Patient Experience as a Source for Understanding the Origins, Impact, and Remediation of Diagnostic Errors

Past research on eliciting patient experience shows the capacity of patients and families to convey accounts about many aspects of their healthcare encounters.⁶ But achieving more consistent reporting of these accounts requires attention to rigor in the design, testing, and implementation of data collection about diagnostic experiences.

For each of these aspects of rigor, additional research could clarify the limits of existing elicitation techniques for diagnostic experiences, while also identifying ways to enhance fidelity in the future. We explore these opportunities further in the second volume of this issue brief, identifying some of the most pressing needs for additional research.

Additional studies could also help map out an agenda for implementing multiple channels of feedback about patients’ diagnostic experiences, embedding those channels within learning health systems used to drawing on diverse patient experiences. We identify below several promising opportunities for future research in each of these domains.

Inviting Participation: Two key issues related to participation merit additional study. First, it is essential to move away from previous practices of asking respondents if they experienced medical or diagnostic “error.” One can frame alternative invitations in many ways, and it remains unclear which approaches encourage the broadest participation and most representative set of reported experiences.

Second, it is also clearly essential to gather accounts from family members (or other close acquaintances) to learn about the care of patients unwilling or unable to recount their own experiences. But not all secondary observers are equally prepared to convey a balanced and complete account of the diagnostic process. It may be useful to clarify criteria for when family respondents are more or less likely to provide thorough accounts. Criteria may include the closeness of caregivers’ relationships with patients or the extent to which they participated in clinical encounters during diagnosis. Research on these criteria could enrich future elicitation efforts.

Drawing on Narrative Accounts To Address Diagnostic Inequities: Past research has documented multiple ethnic and racial disparities in diagnostic performance.⁴⁹ These quantitative studies reveal the magnitude of diagnostic differences but can rarely identify causes. The availability of actionable detail in narrative accounts, by contrast, can illuminate sources of disparities. More compellingly, such details can also facilitate a transformative shift in discourse: from a focus on disparities that is inherently about counting and comparison to one that offers an enriched exploration of diagnostic inequities.

Future research could leverage accounts of patients’ experience toward this end in two ways. The first would use narratives to identify healthcare experiences respondents perceive to be unfair or unjust. The challenge here is to identify how best to encourage patients and family members to report situations in which they have felt unfairly treated, without predetermining specific sorts of “unfairness.” Open-ended narrative inquiries, if carefully worded, have virtues in exploring these matters, because they can allow respondents to define what constitutes evidence of unfairness from their own perspective.

A second approach would look beyond respondents’ own views about fairness, by looking for patterns in the types of diagnostic shortfalls described by different subgroups of patients and families and mapping these into existing conceptual frameworks of healthcare inequities. Both approaches can enrich our understanding of the origins of diagnostic inequities and the points of leverage through which diagnostic experiences can be made more just.

Enhancing the Legitimacy and Future Use of Elicited Patient Feedback: Two key questions need to be explored to more effectively use information in elicited patient accounts to change diagnostic practices. The first question involves identifying the agencies (government or nonprofit) Americans would most trust to field population-based surveys of diagnostic experiences. Thorough narrative responses are predicated on a modicum of trust. But we need research to better understand how responses vary across states, practice settings, and sociodemographics of respondents.

Additional research is also needed about effective feedback techniques when diagnostic episodes extend across multiple settings, as is frequently the case.ⁱⁱ Feedback methods that have been tested to date have been applied to single clinical settings. When multiple settings and multiple clinicians are involved, appropriately assigning responsibility and analyzing improvement possibilities are crucial first steps in assessing appropriate remedial action. To be sure, when settings are not all under the auspices of a single health system, the process of doing so may become quite complicated and may even prove counterproductive. Identifying the most appropriate and effective techniques calls for careful study.

Thus, a variety of promising opportunities are available for future research. Some are more immediately feasible, while others might benefit from findings generated by other studies on the list. To provide a simplified roadmap, we have summarized these options in Exhibit 3.

Exhibit 3. Research priorities related to patient experiences with diagnosis

^ii. Our own unpublished data suggest that about half of the perceived diagnostic shortfalls reported on population-based surveys involve multiple settings, and 30 percent involve three or more settings.