Needs and Perspectives of Cancer Center Stakeholders for Access to Patient Values in the Electronic Health Record

PURPOSE:

High-quality cancer care must incorporate patients' personal values in decision making throughout illness. Unfortunately, patient values are neither consistently elicited nor easily accessible in the electronic health record (EHR). Memorial Sloan Kettering Cancer Center is deploying a major EHR innovation, called the Patient Values Tab, which provides ready access to patients' values and personhood. To inform the Tab's design, we interviewed a large, diverse group of institutional stakeholders to understand their user needs for this Tab.

METHODS:

Qualitative data were collected through semistructured, audio-recorded, in-person, individual interviews. An interdisciplinary team of four coders conducted a process of thematic content analysis. Thematic saturation was achieved, and member checking was performed.

RESULTS:

A total of 110 stakeholders were approached and interviewed. Participants comprised a wide range of disciplines or professions and others involved in hospital and/or clinic administration. Analysis revealed the following themes related to important Tab content: personhood, support system or resources, social history, communication preferences, future planning, end of life, and illness and treatment understanding. Participants also discussed implementation considerations, the Tab's potential to improve communication, and privacy implications.

CONCLUSION:

This study focused on a major EHR innovation to centralize information about values and personhood of patients with cancer. We elicited views of over 100 institutional stakeholders through in-depth interviews that were rigorously analyzed, yielding themes related to content and format that helped guide the Tab's design. The interviews generated a sense of ownership and enthusiasm for the Tab among future users. The Tab's introduction advances the use of the EHR as a driver of the delivery of patient-centered care.

BACKGROUND

Cancer challenges patients to maintain their identity and dignity in the face of treatment interventions, functional limitations, and fundamental changes in relationships.^1-3 High-quality cancer care includes strong support of personhood, that is, the uniqueness and inherent worth of each person, regardless of disease or disability.⁴ This support includes structured opportunities for every patient to articulate personal values—what means most in life and what living well means to the individual—and integration of values into decision making throughout illness.^5-8

Unfortunately, clinical teams often lack information about patients' values.^8-10 Challenges include insufficient training and time to elicit patient values,^8,11-14 fragmented care across settings and providers,^8,15,16 and inefficient processes to document and display this essential information for easy access by all members of the healthcare team.^8,17-20 Prior efforts initially focused on centralizing advance directives (including cardiopulmonary resuscitation preferences),^21,22 with more recent initiatives incorporating electronic templates to document patient values and goals,^23,24 as well as patient-facing websites exploring what matters most in life.^25,26 However, previous work has not encompassed rigorous investigation of the needs of stakeholders for whom ready access to this information is important.

Memorial Sloan Kettering (MSK), a comprehensive cancer center, has already systematized a process for structured, nurse-led discussions of patient values for all patients, shortly after diagnosis, in the outpatient setting.^27,28 The oncology nurse summarizes the patient's values on a templated Values Summary (Fig 1),^27-30 which is verified by the patient, reviewed with the oncologist, and uploaded to the electronic health record (EHR). Items in the Values Summary are based on existing literature including the Serious Illness Conversation Guide^29,30 and our own prior work.^27,28 Goals of care discussions by oncologists, other physicians, and advanced practice providers are also templated for EHR documentation.

FIG 1.

Process of nurse-led values discussions at Memorial Sloan Kettering.^27-30

MSK is now implementing a major innovation, the Patient Values Tab, which provides ready access to these key documents, along with other information elicited by members of the interdisciplinary healthcare team illuminating the patient with cancer as a person with individual values, goals, and preferences. In a single EHR location, this Patient Values Tab organizes, displays, and highlights such information as prominently as laboratory values and other objective data that typically dominate the EHR. It is intended to encourage clinicians to seek, document, and honor this information as a basis for patient-centered care throughout illness.^31,32

To inform the creation of the Values Tab, we conducted in-depth interviews with a large and diverse group of institutional stakeholders, focusing on content they perceived as most important for patient-centered care and how it should be displayed. We chose this qualitative approach to obtain nuanced feedback (grounded in current practice and workflow processes), understand concepts not previously explored, and generate new ideas and suggestions.

METHODS

The MSK Institutional Review Board approved a waiver of informed consent because this research presented no more than minimal risk to participants. Qualitative data were collected through semistructured, in-person, individual interviews with stakeholders (Table 1), who were purposefully sampled to include multiple disciplines and professions, as well as others involved in administration of the hospital and/or clinic. Within each stakeholder group, individuals were selected to represent a range of demographic characteristics (eg, sex and race) and practice experience (eg, time in clinical practice). All participants were recruited by e-mail, and interviews were held in stakeholders' professional offices.

TABLE 1.

Stakeholders by Group (Total N = 110)

Two Supportive Care clinical investigators (A.V.D. and R.A.), trained by MSK's qualitative methods specialist (K.A.L.), conducted all interviews, which were audio-recorded with participant consent. Interviewers used a written question guide (Appendix 1, online only) with questions, probes, and flexibility to elicit additional comments. Items in the guide were generated by institutional leaders in oncology, supportive care, health informatics, and nursing informatics, along with patients and families on MSK's Patient and Family Advisory Council for Quality (PFACQ), and refined with input from the qualitative methods specialist. Prespecified domains included respondents' perspectives on the most important patient values, goals, and preferences; approaches to make this information readily accessible in the EHR; and ways to incorporate this information in care and decision making. For PFACQ participants, items addressed information that patients and informal caregivers would wish to share with the healthcare team through the EHR. Initial questions allowed respondents to freely describe content that would be desirable, whether or not they believed that information was available in the existing EHR. Interviewing continued until full representation of stakeholder groups and thematic saturation were achieved.

Audio-recorded interviews were transcribed and deidentified.³³ Transcripts were imported into ATLAS.ti software v. 7.0 to assist with qualitative analysis and approached using an independent and collaborative process of thematic content analysis.^34-36 An interdisciplinary team of 4 trained coders (supportive care researcher [A.V.D.], transitional-year intern [K.L.C.], chief internal medicine resident [B.K.], and health informatics researcher [C.L.M.]) first reviewed the same subset of 25 transcripts (approximately 20% of the total) to develop the initial coding framework. A codebook was iteratively built using (1) a priori codes based on written guide domains and (2) an initial line-by-line inductive coding process. This codebook was then applied to all transcripts using cycles of coding and constant comparison, with refinement of the codebook to incorporate new insights.^34-36 After development of the initial coding framework, the remaining transcripts were divided among the coding team, which met regularly for consensus discussions. Codes identifying common feedback were grouped into categories to identify major themes, which represented patterned response[s] or meaning within the data set that captures something important about the data in relation to the research question³⁵; these were reported prominently among at least one stakeholder group. We determined the proportion of theme endorsement within each stakeholder group by examining code frequencies. After all transcripts were coded, we conducted a quality assurance check of the data set to confirm that statements were assigned to the appropriate code category before identifying major themes. Member checking was performed with 13 respondents (12%) from varying stakeholder groups to validate the analytic framework. This manuscript follows the consolidated criteria for reporting qualitative research.³⁷

RESULTS

A total of 110 institutional stakeholders were interviewed (Table 1) for a median duration of 27 minutes (range, 7-57). The participants included 67 females and 43 males, whose time working at MSK ranged from 1 month to over 40 years. Respondents were 69% Caucasian (76 respondents), 16% Asian (18 respondents), 6% Latinx (six respondents), and 9% African-American (10 respondents). No stakeholder refused participation or audio-recording. Three participants elected to reflect on the questions and send responses via e-mail. For six interviews, audio-recording was of suboptimal quality, but concurrent handwritten notes taken by the interviewer were available.

Respondents reported that the current EHR lacked important information about patients' values and personhood, and/or this information was difficult to access. They expressed widespread enthusiasm for a centralized EHR feature, for its practical utility and its symbolization of high-quality patient care. One inpatient oncology nurse said, “This would absolutely be my go-to Tab,” whereas a senior nursing administrator commented, “Values directly relate to the patient experience. By providing this kind of care, we are showing our own values as an institution.”

Analysis of qualitative interviews revealed 10 major themes, each of which contained at least one category. Table 2 shows the proportion of each group of stakeholders expressing these themes. Table 3 includes excerpts corresponding to each category.

TABLE 2.

Themes Expressed Within Each Stakeholder Group (Proportions)

TABLE 3.

Illustrative Excerpts Corresponding to Each Category Within Each Theme

Themes related to important Values Tab content included:

Personhood

Ninety-two respondents (84%) supported inclusion of core elements of personhood, such as the patient's personal story, hopes, fears, sources of strength and/or resiliency, and accomplishments. A psychiatrist wanted to know, “What is their life narrative?” whereas a surgical oncologist commented, “You have to know your patient beyond their disease.”

Support System and Resources

Seventy-nine respondents overall (72%), including all patients and caregivers, identified the patient's social support system and other resources as important content. This information would include the patient's social network of family and friends, financial resources, and access to healthcare services including transportation. As a PFACQ member noted, “You have to be prepared if the person says, yeah, I'm really struggling… I'm afraid that I can't pay my bills” to connect each patient to important resources.

Social History

Seventy-eight respondents (71%) endorsed inclusion of aspects of the patient's social history, including cultural and religious practices and beliefs, occupation, members of the patient's household, and important family dynamics. To tailor care to an individual patient, a chaplain inquired, “What is their weekly rhythm in their spiritual practice?”

Communication Preferences

Seventy-six respondents overall (69%), including all patients and caregivers, endorsed the importance of information about the patients' preferences with respect to clinician communication of health-related information. For example, some patients prefer to receive information as a big picture overview versus with a lot of details; some patients prefer to receive information with a family member or friend present versus by themselves first. A hospitalist wanted to know, “Are they a statistics kind of person, a picture kind of person?”

Future Planning

Seventy-one respondents (65%) supported inclusion of specific goals and milestones to which the patient looked forward, as well as preferences regarding fertility preservation and screening for other medical conditions linked to the cancer or treatment. A hospital administrator noted, “We'd like to know if you have a wedding or a graduation or a new baby being born … we've had patients try to go to a wedding and come back … because it was so important to them.”

End of Life

Sixty-three respondents overall (57%), including three out of 10 patients and caregivers (30%), favored display of patients' end-of-life care preferences including resuscitation preferences and other advanced directives (living will and/or healthcare proxy), as well as preferred location and other aspects of end-of-life care. An oncologist was interested in, “Whom they want to be with when they die?”

Illness and Treatment Understanding

Forty-six respondents overall (42%), including one out of 10 patients and caregivers (10%), supported inclusion of the patient's understanding of the expected course of illness and intent of current treatment. This theme was one of the most variable among the stakeholder groups, in terms of response frequencies. A Supportive Care clinician inquired, “What is their understanding of their disease? What is the intent of their treatment—is it curative or palliative?”

Other important themes that arose within the interview process included:

Implementation Considerations

Eighty-nine respondents (81%) identified issues relevant for implementation of the Values Tab, including the importance of updating content, clarifying responsibility for obtaining and updating it, and determining the patient's role in contributing to content. Respondents felt that all team members should play a role in these processes, but that the main role should rest with the primary oncology team because of their unique continuity and connection with the patient. Respondents also highlighted the value of documentation from palliative care, psychiatry, and ethics consultants for understanding the patient as a person and offered operational suggestions about the Values Tab's format. As a paramount consideration, respondents emphasized efficiency within the Values Tab's design: As a hospital medicine physician stated, “I would like for the Tab to involve 1 click, so I can open it, and the key content will be right there.”

Improved Communication and Relationship Building

Without prompting, 42 respondents (38%) shared their belief that this Values Tab would have the potential to improve the relationship between patients and healthcare providers and/or facilitate communication among healthcare team members. As expressed by a senior medical student: “If we did have access to that information before, we could help make the patient feel like we know them really well and we really care about them. All of this is important in building rapport.”

Privacy Implications

Although the interview guide did not explicitly inquire about privacy, some respondents (15 out of 110, 14%) spontaneously expressed concerns that patients might prefer to keep certain information private rather than displayed in the Values Tab. Half of the patient and family participants voiced such concerns, noting that a patient might be uncomfortable sharing such personal information with the full healthcare team. Privacy concerns were alleviated by the knowledge that nurses would review summaries of patients' values with them before inclusion in the EHR.

DISCUSSION

This study focused on a promising EHR innovation to centralize, display, and highlight key information about health-related values, goals, and preferences of the patients with cancer as a basis for respecting their personhood and including them meaningfully in care throughout the illness. To inform the design, development, and initial implementation of this innovation, called the Patient Values Tab, we elicited views of more than 100 institutional stakeholders through audio-recorded, in-depth, semistructured interviews that were analyzed using rigorous qualitative methods. Universally, respondents conveyed great enthusiasm for the utility and importance of the Values Tab as a single, succinct, organized source of information needed for delivery of high-quality patient-centered care. Our analysis identified themes related to content, format, and workflow that helped guide the Values Tab design process, balancing level of interest among the stakeholders with impact on user efficiency and space constraints.

Themes endorsed by the broadest cross section of stakeholders were incorporated prominently within the Values Tab. These included core aspects of personhood, social support system and resources, social history, communication preferences, and information related to future planning. Much of this information is featured in the Values Summary and Goals of Care Discussions sections in the center of the Values Tab (Appendix Fig A1, online only). Additional information is shown in Communication Preferences, Social Information, and Practices and Beliefs sections. The Values Tab displays a range of information captured in a variety of assessments by many members of the interdisciplinary team, including physicians, advanced practice providers, nurses, social workers, chaplains, supportive care consultants, ethicists, psychiatrists, and patients themselves, to illuminate the patient as a person. By prominently highlighting this information, the Values Tab encourages systematic incorporation of patient values into care.

Other themes seemed especially relevant for some stakeholder groups, while being less applicable to others. For example, end-of-life care preferences were prioritized by providers from intensive care, supportive care, oncology, hospital medicine, and pharmacy, presumably because of the frequency and depth of their involvement in care of patients at end of life. Stakeholders also varied in their prioritization of the patient's understanding of the illness course and intent of treatment. Challenges in prognostication and in assessment of illness understanding may be perceived differently across the groups we interviewed, who may also vary in the ways that they use this information for treatment planning purposes.^38-40

Although few other respondents mentioned concerns about patient privacy, half of the patients and caregivers felt that this topic was especially important. We learned from these interviews that patient-centered care entails inquiry into not only patients' values but also sensitivity in sharing such information in the EHR. By requiring patient verification of the Values Summary, we sought to balance respect for both values and privacy.

Efforts in other health systems to enrich the EHR with information about patients' values, goals, and preferences are in various stages of development. Some projects have focused primarily on the creation of orders for life-sustaining treatment (eg, Medical Orders for Life-Sustaining Treatment) and documentation of other advance care planning.^21,22 These initiatives have had a positive impact on advance care planning by older adults.^21,22 More recent initiatives have incorporated the patient's life story as transcribed by volunteers⁴¹ or broader information about the patient's care goals, along with clinician recommendations for next steps during serious illness conversations.⁴² Prior literature has used qualitative methods to investigate stakeholder perceptions about health information technology more generally^43-47 or to enhance understanding of the patient and clinician experience of conversations using a templated Serious Illness Conversation Guide.⁴⁸ Our study takes a further step by engaging a large number of key stakeholders at a comprehensive cancer center to inform the initial design and development of our Values Tab within the EHR. This rigorous approach helped elicit user needs and ensure that the Values Tab was appropriately grounded in existing workflows at our center.

There were several limitations to this qualitative research study. Because we interviewed stakeholders at a single cancer center, the generalizability of our findings to patients with other illnesses, in different geographic areas, is unknown and merits further investigation. However, we did interview a large and diverse group of respondents, patients with cancer often have other comorbidities, and our institution cares for patients not only within New York City but throughout a regional network that encompasses three states, as well as patients from around the world. All stakeholder interviews were conducted in English, and it is possible that non–English-speaking patients and clinicians might respond differently. Others within the stakeholder groups we interviewed may not have had the same views, but sampling achieved diversity both within and across groups and no individual we approached declined participation. We designed our Values Tab in MSK's EHR (ie, Allscripts)⁴⁹; however, this type of display feature can be configured in other EHR software, including EPIC (the predominant platform in the United States). Questions about portability of information in our Values Tab need further investigation.

We chose a qualitative approach to obtain in-depth feedback and ideas on content, format, and process for our Patient Values Tab in the EHR. By interviewing a large and varying sample of key stakeholders and rigorously analyzing their responses, we obtained robust results to inform the Values Tab's development and implementation. In addition, the process of conducting these interviews itself stimulated interest in the Values Tab and the information it displays, generated institutional buy-in, and created a sense of ownership among a broad group of future users, while paving the way for institution-wide implementation. As a Supportive Care clinician commented, our Patient Values Tab is “as much a consciousness-raising exercise as it is a clinical tool.” The impact of this innovation on patients and families facing cancer, the clinicians caring for them, and the health system will be the subject of further research. Already, the introduction of the Values Tab in this comprehensive cancer center signals the importance of patient personhood and values throughout the institution and advances the use of the EHR as a driver of the delivery of patient-centered care.

APPENDIX 1. Written Interview Question Guide

Interviewer's or Moderator's Guide:

“As part of a new quality improvement and research program, we're reaching out to health care providers at Memorial Sloan Kettering (MSK) to understand how we can best design a new electronic system-based ‘Patient Values Tab’ that will help providers better understand their patients' values, goals and preferences. These questions should take no more than 20 minutes of your time. If it's OK for me to audio-record our conversation, we can be sure your comments are recorded accurately. This interview will be anonymous, the transcript will be de-identified and nothing you say will be linked to you by name. You have the option to decline this interview, decline to have the interview audio-recorded, or decline to answer any individual question. Feel free to stop me at any time if you have questions or concerns.”

1. How long have you been working at MSK? What is your current role at MSK?

2. What are the most important things you want to know about your patients' values, goals and preferences? (Probe: By patient values, I mean, very broadly, what the patient considers most important in life. As an example, what does living well mean to the patient at this particular time?) (Probe: By patient goals, I mean, very broadly, what the patient is hoping to accomplish with his or her care.) (Probe: By patient preferences, I mean, very broadly, any preferences regarding receipt of care, receipt of medical information or anything else relevant to the care encounter.)

3. How would you incorporate this type of information in caring for your patients? (Probe: can you tell me more about that?)

4. Do you think that this information is currently being captured in our electronic system? (Probe: can you tell me more about that?)

5. If so, do you feel that this information is easily accessible when you need it?

6. To the extent that the information you would like to have is not currently accessible in our electronic system, can you suggest how or where this information might be captured?

7. Do you use the electronic system to document your patients' values, goals and preferences? If so, which part of the electronic system do you use to document these?

8. Is there anything else that you would like for me to know about your views or experiences related to this topic?

9. Do you have any questions or comments?

“Thank you for taking the time to speak with me today. This information will be shared, in a de-identified and anonymous fashion, with the MSK Health Informatics Committee, to ensure that we are optimizing our ‘Patient Values Tab’ design to the fullest extent. We really appreciate your valuable feedback!”

FIG A1.

A new Patient Values Tab in the electronic health record.

AUTHOR CONTRIBUTIONS

Conception and design: Anjali V. Desai, Rajiv Agarwal, Gilad J. Kuperman, Haley Mittelstaedt, MaryAnn Connor, Camila Bernal, Dana Kramer, Judith E. Nelson

Administrative support: Judith E. Nelson

Collection and assembly of data: Anjali V. Desai, Rajiv Agarwal, MaryAnn Connor, Brittany Katz, Kelsey L. Corrigan, Dana Kramer, Mary Elizabeth Davis, Judith E. Nelson

Data analysis and interpretation: Anjali V. Desai, Rajiv Agarwal, Andrew S. Epstein, Gilad J. Kuperman, Chelsea L. Michael, MaryAnn Connor, Kathleen A. Lynch, Jamie S. Ostroff, Brittany Katz, Kelsey L. Corrigan, Dana Kramer, Mary Elizabeth Davis, Judith E. Nelson

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Needs and Perspectives of Cancer Center Stakeholders for Access to Patient Values in the Electronic Health Record

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).