Admitting when mistakes are made

How has the law changed?

Among the results of the Francis report into the failings at Mid Staffordshire NHS Foundation Trust was the introduction of a statutory duty of candour aimed at obliging NHS organisations in England to be open and honest with patients when things go wrong.

The law applies to organisations, not individuals. Breaches are a criminal offence, which can be punished by fines, but the real damage is reputational. Breaches do not depend on the degree of harm a patient may have suffered: the offence is to have breached the regulations. This is akin to driving at 100 mph on a motorway, which is an offence even if no accident occurred.

But aren’t doctors already obliged to be open and honest with patients?

Yes. A professional duty of candour has existed for many years and is enforceable by the General Medical Council. The GMC and the Nursing and Midwifery Council recently issued new guidance on what this means in practice.3

Critics such as Action against Medical Accidents (AvMA) argue that this obligation was ineffective because the GMC was inconsistent in enforcing it, and doctors were under pressure from hospital management not to speak out. “Doctors were told, ‘Don’t say anything about it’ when things went wrong,” says Peter Walsh, chief executive of AvMA. “It put them in an impossible position.”

So the statutory duty of candour now puts managers in the same position as doctors?

Not quite. The thresholds for action in the two systems are different. The statutory code covers “notifiable safety incidents,” defined by the Care Quality Commission (CQC) as those causing death, severe harm, moderate harm, or prolonged psychological harm.4 Such incidents are estimated to total 100 000 a year, excluding near misses. The GMC code is broader, makes no attempt to define a threshold, and says it is up to professionals to decide whether or not to discuss near misses with patients.

So we have two separate duties of candour that are slightly different?

Yes. Bringing the two into direct alignment was rejected in a review carried out by David Dalton, a senior NHS manager, and Norman Williams, former president of the Royal College of Surgeons of England.5 They said that they could see the force of the argument for having organisations “on the hook” for the same thing as the professionals who work for them, but they did not accept it. “It is possible for the professional and organisational duties to be mutually supportive without applying to exactly the same incidents,” they concluded.

The Medical Defence Union disagrees. “We believe there has never really been a need for a statutory duty of candour,” says Catherine Wilks, deputy head of advisory services at the union. “It is potentially confusing to have different thresholds. But our surveys show that doctors are well aware of their professional obligations, which are very clearly laid out in Good Medical Practice, and that is the one they should follow.”

What obligations does the statutory duty of candour impose?

It says that, “as soon as reasonably practicable” after a notifiable safety incident has occurred, the patient must be told, in person, what happened, why it happened (so far as is known), and what further enquiries are being made. This notification must include an apology and be recorded in a written record.

So doctors have to apologise for making mistakes. That’s a good thing, surely?

Nowhere in the regulations or the guidance does the word mistake appear. The statutory code relates to whether a notifiable safety incident has occurred, not what caused it. This may or may not have been a medical error—it could have been chance or the inherent risks of the procedure or treatment. Thus the doctor may be apologising not for making a mistake but for the fact that medicine is an imperfect art. This means that arguing, “I’m not going to apologise, I didn’t make a mistake,” is not a get-out.

Walsh believes this is wrong. “We at AvMA would never have drafted the rules in this way,” he says. “I don’t think it’s appropriate to enforce apologies—they should come naturally. A box ticking approach devalues the apology, and it’s unhelpful.”

The GMC guidance says an apology is due whenever “things go wrong.” Its definition of things going wrong is wider than the statutory code, so if doctors follow the GMC guidance they could find themselves apologising for things that trusts believe are not notifiable safety incidents, and which they do not report as such.

Apologies remain a controversial matter. “The credibility of apologies depends on their being freely given,” the BMA said in its response to the GMC’s consultation on this topic. If an incident should subsequently lead to a fitness to practise panel, the GMC says it will be influenced by the “insight” a doctor has shown, which could mean, in the BMA’s view, that apologies are being extorted under threat of regulatory pressure.

When the GMC asked, in a separate consultation,6 whether fitness to practise panels should have the power to require a doctor to apologise, opinion was split. Patient organisations said yes, bodies representing doctors said no. Individual doctors voted no by a margin of 161 to 114. The GMC withdrew the proposal. So the position now—paradoxical, some may think—is that doctors should apologise promptly after a safety incident, whether the mistake is theirs or not, but cannot be compelled to apologise if the incident goes to a panel, and they are found at fault.

When should apologies be given?

The statutory code says that the apology should be made “as soon as reasonably practicable,” and the GMC guidance says, “as soon as possible.” But how soon is soon? Neither spells it out. There is a hint in the contract signed by all NHS providers since 2013, which makes candour a contractual obligation. This says that notification of incidents should be within “at most 10 working days” and sooner where possible.

To whom should apologies be made?

To the patient, or “relevant person” in CQC-speak. There is no obligation to tell relatives, except in cases where the patient has died, is under 16, or lacks mental capacity. Other than in these cases, information should only be disclosed to family members or carers when the patient has given express or implied consent, the CQC says. The GMC guidance adds an extra twist by saying that there should be someone available to support the patient—a friend, relative, or professional colleague—but it does not say this person must be physically present.

What if it isn’t yet clear what went wrong?

This isn’t an excuse to procrastinate. The obligation is to share “all you know and believe to be true about what went wrong and why, and what the consequences are likely to be,” the GMC guidance says. If the patient doesn’t want to know any details, and sticks to this position after discussion, then it is not obligatory to burden them. The fact they have demurred must be recorded, and to meet the statutory code the whole process must become part of the written record.

Special difficulties arise in community and mental health settings, as both NHS Providers and the Royal College of Psychiatrists have pointed out. Harm may occur when the doctor or nurse is not even present. “When individuals or their families are distressed because of an instance of harm, disputed ground is likely to be relatively common and boil down to opinion,” NHS Providers says. “It is difficult to fully understand the role of a legal duty of candour in circumstances where one may be dealing with opinion rather than fact.”

The Royal College of Psychiatrists warns, “Giving incorrect information will potentially cause further harm and reduce rather than build trust. There may be circumstances where the professional may be put at personal risk by this action.” Neither the CQC nor the GMC guidance addresses what to do when patients or their relatives become angry or aggressive.

How have other countries tackled this matter?

In Australia a system called open disclosure covers much the same ground as the duty of candour.7 While generally supported by doctors, the medicolegal consequences of disclosure and the need to seek early advice from insurers have proved barriers to universal adoption.

In the United States, says Blair Sadler, who for 26 years was president and chief executive of Rady Children’s Hospital in San Diego, “there are islands of excellence where people have shown compassion under fire and overcome the three main obstacles to transparency—which are fear of litigation, fear of media criticism, and fear of damage to reputation—but far too many that have not embraced that concept.” That doesn’t make them bad places, he adds.

His own experience has not borne out any of these fears. “On balance I believe that if the principles of openness are followed, the incidence of litigation is lower, the risk of being vilified in the press is lower, and the risk of ending your career is lower.” Using the law to reduce the fear, by protecting those who speak out, can be helpful. “But if the only reason is a tick-box exercise, then the risk is that management will obfuscate and do the minimum that’s required.”

In Sadler’s view there are three victims of medical incidents: the patient, the doctor (whose psychological trauma can be severe), and the institution. “It’s important to deal with them in that order,” he says. “Institutions that are failing in openness often put them in the reverse order, and that’s what has to be put right.”