Patients typically sue because they are dissatisfied with the outcome of their medical care. Sometimes the reasons are obvious, such as a surgeon operating on the wrong leg. Sometimes it is outright negligence, such as a doctor not prescribing antibiotics when they were clearly needed. Sometimes the case is coaxed forth by the stereotypical ambulance-chasing lawyer. Often it is simply a bad medical outcome, without any obvious fault or flaw, but that is enough to get the malpractice ball rolling.
But the underlying commonality is almost always a breakdown in doctor-patient communication. A group of researchers analyzed four thousand pages of depositions filed by patients and noted “problematic relationship issues” in almost three-quarters of the cases.1 When they dissected the depositions in more detail, three communication shortcomings arose repeatedly: doctors did not deliver information well, doctors didn’t take patients’ viewpoints seriously, and doctors didn’t understand their patients’ perspectives. (The fourth common theme in these depositions was that many patients felt abandoned by their doctors during their medical care. I suppose one could consider this a form of communications breakdown—a particularly heinous one—but I’ll leave it aside for now.)
It should come as no surprise that doctors and patients see things quite differently when it comes to malpractice. In one large survey of doctors and patients who’d ever been involved in lawsuits,2 97 percent of patients cited physician negligence as their reason for suing. Of doctors who had been involved in lawsuits, a mere 10 percent felt negligence had occurred. By comparison, doctors who had never been sued estimated that negligence occurred in about 50 percent of malpractice cases.
More than three-quarters of doctors, regardless of whether they’d been sued, felt that financial motivation was a primary reason for malpractice suits. Only a fifth of patients agreed with that statement. Doctors and patients also had very different takes on the doctor-patient relationship. The vast majority of doctors were quite confident that they were offering an honest, open relationship, but fewer than half of patients reported experiencing it that way.
In fact, there was only one thing in the entire survey that the doctors and the patients agreed upon: two-thirds of both groups felt that improving doctor-patient communication would decrease lawsuits.
But is that true? Does better communication between doctors and patients make malpractice suits less likely? Using Debra Roter’s RIAS analysis system, linguistics researchers tried to answer this question by examining in depth ten office visits for each of 124 doctors.3 Roughly half the doctors were primary-care doctors (general internists and family doctors). The other half were orthopedic surgeons. They analyzed how these doctors spoke and interacted with their patients to see if any particular behavior could predict which were the doctors who had been sued (based on insurance company records).
Among the primary-care doctors, length of visit was the strongest predictor. Those who had never been sued spent 20 percent more time with their patients. Other notable differences were that doctors who had never been sued spent more time outlining what the patient might expect—both for the clinical condition and for the logistics of the medical evaluation. They also encouraged the patient to talk and frequently asked the patients’ opinions. They checked in to see if the patients understood what was going on. There was also more laughter and humor in these visits.
It can’t necessarily be said that these behaviors prevented lawsuits. Things could work in the reverse—doctors devastated by lawsuits might be more tight-lipped in conversation, more wary of extending conversation beyond the necessary (certainly studies have consistently shown that doctors experience much less joy in medicine after a lawsuit). But thinking back to the analysis of the depositions filed by patients, the prime frustrations were that they were not being heard and that things were not communicated clearly to them. So these observations about the primary-care doctors who were never sued versus those who had been sued make sense. And from an economic standpoint regarding malpractice prevention, a few more minutes talking is a lot cheaper than the billions of dollars spent annually on defensive medicine.
Interestingly, among the orthopedic surgeons, there weren’t major differences in communication styles between those who’d been sued and those who hadn’t. However, another study by the same research group focused specifically on tone of voice in surgeons.4 Ten second clips of the surgeon speaking, taken from the first and last minute of the visit, were analyzed for tone. The tone could be rated from one to seven for each of the following qualities: warm, concerned, interested, satisfied, genuine, professional, competent, sympathetic, anxious, hostile, and dominant. Then, using these ratings, the researchers tried to see which qualities of voice tone could predict which surgeons had been sued in the past. Unsurprisingly, perhaps, it turned out that surgeons who rated highest on the dominant scale and lowest on the concern and warmth scales were those who were most likely to have been sued.
Again, it’s hard to dissect out the cause and effect. Patients may be less likely to sue doctors who are more open to hearing their views, who help them know what to expect, and who have a less overbearing way of speaking. Or, it might be that doctors with those traits are better clinicians and make fewer errors. A third possibility is that none of these traits affects the risk of lawsuit, and it’s just that doctors who have been sued end up less sympathetic, more dominant, and less likely to engage with patients. It’s impossible to know which way the causality runs.
Nevertheless, these studies suggest some good guidelines for doctors: don’t be too overbearing, get a sense of the patient’s perspective, and do your best to let patients know what to expect, including the possible bad outcomes.
Malpractice suits, of course, are only the tip of the iceberg when it comes to medical care that doesn’t turn out the way all parties hope. The scope of medical malfeasance is a far more sprawling field: errors, complications, misdiagnoses, and untoward outcomes. This broader field is more challenging to study because there aren’t clear registries of data, as there are for malpractice suits. But how doctors and patients communicate in all of these circumstances is critically important.
An interesting story about communication in this realm took place a few years ago at a sleepy VA hospital in Lexington, Kentucky. Steve Kraman is a pulmonologist with a specialty in critical-care medicine, and in 1986 he became the chief of staff at the Lexington VA. The hospital had just lost two bruising malpractice suits that cost it $1.5 million plus incalculable effort and pain. Kraman and the hospital’s chief attorney, Ginny Hamm, were put in charge of a risk management program designed to learn about errors early on, before they snowballed into malpractice suits. The committee hung out their shingle and let it be known that they wanted to be informed of any medical errors that occurred in the hospital.
The silence was deafening—and unsurprising.
“Nobody came forward with errors,” said Kraman, a slim, bespectacled physician with a trim mustache and beard, “so we had to do a bit of snooping.” And it didn’t take long to uncover a smoking gun.
An alcoholic woman, admitted with dehydration and electrolyte depletion, had died on a recent night in the hospital. The death was attributed to severe malnutrition brought about by decades of alcohol use and had aroused no suspicions at the time. During the treatment course the patient had been given fluids, along with intravenous potassium to correct a dangerously low potassium level. The resident physician had ordered blood tests that afternoon to recheck the patient’s potassium level, with a plan to decide whether the potassium drip needed to be continued or stopped. The labs were drawn but the resident became swamped with work and forgot to check them. The potassium continued to drip methodically into the patient’s veins.
For entirely different reasons—the patient had complained of palpitations—a Holter monitor happened to be hooked up to the patient. Holter monitors are little boxes that hang around a patient’s neck, recording the EKG for a full day or more. These are usually given to outpatients, so there’s no visible monitor on the outside; it’s just a box that contains the data. When the patient was discovered dead, the Holter monitor was shoved aside as weightier things were attended to.
But the Holter monitor wended its way back to the cardiology lab, and in due time the contents of the device were downloaded. The pages and pages of printout eventually made their way to the top of a staff cardiologist’s pile.
The electrical output of the heart—the EKG—is a delicate, squiggling affair, with waves, peaks, and valleys corresponding to the electrical signaling during the cardiac cycle. The QRS complex is a spike that corresponds to the contraction of the ventricles. The T-wave, a semicircular hump toward the end of the cycle, indicates relaxation. Potassium is critically involved in the electrical conduction of heart cells, so EKG morphology can be a mirror of potassium levels in the blood.
What the cardiologist saw in the Holter reading was a textbook progression of rising potassium levels. First, the smooth T-waves grew peaked. Then the spikes of the QRS complex widened. Then all the spikes and waves began to soften, jellylike, and drift apart from one another, degenerating into an amorphous sine-wave slush. And finally it all collapsed into what doctors charitably call asystole, the featureless horizon of flatline that stretches out to a desolate eternity.
This is not what cardiologists typically view on routine outpatient Holter reports, and one can imagine the gripping fear in the heart of this one as the inexorable march toward death was documented in excruciatingly accurate detail. This patient clearly had died of a hyperkalemic (high-potassium) cardiac arrest, though no one had known it at the time.
When Kraman—a critical-care specialist—saw the Holter, he knew before he’d even exhaled a single breath that the VA had a wrongful death on its hands. You couldn’t have evidence any more incriminating than this minute-to-minute, time-stamped Holter readout. But what to do with this information? The patient’s family had taken her body; the funeral had already occurred. There had been no concerns at the time about the death; both the family and the doctors had accepted the readily available explanation that the patient had died of her severe alcoholism. The desire to let sleeping dogs lie could not be denied.
But there was no question in Kraman’s mind about what to do. The hospital called the family, told them that some new information had come to light about their loved one’s death, and that the hospital staff would like to talk with them. Oh, and they should bring a lawyer.
“We never want to reveal what happened during the initial phone call,” Kraman says, “but we do want to indicate its seriousness so that the family will definitely show up. Telling them to bring a lawyer usually accomplishes that.”
The risk management staff met with the family and explained that they had strong evidence that a medical error had been made and that the hospital was directly responsible for the death. “When we told them,” Kraman recalled, “the family was certainly shocked, but it was their attorney whose jaw dropped. He could not believe we were voluntarily disclosing this information.” The hospital took responsibility for the death, offered a settlement, and within a few weeks there was an agreement.
The fact that a lawsuit had been averted—one the hospital surely would have lost—became the impetus for a new hospital policy mandating full disclosure. The goal was that all errors would be fully and promptly disclosed to the patient or family—most certainly not the standard way most doctors communicated errors to patients.
Every single untoward incident had to be reported. A peer review committee would do an evaluation to decide if there actually was an error. For example, if a patient died of pneumonia despite being given three different antibiotics but the antibiotics had all been appropriate choices, this would not be considered an error. But if a true error were uncovered, the disclosure gears would grind forward. All medical errors—including invisible ones and near misses—would be proactively disclosed to the patients and their families. Sleeping dogs would not be allowed to lie.
Telling patients the truth about errors, even about near misses, is, of course, the right thing to do. Every doctor would say that this is the ethically correct course of action. But I know from brutal experience that this is far from easy.
It was during a long night on call—we were probably up to our eighth or ninth admission that day, but my intern and I had long since given up counting. I was midway through my medical residency at Bellevue and was already a seasoned survivalist; you had to be, otherwise you’d drown in the overwhelming pummel of admissions that streamed in from the ER, day and night.
This admission was a classic eye-roller: a nursing home patient with dementia, sent to our hospital for altered mental status. When you were juggling patients with bleeding ulcers, acute heart failure, fulminant septicemia, raging diabetic infections, and multidrug overdoses, it was hard to get worked up about a demented nonagenarian who someone thought was looking perhaps a wee bit more demented that day. The patient was totally stable, I was told—labs fine, radiology fine.
The trick to surviving on call was to “turf” as quickly and as aggressively as possible. Anyone you could possibly turf to another service—surgery, rehab, psych—was one less on your ever-expanding list. This patient was a perfect candidate for the so-called intermediate-care unit, which was a back corner of the hospital that functioned as a holding station for patients with no active medical issues, patients just awaiting discharge arrangements.
But the covering doc left at 5 p.m., so we’d have to work quickly—assemble the labs, head CT, chest X-ray. Once we’d ruled out any treatable medical explanation for the altered mental status, we could safely turf the patient. It was 4:45 p.m., and I scanned through the labs as I dialed the covering doctor. I quickly ran down the case to her: demented patient, totally stable, labs fine, radiology fine, just needs to get back to his nursing home.
I remember the doctor’s voice so clearly over the phone. She had an Indian lilt to her voice, and she asked me, “You’re sure the labs and scans are normal?” Yes, yes, yes, I pressed her. Totally stable. Everything is fine.
The doctor hesitated; taking a patient so near to closing time wasn’t ideal. I reassured her that the patient was completely stable, that there was no work for her to do, and eventually she relented. The intern and I high-fived each other and then bolted to the ER for the rest of the night, sifting through the admissions that had piled up. There were patients with cellulitis, emphysema, unstable angina, and pancreatitis. Illness never seemed to end, especially during those long nights on call.
It wasn’t until early the next afternoon that I learned that my “totally stable” patient had spent the night in the operating room. It turned out he actually had an intracranial hemorrhage—he was bleeding into his brain. That’s why his mental status was altered. The covering doctor didn’t mince words when she told me how she’d been paged during the night by the radiologist who saw the bleed on the head CT. The patient was whisked out by the neurosurgeons for emergency surgery to drain the blood inside his skull.
My body felt like it was turning to stone as all my metabolic processes ground to a halt. An intracranial bleed? A hemorrhage into the brain? You couldn’t get much worse than missing an intracranial bleed. But I had. I had failed to check the head CT before jettisoning the patient. Someone had said “radiology fine” and I hadn’t looked at the actual scan myself, as I knew I should have.
The covering doctor stormed away before I could even open my mouth, which was just as well because I was incapable of speech. If I could have willed myself to melt away on the spot, to melt myself out of existence—I would have.
In cerebral academic analysis, this incident would be classified as a near miss because the patient’s medical care ultimately proceeded correctly. The multiple layers of care in the hospital ensured that the intracranial bleed was correctly diagnosed and treated. Appropriate medical care had not been impacted, or even delayed really, by my oversight.
But I had still made the error. What if I had prescribed a medication such as aspirin that could have furthered the bleeding? What if it had been earlier in the day and I’d sent the patient back to the nursing home in an ambulette that lacked resuscitation equipment? What if the radiologist hadn’t read the scan until the next day? The “what-ifs” were endless, and endlessly tormenting. My error could easily have led to a fatal outcome. It was nothing but sheer luck that saved the patient from my bungling error.
I was horrified at myself for not performing due diligence for my patient, mortified that I had relied on a verbal report rather than taking the time to examine the CT scan myself. The shame was blistering and all encompassing. No matter how I rationalized to myself that the patient was unharmed and that the system of checks and balances in the hospital had functioned appropriately, I was still faced with the unvarnished truth: I had committed a life-threatening medical error.
I was so ashamed that I didn’t tell anyone about my lapse—not that day and not ever after. I didn’t tell my intern. I didn’t tell my attending. And I certainly couldn’t fathom dragging my sorry self into the patient’s room to come clean. I would sooner knock back a mug of the muck-strewn waters of the East River just outside the hospital than face the patient with my inexcusable oversight. All I wanted to do was barricade myself into a broom closet and mope among the cleaning supplies.
It is precisely for this emotional reason that Kraman’s team would not have the “offending” physician present at the family meeting. “The first ten minutes can be a bit hot,” he said, and this struck me as an affable but staggering understatement. “There can be accusations, anger.” If the doctor got defensive it could defeat the whole purpose of the meeting.
Carol Liebman disagrees with this. Liebman is a clinical professor of law at Columbia University who has been interested in medical mediation as a way to improve patient safety and possibly—as a byproduct—minimize lawsuits. Her group decided to offer free mediation services to hospitals in exchange for being able to study the process and interview the participants.5 The goal was to figure out how communication and mediation could both shorten the process and ease the arduousness that typically attends medical errors.
The parties to a medical error—patients, families, doctors, administrators, lawyers—typically disagree on the key elements of a case, but the one thing they do agree on is that the current malpractice process is inefficient, costly, and just plain dreadful. Like a Band-Aid being haltingly peeled off, the misery seems designed to be meted out in excruciating slow motion. No one ends up pleased, even those who “win” the case. When you talk to people who have prevailed in a malpractice case, they will rarely speak of joy or even satisfaction at winning. It’s more like relief that a marathon of emotional exhaustion has finally ceased. Could mediation offer something better than that protracted agony?
Liebman described a case to me that centered on the placement of a central line. When central lines are placed in either the subclavian or jugular vein, there is always the risk of puncturing the top margins of the lungs, which are parked amid these prominent vessels. A pneumothorax, as it’s called, is a known complication of central lines, though thankfully not common. And if it does happen, the pneumothorax is usually minuscule. A tiny sliver of lung deflates, and then generally reinflates on its own with no notable consequences. But if the pneumothorax is massive—which is extremely rare from a central line, more likely from a motor vehicle accident—enough oxygen supply can be cut off to lead to cardiovascular collapse.
In this case, a patient was getting a subclavian line from a resident physician. Stuart Benson suffered from emphysema, which eats away at the taut orderly air sacs of the lungs, leaving the organs filled with the equivalent of unkempt, oversized, saggy balloons. Emphysematous lungs lack the firm recoil of healthy lungs and possess little physiologic reserve.
Tragically, the tip of the resident’s needle punctured Mr. Benson’s fragile lung, which promptly deflated. The massive pneumothorax acutely sapped his body of oxygen and his heart went into cardiac arrest. Mr. Benson did not survive.
The resident called Mrs. Benson at home and told her to come to the hospital immediately, though he didn’t give her the reason. When she arrived, the attending physician met her outside her husband’s room. The attending informed Mrs. Benson of her husband’s untimely death. After that, though, Mrs. Benson was left in the hallway while the medical team tended to all the things that medical teams attend to after a patient has died. The staff eventually dispersed, the communications petered out, and Mrs. Benson was left stranded in the hallway, literally and figuratively.
That feeling of being stranded in the hallway is regrettably common and is often the prime motivation behind lawsuits. I’ve never met Mrs. Benson, but I can only imagine the enormity of what she was facing. Losing a spouse is already one of life’s most harrowing moments. Grief is emotion enough to consume a person, but to then have anger elbow in on that grief can be corrosive. There is anger at being abandoned by the doctors, anger at not knowing how or why her husband died, anger at having her private grief be contaminated by this unwanted bitterness. I imagine that filing a lawsuit was probably the last thing she’d want to do while mourning the death of her life partner. That she did is a testament to how potent the anger in such situations can be and how long it can fester.
Because of the drawn-out pace of such proceedings, it was nearly a year after Mr. Benson’s death when all the parties finally met in one room. “When the hospital team arrived at the mediation,” Liebman told me, “the widow’s face fell when she saw that the doctor who’d placed the central line wasn’t there.” From Mrs. Benson’s perspective, the hospital’s taking of responsibility was hollow if the person who’d actually caused the error wasn’t part of the human equation.
The mediation process allowed Mrs. Benson to ask questions about the central line and why a resident physician was doing the procedure. It allowed her to express her anger and frustration at how she had been treated. But the lack of communication with the doctor directly responsible was undermining the credibility needed for mediation to work. Trust was fast eroding and it seemed a lawsuit might be inevitable.
At the second mediation session, however, the chief of medicine had something to add. He mentioned that he had seen the resident that morning and had asked him if he remembered the patient—it had been almost a year since the incident, after all. The physician’s response was both incredulous and pained. “Remember him? I think of him every single day. I grieve for him every day.” When this quote was relayed to the widow, the atmosphere changed immediately. Even though the doctor was not present, his emotions—particularly his remorse, his pain, and the enduring ramifications—restored an important degree of humanity to the process, and the widow was more open to what the team had to say. The sides agreed to a fair settlement and a protracted, painful, adversarial fight was averted. There was a financial payment but also a commitment from the hospital to improve its care—something that is a critical ingredient in responding to the pain experienced by families. A safety checklist for central lines was instituted. The hospital also ramped up its training for the staff on how to communicate with family members in the event of a death, so that no one would be left standing in the hallway.
Liebman feels it is critically important for the person who made the error to personally acknowledge the error, take responsibility, and, most critically, apologize. Even when there isn’t a clear-cut cause—as there was in the case of the central line—the spoken words of acknowledgment and apology are critical. Nothing can move forward without them.
I agree with her but I also know how arduous this can be. After I had missed that intracranial bleed I was so overcome with shame that I didn’t have the inner resources to tell even my supervising attending, let alone tell the patient. My error took place before the widespread acceptance of full disclosure, but I knew coming clean was the ethically correct thing to do. I had no doubts about that. Nevertheless, I simply couldn’t. It’s true that the atmosphere of medical training at the time wasn’t conducive to such admissions, but I couldn’t blame it wholly on that. My resistance was internal.
Looking back now, I see the missed opportunities so clearly and so poignantly. Had I felt comfortable enough to tell my attending on that morning, she could have helped me understand how the error occurred and how to prevent it in the future. Even more powerfully, she could have taken our team to the bedside and modeled how a doctor might speak to a patient about an error. So much of what we learn in medical training is promptly forgotten, but this is the type of teaching moment that resonates forever.
Apart from the error itself and the ramifications for the patient and family, there is also the devastating effect on the person who made the error. As a barely-out-of-the-gates doctor, my missing the intracranial bleed was cataclysmic. Until that moment I’d thought I was a reasonably good doctor—no worse or no better than my colleagues—but in a split second that entire persona came shattering down. It wasn’t just that I had made an error; it was that my being a doctor was an error. I felt like such a crashing disappointment to my profession that the only prudent course would be to take down my shingle, such as it was, and barricade myself at a desk job alphabetizing insurance forms, selflessly sparing future innocents my shabby medical ministering.
Sacrificing my medical career—as dramatic and tempting as it felt—was simply not an option, though. Calling in sick for even a single day was practically verboten, because everyone else had to shoulder your work. If you had the audacity and selfishness to quit outright, your fellow residents would probably flay you before your grubby white coat even landed in the laundry bin. The only acceptable reasons to abandon ship were permanent coma, cardiac arrest, or possibly a full-body cast, but even that was dubious.
Unable to commit my noble hari-kari, I simply plowed on. I showed up every day at 7 a.m., rounded on my patients, wrote in the charts, attended noon conferences, but I was numb, paralyzed by an all-consuming shame. Had I been able to tell my attending about my error, maybe she could have eased the dreadfulness of the weeks that followed. I say this with empathy not only for my younger self, but also for the patients I cared for in those weeks. With my soul in a fog and my brain in permanent low gear, who knows how many little things I missed—the slightly depressed bicarb level, the subtle signs of a wound infection, the incrementally prolonged QT interval on an EKG. I’m certain there was a trail of errors in the wake of my missing that intracranial bleed—many of which could have been avoided if I’d been able to speak openly about what had happened. The shame burrowed so deep and with such tenacity that it took me a full twenty years to finally be able to speak of that moment.
Steve Kraman, who is now a professor of medicine at the University of Kentucky, is convinced that the full-disclosure experience can help dissolve the poisonous culture that inhibits people from doing the right thing. “Mostly these are honest mistakes by good people,” he told me, and I found myself grateful for the forgiveness and optimism of those words, even these many years after my error. “Usually they were overworked, or their attention wavered for just a moment.” From the way he described it, I could see how the disclosure-and-restitution process could offer some balm for the person who’d made the error. And, it turned out, lawsuits were indeed fewer.6
The decrease in lawsuits and overall financial payout at the VA garnered a significant amount of media attention. Other hospitals found similar results after implementing comparable programs.7 In fact, an entire organization—Sorry Works!—was formed to promulgate this full-disclosure-and-restitution policy. A self-described “middle ground” for doctors, patients, insurers, lawyers, and hospitals, Sorry Works! is an advocacy group for this new way of thinking. Its website offers testimonials from people for whom an apology averted a lawsuit, and others for whom an apology never materialized and led to bitter litigation.
In reading the plethora of news articles on their website, I was struck by the cheerfully righteous tone as they exclaimed over the counterintuitive conclusion that hospitals have been saving money by full disclosure of error. This is almost divine justice: virtue is rewarded, good guys finish first, doing the right thing brings accolades and cost-savings . . . just like your mother always promised.
What a wonderful conclusion: full disclosure and an apology make a patient less likely to sue. I am wholly supportive of this important work. Lowering the number of lawsuits is salutary for every party involved: doctors, patients, administrators, insurers (well, maybe not for the malpractice lawyers. . .). However, it doesn’t bring the risk to zero. From the hospital’s perspective, decreasing lawsuits from four per year to one per year is a vast improvement. But for the one doctor who is still sued, that 75 percent decrease is meaningless. For that one doctor, it is a 100 percent life-altering experience, almost always devastating.
The hospital and the individual physician have very different agendas when it comes to disclosing errors and apologizing, and they also have fundamentally different things at stake. A hospital is rarely destroyed by a single malpractice suit, but an individual physician surely can be. The emotional and financial toll can permanently scorch a doctor’s life—even if the doctor is acquitted.
The shame of error, with its attendant loss of sense of self, is an individual, human trait. Organizations rarely suffer from shame. After the money is forked over and the publicity dies down, hospitals continue on as they always have. Not so for the individual physician or the nurse; the emotional ravage tunnels deep into the soul. A veteran ICU nurse in Seattle committed suicide after her error led to the death of an eight-month-old baby.8
In the end, though, even if the data convincingly prove that the sophisticated communication skills of disclosure and apology lead to fewer lawsuits, I believe the desired culture of openness will come about only when we address the issue of shame. No matter how rational we doctors claim to be, the fragility of the human heart will override hard data, ethics, and even laws.
When thinking about how to improve communication between doctors and patients regarding medical error, it’s clear we have to attend to the emotions on both sides. But doctors and patients have different stakes and expectations in that inevitable conversation. Patients overwhelmingly say they want as much information as possible. They want to know about any and all errors that may have occurred, even the minor ones. When researchers interviewed patients about this, there was absolute unanimity that all information should be disclosed.9
Physicians interviewed in the same study agreed with the patients, at least in theory. In practice, however, they were more circumspect. Doctors worried a lot about medical error and agreed that errors needed to be disclosed, but they felt strongly that they had to “choose their words carefully.” They agreed they needed to be honest and accurate but they also felt the need to portray the events in the least negative way possible. The fear of a lawsuit is so pervasive that doctors worry any admission of error is tantamount to handing your head to a lawyer on a stainless steel surgical tray.
Researchers in this particular study conducted in-depth focus groups with doctors and patients in the same room. For half the session, the doctors sat in an inner circle discussing the issues while the patients observed from an outer circle. The situation was reversed for the second half of the session. This arrangement led to some of the most interesting observations of the study.
Patients were quite surprised to hear how deeply doctors were upset by errors they’d made. Most had assumed that like illness and death, errors and bad outcomes were all in a day’s work for doctors. The patients were astonished by the depth and duration of anguish that doctors experienced. One patient remarked, “I was really surprised to hear the doctors talk like that. I saw a lot more caring than I expected. . . . You know, most of the time when you see the doctor you don’t get their feelings. Yeah, I was surprised.”
When the circles were reversed, doctors were surprised to learn that assigning blame was by no means the top priority for patients. As with Mrs. Benson, the primary goal was to obtain information. A close second was to know that the doctors and the hospital accepted responsibility and learned from the mistake. Blame was not the focus. This would support the experiences of the Lexington VA and the Sorry Works! group: acknowledgment, meaningful apology, and concrete actions to improve things go a long way toward repairing the harm. In fact, these probably accomplish more than a lawsuit, which is as bruising for the patient as for the doctor.
When it comes to acknowledging and apologizing, what seems to matter is the patient’s perception of what the doctor says, even more than the actual words.10 A group of researchers in Baltimore showed videos of three medical-error enactments to a group of volunteers. The simulated errors included a delay in noticing a mammogram report that indicated a possible cancer, a chemotherapy error in which the dose was ten-fold too high, and a patient who suffered a cardiac arrest after a surgeon was slow to answer a page.
For each medical error scenario, there were several versions of the video so that the researchers could vary whether or not the doctor accepted responsibility and whether or not the doctor apologized. Furthermore, there were two versions of the apology—one in which the doctor gave a generic apology (“I’m sorry that things turned out badly”) and one that included a more directly personal apology (“I’m sorry that I gave the wrong dosage of chemotherapy”).
As expected, the viewers gave higher ratings to the doctors who accepted responsibility and made a personal apology. But the fascinating finding was that what seemed to matter most was whether the viewer perceived these things as having occurred. If he or she perceived that the doctor made a personal apology (even if the video was scripted to be a nonspecific apology), then the viewer rated the doctor as more trustworthy.
Making an apology, though, is never easy. Acknowledging that you’ve made a mistake or fallen short can be painful, especially for doctors. I recently had this experience with a healthy, middle-aged, white woman whom I’d seen just once before, a year earlier, for a routine checkup. I ushered Kim Fuller into my office with a smile, prepared for another straightforward visit. “What can I help you with today?” I asked, as I always do to start a visit.
She stared stonily at the wall in front of her as my question hung, too long, in the air. “I almost didn’t come back today,” she finally said. “I was ready to choose another doctor but I decided to give you a second chance.”
There was a reflexive start in my chest from her unexpectedly harsh reply but I held my composure. I quickly scanned the chart from our last visit to see if there had been anything we’d disagreed about or any unresolved issues. But no, it had been a routine checkup. Nothing beyond standard health screening issues and a run-of-the-mill tension headache.
“Someone had said you were a good doctor,” Ms. Fuller said derisively, still addressing the wall, “but I was not impressed. My previous doctor, even though he was just a resident in training, was much better than you.”
That certainly stung. And I still had no idea what had occurred between us to inspire her vitriol, so I endeavored to stay quiet, to simply listen. “But he graduated,” she continued, “and now I’m stuck with you.” She folded her arms across her chest, pursed her lips, and then was silent.
I sat there trying to figure out the best way to approach this situation. Should I apologize even though I didn’t know what I’d done? Should I admit ignorance and ask directly what the transgression had been? Should I pretend to remember and try to craft a passable apology? I glanced again at the chart to find some clue, any clue, as to what I might have done to upset this patient so much. But I couldn’t find a hint of anything. I decided honesty was better than faking it. I kept my voice as low key as possible and said, “I’m so sorry if there was something that upset you. I’m trying to recall—”
“You don’t remember?” she snapped. “It figures!”
Now I was scraping desperately around in my mind. What had I done? Making a patient livid isn’t something you typically forget but I couldn’t come up with anything for Ms. Fuller. I realized I was going to have to offer blanket contrition, both for whatever had transpired between us and for my inability to remember it.
“I apologize,” I said, as genuinely as I could, “but I honestly don’t recall what happened. Could you tell me?”
Ms. Fuller looked at me for the first time, her eyes taut and icy. The pause before she spoke uncoiled tortuously, serrated with resentment. “You didn’t do a physical exam. All you did was talk to me! That’s all. My old doctor always did a physical exam but you didn’t even bother.”
My jaw actually dropped when she said that. I’d forgotten to do the physical exam? That was a biggie! How could I have done that?
As an internist, I’m well aware of the limits of the modern physical exam. For most medical ailments, nearly all the crucial diagnostic information is gathered in the history, in the conversation between doctor and patient. The physical exam serves mainly to confirm the diagnosis but it nevertheless remains an important part of the interaction between doctor and patient.
I felt like a complete idiot. Looking back at my notes, I honestly couldn’t tell if I had been running late that day, or was distracted, or simply had a momentary lapse. Part of me wanted to defend the academic truth that the physical exam wasn’t really necessary in a healthy patient with no symptoms other than a mild tension headache. I wanted to point out that the risk of a false positive result—finding something you think is abnormal but really isn’t—outweighed any benefits by a long shot, and that by this reasoning you could say a physical exam in a healthy patient like her was more likely to cause harm than offer benefit. Maybe I’d actually given her better medical care by not doing the physical exam. But I knew that wouldn’t go over well. And the unvarnished truth was that I’d neglected a standard part of the medical visit and the patient was appropriately calling me on it. There was really only one option for me.
“I am truly sorry,” I said to her. “I honestly have no idea why I didn’t do a physical exam that day. I really don’t.” I placed my hands down on the desk between us. “I can only offer you an apology, both for my shortcoming that day and for the bad feeling that it left you with.”
Ms. Fuller gave a brief nod of acknowledgment and the muscles of her face softened by a few degrees. Our gazes simultaneously traveled to the exam table and then back to each other, questioning. “It’s up to you,” I said, hesitantly. “I would completely understand if you would feel more comfortable with a new doctor.”
She shrugged. “Like I said, I’m willing to give you a second chance.”
It was an awkward physical exam, no doubt for both of us. I felt like I was being given a test but appreciated the opportunity to face up to a mistake and work through it, however uncomfortable the experience.
Though “objective” measures of “quality” abound in medicine these days, getting meaningful feedback is actually quite rare. Getting it directly from a patient, rather than on a spreadsheet from an institution—or in a subpoena from a court—is rarer still. And even though it’s never pleasant to be reminded of your shortcomings, I ultimately felt quite lucky that this patient had the grit to come back and tell me directly. She could just as easily have moved on to another doctor, and I never would have known.
It makes me wonder how many other times I have disappointed a patient but been completely unaware. I’m sure the number is larger than I’d care to acknowledge, and probably rising as time pressures and documentation requirements mount.
I thanked Ms. Fuller for giving me a second chance and we parted reasonably cordially. But I’ll have to wait a year until her next annual visit to see if I passed the test.