Psychology insights on apologizing to patients

A clinician will make many mistakes during a career practicing medicine. Some mistakes will result in patient harm, wasted resources, or hurt feelings. The adverse consequences of a mistake will also extend back to the clinician and ultimately increase the risk of burnout if not mitigated.1 One protective strategy is to reduce the frequency of mistakes to zero, but such utopian ideals are not realistic.2 Another strategy is to learn how to apologize effectively after a mistake; however, most advice on how to apologize stems from popular press, legal recommendations, ethics analyzes, religious theology, social norms, or community standards and not from scientific evidence.3 Psychology is the science that explores how people perceive, think about, and act toward other people. This science recognizes that interpersonal offenses can occur between a transgressor (also called wrongdoer, harmdoer, offender) and victim (also called sufferer, casualty, target, offended). Examples can include a hurtful comment, betrayal of trust, or interpersonal unfairness. The damaging effects on relationships can be mitigated by an apology, defined as a statement expressing remorse, acknowledging responsibility, and potentially offering restitution. Despite the power of an apology in conflict management, professionals often do not apologize or do not apologize well.4 Traditional training and consensus guidelines encourage clinicians to tackle difficult conversations, including the disclosure of upsetting information.5 These fundamentals rarely mention how to apologize effectively.6, 7 The purpose of this article is to review three unfamiliar findings from psychological science on how to apologize effectively (Table 1). An effective apology, we propose, is an essential clinical skill that does not necessarily materialize from years of clinical practice but might be informed by scientific insights.8 Apologies are hard, yet an awareness of these insights might lead to increased clinician motivation, greater patient satisfaction, and ultimately more healing.9-11 An apology acts as a promise that restores trust after an adverse outcome by fostering expectations of improved future care. Many apologies in medicine, for example, include a commitment to do better for the individual involved and for other patients. When higher expectations go unmet, however, reputation can suffer more than if no apology had been offered.12 This also implies the effectiveness of an apology tends to decrease with repeated use and can backfire if overused (more effort leads to worse outcomes). This counterproductive aspect of an insincere apology, therefore, could easily arise by apologizing repeatedly to the same patient after separate bad experiences. A recent study conducted by the UberX ridesharing platform demonstrated the potential backfire effect from repeated apologies, as assessed by customer spending patterns.13 The study randomly assigned customers who experienced three bad trips to receive an apology and a $5 discount after only the first bad trip or repeatedly after each of the bad trips. Surprisingly, the third apology had a negative effect on short-term spending (1 week) compared to the first apology (7% decrease vs. 2% increase, p < .001). The negative effect also extended to long-term spending (1 year) and the total number of future trips (despite accumulating discounts). This means client satisfaction requires using apologies selectively for unexpected mishaps rather than for repeated adverse events. Undue repetition is not the only way an apology can appear insincere and worse than none at all. For example, a long delay from transgression to apology can make the effort feel contrived or fake.14 In addition, a proactive apology before a transgression can seem manipulative or phoney.15 Poor language is also counterproductive such as passive verbs (“Mistakes were made”), vague clichés (“Sorry you feel hurt”), conditional disclaimers (“I apologize but …”), incomplete points (“Sorry it did not work out.”), vacuous excuses (“It wasn't my fault”), illegitimate reasons (“I was hungover”), or inadequate precision (“Sorry for the situation”).11 An effective apology in a medical setting, therefore, needs to be original, timely, accurate, articulate, and smart. Psychological science indicates an apologizer needs to talk about themselves (remorse, responsibility, reparations) and also fully voice the victim's perspective (emotional feelings, subjective experiences, mental state).16, 17 Articulating an empathetic understanding helps provide validation, enhance trust, bolster a victim's self-esteem, and foster forgiveness. Claiming to understand another person's feelings, however, can fail if the victim's perspective is not characterized correctly.18 Misunderstanding the predicament, therefore, can make the apology seem a hollow ploy from a self-serving ulterior motive that lacks moral authenticity.19 The net result is that an effortful apology might backfire and result in a counter-productive loss of trust. Experiments exploring this nuance use scenarios in which an apologizer is randomly assigned to incorrectly or correctly voice a victim's emotions. In one study involving a financial mistake, for example, the apologizer voiced an accurate emotion aligned to the victim (“I see how angry you feel”) or an inaccurate emotion mismatched to the victim (“I see how dishonoured you feel”).20 As hypothesized, average ratings of trust (7 point scale) were greater after an accurate compared to an inaccurate apology (4.74 vs. 2.22, p < .001). Indeed, an inaccurate apology was worse than an apology that offered no consideration of emotion. More generally, empathy requires an emotional fit; otherwise, the mismatch smacks of hypocrisy and can decrease trust. An effective apology requires an accurate understanding of emotions by the transgressor, yet the success of an apology is determined by the subjective reaction of the victim. This means clinicians must have emotional intelligence. A compassionate tone helps signal shared values by combining negative self-criticism (“I'm sorry that I am late”) and positive appreciation (“Thank you for waiting for me”).21, 22 Non-verbal cues also help create emotional resonance, such as body language aligned to cultural norms.23-25 Ultimately, an apology needs to convey an expression of personalized care and differs from the clinician expressing a statement of regret (inconveniences that are not offenses) or an honest mistake (errors from fallible medical systems).26 Many apologies in everyday life are not heartfelt and the apologizer is not truly sorry. The average recipient, however, is not discriminating about genuine sincerity since such careful judgment requires extra effort, can conflict with self-esteem, may escalate conflict, and risks appearing uncharitable. This tendency toward naive acceptance is similar to common reactions to flattery, where a recipient tends to believe the ingratiator rather than carefully check for authenticity.27 An onlooking companion, however, may be more discerning and not so easily satisfied. As a consequence, a patient's initial apparent satisfaction following an apology might change substantially after talking with friends, family, or another third-party afterwards. A scientific method for studying apologies involves staging a harm, expressing an apology, and gauging how victims or onlookers react. One study used an insult (“quit being a baby”), followed by a coercive point from an assistant posing as a participant (“that was uncalled for, you must apologize”), and ending with a perfunctory apology (“I'm sorry”).28 As later rated by the victim (10-point scale), the coerced apology was more effective than no apology (4.37 vs. 2.94, p = .01) and nearly as good as a spontaneous apology (4.37 vs. 4.34, p > .20). For the average onlooker, however, the coerced apology was no better than no apology (3.62 vs. 3.61, p > .20) and less effective than a spontaneous apology (3.62 vs. 5.26, p < .001). Apparently, an insincere apology can seem different to those addressed directly compared to those observing indirectly. The concept of third-party reinterpretation is especially relevant to medicine since clinicians have limited time with a patient whereas the patient may recount their story many times thereafter.29 This can be especially poignant with lawyers who may argue an apology justifies more compensation despite being initially well-received.30 Further distortions also arise because an apology is rarely documented in sufficient detail in the chart to accurately capture the words, emotions, added team members, and other external factors prevailing at the time. Post-hoc reinterpretation is also important for clinicians themselves, since a transgressor who has their apology rejected tends to hold more anger toward the victim and later tends to become unlikely to reach a satisfying resolution.31 The purpose of presenting evidence from psychological science is not to dissuade clinicians from apologizing by highlighting basic pitfalls; instead, the purpose is exactly the opposite.32 Moreover, a general failure to apologize often occurs because a transgressor does not recognize the offense or morally disengages from the wrongful action.33 The patterns in clinicians include denying responsibility, characterizing the incident as justifiable, minimizing the loss, or blaming events as outside control.34 Of course, developing skills toward making more effective apologies is difficult since talented clinicians may have few opportunities to apologize and some regions provide no legal safeguards for clinicians who provide an ineffective apology (Table 2). More effective apologies have several theoretical advantages relevant in medicine. An effective apology is congruent with modern safety systems when errors are a blend of latent systems factors combined with shared individual lapses. An effective apology agrees with institutional risk management that prioritizes hospital safety, operational compliance, and harm prevention. An effective apology might also reduce medico-legal damages even where such statements are admissible as evidence.35 An effective apology can also be an educational topic for motivated trainees that is taught, learned, improved, and role-modeled. Conversely, the prevailing infrequency of apologies in medicine is a potential common and correctable failure of empathy in patient care. Donald A. Redelmeier: Conceptualization; data curation; formal analysis; formal analysis; funding acquisition; methodology; project administration; resources; supervision; validation; visualization; original draft; revisions. Jada Roach: Data curation; formal analysis; methodology; resources; validation; visualization; and revisions. We thank Vidhi Bhatt, Tom Gilovich, Nick Daneman, Alan Detsky, and Jonathan Wang for helpful suggestions on specific points. The views expressed are those of the authors and do not necessarily reflect the Ontario Ministry of Health. This project was supported by the Canada Research Chair in Medical Decision Sciences, Canadian Institutes of Health Research, PSI Foundation in Ontario, and Sunnybrook Program to Access Research Knowledge for Black and Indigenous Medical Students. Donald Redelmeier serves as the guarantor of the work and accepts full responsibility for the work, the conduct of the work, subsequent access to the data, and the decision to ultimately publish. The authors declare no conflicts of interest.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.