Table of Contents


H3IT Logo

Home Healthcare, Hospice, and Information Technology Innovations Conference

Innovations in Home Healthcare, Hospice, and Information Technology

A forum towards achieving evidence-based diffusion and implementation of innovations

Fri, Nov 3, 2017
Friday, Nov 3, 2017, Washington - D.C.
Proceedings of the Home Healthcare,
Hospice, and Information Technology
Conference
Volume 3
Chicago, IL
November 2016
Steering Committee
Kathryn Bowles, PhD, RN, FAAN General Chair School
of Nursing University of Pennsylvania
George Demiris, PhD, FACMI, Program Committee
Chair School of Nursing & Biomedical and Health In-
formatics, School of Medicine University of Washington
Max Topaz, Organizing Committee Chair School of
Nursing University of Haifa
Güneş Koru, Organizing Committee Chair Department
of Information Systems University of Maryland Balti-
more County
Dari Alhuwail, Local Arrangements Chair Department
of Information Systems University of Maryland Balti-
more County
Ann Horton, Vendor and Government Relationships
Committee Chair Maryland National Capital Home
Care Association
Diane Link, Provider and Consumer Relationships
Committee Chair Link Healthcare Advantage
Richard D. Brennan, Jr. NAHC Liaison National Asso-
ciation of Home Care and Hospice
Program Committee
Dari Alhuwail, University of Maryland Baltimore
County
Kathryn Bowles, University of Pennsylvania
John Cagle, University of Maryland Baltimore
George Demiris, University of Washington (chair)
Birthe Dinesen, Aalborg University
Angelica Herrera, University of Maryland Baltimore
County
Sabine Koch, Karolinska Institute
Güneş Koru, University of Maryland Baltimore County
Robert Lucero, Columbia University
Karen Marek, Arizona State University
Michael Marschollek, Medical School
Karen Monsen, University of Minnesota
Huong Nguyen, Kaiser Permanente Research
Anthony Norcio, University of Maryland Baltimore
County
Debra Oliver, University of Missouri
Guy Pare, HEC Montreal
Kavita Radhakrishnan, University of Texas at Austin
Paulina Sockolow, Drexel University
Max Topaz, University of Pennsylvania
Oleg Zaslavsky, University of Haifa
Editor: Güneş Koru, PhD
Typesetting: Pratik Tamakuwala and Ketan Patil
Sponsors:
2016 - i
Contents
1. Its in the Fridge: The practices of older adults in managing advanced directives and other emergency
information Anne M Turner, Katie Osterhage, Andrea Hartzler and George Demiris . . . . . . . . . . . . . . . . 35
2. Patient Health Goals Elicited During Home Care Admission: A Categorization Paulina Sockolow,
Kavita Radhakrishnan, Edgar Chou . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3. Natural Language Processing and Speech Recognition: Technology Overview and Potential Applica-
tions in Homecare Maxim Topaz, Li Zhou . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4. Engaging Home Care Patients and Caregivers for Better Fall Risk Management: Challenges, Oppor-
tunities, and Leveraging Information Technology Dari Alhuwail, Jennifer Callaghan-Koru, Brandt Braun-
schweig, Gunes Koru . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5. Measure, Share, Improve: Using Performance Dashboards to Impact Home Health Documentation
Times and Quality Juanita Gross, Cheryl Adams, Mark A Bassett . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6. Understanding Information Exchange between Home Care Clients and Aides: Opportunities for
Informatics Laura Kneale, Anne Ordway, Kurt Johnson and George Demiris . . . . . . . . . . . . . . . . . . . . . 42
7. A Feasibility Study Examining Older Adult Needs within Smart Home Sensor Deployments Yong
Choi, George Demiris, Arjmand Samuel, Danny Huang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
8. Accessibility and Beyond: Designing Consumer Health IT for Individuals with Disabilities Rupa S.
Valdez, Melissa R. Lemke, Geoffrey R. Smith, Claire A. Wellbeloved-Stone . . . . . . . . . . . . . . . . . . . . . . . 46
9. Improving Patient Prioritization during Homecare Admission: A Pilot Study Maxim Topaz, Marygrace
Trifilio, Donna Maloney, Kathryn H. Bowles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
10. Supporting Home Care Nurse Decision Making at the Point of Care Through Clinical Dashboard
Design Dawn W Dowding, Nicole Onorato, Yolanda Barron, Jacqueline A. Merrill, Robert J. Rosati, David Russell 49
11. Current Roles and Applications of Electronic Health Record in the Healthcare System Florence F.
Odekunle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
12. Life and Death in a Prescription Bottle: Design of Mobile Health Education to Transform Self-care
in Cancer Medication Regimens Andrew D. Boyd, Claire Heshmat, Ashwin K Nayak, Sandra C. Puri, Scott
M. Wirth, Neeta K. Venepalli, Stephanie Y. Crawford . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2016 - ii
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
It’s in the Fridge: The Practices of Older Adults in
Managing Advanced Directives and Other
Emergency Information
Anne M Turner, Katie Osterhage, Andrea Hartzler, George Demiris
A
lthough there is a growing emphasis on the value of emergency planning through mechanisms such as advanced
directives and standardized emergency forms, research has not examined how older adults gather and manage
this information, or how they share this information within social support networks, retirement communities,
EMS and clinical care systems. Research in this area would be particularly important for ensuring people’s
preferences are known to EMS at the end of their life. As part of our larger AHRQ funded SOARING (Studying Older
Adults & Researching Information Needs & Goals) project, investigating the health information management practices of
older adults, we explored the types and management of advanced directives and emergency information among older adults.
Methods: We conducted 60-90 minute in-depth interviews with 90 older adults (60 years and older). We recruited partici-
pants from retirement homes, senior centers and assisted living facilities. Participants were asked structured and open-ended
questions about their needs and practices around managing personal health information. We recorded, transcribed, and
coded interviews looking for themes using a Grounded Theory approach.
Results: Emergency planning materials were frequently mentioned by older adults in reference to managing personal health
information. The most frequently occurring types of emergency planning materials included: advanced directives, medication
lists, emergency contact information, and standardized POLST (Physician Orders for Life-Sustaining Treatment) forms. 71%
(60 out of 84) of participants indicated that they have some type of emergency information in their place of residence. Of
these people, 43% manage this emergency information independently, and 57% do so with varying levels of involvement from
others, such as family, friends, and/or staff at a retirement community. Emergency planning materials were often initiated by
retirement homes or assisted living facilities. Demographically, advanced planning increased with age (p=.05), and education
level (p=.03). Many participants posted emergency planning materials in their living space; most commonly, this information
was located on or in their refrigerator or on the back of the front door. A variety of reasons were given for why older adults
don’t have emergency planning materials. Many participants mentioned they simply don’t keep emergency information up-
dated. Some explained this was because they are currently healthy and do not anticipate emergencies. Other participants
did not keep emergency information because such preparations make them feel “old”. Others expressed uncertainty in their
decisions regarding advanced directives.
Discussion: Emergency planning materials are an important aspect of health information management among older adults.
Information systems designed to maintain health information should take into consideration the needs and practices of older
adults to maintain advanced directives and other emergency information. Given that many of our participants managed
emergency information collaboratively, an important function of these systems would be to allow older adults to update
information easily as well as to grant viewing/editing capabilities to the person(s) of their choice. Because our participants
primarily maintained emergency planning materials in a physical form, when designing digital systems, consideration must
be given to how this information would be made available to first responders.
Conclusion: Our research indicates that use of emergency planning materials is an important, but understudied, activity
of older adults. Taking into consideration the information management practices and needs of older adults, caregivers, emer-
gency personnel, and other key stakeholders will be critical to designing clear and up-to-date emergency materials which
meet the needs of older adults. Specifically, home care and hospice care providers will benefit from training that addresses
how to acknowledge and integrate these information needs into the care plan. Furthermore, this information can be used to
educate and support family members and other informal caregivers in their care of older adults.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 35
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Patient Health Goals Elicited During Home Care
Admission: A Categorization
Paulina Sockolow, DrPH, MBA, MS
1
, Kavita Radhakrishnan, RN, PhD, MSEE
2
, Edgar Chou MD, MS
1
H
ome care agencies (HCA) have numerous patient engagement opportunities to manage the patient’s symptoms
and provide much needed health services. Despite these services, HCA patient rehospitalization rates exceed
20% for numerous health conditions.
1
Patient engagement entails patient-clinician interaction and patient par-
ticipation in managing his/her health to achieve desired health goals.
2
In home care, a program of patient
self-management goal elicitation with behavioral change was shown to decrease hospital readmissions and improve health
outcomes.
3
Our objective was to categorize elicited patient health goals and identify “clinically informative” goals at a com-
munity based HCA.
Methods: The research team with a HCA partner examined patient goals that admitting clinicians documented in the
point-of-care electronic health record (EHR) during a 5-month pilot project in 2015. The closely-held for-profit HCA oper-
ates over 300 offices in 22 states. Admitting clinicians were employed by the HCA and were predominantly nurses (76%) and
physical therapists (23%). Patient goals were available in a text string in a de-identified Excel file that the HCA extracted
from their EHR. To develop a coding scheme, a researcher (PS) conducted content analysis on patient goal data: 1-assigned
themes to the first 100 patient goals; 2-grouped themes into codes; and 3-specified code categories. A home care nurse (KR)
reviewed the coding scheme. PS assigned a goal code to every 10
th
patient: Sampling was used due to resource constraints.
Records without a patient goal were tabulated. PS added new codes that emerged to the coding scheme that KR reviewed.
Subsequently, KR and the physician researcher (EC) reviewed the coding scheme independently to identify codes that were
informative to their disciplines (clinically important).
Results: Of the 1,763 patient records, 8% had no recorded
patient goal. After content analysis of 122 records, the cod-
ing scheme totaled 20 codes among 3 categories as shown in
the table. In the sample of records with patient goals, there
were 1 to 4 goals documented in each record, for a total cor-
pus of 253 goals. Most goals were phrased in clinician vernac-
ular (e.g., “increased ambulation”) and 6 were in a patient’s
voice (e.g., “to be able to walk again”). Codes identified as
clinically important to both the nurse and physician experts
were equally distributed among the Activities of Daily Living
(ADL) and the Health Management (HM) categories with no
Quality of Life codes selected. There were 5 clinically impor-
tant codes that also occurred most frequently: safety/falls
(ADL, 18%); ambulation (ADL, 9%); ADL activities (ADL,
9%); manage disease process (HM, 9%); knowledge of dis-
ease process (HM, 10%).
Discussion: The absence of the patient’s voice and less
than universal recording of home care patients’ goals indi-
cated differential clinician documentation of elicited patient
goals. Consistent communication of the intent and opera-
tionalization of patient goal elicitation by HCA leadership
may address differential documentation. In addition, clini-
cian training may be advisable to have clinicians understand
why they are asking patients about their goals.
3
Findings
also suggest that the most frequently occurring codes were
codes identified as clinically important for both home care
nurses and primary care physicians. These findings indicate a shared perspective about the importance of specific clinical
information in the treatment of home care patients; however, a Norwegian study found differences in perspectives.
4
Future
research should include perspectives from other disciplines, such as physical therapists.
Conclusion: Research is needed to identify the most effective approach to operationalize patient goal elicitation; clinically
1
Drexel University, Philadelphia, PA
2
The University of Texas, Austin, TX
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 36
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
important goals using a larger group of clinicians; and optimal dissemination of this information in patient care. Useful
research would also be to identify associations between elicited patient goals, nursing interventions, and outcomes.
References
1. Home Health Chartbook 2015: Prepared for the Alliance for Home Health Quality and Innovation. http://ahhqi.org/
images/uploads/AHHQI_2015_Chartbook_FINAL_October.pdf. (Accessed on 05/06/2016). 2015.
2. Sawesi, S, Rashrash, M, Phalakornkule, K, Carpenter, JS, and Jones, JF. The impact of information technology on
patient engagement and health behavior change: a systematic review of the literature. JMIR medical informatics 2016;4.
3. Wagner, EH, Austin, BT, Davis, C, Hindmarsh, M, Schaefer, J, and Bonomi, A. Improving chronic illness care: trans-
lating evidence into action. Health affairs 2001;20:64–78.
4. Hellesø, R and Sogstad, MKR. Hospital nurses’ and physicians’ use of information sources during their production of
discharge summaries: a cross-sectional study. In: Nursing Informatics. 2014:335–341.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 37
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Natural Language Processing and Speech
Recognition: Technology Overview and Potential
Applications in Homecare
Maxim Topaz, PhD, MA, RN
1, 2
and Li Zhou, MD, PhD
1, 2, 3
T
he presentation will discuss several emerging methodologies that have a potential to significantly enhance the
meaningful use of clinical data. First, each technology (natural language processing and speech recognition) will
be briefly introduced with conceptual and practical key concepts. Then, we will overview several recent projects
that use natural language processing to extract meaning from free text clinical narratives, including depression
detection, heart failure self-management status extraction, and using socio-behavioral characteristics to improve readmission
predication models, among others. Applications and unexplored venues of natural language processing in homecare will be
highlighted. We will also review the emerging field of speech recognition and discuss its various potential applications in
homecare, including automated interaction with the patient or tools to facilitate workflows and make clinician’s work more
efficient. Examples throughout the presentation will use projects conducted by Drs. L. Zhou and M. Topaz with MTERMS- a
natural language processing engine developed at the Harvard Medical School and Brigham Women’s Health Hospital (Boston,
USA).
1
Brigham and Women’s Hospital, Boston, USA
2
Harvard Medical School, Boston, USA
3
eCare, Clinical Informatics, Partners Healthcare, Boston, USA
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 38
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Engaging Home Care Patients And Caregivers For
Better Fall Risk Management: Challenges,
Opportunities, And Leveraging Information
Technology
Dari Alhuwail
1
, Jennifer Callaghan-Koru
1
, Brandt Braunschweig
1
, Güneş Koru
1
E
mergency room visits due to falls constitute the largest group of potentially avoidable events in home care.
1
In
addition to increasing healthcare costs,
2
all injuries can easily lead to further serious health problems and even
death for home care patients who are often elderly and vulnerable.
3
Therefore, achieving better fall-risk man-
agement (FRM) becomes critical in improving the quality of care provided by home health agencies (HHAs).
4
In working towards this goal, better engaging home care patients and caregivers in FRM can be both effective and efficient
because they are often the least utilized resources in healthcare.
5,6
Information technology (IT) can play a catalyst and even
an enabler role in improving the quality of care.
7
Therefore, this qualitative research identified the prominent challenges and
opportunities associated with engaging the patients and their caregivers in FRM during home care episodes, and it explored
how IT solutions can be leveraged to positively impact their engagement in FRM. On these topics, there has been a lack of
evidence prior to this study.
Methods: After obtaining the ethics approval, four focus groups were conducted to elicit the perspectives of a professionally
diverse group of home care professionals in Maryland. Twenty participants were recruited based on maximum variation
sampling
8
strategy by considering the participants’ professional background as well the characteristics of their organization
(e.g., size, business model, geographical areas served). The discussion was audio-recorded and run by an experienced facil-
itator assisted by a scribe. Each participant reflected his or her notes individually on the provided handout sheets before
discussing within their focus group. Each focus group summarized their ideas on a flip chart. Raw data were collected
from the participants’ individual handouts, group flip charts, transcripts of the audio recordings, and the scribe notes. The
Framework
9,10
method was used to analyze the raw data resulting in a number of recurring themes.
Results: Challenges and Opportunities The participants noted that while physiological problems such as gait and
balance issues increased fall risks, some patients with such problems also had cognitive issues preventing them from fully
comprehending the FRM advice. This combination requires an even higher degree of caregiver involvement in home care.
Additionally, the participants noted that some of the caregivers did not fully understand their role in FRM or what was
required of them to keep the patients safe either. Among the challenges for FRM was the lack of physical presence of care-
givers with the patient most of the time. The participants also noted that some patients fall because they are in denial of
their physical abilities and limitations. Additionally, cultural differences and language barriers hindered the clinicians from
providing effective FRM advice to both the patients and their caregivers. The participants also reported that the lack of
knowledge and literacy among some patients and caregivers increased fall risks due to their inability to understand and follow
directions. Some home environments did not support effective FRM; these homes had uneven surfaces, poor lighting, and
no hand-rails. Leveraging IT The participants noted that making patient portals available to patients and caregivers on
mobile devices, such as smart-phones, can increase their engagement in FRM. Patient portals have the potential to make the
FRM information easily available as well as keep the patients and their caregivers informed of the care plan and progress.
However, the acquisition, maintenance, and training costs limited the adoption of portals. Cellular network coverage limi-
tations and the Internet affordability were also mentioned as barriers to patient portal use. Sensors, such as those installed
in the home or integrated into devices, can possibly increase engagement by providing the patients and caregivers with data
about activity levels as well as movement and gait patterns. However, cost and privacy issues still impede their adoption to
reduce fall risks. The participants believed that if telehealth solutions can be economically feasible and adopted widely, they
can help clinicians address FRM-related questions in a timely manner, such as those about durable medical equipment use.
Discussion: Similar to other studies,
11–13
the results indicate that for effective engagement in FRM using IT, clinicians
must consider the literacy and comprehension levels of patients and caregivers. HHAs should consider involving the motivated
patients and caregivers more directly in some of the FRM-related quality improvement discussions and meetings. As the IT
adoption for FRM increases, it will be important to provide patients and caregivers proper training and continuous support
to use the adopted solutions.
Conclusion: This study provided evidence about the issues related to effectively engaging patients and caregivers in FRM
during home care episodes. It also identified patient portals, sensors, and telehealth as the most promising solutions to
1
University of Maryland, Baltimore County
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 39
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
increase patient and caregiver engagement in FRM.
References
1. Centers for Medicare and Medicaid Services. Potentially avoidable event measures (PAE): OASIS C Based Home
Health Agency Patient Outcome, Process and Potentially Avoidable Event Reports [Internet]. https : / / www . cms .
gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/OASIS/09aa_hhareports.html. 2016.
2. Titler, M, Dochterman, J, Picone, DM, Everett, L, et al. Cost of hospital care for elderly at risk of falling. Nursing
Economics 2005;23:290.
3. Stevens, JA, Corso, PS, Finkelstein, EA, and Miller, TR. The costs of fatal and non-fatal falls among older adults.
Injury prevention 2006;12:290–295.
4. Ellenbecker, CH, Samia, L, Cushman, MJ, and Alster, K. Patient safety and quality in home health care. 2008.
5. Slack, WV. Cybermedicine: how computing empowers doctors and patients for better health care. 1997.
6. Wiederhold, BK, Riva, G, and Graffigna, G. Ensuring the best care for our increasing aging population: health engage-
ment and positive technology can help patients achieve a more active role in future healthcare. 2013.
7. Alhuwail, D and Koru, G. Leveraging Health Information Technology for Fall-Risk Management in Home Care A
Qualitative Exploration of Clinicians’ Perspectives. Home Health Care Management & Practice 2016;28:241–249.
8. Creswell, JW. Qualitative inquiry and research design: Choosing among five approaches. Sage publications, 2012.
9. Ritchie, J, Lewis, J, Nicholls, CM, Ormston, R, et al. Qualitative research practice: A guide for social science students
and researchers. Sage, 2013.
10. Srivastava, A and Thomson, SB. Framework analysis: a qualitative methodology for applied policy research. 2009.
11. McCormack, L, Thomas, V, Lewis, MA, and Rudd, R. Improving low health literacy and patient engagement: A social
ecological approach. Patient Education and Counseling 2017;100:8–13.
12. Davis, RE, Jacklin, R, Sevdalis, N, and Vincent, CA. Patient involvement in patient safety: what factors influence
patient participation and engagement? Health expectations 2007;10:259–267.
13. Graffigna, G, Barello, S, and Riva, G. How to make health information technology effective: the challenge of patient
engagement. Archives of physical medicine and rehabilitation 2013;94:2034–2035.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 40
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Measure, Share, Improve: Using Performance
Dashboards to Impact Home Health Documentation
Times and Quality
Juanita Gross, Cheryl Adams, Mark A Bassett
I
n the fall of 2014 a pilot project was initiated at Sparta Community Hospital (Sparta, IL) to study the impact
of a new clinical dashboard that displays timely performance measurements for its field clinicians. Recognizing
that “you can’t improve what you don’t measure” the organization worked with its home care software vendor
to develop a new tool for capturing and displaying key performance indicators, like the percentage of TP9
(discharge) visit notes as well as regular visits notes completed during the visit. The results were dramatic: Sparta’s home
health agency increased the rate of in-home completion of regular visit notes by more than 35% (from 60% to over 95%).
Moreover, they reported a marked improvement in documentation quality, including a significantly lower error rate during
the plan of care process. And these were not short-lived gains. Sparta’s improvements in the timeliness and accuracy of
their home health documentation have been lasting (> 2 years). In essence, they have set a “new normal” and established a
significantly higher baseline for quality, accuracy, and timeliness that has made them one of the top-performing agencies in
their region. Upon learning about these results, SwedishAmerican (Rockford, IL) agreed to participate in a follow-up study
to validate the impact of real-time clinical dashboards on clinician performance. This expanded study focuses on an agency
with more than 5,000 regular visits per quarter, with a current completion rate of end-of-day-shift documentation near 68%.
Methods: Swedish-American will discuss with its staff its desire to improve in-shift visit note documentation rates as a means
of improving quality and clinician job satisfaction. Four pairs of clinicians, each with similar documentation completion rates,
will participate in a blind study with a control group. Under the guise that they are testing a software update, one clinician
from each pair will have access to a web-based dashboard displaying their current performance results. The rates of in-shift
documentation completion will be calculated on a weekly basis through the acquisition and analysis of visit metadata, and
transferred to a dashboard that renders overall agency results as well as individual performance results for the four clinician
test subjects. The results will be compared between the partners in each cohort/pair to determine if the mere presence of a
performance dashboard improves their in-shift documentation completion rate.
Results: As in the pilot study, we expect to see measurable, statistically-significant improvements in performance even
amongst the clinicians without dashboards, as a result of the Hawthorne Effect (also known as the “observer effect”) wherein
improvements in performance result simply because the individuals know their performance is being observed. We will also
compare performance improvements between those with access to performance dashboards and those without access to this
data. The study duration will be one calendar quarter (three months), after which time the dashboards will be adopted by
all clinicians. Improvements in documentation timeliness and quality will continue to be measured after completion of the
official study.
Conclusion: Improving quality and efficiency requires first measuring what you wish to improve, then sharing the data with
those in a position to affect the improvement. This study aims to validate positive preliminary findings that suggest significant
quality and efficiency gains are possible when home care field clinicians’ have access to timely personal and organizational
performance data. Data from both studies will be presented along with conclusions about the value and impact of real-time
performance feedback on performance improvement.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 41
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Understanding Information Exchange between
Home Care Clients and Aides: Opportunities for
Informatics
Laura Kneale
1
(lkneale@uw.edu), Anne Ordway
2
(ordwaa@u.washington.edu), Kurt Johnson
1
(kjohnson@uw.edu),
George Demiris
13
(gdemiris@uw.edu)
I
n 2014 it was estimated that over 2 million individuals in the United States used formal personal care services
provided by paid, non-skilled care providers such as home care aides.
1
Although some individuals pay out of
pocket for these services, Medicaid in many states, including Washington State, provide home care aides for
older adults or disabled individuals who need support to stay in their homes. Home care aides in Washington
State support consumers with meal preparation, personal care activities, and light housekeeping.
2
Traditionally, due to their
scope of work, home care aides have not been required to complete comprehensive training; however, in 2011 Washington
State passed a law that required all home care aides to take standardized training courses and pass a certification test.
The purpose of this training is to ensure that all home care aides meet the basic qualifications necessary to carry out their
tasks.
3
Despite the routine nature of the services that they provide, home care aides are considered a significant resource for
individuals that utilize their services.
4
Methods: We performed a secondary data analysis on transcripts from home care client and aide interviews that were
conducted between October 2014 and March 2015. The initial interviews gathered home care client and aide opinions on
the new Washington State home care aide training program. During the initial data collection, the interviews were audio
taped and transcribed. For the secondary data analysis, we identified excerpts from the interviews that discussed technology
use and/or information exchange between home care clients and aides. Each excerpt was coded along a number of salient
dimensions, such as topic, thematic content, and general sentiment.
Results: Twenty-seven participants (17 clients and 10 home care aides) were interviewed. The average age of the home care
aides was 45 years (range 26 to 64 years), and the average age of the home care clients was 53 years (range 31 to 71 years).
Independent and agency-affiliated home care aides were equally represented. Both home care aides and clients considered
communication key to a successful client-aide relationship. Clients and aides regularly exchanged information about home
care schedules, the work of caregiving, and a range of interpersonal topics including family and hobbies. Most of the com-
munication was conducted face to face. Telephones and paper were used to communicate daily schedules, appointments, and
schedule changes. Other forms of technology were not used in information exchange between aides and clients even though
mobile telephones and computers were mentioned as potentially useful tools to support caregiving tasks. Participants also
discussed the challenges with communication and information exchange. Clients expressed frustration with several aspects
of care including having to continually train new home care aides on personal preferences and care needs, and the lack of
notification for last minute service disruptions that can be particularly challenging for clients that relied on aides to support
important activities such as food preparation and grocery shopping. In addition, home care aides also expressed frustration
with the clients’ lack of communication about their individual care needs, and often felt unprepared when arriving at a new
client’s home.
Discussion: Information exchange is critical for the success of the client-aide relationship in home care. Face to face
communication is most often used, however, our findings show that there may be additional opportunities for technology
interventions to increase the efficiency and reach of the information exchange. Informatics could help aides with communica-
tion, documentation, and with tasks related to care facilitating continuity of care and improving patient safety. For example,
technology could be used to help clients clearly express their individual needs and preferences, and communicate these needs
to aides prior to service. This may reduce the uncertainty from both the client’s and the aide’s perspective when starting a
new relationship. In addition, aides could use technology resources to support caregiving tasks such as foot exams for clients
with diabetes.
Conclusion: Technology interventions in home and hospice care have often focused on skilled care providers such as home
health nurses, physicians, and therapists. Our findings indicate that home care aides, due to the increasingly important role
that they play in care in the home, could potentially benefit from informatics tools to increase the efficiency and effectiveness
of their services. More research is needed to better understand the current use of technology in home care encounters, to
specify the needs of home care aides and clients and to identify how technology can support care coordination, continuity of
1
University of Washington, School of Medicine, Biomedical and Health Informatics
2
University of Washington, School of Medicine, Department of Rehabilitation Medicine
3
University of Washington, School of Nursing
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 42
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
care and increase home health aides’ confidence. Furthermore, we need to explore how technology can be integrated into the
home health aide training program.
References
1. Forum on Aging, D et al. The Future of Home Health Care: Workshop Summary. National Academies Press (US), 2015.
2. Services that help an adult remain at home | Washington State Department of Social and Health Services. https :
//www.dshs.wa.gov/altsa/home- and- community- services/services-help- adult- remain- home. (Accessed on
03/13/2017).
3. Office of Program Research Summary of Initiative 1163 Olympia, WA: State of Washington House of Representatives.
2011.
4. Choitz, V, Helmer, M, and Conway, M. Improving Jobs to Improve Care. Washington DC: The Aspen Institute. 2015.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 43
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
A Feasibility Study Examining Older Adult Needs
within Smart Home Sensor Deployments
Yong Choi
1
(yongchoi@uw.edu), George Demiris
1, 2
(gdemiris@uw.edu), Arjmand Samuel
3
(arjmands@microsoft.com), Danny
Huang
2
(v-dannhu@microsoft.com)
A
s older adults age, they are faced with numerous challenges that hinder their independent living such as symp-
toms resulting from chronic health conditions, reduced mobility, social isolation and cognitive decline.
1
Recent
developments in smart home technology designed to detect and record individuals’ activities and status within
their living spaces present a unique opportunity to improve their health and wellness.
2
Although there have
been reported benefits resulting from the use of smart home technology,
3,4
the deployment of sensor systems has often been
limited to laboratory settings due to practical challenges such as managing multiple sensor installations. Furthermore, it
becomes challenging for these new technologies to generate data that can be presented and visualized in a comprehensible
and efficient manner so that older adults themselves can access and interpret information about their activities of daily living.
In this study, we deployed commercially available sensors to community-dwelling older adults’ residences using the Lab of
Things platform for connected devices.
5,6
The primary goal of this study was to understand the information and visualization
needs and preferences of older adults when presented with actual sensor data collected from a 2-month deployment within
the home.
Methods: We are recruiting older adult (65 and older) participants through collaboration with local retirement facilities
in Seattle to take part in a 2-month deployment study. Eligible participants have to be community dwelling older adults
(including residing in an independent living facility), able to give informed consent, and able to read and write English.
The participants are given a choice to choose the sensors they would like to have installed within their home. The sensors
are commercially available and consist of a Door/window sensor and a Multi-sensor (collects data on motion, temperature,
luminosity, and humidity), and a Foscam IP camera. Participants have to select at least one sensor to be eligible for the study.
We use a secure platform, Microsoft’s Lab of Things,
5,6
for collecting and managing data from the sensors. We conduct three
interview sessions one at baseline, one at 1 month, and another at exit. During the interviews we engage participants within
the design of data visualizations derived from the sensor data through a participatory design approach. This involves asking
participants to provide feedback on iterations of the design and also to generate ideas for alternatives to the visualization.
The interviews also seek to gather participant perspectives on sensor use within the home, in particular addressing issues
of intrusiveness, perceived value, and information needs from the visualization. All interviews are audio-recorded and tran-
scribed for content analysis. We also gather demographics and collect participants’ self- reported daily activities (IADL), the
participant’s perceived health and well-being (SF-12), and life-space mobility (LSA), and e-health literacy (eHeals) during
baseline and exit interviews. The University Institutional Review Board approved all study procedures.
Results: The study is currently ongoing. To date five older adults have been recruited and have finished mid-point inter-
views. The mean age (SD) of the participants is 91 (4.9) years old. All of our participants have a Bachelor’s degree or higher.
Our findings so far demonstrate the need to develop interfaces that truly match users’ needs and expectations. Furthermore,
as participants have commented, the processing of visualized information requires that users understand the context of this
assessment (e.g., when data were collected, how they can be used to inform decisions, what the clinical implications of pat-
terns changes may be). We will present recommendations for the design of user interfaces and visualization prototypes based
on the iterative feedback provided by our participants.
Discussion: The innovation brought by the smart home technologies can significantly impact the future of home healthcare.
With more and more older adults showing interest in smart home technologies,
4
the home healthcare industry needs to think
about integrating smart home sensors in their care plan. In order for smart homes to become widely used, we need to ensure
that the information generated by such passive monitoring systems is easily accessible and understood by all stakeholders,
especially older adults, if the use of technology aims to empower them rather than simply monitor every one of their move-
ments.
Conclusion: Despite the potential of smart home technology, there remains a challenge in increasing acceptance and usage
of these technologies particularly among older adults. Our work highlights that the effective visualization of smart home
data is key to the success and ultimate adoption of smart home applications to support aging in place.
1
University of Washington, School of Medicine, Biomedical and Health Informatics
2
University of Washington, School of Nursing
3
Microsoft Research, Seattle, WA
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 44
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
References
1. Grembowski, D, Patrick, D, Diehr, P, Durham, M, Beresford, S, Kay, E, and Hecht, J. Self-efficacy and health behavior
among older adults. Journal of health and social behavior 1993:89–104.
2. Demiris, G, Hensel, BK, et al. Technologies for an aging society: a systematic review of “smart home” applications.
Yearb Med Inform 2008;3:33–40.
3. Lee, ML and Dey, AK. Embedded assessment of aging adults: A concept validation with stakeholders. In: Pervasive
Computing Technologies for Healthcare (PervasiveHealth), 2010 4th International Conference on-NO PERMISSIONS.
IEEE. 2010:1–8.
4. Demiris, G, Rantz, MJ, Aud, MA, Marek, KD, Tyrer, HW, Skubic, M, and Hussam, AA. Older adults’ attitudes towards
and perceptions of ‘smart home’technologies: a pilot study. Medical informatics and the Internet in medicine 2004;29:87–
94.
5. Brush, A, Jung, J, Mahajan, R, and Scott, J. HomeLab: shared infrastructure for home technology field studies. In:
Proceedings of the 2012 ACM Conference on Ubiquitous Computing. ACM. 2012:1108–1113.
6. Brush, A, Filippov, E, Huang, D, et al. Lab of things: a platform for conducting studies with connected devices in multiple
homes. In: Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct publication. ACM.
2013:35–38.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 45
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Accessibility and Beyond: Designing Consumer
Health IT for Individuals with Disabilities
Rupa S. Valdez, PhD
1
(rsv9d@virginia.edu), Melissa R. Lemke, MS
2
(lemkemr@humanabilitydesigns.com), Geoffrey R. Smith,
II, MD
1
(gs4gd@virginia.edu), Claire A. Wellbeloved-Stone, MPH
1
(caw7fd@virginia.edu)
A
s health care shifts to home- and community-based settings
1,2
the roles of both providers and patients are trans-
forming.
3,4
There has been an increasing focus on consumer health IT (CHIT) (i.e., electronic technology used
by laypeople to support health and health care management) to address this shift in health care.
5–7
People
living with physical, cognitive, and/or sensory disabilities face challenges engaging in CHIT and have not been
a demographic focus of developers.
8–10
Approximately one-fifth of the non-institutionalized U.S. population self-identifies as
disabled.
11
With such a large portion of the population identifying as disabled it will be necessary to develop CHIT with this
demographic in mind. This study aims to assess the needs of individuals with disabilities for CHIT focused on one technology
platform (mHealth) and one self-management task (health information communication with social network members). This
will be accomplished through a multi-method study that explicates individuals with disabilities’ existing health information
communication practices and the accessibility, usability, and usefulness of three existing mHealth solutions incorporating
health information communication functionalities.
Methods: There are three phases in this two-year longitudinal study: (1) interview-based exploration of existing health
information communication practices, (2) task analysis and journal-based exploration of challenges to using existing con-
sumer health IT incorporating health information communication functionalities, and (3) design session-based exploration of
potential design solutions. The study will involve 60 participants living with physical (n=20), sensory (n=20), and cognitive
(n=20) disabilities. In Phase 1 participants are involved in a semi-structured interview about their personal network and
health information sharing—including use of CHIT for health information communication. In Phase 2 participants interact
with three existing mHealth apps focused on health information communication on either an iPad or iPhone: (1) Microsoft
HealthVault (non-tethered PHR), (2) Epic’s MyChart (tethered PHR), and (3) CaringBridge (social support). In Phase 3
participants attend group design sessions to discuss mHealth app design focused on health information communication as
well as accessibility challenges of mHealth apps and how to overcome them in future design.
Results: To date we have screened 29 individuals, interviewed XX, and conducted task analysis with 4. Initial analysis of
interviews has shown that most participants rarely use CHIT to communicate health information. Instead, they mainly use
the telephone or communicate in person. Their networks tend to be small (range: n=3-47, mean: n=11, median: n=8).
Many participants are not opposed to using CHIT, but have never learned how or do not have access. The pilot data from
Phase 2 led to initial findings about the usability and accessibility of the three mHealth apps. HealthVault and MyChart do
not have iPad versions. It is not possible to click the sign up button in HealthVault on the iPad, making it non-functional on
the iPad. Additionally, there are limitations with the voiceover function’s interface with the apps. For example, the terms
and conditions text was not read out on the CaringBridge app, there was no voiceover indication when a pop-up appeared,
and it was difficult for the user to tell what she had entered into the email text field. By the symposium we will have a more
robust dataset and more results to present.
Discussion: The initial results show that there are still large gaps between CHIT developers and the needs of individuals
with physical, cognitive, and/or sensory disabilities. After collecting and analyzing data from all 60 participants we will
develop guidelines for app development that address this population’s needs. Simple modifications, such as ensuring that alt
text is available for photos and that the screen flips orientation to landscape, could reduce user burden. A limitation of the
study is that the design guidance to be generated is anchored in one technology platform and three specific mHealth apps.
Conclusion: Engaging with people living with physical, cognitive, and/or sensory disabilities will allow us insight into user
needs and preferences for CHIT. Through mixed methods we hope to use existing mHealth apps to inform design guidance
and increase the accessibility and usability of future mHealth apps as health care shifts to home- and community-based
settings.
Acknowledgments: This research is supported by the Agency for Healthcare Research and Quality (AHRQ) under award
number 1 R21 HS023849-01. The content is solely the responsibility of the authors and does not necessarily represent the
official views of AHRQ.
1
University of Virginia, Charlottesville, VA
2
Human Ability Designs, Canton, MI
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 46
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
References
1. Bureau of Labor Statistics. Career guide to industrie. 2010-11.
2. Home Care & Hospice, NA for et al. Basic statistics about home care. Washington, DC: National Association for Home
Care & Hospice 2008:1–14.
3. Mandl, KD and Kohane, IS. Tectonic shifts in the health information economy. The New England journal of medicine
2008;358:1732.
4. Kaplan, B and Brennan, PF. Consumer informatics supporting patients as co-producers of quality. Journal of the
American Medical Informatics Association 2001;8:309–316.
5. Gibbons, MC, Wilson, RF, Samal, L, et al. Impact of consumer health informatics applications. 2009.
6. Or, CK and Karsh, BT. A systematic review of patient acceptance of consumer health information technology. Journal
of the American Medical Informatics Association 2009;16:550–560.
7. Keselman, A, Logan, R, Smith, CA, Leroy, G, and Zeng-Treitler, Q. Developing informatics tools and strategies for
consumer-centered health communication. Journal of the American Medical Informatics Association 2008;15:473–483.
8. Gell, NM, Rosenberg, DE, Demiris, G, LaCroix, AZ, and Patel, KV. Patterns of technology use among older adults
with and without disabilities. The Gerontologist 2013:gnt166.
9. Archer, N, Keshavjee, K, Demers, C, and Lee, R. Online self-management interventions for chronically ill patients:
cognitive impairment and technology issues. International journal of medical informatics 2014;83:264–272.
10. Goldberg, L, Lide, B, Lowry, S, Massett, HA, O’Connell, T, Preece, J, Quesenbery, W, and Shneiderman, B. Usability
and accessibility in consumer health informatics: current trends and future challenges. American journal of preventive
medicine 2011;40:S187–S197.
11. Brault, MW et al. Americans with disabilities: 2010. Current population reports 2012;7:–131.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 47
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Improving Patient Prioritization during Homecare
Admission: A Pilot Study
Maxim Topaz
1, 2
, Marygrace Trifilio
3
, Donna Maloney
2
, Kathryn H. Bowles
2, 4
U
p to half of hospitalizations happen within the first two weeks of homecare services [e.g.,
1,2
]. Early targeted
allocation of services for high risk patients has been shown to significantly reduce 30-day readmissions for heart
failure patients.
3
Recently, we developed a tool called PREVENT to facilitate decision making on patient prior-
itization for the first homecare visit during homecare admission. This pilot study aimed to test the PREVENT
tool and determine its effect on the timing of the first nursing visit from hospital discharge (i.e. whether high risk patients
were prioritized for care) and on average readmissions rates and time to readmission.
Methods: This pre-post, quasi-experimental, pilot study was conducted at a large, homecare agency in NewYork (NY, USA)
with 176 patients admitted to homecare after a hospital stay. In the pre-experimental phase, we calculated the PREVENT
priority score on 90 randomly selected patients but did not share the scores with the intake nurses. Prior to the post phase,
we educated the intake nurses and regional teams about the PREVENT tool and asked them to prioritize the first visit for
patients who scored high risk on the PREVENT tool. The PREVENT score was then computed by intake nurses for 86
randomly selected patients and visit priority (high or medium/low) was communicated to the regional teams responsible for
patient admission. Timing of the first homecare visit and hospital admission information were extracted from the homecare
administrative records. This study received IRB approval from the homecare organization.
Results: On average, patients in both phases were seen within two days of hospital discharge. In the pre-experimental phase,
72% of patients were high priority compared to 78% patients in the experimental phase (p =.35). During the pre-experimental
phase, both high and medium/low priority patients were admitted to homecare on average 2.2 days after hospital discharge
whereas in the experimental phase, high risk patients were admitted one-half day sooner (1.8 days) and medium/low priority
patients within 2.6 days. Thirty-four percent of patients were readmitted within an average of 21.9 days (SD = 15) in the
pre-experimental phase versus 30% of patients in the experimental phase within an average of 26.5 days (SD =18.8). Further,
hospital admission rates decreased in both high risk (32.8% vs. 36.9%) and medium/low risk patients (21% versus 28%)
between the pre and post experimental phases. Although none of the outcomes were statistically significantly different, all
outcomes trended in the expected direction.
Discussion: In the experimental phase, high risk patients were admitted to homecare almost one day sooner than medium/low
risk patients, reflecting changes in nurses’ admission practices and almost one half a day was shaved off the wait time for
high risk patients. The study successfully tested the feasibility and workflow for administering and delivering the PREVENT
decision support intervention. Hospitalization outcomes all trend toward a positive effect of the PREVENT tool, however
further study is needed with a larger sample under randomized conditions to eliminate confounders.
Conclusion: This pilot study of patient prioritization for the first homecare nursing visit showed promising results. After
applying and sharing the PREVENT tool with the nurses, high priory patients were seen sooner and overall hospital ad-
mission rates decreased. Future work is necessary to validate these results using a larger sample in a randomized controlled
trial. Combining home visit prioritization with other early interventions such as early followup doctor visits should be further
explored.
References
1. Berkowitz, SA and Anderson, GF. Medicare beneficiaries most likely to be readmitted. Journal of hospital medicine
2013;8:639–641.
2. O’connor, M, Hanlon, A, and Bowles, KH. Impact of frontloading of skilled nursing visits on the incidence of 30-day
hospital readmission. Geriatric Nursing 2014;35:S37–S44.
3. Murtaugh, CM, Deb, P, Zhu, C, et al. Reducing Readmissions among Heart Failure Patients Discharged to Home Health
Care: Effectiveness of Early and Intensive Nursing Services and Early Physician Follow-Up. Health Services Research
2016.
1
Brigham and Women’s Hospital, Boston, MA, USA
2
Harvard Medical School, Boston, MA, USA
3
Visiting Nurse Service of New York, NY, USA
4
School of Nursing, University of Pennsylvania , PA, USA
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 48
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
Supporting Home Care Nurse Decision Making at
the Point of Care Through Clinical Dashboard
Design
Dawn W Dowding, PhD, RN
1, 2
, Nicole Onorato, BS
2
, Yolanda Barrón, MS
2
, Jacqueline A. Merrill, PhD, RN
1
, Robert J.
Rosati, PhD
3
, David Russell, PhD
2
F
eedback provided to clinicians on their performance is important for improving health care quality
1
and is
a key component of the IHI triple aim initiative.
2
Dashboards are a form of Health Information Technology
(HIT) that display information in a visualized format which can be used to help provide feedback on quality
performance measures. In this presentation we will present the preliminary results from the first phase of our
study, which is focused on (a) identifying existing quality measures related to the care of patients with congestive heart failure
(CHF) that are relevant to home care nurses and that are under their control (actionable) and (b) to explore if and how
nurses’ numeracy and graph literacy impact their ability to comprehend data presented in a visualized format. The results
of this phase will be used to develop a prototype dashboard for home care nurses at the point of care to help implement
evidence based guidelines for the care of CHF patients.
Methods: To identify existing quality indicators that were meaningful and actionable by home care nurses, 6 focus groups
were conducted with 61 nurses working in a large not for profit home care agency in the Northeast region of the United
States between November 2015 and February 2016. Focus group participants were provided with a list of 23 statements
derived from evidence based practice guidelines on the management of patients with heart failure and asked to identify the
top 5 statements that they felt were a priority in terms of receiving feedback and rank them from 1 (top priority) to 5 (least
priority). The focus group discussion then explored the rationale for priority rankings and how a dashboard could be designed
to provide that feedback. Data was analyzed using thematic analysis.
To explore nurses’ numeracy and graph literacy and their ability to comprehend visualized data a multi-factorial experimental
research design using an online survey was used. Graph literacy was measured using the graph literacy scale
3
which was
developed specifically for the health domain and measures graph reading skills and comprehension across different types of
graphs. Numeracy was measured using the expanded numeracy scale.
4
196 nurses from two home care agencies located in
the North East region of the USA were randomly allocated to 1 of 4 experimental conditions. Outcomes include knowledge
and understanding of the information presented in the visualized dashboard.
Results: Quality indicators related to the tracking of vital signs, symptoms and weight changes were ranked the highest
by nurses (e.g. identification of weight gain). The second highest ranked quality indicator related to ensuring a patient had
received education to support self-management. Themes arising from the discussions included how feedback could improve
workflow and communication between visits.
Nurses answered approximately 10 of 13 graph literacy items and 7 of 8 numeracy items correctly-slightly higher than average
scores for the U.S. population. Across the whole sample, nurses most easily understood information presented in the format
of a bar graph. There was an interaction between numeracy, graph literacy and comprehension. Nurses with low numeracy
were less able to interpret line graphs, those with low graph literacy were less able to interpret spider graphs, and those with
low literacy and numeracy were less able to understand information presented as a table.
Discussion: This study has identified specific elements of feedback on the care of CHF patients that home care nurses
would find valuable for improving care. These elements of care are exclusively at individual patient level, and are required
by nurses in real time. The findings demonstrated that nurses’ numeracy and graph literacy have a significant impact on
their comprehension of information presented in visual formats.
Conclusion: The results will be used to develop dashboards that provide feedback on quality indicators to home care nurses,
in real time, at the point of care. The dashboards will be dynamic; presenting the same information in different formats,
to enable nurses’ to comprehend the data effectively. Future research will evaluate the effectiveness of the dashboards in
improving care processes and patient outcomes.
1
Columbia University School of Nursing, New York, NY
2
Visiting Nurse Service of New York, NY
3
VNA Health Group, NJ
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 49
H3IT: Home Healthcare, Hospice, and Information Technology Conference Chicago, IL, 2016
References
1. Ivers, N, Jamtvedt, G, Flottorp, S, et al. Audit and feedback: effects on professional practice and healthcare outcomes.
The Cochrane Library 2012.
2. Berwick, DM, Nolan, TW, and Whittington, J. The triple aim: care, health, and cost. Health affairs 2008;27:759–769.
3. Galesic, M and Garcia-Retamero, R. Graph literacy: A cross-cultural comparison. Medical Decision Making 2011;31:444–
457.
4. Lipkus, IM, Samsa, G, and Rimer, BK. General performance on a numeracy scale among highly educated samples.
Medical decision making 2001;21:37–44.
Copyright © 2016 by Maryland Health Information Technology LLC Creative Commons License c bn d 50