Letters

Giving medicine a fair trial

Patients' preferences should be assessed

Patients' altruism should be appreciated

Patients' perspective must be acknowledged

Blanket enthusiasm for trials won't help

It needs to be established whether patients really fare better in trials

Improving patients' access to information may help

Patients' preferences should be assessed

EDITORAshcroft in his editorial calls for trials that do not second guess what patients want.¹ We agree that sometimes the search for the perfect design of a clinical trial is impractical. Ashcroft argues that if we are uncertain about treatments then the best treatment for the patient is the trial.

Although trials should be simple, timely, and well designed to answer well posed questions, we cannot agree that they should not assess what patients prefer. The preference of a patient deserves special emphasis when diseases or treatments affect quality of life, when the treatment entails risks or side effects, or when the choice between treatments is a "close call."² This is particularly relevant in chronic diseases where a particular symptom such as pain is the problem. The balance between the efficacy and the profile of side effects of the treatment in relation to the overall pain experience is one that only patients can judge and may be more accurate than pain measures alone. Preference for a particular treatment may promote compliance and contribute to its success. We understand Ashcroft's difficulty that endpoints relevant to patientsfor example, quality of lifemake trials harder to run and take longer to implement, but we cannot agree that the results are harder to generalise and apply.

Those who apply a true evidence based approach include only studies that are randomised, double blind, and placebo controlled in systematic reviews. The goal of good clinical trial design is to eliminate chance and bias. Without randomisation, treatment effects are exaggerated up to 40%; without effective blinding, exaggeration may reach 20%.³ The larger the sample population the more likely the results are to be credible; small trials overestimate treatment effect by as much as 30%.⁴ These facts constitute a hierarchy that has not yet been recognised in levels of evidence attributed to quality. Trials attempting perfect design may fail to yield clinically useful results, raising ethical questions of enrolling patients into studies doomed to fail.

Although we embrace the gold standard of evidence based medicine, we must employ some common sense. Jadad, a well known proponent of evidence based medicine, recently published 10 challenges for clinical trials in pain relief.⁵ These emphasise that more trials should be clinically relevant and more collaboration used over sample size, acknowledging the importance of integrating the findings from clinical trials with other types of research that must be balanced by individual values, preferences, and circumstances. This pragmatic approach requires unprecedented commitment from clinicians, research funders, journal editors, policymakers, journalists, and patients.

Laurie Allan, director chronic pain services. 
Lance Tooke, clinical research fellow, pain management. 
Northwick Park Hospital, Harrow, Middlesex HA1 3UJ

1.	Ashcroft R. Giving medicine a fair trial. BMJ 2000; 320: 1686[Free Full Text]. (24 June.)
2.	Owens DK. Patient preferences in the development of practice guidelines. Spine 1998; 23: 1073-1019[CrossRef][Medline].
3.	Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empiral evidence of bias: dimensions of methodological quality associated with estimates of treatment effect in control trials. JAMA 1995; 273: 408-412[Abstract].
4.	Moore RA, Gavaghan D, Tramer MR, Collins SL, McQuay HJ. Size is everythinglarge amounts of information are needed to overcome random effects in estimating direction and magnitude of treatment effects. Pain 1998; 78: 209-216[CrossRef][Medline].
5.	Jadad AR, Cepeda MS. Pain: clinical updates. International Association for the Study of Pain 1999;11(2).

Patients' altruism should be appreciated

EDITORAshcroft in his editorial argues for wider acceptance of the need for randomised clinical trials.¹ His case is founded on the idea of uncertainty, and herein lies both the strength and the weakness of the argument. Not only does the rest of Ashcroft's argument flow from this central point, but it also limits the argument to very few clinical trials.

Clinical trials are not conducted in a vacuum of knowledge about the treatment options available. In the case of diabetes doctors already have a fair amount of trial based information about what can be expected from each of many treatment options. A patient may be reasonably stabilised if treated with one of them. Into this scenario a proposed clinical trial is brought, of a new hypoglycaemic agent, with either a placebo or a current drug as the comparator. Is the best option for the patient to enter this trial, as Ashcroft suggests?

Most trials that come before research ethics committees are of this nature. Each involves a chronic condition (asthma, diabetes, epilepsy, hypertension, hypercholesterolaemia, depression, and psychosis) and a potential new wonder drug and is the latest in a series. Patients have to balance the relative certainty of current best treatment against the unknown potential benefits and harms of entering the trial. The harm is not all hypothetical either. Promising new hypoglycaemic agents have been withdrawn because of liver toxicity, and much vaunted antipsychotic drugs have been discovered to have serious cardiac toxicity.

Seldom do patients have everything to gain and nothing to lose by entering a trial. Future patients, on the other hand, would certainly benefit from the knowledge gained from a trial, whatever the outcome. For this reason I do not accept Ashcroft's contention that it is misleading to believe that trials are run to benefit future patients, nor that the trial is the treatment. The only clinical situation when this is true is when we are dealing with a serious condition for which there is no treatment of any degree of efficacy and doing nothing means either certain death or serious disability. In all other situations, patients who volunteer are trading a degree of certainty under the current best practice regimen for an uncertain balance of risk versus benefit under the trial protocol. This is altruism at its best, and we ought to recognise it.

Jammi N Rao, chairman, West Midlands Multicentre Research Ethics Committee. 
Sandwell Health Authority, West Bromwich B70 9LD jammi{at}bharat.demon.co.uk

Patients' perspective must be acknowledged

EDITORWe have been using Ashcroft's arguments for 20 years, trying to convince clinicians to start randomised trials instead of using unproved treatments.¹ In recent years, however, we have come to realise that things are not that simple. The argument "either you know or you don't" often does not work, since for many new (and old) treatments some evidence of efficacy is available, albeit imperfect, inconclusive, and unreliable.

Patients' decisions on treatment cannot be based solely on the results of "perfect" studies, but take into account also the size of the potential benefit associated with each therapeutic option, modulated by the personal attitudes. This was best illustrated a couple of years ago by the brother of a patient with terminal cancer, who was seeking the Di Bella treatment² for his sister. He said that he knew that the chances that this treatment works are almost nonebut might we not all be wrong? If we were to realise in a few years from now that the treatment can save lives, it would be too late for the sister. There are patients who claim they have been cured by the Di Bella treatment.

Many patients cannot wait for definitive evidence, and they want to select their treatment on the basis of whatever evidence is available, including reports on the media, patients' stories, and such like. The widespread use of unproved treatments outside randomised trials should not be labelled as entirely irrational: in many progressive diseases for which treatments of proved efficacy do not exist or are unsatisfactory (for example, many rare or advanced tumours), the randomised trial implies the possibility of not receiving the experimental treatmentthat is, of not exploiting the only chance, small as it may be, of a substantial benefit.

The case of high dose chemotherapy in breast cancer, routinely used while the clinical trials that eventually showed its lack of efficacy were still ongoing, is paradigmatic.³ What can we do to protect patients from false (and often expensive) hopes and useless or even harmful treatments, while assessing promising treatments? We are not proposing to abandon the randomised trial as the model for the assessment of medical procedures and to give up evidence based medicine. We should, however, start to reflect critically on the current methods of clinical trials from the patients' perspective, which often may differ from, and yet be as rational as, the scientific perspective.

Paolo Bruzzi, epidemiologist. 
bruzzi{at}ermes.cba.unige.it

Massimo Costantini, epidemiologist. 
Unit of Clinical Epidemiology and Trials, National Institute for Cancer Research, 16132 Genoa, Italy

Blanket enthusiasm for trials won't help

EDITORAshcroft in his editorial argues that patients in well designed and timely clinical trials do better than those not participating in trials, andconsidering also the inherent uncertainty medicine is built ontrials in themselves are therefore treatments for patients.¹ This means that there is a moral obligation for doctors and investigators to offer trials to all patients since it is unethical to offer an inferior treatmentnamely, the currently accepted standard treatment. If we find a medical condition for which there is no alternative (experimental) intervention that could serve as a control in a trial, our task is simply to find one to enable patients to be enrolled in trials.

What does that all mean?

Research ethics committees may be disbanded since there is no extra risk that research subjects should be protected from. Instead, those patients who do not have the choice to participate in trials need additional protection; therefore committees for protecting the rights and interests of getting standard treatment need to be established. The standards for informed consent for regular treatment must be higher than those for trials from now on.

Hospitals not offering clinical trials should be forced to change their practice, and those not willing or able to do it need additional supervision and permanent monitoring from governmental and professional organisations. The Helsinki Declaration should be revised or eliminated, and guidelines for protecting patients from the risks associated with standard treatment must be formulated.

Our whole way of thinking must be changed with regard to standard and trial treatment. And certainly, those who suffered or even died just because they participated in well designed and timely trials must be forgotten.

These are absurd consequences of Ashcroft's statement. Although it might be true that some patients do better in trials, this does not mean that all patients do so. Why this is so and how it may improve the standard of care may be the subject of future research. But a blanket enthusiasm for trials won't help.

Imre Szebik, postdoctoral fellow. 
Clinical Trials Research Group, Biomedical Ethics Unit, McGill University, Montreal, Quebec, Canada H3A1 W9 iszebik{at}med.mcgill.ca

It needs to be established whether patients really fare better in trials

EDITORAshcroft in his editorial argues that individual patient outcomes are improved as a consequence of being treated as part of a randomised trial, and this alone is a reason why patients should consent to such treatment.¹ Although the outcomes of participants of trials are often better than those of non-participating patients, probable explanations for this include selection bias (patients with a poorer prognosis are less often offered, or accept, randomisation).²

Recently the Cancer Foundation of Western Australia ran a full page advertisement in the state's newspaper, taking Ashcroft's arguments directly to the public.³ Beneath an eye catching photo of shark fins circling in a petri dish, the text states that patients participating in trials usually do better than those who are not. The other text of the advertisement uses such similar format and words to the recent BMJ commentary that it is almost certainly the source. As the public is likely to accept such statements at face value this may raise ethical concerns. ^{4 5}

Whenever I have obtained a patient's consent for treatment on a trial I have reassured them that their treatment was not going to be worse if they declined to participate. Are there people who believe that oncologists should tell patients that it will be?

Sean Bydder, registrar. 
Department of Radiation Oncology, Sir Charles Gairdner Hospital, Nedlands, Western Australia 6008, Australia s_bydder{at}hotmail.com

Improving patients' access to information may help

EDITORWe respond to the debate on clinical trials prompted by the editorial by Ashcroft.¹ At the Cancer Foundation of Western Australia we believe that progress in the management of cancer depends in part on the participation of greater numbers of patients in clinical trials. In conjunction with the Western Australian Clinical Oncology Group we are therefore enthusiastic to see improvement to the current poor rate or accrual into trials.

As part of our public education initiatives an annual cancer update campaign is conducted, during which invited speakers address community and professional audiences on issues of relevance to cancer control.

As part of this year's campaign in July 2000 we aimed to increase awareness of and knowledge about clinical trials. Part of the programme entailed placing a full page advertisement on the value of clinical trials in the largest selling newspaper in the state.² Other electronic and print publicity was generated. This coincided with the release of a patient brochure and booklet aimed at improving patients' access to information on clinical trials.

These measures attracted substantial attention. Over 220 people attended the lecture, presented by Professor Konrad Jamrozik from the University of Western Australia, and several said afterwards that they had changed their attitude towards clinical trials as a result of attending the lecture.

The advertisement also attracted 146 telephone enquiries to our cancer helpline. Fallowfield has reported that doctors included clinical trials among a list of the five most difficult areas of discussion during patient consultations.³

If we are able to prompt further discussion between clinician and patient about the issue of clinical trials, we believe we have made a contribution towards our cancer control objective. The outcome of this discussion is a matter for patients and their doctors.

With regard to Bydder's reassurance to his patients that their treatment would not be worse if they declined to participate in a trial⁴ our question is: how does he know?

Terry Slevin, director of education and research. 
Cancer Foundation of Western Australia Terry{at}cancerwa.asn.au

Paul Katris, executive officer. 
Western Australian Clinical Oncology Group, Subiaco, Western Australia 6008, Australia