Keywords
shared access - human centered design - consumer health informatics - caregivers -
patient portal
Background and Significance
Background and Significance
Patient portals help patients view their health information and navigate electronic
care delivery demands.[1] Patient portals have become a mainstream in care delivery interactions.[2]
[3] Yet millions of people with complex health needs are unable to independently manage
their care.[4]
[5] Care partners—family members, friends, and community members who assist patients
with daily activities—often find it difficult to access critical information about
a patient's health and treatment.[6] Many health systems offer “shared access,” a functionality that allows patients
to formally register a care partner to access their portal account using unique login
credentials (username and password),[7] thus respecting patient preferences and need to involve other individuals in their
care.[6] Shared access offers a range of benefits for care coordination and communication
between care partners and clinicians;[8] however, adoption of this functionality has been limited and implementing effective
processes at scale requires consideration of state laws, electronic health record
(EHR) systems, and clinician opinions.[9]
[10] Portal user interfaces and registration processes are complicated and cumbersome,[8] and clinicians and staff cite complex registration processes, and infrequent use
of shared access features such as tailoring care partner access permissions, scheduling,
and messaging.[8]
[11]
[12] Additional complications include the lack of knowledge about registration processes
by clinical staff, patient lack of clarity about the permissions and confidentiality
granted with shared access, and concerns about equitable uptake of shared access for
patients without traditional access to resources (also known as the “digital divide”).[6]
[12]
[13]
[14] Despite a significant growth of patient portal usage since 2020,[15] racial disparities persist where Black and Hispanic patients are 5% less likely
to be offered access to the portal than White patients, while being 8% less likely
to use the technology.[16]
[17] Overall, adoption of shared access is limited and gaps in implementation strategies
merit attention.
Evidence-based guidelines and methodologies may support health care organizations
in overcoming challenges in shared access uptake, however, implementing theoretical
best practices in routine care is difficult.[18] There is an increasing call for research designs that engage multiple approaches
to increase implementation efficacy including designs, which acknowledge the dynamic
and relational nature of research and practice.[19]
[20]
[21] This is aligned with contemporary views on research priorities to support patient
portal use and registration, specifically for closing gaps in adoption for diverse
populations including among care partners.[22] One method is to develop and test implementation strategies in the environments
that are intended to adopt them using human-centered design (HCD).[23] As defined by the International Organization for Standardization, HCD is a method
that “aims to make systems usable and useful by focusing on the users, their needs
and requirements…by applying human factors/ergonomics, usability knowledge, and techniques.”[24] HCD methods encourage participatory approaches that respect unique perspectives
from individuals whose experiences have been affected by a given issue.[12]
[24]
[25]
[26] HCD processes result in “products” that are meaningful to end-users[27] and are developed by using cycles of rapid prototyping and iterative refinement.[26] By engaging clinicians, patients, and care partners, HCD may highlight useful insights
for improving shared access derived from end-user feedback.
We sought to explore care partner perspectives on engagement with patient portals,
in conjunction with findings from a multisite demonstration of a quality improvement
project, and explored additional solutions that could increase shared access uptake.
The parent quality improvement project drew on insights from an investigation into
systemic challenges of shared access,[8]
[28] and a survey of participants attempting to sign up for shared access.[29] Additionally, a set of patient portal educational materials (brochures, posters,
website, and implementation toolkits) was co-designed with patients, care partners,
clinicians, and staff from three partner health care organizations.[12]
Objectives
The objectives of the HCD process and prototypes described in this manuscript are
intended to build upon the parent study by:
-
Collecting feedback on shared access processes and educational materials developed
in our previous project from key informants outside of our partner health organizations
and
-
Describing a design methodology for evaluating key informant feedback to recommend
organizational strategies to increase public awareness and use of shared portal access
functionality.
Methods
We used the Double Diamond Model (DDM) process of HCD as a framework to guide our
approach to eliciting feedback from key informants. Originally developed by the British
Design Council in 2004,[30] the DDM is a standardized design process that places a priority on contextual factors
to identify core issues and find solutions. The DDM was selected for this project
for its emphasis on innovative thinking and on flexible methodological processes allowing
rapid response to and iteration from evolving feedback.[31] The DDM was prioritized because it is a simple process that allows respondents to
effectively participate in the design process regardless of educational or theoretical
background,[32] and it integrates the lived experience perspective in the co-evaluation and co-creation
of solutions that are meaningful to end users. The DDM consists of four phases: Discover, Define, Develop, and Deliver. The Discover phase prioritizes collecting community insights and ideas through interviews with
an emphasis on “divergent thinking,” or outside-the-box thinking, for solutions and
innovation.[31]
[33] The Define phase involves homing in on the definition of a problem based on feedback from community
members, as well as systematically synthesizing information. Inductive coding[34]
[35] is used to summarize feedback and thematically link these with research objectives
to develop a model or theory based on experiences and processes within the text and
observations from respondents. The Develop stage is used to formulate potential solutions for discussion and exploration through
repeated rounds of iteration. Finally, the Deliver phase refers to selecting a single solution and evaluating it while removing and
improving individual components that do not function as intended. The DDM focuses
on stages of both divergent (Discover, Develop) and convergent thinking (Define, Deliver) which are used to prioritize the greatest spread of ideas, and then collating these
ideas into a unified, comprehensive solution.[31]
Interviews
Key informant research is the involvement of patients and other community partners
in the research process to highlight the importance of “incorporating the knowledge
and experiences of those closest to the issue being researched.”[36]
[37] In our process, we recruited key informants whose background, roles, and experiences
mirrored the perspectives captured in the parent quality improvement project, but
included individuals broadly recruited from outside of the three partner health organizations.
Invitations for focus groups were sent to potential key informants, and individuals
who were unable to attend a focus group we offered an option to participate in 1-on-1
conversations. Informants included:
-
Patients and/or care partners who expressed interest in shared access for themselves (as patients) or for another person (e.g., family member).
-
Project implementers (including physicians) who had decisional power in their roles to lead new health
IT-related initiatives. These individuals held titles like chief medical information
officer or were involved in patient portal decisions and had experience with leading
change management initiatives.
-
Connectors/influencers (including physicians) who were broad, influential advocates with strong connections
to health systems. These individuals were health care professionals and/or executives
knowledgeable about the history, policy and research around patient access to information
(e.g., HIPAA, open notes, 21st Century Cures Act, information blocking rule).
-
Marketing and communications professionals who worked inside of health systems or were communications professionals for health
IT professional societies. They often created materials and led communications campaigns
about adopting shared access messaging.
Patient and care partner participants were identified through the existing Coalition
for Care Partners network—a joint collaboration between Johns Hopkins School of Public
Health and OpenNotes at Beth Israel Deaconess Medical Center—and social media channels
run by the OpenNotes communications and dissemination team.[38]
[39] OpenNotes has had more than a decade of experience in research implementation and
dissemination and maintains a strong network of patient, care partner, and clinician
collaborators.[40] As a result, OpenNotes was able to identify implementers and connectors through
existing relationships and worked with Johns Hopkins colleagues to obtain referrals
to additional implementers and connectors. Similarly, marketing/communications staff
were identified through snowball sampling within the OpenNotes and Johns Hopkins communications
staff and referrals from colleagues.
The interview guides were designed to assess the most critical pain points around
shared access for each group of key informants and illicit creative solutions to those
challenges. Additional questions were designed to evaluate the educational materials
developed as part of the parent quality improvement project.[12] Focus group/interview questions included:
-
Is shared access important to you? Why or why not?
-
Have you faced any challenges in learning about or setting up proxy access?
-
What might prevent people from setting up proxy access?
-
Beyond the materials reviewed today, are there other ways we can promote proxy access?
While some questions were consistent across groups, interview guides were tailored
for each type of key informant. Interviews drew on knowledge of the individual's background
and/or role and prioritized personal experiences regarding the topic of shared access
to the patient portal.
All participants were sent information about the interviews and their role in the
research in a prospectus agreement shared 48 hours ahead of a focus group or interview.
This information was also reviewed at the start of each interview session. Transcripts
of the conversations were recorded, with consent, for analytic purposes only. In addition,
at least one team member was available to take minutes at each session. Patient and
care partner participants were offered $100 in honoraria. This work was reviewed and
approved by the Beth Israel Deaconess Medical Center Committee on Clinical Investigations
and was assigned protocol no.: 2022P000020.
Mural
To plan, prototype, organize, and synthesize the large number of insights developed
during focus group and individual interview sessions with end users, the study team
utilized Mural.[41] Mural is a virtual and visual workspace designed to facilitate efficient teamwork
allowing for collaborative participation and facilitation using outlining, diagramming,
and workflow tools. Logical relationships between observations and themes were visualized
in Mural using a variety of approaches such as idea matrices and affinity mapping.
Affinity mapping is a design approach used to visually interpret data to reveal connections
and themes through collaborative discussion.[42] This method favors collaboration over structure that allows users greater flexibility
to synthesize multiple data sources into agreed upon categorizations.
Results
Feedback from key informants was synthesized according to prevalent themes with the
goal of identifying potential solutions, which were rank-prioritized according to
the highest probability of effect and success.
Discover
In total, 47 key informants were recruited to provide feedback ([Table 1]). Of the informants, 20 participated in one-on-one interviews (each 30 minutes in
length), and five focus groups were conducted with four to six participants (each
90 minutes in length).
Table 1
Focus groups and 1-on-1 interview participation
|
Partner roles (personas)
|
No. in focus group
|
No. 1-on-1 interviews
|
Total participants
|
|
Project implementers
|
5
|
7
|
12
|
|
Health care marketing and communications staff
|
4
|
2
|
6
|
|
Patients
|
6
|
3
|
10
|
|
Care partners
|
5
|
6
|
11
|
|
Connectors and influencers
|
6
|
2
|
8
|
Project Implementers, which among them included nine physicians, emphasized why shared access was important
and highlighted systemic strategies for increasing uptake, such as mandatory sign-up;
dedicated resources for patients who want shared access; removing logistical barriers
and increasing ease of granting shared access; standardization of terminology (proxy
vs. shared access); and creating a spectrum of solutions relevant to health systems
at varying levels of willingness/ability to enact a change.
Marketing and Communication professionals emphasized the importance of standardized terminology (e.g., “shared
access” vs. “proxy access”) to help with their efforts. Additionally, they desired
clearer messaging to patients about the value of shared access; easy-to-understand
and follow instructions with pictures; wanted to see care partners prioritized in
messaging about the importance of shared access; and stated that educating clinic
staff about shared access would make the concept more relevant to their own lives
and increase personal buy-in and investment.
Patients endorsed guided facilitation of enrollment in shared access, such as staff-guided
step-by-step instructions; reminders for patients at multiple contact points (e.g.,
in the waiting room, after visits, through portal messages); and automated reminders
within the patient portal. Patients recognized inconsistencies of shared access messaging
across health systems and agreed that varied terms should be standardized (i.e., “shared
access” instead of “proxy access”).
Care Partners emphasized frustration with interoperability and changing requirements for shared
access between health care systems and expressed the need for a more convincing reason
to
not use patient login credentials when accessing a loved one's portal information. Care partners
suggested a preference for account management tools similar to existing applications,
such as Netflix (i.e., one main account with sub-account designations), with granular
controls for patients to decide who has specific access permissions to their records.
Connectors, which included three physicians, named additional interest groups and stakeholders
who could advance shared access through advocacy, policy, marketing, data analysis,
and economics; endorsed monetary incentives for health organizations to increase shared
access; and noted innovative technology companies may provide inspiration in the design
of a universal system of shared access based on modern accessibility standards.
Define
After completing focus groups and 1-on-1 interviews, we used affinity mapping and
inductive coding to organize feedback into thematic groupings to identify patterns
of feedback across all the information collected. The study team used the results
from these processes to define themes and contextualize the most prevalent viewpoints
from informants ([Table 2]). These themes indicated that solutions for increasing the utilization of shared
access need to: (1) include acknowledgement of clinician and staff overload; (2) rely
on simple and effective processes and workflows that clearly demonstrate value and
ease of use; (3) utilize technology and existing functions within the EHR system and
connected organizational resources; and (4) function as a natural extension of clinician–patient
relationships that invoke emotional understanding and safety in users through storytelling
and “warm handoffs.”
Table 2
Human-centered design focus group themes
|
Acknowledging staff burnout
|
Reducing complexity
|
Rethinking shared access intervention adoption process
|
Facilitating shared access adoption beyond materials
|
|
• Sustainable intervention for overworked staff
• Supply necessary support and resources
|
• Create a workflow that is easier than password sharing
• Demonstrate the benefits of separate log-in information for patient the care partner
• Expand privacy settings with clear explanations
|
• Successful adoption of intervention without a research team
• Inclusion of technology for buy-in
• Involve EHR systems and legislation
|
• Warm handoff (via an instructional video or staff member)
• Inclusion of storytelling and emotional drivers
|
Abbreviation: EHR, electronic health record.
Develop
The study team collaborated regularly via Mural to collaborate and draft potential
solutions responding to the themes synthesized during the Define phase. Proposed solutions were modified, combined, or eliminated through iterative
discussion and consensus among team members until the remaining options represented
ideas that were sufficiently distinct and responsive to key themes. An idea matrix
was developed to organize a full set of 13 potential solutions according to the magnitude
of the potential effect on shared access uptake for patients and utility for the health
system, and the amount of effort each solution would require to implement ([Fig. 1]). The HCD study team presented the idea matrix in an all-site meeting consisting
of leaders from across three-partner health care organizations involved in the parent
quality improvement study.[12] Attendees were asked to provide feedback on the feasibility of the proposed solutions
and the ease of implementation within their existing health system context. The HCD
team presented each idea on the matrix starting with the most effective solution,
which would take the most effort, down to the least effective solution, which would
require the least effort. Each idea was presented, and the HCD team detailed their
rationale for placement on the matrix. The attendees were asked to discuss each idea
with specific consideration to the feasibility of the idea, its potential to change
shared access behavior, the ease of implementation within their existing health system
context, and the cost to implement. The solution with the greatest effect that would
require a high amount of effort was wholesale system changes enacted through EHR vendors
(derived from “Rethinking Shared Access Intervention Adoption Process” in [Table 2]), which the group recognized as not being feasible within the scope or budget of
this project. After presenting all possible ideas to leadership, a step-by-step video
on how to grant shared access through the MyChart patient portal was selected (derived
from “Facilitating Shared Access Adoption Beyond Materials” in [Table 2]) due to the ease of prototyping using consumer grade technology and software (feasibility),
available resources at partner sites (potential to change), and similar shared access
videos which could be modified and used to guide progress (cost to implement).
Fig. 1 Shared access prioritization idea matrix.
Deliver
The study team developed a low fidelity prototype of the shared access video. Prototype
activities included a draft script and rough storyboard for an instructional video.
We used an existing shared access instructional video created by Epic[43] as a reference; however, we modified the reference to place a priority on patient
perspectives, health system-specific EHR functionality, customization, and branding.
Using a preexisting research relationship developed as a demonstration site during
the parent study, the study team initiated a collaboration with Providence health
system.[12] Leadership at Providence Institute for Human Caring supported the implementation
of the instructional video with live navigation of their portal narrated by a Providence
clinician ([Fig. 2]).
Fig. 2 Screenshot of the official shared access instructional video on the Providence YouTube
page.
A video script was developed to contextualize instructions from a clinician perspective
to replicate a trusted patient-provider relationship. In partnership with the study
team, leadership at Providence selected a clinician—who also had experience using
shared access as a care partner. The study team included the clinician in the design
and review process for the script to increase accuracy, simplify language for patients,
and add personal insights to develop storytelling elements. Providence asked their
internal marketing and communications experts—some of whom had participated in the
focus groups described earlier—to co-design the implementation with the study team.
Storyboards for the video were developed and shared for review, and the study team
provided feedback to the marketing team.
The study team revised the script to align with the final storyboard. Providence engaged
an in-house staff for production, including a professional videographer and makeup
artist. The filming of framing and education segments with the clinician took place
in October 2023. Click-by-click navigation on the health system portal was captured
for shared access sign-up to visually model the exact steps patients would need to
take to complete the process on the Providence portal. A voiceover by the physician
featured in the video was later added to these segments to explain each action in
full detail. A draft video underwent three rounds of review, with editing supervised
by health system leadership and the study team, before the video was finalized. Edits
mainly included notes for pacing of the video and the length of time portal instructions
were shown on screen. The video was released to the Providence YouTube channel to
coincide with National Caregivers' Month in November and can be found at the cited
link.[44]
Discussion
This study offers a unique contribution to the field of clinical informatics by applying
HCD to the specific challenge of shared access to patient portals—an area that has
received limited attention in both HCD and implementation science literature.[12]
[45] Unlike traditional “proxy access” models, our approach reframes the concept as “shared
access,” emphasizing relational dynamics and patient empowerment. Existing dissemination
and implementation strategies prioritize engagement with end users, dissemination
agents, and interpersonal communication.[45] HCD is not well-integrated into the mainstream development of implementation science;
however, it is an area of study which is growing in importance.[19] This work collected perspectives about shared access from a diverse group of key
informants and evaluated a prior quality improvement project.[9] The authors used the DDM to understand contextual factors, help identify core issues,
and develop solutions for improving receptivity to shared access. While our focus
group participants shared favorable views of shared access materials from the parent
study, additional ideas were suggested for further exploration. Several solutions
were derived from the HCD process, and we were successful in partnering with a large
not for profit U.S. health system to rapidly prototype and implement a guided, simplified
instructional video for patients seeking to learn about and obtain shared access to
their patient health records.
This project is a novel application of HCD, which has not been richly applied to the
challenge of shared access to the patient portal and is not often applied in implementation
science in general. This approach allowed us to create prototype ideas that were attractive
to real world care delivery organizations and addressed feedback from key informant
perspectives at the organizational and patient/care partner level. The video prototype
addressed the 4 key themes derived from informant interviews by: (1) acknowledging
clinic personnel overload and lack of dedicated time (Acknowledging Staff Burnout);
(2) presenting helpful information in a clear and understandable manner (Reducing
Complexity); (3) utilizing existing technology that is easily accessed (Rethinking
Shared Access Intervention Adoption Process); and (4) connecting with patients by
using a practicing clinician within Providence health system in the video (Facilitating
Shared Access Adoption Beyond Materials). After the video was published online, additional
health systems expressed interest in reproducing its content, which suggests potential
to further disseminate the shared access video outside of the parent study project.
This work validates previous research conducted by the parent quality improvement
study[12] who also derived insights from informants for topics such as cumbersome technical
processes for obtaining shared access, the need for clear instructions, and considerations
for overworked clinical staff. Outcomes from the parent study also showed that many
patients were unaware of shared access and that the quality improvement study did
not significantly increase new shared access registration.[28] The video prototype responds to a need for additional solutions to facilitate and
encourage shared access and is intended to be repurposed by other health systems through
simple adaptation of branding, color palettes, and most importantly, functionality
specific to each EHR. The video prototype expands future directions outlined in previous
work and provides a targeted and specific product that educates and instructs patients
and their care partners on the exact process for signing up for shared access.
Two major lessons were learned from this project. First, health systems support efforts
to engage in relatively low-effort initiatives that have potential for great effect
and seem to show willingness to collaborate to incorporate key informant perspectives.
Second, feedback from informant interviews can be leveraged to improve messaging to
patients and care partners. For instance, participants from informant groups indicated
the need for instructional resources, easy-to-use privacy controls, and personal,
guided facilitation as well as a movement away from confusing terms such as “proxy
access.” The video prototype includes visual, step-by-step instructions delivered
by a relatable clinician/care partner, combined with a simplified script that avoids
confusing terminology and was co-written by the health system, an experienced clinician,
and a HCD research team.
There are several limitations to this research. We initially aimed to recruit eight
participants per focus group, targeting a total of 40 participants. However, coordinating
group sessions with diverse stakeholders proved challenging ([Table 1]). To address potential gaps in representation, we conducted an additional 20 one-on-one
interviews to capture a broader range of perspectives. Ultimately, we included insights
from 46 individuals. While this approach allowed us to meet our overall recruitment
goal, we acknowledge that the shift from group discussions to individual interviews
may have influenced the nature of the information shared with the HCD team. Additionally,
while participants were recruited through institutional networks and social media,
we recognize the potential for selection bias toward individuals who are already familiar
with or interested in patient portals, which may limit generalizability. While this
project successfully evaluated the educational materials from the parent quality improvement
study through informant interviews, it evolved beyond evaluation into exploring additional
solutions for shared access beyond what was originally developed. Future efforts to
design digital health interventions may find value in HCD methods that encourage evolution
of ideas and rapid cycle prototyping. Ultimately, systematic change is necessary to
address complex sign-up procedures that require extensive forms to be completed in-person
or confusing multistep authorization procedures that may be burdensome for some patients.
The video prototype attempts to respond to these issues by being tailored to the health
system for which it is implemented. Clinics adopting a similar video need step-by-step
instructions specific to their own shared access processes, and all organizations
adapting this approach would need to periodically revisit the video to ensure accuracy
to current policies and EHR functionality. An instructional video may not benefit
all patients including those who are not English-speaking, and it is unclear at present
the full effect of the video prototype. This study was not designed to evaluate whether
proposed solutions affect care partner engagement with shared access functionality.
As such, additional work should be done to target implementation of these videos to
track adoption, engagement, sustainability, and a strategy for visibility to patients
across the system. Future research should prioritize multimodal interventions for
shared access including print materials, instructional videos, interoperable EHR standards
and practices, and advocacy to state-level agencies and health systems to emphasize
the barriers to shared access.
Conclusion
We successfully evaluated a multisite quality improvement study by utilizing an HCD
methodology intended to evaluate key informant feedback on organizational strategies
to increase public awareness and use of shared portal access functionality. This research
demonstrates a practical process for eliciting, codifying, and rapidly developing
efficient and effective responses to key informants on the topic of shared access
using HCD methods. We demonstrate the feasibility and importance of working with clinicians,
patients, care partners, and health systems to utilize HCD methodologies that help
develop thematic analysis and identify solutions that can be adopted with minimal
barriers. By co-designing a low-burden, scalable instructional video with input from
patients, care partners, and health system stakeholders, we demonstrate how HCD can
be used not only to identify barriers but also to rapidly prototype and implement
solutions that are adaptable across diverse care settings. This work bridges a critical
gap between user-centered design and real-world implementation, offering a replicable
model for future digital health interventions.
Clinical Relevance Statement
Clinical Relevance Statement
This study demonstrates how HCD methods can be used to identify novel solutions for
simplifying shared access to patient portals and benefit patients and care partners.
The HCD process resulted in a prototype video developed in collaboration with a health
system to address barriers to adoption that were revealed through informant interviews
such as complex registration processes and low adoption rates. HCD methodologies can
help health systems identify practical solutions for improving care coordination through
patient portals.
Multiple-Choice Questions
Multiple-Choice Questions
-
What was the primary goal of applying Human-Centered Design (HCD) in this study?
-
Increase health system revenue
-
Improve the design of patient portal interfaces
-
Utilize HCD methods to synthesize, understand, identify, and develop solutions for
shared access
-
Enhance patient access to test results
Correct Answer: The correct answer is option c. This study aimed to describe a methodology for evaluating
key informant feedback about shared portal access by applying HCD methodologies such
as the Double Diamond Model.
-
Which group was not included in the key informants interviewed in this study?
-
Patients and care partners
-
Health care marketing and communications professionals
-
Health system executives
-
The payors of health services (e.g., insurers, CMS)
Correct Answer: The correct answer is option d. Payors were not included. This study aimed to evaluate
feedback from end users of shared portal access functionality including patients,
organizational leadership, and marketing.
-
What is one of the major challenges the study identified in adopting shared access
to patient portals?
-
Lack of clinician interest
-
Complexity of portal interfaces and registration processes
-
High costs associated with shared access technology
-
Insufficient EHR functionality
Correct Answer: The correct answer is option b. Complexity of portal interfaces and registration
processes were most relevant to key informants in systems where shared portal access
functionality exists but is not widely adopted.
-
How did the researchers address the challenge of complex registration processes for
shared access?
-
By simplifying the portal software
-
By developing an instructional video prototype
-
By conducting additional user interface training for clinicians
-
By reducing the number of required forms
Correct Answer: The correct answer is option b. The development of an instructional video prototype
was selected by health organizational leadership due to ease of prototyping using
consumer grade technology (feasibility), available resources at partner sites (potential
to change), and existence of similar products which could be adapted (cost to implement).
-
What human-centered design methodology did the study use to guide the development
process?
-
Lean Startup
-
Six Sigma
-
Double Diamond Model
-
Agile Development
Correct Answer: The correct answer is option c. The Double Diamond Model was selected for its emphasis
on flexible methodological processes for rapid iteration. This model allowed for meaningful
participation of key informants in the design process without the need for specialized
education or training.
