10
The Future of Medicine
Healthcare is serious business. Sometimes, it is deadly serious. It is also big business. Regardless of the way that healthcare is provided and financed (private or public, via taxes, insurance, or dirty cash), most civilised countries spend somewhere in the region of 10 per cent of their gross domestic product (GDP) on the health of their citizens. This is an astounding proportion of national spending. It means that one dollar (or pound or euro) of every ten that the nation earns goes on healthcare.
In the case of Britain, the NHS budget is currently around £130 billion per year, or £247,337 every minute. Despite this apparent profligacy, Britain is considered to be a relative skinflint in this area of public expenditure, in that it spends only a miserly 9 per cent of its GDP on health.* France is on around 12 per cent, and the USA is on a whopping 18 per cent of GDP. The trend for all of these percentages goes inexorably upwards as nations become richer, and their citizens live longer and expect more and better healthcare.
[* Though it is not alone: Portugal, Finland, and Australia, for instance, have similar figures.]
These massive figures and their huge impact on the budgets of nations and the pockets of taxpayers inevitably mean that healthcare must be managed in some way to ensure that the prodigious sums of money spent on it at least yield a reasonable outcome for the patients. This is purely from a financial point of view. Of course, the case for decent and intelligent management of healthcare goes beyond the mere financial: the outcomes affect people, their health, lives, and happiness, and, by implication, the lives and happiness of those around them.
So the managers and the management consultants entered the fray. To begin with, they were driven primarily or perhaps solely by a desire to reduce the costs of healthcare, and they predictably started by looking at the bottom line.
During the early years of the Thatcher government, spending on healthcare in the UK barely exceeded 3 per cent of GDP. This meant that Britain was spending far, far less than comparable developed countries. Waiting lists were interminably long, many patients did not receive basic treatment, and those who did had to wait a very long time for such treatment, sometimes as much as one or two years. We are not talking about the luxury end of medical treatment here, but about the bread and butter of therapies that the citizens of wealthy economies have come to expect: heart surgery, hip replacement, prostate operations, and so on. The conservative government under Thatcher decided that it would be a good idea to overhaul the health service, and make it more efficient, so as to obtain the best possible outcomes for the nation’s tax contributions: value for money. What that government had not (or pretended it had not) realised was that it already owned, by and large, an excellent and highly efficient national health service, whose primary problem was starvation of resources, and what it had set out to do in overhauling the management of the service was akin to the driver of a fantastic supercar who, on having difficulties in starting the engine one morning, proceeds to dismantle and rebuild the engine without glancing at the fuel gauge reading ‘EMPTY’.
So the government decided to change the entire structure of the health service, and to find ways of reducing costs by having expensive hospitals learn from cheaper ones. To this end, the government hired the services of a firm of very expensive and successful accountants and management consultants to improve the efficiency of the service, and these clever folk thought the best place to start would be cardiac surgery, because the specialty did relatively few types of operations in relatively large numbers, counted these operations, and had a reasonable idea of their costs and success rates. One of their first investigations was to compare the costs of cardiac surgery in two hospitals located in the same region of England. We shall call them hospital A and hospital B. The accountants calculated the costs of standard heart operations in these two hospitals for three broad types of operation: the first was CABG, the second included straightforward heart valve operations, and the third consisted of more difficult stuff: complex and combined operations, and ‘redo’ procedures (done on patients who have already had heart operations previously, which makes everything more hazardous). This is what they found:
|
Hospital A |
Hospital B |
|
|
CABG |
£5,600 |
£6,800 |
|
valve |
£6,400 |
£8,700 |
|
redo/complex |
£9,950 |
£4,800 |
The accountants looked at these figures, and drew what conclusions they could from them. They saw that hospital A was more cost-effective in simple CABG and valve operations, but that hospital B was more cost-effective in complicated and redo operations. They recommended that hospital A should stick to the basics, and that hospital B should be developed into a specialist tertiary-referral centre for the really complicated stuff, as it was so cost-effective in such cases.
Of course, these clever accountants never looked beyond the bottom line. Had they bothered to check the clinical outcomes in these two hospitals, they would have found a different story. Here are the mortality rates for the same operations in the two hospitals:
|
Hospital A |
Hospital B |
|
|
CABG |
3% |
5% |
|
valve |
6% |
11% |
|
redo/complex |
9% |
65%* |
* of whom 40 per cent died on the table
Therefore, the reason hospital B was so ‘cost-effective’ in treating complex patients is a simple one: many died on the operating table. In a high-risk patient, this is a very inexpensive pathway: it results in a very short hospital stay, no intensive care bills, and no recovery costs. The lesson from this appalling exercise is simple, and it is one emphasised earlier in this book. If we are interested in measuring and managing healthcare, we must measure those things that matter. Survival is an outcome of healthcare that should at least be as important as cost. In fact, there is very little point in measuring the cost of any intervention or other form of therapy until we know for sure that the therapy is effective. Once we know that, we can assess its cost-effectiveness and compare it to other forms of treatment.
It is now more than a quarter of a century since the escapades of those particular accountants, and we have learned much about the measurement of the quality of care. We still, sometimes, make the same old mistake of according too much importance to what is measurable. There is also room for improvement in our efforts to finds robust ways of measuring that which is important. Despite these advances, we in healthcare continue to sin against the above rule, and one of the biggest errors committed by the bodies that seek to regulate medical care is a very elementary one: an obsession with process over outcome.
Let me explain. What is important in healthcare for the patient (the person who matters) is the outcome of treatment, and not the process by which the treatment is given. Unfortunately, outcomes can sometimes be difficult to measure. They are affected by risk profile and are subject to statistical vagaries. Process, however, is very easy to measure (especially if you choose those bits of it which are easily measurable). This means that scores of bodies regulating healthcare can introduce myriad rules about the way things are done, from the method of greeting a patient on arrival, the way a consent form is filled in, the colour of uniforms of healthcare staff, whether all doctors and nurses wear badges, whether dozens of tedious forms are filled in correctly, all the way to the colour of the cups and saucers used in the out-patient department. In the UK, the Department of Health has even dictated the type of font used in hospital signs to direct patients and visitors to various departments, and even mandated the colour and shape of headers used in the paper correspondence between doctors. Unfortunately, it does not stop there.
Under the Labour government at the turn of the millennium, the Department of Health in the UK decided that CABG should be a priority treatment, and introduced a target for managers: no patient on the waiting list for CABG should wait more than 13 weeks. Many in the developed world would choke on their coffee if they read about such an apparently unambitious target, and would be surprised that any wait whatsoever is considered acceptable for treatment that can be lifesaving in some patients, but you have to remember that, under the previous Conservative government, waiting times for CABG were often longer than 13 months.
So far, so good: reducing waiting times for CABG is a good thing — or so it would seem at first, but let us consider the different types of CABG. At one end of the spectrum, a patient can have a single coronary artery that has been completely blocked. This patient has angina, and wants a CABG to relieve the pain and restore a normal quality of life. As the artery has been completely blocked, it cannot get worse, and that patient is safe no matter how long the wait is. At the other end of the spectrum, another patient has 99 per cent narrowings in all his coronary arteries, including what is called the left main stem, which is the very beginning of the all-important left coronary artery, before a single branch has taken off. If the left main stem is blocked, two-thirds of the heart muscle dies, and so does the patient. In more than 20 years of practice, I have only ever seen two patients survive a blockage of the left main stem: after such a blockage, there is simply not enough heart muscle left to sustain the circulation. A tight narrowing in the left main stem is the sword of Damocles. This patient is also treated with a CABG. The Department of Health target does not distinguish between the two.
Let us now look at another condition: tight narrowing of the aortic valve. Wear and tear, among other things, causes the aortic valve on the exit from the heart to become narrowed. The heart copes by working extra hard to push the blood through, and the patient feels fine for years. Sooner or later, the heart runs out of energy reserves, and begins to fail. The patient then becomes short of breath, and develops other symptoms. When that happens, it means that the heart is at the end of its tether, and heart failure and death are just around the corner. This is another sword of Damocles.
A patient who has both a tight narrowing in the left main stem coronary artery and a tight narrowing in the aortic valve has two swords of Damocles above his head, and the strings that hold them are seriously frayed. This patient needs both an aortic valve replacement and a CABG, and he really needs them quickly, before one of the swords plunges into him. Under the Department of Health’s 13-week CABG target, this patient is pushed to the back of the queue to make way for the CABG patient with a single blocked artery, for a very simple reason: a combined aortic valve operation plus CABG is not considered a CABG. Has this happened in real life? Yes, it has. Have patients died as a result? Yes, they have.
The above is a simple example of a target, based on what is measurable and introduced in good faith, resulting in worse outcomes for patients. Once again, obsession with process has trumped care about outcomes.
This type of distortion, which negates what matters by paying attention to what does not matter, pales into utter insignificance when one looks at what happened at a hospital in Staffordshire. The ‘Mid Staffordshire NHS Foundation Trust’ is a stupid managerial name for a hospital in the town of Stafford in England. This hospital failed abysmally in its duty of care to patients. Standards were so poor that it is estimated that several hundred patients (as many as 1,200 by some estimates) died needlessly under the ‘care’ of this hospital. The scandal was so large that a major parliamentary inquiry was launched, led by Sir Robert Francis. The final report of the Francis inquiry was published in February 2013, and ran to 1,781 pages. It identified many failings, poor standards, and almost wilful, downright cruelty to patients. One of the most prominent among these failings was the slavish adherence to ill-conceived targets. Sir Robert Francis wrote (my italic emphasis):
This was primarily caused by a serious failure on the part of a provider Trust Board. It did not listen sufficiently to its patients and staff or ensure the correction of deficiencies brought to the Trust’s attention. Above all, it failed to tackle an insidious negative culture involving a tolerance of poor standards and a disengagement from managerial and leadership responsibilities. This failure was in part the consequence of allowing a focus on reaching national access targets, achieving financial balance and seeking foundation trust status to be at the cost of delivering acceptable standards of care.
Mid Staffordshire was an extreme example where obsession with process has undermined care about outcomes, but we unfortunately still see such distortions to a greater or lesser degree throughout healthcare. These distortions and their consequences illustrate the hidden and unexpected dangers of allowing managers and politicians to choose what is important about healthcare. If we clinicians want to protect our patients and hospitals from the deleterious effects of such distortions, we have to determine ourselves what we are trying to achieve, find a way of measuring whether or not we achieve it, and make that the criterion by which our performance is measured. In other words: make what’s important measurable!
Despite these horror stories, at least in my specialty of cardiac surgery, we have made some important early progress. We are now good at measuring an outcome that really matters: survival. We can compare it with predicted survival, adjusting all measurements for each patient’s risk profile, and we can come up with sensible, statistically sound comparisons to guide us towards identifying problems, fixing them, and running a rudimentary quality-control programme. It is true that we still tend to focus largely on the important but somewhat crude outcomes of death and survival, but at least we do recognise that other clinical outcomes also matter. Now that the immediate mortality rate from heart surgery is vanishingly small for most patients, perhaps it is time we started to use other measures of successful outcome, such as the patient being free from symptoms and the need for future interventions, having a good quality of life, and surviving not just the operation, but in the long term. We may, also, now begin to look seriously at the cost of treatment, so that we can deliver true healthcare improvements at the lowest possible price.
Medical practice as a whole also now needs to embrace the concepts that I outline in this book. The bulk of our discussion so far has concentrated on heart surgery. This is for several reasons. First, I am heart surgeon, and therefore know a little more about the topic than other medical specialties. Second, heart surgery has blazed the trail in quality monitoring, and, where it has led, others are following or will follow. Third, measuring outcomes in heart surgery is relatively straightforward, because there is one outcome that is simple, objective, directly related to the operation, and difficult to quibble about: death (or survival, if you prefer). So, if heart surgery patients come in alive, then they leave the hospital either alive or dead. There is no halfway house, and counting the live ones on exit is very easy indeed.
The trouble with specialists in other fields of medicine is that they (or so they claim) have no such clearly defined outcome measure. How on Earth would you measure the outcomes of treatments given by a psychiatrist, or a chest physician, or a kidney specialist, or even a bone surgeon operating on dodgy hip joints? Impossible, say the medical specialists. Not at all, say I.
One patient I will never forget was the beautiful, slim, blonde-haired 16-year-old girl who walked cheerfully into the paediatric outpatient clinic in Bristol when I was a medical student. She was dressed in horse-riding gear and was accompanied by her mother, and a group of four of us medical students were sitting in on the clinic to observe and learn. The consultant talked to her and her mother, examined her chest, briefly looked at the results of her lung function tests, and discussed antibiotics with her. He then said goodbye to them with a smile, and asked her to make an appointment in six months’ time. Mother and daughter then cheerfully walked out, and, as soon as the door closed behind them, the consultant’s smile disappeared. He sombrely turned to us and said that the elegant young patient we had just seen would not attend the next appointment, simply because she would be dead before then. She had cystic fibrosis, and almost no lung function left, and her next chest infection would be fatal.
Cystic fibrosis is a terrible condition. It is a congenital illness due to a faulty gene that is quite common, in that one in 25 of the population carry it. Fortunately, it is not a dominant mutation, which means that, if you simply carry the faulty gene, having inherited it from one of your parents, you do not acquire the disease. You only have cystic fibrosis if you inherit the faulty gene from both parents. When you are conceived, you get half a set of genes from your father and half from your mother, and that combination is what makes you (and makes you unique, unless there is an identical twin). When a child is born to two parents of whom just one (mother or father) is a carrier of the cystic fibrosis gene, there is no possibility of the child having cystic fibrosis. This is why:
Father’s genes: N (normal) plus CF (faulty)
Mother’s genes: N (normal) plus N (normal)
Mixing N-CF with N-N has these four possibilities:
N-N N-N CF-N CF-N
This means that, on average, half the children produced would be carriers, but not one would have the actual disease, because you need the combination CF-CF to have cystic fibrosis.
But if both the father and the mother are carriers, the situation changes:
Father’s genes: N (normal) plus CF (faulty)
Mother’s genes: N (normal) plus CF (faulty)
Mixing N-CF with N-CF has these four possibilities:
N-N N-CF CF-N CF-CF
This means that, if two carriers get together and make babies, on average, one in four will be normal, two will be carriers, and one in four will be unlucky and have the actual disease.
Cystic fibrosis affects many organs, but the most vulnerable of all are the lungs. Patients find it difficult to clear lung secretions, so that these build up and clog the airways, leading to infection after infection, each slowly but inexorably leaving some lung damage. In the 1960s, few children with cystic fibrosis made it past the age of ten. With massive effort from parents, physiotherapists, nurses, and doctors over the past few decades, survival has improved tremendously. Nowadays, cystic fibrosis patients receive regular and vigorous physiotherapy every single day, any lung infection they develop is aggressively treated, and they are kept under close review, with care given by parents, teachers, and a dedicated healthcare team in specialist clinics. With this treatment, survival is now well into the fourth decade, and many children with cystic fibrosis now can expect to live well beyond, into their 40s.
Why am I telling you about cystic fibrosis? The reason is simple. It is one of the specialties in which one would have thought that quality monitoring is impossible. The reasons, or excuses, are legion and the following have all been said by doctors specialising in this field and in other areas of medicine:
In fact, none of these excuses holds water. Even a clinic that provides medical care to cystic fibrosis patients can have outcome measures. All we need to do is to ask a few basic questions:
In fact, one could easily argue (and I often do) that if doctors do not know what they are trying to achieve, have no way of measuring whether it has been achieved, and no idea if the achievement is good enough, then what on Earth are they doing in medicine? They’d be better employed selling snake oil.
Let us come back to cystic fibrosis: once these three simple questions are answered, how to monitor becomes self-evident. So here goes:
What is the service trying to achieve?
We know that cystic fibrosis cannot be cured, and that the progress of the disease cannot be stopped, so the best that can hoped for is a slowing down of the inevitable deterioration. Put on the spot, doctors specialising in cystic fibrosis agree with this statement.
Is there a reliable way of measuring that achievement?
Of course there is: all cystic fibrosis patients are monitored using lung function tests (in which they blow into tubes so that the capacity and effectiveness of their lungs can be measured). These measurements are routinely taken at frequent intervals in the cystic fibrosis clinic, and they are safe and non-invasive tests.
Is there an acceptable target to aim for?
Here we need to defer to experts in the field. They have agreed that, in general, a good thing to aim for is that lung function, as measured by the tests, should not deteriorate by more than 5 per cent per year. Achieving that would count as a success of the treatment.
The above is precisely what Dr Diana Bilton, who ran the cystic fibrosis service at Papworth Hospital in the 1990s, worked out and implemented. To the best of my knowledge, the service was the first non-surgical service in the world to introduce robust, objective quality monitoring with an actual, credible, relevant, and objective performance indicator. She proved that such practice is possible outside the narrow field of heart surgery and the wider one of surgery in general.
The value of such an approach is manifold. First, it informs the doctors, nurses, and other healthcare professionals how well they are doing. Second, it allows for the detection and correction of any underperformance (if you don’t even know it’s broke, there is no incentive to try to fix it). Third, it allows units to compare their performance, so that those doing less well can learn from those doing better. Finally, it allows for the assessment and audit of any treatment modification or innovation. Shouldn’t all doctors in all specialties be doing this?
So, where are we now?
I am a tiny, insignificant, ignorant bit of carbon.
I have one life, and it is short
And unimportant
But thanks to recent scientific advances
I get to live twice as long as my great great great great uncleses and auntses.
Twice as long to live this life of mine
Twice as long to love this wife of mine
Twice as many years of friends and wine
Of sharing curries and getting shitty
At good-looking hippies
With fairies on their spines
And butterflies on their titties.
These lines are by the Australian comedian and musician Tim Minchin, from his beat poem ‘Storm’, possibly one the most eloquent (and funniest) eulogies to science and modern medicine that I have ever heard. It is also a viciously sharp and effective attack on alternative medicine and unscientific thinking in general. It makes you both think and laugh out loud at the same time.
A long time ago, medical treatment was simple and cheap, much of it was safe, and most of it was utterly useless. Today, medical treatment is complex and expensive, some of it can sometimes be highly dangerous, but most of it is very effective. The progress made by medicine and surgery in the past few decades has been awesome in depth, range, efficacy, and complexity.
That said, much of our enviable good health and longevity derives not from high-tech medical intervention, but from basic public health measures, such as safe and nutritious food, clean water, decent sewerage, healthy living environments, mass vaccination, and not smoking. More can be ascribed to basic healthcare such as antibiotics for infections. Quite a bit of our good health, however, comes from modern, high-tech, and expensive medicine. The contribution of modern medicine to the overall health standard of the population is still relatively small when compared with more basic public health measures, but it is undoubtedly growing.
Heart surgery alone can now fix what was unfixable with operations that were unthinkable just 30 years ago. In fact, with the rare exception of heart failure, there is now a surgical solution to almost every cardiac problem, from coronary disease to leaky valves and enlarged aortas, all the way even to surgical treatment of electrical rhythm disturbances of the heart. Even heart failure may soon become a thing of the past when the artificial heart becomes reliable and inexpensive, and that surely is only a matter of technology and time. Progress in other fields of medicine has been similarly exciting, with keyhole surgery to fix many problems in the abdomen, phenomenal orthopaedic treatments to replace worn-out joints, kidney transplants for renal failure, liver transplants for hepatic failure, and eye surgery to treat cataracts and refraction abnormalities. Even the brain, that impossibly complex organ, is beginning to yield to surgical attempts to fix it. The results of cancer treatment are massively better than they were 50 years ago, and many cancers can now be looked at as a chronic disease rather than an imminent death sentence.
Chances are that you, if you live long enough, will need and benefit hugely from one or the other of these high-tech medical solutions to problems. A substantial proportion of us, if we reach a ripe old age, will have a worn aortic valve replaced, receive a pacemaker, have a coronary bypass, and so on, and that is just in my field. More will have joint replacements, prostate treatment, operations for cancer, and procedures on the arteries and veins of the body outside the heart.
In this book, I have deliberately drawn attention to the foibles of my profession. We often work with incomplete medical knowledge. We occasionally make stupid decisions. We take unnecessary risks with our patients’ lives and wellbeing. The good news is, we are finally embracing the concept of ‘total quality management’, albeit half a century after the Japanese carmaker Toyota incorporated it into its production systems, and we are beginning to understand risk management and safety protocols two decades after civil aviation. We still have a long way to go.
Despite this, I believe that medicine has never been better than it is today. Our treatments are mostly effective. We can make patients feel better. We can help them live longer. Our practice is largely based on evidence. Quality assurance is rapidly becoming integral to our processes. We have started to learn from crashes and near misses, improving our methods and our systems. And there is increasing transparency in our work. In every single aspect of the improvements listed above, cardiac surgery has played the lead role. For that, my colleagues and I may perhaps be forgiven the indulgence of a little pride.
Tomorrow, or sometime in the future, you or someone you love will perhaps need a heart operation. My hope is that the material in this book will help you to ask the right questions, consider the options, and reach the right decision. If you decide to go ahead and place your life in my hands, or in the hands of one of my colleagues, you can do so with confidence. Medicine has never been as good as it is now, and nowhere is that truer than in the specialty of heart surgery.