Proceedings of a Workshop—in Brief
Convened March 2, 2026
The Forum on Advancing Diagnostic Excellence at the National Academies of Sciences, Engineering, and Medicine hosted a hybrid public workshop on March 2, 2026, to explore opportunities for engaging patients, caregivers, and families in the diagnostic process in order to improve patient safety and reduce diagnostic errors.1,2 The workshop examined potential strategies to engage patients throughout the diagnostic process, which included strengthening patients' and caregivers' capacity to participate actively in diagnosis, enhancing clinicians' communication and partnership with patients and families, and integrating patients and caregivers more intentionally into health system design and improvement efforts. In the opening remarks, Kristen Miller of MedStar Health described how the workshop builds on the consensus report Improving Diagnosis in Health Care (NASEM, 2015) and a subsequent workshop series on Advancing Diagnostic Excellence.3 She highlighted the consensus report's key message saying that "improving diagnosis requires a shared commitment from health care professionals and organizations, researchers and policymakers, and perhaps most importantly patients, caregivers, and family members whose insights and observations are central to accurate and timely diagnosis."
Cristina Gonzalez from NYU Grossman School of Medicine described diagnosis as an iterative process that unfolds over time, beginning when a patient first experiences a symptom or concern and continues through a series of interconnected steps (see Figure1). Within this process, she framed patient engagement as an active partnership in which patients, caregivers, and families are supported in sharing critical information, asking questions, expressing concerns and goals, participating in decision making, and following up when symptoms change. She emphasized that engaging patients throughout the diagnostic process helps ensure the diagnostic team has a holistic and more accurate picture, which can support more timely and accurate diagnosis. This Proceedings of a Workshop—in Brief highlights the presentations and discussions that occurred at the workshop.4
1 The workshop agenda and presentations are available at: https://www.nationalacademies.org/projects/HMD-HCS-25-16/event/45837 (accessed March 30, 2026).
2 Achieving excellence in diagnosis refers to going beyond avoiding medical errors to ensure diagnosis is safe, timely, effective, efficient, equitable, and patient-centered.
3 More information is available at https://www.nationalacademies.org/projects/HMD-HCS-20-02 (accessed April 1, 2026).
4 This Proceedings of a Workshop—in Brief is not intended to provide a comprehensive summary of information shared during the workshop. The information summarized here reflects the knowledge and opinions of the individual workshop participants and should not be seen as a consensus of the workshop participants, the planning committee, or the National Academies of Sciences, Engineering, and Medicine.

SOURCE: Presented by Cristina Gonzalez, March 2, 2026. NASEM, 2015.
Ilene Corina, president of the Pulse Center for Patient Safety Education and Advocacy, shared how the loss of her son due to a diagnostic error inspired her work in patient safety education and support. Centering her remarks on patient engagement, Corina discussed the importance of patients taking an active role in their care and communicating effectively with clinicians to support accurate diagnosis. She highlighted Pulse's Think S.T.A.R.S. (specifics, treatment, associated symptoms, relieve or provoke the symptoms, and severity) toolkit which was designed to help patients communicate symptoms more clearly. The toolkit encourages patients to describe the specific location of pain, details about treatments and medications, any associated symptoms, factors that relieve or worsen the symptoms, and the severity of the symptoms. Corina also highlighted supporting individuals to take charge of their health care decisions through informed decision making. She pointed to practical strategies from the TakeCHARGE Campaign, including understanding and completing advance directives, keeping a record of medical history and current medications, preparing for doctor visits with a list of questions, helping prevent infections by asking caregivers and health care providers to wash their hands, and using a patient advocate when needed. Across these examples, Corina framed communication as an essential skill that health systems and organizations could help patients develop to reduce diagnostic errors.
Phalana Tiller, vice president at Bendable Labs, reflected on her diagnostic journey as a cancer survivor, including a clinical encounter in which her concern about a breast lump was initially dismissed. From that experience, she discussed learning important lessons about self-advocacy and the value of seeking a second opinion. Tiller explained that a diagnosis is strengthened when patients feel encouraged to question an initial assessment, ask for clarification, and revisit concerns when something does not feel right. She emphasized that meaningful patient engagement depends on trust—in both directions—in which clinicians need to trust patients as experts in their own experiences, and patients need to trust that clinicians are listening, respectful, and willing to partner with them. Building trust between patients and clinicians is essential to more accurate diagnosis, Tiller said.
The first panel featured individuals discussing engagement opportunities for patients, caregivers, and families. Hugo Campos, a patient advocate and health care researcher, described his experiences using artificial intelligence (AI) tools to actively engage in diagnosis. He discussed using ambient
shifts asking the patient "What is wrong with you?" to "What happened to you?"
Washington and Williams specifically described how these barriers can impact early stroke detection and treatment. Washington explained that delayed arrivals for stroke treatment are often seen as a patient behavior problem, but community engagement revealed that limited understanding of stroke symptoms, mistrust of the health care system, financial barriers, and different perceptions of when to seek care can lead to delays. Williams described his team's National Institutes of Health (NIH)-funded randomized trial in which a hip-hop based stroke preparedness intervention was provided for school children and families in Central Harlem (Williams et al., 2018). They created a child-mediated health communication model in which critical health education flowed from children to parents. He reported that this school-based strategy was associated with increased stroke treatment rates showing that innovative approaches incorporating a community's lived experiences can improve diagnostic outcomes.
Washington stressed the need for authentic engagement, which includes meeting people in trusted community settings, listening to communities discuss their own needs, and adapting health care approaches accordingly. Both Washington and Williams emphasized co-design, community-based participatory research, and team-based care as opportunities. Williams highlighted the value of community health workers, patient advocates, and care navigators as diagnostic team members who can often communicate more effectively with patients and extend support beyond the clinic.
Alexis Snyder, a patient and community engagement specialist, shared her personal experience as a caregiver to her daughter, who lives with chronic complex health needs, and described approaches to partner with caregivers to strengthen engagement and enhance the diagnostic process. Snyder described the long diagnostic journey her family faced, noting that "living with the uncertainty of a diagnosis, and multiple misdiagnoses for over seventeen years, could have been shortened if not only her [daughter's] voice as a patient was heard more often but if the caregiver voice was not dismissed or overlooked." She explained that even without a diagnosis, her daughter's symptoms might have been better managed and her quality of life improved if patient and caregiver observations had been heard and valued earlier.
Snyder described engagement as an active partnership among patients, families, and clinicians that can help improve health care outcomes. She highlighted the crucial role caregivers play, including helping patients prepare for visits, providing real-time observations and insights that help bridge gaps, facilitating communication with clinicians to help avoid misunderstandings, helping patients understand information and remember details from visits, supporting shared decision making to ensure diagnosis is patient-centered, and helping patients follow through at home with care plans and treatments. She emphasized that active caregiver engagement can build trust and stronger patient-clinician relationships, help manage chronic illness, increase patient satisfaction and clinical outcomes, reduce the risk of complications and avoidable hospitalizations, and provide emotional support that decreases patient anxiety and helps patients stay engaged in their care.
To better engage caregivers, Snyder offered several strategies: identify the caregiver by asking who they are rather than making assumptions; ensure they are included in the room and not excluded from visits; make eye contact and include them in the conversation; validate their role and the difficulty of caregiving; seek their observations and input rather than waiting for them to speak up; ask if they have questions or concerns; engage them in shared decision making; avoid medical jargon and ensure instructions are understood; thank them for being part of the team; and encourage continued communication. She emphasized the importance of including caregivers in the room and not excluding them from visits, noting that she had attended many appointments with her family members where she was told the room was too small or that she could wait in the waiting room. Referring to a cartoon in which a caregiver brings their own chair because only one chair is available, Snyder illustrated how caregivers are often treated as bystanders rather than as members of the diagnostic team. Snyder also stressed that successful caregiver engagement includes recognizing that caregivers themselves are at risk for health problems, stress, and depression. She encouraged clinicians to practice empathy, acknowledge the emotional burden of caregiving, ask how the caregiver is doing, connect them with support groups, social workers,
and educational resources, and help with coordination rather than leaving everything to the caregiver. Snyder concluded by saying, "be kind, be patient, don't be afraid of caregiver input, and please don't make me bring my own chair."
Allyson Bontempo, faculty affiliate with Rutgers University, focused on strategies for strengthening patient-clinician communication to prevent dismissal of patients' symptoms and concerns. She described the medical encounter as the creation and exchange of stories between patient and clinician framing it as a co-creation of diagnosis between patients and clinicians. She identified validation as a key strategy and explained that validation can be communicated explicitly through verbal comments, which she referred to as topographical validation, or implicitly through responses that imply one takes the other's experience as valid, which she referred to as functional validation. In the context of diagnosis, she stated that functional validation includes information gathering strategies such as taking a clinical history, conducting a physician examination, completing diagnostic testing, and engaging in referral and consultation.
Bontempo described several levels of validation. The first level involves listening to and observing what the patient is saying, feeling, and doing and making an active effort to understand what is being communicated. This demonstrates interest in the patient, she said, and many patients describe feeling dismissed when clinicians do not seem interested in learning more about their symptoms. The second level involves accurately reflecting back to patients what they have said to show comprehension and arrive at a shared understanding of the material. The third level is the articulation to the patient of the clinician's understanding of aspects of the patient's experience and response to events that have not been verbalized. The fourth level is validating behavior in terms of sufficient causes by communicating that there is some cause for what the patient is experiencing, even if it is not yet known. The final level is the validation of the patient's behavior as reasonable in terms of current events and the patient's health care goals.
Bontempo stated patient safety scholars and patients are increasingly advocating for disclosure of uncertainty in the patient–clinician encounter. She noted that one study found that clinician disclosure of diagnostic uncertainty can increase trust in clinicians and facilitate shared decision making (Bontempo, 2023). Bontempo encouraged accompanying the disclosure of uncertainty with a verbal commitment to follow up on the patient's behalf, which can include consulting with colleagues, conducting further research, referring patients to a specialist, and scheduling follow-up appointments. She observed that these actions can be validating because they allow patients' symptoms to be recognized as real despite the absence of a known diagnosis. Finally, she cautioned against reassuring patients that nothing is wrong, especially early in the appointment, because patients may perceive it as dismissive. She also discouraged using mental illness as a scapegoat diagnosis when a medical cause cannot be demonstrated.
Naykky Singh Ospina, associate professor of medicine at the University of Florida, presented on the clinician's role in aligning patients' goals with diagnostic care. She noted that patient-centered diagnosis does not happen organically in the current health care system and that, although many forces shape the diagnostic pathway, "the patient should be at the center." Clinicians, she said, have significant influence over whether patient-centered diagnosis occurs during a clinical encounter and can either enable or obstruct patient-centered diagnosis. Once clinicians recognize this influence, she highlighted, the goal is to co-create a diagnostic plan of care with the patient, which includes understanding the patient's clinical situation and what matters most to them.
To illustrate clinicians' influence, Singh Ospina noted that clinicians often set the agenda and control the pace of conversation during clinical visits. She mentioned that clinicians explicitly ask patients why they are there or elicit their concerns in only 36 percent of visits (Singh Ospina et al., 2019). Even when the patient's agenda is elicited, the first interruption occurs after just 11 seconds (Singh Ospina et al., 2019). Singh Ospina outlined three approaches to support patient-centered diagnosis. First, recognizing the diagnostic situation involves eliciting the patient's agenda and listening without interruption to understand the problem. Second, prioritizing diagnostic situations, which includes considering the patient's input, comorbidities, and other circumstances, so that care aligns with what matters most to the patient. Third, co-developing a feasible, desirable, and evidence-based plan with the patient is important. She also discussed communication about diagnostic evaluation, including helping patients understand why a test or referral is needed,
what it entails, the logistical details, and how results and next steps will be communicated. Although clinicians explain why a test is needed, she noted they need improvement in explaining what the test will involve, how results will be shared, and how to address common "what if" scenarios that patients face.
Finally, Singh Ospina discussed ways to remove barriers to patient-centered diagnosis, including value redesign that prioritizes care over productivity, workflow redesign that builds in time for diagnostic conversations, training and culture change that emphasizes co-creation of care plans, and focused diagnostic interventions for subspecialists to reduce distractions. She concluded that "the clinical encounter is where policy and evidence can become care."
During the discussion, the panel speakers considered how AI might impact patient engagement in the diagnostic process. Singh Ospina said that although AI can be useful for some tasks, its accuracy in assisting patients can vary widely, and it is another intervention that will need science to guide its implementation in medicine. Snyder added that AI can help scan large amounts of information in the electronic health record and help patients and clinicians organize complex histories before a visit but cautioned that its usefulness depends on the quality and accuracy of the information it receives. When that information is incomplete or inaccurate, she said AI may perpetuate errors. Panelists also discussed that effective patient engagement requires communication that makes patients feel heard, respected, and included in the diagnostic process. Bontempo proposed agenda setting at the start of a visit as a practical way to surface patient concerns early and use the limited time more effectively. Snyder emphasized listening and validation, noting that clinicians may reach a diagnosis more quickly when they treat patient and caregiver observations as critical information. Overall, the panel highlighted potential opportunities that may strengthen a clinician's role in patient engagement through thoughtful use of AI, clear communication strategies, and a stronger commitment to listening and partnership.
Nancy Netherland, caregiver engagement specialist at Family Voices of California, described her experiences as a caregiver to her two daughters with special health needs and provided strategies for designing health systems to empower patients in the diagnostic process. "The diagnostic process is not one thing. It is not one journey. In my family, it has been two radically different experiences." Netherland noted that diagnosis in pediatrics requires a different approach developmentally, cognitively, and relationally, yet health systems do not always reflect that reality. She emphasized that in pediatric medicine, the patient rarely arrives alone but instead as part of a caregiver–child dyad. The caregiver, she said, can provide the most consistent longitudinal data source in the diagnostic process. When caregivers are dismissed, irreplaceable information can be lost; when they are treated as partners, better diagnoses can be achieved.
Netherland described the contrasting diagnostic journeys of her two daughters. Her first child received a definitive and early diagnosis; however, her second child's journey was very different. By age three or four, she was in the bottom five percent of growth charts, had elevated inflammatory markers, recurrent fevers, ulcers, and significant pain, yet a gastroenterologist specialist interpreted these symptoms as evidence of maternal anxiety. Netherland said it took a child psychologist, pediatric psychiatrist, a multidisciplinary family meeting, extended hospitalizations, repeated requests from the pediatrician, and discussion about possible feeding tube placement before an endoscopy was performed and a diagnosis was made. She said her daughter's health declined for two years, while both her daughter's symptoms and her own observations were dismissed. She added that finding clinicians who treated both her and her daughter as partners changed and improved the diagnostic process.
Netherland also explained that pediatric symptom-reporting tools essential to diagnosis were not designed for children, especially not for her daughter, who had language limitations and learning differences. In response, her family created their own forms that centered her child's voice and experience by using images and illustrations her child could comprehend to track symptoms. She also discussed creating spaces where her children could ask clinicians questions through "gratitude tools" to build rapport and comfort, foster kid-centered dialogue, and surface clinical information that standard symptom checklists might miss. She further noted that trauma shaped how children participate in diagnosis, leading her family to create additional tools for moments when a
child was too overwhelmed to speak. Netherland concluded, "Diagnostic excellence in pediatrics requires that we stop treating the caregiver–child dyad as noise or an anomaly and instead recognize it as the most consistent longitudinal data source, especially in complex systems with multiple silos of care."
Sheila Moroney, executive director of Patient Revolution, discussed empowering patients in diagnosis through careful and kind system design. She described stories of two highly engaged patients who were eager to partner with their care teams, yet both struggled to be heard. Moroney described Liel, whose diagnostic odyssey began at age three and lasted 18 years. Over the course of Liel's life, she received multiple diagnoses, but her health sharply declined at age 18, leading to dozens of hospitalizations and surgical procedures. To help clinicians see patterns in her condition, Liel compiled lab results over a seven-year period from eight patient portals into a 27-page spreadsheet, highlighting abnormal lab values in red, and brought this comprehensive notebook detailing her diagnostic journey to each of her clinical encounters. Moroney said clinicians still dismissed Liel's concerns and sometimes attributed her symptoms as manifestations of mental health issues; Liel died at age 21, after finally receiving a diagnosis of a rare mitochondrial disease. Moroney also shared Earl's story. Earl, an older adult, experienced declining health and repeatedly told clinicians, "Something is not right. I cannot get my air." Although he raised this concern consistently, it took nine months for him to get a CT scan which revealed metastatic lung cancer, and he died three weeks later.
Moroney reflected on both Liel and Earl emphasizing they were informed, persistent, and motivated to get the right diagnosis. She noted the system failed to recognize and respond to the clues that they provided and highlighted building on these failures through a careful and kind system. Signs of care in a careful and kind system include seeing people in high definition, understanding both their biology and biography, minimizing disruption in their lives, unhurried conversations, and being responsive. She stressed that the quality of time matters as much as the quantity, which means reducing distractions, ensuring clinicians have what they need in the room, and asking patients, "What matters to you?" Moroney concluded, "eliciting stories and input from patients to understand the situation, not just the problem, is truly the essence of shared decision making and necessary for diagnostic excellence."
Jane Evered, a nurse and health experiences researcher at the University of Wisconsin-Madison, described how her team uses systematically collected patient and care partner narratives to learn about their experiences and improve diagnostic care. Using mixed methods approaches and large, diverse samples, the team studies patterns in patient experiences of diagnostic problems, mistakes, and remedies. Evered emphasized that patients and care partners with lived experiences are often the consistent connecting thread across fragmented diagnostic journeys and have many creative ideas for system change.
She described the Narrative Elicitation Protocol that elicits people's experiences with diagnostic problems and mistakes, such as what happened, how communication unfolded, what helped or worsened the situation, what patients wished had been done differently, the aftermath, effects on future care-seeking, whether they had someone accompany them, and whether identity shaped the experience. Across the three surveys, more than 5,000 narratives were collected. Evered noted that evidence shows narrative elicitation tools are an important form of patient engagement and empowerment that can generate transferable knowledge across representative national samples and foster more engagement and improved patient experience scores in a virtuous cycle (Grob et al., 2019, 2024; Martino et al., 2023).
These narratives pointed to several recurring structural contributors to diagnostic problems, including short visit times, siloed specialty care, triage and scheduling breakdowns, and clinical guidelines that may not fit all populations. Evered's team also analyzed the narratives to understand the process of patients identifying a person they could rely on as a diagnostic guide to serve as a consistent source of guidance and reliable support during the diagnostic process. She noted qualitative findings suggested that meaningful diagnostic guides, including clinicians or specialists, and sometimes family and friends, are knowledgeable and willing to keep learning, listen deeply without rushing or making assumptions, take concerns seriously, provide ongoing accompaniment and coordination, and demonstrate trustworthiness. Evered reported that regression analyses suggest that clinician diagnostic guides
are associated with resolving diagnostic mishaps, buffering short- and long-term loss of trust in clinicians, and lowering future avoidance of medical care after a diagnostic problem. She provided considerations for health systems to better support access to diagnostic guides, coordinate across teams and institutions, and maintain structures that help patients stay connected throughout the diagnostic process.
Donna Galbreath, senior medical director at Southcentral Foundation, presented on customer ownership in Southcentral Foundation's Nuka System of Care. Through the Indian Self-Determination and Education Assistance Act of 1975, the federal government allowed Indian tribes greater autonomy and responsibility over programs and services delivering health care to Alaska Native people. Galbreath emphasized that the government recognized the value of people being involved in and owning their health care to improve overall health. She said Alaska Natives chose to assume this responsibility and are referred to as customer-owners rather than patients. She noted that Southcentral Foundation serves 229 federally recognized tribes across Alaska and co-manages the Alaska Native Medical Center, a tertiary hospital serving the entire state.
She described Southcentral Foundation's approach as grounded in a strong foundation shaped by customer-owner feedback and guidance, with relationships at the center of the Nuka System of Care. The model was developed in response to what customer-owners and employees said they wanted: respectful interactions, less waiting, more input in decisions, continuity with their own health care provider, and better facilities. Customer-owners emphasized not only their own care but also the health of families and communities, which led to programs centered on elders, veterans, and family wellness. Galbreath also highlighted the organization's core concepts training for all employees, which reinforces the system's operating principles, relationship-based care, listening to people's stories, and recognizing the strengths each person brings. She also described integrated primary care teams, consisting of the health care provider, the case manager, case manager support staff, and certified medical assistants, all working together to improve care.
Responding to customer-owner feedback about access and ease of use, Nuka offers same-day access to health care providers and integrated support from behavioral health consultants, pharmacists, dietitians, pediatricians, midwives, and social workers. These services are embedded alongside care teams so customer-owners can often receive additional support during the same visit. Galbreath highlighted that listening to customer-owners drives continuous improvement and has contributed to strong results, including a 44 percent reduction in emergency room visits between 2000 and 2004; a 43 percent reduction in inpatient discharges between 2000 and 2003; a 33 percent reduction in outpatient visits between 1999 and 2023; 98 percent customer-owner satisfaction and 90 percent employee satisfaction in 2024; and Healthcare Effectiveness Data and Information Set (HEDIS) quality scores above the 75th percentile.7 She stressed that customer-owners are active partners in their health care, seeking information, asking questions, and discussing treatment options with health care providers rather than simply receiving instructions. Galbreath described the customer-owner voices as continuing to shape the system through governing boards, advisory committees, the Elder Council, a 24-hour hotline, and regular surveys.
During the discussion, Netherland, Moroney, Evered, and Galbreath emphasized intentionally integrating patients and caregivers into health systems to better support diagnosis. Netherland focused on embedding patient partnership into formal decision-making structures, such as leadership councils, advisory groups, managed care plans, and innovation centers. She encouraged health systems to create advisory opportunities and invite caregivers and patients to participate in improvement efforts. She also highlighted the need for stronger patient-centered research to build the evidence base for integrating patients at the systems level.
Galbreath described how Southcentral Foundation's Nuka model supports this integration through short, co-designed surveys that ask customer-owners whether they felt involved in their care, respected, and treated in a culturally appropriate way. She highlighted the role of infrastructure and accountability mechanisms in helping the system continuously improve in response to customer-owner needs. Health provider training and transparent health provider performance data, she said, have helped reinforce customer-owner partnership as a standard of care.
7 For more information, see https://scfnuka.com/about-us/ (accessed June 11, 2026).
The panelists also discussed broader organizational approaches to supporting patients in diagnosis. Moroney said her organization, the Patient Revolution, seeks to raise awareness and equip people with tools for patient empowerment. She suggested that training in careful and kind care principles extend beyond future clinicians and nurses to include health care administrators who design, lead, and steward care systems. To sustain a careful and kind system, Moroney emphasized bringing together insurers, regulators, elected officials, administrators, clinicians, and patients to define shared goals and clarify each group's role. Evered added that patients often feel abandoned after diagnostic problems because they do not feel heard and mistakes are often not acknowledged. She suggested that systems can better support patients in taking ownership of their health, especially when barriers arise, such as difficulty obtaining a second opinion or changing clinicians. Overall, this session's panelists centered their discussion on health systems designed to support patients through partnership, transparency, and shared responsibility.
In the final session, moderated by Charity Watkins of North Carolina Central University and Saul Weingart of Tufts Medical Center, panelists reflected on key themes from the workshop. The panel included Carole Hemmelgarn of MedStar Health and Patients for Patient Safety US, Julia Adler-Milstein of University of California San Francisco, Karen Cosby of Rush University, Gordon Schiff of Brigham and Women's Hospital, and earlier workshop panelist Phalana Tiller.
Several panelists described meaningful engagement as partnering with patients, caregivers, and families throughout the diagnostic process, with particular emphasis on communication, listening, and trust. Hemmelgarn highlighted the power of stories, noting that "storytelling is data with a soul," and explained that diagnosis depends on patients having enough time to explain what is happening to them. Schiff described diagnostic communication as bidirectional between a patient and clinician. He emphasized listening to patients in a non-dismissive, empathetic way followed by open communication about uncertainty that is transparent and non-defensive. Hemmelgarn also noted that patient engagement can be supported through verbal and nonverbal tools that encourage patients and caregivers to ask questions, raise concerns, and participate more fully in diagnostic encounters.
A few panelists also discussed the role of culture, access, and system navigation in shaping patient engagement. Tiller emphasized the importance of culturally relevant approaches and reflected on examples raised earlier in the workshop, including the use of nontraditional messengers and community-based supports to meet patients where they are. She also challenged conventional uses of the term ‘health literacy,' asking whether patients are being asked to be "literate enough to be compliant" rather than empowered. Cosby also discussed health literacy but highlighted the need for health system competency to help patients understand how to navigate a fragmented and often unfriendly health care system.
Infrastructure and system design were also described as key elements in improving patient engagement. Referring to the current diagnostic framework model, Cosby suggested developing a broader framework that includes what happens before a patient seeks care and after they leave the system. She noted this phase in the diagnostic process is critical because many diagnostic delays and misses occur before a patient even decides to seek care. Schiff similarly described the need for proactive, systematic redesign in which patients have a defined role in follow-up, communication of symptoms, understanding contingencies, and navigating referrals and diagnostic testing supported by trust, continuity, and changing power dynamics.
Panelists discussed both the opportunities and risks of using technology, including AI, to support patient engagement. Hemmelgarn described AI as "here to stay" and suggested that AI may help patients better understand their symptoms and enter diagnostic encounters with more information, helping to narrow the traditional power differential. Adler-Milstein reflected on the workshop's discussion of both direct and indirect benefits of these tools. She described an indirect benefit as reducing clinicians' busy work, so they have more time to engage deeply with patients and their stories. She described a direct benefit as using technology to process vast quantities of information, especially in cases of complex diagnoses, to support a more timely and accurate diagnostic trajectory that enhances the patient's experience of the diagnostic process. At the same time, she cautioned
that patients and caregivers may not be able to distinguish trustworthy from untrustworthy AI-generated guidance. She emphasized supporting patients' use of tools so the information they receive is accurate and productive, and she noted that patients cannot be expected to know how to interact with these tools using the kinds of prompts, language, or terminology they often require and that may not be easily understood.
Cosby and Adler-Milstein also highlighted accountability as an important area for improving diagnosis. Adler-Milstein noted that there is still no clear accountability for the patient's experience of the diagnostic process. She suggested adding patient-reported measures of diagnostic experience to quality assessment and linking payment models to those measures. She also pointed to the importance of capturing patient symptom data more effectively, including in patients' own words to improve future tools and systems.
In closing, Watkins and Weingart noted that panelists highlighted opportunities to redesign health systems so that patients are respected, heard, supported, and included throughout the diagnostic process. Across the discussion, panelists pointed to the need to align incentives and accountability with what matters to patients and to ensure that communication, culture, infrastructure, and technology work together to advance diagnostic excellence.
Suggestions from workshop participants for engaging patients to improve diagnostic excellence are outlined in Box 1.
NOTE: This list is the rapporteurs' summary of points made by the individual speakers identified, and the statements have not been endorsed or verified by the National Academies of Sciences, Engineering, and Medicine. They are not intended to reflect a consensus among workshop participants.
Bontempo, A. C. 2023. Patient attitudes toward clinicians' communication of diagnostic uncertainty and its impact on patient trust. SSM-Qualitative Research in Health 3:100214.
Grob, R., Y. S. Lee, D. Shaller, E. Warne, S. Matta, M. Schlesinger, and I. M. Nembhard. 2024. "Nothing is more powerful than words": How patient experience narratives enable improvement. Quality Management in Healthcare 33(3):149–59.
Grob, R., M. Schlesinger, L. R. Barre, N. Bardach, T. Lagu, D. Shaller, A. M. Parker, S. C. Martino, M. L. Finucane, J. L. Cerully, and A. Palimaru. 2019. What words convey: The potential for patient narratives to inform quality improvement. The Milbank Quarterly 97(1):176–227.
Khan, A., M. Coffey, K. P. Litterer, J. D. Baird, S. L. Furtak, B. M. Garcia, M. Ashland, S. Calaman, N. C. Kuzma, J. K. O'Toole, A. Patel, and C. E. Yu. 2017. Families as partners in hospital error and adverse event surveillance. JAMA Pediatrics 171(4):372–81.
Martino, S. C., K. A. Reynolds, R. Grob, A. I. Palimaru, S. Zelazny, M. E. Slaughter, L. Rybowski, A. M. Parker, S. L. Toomey, M. A. Schuster, and M. Schlesinger. 2023. Evaluation of a protocol for eliciting narrative accounts of pediatric inpatient experiences of care. Health Services Research 58(2):271–81.
NASEM (National Academies of Sciences, Engineering, and Medicine). 2015. Improving diagnosis in health care. Edited by E. P. Balogh, B. T. Miller, and J. R. Ball. Washington, DC: The National Academies Press.
Singh Ospina, N., K. A. Phillips, R. Rodriguez-Gutierrez, A. Castaneda-Guarderas, M. R. Gionfriddo, M. E. Branda, and V. M. Montori 2019. Eliciting the patient's agenda-secondary analysis of recorded clinical encounters. Journal of General Internal Medicine 34(1):36–40.
Williams, O., E. Leighton-Herrmann Quinn, J. Teresi, J. P. Eimicke, J. Kong, G. Ogedegbe, and J. Noble. 2018. Improving community stroke preparedness in the HHS (Hip-Hop Stroke) randomized clinical trial. Stroke 49(4):972–9.
Disclaimer: This Proceedings of a Workshop—in Brief was prepared by Jennifer Lalitha Flaubert and Adrienne Formentos as a factual summary of what occurred at the workshop. The statements made are those of the rapporteurs or individual workshop participants and do not necessarily represent the views of all workshop participants; the planning committee; or the National Academies of Sciences, Engineering, and Medicine.
Planning Committee:Cristina Gonzalez (Cochair), NYU Grossman School of Medicine; Kristen Miller (Cochair), MedStar Health; Gene Harkless, University of New Hampshire; Helen Haskell, Mothers Against Medical Error; Kathryn McDonald, Johns Hopkins University; Cheryl Phillips, John A. Hartford Foundation; Mark Schlesinger, Yale University; Anjana Sharma, University of California, San Francisco; Susan Sheridan, Patients for Patient Safety US; Charity Watkins, North Carolina Central University; Saul Weingart, Tufts Medical Center. The National Academies' planning committees are solely responsible for organizing the workshop, identifying topics, and choosing speakers. Responsibility for the final content rests entirely with the rapporteurs and the National Academies.
Reviewers: To ensure that it meets institutional standards for quality and objectivity, this Proceedings of a Workshop—in Brief was reviewed by Karen Cosby, Rush University and Cook County Health (Retired); Traber D. Giardina, Baylor College of Medicine; Michael E. DeBakey, VA Medical Center; and Alexis Snyder, Independent Consultant. Kirsten Sampson-Snyder, National Academies of Sciences, Engineering, and Medicine, served as the review monitor.
Sponsors: This workshop was supported by the American Association of Nurse Practitioners; American Academy of Physician Associates; American Board of Emergency Medicine; American Board of Internal Medicine; American College of Radiology; College of American Pathologists; The Doctors Company; The Gordon and Betty Moore Foundation; The John A. Hartford Foundation; The Mont Fund; Patients for Patient Safety US; and Radiological Society of North America. Any opinions, findings, conclusions, or recommendations expressed in this publication do not necessarily reflect the views of any organization or agency that provided support for the project.
Staff:Jennifer Lalitha Flaubert, Adrienne Formentos, Torrie Brown, Joseph Goodman, Tracy Lustig, and Sharyl Nass. Health Care and Public Health Program Area, Center for Health, People, and Places, National Academies of Sciences, Engineering, and Medicine.
Suggested citation: National Academies of Sciences, Engineering, and Medicine. 2026. Engaging Patients to Advance Diagnostic Excellence: Proceedings of a Workshop—in Brief. Washington, DC: National Academies Press. https://doi.org/10.17226/29483
Copyright 2026 by the National Academy of Sciences. All rights reserved.