Palliative Care Clinicians’ Perspectives of an Integrated Caregiver Patient-Portal System in Oncology

Abstract

Background.

Despite recommendations and policies to integrate family caregivers into U.S. healthcare settings, caregivers are not systematically involved. Thus, we developed an Integrated Caregiver Patient-Portal system that (1) allows a patient to identify their primary caregiver and their communication preferences; (2) connects the caregiver to a unique portal page to indicate their needs; and 3) informs the care team of dyad responses to aid in integrating the caregiver.

Objective.

This formative research explored palliative care clinicians’ perceptions of the system to inform refinements before usability testing.

Methods.

We conducted two focus groups with palliative care clinicians (N=11) at an NCI-designated cancer center. Transcripts were analyzed using an integrated approach to specify system benefits and concerns.

Results.

The most referenced benefits included: Learning information that they might not have known without the system; giving caregivers a voice or a way to express needs; and supporting an ideal model of care. Top concerns included lacking capacity to respond; needing to clarify clinician roles and expertise in responding to caregivers’ needs; and ensuring ongoing system adaptability.

Conclusions.

The clinicians’ feedback resulted in revisions including: (1) modifying the caregiver questions; (2) integrating social workers in the response; and (3) adding a text-based report to the care team.

Implications for Practice.

This formative research provided valuable feedback for portal development, and also contributes more broadly to recommendations related to integrating caregivers in healthcare.

Foundational:

This research provides practical and logistical relevance to the discussion on how to integrate caregivers into clinical care.

Introduction

The American Society of Clinical Oncology (ASCO) recommends palliative care services when a patient is diagnosed with advanced cancer or has high symptom burden.^{1, 2} ASCO uses the National Consensus Project definition of palliative care, which includes a patient and family-centered care approach that seeks to optimize quality of life and address physical, intellectual, emotional, social, and spiritual needs.² This comprehensive, patient and family, approach is essential given the strain experienced by family or informal caregivers. As evidenced by studies conducted in various global contexts (i.e., United States [US], Iran),^{3, 4} caregivers are at heightened risk for adverse mental health outcomes when a cancer patient declines in functioning or has high symptom distress, or if the caregiver has poor prognostic understanding. Also based on global findings, it is shown that a caregiver’s mental health can have important implications for their physical health outcomes,^{3, 5, 6} the patient’s emotional well-being,^7–9 and might contribute to heightened health system spending due to the potential of lower quality patient care, greater use of formal services, or heightened health services use by caregivers.^10–14

Based on a recent review of predominately US-based articles, consideration of past review articles, and clinical experience, Alam and colleagues concluded that identifying caregivers early in the palliative care process, recognizing their needs, referring them to services, and providing them with education and, when needed, bereavement support might limit or reduce depression throughout and after caregiving.⁴ Specifically, they offer a Consider, Assess, Respond, Educate, and Support (CARES) framework in palliative-care. This CARES framework aligns with several US-based clinical recommendations and policies to integrate and support caregivers in healthcare,^{12, 15} and provide broad “infrastructure” support to deliver person- and family-centered care.¹⁶ These recommendations to intentionally involve caregivers in healthcare settings align with US state-level laws designed to identify a caregiver in a patient’s electronic medical record and to train caregivers at hospital discharge (i.e., the Caregiver Advise, Record, Enable–C.A.R.E–Act).¹⁷

In light of these US-based recommendations and policies, our team developed an integrated caregiver patient-portal system (herein referred to as a “patient-caregiver portal”) to systematically involve caregivers in healthcare processes by leveraging an existing patient portal system at a National Cancer Institute (NCI)-designated comprehensive cancer center in Philadelphia, Pennsylvania (PA). Importantly, use of patient portals is increasing in the US and among other higher income countries with more advanced healthcare systems.¹⁸ Thus, following further dissemination and testing in this US-context, this tool might be valuable for many contexts and geographic regions where caregiving needs are also high.¹⁹ However, broad dissemination might depend on health system capacity as this method is based on use of an IT-based patient portal system for systematic engagement of caregivers, assessment of need, and delivery of resources.

This patient portal-based system has a number of capabilities, including: (1) allowing a patient to identify their primary caregiver and their communication preferences with that caregiver in the healthcare setting (i.e., share communication, delegate communication, or communicate independently), (2) connecting the caregiver to a unique portal page to indicate their needs, and 3) informing the healthcare team of patient and caregiver responses to aid in integrating and supporting the caregiver. In this initial system, the caregiver assessment included questions about caregivers’ capacity (i.e., caregiving preparation), the level of help needed related to specified care tasks, caregivers’ current level of distress^{20, 21} and caregiver strain across four domains (e.g., social, physical, emotional, and financial).^20–24

A description of the system and the formative research design has been previously reported.²⁵ In short, the system and this research is guided by the Patient and Family Engaged Care Framework,²⁶ the Roadmap for Patient and Family Engagement in Health Care,²⁷ and Clayman’s Autonomy Framework²⁸ along with related work.^{29, 30} Highlights of these frameworks include (1) engagement of key stakeholders; (2) the use and evaluation of technology to enhance involvement; and, (3) consideration of clinical benefit. For example, the National Academy of Medicine’s framework for patient and family engaged care notes that those who interact with the healthcare system must guide process change.²⁶

We have previously reported on patient and caregiver feedback of the system as essential stakeholders.²⁵ The purpose of the present study was to explore palliative care clinicians’ perceptions (i.e., concerns and benefits) of the system as they represent key stakeholders to implementing this system at the cancer center. Specifically, it was perceived that clinician feedback would aid in recognizing system or process limitations and offer problem-solving prior to testing system use among patients and their caregivers in palliative oncology. This approach aligns with a knowledge-to-action framework for qualitative analysis, and, thus, we will share resulting changes due to this feedback.^31–33

Methods

This study involved focus group sessions with palliative care clinicians at an NCI-designated cancer center in Philadelphia, PA. Through the focus group sessions, we sought to explore clinicians’ concerns and benefits of the system to inform modifications to be made prior to usability testing among patients receiving palliatve care and their caregivers. This approach is similar to pretesting a survey prior to distributing it among a larger sample, or the use of cognitive interviewing to understand reactions to survey items,³⁴ and the goal was not to achieve saturation in qualitative analysis.

Procedures and Measurement

We conducted two focus group sessions with the palliative care clinicians at the participating cancer center (N=11; Group 1: n=7, Group 2: n=4). Sessions were led by two study team members with training and experience in qualitative methodology and analysis. For each focus group, clinicians were shown a short video (4 minutes) describing the system and were prompted to respond to open-ended questions about the perceived helpfulness of the system overall and the specific capabilities (i.e., allowing the patient to indicate a caregiver and communication preferences; allowing the caregiver to indicate needs; informing the care team of responses). The study was approved by the cancer center’s Institutional Review Board, and each clinician provided written informed consent prior to participation. Both focus group sessions were audio-recorded with participants’ permission.

Analysis

Qualitative analysis of the focus group transcripts involved an integrated (deductive and inductive) approach.^{33, 35} Specifically, this approach involved categorizing participant statements as either a “benefit” or a “concern” along with a description or code to form subthemes. This integrated qualitative approach for analysis is effective and efficient when seeking a defined purpose,^{33, 35} such as our goal of identifying common concerns in order to inform possible modifications. Comments unable to be coded as a concern or a benefit were coded as a neutral comment.

The primary rater who was involved in conducting the focus groups checked the focus group transcripts for accuracy by reviewing them while listening to the audio recordings. To analyze the sessions, the primary rater read through the transcripts and then coded statements broadly as either a benefit or concern. Text identified as a benefit or concern was further described (i.e., proposed subthemes) to give more depth and understanding and to establish a codebook for secondary coding. This codebook was used for coding by the secondary rater for independent assessment of the transcripts. The secondary rater did not participate in the focus group sessions. Prior to full secondary coding, both raters discussed any concerns with the codebook and made minor adjustments if needed (e.g., modifying the description of a subtheme for clarity).

The benefit and concern subthemes are detailed in Tables 1 and 2, respectively. Coding was conducted using NVivo 12 in order to assess inter-rater reliability with the kappa coefficient calculated by the software. Kappa coefficients indicated moderate-excellent inter-rater reliability for the coding, including a kappa of .795 for coding agreement.³⁶ The primary goal was to identify concerns regarding the system prior to progressing to the usability testing stage. Identified concerns were discussed with the research team, including web technologies experts, to explore possible modifications.

Table 1.

Benefit Subthemes, Descriptions, Exemplars, and Frequency

Subtheme	Description	Example Quote	Frequency Mentioned (n)
Provides New Information	Provides the care team with information about the patient and caregiver that might not be known without the system.	You always see that person, that caregiver is like being the kind of rescuer for the patient. You're happy for the patient. Did they have someone but you're not always aware of what that caregiver’s barriers are like, what their life is, and what they have to deal with to care for that patient. So I think that would be helpful.	8
Gives Voice	Allows the caregiver to express needs themselves.	I think the system kind of formalizes it a little bit, get on paper and they can sort of, mark things that they feel distressed about…	7
Ideal	The system model or concept of attending to caregiver needs is an ideal in terms of care.	In an ideal world, I want to know about these things [Caregiver distress and needs]. I want to be able to support people. It’s just a balance.	7
Avoids Assumptions	Avoids making assumptions about who the caregiver is for the patient or their preferences.	We already kind of do all this, but it will help with those assumptions. So that’s a positive. It’ll trigger questioning instead of the assumption making.	6
Connects Caregiver	Identifies and connects caregiver with the patient’s care team.	I think it’s extremely helpful to know who the caregiver is most closely involved in that patient’s care. We oftentimes have to reach out for various reasons. So to have that information is helpful. And to know that we have permission to talk with them too.	5
Question Relevance	Questions asked to patients and caregivers would be helpful for care.	How much do you want your caregiver be involved? How do you like to communicate? I like those questions.	4
Triage Caregiver Response	Allows care team to assess severity of caregiver needs and allocate or respond accordingly.	This could identify needs that then helps you route people correctly.	3
Portal Use	Use of current portal as part of patient care.	All our work [referring to social work department only] is generated from the portal. So our work orders are in the portal, scheduling rides for people are through the portal consults through the portal.	3
Tailored	System allows understanding of needs so response can be tailored to individual caregiver.	You answer these questions, emotional problems or physical stress and it’s a one-page handout about caregiver's stress and make sure you're taking care of yourself.	3
Validates	Validates any clinician assumptions or concerns about patients or caregivers in terms of caregiver-patient dynamics.	I'm guessing that this would kind of validate for the large majority of our patients what we already know.	2

Subthemes in bold were referenced in both focus group sessions.

Table 2.

Concern Subthemes, Descriptions, Exemplars and Frequency

Subtheme	Description	Example Quotes	Frequency Mentioned (n)
Capacity to Respond	Care team’s perception about capacity (e.g., resources or time) to respond to caregiver needs.	We do ask a lot of these questions in a roundabout way and it’s great, but I don’t know how to help sometimes, which is terrible because then you feel helpless, you know? But we're trained to ask these questions [to caregivers] from a palliative care perspective.	18
Roles	Consideration of clinical role or expertise for caregiver response.	Anything that has to do with emotional coping skills and that sort of thing is social work [a separate department]. Then the specifics of what’s causing them to have difficulty coping, whether it be financial…we would certainly take care of the physical symptoms.	15
Adaptability	Suggested additions/changes to system now or in future.	I would agree an email [email report to clinicians]. I think that would probably be the best. [I] can copy and paste because it is a pain when you're in the portal and then you have to go to Epic [the electronic medical record system] and paste it, unless you had three monitors.	14
Lack of Portal Use	Lack of use of the current portal system.	The only thing I go on there [patient and staff portal] for, for patients is if I want to know who their social worker is, I go to the schedule page and the schedule view tells me who their social worker is. That’s about it.	13
Internal Collaboration	Ability or structural limitations in collaborating or need for collaboration.	I look at about like half of those questions [on the caregivers survey], I'd like social work [to] help.	12
Patient Centeredness	Prioritization of patient well-being and patient-provider relationship given restraints on time or resources.	The other problem that I see with it is that it kind of shifts the focus from the patient onto the caregiver. That may cause some difficulties because of the whole thing about supporting patient autonomy and you are kind of like inserting yourself into a relationship.	11
Caregiver Decline	Assuring that a designated caregiver can decline.	There should probably be something [for the caregiver to] say no, I'm not or I don’t feel confident. I don’t want to be [the designated caregiver].	10
Legal	Legal concerns regarding communication.	Documenting [caregiver needs] kind of puts a little pressure on us as providers to be responsible for yet another thing.	8
Already Done	Similar processes are done organically or informally.	We're doing it sort of informally here and there, but it’s not systematized.	7
Volume	Potential of high caregiver needs reported and strain on clinician workload.	I don’t know, it just scares me. The flood gates that might be open[ed].	7
EMR Compatibility	Lack of direct compatibility with the electronic medical record.	This is very powerful information, right? This should be part of the Epic note, if we get some kind of a narrative summarizing this, that something, even if I get it through email and I could copy and paste and put it into the Epic note.	5
Potential to Overlook	Caregiver need may become lost and unaddressed depending on timing of response.	I wouldn’t want to miss somebody if they're really, really distressed and I don’t know about it.	4
Question Relevance	Questions may not be relevant or could be problematic.	I wonder if we should have some more questions about what’s the best way to communicate because I think that comes up frequently.	4
External Limitations	Inability to address caregiver needs due to regulatory limitations/scope of practice.	I'm looking at nursing related skills [questions]. All the people that would have to care for a central line, they probably already have a home care nurse. And if they don’t have anything [it is because] they don’t typically qualify for that. So then it’s like putting out the question but then not being able to do anything about it…So I feel like there's not a ton of stuff for us to offer besides talking to them again about it over the phone or giving them more information. Most people don’t qualify for a lot of home care stuff for certain reasons via insurance.	3
Internal Practice Logistics	Internal clinic/office logistics might need to align.	I think it kind of comes down to the nitty gritty of how we're going to get this information. If it’s going to be an email before we see the patient and follow up, that’s ideal because then I can like sort of be primed and then address it at the visit versus if there's a distress level that super high that I need to act on it immediately.	3
Patient-Caregiver Dynamics	Patient-caregiver dynamics might be complicated.	I see a lot of people don’t have the traditional nuclear family structure of, okay, this is my loving spouse for x amount of years. There's a lot of like, here's a random cousin in Cleveland. We see people with a lot of social challenges.	2

Subthemes in bold were referenced in both focus group sessions.

Results

The palliative care clinicians (N=11) were predominately female (81.8%) and included doctors, nurses, nurse practitioners, and those in social work. The focus groups most often revealed concerns followed by benefits and neutral comments, respectively. These benefits, concerns, and neutral comments are summarized below.

Benefits

Benefits of the system were synthesized into 11 benefit subthemes and a majority (n=7) of these subthemes were noted in both focus group sessions. Table 1 presents the benefit subthemes with operational definitions, how often each subtheme was referenced or mentioned during sessions, and exemplar quotations. The most frequently mentioned benefits referred to: (1) learning information about the patient and caregiver that they might not have known without the system (Provides New Information); (2) allowing the caregiver to voice or express needs (Gives Voice); (3) being an ideal model of care that considers the patient and caregiver (Ideal); and (4) avoiding making assumptions about who might be the primary caregiver (Avoids Assumptions).

Of note were the clinicians’ positive perceptions of the benefit of giving voice to caregivers and gaining new information about a patient’s and caregiver’s preferences and needs, as well as the system’s contribution to patient- and family-centered model of care. Though the clinicians expressed that such a system would be beneficial in allowing caregivers to express their needs and would be ideal, it was also suggested that such a model might be difficult to implement. For example, one clinician specifically stated,

“In an ideal world, I want to know about these things [caregiver distress and needs]. I want to be able to support people. It’s just a balance.”

Many comments pertained to the system helping clinicians avoid making assumptions about patients’ preferences or who might be the acting caregiver, or validating assumptions if they are made. Similarly, as noted, the system starts with patient preferences related to a primary caregiver and allows for expression of preferred patient-caregiver healthcare communication. One clinician commented that they liked the questions asking about the patient’s communication preferences. There were also positive comments related to the potential of the system to facilitate response to caregivers’ needs, including the ability to prioritize top concerns and tailor information to support caregivers based on their expressed needs.

As shown in Table 1, additional benefits pertained to (1) identifying and connecting caregivers with the patient’s care team; (2) allowing the care team to assess severity of caregiver needs and to respond accordingly; (3) some familiarity with or existing use of the cancer center’s portal system; (4) allowing for understanding of caregiver’s needs so that the response can be tailored to individual caregivers; and (5) validating clinician assumptions or concerns about patients or caregivers in terms of caregiver-patient dynamics.

Concerns

Concerns were synthesized into 17 subthemes and 12 of these subthemes were noted across both focus group sessions. Table 2 presents all concern subthemes with operational definitions, how often each subtheme was referenced or mentioned during sessions, and exemplar quotations. The most frequently noted concerns referred to: (1) the clinicians’ capacity to respond to caregiver needs (e.g., limited resources or time) (Capacity to Respond); (2) consideration of the clinical role or training required for responding to caregiver needs (Roles); (3) the ongoing adaptability of the system in terms of evolving the system over time (Adaptability); and (4) limited use of the portal among clinicians for patient care (Lack of Portal Use).

Though the intention and desire to assist caregivers was expressed by clinicians (see Table 1), the capacity to respond and lack of resources might impede implementation. For example, one clinician stated:

“We do ask a lot of these questions in a roundabout way and it’s great, but I don’t know how to help sometimes, which is terrible because then you feel helpless, you know? But we’re trained to ask these questions [to caregivers] from a palliative care perspective.”

Related to capacity to respond is also the high volume of needs expected among caregivers. One clinician commented, “I don’t know, it just scares me. The flood gates that might be open[ed]” referring to caregivers needing information and/or support.

Also evident was a need for internal collaboration to effectively implement a patient-and family-centered model of care through this system. The clinicians noted needing clarity on roles to effectively respond to specific caregiver needs and that certain departments have the specific expertise to best meet certain types of need – as reflected by the following quote:

“Anything that has to do with emotional coping skills and that sort of thing is social work. Then the specifics of what’s causing them to have difficulty coping, whether it be financial…we would certainly take care of the physical symptoms.”

Similarly, in addition to role distinction, knowing how and when to work together to effectively meet caregiver needs is crucial. For example, one clinician noted the following when commenting on the caregiver survey questions, “I look at about [like] half of those questions, I’d like social work [to] help.” Improving collaboration is viewed as a necessary component of the patient- and family-centered care model and this system if implemented.³⁷

Less frequently noted by the clinicians, but still important, are the challenging dynamics that may be evident between a patient and their caregiver(s). There was concern about the existence of challenging or strained relationships that might exist between a patient and their caregiver and the need to allow a caregiver to decline to be involved.

Concerns were also evident related to specifics of the portal design and flow, including wanting to ensure ongoing adaptability, lack of portal use by most of the clinicians in the focus group, legal concerns due to documentation, lack of direct compatibility with the EMR, and wanting to ensure that the timing of the information to the clinicians would match patient appointments so that an appointment would not occur prior to receiving the information. With respect to connecting to the EMR, one clinician noted:

“This is very powerful information, right? This should be part of the Epic note [the Center’s current EMR system], if we get some kind of a narrative summarizing this, even if I get it through email and I could copy and paste and put it into the Epic note.”

Additional concerns are described in Table 2.

Neutral Comments

Neutral comments, neither a benefit or concern, were specific to the implementation of the system and included suggestions that: (1) the patient and caregiver responses be delivered electronically via email and/or printed report in addition to being available in the portal; and (2) the patient and caregiver reports be sent to the primary clinician and the administrative lead for the care team and social work.

Discussion

The purpose of this study was to explore clinician perceptions of the patient-caregiver portal in palliative oncology care at a US cancer center in the Philadelphia, PA, region. Clinicians’ feedback would be used to refine the system prior to usability testing at the cancer center. Though the primary goal was to refine the system for implementation, it is also worthwhile to share the perceived benefits and concerns of such a system in light of practice and policy recommendations that increasingly encourage the integration of caregivers in clinical care.^{16, 17} A system such as this or others in development would benefit from this clinicians’ feedback given they engage directly and often with patients and their caregivers.

Despite calls to integrate caregivers in healthcare in the US, the application of systematic approaches remains a gap in this field. Indeed, it has been recently reported that a minority (40%) of 204 oncology practices participating in the US National Cancer Institute Clinical Oncology Research Program survey indicated having a process to systematically identify caregivers.³⁸ And, although a majority (76%) of these practices ultimately assessed caregiver needs,³⁸ a lower percentage (63.7%) had supportive services available for caregivers.³⁸ Thus, this highlights that not only are systems lacking, but that resources for caregivers are scarce. The findings from the present study provide important practical perspective on implementing a systematic model to involve caregivers in the US given recent policy and practice recommendations. Findings might also have global relevance as many low to middle-income countries experience high rates of cancer among the population, while the healthcare systems have less capacity and, thus, even higher reliance on informal or family caregivers. Ultimately, with further testing, this Internet-based system or otherwise if needed (e.g., paper-based needs assessment) could be of high benefit to caregivers globally.

In summary, the positive perceptions of the system pertained to giving voice to caregivers, gaining new information about a patient’s and caregiver’s preferences and needs, and offering a patient- and family-centered model of care as well as. However, when considering addressing caregiver needs, it might be essential to strike an appropriate balance in recognizing the needs of the caregiver but still prioritizing the patient or affirming their autonomy or independence from the caregiver.

Concerns raised by the clinicians reinforces reported findings of a great need to connect to existing resources for the patient and caregiver population.^{37, 38} There are several national organizations currently disseminating information resources specifically tailored for cancer caregivers, including the National Cancer Institute (NCI), American Cancer Society (ACS), and the Cancer Support Community (CSC). For example, this includes NCI and ACS’s Springboard Beyond Cancer (https://survivorship.cancer.gov/springboard), ACS’s Caregiver Support Videos https://www.cancer.org/content/cancer/en/treatment/caregivers/caregiver-support-videos.html), and CSC’s Support from a Distance Comprehensive Caregiving Guide (https://www.cancersupportcommunity.org/sites/default/files/fields/resource/file/2018–03/support_from_a_distance_2018.pdf).

In addition to external collaboration, there is a need for internal collaboration to effectively implement a patient-and family-centered model of care. Similarly, in addition to role distinction, knowing how and when to work together to effectively meet caregiver needs is crucial. Less frequently noted topic by the clinicians, but still important, are the challenging dynamics that may be evident between a patient and their caregiver(s).

Knowledge-to-Action

Given this feedback from the clinicians, there were several modifications made to the system prior to usability testing. First, we modified the number and type of questions included in the caregiver survey given concerns about the clinician’s capacity to respond, their clinical role or expertise in responding, and the possible volume of caregiver needs that might arise. The caregiver now includes items on the level (1: Not at all – 5: Very much) of emotional stress, physical strain, and financial strain;³⁹ need for information to address emotional stress, physical strain and financial strain; information on communicating with the patient’s doctor or care team; and information about managing patient symptoms. The selection of these questions were based on focus group input and prior literature recommending that assessments in clinics be concise (i.e., low number of questions), related to constructs of quality of life, and actionable.^{12, 24} To assist in responsiveness to caregiver needs, we also integrated the Department of Social Work into the procedures, so that they could receive patient and caregiver responses and respond to caregivers’ who reported heightened strain in any of the three domains (physical, emotional, or financial stress/strain).

Furthermore, due to the existence of multiple, fragmented systems in clinical care and clinicians’ limited use of the portal, we revised the system to provide a summary of responses to the primary clinician within an email format that is compliant with US patient privacy laws such as the Health Insurance Portability and Accountability Act [HIPAA]. The patient portal system for which the system is based off of originally incorporated viewing of responses in the staff patient portal view. This staff view of the portal is one way that allows clinicians to interact with patients, but also is a tool used at the cancer center for additional work-related tasks (e.g., accessing library resources, room reservations), and, thus, potentially a system broadly used. The finding that several clinicians report limited use of the portal was beneficial for the web technologies team as it allowed for considerations and adjustments for the patient portal in general (e.g., increasing onboarding and training); while, specific to this project, it allowed for consideration of additional modes of information delivery in a way that was responsive to clinicians’ preferences and unique workflow.

Indeed, this revised strategy better fit the clinical flow so that the resulting information could be easily copied into the electronic medical record (EMR) if preferred. It was also indicated that a direct transfer of patient/caregiver responses from the portal to the EMR would be ideal in the future and is an ongoing goal. However, a recent commentary notes the potential challenges of including caregiver reports in patient EMRs⁴⁰, and thus, maintaining a separate file might be ideal and potentially feasible.

Conclusion

Findings from the present study are a necessary and important step in developing systems that engage caregivers and assess their needs systematically through existing and widely-used technology. Identifying and assisting caregivers is increasing recognized as important in preventing or reversing adverse outcomes among caregivers as well as improving care recipient outcomes (e.g., quality of life). Given that caregivers often attend appointments and communicate with healthcare professionals,²² healthcare appointments represent an important touch point for assessing caregivers’ skill-related and personal needs. Recommendations to intentionally involve caregivers in healthcare settings align with US-based state-level statutes designed to identify a caregiver in a patient’s electronic medical record and to train caregivers at hospital discharge (i.e., the Caregiver Advise, Record, Enable–C.A.R.E–Act).¹⁷

Despite this progress in the US, not all healthcare settings intentionally nor systematically involve caregivers. Asking clinicians about their perceptions of this patient-caregiver portal is vital to understand implementation barriers and allowed us to make proactive changes to refine the system prior to widespread patient and caregiver use. Clinicians’ comments reflected a desire to assist caregivers, but also acknowledged difficulties, such as the limited capacity to respond the volume of needs of caregivers. Input from these stakeholders resulted in changes that will increase the feasibility of implementation from a clinical perspective, such as keeping the caregiver assessment short and adapting the transfer of responses to the appropriate care team member. This feedback allowed us to identify ways to modify the system prior to user testing, and informs the long-term development and implementation of the system in oncology and beyond.

It is important to note that the number of clinicians included in the focus groups was low; however, the goal was not to reach saturation of themes, but to explore responses to the prototype system among clinicians (i.e., potential users) at a participating cancer center. One change that resulted from this analysis was modifying the number and type of caregiver questions, including adding a question on communication. We also incorporated an additional option for clinicians to add patient and caregiver responses to the medical chart and involved social work to facilitate timely responses to caregivers. Ongoing work involves finalizing the usability testing and, following that, building out a resourced version to be evaluated among a larger sample of cancer patients and caregivers.