“El usuario de la Telemedicina: Rol del factor humano”

© 2014 UA Board of Regents

Making Telehealth work:

it’s the people & procedures!

Elizabeth A. Krupinski, PhD

University of Arizona


What is Human Factors?

• Examine social, psychological, biological & physical components that inform design, development & operation products or systems

• Goal = optimize performance, user safety, satisfaction with ultimate goal maximizing benefits user experience


HF Is a Process

Tech Acceptance Model 3

• Efficiency, safety, user satisfaction impact service delivery & TA

• TA = behavioral intention mediated by perceived usefulness & perceived ease use

• TA = determining factor use behavior & ultimate adoption

o Case scenarios o Surveys & Questionnaires o Interviews o Focus groups o Observations (ethnographic analyses) o Task, methods & user analyses o Time & time-motion studies o Anthropotechnology (cultural x-fer) o Many others!

Approaches


Common Methods

• Ethnographic analysis: observe uses technology/system practical environment; qualitative/observational in “real world”;

• Focus Groups: qualitative; elicit opinions process/technology; n = small subject to bias; useful any time

• Prototyping: iterative @ several stages with various methods; $$$ = do early


Common Tools

• Surveys & Questionnaires • Interviews • Focus groups • Observations (ethnographic analyses) • Task, methods & user analyses • Time & time-motion studies • Anthropotechnology (cultural x-fer tech) • Many others!


Designing Surveys & Questionnaires

• Review what already exists

• Check IRB & Human Subjects

• Start with easy & comfortable to answer - screening to find out if asking right people & people less likely terminate once put in work

• Clear, precise, short, easy to read questions ]

• Predefined answers provided & appropriate to question

• Room for additional information

• Pilot & validation


Screening Questions

• Which of the following devices do you own? Check all that apply • Smart Phone • iPad • Heart rate monitor watch • Blood glucose monitor • Step monitor • Bluetooth phone speaker in your car


Instruments & Responses

• Major types instruments • Attitudes & opinions • Tests of knowledge • Programs, services, tools, devices • Self-reported behaviors • Demographics

• Major types responses • Likert-scale & semantic differential • Multiple choice • True or False • Rank order • Open-ended


Useful Tips

• Anonymous vs not

• Introduction & instructions

• Design for intended sample/population

• Avoid unnecessary & repetitive questions

• Short better than long surveys (~ 10 min)

• Limit types of scales, always same direction

• Avoid inconsistency (minutes & hours)

• One question at a time & logical flow

• Try to avoid relying on memory


Bad Questions

• 1) How many hours a day do you monitor your HR? • □ 0 – 59 minutes • □ 1 – 5 hours • □ 6 – 1- hours

• 2) Step monitors are expensive & a waste of time.

st agree agree neutral disagree st disagree

• 3) How many servings fruit did you eat last Mon?

0 1 2 3 4 5 6 7 8 9 10

• 4) How many teenage children do you have? 1 2 3 4 5 6 7 8

5 4 3 2 1

Satisfaction How satisfied are you with….?

Very Satisfied Satisfied Neither Satisfied nor Dissatisfied

Dissatisfied Very Dissatisfied

Agreement How much do you agree that…?

Strongly Agree Agree Neither Agree nor Disagree

Disagree Strongly Disagree

Extent To what extent do you…?

To a Large Extent To a Moderate Extent

To Some Extent To a Little Extent Not at All

Helpfulness How helpful is…? Very Helpful Somewhat Helpful

Neither Not so Helpful Not at All Helpful

Interest How interested in…?

Considerable Interest

Moderate Interest Some Interest Little Interest No Interest

Relative Quantity Should..do more or less of…?

Much More Somewhat More Fine as Is Somewhat Less Much Less

Importance How important is…?

Very Important Somewhat Important

Neither Important not Unimportant

Somewhat Unimportant

Very Unimportant

Quality Rating Please rate quality of…

Excellent Above Average Average Below Average Poor


Avoiding Bias Etc.

• Leading questions • Built-in assumptions • Double-barreled questions • Leaving out response choices • Social desirability • Unrepresentative samples • Under-coverage • Non-response bias • Voluntary response bias • Use skip logic or conditional branching


Bad Questions

1)People who use the Wonder Corp. HR monitor watch report they love it. How likely are you to use the Wonder Corp. HR monitor?

v. likely likely neutral unlikely v. unlikely

2) How likely is it that you would use a heart rate monitor you could wear as a watch?

v. likely likely unlikely v. unlikely

Assessing User Satisfaction


Cognitive Walkthrough

• 1+ evaluators inspect user interface • Do set tasks & evaluate ease learning &

understandability • Interface = mock-up, prototype, developed • Input = user profile, users' knowledge task domain

& task cases • Evaluators = HF engineers, software developers,

marketing, etc. • Best in design stage also in code, test, &

deployment



• Define input

• Who are users?

• What tasks analyzed?

• What is correct action sequence/task?

• How interface defined?



• Analysis Phase • Will users try to achieve right effect? • Will users notice correct action available? • Will user associate correct action with effect

to be achieved? • If correct action performed will user

see progress being made towards task solution?


Heuristic Evaluations

• Used evaluate human-computer interaction

• Alternative to using actual users

• Economical

• 3-5 experienced evaluators

• Apply established guidelines/heuristics to review of system


10 Usability Heuristics

• Visibility system status • Match between system & real world • User control & freedom • Consistency & standards • Error prevention • Recognition rather than recall • Flexibility & efficiency use • Aesthetic & minimalist design • Help users recognize, diagnose & recover from errors • Help & documentation


Contextual Inquiries • Fly on wall: Observe & record behavior in context, without

interfering activities

•Day in life: Catalog activities & contexts users experience entire day

•Personal inventory: Document things people ID important

• Shadowing: Tag along & observe routines, interactions, contexts & ask questions to ensure understand what users do & why


Systems to Evaluate • Patient-Clinician Dynamics

• Communication, relationships etc. • Cultural Issues • Training • Selection of appropriate individuals

• Organizational factors • Administrative, technical & provider staff • Patient & provider location issues & logistics • Legal & billing factors

• Environmental factors • It’s not just bandwidth - display, interface, seating,

lighting, colors, temperature, noise, etc.

Why study HF?

o Understand end results for particular healthcare practices & interventions o Develop better ways monitor & improve quality care o Provide evidence benefits, risks, & results treatments & make more informed decisions o Identify effective strategies improve quality & value care

o Efficient care processes o Effective communication between HC providers o Understanding patient’s current medical condition o Implement effective & sustainable RCA solutions o Reduced risk medical device use error o Easier to use (or more intuitive) devices o Reduced risk health IT-related use error o Easier to use (or more intuitive) health IT o Reduced need for training o Easier repair & maintenance o Cost savings via prevention/mitigation adverse events o Safer working conditions o Improved patient outcomes

Natl Ctr Human Factors Engineering in HC http://medicalhumanfactors.net/

What are Benefits?

http://medicalhumanfactors.net/


Some Useful Surveys • TM Satisfaction & Usefulness Survey Bakken et al.

JAMIA 2006;13:660-667

• Technology Readiness Index (Tri) Parasuraman J Service Res 2000;2:307-320

• Structure-Process-Outcome Framework Angst et al. J Mngt Info Sys 2012;29:257-292

• Health IT Usability Evaluation Model Brown et al. J Biomed Informat 2013;In Press Online First

• Health Education Impact Questionnaire Osborne et al. Pt Educ Couns 2007;66:192-201


Useful Tips

• Always consider system as whole

• Narrow hypotheses discrete & concrete questions; focus outcomes

• Use qualitative & quantitative tools

• Utilize control groups when feasible

• Larger sample sizes = better

• Consider longitudinal studies


Case scenario: Telehealth & aging patients


Two Trends • Biomedicalization

• Tendency to define any emotional, mental & physical process as medical problems

• In society that positions youth & youthful bodies as norm – changes associated with aging often labeled pathological & need to “restore” to youth


Two Trends

• Aging natural & variable

• Transform & creating technologies, architecture, etc. to accommodate changes in emotional, mental & physical changes

• Aging body as enabled rather than seeking to transform from inside out


Provider Attitudes

• Aging Semantic Differential

• Maxwell-Sullivan Attitude Survey

• UCLA Geriatrics Attitude Survey

• Implicit Association Test

• Health Professionals Beliefs & Opinions About the Elderly

• Kogan Scale

• AGED Inventory


Provider Attitudes

• Vary considerably: type provider, age experience, gender etc.

• Overall more negative than positive

• Gonzales (2010): med students often have stereotypic attitudes towards older adults

• Gunderson (2005): rural FL physicians caring elderly ageist perceptions, esp against > 85 & nursing home populations

• Impacts quality, options considered etc.


Stereotypes & Tech

• Marginalized from design & development

• Assigned object rather than subject role

• Do come in daily contact with technology but commonly represented as lacking skills & comprehension to accept, negotiate & assess risks/benefits


Acceptance e-Health • Jung & Loria 2010 J Multidisc HC

• Technology Acceptance Model (TAM)

• 12 Swedish subjects > 45 in-depth interviews

• 3 services: online health guide, e-prescriptions, ask-the-doctor online service

• All positive attitudes towards using, usefulness, ease of use, intention future use

• Main reason not used previously – simple lack of awareness that they exist!


Telemonitoring • Steele et al. 2009 Intl J Med Informatics

21 Australians > 65 living independently

Independence Perception nursing homes -

Significance independence + 8.5/10 score

Impact QOL Benefits & change lifestyle + safety net, not interfere

Concerns Cost, social implication, adherence, healthy, privacy, anxiety, system reliability

-

Personal preferences Previous experience Training willingness

- +

Design preferences Wearable Ambient Embedded (under skin)

- - (cost) +

External factors Support Type care Housing Who subsidize

+ & -


Doing it My Way • Loe 2010 Sociol Health & Illness

• 10 women 90 – 96 aging in place interviewed

• Technology for mobility – walking, driving etc.

• Telephone top technology – emergency, stay connected, feel needed, relieve isolation

• Grass-roots health monitoring networks

• Computers, tv, radio – in control & connected

• Technology – use relieve fear injury, helpless

• Often choose NOT to use technology, medications etc. – maintain self-determination


Rejecting Technology

• Copelton 2010 Sociol Health & Illness

• Hospital-sponsored walking group 50-79 yo

• Free pedometer, manual, activity log, resistance band, t-shirt

• Adopt goals & chart progress pedometer

• Revised! No goals, records, pedometers!

• Pedometers a turn-off

• Social aspect main reason attend & return

• Pedometers = competition, hierarchy, threat to sociability


Obviously Not For Me • Neven 2010 Sociol Health & Illness

• Test robot prototype enhance health aging

• 62 – 79 yo in lab tests & interviews

• Liked it, saw use for it, interacted with it

• But – not for me!

• Made prospective users into someone else - housebound, old, lonely, feeble, in need care & attention

• Saw selves as just the opposite


Implications

• Attitudes matter – patients & providers

• Need to educate & transform stereotypes is key to transitioning delivery of healthcare to patients via telemedicine

• Reasons for acceptance or rejection of telemedicine systems/technology may not always be what you would expect

• People really only as young/old as they feel!

• Adapt & transform technology - don’t target


Do Not Underestimate!

Thank you! Questions?

[email protected]

mailto:[email protected]

“El usuario de la Telemedicina: Rol del factor humano”

Documents

ua board of regents

time prototyping

waste of time

user safety

time logical flow

surveys questionnaires

long surveys

various methods