Acceptance of a Digital Paper-based Diabetes Diary Andr Calero - - PowerPoint PPT Presentation

Acceptance of a Digital Paper-based Diabetes Diary

André Calero Valdez & Martina Ziefle

Human-Computer Interaction Center, RWTH-Aachen University

Diabetes is on the rise

• Over 10% of the population now have

diabetes, 27% have pre-diabetes

• Plethora of additional costs
• Damages to

§ nervous system § cardiovascular system

• Therapy requires
• Adherence
• Tracking of progress (diary)
• Therapy is replacement for

cybernetic system: Pancreas

What is the problem? Blood sugar must be in a very tight corridor

• The pancreas regulates blood sugar
• It decreases blood sugar after meals by secreting insulin
• It increases blood sugar after fasting by secreting glucagon
• If large amounts of sugar are consumed, large amounts of insulin are secreted
• Cells become resistant to this higher dose of insulin
• Even more insulin is required to lower the blood sugar
• Repeat, until pancreas can no longer secret enough insulin à Diabetes Type 2
• High blood glucose levels, glucose reacts with proteins and lipids
• Create advanced-gylcation end-products
• Leads to cell-ageing, arterial stiffness, inflammation, etc.

Therapy of Diabetes Application of medicine to reduce blood sugar

• Change diet, physical activity, reduce blood sugar intake, supply additional insulin
• Challenge:
• Every patient is different and reacts differently to food, activity, medicine at different times
• f days
• Diary keeping is used to measure bodily parameters to optimize therapy
• Main research focus:
• Many older patients do not keep a good diary, if so paper-based
• Adoption of technology is diverse
• Integrating technology into existing behavior

LiveScribe Diabetes Diary Digital pen-based diary

• Livescribe is a pen with an infrared-camera
• Tracks what you write with integrated OCR
• Has a Microphone and loudspeaker integrated
• Extensible using penlets
• Requires specialized paper

Kapitel 21

LiveScribe Diabetes Diary Digital pen-based diary

• The diabetes diary looks like a normal paper-based

diary

• But, when using the livescribe pen becomes interactive
• Entries are digitally stored
• Pen gives recommendations (by voice)
• Date entry is customizable (create shorthand)

Kapitel 21

2IU basal. 9:30 one roll 2bu. 11:00 one hour walking. pear? 3:00pm forest. 120 bs.

User study to evaluate the pen and diary Task-based Study

• Seven tasks (4 introductory)
• Opening the help function
• Writing a sentence (I am taking part in an experiment) – Feedback “Activity logged”
• Log a glucose measurement – Feedback 100mg/dl
• Log insulin dosage
• Log a whole day (what did you eat? What glucose measurements did you recall?)
• Use the diary to retrieve nutrition facts
• Create and use your own abbreviation
• Everything was video-taped for later analysis and qualitative data

Research Questions What determines acceptance of such a device?

• Measured variables
• Age, Gender
• Diabetes Type and duration
• Expertise with computerized technology
• Computer self efficacy
• Task Effectiveness and efficiency
• Perceived Ease of Use and Usefulness (TAM Models)
• Behavioral Intention

Hypotheses

Sample Description Study in a diabetes clinic + control group

• 27 participants
• Three age groups
• Young M=25, SD=2.7, N=10
• Medium M=34, SD=5.7, N=10
• Older M=62, SD=5.7, N=10
• Diabetics Mean Age M=49, SD=17.7
• Non-Diabetics M=33, SD=12.7
• 11 Male, 16 Female

User Diversity in our Sample High levels of expertise

1 2 3 4 5 6 <= 28 29-41 41+ Level of Agreement Age Groups

Technology Expertise and Computer Self-Efficacy

EHT ECT CSE

Task Completion More complex tasks were completed less well by older users

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% <= 28 29-41 41+ Success Rate (%) Age Groups

Task Effectiveness

Task 1 Task 2 Task 3 Task 4 Task 5 Task 6 Task 7

How long do users take? Similar patterns across age groups

50 100 150 200 250 300 350 400 450 <= 28 29-41 41+ Time on Task (s) Age Groups

Task Efficiency

Task 1 Task 2 Task 3 Task 4 Task 5 Task 6 Task 7

Prototype Evaluation Very good evaluation of features

5.54 5.08 5.08 5 4.88 5.08 4.5 4.7 5.19 5.62 5.62 5.62 5.62 5.27 5.32 5.32 5.37 5.21 1 2 3 4 5 6 Help function Logging of blood glucose measurements Logging of consumed BE-units Logging of insulin dosage Logging of activities Looking up of nutritional facts Usage of abbreviations Editing with Diabetto Desktop Printing with Diabetto Desktop Level of Agreement

Means of PEU/PU for Individual Functions

PU PEU

Acceptance of the digital diary High Behavioral Intention in all user groups

1 2 3 4 5 6 <= 28 29-41 41+ Level of Agreement Age Groups

Pen Evaluation

PEU PU BI

Hypotheses

Correlation and LR Results Unexpected results

Caveats and Limitations Prototypical Evaluation and User Group

• Best case users
• Willing to try something new, high technical expertise
• Wizard of Oz-Style Experiment, no real “misunderstandings”
• Scenario driven research
• Some errors occurred in OCR
• No long-term evaluation, no real critical incidents
• Technology and guided use (an experimenter was present)

Discussion Qualitative Insights

• Integration into technology feels very unobtrusive
• “It’s just a pen, and it even feels good in my hand.”
• “I like the voice, it’s so nice and calm.”
• The technology was perceived as ambient and assistive
• Hidden inside the pen
• It felt “smart” and responsive
• No additional things to learn
• Integration of technology inside existing processes is helpful!
• Minimal change of behavior
• No influence of user diversity on acceptance variables

Conclusion Thank you very much for your attention

• Diabetes is rising and very costly
• Therapy is highly individual and requires “customization”
• Digital Diabetes Diary using a Livescribe Pen
• Integrates into existing process (Paper-based diary)
• No influence of user diversity on acceptance!
• Perception of ambient “non-technology”