By Richard Smith

I'm thinking of starting a new political party. Its working title is the "Life is hard; we have no solutions" party. It will work in a completely different way from traditional parties. They promise everything. We will promise nothing. Our manifesto, which is still under development, is shown in the box on page xiii. It seems unlikely that we will come to power soon, but we will eventually when the electorate tires of the political cycle of a party promising everything, getting elected, failing to deliver, and being replaced by its opponents who in their turn promise everything. One of the stimuli to the new party is this magnificent book. It grew out of our uncertainty and confusion at the BMJ, and it's a book that provides no easy answers but will hugely deepen your understanding of informed consent. The book is a tribute to sharing uncertainty and confusion.

The BMJ had to decide three years ago whether to publish two original studies where the participants had not given informed consent. Publication is of course a binary event. Either we published or we didn't publish. We had intense arguments over what to do, and several advisers told us that we would be disgraced if we published. Nevertheless, we thought that there was much to be gained and little to be lost by sharing our uncertainties. I hesitate to say that we were right, but this book had its roots in our uncertainty.

We published the studies together with a series of commentaries, and we invited Len Doyal, the five star ethicist, and Jeff Tobias, the clinician at the coal-face, to write opposing articles on whether we were right to publish some studies that didn't include fully informed consent. They thus started as opponents but then came together to edit this rich book. The book is made especially compelling and interesting by their dialectic running right through it.

The issue of the BMJ that included this collection of material produced a flood of well argued correspondence, and we recognised that we had touched a nerve. The BMJ issue was followed by a conference, where the focus began to shift from informed consent in research to informed consent in clinical practice and teaching. It became very clear that standards in clinical practice and teaching are considerably below those in research. Yet the spotlight continues to be shone on research because it must be approved by ethics committees and the results are held up to the world to be scrutinised. Clinical practice and teaching have nothing like the same exposure.

This book gathers together much of the material that arose from the BMJ and the conference and moves it on. Our small scale confusion is embedded in its historical context and then developed with a series of profound pieces on all aspects of informed consent.

An issue that becomes more central every day Books take a long time to put together, and publishers always worry that the world may have lost interest in the subject of their book by the time that it is published. Here, I would argue, the opposite has happened. With every day that goes by the issue of informed consent becomes more central in health care. The driver is the changing relationship between clinicians and patients.

Clinicians were like politicians. They knew best. They would solve the patient's problems. So long as patients and society trusted them to do what was best then they would get on with their mysterious processes, including some judicious research without consent, and everything would be alright. In those days trust and ignorance went together. Now that's unacceptable. Patients are better educated. Deference is outdated. A series of scandals have exposed serious abuses within medicine and research. We now live in a world where what is closed is immediately suspect. Lack of transparency implies incompetence, corruption, or bias.

So clinicians increasingly relate to patients as partners, and there is growing evidence that this partnership brings better clinical outcomes, greater patient satisfaction, and reduced costs. It can also be beneficial to clinicians. They do not have to pretend to know what they do not know and be able to do what they cannot do.

A successful partnership depends on effective interchange of information. If the doctor keeps something from the patient then trust is likely to be broken, particularly if the doctor does something to the patient without consent. Trust comes now not from ignorance but from sharing uncertainty. Here I must confess that this is more the view of an informed patient than a practising doctor. It's 20 years since I saw patients, and some of my friends who are seeing patients everyday tell me that I'm out of touch. I've spent too much time with ethicists and the chattering classes. I've forgotten the realities of practice, which include dealing with frightened and impaired patients who don't want to be overloaded with information and patients who for one reason or another have great difficulties handling complex information.

But even if it's true that some patients chose to know little, most do want to know – and there is no going back. Informed consent is coming ever more to the fore, and often it's a battleground. I'm writing this in July 2000, and one news page in this week's BMJ contains two stories that relate directly to consent. A new federal agency in the United States has stopped all government sponsored clinical trials at the University of Oklahoma Health Sciences Center at Tulsa. Researchers conducting a trial of a vaccine for malignant melanoma had failed to tell the patients of the vaccine's safety problems. They had also allegedly misled candidates for the trial by telling them that the vaccine might reduce the size of their tumours when the aim of the study was to test the toxicity of the vaccine.

The story underneath tells how researchers in Britain are having great difficulty recruiting patients to a big trial that aims to assess the effectiveness of chemotherapy in non-small cell lung cancer. Of 253 eligible candidates only 63 (25%) have agreed to enter the trial. Patients seem less willing than in previous studies to accept a high degree of toxicity. The possibility arises in my mind that these patients may be being told more about the possible side effects than participants in previous trials.

Clinicians and researchers

Some of those who read this book, particularly those who are researchers, will feel sympathy for the researchers in the lung trial. They are addressing a difficult and important clinical problem. Many patients with non-small cell lung cancer are treated every day all over the world, yet, as the lead researcher in the trial says, considerable uncertainty remains over the effectiveness of adding chemotherapy over standard treatment. That's why they are doing a trial.

But what do clinicians all over the world say to these patients? It's unlikely, I suggest, that they share all the uncertainty with their patients. The clinicians may pretend that they know more than they do, or most likely – they pretend to themselves. They can then with a good conscience recommend a particular line of treatment. If, however, they choose to acknowledge the uncertainty (and perhaps join my new political party) then they are into something difficult. They will have to share their uncertainty with the patients. They might also feel an ethical obligation to start a trial to answer their uncertainty. Then they have to design a protocol, seek approval from an ethics committee, and seek from patients not only consent for treatment but also for randomisation. Researchers, many of whom are clinicians, resent what they see as these "double standards". Some want to be able to adopt the "lower" standards that apply to the clinicians, but I think they misread the zeitgeist. What is more likely is that clinicians will have to learn to share their uncertainty.

From Pappworth to COPE

This territory in health care where all is uncertain and where clinicians and researchers are both found is the source of many difficulties. Clinicians want to help patients. Researchers want to reduce the uncertainty. Patients would like things to be certain but mostly, evidence suggests, want to know about the uncertainty. It's easy to take a wrong step in such territory, which covers much of health care.

I thought about this territory when I read the beginning of Maurice Pappworth's preface (reproduced here) to his important book Human Guinea Pigs in which he exposed the wide extent of experimentation without consent in Britain (p. 39). The book was published in 1967, but his words are as applicable now as then.

"The main purpose of this book is to show that the ethical problems arising from human experimentation have become one of the cardinal issues of our time…I believe that only by frank discussion among informed people, lay as well as medical, can a solution be reached."

To bring the words completely up to date I would broaden Pappworth's phrase about "ethical problems arising from human experimentation" to "ethical problems arising from managing uncertainty in clinical practice and research". It also sounds old fashioned to me to think that "a solution can be reached". In the postmodern world, all truths are provisional and all solutions are not only incomplete but also suspect.

Thirty years after Pappworth wrote his book, the Committee on Publication Ethics was formed, and anybody reading COPE's annual reports might be reminded of Pappworth's book. COPE advises editors of medical journals on ethical problems thrown up by papers they are considering or have published, and the annual reports describe these cases. The cases are rarely as shocking as those included in Human Guinea Pigs, but they have similarities. I have dealt with several cases where clinicians have developed theories on new treatments, including surgical treatments, and have tried them out on patients without getting any approval from an ethics committee. They have also conducted their experiments in ways that I believe do not allow for any confident conclusion, which in itself raises ethical problems. Research that is scientifically unsound may be regarded as automatically unsound ethically. When I challenge these authors they usually tell me that patients have consented to the treatment. Clearly they have in the sense that they have taken the tablets or undergone the operation, but did they understand that they were part of an unscientific experiment? I doubt they did because surely then they would not have consented. But then the doctors did not regard what they were doing as experiments, even though they have written up their studies for the BMJ. They think of them as accounts of clinical practice. I think of them as explorations of uncertainty, and I think that they should share their uncertainty with the patients. They think I'm an ethically obsessed editor. We don't agree. Readers of this book are not expected to agree with everything it contains. Indeed, they couldn't because the book is full of disagreement. I hope, however, that readers will agree that informed consent is a central issue for health care, that the problems are real and pressing, that no solution will be found, and that this is a marvellous book.

Richard Smith

Editor, BMJ

Competing interest statement: I am the chief executive of the BMJ Publishing Group, which is publishing this book, and I am the editor of the BMJ, from which some of the material comes. I am, however, paid a fixed salary and will not benefit financially from any success this book might have.

Manifesto of the "Life is hard; we have no solutions" party

Death is inevitable. Prepare for it.

We live in a world of competing sorrows.

You will rarely understand and will be constantly misunderstood. You identify the problems. You do something about them, recognising that all solutions tackle only part of the problem and give rise to other problems. Life is full of inevitable contradictions.

When things go wrong, as they will, little is gained by blaming somebody.