We need to talk: Provider conversations with peers and patients about a medical error

Abstract

Purpose

Although open communication with patients is the established best practice after a medical error, healthcare providers’ conversations with each other in these circumstances are less studied. We identified and compared what providers identified as the most important thing to say to their peer and to the patient after a medical error.

Methods and materials: This study surveyed providers about the most important thing they would say to their peers and patient regarding a hypothetical scenario depicting a delayed diagnosis of cancer. Participants included primary care physicians, oncologists, and oncology nurses. Direct content analysis was used to identify major themes and the McNemar test was used to evaluate significant differences in the providers’ references to major themes (p > 0.05).

Results

A total of 303 providers produced valid responses. Four major themes emerged: (1) information sharing; (2) emotion handling; (3) preventing recurrences; and (4) responsibility. While the majority of provider responses included information sharing, fewer than one-third described the event as an error. Significantly, fewer providers addressed emotion with their peer than with the patient (10% vs. 54%, p ≤ 0.001). Providers were more likely to bring up prevention of recurrences with their peer than with the patient (43% vs. 19%, p ≤ 0.001). Approximately one-quarter of providers addressed responsibility with the peer and patient (25% vs. 26%, p = 0.707), although fewer than 10% acknowledged personal responsibility for the error in either context.

Conclusion

Providers approach conversations about medical errors with a peer differently than with patients and may benefit from additional communication training or support.

Keywords

Provider conversations peer conversations medical error communication delayed diagnosis

Introduction

When a medical error occurs, two distinct sets of challenging conversations become necessary: among the involved providers and with the patient. When multiple providers are involved, conversations between physicians clarify what happened and how best to approach the patient.¹ Substantial research has addressed what providers should say to patients about a medical error, but little is known about conversations about errors among providers.^1–5 While both types of conversations can reveal details about the event and have strong emotional undertones, they can be expected to be different given the distinct types of relationships (professional vs. therapeutic) involved. After a medical error, patients appear to respond most favorably when providers approach them with a statement of regret, an explanation of what occurred, acceptance of responsibility, and a commitment to preventing recurrences.^2,4,6,7 While some of these elements may also exist in provider conversations, these conversations are uniquely influenced by fears of damaging interprofessional relationships, experiencing retaliation, or affecting a peer’s professional reputation.^6,8,9

At present, Communication and Resolution Programs (CRPs) are being widely implemented to help support an appropriate response to adverse events.¹⁰ With CRPs, provider conversations about medical error will become more common, and understanding what providers discuss amongst themselves and with their patients after a medical error can inform the development of training programs to improve open communication. The present study explored and compared thematic similarities and differences in what providers considered was the most important thing they would say to their peers and the patient after a medical error.

Methods and materials

We created a hypothetical but realistic vignette about a young woman diagnosed with locally advanced breast cancer who sought care from multiple providers before receiving a diagnosis six months after presenting with a complaint of a breast lump. We chose this case as it involves multiple providers with different roles and responsibilities, and thus highly relevant to other error situations. We asked open-ended questions to primary care physicians (PCPs), oncologists, and oncology nurses to understand how they would respond to peers and the patient about a delayed diagnosis of cancer. The study protocol was approved by the University of Washington Institutional Review Board (Application #36619), Fallon Community Health Plan Institutional Review Board (Project #1184), and the Kaiser Permanente Georgia Institutional Review Board (Protocol #GA-08JCavi-01). Completion of the survey served as implied consent. This study (P20CA137219) was funded by the National Cancer Institute, which had no role in its design, conduct, or reporting.

Study design, setting, and participants

Healthcare providers (PCPs, oncologists, and oncology nurses) at three sites that participate in the Cancer Communication Research Center of the Health Care Systems Research Network were surveyed between June and September 2011. The Cancer Communication Research Center is a consortium of research organizations affiliated with healthcare delivery systems that works to improve cancer care through population-based research. The three participating sites were located at Washington, Massachusetts, and Georgia. The survey was pre-tested with providers outside the study population. The research team administered the final version using a modified tailored design method.¹¹ Eligible participants received a questionnaire along with a letter from a leader in their organization, a return envelope, and an incentive valued at approximately $20 that varied depending upon site policy. Up to three reminders were sent to non-responders. Wording varied slightly to reflect the respondents’ clinical roles. The questionnaires contained clinical vignettes followed by multiple-choice and open-ended questions exploring the providers’ reactions and opinions. Multiple choice questions evaluated factors influencing the provider’s intent to disclose a medical error (example: What would you most likely say regarding an apology or to express regret? (a) I would not volunteer that I was sorry or apologize. (b) “I am sorry about what happened to you.” (c) “I am sorry that you were not diagnosed sooner.” (d) “I am sorry that I did not examine your breast sooner.”). Analysis of these multiple choice questions have been published previously.¹²

In this study, we analyzed responses to two open-ended questions that followed a vignette describing a delay in the diagnosis of breast cancer. The provider is asked to imagine covering for a peer whose patient has had several visits over two months for hypertension management. During the visit, the patient mentions that a breast lump, previously evaluated by the peer, has grown. The provider’s chart review reveals their peer’s documentation of a cyst-like lump in the left breast six months ago with recommendation of a repeat exam. The patient never made the follow-up appointment after the initial complaint of a breast lump and subsequent notes from the primary provider do not mention the breast lump. The covering provider orders a biopsy which confirms the presence of locally advanced breast cancer. Two questions were posed: “What are the most important things you would say to your colleague in this situation?” and “What is the most important thing that you would tell this patient about what happened?” For nurse participants, the wording of the survey questions differed slightly: “What are the most important things you would say to the physician in this situation?” and “What is the most important thing that this patient should be told about what happened?”

Data analysis

Responses included verbatim quotes from providers and descriptions of their communication strategies. Both types of responses were coded using an identical approach to direct content analysis. We developed an initial coding framework using inductive thematic analysis. Three authors (TD, JZ, KM) reviewed and revised the coding tool based on thematic similarity. The final coding scheme was created through an iterative process consisting of review, coding, and discussion. Two coders (TD, JZ) independently coded 10% of the responses and discussed discrepancies until the kappa statistic for inter-rater reliability was at least 0.75. The two coders independently coded a share of the responses, and 20% were coded by both to ensure continued agreement. Each unique concept in a response was assigned a code, yielding an average of two codes per response (range 1–7).

Statistical analysis

We calculated the percentage of providers referring to each major theme and subtheme in their responses. We used the McNemar test to evaluate whether providers’ references to major themes differed when talking to a peer vs. the patient.¹³ Data analysis and summarization of the codes were performed using Statistical Package for the Social Sciences (Version 24) and Microsoft Excel (Version 15.25.1).

Results

Out of 759 surveys, 404 were returned for a response rate of 53%; 303 providers produced valid responses to both open-ended questions. Invalid responses included no response and uncodable responses, such as “??” or “not sure”. Provider characteristics are summarized in Table 1.

Table 1.

Provider demographics.

Demographics	Providers responding to both questions
Total sample (n)	303 (100%)
Type of provider
Primary care provider	256 (84.5%)
Oncology nurse	28 (9.2%)
Oncologist	19 (6.3%)
Gender
Female	161 (53.1%)
Male	142 (46.9%)
Years in practice
<10 years	77 (25.4%)
11–20 years	71 (23.4%)
>21 years	150 (49.5%)
Unknown	5 (1.7%)

Four major themes emerged about the most important thing providers would say to their peer and the patient after the medical error: (1) information sharing; (2) emotion handling; (3) preventing recurrences; and (4) responsibility. Providers approached the topics of information sharing, emotion handling, and preventing recurrences with their peer and the patient in significantly different ways (p > 0.05).

Table 2 provides illustrative quotes for each major theme and subtheme present in at least 10% of responses to the peer or the patient.

Table 2.

Most important themes raised in responding to peers and patients about medical error.

Subthemes	Responding to the peer	Responding to the patient
Information sharing
Describe the event as an error	“I think we failed this woman but hope that we can learn from the error.” (PCP)	“An error was made and we will try to prevent it from happening again to another patient.” (PCP)
Provide an objective description of facts	“I wanted to inform you of what transpired since the initial visit regarding a lump.” (PCP)	“Changes in breast exam, especially those noted by a patient, may be a sign of cancer.” (ONC)
Identify cause of error	“Failure to list the breast lump as an active problem led to a delay in diagnosis.” (ONC)	“This problem should have been noted in the Problem List so all involved in your care would know to follow up on this.” (PCP)
Discuss error mitigation	“Contact risk management [and] don’t alter the record.” (ONC)	“We will be diligent about getting her the best care and develop communication to ensure she understands any recommendation we have.” (PCP)
Emotion handling
Expressions of apology or regret	“I want you to be aware of what happened. I’m sorry for my role in not noticing the needed follow up. I think it would be useful for you to contact the patient and apologize also.” (PCP)	“I’m sorry we did not diagnose this sooner but I hope that we can get you in with the surgeon and oncologist quickly so that we can start treatment right away.” (PCP)
Empathy and reassurance	“We all make mistakes. Let’s learn from this one.” (PCP)	“We’re going to take good care of you. I'm confident all will be well.” (ONC)
Preventing recurrences
Individual behavior change	“Be more proactive in reviewing charts and ordering images for breast issues.” (RN)	“Schedule follow up appointments as directed. Never assume no news is good news. Advocate for yourself always.” (RN)
Health system change	“Let us look at what we can set up in our system to help capture this in the future.” (RN)	“We will look at this situation and figure out a system to ensure a delay like this doesn't happen in the future.” (RN)
Responsibility
Peer is responsible	“You should have referred this patient for a mammogram at her first visit. You should have flagged on her problem list that she had a breast mass so that any other providers who may see her can become aware. [You] should have explained to patient the potential severeness of this finding and scheduled a definite date for her to return for re-examination.” (PCP)	“the covering MD did not review the chart thoroughly.” (RN)
Patient bears responsibility	“[Your] patient did not schedule follow up as recommended.” (PCP)	“Don’t assume that no news is good news. Please stay on top of an issue as important as this one and don't let your doctor forget to follow up on issues as important as this one.” (PCP)

ONC: oncologist; PCP: primary care provider; RN: nurse.

Information sharing

The majority of providers said it was most important to share information about the error when addressing both their peer (62%) and the patient (72%) (p = .004). Four key subthemes emerged: (1) describing the event as an error; (2) providing an objective description of the facts; (3) identifying the cause of the error; and (4) discussing error mitigation (Table 2).

Fewer providers stated that an error had occurred when addressing their peer (10%) than with the patient (30%). Providers who described the event as an error most commonly used the words “delay,” “mistake,” “problem,” and “error”.

Twice as many providers indicated that it was most important to provide an objective description of the events leading to the error when addressing their peer (20%) than with the patient (10%). Some providers expressed a desire to provide information in a “non-judgmental way” to their peers.

Approximately one-quarter of all providers indicated that it was most important to identify the cause of the error when addressing their peer (23%) and the patient (26%). Providers most commonly cited communication breakdowns, process errors, and clinical errors as the cause of the error.

Fewer than one-quarter of providers indicated that it was most important to discuss error mitigation with either their peer (22%) or the patient (15%). With peers, these providers discussed mechanisms for reporting the error, recommended that the peer contact the patient directly, and/or suggested a formal case review. When addressing the patient, they focused on the cancer treatment plan and less commonly discussed reporting the error or conducting a case review.

Emotion handling

Significantly, fewer providers stated that it was most important to address the emotional dimensions of the error when talking to their peer (10%) than with the patient (54%) (p ≤ 0.001). Two key subthemes were identified: (1) expressions of apology or regret and (2) empathy and reassurance (Table 2).

Providers expressed an apology or regret less frequently to their peer (2%) than the patient (36%). Only six providers used the word “apology” or “sorry” with their peer, while nearly one-third of providers said it was most important to offer an apology or expression of regret to the patient.

Fewer than one-quarter of providers expressed empathy or reassurance in their statements to either their peer (9%) or the patient (22%). With peers, they acknowledged the emotional burden of the medical error, shared similar experiences, minimized the error, and legitimized their peer’s actions. Such expressions to patients included promising support, encouraging trust of the healthcare team, and focusing on the positive aspect of cancer detection.

Preventing recurrences

More providers responded that it was most important to address the topic of preventing recurrences of the error with their peer (43%) than with the patient (19%) (p ≤ 0.001). Two subthemes emerged: (1) the need for individual (patient or provider) behavior change and (2) the need for health system changes (Table 2).

Providers generally addressed error prevention with their peer by focusing on the peer’s need for individual behavior change (27%). They made recommendations regarding their clinical practice, communication behaviors, and practices to ensure patient follow-up.

Providers discussing error prevention with patients generally emphasized the need for health system improvements (14%). Fewer providers discussed individual behavior change with patients (5%), but when they did, they encouraged assumption of responsibility for their care and improved communication with their healthcare team.

Responsibility for the error

Approximately, one-quarter of providers said it was most important to address responsibility for error when talking to their peer (25%) and the patient (26%) (p = 0.707). Two key subthemes emerged: (1) the peer bears full or partial responsibility for the error and (2) the patient bears full or partial responsibility for the error (Table 2).

Providers acknowledged personal responsibility for the error in fewer than 10% of statements to either the patient or the peer. Providers’ statements to their peer most commonly highlighted the peer’s missteps (12%), such as failure to order imaging at the initial visit, failure to ensure follow-up, and poor documentation. One frustrated PCP wrote, “Geez! Get a mammogram at least. Use a tickler to bring patient back. They don't come back on their own.”

Providers’ statements to the patient focused on the patient’s role in causing the delay (11%), including failure to express ongoing concerns to each provider, and stressed the patient’s responsibility for their own health. One oncologist said, “Communication is key. You must communicate all health concerns and never assume your current provider knows something that you should divulge. Time to work together and focus on what the next step is here.”

Discussion

This study used open-ended questions to elicit the words and themes that providers prioritize when communicating with their peers and patients about a medical error. Our study demonstrates that providers approach discussions about medical errors with their peers and their patients in significantly different ways, especially regarding information sharing, emotion handling, and prevention of recurrences. Given that both types of conversations are challenging and yet necessary to ensure individual and organizational learning, these findings suggest the need for a deeper understanding of how relationships and conversation contexts influence provider discussions about medical errors.

Our study found the greatest disparity in how providers addressed the emotional dimension of error disclosure conversations. Only 10% of providers expressed an apology, empathy, or reassurance with their peer, compared with more than 50% of providers who responded that emotion handling was the most important thing they would address with the patient when discussing a medical error. Providers may find apologizing to a peer irrelevant or challenging, particularly in scenarios when culpability is shared. Similarly, while empathy is well accepted within a doctor–patient relationship, expressions of emotions between medical providers may be perceived by as unprofessional and thus avoided.¹⁴ Yet clinicians involved in medical errors often experience significant emotional distress^15–17 and many prefer the support of their peers over support from institutional or mental health professionals in such circumstances.¹⁸ Further research is needed to understand how providers can best address the emotional complexities of peer conversations.

We found that fewer than half the number of providers who said it was most important to address prevention of recurrences with their peer addressed error prevention measures in their responses to the patient. This is consistent with prior observations that providers do not address prevention of recurrences with patients despite clear evidence that patients want to know not only an error’s cause and consequences but also error prevention measures.^3,7 One reason for this may be that by highlighting a peer’s missteps, providers fear they will undermine the patient’s relationship with their primary physician or trigger litigation.¹

While the majority of providers emphasized the importance of sharing information about the event with both their peer and the patient, a minority acknowledged that an error had occurred. Prior studies have demonstrated that healthcare providers struggle with open disclosure and that discussions about medical errors with healthcare providers often fall short of patient and family expectations.¹⁹ Several factors affect providers’ willingness to disclose medical errors to patients, including the degree of responsibility they feel for the error, pressure from malpractice insurers, and varying perceptions of what constitutes a medial error.^3,7 Our study found that providers are less likely to use the term “error” or synonyms of “error” with their peers, instead using objective descriptions of the event. This is consistent with existing literature that suggests that peer feedback may be constrained by a reluctance to assign blame or potentially offend a peer.⁸

Our data show that in complex medical situations involving multiple providers, discussions of responsibility for an error, particularly personal responsibility, are not prioritized. Reasons for this may include ambiguity regarding fault or a sense of powerlessness to address the error. Given that healthcare is increasingly delivered in teams, institutions must encourage providers to adopt a sense of collective accountability to respond to errors as unified teams.²⁰ For this to occur, providers must feel confident in approaching peers to discuss errors. Institutions have largely relied on formal mechanisms of peer feedback after a medical error, including mortality and morbidity conferences, autopsy conferences, and chart audits. Yet evidence suggests that providers favor direct provider-to-provider feedback in informal settings and that such peer feedback may more effectively stimulate change.²¹ To help providers navigate these difficult conversations, future studies should focus on identifying the specific challenges providers face when discussing medical errors and work to establish best practices to guide providers during peer conversations about medical errors.

There are several important limitations to this study. The distribution of providers was skewed heavily towards PCPs, precluding definitive observations about interprofessional conversations. Our analysis was also limited by the intrinsic weakness of inductive thematic analysis. With this approach, nuances in the data, including infrequent but important themes, may have been missed. As this study utilized a survey with a hypothetical scenario to elicit provider responses which may not accurately reflect actual provider practice, future studies should consider engaging providers in focus groups and direct interviews to overcome these limitations. Finally, the wordings of the two survey questions differed slightly (respondents were asked about the most important “things” they would say to their peer vs. the most important “thing” they would say to the patient) which may have contributed to a bias favoring more peer responses.

Conclusion

This is the first study to examine the key themes identified by providers as most important in responding to a peer about a medical error. The significant differences in how providers approach these conversations with peers compared to patients highlight the need for peer support programs to facilitate these difficult conversations among providers, enabling them to be better prepared to address patient concerns about medical errors.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study (P20CA137219) was funded by the National Cancer Institute, which had no role in its design, conduct, or reporting.

References

Gallagher

Mello

Levinson

, et al. Talking with patients about other clinicians’ errors. N Engl J Med 2013; 369: 1752–1757.

Truog

Browning

Johnson

, et al. Talking with patients and families about medical error: a guide for education and practice. Baltimore: Johns Hopkins University Press.

Gallagher

Garbutt

Waterman

, et al. Choosing your words carefully: how physicians would disclose harmful medical errors to patients. Arch Intern Med 2006; 166: 1585–1593.

Mazor

Greene

Roblin

, et al. More than words: patients’ views on apology and disclosure when things go wrong in cancer care. Patient Educ Couns 2013; 90: 341–346.

Apology in medical practice. Health care safety. JAMA. The JAMA network, http://jamanetwork.com/journals/jama/fullarticle/203361 (accessed 3 July 2017).

Surbone

Onclogists’ difficulties in facing and disclosing medical errors: suggestions for the clinic. Am Soc Clin Oncol Educ Book 2012; e24–e27.

Gallagher

Waterman

Ebers

, et al. Patients’ and physicians’ attitudes regarding the disclosure of medical errors. JAMA 2003; 289: 1001–1007.

Lipitz-Snyderman

Kale

Robbins

, et al. Peers without fears? Barriers to effective communication among primary care physicians and oncologists about diagnostic delays in cancer. BMJ Qual Saf 2017; 11: 892–898.

Kaldjian

Forman-Hoffman

Jones

, et al. Do faculty and resident physicians discuss their medical errors? J Med Ethics 2008; 34: 717–722.

10.

McDonald

Van Niel

Gocke

, et al. Implementing communication and resolution programs: lessons learned from the first 200 hospitals. J Patient Saf Risk Manag 2018; 23: 73–78.

11.

Dillman

Smyth

Melani

Internet, phone, mail, and mixed-mode surveys : the tailored design method. 4th ed. Hoboken, NJ: Wiley, 2014.

12.

Mazor

Roblin

Greene

, et al. Primary care physicians’ willingness to disclose oncology errors involving multiple providers to patients. BMJ Qual Saf 2016; 25: 787–795.

13.

Mackridge

Rowe

(eds). A practical approach to using statistics in health research: from planning to reporting. Newark: John Wiley & Sons, Inc., 2018.

14.

Kerasidou

Horn

Making space for empathy: supporting doctors in the emotional labour of clinical care. BMC Med Ethics 2016; 17: 8.

15.

Wears

AW.

Dealing with failure: the aftermath of errors and adverse events. Ann Emerg Med 2002; 39: 344–346.

16.

AW.

Medical error: the second victim: the doctor who makes the mistake needs help too. BMJ 2000; 320: 726–727.

17.

Busch

Moretti

Purgato

, et al. Psychological and psychosomatic symptoms of second victims of adverse events: a systematic review and meta-analysis. J Patient Saf. Epub ahead of print 26 2019. DOI: 10.1097/PTS.0000000000000589.

18.

Y-Y

Fix

Hevelone

, et al. Physicians’ needs in coping with emotional stressors: the case for peer support. Arch Surg 2012; 147: 212–217.

19.

Iedema

Allen

Britton

, et al. Patients’ and family members’ views on how clinicians enact and how they should enact incident disclosure: the “100 patient stories” qualitative study. BMJ 2011; 343: d4423.

20.

Bell

Delbanco

Anderson-Shaw

, et al. Accountability for medical error: moving beyond blame to advocacy. Chest 2011; 140: 519–526.

21.

Ivers

Jamtvedt

Flottorp

, et al. Audit and feedback: effects on professional practice and healthcare outcomes. Cochrane Database Syst Rev 2012; CD000259