Intended for healthcare professionals

Education And Debate

Five times: coincidence or something more serious?What should a junior doctor have done?You cannot expect people to be heroesPut out the fire or risk an infernoPresent system of whistleblowing is unsatisfactory

BMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7146.1736 (Published 06 June 1998) Cite this as: BMJ 1998;316:1736

Five times: coincidence or something more serious?

See Editor's choice

The anonymous article below was sent to us by a doctor outlining the concerns he had about the competence of a surgeon he once worked with when he was a junior doctor. We asked four other doctors what the junior should have done, what they would have done had they been approached by the junior, and what the implications are for the regulation of medicine.

Perioperative mortality (death within 28 days of an operation) has became a key surgical phrase in the past decade, particularly after the publication of the first report of the confidential inquiry into perioperative deaths. This document detailed a variety of surgical and anaesthetic disasters, and, although it pointed out that many perioperative deaths were and remain unavoidable, there were contributory factors such as inadequate hospital facilities, poor supervision of junior doctors, and inappropriate surgery in severely ill patients.

This and subsequent reports, together with regular intradepartmental and interdepartmental audits, have raised the awareness of perioperative mortality. All operative deaths should now be discussed to discover if care could have been improved or death avoided. I have been fortunate to be a surgical trainee in these more enlightened times. Usually, the audits I have attended have had an average of one death every six months from routine general surgery lists (somewhat more from emergency surgery), and even fewer during my five years in specialist training. With one exception: during a six month period on one firm, five patients on routine lists died from a variety of reasons. All of these patients were led to believe that their conditions would be substantially improved if not cured by the surgery, and yet within a matter of days they were dead. I felt at the time that certain questions were overlooked, if not ignored. My polite queries to the consultant staff were brushed aside, and the surgeon allowed to continue (with more unquestioned deaths) until his eventual retirement.

Memories of the patients, and their families, have stayed with me, and I now wish the problems to be exposed to wider scrutiny. Am I being paranoid or too sensitive? Or am I raising legitimate problems associated with a certain brand of surgeon that was supposed to have been swept away with the advent of the modern NHS—surgeons who believe they cannot be questioned and that their techniques and beliefs are always right?

Each of the cases raised different questions, although all but one of the patients had cancer. One patient died of unexpected medical complications after routine surgery (could the preoperative work up have been improved?); another died of metastatic cancer which the operation could never have cured (it should not have been performed); and the other three deaths were totally unexpected. In one of these cases a necropsy was not requested, so we learnt nothing and realistically should not have issued a death certificate; in another case no cancer was found in the removed organ (should the operation have been performed?); and the last patient died from a presumed iatrogenic complication.

The surgeon may have been unlucky, though I feel that the deaths must be seen in the wider context. They all occurred within 16 weeks of each other, and within my six month rotation nine other major operations were performed, of which five involved major and potentially avoidable complications. Both the junior medical and nursing staff were concerned about obtaining the consent of patients for major surgery as there seemed no guarantee that they would do well. And the problems continued after my spell on the unit.

Criticism of the surgeon at the time was difficult. He was rarely on the unit and planned and assessed his major cases personally, rarely involving either his consultant colleagues or the junior staff. Disagreeing over patient management was not an option as I needed a report at the end of my stint that would be filed in my training record. I ensured that the cases were aired at the monthly audit meetings, but the surgeon concerned rarely attended these and the meetings were treated with little interest by the other consultants (formal meetings have subsequently been dropped, contrary to royal college guidelines). All I have done is keep a diary of the events, file the worry on my six month assessment form and discuss the problems with colleagues and friends.

What should a junior doctor have done?

  1. Miles Irving, professor of surgery
  1. Department of Surgery, Clinical Sciences Building, Hope Hospital, Salford M6 8HD
  2. Institute for Healthcare Improvement, 135 Francis Street, Boston, MA 02215, USA
  3. Faculty of Medicine and Health Sciences, University of Nottingham, University Hospital, Queen's Medical Centre, Nottingham NG7 2UH
  4. St George's Hospital, London SW17 0QT

    What we do not know from this account is whether the author chanced upon a surgeon with a long record of poor performance or whether he arrived on the firm when performance had just started deteriorating. I suspect the latter, for the gossip network usually forewarns trainees of firms with poorly performing surgeons or other difficulties. In either circumstance, however, action was urgently needed, not only for the sake of the patients but because such deterioration in performance can be the first indication of a surgeon's physical or psychiatric illness.

    So what should a junior doctor do in such circumstances? I can answer this question from personal experience, although admittedly as an intermediary, rather than the observer of the events in question. When I was a senior registrar, I was consulted by a senior house officer on the surgical rotation about a consultant whose operative techniques and results were a cause for concern to the trainee, in much the same way as described above. There had also been some behavioural change in the consultant, so I had no hesitation in alerting my own consultant, who in turn consulted the professional panel often referred to as the “three wise men.” The result was a dignified removal from practice of the person concerned on the grounds of ill health, treatment for the underlying illness, and his return to practice in a non-operating role.

    In a nutshell, therefore, what the author should have done was to use an existing mechanism described in Annex D of the Department of Health guidelines HC(90)91 as “pre-disciplinary procedure,” namely the special professional panel whose importance has been reinforced by a government circular, HSG(94)49.2 Of course, the author, like many doctors, may not have known about the existence of this panel.

    I believe the real issue is how we as a profession act to stop the type of situation witnessed by the author developing into a crisis. Lunchtime gossip and oblique comments at audit meetings rarely succeed, and it becomes a disaster for all concerned if the situation progresses to the point where the results are so bad that medicolegal claims or hospital statistics force management to suspend the doctor in question. Referral to senior respected clinicians in a hospital to deal with the situation has to be the most logical and humane way for early intervention in those whose performance is failing, not least because such failure is often a sign of illness rather than innate incompetence. In those few in whom it is arrogance and incompetence, the persuasive powers of those chosen by the consultant body to act as their professional panel should, in most cases, be sufficient to commence remedial action. The professional panel can suspend doctors and refer them to the General Medical Council, but these powers should rarely be required.

    All of us in consultant posts must acknowledge that age and illness can impair our activity and our judgment, and do so in an insidious fashion. My own approach has been to make my junior consultant colleagues promise that they will tell me if they see my performance deteriorating, even if it is not reflected in audit figures, and I have anticipated these problems by refusing to undertake any major surgical procedures without one of my consultant colleagues operating with me. In time, more sophisticated approaches to the problem, such as those described by de Leval,3 may be a surer way of determining the onset of poor performance, but, until such mechanisms are available, common sense has to prevail.

    Anxieties provoked by recognition of deteriorating performance should be countered by knowing that experienced consultants can offer much in terms of training junior surgeons and teaching undergraduates. Whispering campaigns about competence can thrive only in circumstances where it is not possible to openly discuss cases in an audit attended by all the medical practitioners involved and senior nursing colleagues. In the end, the cathartic effect of standing up in front of your peers explaining problems and asking for advice is the best way of preventing gossips.

    References

    1. 1.
    2. 2.
    3. 3.

    You cannot expect people to be heroes

    1. Donald M Berwick, president
    1. Department of Surgery, Clinical Sciences Building, Hope Hospital, Salford M6 8HD
    2. Institute for Healthcare Improvement, 135 Francis Street, Boston, MA 02215, USA
    3. Faculty of Medicine and Health Sciences, University of Nottingham, University Hospital, Queen's Medical Centre, Nottingham NG7 2UH
    4. St George's Hospital, London SW17 0QT

      The author, like all who give care, should have done what is best for the patients. By this standard, the proper course of action is clear. Having reason to believe that harm was being done and likely to be repeated, his duty was to report it to those capable of investigation, understanding, and remedy. But that answer, however morally satisfying, is far from adequate. Merely to assert the author's duty as our response to his dilemma ignores that acting on that duty would have required heroism on his part. This is unreasonable. We should applaud heroes, and hope that they are among us, but to base our hope of remedy in ordinary systems on the existence of extraordinary courage is insufficient.

      Rather than asking what the author should have done, and toasting our high minded answer, let us ask a more enlightening question: given the circumstances, what would be the behaviour of a person of ordinary moral character—not a hero, but the rest of us? If we find the prediction unsatisfactory not in the best interests of our patients—then our problem is not lack of heroism, it is deficiency in leadership.

      So far as I know, all modern, effective systems to assure and improve safety involve a culture in which the reporting of error or apparent error is a valued and positive act, which leads, not to blame, but to curiosity and study. The alternative to discussing possible error is to surrender to error. We must seek a wise middle ground. Ignoring harm is not acceptable. The context of the author's work—a context for which, not he, but his senior leaders must be accountable—did not support the first, essential step in improving safety, disclosure of risk.

      Modern aircraft travel is extremely safe, in part because carefully managed safety systems have invested in “cockpit resource management” (CRM). This focuses on communication among pilots, copilots, navigators, flight attendants, ground controllers, and all who may contribute to the “resource” of knowledge in the interests of safe flight. Three quarters of aircraft crashes are attributable to communication flaws; someone knew something that could have prevented the disaster, but the information never reached the right people. Often, the barrier to communication is the “authority gradient”; a lower status person (say, the copilot) is unable to inform a higher status person (say, the pilot) that the wings are loaded with ice. Either the copilot does not speak or the pilot does not listen. Sometimes, the communication is flawed because it is inappropriately “mitigated” (“Uh…. Do you think we should check the wings?” instead of “There is too much ice on the wings”). Airlines work hard to establish cockpit cultures in which unmitigated communication against the authority gradient is expected and requested by those in the highest status level. Therefore, aircraft fly more safely.

      Had he been a new copilot, the author would have been trained to speak up. His superiors would have both praised him for doing so and taken his information into a sophisticated and respectful system of investigation and remedy. Until healthcare leaders—people like the chair of the clinical service on which the author was training—do the same, future trainees will feel his pain again.

      One final word on error itself. The story suggests that the surgeon may have passed beyond his time of competence. That may be so, but most avoidable errors in health care are not due to the incompetence of individuals. Most errors in health care have the same sources as most errors in other complex systems—poor designs. To reduce errors requires redesign of work processes. Once again, the burden falls to leaders to mobilise the will, resources, and knowledge to change healthcare processes to make them safer. Only when leaders take this task seriously will ordinary people be protected against the scarring pain of being trapped within sight of hazards that they are helpless, without unreasonable self sacrifice, to prevent.

      Put out the fire or risk an inferno

      1. Peter Rubin, dean
      1. Department of Surgery, Clinical Sciences Building, Hope Hospital, Salford M6 8HD
      2. Institute for Healthcare Improvement, 135 Francis Street, Boston, MA 02215, USA
      3. Faculty of Medicine and Health Sciences, University of Nottingham, University Hospital, Queen's Medical Centre, Nottingham NG7 2UH
      4. St George's Hospital, London SW17 0QT

        Two things struck me about this account. One was that the author thought there was something wrong but no one seemed to care. The other was that, on the evidence given, it is hard to say if there was a problem because the deaths were from different and apparently unrelated causes. If I had been given this information (I know I'm a physician, but in my experience it is not uncommon for junior staff to look outside their specialty when seeking help with this kind of issue) I would ask the junior doctor if there seemed to be an explanation—such as ill health or a drink problem. I would then rapidly share the information with the relevant clinical director but would protect the identity of my source.

        A review of the case notes should be quickly undertaken to see if—without that great diagnostic tool of hindsight—the five patients seemed to be a high operative risk. Also, looking at the surgeon's record over a longer period would be important to provide perspective, but this should not be a trawl for individual cases that proved a particular point: we all have cases that go wrong for whatever reason. Several scenarios are then possible: the surgeon may have had an unusually low perioperative death rate in recent years, the cases did not look special preoperatively, and these deaths may have been just chance; the surgeon may have had an average death rate, the operations should not have been performed, and this could reflect a subtle change in his health; or a review may reveal a dreadful record (as implied here).

        If there seemed not to be a problem, I would feed this back in person to the junior doctor. But if there is cause for concern the matter should be raised sympathetically with the consultant, who, pending further inquiry, should suspend operating if there is a possibility of other patients being harmed.

        Dealing with doctors (or medical students) who may be falling below the standards that the public have a right to expect is never easy—I have done it often enough to know. Insight is often lacking, and excuses abound. But there is a simple bottom line: put out the fire now or be prepared to tackle a raging inferno if you don't.

        Present system of whistleblowing is unsatisfactory

        1. Tom Treasure, professor of cardiothoracic surgery
        1. Department of Surgery, Clinical Sciences Building, Hope Hospital, Salford M6 8HD
        2. Institute for Healthcare Improvement, 135 Francis Street, Boston, MA 02215, USA
        3. Faculty of Medicine and Health Sciences, University of Nottingham, University Hospital, Queen's Medical Centre, Nottingham NG7 2UH
        4. St George's Hospital, London SW17 0QT

          The author plaintively expresses the dilemma he faced when he suspected that his chief's operations were causing needless deaths. His attempts to bring this into the open through the usual channels of audit failed, and subtle approaches to other doctors were brushed aside, but it was probably fear for his career that curtailed his criticism.

          Contemporary advice to him seems clear. The General Medical Council's position is: “You must protect patients when you believe that a doctor's or other colleague's health, conduct or performance is a threat to them.”1 The Senate of Surgery of Great Britain and Ireland (which comprises the four royal colleges of surgeons and 12 specialist surgical associations and faculties) advises, under the headline “Potentially Harmful Surgeons,” that we should “Take appropriate remedial steps to bring performance to an acceptable standard where audit reveals that the existing standard of a surgeon's care is consistently unacceptable.”2 The BMA takes a similar view and advises that “whistle blowing may well include revealing the incompetence or bad practice of medical colleagues” and “to fail to ‘blow the whistle’ on poorly performing colleagues is clearly no longer acceptable.”3

          But what if, as a result, the alleged malpractitioner becomes the victim of a malicious attack? (see p 1756)4 I have seen several cases in which the letters, reports, and witness statements include vindictive and unsubstantiated allegations way beyond what was needed to establish the facts of the case. They were written about senior colleagues, who had to undertake responsibilities, make decisions, and maintain skills way beyond anything contributed by those now ready to join the lynch mob. The General Medical Council warns of this: “Before taking action, you should do your best to find out the facts.”1

          Once started, it is hard to stop the process, and it may gather its own momentum. In the case of deaths of children undergoing surgery for complex congenital heart disease at Bristol Royal Infirmary,5 the medical director and the chief executive were subjected to one of the most public and harrowing investigations that anyone can recall. As a result, once a doubt is raised, chiefs of hospital trusts, fearing that any subsequent adverse events will be brought to their door, may use their power to suspend. But suspension is not a “neutral act.”4 It damages its victim, and yet in possibly 80% of occasions the original allegation is unsubstantiated and the suspension is lifted.4 Perhaps we will see the pendulum swinging too fast and too hard over, from a tradition of closing ranks to a time when we will all go in fear of informers.

          Setting that extreme fear aside, there comes a time when you must speak out. My advice to the would-be whistleblower is to first test the evidence and the arguments on a wise and sufficiently senior colleague. Beware of gossips, for they have their own agendas. Next remember that you do not have to do what they say; you only have to consider their arguments. There may be a simple, quick, and obvious answer, but be cautious if the advice comes too glibly. In deciding whether to speak out, to my mind the test is this. Is it likely that, in years to come, you will regret not doing so? If so, do it now.

          And then to whom should you go? On this point there is no clear advice. I would prefer a senior medical colleague than a manager because there will be a commonness of understanding of the issues. Provided there is no doubt that the substance of the concern has been received, it may be best for the whistleblower to then go quiet. If you do not find support, are you sure enough of the facts and of your motives to run a single handed crusade? For the recipient of such an approach, my main comment is “Don't shoot the messenger” (John Nunn, personal communication). At present the system is unsatisfactory in dealing with such problems. It will never be easy, but it could be better.

          References

          1. 1.
          2. 2.
          3. 3.
          4. 4.
          5. 5.