If reading this post doesn’t give you a headache, you weren’t concentrating hard enough. 

I once anaesthetised a man for a total knee replacement. Mark wasn’t the fittest guy in the world. He was an obese, lifelong smoker with COPD who was breathless on gentle exertion. He also suffered from constant agonising knee pain that kept him awake at night and generally made his life miserable. His orthopaedic consultant told him a total knee replacement was his best option. 

I first met Mark on the day of surgery, as I did with nearly all the patients I anaesthetised as a consultant. He had attended the nurse-led preoperative assessment clinic and hadn’t been flagged up as being at particularly high risk. In fact, Mark’s chest consultant had documented in the notes that Mark’s condition was stable, and he didn’t envisage any difficulties with anaesthesia for the proposed surgery. Of course, Mark’s chest consultant had previously anaesthetised the grand total of zero patients to date and wouldn’t be anaesthetising Mark either. 

Mark immediately set my internal alarm bell ringing. After fully assessing him, I guestimated his risk of potentially life-threatening complications to be between 1 and 5 in 100. I tried to talk him out of going ahead with surgery, at least today. I felt he would benefit from time to reflect on the risks we had discussed and from the opportunity to talk his decision over with friends and family. In the interim, he could try to lose some of his excess weight and stop smoking. Mark’s mind was set, however. He insisted on going ahead. He had mental capacity, so I had little option but to accept his decision. I documented our discussion in his notes. 

That afternoon I gave Mark a spinal anaesthetic. All went well. Until he dropped down dead on the orthopaedic ward at about three o’clock the next morning. 

A queue of people lined up afterwards to ask the question: Why had Mark died? Heading the list of interrogators were Mark’s wife, his two daughters, the orthopaedic consultant surgeon and HM Coroner. They all deserved an answer. The answer, formally established at the inquest some weeks later, was that Mark had died of a heart attack. To me, that wasn’t the crux of the matter, however. I’d anaesthetised many other patients over the years with a similar or even higher estimated perioperative mortality risk, and they had sailed through their total knee replacement operations. Yes, Mark had died of a heart attack, but more crucially, he had fallen foul of random chance. He’d pushed his luck, but Lady Luck had not been on his side. The odds had not been forever in his favour. He had thrown the dice and crapped out. 

What is informed consent?

 What does it mean to give informed consent? And when a doctor tells you that there is a “high”, “small”, “low”, “rare” or “very rare” risk of something going horribly wrong, what do those words actually mean? 

I spent many years asking patients to consent to their anaesthetics. I explained the options available to them, each with its own pros and cons. The patients listened. They asked questions if they weren’t sure of something. I asked them how they wanted to proceed. They agreed to one of the options. Simples. 

You might have noticed that nowhere in the short story about Mark did I mention asking him to sign his life away on the dotted line of my anaesthetic consent form. One of the quirks of UK medical practice is that there isn’t usually a specific anaesthetic consent form for you to sign. I’ll repeat that: there is no form for you and your anaesthetist to sign that details what you discussed and what you consented to! Given that there is no form, you don’t get to keep a copy for future reference. 

When I first qualified as a doctor sometime during the Jurassic period, anaesthesia consent documentation was laughably succinct and treated more as an administrative box-ticking exercise for the hospital’s files than as a legal document describing a contract between the hospital and the patient. It comprised a small section buried within the surgical consent form (that was itself brief) that referred to you having some sort of anaesthetic or other, though it was light on detail. This wasn’t surprising because the person asking for your signature (usually a lowly surgical house officer) knew the square root of sod all about anaesthesia, giving anaesthetics and the risks versus benefits of different anaesthetic techniques. 

The NHS revamped its process for obtaining informed consent for both anaesthesia and surgery, and for recording that due process had been followed, in the early 2000s after realising somewhat belatedly that clinical practice in this area was in dire need of some serious updating. The stated aim was to improve shared decision-making. The new consent form was a more detailed, multi-page booklet with sections explicitly addressing side effects, risks and alternative treatments. In addition to making consent forms more comprehensive and tailored to the specific surgical procedure being proposed, a key recommendation of the new guidance was to obtain consent before the day of surgery wherever possible to afford patients time to reflect on the risks and benefits. A cooling-off period, if you like. Most hospitals adopted the recommendation, although obtaining consent on the day of surgery wasn’t formally outlawed until 2016. 

I hadn’t been a consultant for very long when these new consent rules came into force. There had been talk of introducing a bespoke anaesthesia consent form at the same time. It looked like it would go ahead, but then the idea was suddenly ditched at the last minute. The penny had dropped with the bigwigs in the Department of Health that if a separate anaesthetic form became a de facto standard of care, and there was a recommendation that patients’ consent be obtained before the day of surgery, the whole practice of anaesthesia in the NHS would have to change radically. Instead of patients routinely meeting their anaesthetists for the first time on the day of surgery, special anaesthetic consent clinics would need to be established at huge expense. It was a good idea, of course, but no wonder the innovation was jettisoned. Quality initiatives like that can wait until the NHS is flush with money. Which is why, even today, if you go into hospital for an operation, you won’t be given a consent form to sign by your anaesthetist. 

While that might seem a trifle dodgy, informed consent isn’t only about getting patients to sign a form, no matter how many pages it contains. Although a consent form documents what the doctor or nurse said was discussed and confirms that the patient signed it, the patient’s signature does not prove that informed consent took place. Any patient can sign a consent form without understanding what is being proposed or the risk attached to it. A patient’s signature per se is worthless if it can be reasonably shown that the process of obtaining informed consent was inadequate or otherwise flawed. 

What are the essential elements of (legally valid) informed consent?

 Informed consent is a conversation.

 1.      Nothing about obtaining informed consent should be automatic. Every patient is different, and the GMC expects doctors to find out what matters to individual patients and explain things in a way they can understand.

2.      The doctor needs to establish what the patient understands about their condition and what is being proposed – an operation, a drug therapy, some other medical intervention.

3.      The patient needs to understand what treatment involves, its intended benefits, and the risks of important side effects and complications.

4.      The patient needs to know what alternatives exist to the proposed treatment and how they compare with the proposed treatment in terms of efficacy and side effects. One alternative patients must always be reminded of is that of doing nothing. Just because you have consulted a doctor or a surgeon does not mean you have to proceed with anything. You can, and should, change your mind about going ahead, even on the day of surgery itself, if you are unsure. It’s only a postponement and not a permanent cancellation until/unless you decide to cancel.

5.      There should be ample opportunity to ask questions.

6.      A doctor may recommend a treatment option if a patient has been bamboozled by the available options, is genuinely unsure of the way forward and asks what the doctor thinks is best for them. It’s perfectly reasonable for a doctor to say, ‘If I were in your situation, I would choose this option and here’s why.’ Indeed, I said words to that effect on countless occasions during my career. If appropriate, I would also often suggest postponing surgery while they properly made up their mind. Whenever possible, I always liked to include a friend or family member in these types of discussions.

7.      The doctor must ensure the patient makes the final decision, even if it is on recommendation.

 What do patients want to know? How much information do they need to make an informed decision?

 The traditional paternalistic model of the doctor-patient relationship, the one that preaches that doctor knows best (the good old days), is out. Patient-centred care with shared decision-making is in. Hurray! Quite right, too, I hear you all cry. And I wouldn’t disagree. However, I often found that many patients, particularly older ones, hankered after the old system. How often did I hear the refrain, ‘I don’t want to know any of the details, doctor, you just do what you need to do.’ About as often as I heard, ‘I don’t understand what you are saying. How can you expect me to choose? You’re the doctor.’ 

There was often a tension, it seemed to me, between what patient advocacy groups proclaimed patients wanted to know – nay, insisted on knowing – and how little information many patients told me they actually wanted to know. That is, until it all went tits up and the patient sought legal redress. Now, according to their solicitor, my patient had wanted to know every single last detail about everything. As I apparently hadn’t been forthcoming with enough material to fill a five-year medical school curriculum in the ten or fifteen minutes available during my preoperative visit, their client’s informed consent was invalid. Can they have £100,000 compensation, please? Do I sound cynical?

 Be all that as it may, I do agree that in an ideal world, the best approach in any doctor-patient consultation is one of an equal partnership. The doctor brings years of medical knowledge and experience to the table. The patient has unique knowledge of themself: their priorities, circumstances, hopes, fears, and, crucially, their understanding of and attitude to risk. Good decision-making harnesses the synergy between these two perspectives. It allows the doctor to conclude, ‘And so, for all these reasons, I recommend Treatment X,’ and for the patient to decide whether or not to accept the recommendation or come up with an alternative plan.

 An example of patient perspective and how it can affect decision-making is illustrated by two different patients undergoing shoulder surgery. In each case, the anaesthetist recommends an interscalene brachial plexus local anaesthetic block for postoperative pain relief. The first patient is an elderly man who likes line dancing and playing chess. The second patient is a young concert pianist at the peak of their musical powers. The risk of a significant, permanent neurological injury affecting their upper limb is quoted as not greater than 1 in 1000. The dancing chess player immediately consents to the block, enticed by the welcome prospect of little or no immediate postoperative pain. The concert pianist withholds their consent to a block, fearing that even a 1 in 1000 risk is too high to gamble their passion and livelihood on.

 How much do patients understand?

 The answer to this question is pivotal to obtaining informed consent. A presumably intelligent senior orthopaedic consultant surgeon (no sniggering at the back from anyone reading this who has worked in theatre with orthopaedic surgeons) who develops severe symptoms and signs of spinal nerve root compression due to a prolapsed intervertebral disc probably understands the risks and benefits of surgical treatment better than a random man or woman in the street. Yet the anaesthetist and surgeon involved in the case must obtain the same informed consent in each case.

 I used to try to gauge my patients’ knowledge and understanding, and pitch my approach at what I thought was the right level. This meant some patients received more information than others, had a better grasp of the key issues than others and, as a result, gave more informed consent than others. I am sure that, in many instances, the consent I obtained from patients would probably have been found wanting if something had gone badly wrong and the consent process had been subsequently tested in court.

 Quantifying risk

 The law says a doctor has a duty to take reasonable care to ensure a patient is aware of the material risks of a proposed treatment and of reasonable alternative treatments, including the risk of doing nothing. A risk is material if a reasonable person would be likely to attach significance to it. The law also says that whether a risk is material cannot simply be reduced to percentages. For example, a 1% risk of dying is not the same as a 1% risk of a dicky tummy.

 In my humble opinion, analysing risk is a skill most people aren’t proficient in. Let me rephrase that: most people are bleeding useless at analysing risk. Many doctors are hopeless, too, especially surgeons. But then risk assessment is a complex, uncertain business involving the mathematics of statistical probabilities and random chance.

 What does, say, a 1 in 100 chance of dying from an operation even mean? Simply stated, it means that, on average, if 100 people with a 1% risk of dying undergo surgery, ninety-nine will survive and one will die. Note the words “on average”. If I toss a coin 100 times, on average, I will observe fifty heads and fifty tails. However, in practice, I’m much more likely to observe forty-something heads versus fifty-something tails (or vice versa). Even a hundred consecutive heads or tails is possible, albeit extremely unlikely. If you’re interested, the actual probability of obtaining exactly fifty heads and fifty tails if you toss a coin 100 times is a paltry 8%, or 1 in 12.5.

 Similarly, returning to our example of operating on a series of 100 patients with a 1 in 100 risk of dying, we shouldn’t be surprised if two or three or even more patients died. It might also be the case that none of the patients ends up dead. The chance of zero deaths occurring in a series of 100 patients with a 1 in 100 risk of death is about 1 in 3. Honestly, it’s true. Confusing, isn’t it?

 Also, a 1% risk of dying obviously doesn’t mean that you, as an individual, will end up 99% alive and 1% dead. The outcome for an individual is dichotomous – you are either 100% dead or 100% alive. Because people understand dichotomies better than probabilities, they often translate percentages into binary outcomes. If an event has a greater than 50% chance of occurring, it’s probably going to happen, and if it doesn’t, we want an explanation for why it didn’t. Conversely, if an event has less than a 50% chance of occurring, it’s probably not going to happen. If it does, we again want to know why. The closer an event’s probability is to either 100% or 0%, the more certain we are of our false dichotomy. “Probably” equates to “definitely”; “probably not” equates to “definitely not”.

 Another important consideration when trying to make sense of percentage risks is whether a 1% risk of dying is unacceptably high or acceptably low. The answer to the question depends. Suppose you’re a healthy 20-year-old with your whole life ahead of you. The mortality risk of having your painful wisdom teeth extracted under general anaesthetic is estimated by your dentist at 1% (keep your hair on, it’s not anything like that high, but we’re just supposing). In that case, you might reasonably conclude that the risk of dying is unacceptably high, and you’ll jolly well put up with your toothache, thank you very much. On the other hand, if you’re a frail 80-year-old with locally advanced kidney cancer and your anaesthetist quotes the same perioperative mortality risk to you (it’s actually significantly higher than this, but we’re still just supposing), you should have no hesitation in going right ahead. Even the more accurate risk of death of nearer 5% is better than the alternative – almost certainly dying of cancer in the coming months.

 Which side of the coin you or your doctor focuses on also makes a big difference. If your surgeon tells you that your risk of dying as a result of surgery is 1% (1 in 100), that may elicit a different reaction from saying that you have a 99% (99 in 100) chance of getting through your operation. Both statements are mathematically identical.

 If neither you nor your doctor is good with numbers, this same situation could be expressed in either of these ways:

 1.      Doctor purses his lips and shakes his head. ‘Just so you’re clear, there is a real risk you could die.’

2.      Doctor smiles reassuringly. ‘There’s always a small risk of death, of course, but I wouldn’t worry too much about it. You’ll almost certainly sail through your operation without any problems.’

Both statements are equally true but elicit a different response.

 Words versus numbers

 Even though mathematics is a precise, unambiguous science, many people don’t understand what percentages mean, so expressing risk in percentage terms is meaningless to them. A 1 in 10 or a 1 in 100 risk is perhaps easier to appreciate, but the mere mention of numbers is off-putting to many people, causing their eyes to glaze over and their brains to fog. Even when people do have some handle on what numbers mean, numbers tend to become abstract when applied to an individual’s perception of their own perioperative risk. Theoretical. They apply to other people, not us. Or, people invoke the false dichotomy principle. We tell ourselves we can safely assume anything significantly less than a 1 in 2 risk is unlikely. So unlikely, we can disregard it. And anything significantly higher than a 1 in 2 risk is likely. So likely, we can assume it definitely will happen.

 Are words any better? What is the difference between saying the risk of something is low, small, uncommon, unlikely, or rare? The problem is that these words don’t have a universally agreed meaning. If I tell ten patients that a complication is rare, I may receive ten different answers about what they think the numerical chance of it happening is. In one published study, the range of interpretation of the word “rare “ ranged from 2-21%, while “common” ranged from 24-71%.

 An official NHS-approved definition of how risky something is exists that correlates verbal definition with mathematical probability:

 Very common – More than 1 in 10

Common – Between 1 in 10 and 1 in 100

Uncommon – Between 1 in 100 and 1 in 1,000

Rare – Between 1 in 1,000 and 1 in 10,000

Very rare – Less than 1 in 10,000

 Is that helpful? I’m not so sure.

 Relative risk versus absolute risk

 If you’re not a nerd, feel free to skip this section.

 Let’s get back to numbers. I said earlier that many people struggle with the mathematics of probability, without which it’s impossible to gauge how serious a perioperative risk is in terms of how likely it is to happen to them. As if getting your head around the concept of risk is bad enough, it’s about to get worse. There are two types of risk to consider: absolute and relative. Misusing these terms (either deliberately or through ignorance) can distort risk assessment. Let me give you an example.

 The manufacturer of Drug A claims that their product reduces the risk of dying after contracting the dreaded lurgy by 15%. The manufacturer of Drug B counterclaims that their product reduces the risk of dying after contracting the dreaded lurgy by 50%. You come down with the dreaded lurgy. Which drug do you opt to take?

 Before giving me your final answer, I’ll provide you with a bit more information:

The mortality (death rate) of the dreaded lurgy if left untreated is 20% (1 in 5 people die).

The mortality (death rate) of the dreaded lurgy treated with Drug A is 5% (1 in 20 people die).

The mortality (death rate) of the dreaded lurgy treated with Drug B is 10% (1 in 10 people die).

 With this additional information, the choice is clear and the opposite of what we were going to do originally. How can the manufacturers of Drug B get away with such false claims? The answer is that they have not made a false claim. They’ve merely been fast and loose with their terminology.

 Drug A produces an absolute reduction of the risk of death of 15% (the difference between 20 and 5).

Drug B produces an absolute reduction of the risk of death of 10% (the difference between 20 and 10).

Drug A produces a relative reduction of the risk of death of 75% (the drop of 15 expressed as a percentage of the starting point of 20).

Drug B produces a relative reduction of the risk of death of 50% (the drop of 10 expressed as a percentage of the starting point of 20).

 The manufacturers of Drug A presented their efficacy data as an absolute risk reduction, whereas the manufacturers of Drug B presented their efficacy data as a relative risk reduction.

 The difference between absolute risk and relative risk matters if you’re having to decide on a particular treatment. Let’s say there’s a tablet you can take for a condition you have developed. Your doctor says drug treatment reduces your risk of dying of the condition in the next few years by 50%. He also says 10% of patients have a stroke as a side effect of treatment. What do you decide?

 You have learnt that you can’t make a rational decision until you know whether or not your doctor is talking about an absolute or relative reduction in your risk of dying by taking the tablets. In response to your request for clarification, your doctor informs you that your absolute risk of dying from the condition is 1 in 1,000,000 (1 in a million) without treatment. The tablet reduces this risk by 50% (halves it, in other words) to 1 in 2,000,000 (1 in 2 million). This equates to an absolute risk reduction of only 1 in 1,000,000 (0.0001%). You conclude that the 1 in 10 (10%) risk of a stroke far outweighs the potential benefit of taking the tablets and decline the offer of treatment.

 Final thoughts

 Each and every time we choose to take a drug, have an invasive diagnostic procedure or undergo surgery, we roll the dice. Most of the time, the dice has many sides, and the risk of the wrong number coming up is correspondingly low. Sometimes, though, the dice has relatively few sides, which means the chance of the wrong number coming up is correspondingly high. If we choose to throw the dice, the wrong number can come up no matter how many sides it has. Before making your final decision, you’ve got to ask yourself one question: ‘Do I feel lucky?’ Well, do ya, punk?

 I don’t tell you this to scare you off from getting treatment for something that needs treating. I have had several surgical procedures in my life under anaesthetic, including a major back operation. One of the risks of this operation is permanent paraplegia. I assessed the risks and benefits of treatment and made my decision accordingly.

 When conducted in accordance with best-practice guidelines, the informed consent process ensures you receive all the information you need, in a form you can understand, to make the right decision about your treatment. The process clarifies and quantifies risks and benefits. All alternative approaches to your condition are considered. You can ask questions and are given ample time and opportunity to fully consider your options.

 At the end of the day, doctors advise, but patients decide.