefit participants with visual impairment or
farsightedness; larger target size could benefit participants with dexterity issues or large fingers.
Designing and developing personalizable and fully accessible iMHere apps will be the next step in
improving the accessibility of these mHealth tools.
8 Perspectives in Health Information Management, Spring 2015

Acknowledgments

This research was funded in part by the National Institute on Disability and Rehabilitation Research
(NIDRR) grants no. H133E140039 (RERC on ICT Access) and no. H133A140005 (Self-management
using mHealth), and by the National Institute of Health (NIH) grant no. 1R21HD071810-01A1. The
views herein reflect only those of the authors and not of any funding agency or any other entity.

Daihua X. Yu, PhD, is a research fellow in the Department of Health Information Management at the
University of Pittsburgh School of Health and Rehabilitation Sciences in Pittsburgh, PA.
Bambang Parmanto, PhD, is a professor in the Department of Health Information Management at the
University of Pittsburgh School of Health and Rehabilitation Sciences in Pittsburgh, PA.
Brad E. Dicianno, MD, is an associate professor in the Department of Physical Medicine and
Rehabilitation at the University of Pittsburgh Medical Center in Pittsburgh, PA.
Gede Pramana, MS, is a predoctoral fellow in the Department of Health Information Management at
the University of Pittsburgh School of Health and Rehabilitation Sciences in Pittsburgh, PA.
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

Notes

1. Vital Wave Consulting. mHealth for Development: The Opportunity of Mobile

Technology for Healthcare in the Developing World. Washington, DC, and Berkshire,
UK: UN Foundation–Vodafone Foundation, 2009.
2. Kosaraju, Akhila, Cynthia R. Barrigan, Ronald K. Poropatich, and Samuel Ward
Casscells. “Use of Mobile Phones as a Tool for United States Health Diplomacy
Abroad.” Telemedicine and e-Health 16, no. 2 (2010): 218–22.
3. Barton, Amy J. “Using Mobile Health Technology to Enhance Patient-centered Care.”
Clinical Nurse Specialist 24, no. 5 (2010): 233–34.
4. Vital Wave Consulting. mHealth for Development: The Opportunity of Mobile
Technology for Healthcare in the Developing World.
5. Boyer, Edward W., David Smelson, Richard Fletcher, Douglas Ziedonis, and Rosalind
W. Picard. “Wireless Technologies, Ubiquitous Computing and Mobile Health:
Application to Drug Abuse Treatment and Compliance with HIV Therapies.” Journal of
Medical Toxicology 6, no. 2 (2010): 212–16.
6. Han, Dongsoo, Minkyu Lee, and Sungjoon Park. “THE-MUSS: Mobile U-health Service
System.” Computer Methods and Programs in Biomedicine 97, no. 2 (2010): 178–88.
7. Wu, Shin-Yi, and Anthony Green. Projection of Chronic Illness Prevalence and Cost
Inflation. Santa Monica, CA: RAND Health, 2000.
8. Anderson, Gerard. Chronic Conditions: Making the Case for Ongoing Care. Robert
Wood Johnson Foundation, 2004.
9. Lorig, Kate R., and Halsted R. Holman. “Self-Management Education: History,
Definition, Outcomes, and Mechanisms.” Annals of Behavioral Medicine 26, no. 1
(2003): 1–7.
10. Clark, Noreen M. “Management of Chronic Disease by Patients.” Annual Review of
Public Health 24, no. 1 (2003): 289–313.
11. National Institute of Neurological Disorders and Stroke. “Spina Bifida Fact Sheet.”
Available at http://www.ninds.nih.gov/disorders/spina_bifida/detail_spina_bifida.htm
(accessed February 16, 2014).
12. Dicianno, Brad E., and Richard Wilson. “Hospitalizations of Adults with Spina Bifida
and Congenital Spinal Cord Anomalies.” Archives of Physical Medicine and
Rehabilitation 91, no. 4 (2010): 529–35.
13. Ouyang, Lijing, Scott D. Grosse, Brian S. Armour, and Norman J. Waitzman. “Health
Care Expenditures of Children and Adults with Spina Bifida in a Privately Insured US
Population.” Birth Defects Research Part A: Clinical and Molecular Teratology 79, no. 7
(2007): 552–58.
14. Dicianno, Brad E., P. Peele, J. Lovelace, Andrea Fairman, D. Smyers, M. Halgas, K.
Burkholder, and M. L. Boninger. “Specialty Medical Homes and Wellness Services in
Congenital and Acquired Spinal Cord Injury.” American Medical Group Association
Compendium of Chronic Care Practices (2012).
15. Ibid.
16. Zemper, Eric D., Denise G. Tate, Sunny Roller, Martin Forchheimer, Anthony Chiodo,
Virginia S. Nelson, and William Scelza. “Assessment of a Holistic Wellness Program for
Persons with Spinal Cord Injury.” American Journal of Physical Medicine &
Rehabilitation 82, no. 12 (2003): 957–68.
10 Perspectives in Health Information Management, Spring 2015

17. Stuifbergen, A. K., and H. Becker. “Health Promotion Practices in Women with Multiple
Sclerosis: Increasing Quality and Years of Healthy Life.” Physical Medicine and
Rehabilitation Clinics of North America 12, no. 1 (2001): 9–22.
18. Ipsen, Catherine, Craig Ravesloot, Nancy Arnold, and Tom Seekins. “Working Well with
a Disability: Health Promotion as a Means to Employment.” Rehabilitation Psychology
57, no. 3 (2012): 187.
19. Ravesloot, C. H., T. Seekins, T. Cahill, S. Lindgren, D. E. Nary, and G. White. “Health
Promotion for People with Disabilities: Development and Evaluation of the Living Well
with a Disability Program.” Health Education Research 22, no. 4 (2007): 522–31.
20. Ravesloot, Craig, Casey Ruggiero, Catherine Ipsen, Meg Traci, Tom Seekins, Tracy
Boehm, Desirae Ware-Backs, and Bethany Rigles. “Disability and Health Behavior
Change.” Disability and Health Journal 4, no. 1 (2011): 19–23.
21. Wagner, Edward H., Susan M. Bennett, Brian T. Austin, Sarah M. Greene, Judith K.
Schaefer, and Michael Vonkorff. “Finding Common Ground: Patient-centeredness and
Evidence-based Chronic Illness Care.” Journal of Alternative & Complementary
Medicine 11, suppl. 1 (2005): s-7.
22. Bodenheimer, Thomas, Edward H. Wagner, and Kevin Grumbach. “Improving Primary
Care for Patients with Chronic Illness: The Chronic Care Model, Part 2.” Journal of the
American Medical Association 288, no. 15 (2002): 1909–14.
23. Holman, Halsted. “Chronic Disease—the Need for a New Clinical Education.” Journal of
the American Medical Association 292, no. 9 (2004): 1057–59.
24. Parmanto, Bambang, Gede Pramana, Daihua X. Yu, Andrea D. Fairman, Brad E.
Dicianno, and Michael P. McCue. “iMHere: A Novel mHealth System for Supporting
Self-Care in Management of Complex and Chronic Conditions.” JMIR mHealth and
uHealth 1, no. 2 (2013): e10.
25. Ibid.
26. Turner, Carl W., James R. Lewis, and Jakob Nielson. “Determining Usability Test
Sample Size.” In Waldemar Karwowski (Editor), International Encyclopedia of
Ergonomics and Human Factors, vol. 3. Boca Raton, FL: CRC Press, 2006.
27. Lewis, James R. “Sample Sizes for Usability Tests: Mostly Math, Not Magic.”
Interactions 13, no. 6 (2006): 29–33.
28. Hertzum, Morten, and Niels E. Jacobsen. “The Evaluator Effect: A Chilling Fact about
Usability Evaluation Methods.” International Journal of Human-Computer Interaction
15, no. 1 (2003): 183–204.
29. Turner, Carl W., James R. Lewis, and Jakob Nielson. “Determining Usability Test
Sample Size.”
30. American Psychiatric Association. Diagnostic and Statistical Manual of Mental
Disorders, 4th ed., text revision. Washington, DC: American Psychiatric Association,
2000.
31. Parmanto, Bambang, Gede Pramana, Daihua X. Yu, Andrea D. Fairman, Brad E.
Dicianno, and Michael P. McCue. “iMHere: A Novel mHealth System for Supporting
Self-Care in Management of Complex and Chronic Conditions.”
32. Parmanto, Bambang, Andi Saptono, Gede Pramana, Wayan Pulantara, Richard M.
Schein, Mark R. Schmeler, Michael P. McCue, and David M. Brienza. “VISYTER:
Versatile and Integrated System for Telerehabilitation.” Telemedicine and e-Health 16,
no. 9 (2010): 939–44.
33. Dix, Alan. “Human-Computer Interaction.” In Ling Liu and M. Tamer Özsu (Editors),
Encyclopedia of Database Systems. New York: Springer, 2009, 1327–31.
34. Holman, Halsted. “Chronic Disease—the Need for a New Clinical Education.”
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

35. Brewster, Stephen. “Overcoming the Lack of Screen Space on Mobile Computers.”
Personal and Ubiquitous Computing 6, no. 3 (2002): 188–205.
36. Abascal, Julio, and Anton Civit. “Universal Access to Mobile Telephony as a Way to
Enhance the Autonomy of Elderly People.” In Proceedings of the 2001 EC/NSF
Workshop on Universal Accessibility of Ubiquitous Computing: Providing for the
Elderly. New York, NY: ACM, 2001.
37. Kane, Shaun K, Chandrika Jayant, and Jacob O. Wobbrock. “Freedom to Roam: A Study
of Mobile Device Adoption and Accessibility for People with Visual and Motor
Disabilities.” In Proceedings of the 11th International ACM SIGACCESS Conference on
Computers and Accessibility. New York, NY: ACM, 2009.
38. Kurniawan, Sri, Murni Mahmud, and Yanuar Nugroho. “A Study of the Use of Mobile
Phones by Older Persons.” In CHI ’06 Extended Abstracts on Human Factors in
Computing Systems. New York, NY: ACM, 2006.
39. Amar, Robert, Steven Dow, Richard Gordon, Muhammad R. Hamid, and Chad Seller.
“Mobile ADVICE: An Accessible Device for Visually Impaired Capability
Enhancement.” In CHI ’03 Extended Abstracts on Human Factors in Computing Systems.
New York, NY: ACM, 2003.
40. Astrauskas, Michael J., John A. Black Jr., and Sethuraman Panchanathan. “A
Demonstration of Phototacs: A Simple Image-based Phone Dialing Interface for People
with Cognitive or Visual Impairments.” In Proceedings of the 10th International ACM
SIGACCESS Conference on Computers and Accessibility. New York, NY: ACM, 2008.
41. Kane, Shaun K., Jeffrey P. Bigham, and Jacob O. Wobbrock. “Slide Rule: Making
Mobile Touch Screens Accessible to Blind People Using Multi-touch Interaction
Techniques.” In Proceedings of the 10th International ACM SIGACCESS Conference on
Computers and Accessibility. New York, NY: ACM, 2008.
42. Li, Kevin A., Patrick Baudisch, and Ken Hinckley. “Blindsight: Eyes-Free Access to
Mobile Phones.” In Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems. New York, NY: ACM, 2008.
43. Pirhonen, Antti, Stephen Brewster, and Christopher Holguin. “Gestural and Audio
Metaphors as a Means of Control for Mobile Devices.” In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems. New York, NY: ACM, 2002.
44. Brewster, Stephen A., and Peter G. Cryer. “Maximising Screen-Space on Mobile
Computing Devices.” In CHI ’99 Extended Abstracts on Human Factors in Computing
Systems. New York, NY: ACM, 1999.
45. Watanabe, Tetsuya, Manabi Miyagi, Kazunori Minatani, and Hideji Nagaoka. “A Survey
on the Use of Mobile Phones by Visually Impaired Persons in Japan.” In Klaus
Miesenberger , Joachim Klaus , Wolfgang Zagler, and Arthur Karshmer (Editors),
Computers Helping People with Special Needs: 11th International Conference, ICCHP
2008. New York: Springer Berlin Heidelberg, 2008, 1081–84.
12 Perspectives in Health Information Management, Spring 2015

Figure 1

Architecture of the iMHere System

Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

Figure 2

Average Time in Seconds to Complete Tasks in Three Tests

14 Perspectives in Health Information Management, Spring 2015

Figure 3

Telehealth Usability Questionnaire (TUQ) Factors and Scores

Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

Table 1

Average Time in Seconds for Experienced and Inexperienced Users to Complete Tasks

Experienced Inexperienced
Tasks
Participants Participants
Schedule medication 131 106
Schedule mood alert 46 33
Respond to skincare alert 97 92
Respond to mood alert 39 20
Respond to BMQ alert 7 5
Respond to TeleCath 26 29
once-per-day alert
Average 58 48
16 Perspectives in Health Information Management, Spring 2015

Table 2

Mistakes, Steps, and Error Rate

Total No. Average

Total No. Error
Tasks of No. of
of Steps Rate
Mistakes Stepsa (SD)
Schedule medication 16 310 5.2% 17 (4.08)
Schedule mood alert 2 222 0.9% 12 (4.55)
Respond to skincare alert 10 192 5.2% 11 (2.40)
Respond to mood alert 0 216 0.0% 12 (0.00)
Respond to BMQ alert 0 126 0.0% 7 (1.08)
Respond to TeleCath
0 18 0.0% 1 (0.00)
once-per-day alert
Total 28 1,084 2.6% 60
a
Average number of steps = total number of steps / (6 participants × 3 tests)
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

Table 3

List of Mistakes and Difficulty-on-Performance (DP) Scores

MyMeds Participant DP Score Summarya

P01 2
Confused with “Alias” (Type A) MyMeds:
P02 1
• Total MS =
Saved without scheduling an alert (Type B) P06 2
10
P01 3
• Total DP = 20
P02 2
Forgot to save alias and notes (Type B) SkinCare:
P05 2
P06 2 • Total MS = 4
Test
• Total DP = 7
1 Forgot to click “+” to add a new item (Type B) P03 2
Problem selecting the medication from the list (Type B) P02 2
Problem selecting the medication name to expand the Total for Test 1:
P02 2 • Total MS =
detailed lists (Type C)
SkinCare 14
Saved without finishing the form (Type B) P01 2 • Total DP = 27
P02 1
Forgot to click “+” to add new (Type B) P04 2
P05 2
MyMeds MyMeds:
Saved without scheduling an alert (Type B) P01 2 • Total MS = 5
Confused with “once per week” (Type A) P01 2 • Total DP = 10
P01 2
Forgot to save alias and notes (Type B) P05 2 Mood:
P06 2 • Total MS = 2
Mood • Total DP = 3
Test
2 Clicked save button twice (Type C) P01 1 SkinCare:
Confused with “Start Date” (Type A) P04 2 • Total MS = 5
Skincare • Total DP = 6
Saved without finishing the form (Type B) P01 2
P02 1 Total for Test 2:
P04 1 • Total MS =
Forgot to click “+” to add new (Type B) 12
P05 1
P06 1 • Total DP = 19
MyMeds MyMeds:
• Total MS = 1
Saved without scheduling an alert (Type B) P01 1 • Total DP = 1
SkinCare:
Test • Total MS = 1
3 Skincare
• Total DP = 1

Total for Test 3:

Forgot to click “+” to add new (Type B) P03 1
• Total MS = 2
• Total DP = 2
a
Total MS: total number of mistakes; Total DP: total difficulty-on-performance score.
18 Perspectives in Health Information Management, Spring 2015

Table 4

Telehealth Usability Questionnaire (TUQ) Score, Average Time to Complete Tasks, and Total
Number of Mistakes

Participant TUQ Score Average Time (seconds) Mistakes

Participant 1 5.65 90 9
Participant 2 6.95 75 6
Participant 3 6.70 52 2
Participant 4 6.90 48 3
Participant 5 5.90 34 4
Participant 6 7.00 38 4
Average 6.52 56 5
Accessibility of mHealth Self-Care Apps for Individuals with Spina Bifida

Table 5
Participants’ Preferences of iMHere Apps

Number and Percentage of

iMHere App Participants Who
Considered the App Useful
Skincare 6 (100%)
MyMeds 4 (67%)
Mood 3 (50%)
TeleCath 3 (50%)
BMQ 2 (33%)