Patient Preferences in Cases of Inter-system Medical Error Discovery (IMED)

Introduction

Transparent communication is a major tenet of the optimal response to medical errors.¹ Major professional organizations, including the American Medical Association (AMA), consider disclosure to be a duty of the physician responsible for the error, and physicians agree in principle that errors should be disclosed to patients.² Most importantly, patients and families are unanimous in their desire for disclosure, citing a preference to be informed and maintain autonomy regarding future care.^3,4

While a responsible physician discovering her/his own error has a relatively clear duty and process for disclosure, the optimal process for disclosure of inter-system errors discovered by another physician is uncertain. In cases of inter-system medical error discovery (IMED), in which an error is discovered by a provider in a system different from the one in which it occurred and is not known to the responsible provider, optimal disclosure practices are particularly complex. The triangulation of relationships between two physicians and a shared patient introduces novel disclosure concerns, but inter-system errors are particularly fraught because they lack the established channels of communication that exist between physicians within the same system, and they are more likely to jeopardize future referrals.⁵ Our previous work has shown that physicians often fail to disclose other physicians’ errors to patients when the two providers work in different systems.⁶ They fear potential legal repercussions, increasing patient distress, and damaging professional relationships. Since patients’ perspectives about IMED disclosure are currently unknown, this study is a necessary response to the physician interviews.

Understanding patient preferences may provide an impetus for physicians to change their disclosure practices. This qualitative study seeks to determine patients’ preferences for disclosure in cases of IMED, their reasons behind those preferences, actions they would likely take, and factors that would affect their preferences and actions. Our specific research questions were: 1) Do patients’ preferences for disclosure differ in cases of IMED compared to instances self-discovered error disclosure? 2) If patients desire disclosure in cases of IMED, what are their reasons for that preference, and what can mediate their preference? 3) In cases of IMED disclosure, what actions would patients likely take, and what would moderate those actions?

Methods

We conducted semi-structured telephone interviews from January to March of 2018 with 30 patient volunteers in Michigan. The interviews sought to explore participants’ perspectives as patients on disclosure of medical errors, particularly when the error is discovered between different health systems. The Institutional Review Board approved the study. This study is reported according to the Consolidated Criteria for Reporting Qualitative Research.⁷

Participants were eligible if they were at least 18 years of age, spoke English, and had seen a medical provider within the last two years. They were recruited via the UMHealthResearch.org study recruitment website. Among the many volunteers expressing interest within the first month of enrollment (N=359), purposive sampling was used to select a diversity of participants with respect to self-reported gender, age, and race/ethnicity. Minority racial/ethnic groups were oversampled, in order to provide adequate representations of perspectives from these groups. A $20 gift card was offered to those completing the study.

Preliminary drafts of the semi-structured interview guide were discussed with three members of the public, as well as with other experts in ethics, social sciences and qualitative methods, in order to improve clarity and refine the design. Probes were added to the interview guide after 1, 3, and 14 interviews to iteratively and more systematically explore themes mentioned by participants. Interviews were structured to explore participants’ views on disclosure of medical errors in two hypothetical scenarios: 1) when a physician discovers their own error, i.e., the self-discovered scenario, and 2) when a physician discovers an error made by another physician in another health system, i.e., the IMED scenario. A similar series of questions was asked regarding both scenarios, with a few additional questions that were only applicable to the IMED scenario (see Supplementary Material A). We chose this parallel design to more clearly present the IMED topic and for comparison, since it is well established that most patients and families strongly prefer disclosure of self-discovered medical errors. Interviewers were free to ask additional probing questions to elicit participants’ reasoning.

Verbal consent was given prior to each interview. Interviews began by defining a medical error in lay terms. Then, a hypothetical medical error vignette was read aloud, wherein the primary care physician responsible for the error (failure to order a biopsy as recommended after a mammogram) discovered it herself/himself. Later, an IMED vignette was read aloud wherein the patient discovers a breast mass during a self-exam and self-refers to a specialist, who then discovers the primary care physician’s prior error (again, failure to order a biopsy). These parallel vignettes were designed to avoid complicating questions about the logistics of referral and to standardize the findings. At the end of the questionnaire, participants were asked to agree or disagree on a Likert scale with the statement, “All in all, doctors are totally honest with their patients.” Immediately following the interview, a brief demographic questionnaire was conducted (see Supplementary Material B). Member checking was not part of the study design.

Three researchers (JM, AA, AS) conducted interviews with 30 participants, most approximating 30 minutes in length. JM is a masters-level trained, experienced interviewer and she personally trained AA and AS and supervised their initial interviews. Interviews were transcribed verbatim, de-identified, reviewed for quality control, and imported to NVivo 11 (QSR International) qualitative data analysis software. Thematic coding was conducted concurrently with interviews in order to iteratively assess recruitment needs. Interviews were conducted well beyond thematic saturation, defined as the time when no new themes or major dimensions of themes were identified by interviewers or coders.⁸ A codebook with detailed definitions and cross-references was developed (see Supplementary Material C). Some potential codes were anticipated a priori based on previous research on error disclosure, while others were developed inductively. Each interview was coded independently by one researcher (JM) as well as by one of three other researchers (AA, AS, LD) with consensus reached through discussion.

Analysis proceeded following the inductive approach of interpretive description.⁹ Interpretive description is a constructivist and pragmatic methodology that seeks to understand clinical phenomena by interpreting participants’ subjective experiences with the aim of informing clinical understanding. The analysis included biweekly research team meetings to compare cases and discuss newly-identified themes using the constant comparative method.⁹ Questionnaire data were imported to NVivo 11 to facilitate exploration of relationships among case variables and qualitative coding. In exploratory fashion, we tested hypotheses about relationships among themes, codes, and cases through data queries in NVivo. Qualitative and categorical data were abstracted to an extensive matrix to facilitate comparison of themes between cases and between the two scenarios. We discussed the significance of prevalence and frequency in the data while also examining contrary and outlying cases to seek deeper understanding and to check researcher bias.

Results

The sample (N=30) was 50% female, 47% racial or ethnic minorities (oversampled), and had an average age of 46 years (standard deviation 16 years). This sample was highly educated, with 80% holding at least a 4-year college degree, compared to an average of 30% of Americans. One third (n=10) of the sample had experience in healthcare-related fields, compared to a national average of 9%.¹⁰ Technicians were the most common occupation in this latter group. Although no one was asked about their own experience with medical errors, 10 participants volunteered that they had past experience with medical errors as a patient or family member. Nearly every participant that volunteered this information also, at some point in the interview, expressed the sentiment that everyone makes mistakes, and responsible physicians should be shown some type of leniency. Two participants strongly agreed and two disagreed with the statement “All in all, doctors are totally honest with their patients” (Table 1). These four participants were deliberately examined for distinct patterns of responses, and no distinct differences from other participants were found. Coding was also analyzed by demographics and no strong patterns were noted. Age may have influenced participants’ reactions to some extent, with more younger participants reporting the time since the error occurred as a major factor that would moderate their actions, and more older participants describing their relationship with their physicians as an important driver of their reactions to error disclosure. Additionally, more educated (college and beyond) participants cited the disclosing doctor’s demeanor and behavior as a factor in their actions, and both younger and more educated participants were more likely to consider investigating the situation surrounding an error once it was disclosed to them.

Table 1:

Patient Questionnaire Responses

Patient Demographic Characteristics	N=30
Age (SD, Range)	46.1 (16.9, 18–76)
Sex (Female)	15 (50%)
Race*
White	16 (53%)
Black	6 (20%)
Hispanic/Latino	3 (10%)
Asian/Asian American	3 (10%)
Pacific Islander	1 (3%)
American Indian/Alaska Native	2 (7%)
Middle Eastern	2 (7%)
Other	5 (17%)
Marital Status
Married	9 (30%)
Divorced	8 (27%)
Never Married	13 (43%)
Employment Status
Working full time	14 (47%)
Working part time	5 (17%)
Temporarily not working	2 (7%)
Disabled	5 (17%)
Retired	4 (13%)
Education
High School Graduate/GED	2 (7%)
Some College/2-Year Degree	4 (13%)
4-Year College Degree	12 (40%)
More than a 4-Year College Degree	12 (40%)
Medical Professional	10 (33%)
Physician Trust: “All in all, doctors are totally honest with their patients.”	N=30
Strongly Agree	2 (7%)
Agree	21 (70%)
Neutral	5 (17%)
Disagree	2 (7%)
Strongly Disagree	0 (0%)

^*Multiple participants shared more than one racial/ethnic identity

Disclosure Preference

Participants were unanimous in considering disclosure in cases of IMED essentially equivalent to disclosure in cases of self-discovered errors, and all participants preferred that errors be disclosed to them in both situations.

“I think bottom line it’s the same problem, it’s a medical error. It doesn’t really matter who told it to me.” (Participant 30)

“It doesn’t matter if a new provider or an old provider discovers this error, as long as the error is discovered and some kind of action is taken to prevent this from happening again.” (Participant 20)

Participants felt strongly that severe, costly, painful, or repeated errors should be disclosed. They preferred disclosure even in instances where the circumstances surrounding the error were somewhat uncertain. The only prevailing exceptions to universal disclosure were cases of small errors with no impact on a patients’ wellbeing. Approximately one third of participants expressed that it is not necessary for physicians to disclose errors with no effect on current health or future care (e.g., ordering duplicate or unnecessary lab tests). Participants would not mind if these errors were disclosed, but did not feel the same strong need for disclosure in these instances.

“If it doesn’t change the way you’re cared for or if you don’t need medical treatment right away, then I don’t see the reason to tell them, because all you’re going to do is make people mad.” (Participant 10)

“I can’t come up with a valid reason, aside from if it didn’t impact my future care or financially, that I wouldn’t want to know about a past medical error from a current or new doctor.” (Participant 23)

Participants broadly felt that concern about causing patient anxiety was not a valid reason to withhold disclosure. However, a minority of participants (n=5) said that withholding information in cases of IMED with particularly anxious patients might occasionally be appropriate (Table 4).

Table 4:

Patient Responses to Physician Barriers to Disclosure

Theme	Quote
Physicians Assess No Benefit	I think it’s not your position to tell the patient things based on how you think they would feel. (Participant 30)
	You’re employing them to help you make decisions, and not make decisions for you. I think it’s a team sport and I think it’s up to the patient and the doctor to decide what’s significant and what’s not significant in their care. (Participant 29)
	For a doctor to believe that they are in a position to determine whether or not it would be beneficial to a patient might not reflect what the patient actually believes. (Participant 17)
	It shouldn’t be up to the physician whether or not I would disagree. What he thinks is not pertinent to my health. Anything that he discovers or any mistake he makes is absolutely pertinent to me, and I would want to know. (Participant 21)
Physicians’ Concern for Increasing Patient Anxiety	As a patient with an anxiety disorder, because that’s actually me, I absolutely would still want to be informed. (Participant 28)
	I already have some anxiety about doctors. I don’t think it would increase by knowing about a mistake. (Participant 25)
	The exceptions would be, if somebody’s really, really in a frail state, maybe just had a nervous breakdown or something like that, maybe then I would say no. But other than … most cases I think yes. (Participant 16)
	With my 25-year doctor, I’m okay with (not disclosing), because she knows me. She knows it might just be extra in my head. (…) Now, a new doctor, you don’t know me that well. I feel like you don’t know me that well to make that call. (Participant 19)

Reasons for Preferring Disclosure

Participants gave multiple reasons for preferring disclosure of IMED, the three most common being: (1) they inherently value knowing everything possible about their own health, (2) they believe that physicians’ ethical code mandates disclosure, and (3) they want to fully participate in future medical care (Table 2). The theme that there is an inherent value to knowing everything about one’s own health was the most common reason for preferring disclosure. This reason was unique in that it was relatively unmediated by other factors or actions in the disclosure process. Participants described an intrinsic desire to be completely informed about their own health. Occasionally this reason was connected to a desire to understand past events. For example, some participants wanted to know about the timing of the error or its inevitability in order to reconcile with events of the past.

Table 2:

Summary of Reasons for Preferring Disclosure

Reason	Excerpts
Knowledge of One’s Own Health	I’d like to know as much information as possible about any error that the doctor commits pertaining to me. (Participant 13)
	And whether or not it impacted me, fortunately it didn’t but the next time it might. It’s all part of decision making and continuing a relationship, and I want the information. (Participant 11)
	The patient goes to the doctor seeking medical treatment, so anything that happens to them, they should know what happened. And it’s their own body, they should know. (Participant 30)
Physicians Should be Honest	I think if the second doctor noticed an error, I think it might be their responsibility to say something about it in that case. (Participant 17)
	I think the doctor should be okay with having difficult conversations with their patients and that they should still tell them. (Participant 28)
	A person has a right to know, and I think they certainly should be able to choose and not have somebody else be - it kind of sounds paternalistic in a way, or maternalistic, on the doctor’s part. (Participant 1)
	As a doctor it’s your obligation to put the patient first. So I think that no matter the professional repercussions, if you think that a doctor messed up pretty seriously and it really does impact this patient’s health, it is your responsibility as a doctor to [disclose] that. (Participant 8)
Prepare for Future Care	It gives me more information about how to make future decisions about my healthcare. (Participant 1)
	You need to be able to evaluate your doctor and know what he or she doesn’t know. So, I think it would be nice to know that this person doesn’t know a whole lot about this particular area. (Participant 15)
	Especially if it’s with an illness or sickness that’s going to be more long term, (…) because some illnesses you’re moving from different doctors, the hospitals involved, a lot of different moving parts. As much context and knowledge about what’s going on as possible I would see as a benefit. (Participant 25)

The second major reason for preferring disclosure is related to participants’ assessment of the ethical values that physicians should uphold. Most participants believed that physicians should value honesty and transparency in the patient-physician relationship, and that they should act in accordance with these values. This was expressed as both a provider-centered preference (physicians should be honest) and a patient-centered preference (patients have a right to know about their health).

The final major reason that participants preferred disclosure was a desire to participate in their future medical care as autonomous decision makers. Awareness of errors would allow participants the choice to be more watchful about their future healthcare and to more accurately assess the competence and experience of their physician. Additionally, there were a number of less prevalent reasons for disclosure. One third of participants wanted disclosure in order to help prevent other patients from experiencing similar errors. Some participants expressed a desire to recover costs incurred in the course of diagnosis and treatment. Others hoped that disclosure would facilitate a physician’s ongoing education or improve their practice generally. A few believed that error disclosure would help vulnerable patients with fewer resources obtain better medical care.

Actions After IMED Disclosure

If IMED were disclosed, patients would consider a range of actions including confronting the responsible physician, changing providers, reporting the error to a regulatory body, or pursuing legal action (Table 5). The severity of the error, its impact on future care needs, and the monetary cost of addressing it all influenced potential actions. Whether an error was egregious, or would be avoided by most physicians in the same situation, was also an important moderating factor in their actions. About half of participants reported that they would likely take no punitive or disciplinary action, and were instead primarily concerned with addressing the repercussions of the error. These participants simply desired information, resolution, and future care that addressed the error’s effect on their health.

Table 5:

Summary of Patient Actions in Cases of IMED and Their Moderating Factors

Action
Consider Legal Action	I think if the physician was actively diminishing their responsibility, or hiding it, (…) I think that there would be times that legal action would be appropriate. (…) I definitely think that if they played the game, we would play the game back. (Participant 29)
	Well, the greater the impact to my health, the more likely I might be to change physicians (…) And it also would impact whether or not I might choose to seek financial compensation, to sue or whatever. (Participant 5)
	If it was a major error, one that resulted in a negative health outcome, I probably wouldn’t trust them and I might take legal action with the first doctor. (Participant 8)
	Personally I would like to see a little tempering of the law as a process. That if the doctor does acknowledge medical errors that any legal ramifications would be able to be knocked down a few, but that if he does not acknowledge medical errors, and then it’s proven that there is a medical error, then he’s more liable if he doesn’t take medical responsibility. (Participant 7)
Change Doctor/Report	Depending on the response of the doctor, the doctor seems like he’s conscientious and looking to figure out what we can do here, I’d probably go along with that. But if it looks like he’s just trying to brush it off and move on to the next patient, then I would not go to that doctor anymore. (Participant 16)
	If it was something more minor, I might be more worried at first when I see them, but I don’t think I would quit seeing him if it was something minor. (Participant 27)
	I would like to have the option of choosing a different provider based on the first doctor’s mistake. Whereas if I didn’t know, I might continue seeing them and he or she might make further mistakes. (Participant 25)
Confront Responsible Physician	Again, it depends on the severity. If it was minimal, I’d just make my next scheduled regular exam, and go, ‘Hey, I just want to acknowledge that that happened, and I’m still here,’ you know? And give them an opportunity to say whatever they’re going to say about it. (Participant 22)
	And the next time I saw them, I mean, it would be very difficult to confront the primary care without some discussion. You can’t just ignore it. (Participant 2)
	I would talk to the doctor about it and see what he or she says and if I’m not comfortable with that, I’d either go higher up the ladder or I would pursue another doctor’s opinion. (Participant 18)

Legal action was most likely in cases where the error had severe consequences or when it was discovered after concealment. Participants were most interested in litigating the behaviors of the responsible physician, and in no circumstance where the discovering physician practiced timely and complete disclosure would they plan to take legal action against her/him. The primary goal of legal action was to recoup costs incurred by the patient due to the error. Moreover, there was empathy for the fallibility of the responsible physician, as well as for the difficulty of the discovering physician’s decision to disclose (Table 3). Some participants even suggested a degree of legal protection for physicians who were proactive about disclosure (Table 5).

Table 3:

Themes Related to Trust, Human Error, and Concealment

Change in Trust	It would just make me feel more confident in my new provider that they, one, notice the mistake that was made and two, acted on it. They were honest with it. (Participant 9)
	If someone was able to tell me there was a medical error, I don’t know; it would just make me feel like I could trust that person, even though it’s not a good thing. (Participant 4)
	It goes back to trust. You definitely want to have trust in your doctor in knowing that they’re going to do the right thing for you, and if they make a big mistake, you lose all your trust. But if it’s something little, people make mistakes all the time. If it’s something small, to me that wouldn’t be a big deal. (Participant 10)
	It would probably lessen my trust in the first doctor and increase my trust in the second doctor. (Participant 30)
Empathy for Human Error	For my own I guess validation that everybody can be fallible, everybody makes mistakes. (…) There are oversights, things do happen. That’s what I would want to know. At least if there was an error made own up to it. (Participant 3)
	Again, if I’m still dealing with my same doctor, emotionally I’d be a little bit distraught, but honestly I would stay with my doctor. People are human. They make mistakes, or whatever, and I’ve been with her 20, 25 years and she’s never made a mistake, but this was obviously something that she didn’t catch. (Participant 19)
	It would make me feel good if doctors were open with me about things. Everybody makes mistakes but to not own up to your mistakes, I see that as a problem. (Participant 26)
	Usually medical errors are committed because someone is not perfect, in knowledge, in understanding, in perception, and that kind of describes all humans, and so we need to keep growing, and sharpening each other up. We brush up against each other and sharpen up on our expertise. (Participant 7)
Discovery After Concealment	A mistake will undermine trust for a short amount of time, but I think a hiding of an error is a whole different set of subjects. (Participant 29)
	I’d rather know the truth than somehow find out later and feel like not only one but two doctors have betrayed me and have hidden the truth from me, and then that would make it much worse. No, I’d rather know, because that’s like the good old boys’ club. (Participant 9)
	If someone doesn’t tell you the truth and you find it out through some way else or someone else, then it makes you question that doctor and why did they withhold that from you in the first place? (Participant 26)
	If I found out on my own or from a different doctor that I had suffered damage from an error in healthcare, it’s going to be worse for you than if you tell me. (Participant 5)

Many participants would consider changing providers (i.e., ending care with the responsible physician) if the discovering physician disclosed IMED. This was most frequently moderated by the severity of the error, whether it was a recurring error, and the interpersonal dynamic between the patient and the responsible physician. This latter factor was influenced by their previous relationship and how forthcoming and concerned the responsible physician seemed when they discussed the error with the patient.

Participants were also likely to investigate the error more thoroughly. This action tended to take one of two forms: participants might confront the responsible physician or potentially report them to a medical review board at the hospital or state level. Through this latter action, some participants hoped to elucidate whether the physician had made other medical errors. Similarly, in confronting the physician participants wanted to better understand the circumstances surrounding the error, from the logistical aspects of healthcare provision to the physician’s reasoning. Less prevalent potential actions included informing family and friends of the responsible physician’s error and leaving a negative online review of that physician. Conversely, a few participants were more reassured by a physician who had made a recent error. They posited that the physician would likely be more careful and vigilant in their future care.

Patient Perspectives on the Disclosure Process and Feedback

Participants were asked whether they would have a preference regarding which physician disclosed the error to them in cases of IMED. There was not a clear consensus amongst the responses. Some patients prioritized timely disclosure, i.e., whichever physician found the error first should be the one to disclose it. Others felt that the physician with whom they had an established relationship would be their preferred discloser, while still others had no preference. A major preference was for both physicians to participate in the disclosure process in some manner, since that could facilitate the most clear and complete exchange of information. A common train of logic for preferring both physicians’ participation was that the first physician would have the opportunity to take ownership of their mistake, while the second physician would be able to act as a neutral arbiter of the medical facts. Patients acknowledged that it would be logistically difficult, and potentially interpersonally uncomfortable, to have both physicians in the same room for this process, so they frequently suggested conversations with their physicians could occur sequentially.

Participants recognized that feedback would be inextricably connected to patient disclosure practices. Participants felt that feedback from the discovering physician to the responsible physician should be encouraged in cases of IMED, with a few expressing the assumption that such feedback is already standard practice. Again, some suggested that feedback might take place prior to disclosure, so that the discovering physician could clarify ambiguities. They suggested that this order of communication could give the responsible physician an opportunity to disclose their own error to the shared patient. Some participants thought that feedback between physicians should contain the same information as disclosure to the patient. Most participants explicitly acknowledged that the discovering physician might encounter legal or professional repercussions if they provide feedback to the responsible physician. However, no participant thought that these possible consequences justified withholding disclosure to patients or withholding feedback to responsible physicians. Participants distinguished cover-ups orchestrated within a network of colluding physicians from the type of professional communication necessary for feedback, abhorring the first and applauding the second. Participants describe significant distress, anger, and mistrust at the prospect of discovering an error after their physician(s) concealed it (Table 3).

Expanding on the theme of trust, participants felt that disclosure of an IMED error would increase their trust in the disclosing physician and decrease their trust in the physician who made the error. They tended to distinguish between interpersonal trust (i.e., “I trust my physician to make correct moral choices”), and clinical trust in their physician’s competence (i.e., “I trust my physician to make correct clinical choices”). They reported that disclosure would increase their trust in the discovering physician’s interpersonal and/or clinical judgment, while it would decrease their trust in the responsible physician’s clinical competence. However, participants again expressed sympathetic understanding that physicians are fallible and medical errors are an expected reality in receiving care.

Discussion

This study has three major findings. First, patients considered IMED essentially equivalent to self-discovered errors, and strongly preferred disclosure in both scenarios. Second, patients prefer disclosure because they expect to be fully informed about their own health, believe that physicians should practice transparency, and want to maintain autonomy over future care decisions. Finally, patients said they would consider confronting the responsible physician, changing providers, or pursuing legal action after IMED disclosure, depending on the severity of the error and the transparency of the disclosure process, with legal action typically reserved for cases of debilitating, irreversible errors.

The Institute for Healthcare Improvement, a leading patient safety organization, supports the notion that IMED is a common type of error, but direct measurement is difficult given no agreed upon definition or system of reporting.¹¹ This study explores patient preferences within this underexplored domain of error disclosure. While disclosure is considered a key tenet in the optimal response to medical errors, in cases of IMED, discovering physicians are uncertain of their obligations and have previously expressed uncertainty regarding patient preferences and values.⁶ Importantly, these results confirm previous findings that patients prefer and expect disclosure of errors, and this preference is now demonstrated to hold true in cases of IMED. Patients expect error disclosure from all of their physicians, and honesty about errors enables patients to be informed and autonomous.¹² These expectations are based on values that are constant across error scenarios. This study suggests that strong patient-physician relationships, founded on trust and transparent disclosure practices, can weather inter-system clinical mistakes, but concealing mistakes can break down the therapeutic relationship.¹³

Comparing the results of this study with previous work, patients do not see physicians’ justifications for withholding disclosure as legitimate. Some specialists have reported that they would withhold disclosure out of concern for increasing patient anxiety and an assumption that there is no benefit to disclosure after the fact. However, this study indicates that patients strongly prefer disclosure regardless of potential anxiety. Furthermore, their intrinsic desire for knowledge about their own health contradicts physician claims that error disclosure has no tangible benefit (Table 4). Physician decision making about IMED disclosure should thoughtfully weigh these patient perspectives.¹⁴ The preferences expressed were consistent and reasonable, and they should catalyze efforts to encourage disclosure and feedback of IMED errors.

Additional obstacles to IMED disclosure include physicians’ fear of jeopardizing referral relationships, as well as their fear of involvement in legal action patients take against responsible physicians. Patients expressed considerable appreciation for the nuance of these interactions and stated that they would not be primarily interested in litigating or punishing the responsible physician. They also encouraged collegial communication between the two physicians as a means for advancing patient safety. Additional research assessing preferences of referring physicians is needed to address these obstacles to disclosure, but the patient perspective is nonetheless an important part of this narrative.¹⁵ Our group has proposed a potential mechanism for inter-system feedback, which would ideally involve legally privileged, physician-to-physician communication first, to clarify clinical facts and plan for disclosure to the shared patient.¹⁶ There are a limited number of similar systems that already exist, such as feedback within quality collaboratives and between trauma centers, as well as Patient Safety Organizations.^17,18 The organizational strategies and collaborative principles behind these systems may be extended to IMED scenarios.

This study has several limitations. First, these qualitative results cannot be generalized to a broader patient population. Although we purposively sampled by age, gender, and race/ethnicity for demographic diversity, results may be biased by self-selection for study participation, high education levels, and lack of geographic diversity among participants. Surveys are a vital next step in investigating specific predictors and modifiers of patient and provider preferences around IMED. Particular variables in case scenarios, such as error severity, emotional distress, and disclosure logistics, are best investigated via these quantitative tools. Also, participants were asked to predict their preferences and behaviors in response to hypothetical vignettes rather than actual experiences. Additionally, the medical error vignette, while deliberately designed to control for as many relevant variables as possible, did describe a medical problem that primarily impacts women; this may have biased the responses in an unpredictable way. Future research with patients who have experienced IMED could test and supplement this study’s conclusions.

Conclusions

Physicians’ concerns that patients may be needlessly distressed following IMED disclosure are superseded by patients’ preferences. Patients desire disclosure in IMED in order to be informed and to make empowered decisions about their healthcare. In these instances, they would consider taking actions proportional to the harm caused by the error and influenced by the transparency of the disclosure process. These findings should be a serious consideration in physicians’ decision-making and systems’ policymaking about IMED. Overall, this study provides a convincing impetus to devise systems-level solutions to enable IMED disclosure.