Information science

In January 2002 the eMJA website began providing open access to the full text of each issue of the Journal as soon as it is published. Since then 3000–7000 people have visited the eMJA every day — not just Australian doctors and citizens, but people from all over the world. Apart from the altruistic glow that comes from subsidising this access, what is its value to the members of the Australian Medical Association (AMA) and other subscribers who fund the considerable cost of publishing the MJA? Like everyone else, they get immediate access to the Journal via the Internet without having to fuss around with passwords and usernames. For many purposes, the Internet Journal may be more convenient than using the printed edition: back issues never get mislaid, and there is a useful search tool that makes it easier to track down a half-remembered article. The Internet edition includes links to related articles in the MJA, and to other journals via PubMed, the US National Library of Medicine's online index of the medical literature. And it is a lot easier to copy the citation details and other text from the Internet edition than from paper. However, should these conveniences be restricted to those who pay for the Journal? Most medical journals on the Web, including the "big four" — the New England Journal of Medicine, The Lancet, the Journal of the American Medical Association and Annals of Internal Medicine — restrict access to paying customers. There are exceptions, including the BMJ (which has provided full text free since 1998), the Canadian Medical Association Journal, and CA: A Cancer Journal for Clinicians. There are academic arguments that the scientific medical literature should be freely available because of its importance in medical advancement and education — ultimately, because of its importance in saving lives and ameliorating suffering. There are pragmatic counterarguments that producing a journal is expensive, and that the people who use it should pay for it. Both views were aired at length in the debate inspired by Harold Varmus, Director of the US National Institutes of Health, when he proposed a free electronic archive of biomedical research.1 Many journals are now finding compromises between the idealistic and the commercial arguments. Some release their archives for free access six months after publication, reserving the up-to-the-minute content for paying customers (eg, Antimicrobial Agents and Chemotherapy, Molecular and Cellular Biology, Proceedings of the National Academy of Sciences). Some journals make content freely available only in the developing world, where medical information is hardest to afford and also most needed (eg, Archives of Disease in Childhood, Journal of Bone and Joint Surgery, and the Journal of Clinical Investigation). We at the MJA need to know what our paying readers think about this. Producing the eMJA costs about $150 000 a year, or about $5.60 per AMA member annually (this is on top of the much greater cost of producing the MJA itself). Do AMA members wish to emulate the example of members of the British Medical Association in supporting an Internet journal that is a gift to the world? Should free access be limited by time or place? How many subscribers and AMA members are happy to receive the Journal only in electronic form, thereby saving the costs of printing and post? Meanwhile, we are developing new features for the eMJA, such as Internet-based manuscript submission and peer review; discussion lists attached to articles for rapid postpublication debate; and articles in a format suitable for downloading to personal digital assistants ("palmtop computers"). In these developments too we ask for guidance from AMA members. What features are most useful in the eMJA? What would you most like us to add? eMJA vital statistics 3000–7000 visitors every day Three most accessed articles published in 2002 (to end October): 1. Chronic fatigue syndrome. Clinical practice guidelines — 2002. Med J Aust 2002; 176 (Suppl 6 May): S17-S55. <http://www.mja.com.au/public/guides/cfs/cfs2.html>. Accessed 9845 times in six months. 2. McLaren B, Shelley JM. Reported management of early-pregnancy bleeding and miscarriage by general practitioners in Victoria. Med J Aust 2002; 176: 63-66. <http://www.mja.com.au/public/issues/176_02_210102/mcl10306_fm.html>. Accessed 7532 times in nine months. 3. McGrath BP, on behalf of the National Blood Pressure Advisory Committee of the National Heart Foundation of Australia. Ambulatory blood pressure monitoring. Med J Aust 2002; 176: 588-592. <http://www.mja.com.au/public/issues/176_12_170602/mcg10817_fm.html>. Accessed 4920 times in four months. Most accessed article published online, 1996 to 2002: Kidson W. Polycystic ovary syndrome: a new direction in treatment. Med J Aust 1998; 169: 537-540 <http://www.mja.com.au/public/issues/nov16/kidson/kidson.html>. Accessed 38 188 times in 2002. Estimated origin of visitors, based on data for visitors from known internet domains: Australia 68%, North America 16%, Continental Europe 5%, South America 4%, Asia 3%, United Kingdom 3%, Oceania 1%. We would like your views on the eMJA and the access policy Contact: Craig Bingham, Communications Development Manager, MJA Email: cbingham at ampco dotcom dotau. Post: Locked Bag 3030, Strawberry Hills, NSW 2012

Craig M Bingham BA(Hons), DipEd · Martin B Van Der Weyden MD, FRACP, FRCPA

A dream becoming a reality The Cochrane Library is now available free to all Australians who have Internet access. At the 3rd Annual Meeting for Australasian Contributors to the Cochrane Collaboration, held in Melbourne in October this year, the Federal Minister for Health and Ageing announced this important milestone — universal access to high-quality health information. Just as ready access to "clean" drinking water has come to be seen as a public health milestone of the 19th century, so, in the future, ready access to "clean" health information might well be dubbed as a major public health achievement of this century. The United Kingdom and Ireland, as well as several other countries in Europe, also have free access and the Library is also available free to developing countries. Such access will have a direct impact on satisfying clinicians' daily information needs, and an indirect impact on their practices through the information accessed by patients, consumer organisations, policymakers, and others. So where will this lead us? To understand some of the implications requires an understanding of both the history and the future of the Cochrane Collaboration.1 In 1979, a challenge came from the UK epidemiologist Archie Cochrane, who stated: "It is surely a great criticism of our profession that we have not organised a critical summary, by speciality or subspeciality, adapted periodically, of all relevant randomised controlled trials."2 Meeting Cochrane's challenge required two important steps: (i) assembling all controlled trials in one database, and (ii) completing and maintaining systematic reviews of these trials. The first step corresponds to the Cochrane Central Register of Controlled Trials, which now includes over 350 000 trials, and the second to the Cochrane Database of Systematic Reviews, which currently contains 1456 complete Cochrane reviews, and the protocols for 1101 planned reviews. How did the Cochrane Library come to be? In Cochrane's 1979 article1 he challenged the medical profession in general, but singled out obstetrics as the specialty most in need of an evidence base from controlled trials. When Iain Chalmers (who had worked in obstetrics) became Director of the UK National Perinatal Epidemiology Unit in 1978, he initiated a classified bibliography of randomised trials of interventions in pregnancy, childbirth and early infancy, using both electronic searches and manual searches of over 60 journals. This bibliography provided the raw material for an international collaboration to prepare systematic reviews, which were eventually published in 1989 in a seminal, two-volume, 1500-page book entitled Effective care in pregnancy and childbirth.3 The book concluded with a chapter summarising which interventions (of the 283 assessed) were supported by reasonably strong research evidence (100 were deemed effective, 36 promising, 86 of unknown benefit and 61 so unlikely to be useful that they should be abandoned). Importantly, not only was a paperback summary prepared for women published concurrently, but also a six-monthly electronic update of the systematic reviews — the Oxford Database of Perinatal Trials. This "pilot" project was well received. In 1992, the UK National Health Service provided crucial support for Chalmers to work with others to extend the process to other areas of healthcare. Recognising that the work could not be done by a single group or country, the international Cochrane Collaboration was founded in 1993 at the first Cochrane Colloquium, held in Oxford. Two years later the Cochrane Database of Systematic Reviews was launched, and the late Chris Silagy, inaugural Director of the Australasian Cochrane Centre, became the first elected Chair of the Steering Group, guiding the growth of reviews prepared by members of 49 Collaborative Review Groups, which, collectively, are responsible for covering most health problems. At the 1995 Cochrane Colloquium, David Sackett described the Cochrane Collaboration as a plane that took off while it was still being built. In 2002, the Collaboration is flying at a respectable altitude. But what still needs to be done for Cochrane's dream to become a reality? One sobering fact is that less than 10% of more than 350 000 published controlled trials have been synthesised within Cochrane reviews. To complete the journey will require three things: Review efforts will need to be sustained and extended by appropriate support and training in systematic reviewing. The science of research synthesis will need to develop an academic capacity and infrastructure equivalent to those of other fields of specialised medical endeavour. To maximise clinical relevance and uptake, the Library interface and reviews have to become more attuned to the needs of users and the users need to be more sophisticated in applying evidence. Reviewers will need to have access to all trials, not just those published in electronically indexed journals. Registration of all trials at inception is necessary, and is becoming a reality (see the meta-register at www.controlled-trials.com). Finally, we need to recognise that the Cochrane Library synthesises only intervention studies. At the 1996 Cochrane Colloquium, Hilda Bastian, chair of the Cochrane Consumer Network, suggested "people often ask if we can afford to extend the Collaboration beyond the RCT; we also need to consider whether we can afford not to do so". We will also need more systematic use of non-randomised study data on harms or treatments, and equivalent collaborations for systematic reviews of the accuracy of diagnostic tests, the natural history and prognosis of disease, and other types of clinical questions (see, for example, a description of Bayes Library of Diagnostic Studies and Reviews: <www.bice.ch/engl/content_e/bayes_library.htm>). In brief, although the announcement of free access to the Cochrane Library is an important and welcome milestone, much work remains to be done to make best use of the presently available results of clinical research relevant to the wellbeing of users of health services. Sceptics might ask what evidence there is that ready access to such databases will make a difference. Free access might be seen as an ethical obligation — in return for the public's participation and direct or indirect funding of research. However, adequate evaluation is also essential to both assess the impact and to guide improvements. The National Institute of Clinical Studies (<www.nicsl.com.au/>), which brokered the free Cochrane Library access, will also undertake an evaluation that will include process measures, such as who is accessing what, and a more detailed study of the difficulties and needs of end-users. However, we should recognise that information access is a necessary, but not sufficient, condition to bridge the gap between research and practice. It may be far from enough. Hence, more detailed evaluation might be planned to explore factors such as the impact of clinicians' skills in using evidence, attitudes to applying evidence, structural barriers to using proven interventions, and problems in matching evidence to individual patients' needs that limit any potential benefits of free access.

Paul P Glasziou PhD, FRACGP

Randomisation is the process of assigning clinical trial participants to treatment groups. Randomisation gives each participant a known (usually equal) chance of being assigned to any of the groups. Successful randomisation requires that group assignment cannot be predicted in advance. Why randomise?If, at the end of a clinical trial, a difference in outcomes occurs between two treatment groups (say, intervention and control) possible explanations for this difference would include: the intervention exhibits a real effect; the outcome difference is solely due to chance; or there is a systematic difference (or bias) between the groups due to factors other than the intervention. Randomisation aims to obviate the third possibility. Allocation of participants to specific treatment groups in a random fashion ensures that each group is, on average, as alike as possible to the other group(s). The process of randomisation aims to ensure similar levels of all risk factors in each group; not only known, but also unknown, characteristics are rendered comparable, resulting in similar numbers or levels of outcomes in each group, except for either the play of chance or a real effect of the intervention(s). Statistical analyses of clinical trials assume that randomisation was used and was "successful". The analytic tests used give the likelihood of chance explaining a difference of at least the magnitude observed. If this likelihood is small, we conclude that the observed difference was due to a real effect of the intervention. Successful randomisation allows for valid statistical interpretation of "raw" results (ie, estimates that are unadjusted for other patient characteristics). However, successful randomisation does not guarantee perfect balance in risk factors between groups (due to the play of chance), so adjusted analyses can also help in further interpretation of outcome results. In a clinical trial report, it is important to document that random allocation of treatment assignment was successfully achieved. The CONSORT statement1 suggests that the sequence generation, allocation concealment and implementation be reported (Box 1). Sequence generationSimple randomisationSimple randomisation is the most basic method of random treatment assignment. This can be thought of as tossing a coin for each trial participant, A being allocated with "heads", B with "tails". However, it is not usually performed using a real coin-toss, as issues of concealment, validation and reproducibility arise (see below). Simple randomisation is usually achieved using a sequence of random numbers from a statistical textbook, or a computer-generated sequence. Permuted block randomisationIn a large trial (at least 1000 subjects), simple randomisation should give a balance in number of patients allocated to each of the groups in the trial, but for a "small" study the numbers allocated to each group may not be well balanced. In small trials, to maintain good balance, blocked randomisation may be used. "Blocks" having equal numbers of As and Bs (A = intervention and B = control, for example) are used, with the order of treatments within the block being randomly permuted (Box 2). A block of four has six different possible arrangements of two As and two Bs.2,3 A random number sequence is used to choose a particular block, which sets the allocation order for the first four subjects. Similarly, treatment group is allocated to the next four patients in the order specified by the next randomly selected block. The process is then repeated. Permuted block randomisation ensures treatment group numbers are evenly balanced at the end of each block. Stratified allocationStratified block randomisation can further restrict chance imbalances to ensure the treatment groups are as alike as possible for selected prognostic variables or other patient factors. A set of permuted blocks is generated for each combination of prognostic factors. For example, in a trial of chemotherapy for breast cancer, suitable stratification factors might be menopausal status and oestrogen-receptor status. A set of permuted blocks is generated for those women who are premenopausal and oestrogen-receptor negative, another set for those who are premenopausal and oestrogen-receptor positive, and so on. Stratification can add to the credibility of a trial, as it ensures treatment balance on these known prognostic factors, allowing easy interpretation of outcomes without adjustment. Dynamic (adaptive) random allocation methodsSimple and block randomisation methods are defined, and allocation sequences set up, before the start of the trial. In contrast, dynamic randomisation methods allocate patients to treatment group by checking the allocation of similar patients already randomised, and allocating the next treatment group "live" to best balance the treatment groups across all stratification variables. Minimisation3 is one such method, and can be implemented using a manual card system, but dynamic methods are best implemented on computer (Box 3). Inappropriate randomisation methodsMethods of allocation such as alternate allocation to treatment group, or methods based on patient characteristics such as date of birth, order of entry into the clinic or day of clinic attendance, are not reliably random. Such allocation sequences are predictable, and not easily concealed, thus reducing the guarantee that allocation has indeed been random, and that no potential subjects have been excluded by foreknowledge of the intervention. Concealment of the allocation processIt is very important that those responsible for recruiting people into a trial are unaware of the group to which a participant will be allocated, should that subject agree to be in the study. This avoids both conscious and unconscious selection of patients into the study. "Allocation concealment" is the term used to describe this process and underpins successful randomisation strategies.4,5 For multicentre clinical trials, central randomisation by telephone, interactive voice response system, fax or the Internet are ideal methods for allocation concealment. The clinician or data manager at the participating site assesses eligibility, gains consent, and makes the decision to enrol a patient, then calls the randomisation service to get the treatment allocation. Central randomisation also enables trial coordinators to monitor randomisation rates, and have a record of all allocated patients for potential follow-up. For single-centre clinical trials, it is usually possible to identify a staff member not involved with the trial who can keep the randomisation list or envelopes, preferably in a location away from the clinic or ward where patients are being assessed. For example, pharmacy staff may be able to undertake randomisation. They should be instructed to keep the list private, and to only reveal a treatment allocation after receiving information demonstrating that the patient is eligible and has consented to the trial. In situations where remote randomisation may not be feasible or desirable, a set of tamper-evident envelopes may be provided to each participating site. The envelopes should look identical, and each should have the trial identification and a sequential number on it. Inside is the treatment allocation and usually a trial identifier for the patient (eg, unique sequential number). After assessing eligibility and consent, as described above, the next envelope in sequence is opened. Care needs to be taken that the envelopes are opaque and well sealed, and that the sequence of opening the envelopes is monitored regularly. For example, the patient identifiers could be written on the envelope, and the contents of the envelope, along with the date and time of randomisation, transcribed to the randomisation form where eligibility assessment was recorded. Stratified randomisation is still possible using randomisation envelopes by having a set of envelopes for each combination of stratification factors. A screening log should be considered to help ensure that eligible patients were not missed, and were not excluded on the basis of study staff somehow knowing the next treatment allocation. Concealment through sequence generationAllocation concealment may be thwarted by an inappropriate choice of randomisation sequence generation. For example, a permuted block design with a fixed block size of four, in an unblinded study6 where treatment group is revealed at the time of randomisation, may make it easy to predict the next allocation once three patients have been randomised. For this reason, details of block size should not be revealed to investigators or other study staff. A varying block size can also be used (eg, blocks of size 4, 6 and 8 randomly arranged). Dynamic allocation methods provide a more secure method of allocation concealment. ImplementationThe trial statistician (or others not directly involved in recruiting patients to the trial) commonly generates the randomisation sequence. Methods that allow a permanent record of the sequence created are important to validate its randomness later if required (whereas a coin toss can be replaced without record). A clinical trial report should clarify who generated the sequence, the method used, and how concealment was achieved and monitored. There should be some demonstration that randomisation was successful. This is usually achieved by providing a table in a report comparing the major baseline demographic and prognostic characteristics of the two treatment groups. 1: CONSORT checklist of items to include when reporting a trial1 Section and topic Item no. Descriptor Methods Randomisation Sequence generation 8 Method used to generate the random allocation sequence, including details of any restriction (eg, blocking, stratification) Allocation concealment 9 Method used to implement the random allocation sequence (eg, numbered containers or central telephone), clarifying whether the sequence was concealed until interventions were assigned Implementation 10 Who generated the allocation sequence, who enrolled participants, and who assigned participants to their groups 2: The permuted block method of randomisation for a block size of four, with A and B being treatment groups (A = intervention and B = control, for example) A random number sequence is generated from a statistical textbook or computer. Each possible permuted block is assigned a number (1 to 6 in the above example). Using each number in the random number sequence in turn selects the next block, determining the next four participant allocations. Numbers in the random number sequence greater than the number of permuted block combinations (7, 8, 9 and 0 in the above example) are not used to select blocks. 3: Example of randomisation using the minimisation method in a trial of chemotherapy for breast cancer, with stratification factors of clinic site, oestrogen receptor status (ER+ or ER–) and menopausal status Status after 34 participants have been randomised to the trial Characteristic Treatment A Treatment B Site 1 7 8 Site 2 10 9 ER+ 5 6 ER– 12 11 Premenopausal 8 9 Postmenopausal 9 8 Total 17 17 The next participant (no. 35) is from Site 2, ER+, postmenopausal. Subtotals for treatment allocation to this profile of characteristics are 10 + 5 + 9 = 24 for Treatment A and 9 + 6 + 8 = 23 for Treatment B (note subjects are counted more than once). Participant no. 35 would therefore be allocated to Treatment B. When the tallies on A and B are equal within a profile, the next participant is randomly allocated. This process is equivalent to a permuted block size of two within the profile. 4: Checklist for choosing a randomisation strategy How many subjects and clinical sites are planned? Are 24-hour randomisation services required? How will randomisation be implemented: central, remote local, bedside? Who will generate the sequence and by which method: random number lists, computer? Is a stratified or simple randomisation needed? If stratified, how many strata and levels within each stratum are required? What balancing strategy should be chosen: simple, permuted blocks, minimisation? What measures will be taken to guarantee allocation concealment? Who is going to monitor successful implementation (the balance of treatment allocation, unblinding rates) during recruitment? Should a screening log of eligible subjects be collected to ensure patients are not excluded by foreknowledge of treatment allocation? Queensland Clinical Trials Centre, University of Queensland, Herston, QLD. Elaine M Beller, MAppStat, Director of Biostatistics. NHMRC Clinical Trials Centre, University of Sydney, Camperdown, NSW. Val Gebski, MStat, Principal Research Fellow; Anthony C Keech, FRACP, MScEpid, Deputy Director. Correspondence: Associate Professor Anthony C Keech, NHMRC Clinical Trials Centre, University of Sydney, Locked Bag 77, Camperdown, NSW 1450. enquiryATctc.usyd.edu.au AntiSpam note: To avoid spam, authors' email addresses are written with AT in place of the usual symbol, and we have removed "mail to" links. Replace AT with the correct symbol to get a valid address.

Elaine M Beller MAppStat · Val Gebski MStat · Anthony C Keech FRACP, MScEpid

To the Editor: Thanks for the interesting opinion pieces by Reid1 and Paterson.2 Would I be correct in assuming that they are the Reid and Paterson who were formerly health bureaucrats in New South Wales and Victoria, respectively? May I suggest that your readers, especially those interstate and overseas, would have been better informed on the import of these articles if you had made some editorial mention of this fact? In this era of "career-hopping" between industry, government and academia, your readers, if they are to intelligently interpret an opinion piece, need to know more than merely the present position held by the author. You rightly ask about the "competing interests" of contributors of research articles. Perhaps the writers of opinion pieces should declare their background?

Peter Arnold

In reply: As always, I appreciate Arnold's input, and he is right yet again: M Reid was Director-General, New South Wales Health, 1995–2001, and J Paterson was Secretary, Health and Community Services, Victoria, 1992–1996. This information was conveyed in the author's details for Paterson, but inexplicably not for Reid. The Journal asks its contributors to declare "competing interests"; that is, disclosure of "any situation in which an individual . . . might be influenced . . . by financial or personal factors that involve self-interest".1 Most journals, including the MJA, choose to focus on competing financial interests, but an ongoing quandary is where to draw the line in the sand of competing interests — should they be religiosity, sexuality, consultancy within the political or health bureaucracy, or positions on committees or advisory boards, and so on? Arnold wishes to move to a higher plane through disclosure of relevant areas of contributors' life stories, presumably to alert readers to the potential for bias. But might not the publication of a contributor's relevant career prejudice the response of the reader? Kenneth Rothman, editor of the journal Epidemiology, has argued that objectivity in communication "depends on each contribution receiving its due regard, whatever the motivations for bringing it. It depends on judging a work on its merits, rather than on the inferred state of mind of the author".1 The contributions by Reid and Paterson were published under the Journal's For Debate banner. I was hoping for a debate on the messages rather than the messengers.

Martin B Van Der Weyden

Conference presentations are preliminary findings which should be interpreted with caution by the media, health professionals and the public When organisers began planning the XXIXth International Congress of Ophthalmology, held in Sydney earlier this year, an early consideration was how to promote media coverage of the conference, with the aim of raising public awareness of the specialty of ophthalmology and eye health more generally. A company which specialises in media relations for medical conferences was retained to work with the conference scientific program committee to develop a media strategy. As a result of the press releases issued, there were more than 520 news reports in print, broadcast and online media in Australia and overseas, including substantial stories in major media outlets. This is not an unusual scenario. Australian journalists are often approached to run stories arising out of conferences. In Europe and North America, where there is a larger market for such stories and a more established tradition of specialist medical and scientific reporting, media management of medical and scientific conferences is even bigger business. Such media management can have advantages for conference organisers, sponsors, participants, the media and the public: Conference organisers may wish to encourage media coverage as a way of promoting greater awareness of their profession or of particular health issues, and prefer to guide the media agenda so they are not left on the back foot, responding ad hoc to journalists' demands. Presenters may welcome media coverage to promote awareness of their work or professional interests. Corporate interests, conference funders and sponsors, and institutions such as universities, hospitals and research centres, often actively encourage such publicity. Indeed, both corporate and non-corporate interests have paid the expenses of Australian journalists to attend health and medical conferences, with the aim of promoting coverage. The media, driven by the community's thirst for health and medical news, finds conferences newsworthy on several grounds. Often they provide the first airing of research not yet formally published, and the timeliness of the presentation provides an additional "news hook". Also, conferences often provide a rare opportunity to make contact with and interview leading experts from around the world. Conference reports can contribute to public good, by alerting policy makers, researchers, health professionals and the general public to important new developments. However, there can be a downside. As recently noted, media coverage of new research often does not reflect an evidence-based assessment of its significance; for example, the media is less likely to report randomised trials, relative to observational studies.1 This illustrates the priorities which drive news gathering; one of the media's main criteria for story selection, particularly in health, is whether the story is likely to interest its audiences. Study methodology is far less likely to influence story selection. As well, most journalists and news managers have not been trained in understanding the relative merits of different types of scientific evidence. Other difficulties include the particular circumstances of conferences making it difficult for the media to scrutinise the validity of research findings or researchers' claims, and, as the results have usually not been published, difficulty in obtaining informed comments from other sources. The practical constraints of covering a conference and meeting deadlines may also encourage the media to rely on a single source. Journalists often have to rely on what presenters say about their findings without having access to the data or other information useful in assessing such comments. They may cover conferences without actually attending the presentation itself, and so not have the benefit of comments or criticism from an informed audience. Further, the conference presentation may vary from what journalists are told in an interview. The media may cover research findings in advance of their presentation — this adds to their newsworthiness in the media competition to be first. But it also amplifies all the problems mentioned above, as illustrated in a recent conference preview in The Bulletin.2 The article was based on several interviews and an abstract released in advance of the conference. By the time the study findings were actually presented, they had been revised because of further analysis. Even if conferences have a legitimate scientific objective, they may not follow a sound scientific review process. Just because a paper has been accepted for presentation does not mean it will have scientific merit, although it may still end up on the front pages of the newspapers. A recent report of media coverage of scientific meetings in the United States raised similar concerns, finding that many of the presentations covered by the media were not subsequently published in journals, raising questions about their scientific merit.3 The report concluded that press coverage of scientific meetings often did not make it clear that these were preliminary findings of uncertain validity. As a consequence, patients may experience undue hope or anxiety or may seek unproved, useless or even dangerous interventions. The authors urged conference organisers to be cautious in their promotions; researchers to emphasise the limitations of their work when being interviewed; and the media to emphasise the preliminary nature of conference presentations. Many in the media might counter that they are not in the business of health education or promotion. But if the media's role includes providing independent, critical and balanced news coverage, journalists and news managers should be careful to apply the same standards of scrutiny to conference presentations as to other sources of news. Finally, the media's audiences — whether the general public or health professionals — should approach news reports of conference proceedings, and indeed all sources of health and medical information, with a sensible caution.

Melissa Sweet MA

Research papers should have a short section on how the results should be communicated to the public By all accounts, many of the half million Australian women who regularly take combined oestrogen and progestin hormone replacement therapy (HRT) were alarmed by the news on Wednesday, 10 July 2002, reporting that a United States study had shown HRT to increase the risk of breast cancer by 26%, as well as causing more vascular disease. Subsequently, numerous media reports, based on press releases from organisations such as the US National Institutes of Health (NIH)1 and the Cancer Council of New South Wales,2 continued to highlight the apparently large increases in risks caused by combined HRT and called for restrictions on the use of this treatment. General practitioners and cancer help-lines were inundated by enquiries from frightened women and reports of mass withdrawals from therapy soon followed. The source of this concern was the early termination of the NIH-funded Women's Health Initiative (WHI) trial comparing combined HRT and placebo among healthy postmenopausal women. The study was stopped after five years by an independent Safety and Data Monitoring Committee when a predetermined safety boundary for the risk of invasive breast cancer was crossed at an interim analysis. The report of the trial, published in JAMA,3 suggested that women allocated to combined HRT experienced increased risks of invasive breast cancer, coronary heart disease, stroke and venous thromboembolism, and decreased risks of colorectal cancer and hip fracture (Box 1). It was argued that, when all these outcomes were summed in a "global index", the adverse effects outweighed the benefits. However, treatment effects on only two of the outcomes — fractures and venous thromboembolism — met conventional criteria for statistical significance when appropriate (and prespecified) account was taken of the multiplicity of statistical tests performed. Even without adjustment for the dozens of statistical tests, the 95% confidence intervals for each of the other outcomes reportedly affected by combined HRT (including the global index) were consistent with a broad range of possible effects, including little or no effect. For example, any effect on the relative risk of invasive breast cancer appeared to lie somewhere in the range from no effect to an increase of about a half to two-thirds, whereas any effect on the absolute risk of the same outcome appeared to lie somewhere in the range from no excess cases to about 17 extra cases per 10 000 women per year. This very large degree of uncertainty about the true size (and arguably the existence) of most of the treatment effects reported is not reflected in any of the press releases we have seen, including that from JAMA,4 all of which report apparently precise estimates of the excess risks. However, the controversy that followed publicity about the results of the trial did not reflect concerns about the strength of the evidence, but rather dissatisfaction with the way in which the study outcomes were described. While the original report published in JAMA provided estimates of both relative-risk and absolute-risk differences, press releases from most sources focused on the relative increases in risk — in particular, the 26% increase in invasive breast cancer. That this increase in relative risk reflected a difference in incidence of eight cases per 10 000 women per year was much less emphasised. It was suggested by the economics editor of the Sydney Morning Herald that the reported 26% increase was likely to have been misinterpreted by many women as meaning that combined HRT conferred a one-in-four chance of developing invasive breast cancer.5 Others argued that the use of relative risks in press releases was a deliberate effort to dramatise results that would appear much less newsworthy if described in absolute terms. The same commentators suggested that press releases should focus instead on absolute treatment effects, as these are of most direct relevance to the advice provided by doctors and the decisions made by women. Are these criticisms justified? Certainly, there is little doubt that the way in which risk data are presented influences treatment preferences.6-9 In a recent randomised trial in which general practitioners were asked whether they would prescribe a preventive treatment that had negligible side effects, 91% of those given information about relative risks alone said they would do so, compared with 63% of those given information about absolute risks.6 Other studies suggest that the way in which risk information is presented to consumers can generate even greater divergence in preferences.7 Should we therefore abandon the use of relative risks entirely in interpreting the results of clinical trials? Almost certainly not — although a strong case can be made for not allowing relative risks to dominate press releases without appropriate reference to absolute risks. Arguably, each has a place in communications to doctors and patients, and neither should be relied upon exclusively, as both have strengths and weaknesses. For example, while relative risks are usually generalisable to a variety of different patient subgroups (since the proportional effects of treatments are often broadly similar in most major patient subgroups), absolute risks are not (since absolute effects are determined in large part by background disease risks, which can vary substantially). Conversely, relative-risk estimates do not provide sufficient information for assessing the ratio of benefit to harm, as this can only be calculated from estimates of absolute treatment effects. Given the obvious complexity of identifying and delivering the most appropriate message to consumers (whether doctors or patients), medical journals might well consider taking a more substantive role in overseeing the broader dissemination of information about the results of major randomised trials. At a recent seminar ("The HRT debate: how should the new evidence affect policy?") conducted by the Australian Health Policy Institute at the University of Sydney, it was suggested that research papers should have a short section on how the results should be communicated to the public (Sally Crossing, Chair, Breast Cancer Action Group NSW, personal communication). Journals could assume more responsibility in two ways. Firstly, by ensuring compliance with a checklist of essential statistical components to be included in press releases issued by journals (Box 2); and secondly, by publishing a section within the main journal article that summarises the key messages for consumers, with reference to the same checklist. Such a checklist should include requirements for information about absolute as well as relative treatment effects, and for information about the full range of possible effects consistent with the observed result. If journals were to adopt this policy, it would be less likely that consumers would be misled, unintentionally or otherwise, by information released through the press. One can only speculate as to whether providing such information after the termination of the WHI would have altered the subsequent 30% fall in sales of the most commonly prescribed HRT preparations in Australia.10 1: Main results of the Women's Health Initiative trial of oestrogen plus progestin in healthy postmenopausal women3 Outcome Hazard ratio* Adjusted 95% CI† Unadjusted 95% CI Cardiovascular disease 1.22 1.00–1.49 1.09–1.36 Coronary heart disease 1.29 0.85–1.97 1.02–1.63 Stroke 1.41 0.86–2.31 1.07–1.85 Venous thromboembolism 2.11 1.26–3.55 1.58–2.82 Cancer 1.03 0.86–1.22 0.90–1.17 Invasive breast 1.26 0.83–1.92 1.00–1.59 Endometrial 0.83 0.29–2.32 0.47–1.47 Colorectal 0.63 0.32–1.24 0.43–0.92 Fractures 0.76 0.63–0.92 0.69–0.85 Hip 0.66 0.33–1.33 0.45–0.98 Vertebral 0.66 0.32–1.34 0.44–0.98 Deaths from other causes 0.92 0.62–1.35 0.74–1.14 Total deaths 0.98 0.70–1.37 0.82–1.18 Global index‡ 1.15 0.95–1.39 1.03–1.28 * Hazard ratios from Cox regression analyses of outcome among 8506 women randomly allocated to oestrogen plus progestin and 8102 women allocated to placebo. † Adjusted using group sequential methods to correct for multiple analyses over time. ‡ First event for each participant from among the following: coronary heart disease, stroke, pulmonary embolism, breast cancer, endometrial cancer, colorectal cancer, hip fracture, and death from other causes. 2: Essential statistical components for medical journal press releases describing the results of randomised clinical trials A. Provide estimates of absolute treatment effect in addition to estimates of relative treatment effect Estimates of relative treatment effect should not be provided without accompanying information about absolute treatment effect (or, at least, absolute disease rates). If the rates observed in the trial are substantively different from absolute disease rates in major patient subgroups, the limited generalisability of the observed absolute treatment effects should be acknowledged. For example, among perimenopausal women beginning hormone replacement therapy (HRT), whose average age is 10–15 years younger than those recruited to the Women's Health Initiative (WHI), any absolute increase in invasive breast cancer incidence is likely to be less than that observed in WHI, as breast cancer rates are strongly age related. B. Describe the full range of possible effects consistent with the observed result Avoid inappropriate focus on point estimates of either relative or absolute effect when confidence intervals indicate a broad range of potential effects. For example, the WHI result for invasive breast cancer risk was reported in press releases as a 26% increase in relative risk (and, occasionally, as an absolute excess of 8 cases per 10 000 women per year). However, the observed result is consistent with no increase in risk, as well as with an increase in relative risk of half to two-thirds and an increase in absolute risk of up to about 17 cases per 10 000 women per year (based on unadjusted 95% confidence intervals).

Anushka Patel MB BS, MS, FRACP · Robyn Norton PhD, MPH · Stephen MacMahon PhD, FACC, FAHA

At the end of May this year the Rheumatology Associations of Australia and New Zealand hosted a combined scientific conference in Christchurch, New Zealand. Notable among the presentations was a retrospective observational study on the association between gastrointestinal (GIT) bleeding and the use of aspirin, non-steroidal anti-inflammatory drugs (NSAIDs), or cyclo-oxygenase II (COX-II) inhibitors (Box 1). The essential finding of the study was that among 20 patients with bleeding related to peptic ulcer or oesophagitis, six were taking low-dose aspirin, four were taking COX-II inhibitors and one an NSAID. The researchers concluded "that aspirin is still more commonly associated with GIT bleeding than conventional NSAIDs and COX-II inhibitors". One week later, the headline "Doctors warn: just one tablet of aspirin a day may be enough to do you serious harm" appeared on page one of the respected morning broadsheet the Sydney Morning Herald. The accompanying story, reprinted in Box 2, related that a "world first study" from the Prince of Wales Hospital in Sydney "found that: where any drug was implicated in [GIT] haemorrhage, low-dose aspirin was consistently the likeliest culprit — ahead of anti-inflammatory drugs which are more often blamed for the condition. All 120 patients eventually recovered but many had to have transfusion, adrenaline shots or surgery." The chairman of rheumatology at the Prince of Wales Hospital was quoted as saying that "It is of some considerable concern that aspirin, even at 100 mg a day, is likely to be associated with gastrointestinal hemorrhage", adding, "When patients are admitted vomiting and passing blood, it's pretty dramatic". Such a blood, guts and fear story proved irresistible to the media. Before the day was out television news and radio commentators carried the story with such graphic gusto as "new research shows that even small doses of aspirin can cause potentially fatal stomach bleeding", or "a daily dose of aspirin to thin the blood to avoid heart disease or stroke may be deadly, causing ulcer bleeding". On the day, the story was carried by at least 20 media outlets, including most of the prime-time television evening news broadcasts in eastern Australia (information gathered by Media Monitors Australia). The process of transferring information from the researchers to the newspaper had transformed a conference poster of a retrospective observational study of only 20 patients into sensational medical news! This breaking of "a world first study" undoubtedly caused disquiet and despair in the community, as people with heart or cerebrovascular disease were faced with the decision as to whether they should continue taking aspirin to prevent a heart attack or stroke, or stop taking it to prevent a "fatal" stomach bleed. This is an impossible choice in the absence of vital information, such as the absolute risk of GIT bleeding and the benefit-to-harm ratio for low-dose aspirin. Their doctors were also impotent, as they were not privy to information on which the news story was based. Because of the implications of such a sensational story, the Journal sought to put the report into context, approaching both a cardiologist and the story's creators for comment (see the letters following this article). In response to this request, the medical researchers raised the philosophical question of when a scientific fact becomes established and accepted as fact, and whether this process requires the blessing of peer review. The journalists stressed their responsibility to report medical matters and pointed out that journalistic judgement takes precedence over scientific or evidential rigour. It should be no surprise that the research on which this medical news was based has yet to appear in the scientific literature. Does this matter? The timing of wide dissemination of medical research outcomes to other researchers, clinicians and the public remains a contentious issue.1-4 Journalists want to be the first to break important medical news, and their editors know that there is an insatiable public appetite for health matters, more so if dressed up as "miraculous cures" or "dire threats".5 Researchers also want to be first in reporting new findings, as this is critical for their research funding, professional advancement and peer approval. Most do this through respected peer-reviewed journals. Despite these similar aspirations, there exists an inherent tension between the two cultures: "Media constraints of time, brevity and simplicity preclude careful documentation, nuance positions and precautionary qualifications that scientists feel are necessary to present their work."2 And at the centre of this tension is what the participants perceive science to be. Frank Davidoff, the emeritus editor of the Annals of Internal Medicine, argues that science does not exist until it is published.6 All that takes place before publication is part of the doing of science: the critique of ideas, methods, data and hypotheses, either informally through conversations in cafes and corridors, or more formally from the podiums of clinical or scientific meetings. The science becomes established when it has stood up to the scrutiny of peer review and the quality filters of the editorial process and is "in print ".6 Being in print in the press is not science, it is story telling. Furthermore, press coverage before publication may well do the community a disservice, as the imprimatur of the press conveys a sense that the information is valued, accurate and widely accepted. But the softness of prepublication information is highlighted by the fact that almost half of the abstracts of scientific or clinical meetings are not published,7 and a quarter of meeting abstracts that receive prominent press coverage share the same fate.8 One feature of medical reporting is sensationalism.5,9-11 This phenomenon has been attributed to "miscommunication" arising from the different cultures and styles of science and journalism.10 However, at times both parties may benefit from sensational stories.11 The reporter has a front-page story, and the researchers (and their institutions) bask in the publicity. But, more importantly, sensationalism may occur because no one is responsible or accountable "for the results of transactions between reporters and researchers in the way that an editor is responsible for the peer review process and content of a scientific journal . . . In the wake of a sensationalised medical story, the involved parties can always point the finger at each other".11 This lack of accountability has prompted a call for a watchdog to spotlight both good and bad medical stories.11 Robotham and Whitehead stress that "a basic tenet of journalistic ethics is that journalists should be independent" (see the letters following this article). But with independence comes responsibility and accountability, and this applies on both sides of the fence: both to reporters and researchers and their respective institutions. In the absence of such accountability the public will continue to be bombarded by the touting of "miracle cures" — which, on closer inspection, are years away or apply only to rats; or the proclaiming of "dire threats" to health, which, on closer scrutiny, are tentative at best and are usually followed by a dissenting story and the inevitable backflip. Subjecting the public to a roller coaster ride of excitement and disappointment or anxiety and anger will eventually engender cynicism for medical research and its reporting, and this would be a shame. After all, the aim of both parties is to publish and report medical science for the good of all. 1: The impact of cyclo-oxygenase II (COX-II) inhibitors on gastrointestinal (GIT) bleeding Background: Our previous study demonstrated that aspirin is a more commonly associated risk factor for GIT bleeding than both conventional NSAIDs and COX-II inhibitors. This study reports the impact of COX-II inhibitors on GIT bleeding in a third time period (2001) following the listing of a second COX-II inhibitor (rofecoxib). Method: Medical records of patients admitted to Prince of Wales Hospital, Sydney, with GIT bleeding or those requiring a gastroscopy during the period of 1 March to 31 August 2001 were reviewed. Results: 31 out of 321 patients have met the criteria. Of these 20 had bleeding related to peptic ulcer disease or reflux oesophagitis. Of this group, four were on a COX-II inhibitor, six were taking low dose aspirin (100–150 mg/d) and one was on a conventional NSAID. One patient was on clopidogrel only and another on warfarin only. Two patients were on prednisolone. Only two patients had a previous history of peptic ulcer disease. Overall in both studies of the three time periods, 35% were on aspirin, 20% on NSAID's and 11% were on COX-II inhibitors. A very small percentage were on warfarin, clopidogrel or prednisolone, and the remainder were not on any medications known to be associated with increased risk of GIT bleeding. Conclusion: Our most recent data confirms our previous study that aspirin is still more commonly associated with GIT bleeding than conventional NSAID's and COX-II inhibitors. J Bertouch, L Lee, H P McNeill, T Bolin Prince of Wales Hospital, Australia 2: Doctors warn: just one tablet of aspirin a day may be enough to do you serious harm Even very low doses of aspirin designed to prevent stroke and heart attack can cause life-threatening stomach bleeding, says a Sydney study that challenges the widespread use of the blood-thinning therapy. The world-first study looked at all patients admitted to Prince of Wales Hospital with gastrointestinal bleeding during three six-month periods between 1999 and 2001. It found that where any drug was implicated, low-dose aspirin was consistently the likeliest culprit — ahead of anti-inflammatory drugs, which are more often blamed for the condition. All 120 patients eventually recovered, but many had to have transfusions, adrenaline shots or surgery to stem the bleeding. The hospital's chairman of rheumatology, Jim Bertouch, said the results indicated the preventive aspirin doses most commonly used in Australia — a single daily tablet of 100–150 milligrams — may be too high. He suggested that doses of 75 mg, already used in Britain, might be more appropriate. Low-dose aspirin was implicated in up to a third of the gut hemorrhages diagnosed during the study, Dr Bertouch said. By contrast, arthritis medications, previously maligned as likely to cause stomach bleeding, caused fewer bleeds, about a quarter of the total. But the new COX-2 inhibitor arthritis drugs, which include the popular Celebrex — heavily promoted as safer for the stomach than older anti-inflammatories — were still involved in one in eight hemorrhages. The rest of the bleeds were either not linked with any drugs, or with a range of other drugs. "It is of some considerable concern that aspirin, even at 100 mg a day, is likely to be associated with gastrointestinal hemorrhage", said Dr Bertouch, who presented the research at a rheumatology conference in New Zealand last month. "When patients are admitted vomiting and passing blood, it's pretty dramatic". Terry Bolin, chairman of the Prince of Wales Hospital's Gastrointestinal and Liver Unit and a co-author of the study, said "We've known for years that aspirin was a risk but we were unaware that the really low dose aspirin had the same risk." Doctors should check whether patients were carriers of H. pylori bacteria that precede stomach ulcers, because this further increased bleeding risk from aspirin and other drugs, Associate Professor Bolin said. Associate Professor Con Aroney, a spokesman for the National Heart Foundation, said the benefits of aspirin outweighed the risks for people with a history of coronary heart disease or stroke. But daily aspirin was not recommended to prevent heart disease in healthy people. Julie Robotham Medical Writer, Sydney Morning Herald (June 7, 2002: 1)

Martin B Van Der Weyden MD, FRACP, FRCPA

To the Editor: The front-page article in the Sydney Morning Herald on 7 June this year1 highlights the problem of premature media coverage of a scientific presentation,2 potentially causing distress and confusion. Without being subjected to full peer-review and unavailable for analysis in its full published form, such data should not be presented to the public as scientific fact, and should not be sensationalised so as to encourage patients and doctors to change management. A small single-centre observational study is regarded as Level 4 evidence and cannot be used to recommend a change in management. At most, such data might be considered hypothesis-generating and used as the basis for a properly conducted clinical trial. In a meta-analysis of 70 000 "high risk" patients, antiplatelet therapy, mainly with aspirin, reduced rates of stroke, myocardial infarction and vascular death by 25%.3 Aspirin also reduced by almost half the rate of graft occlusion after coronary bypass surgery.4 The press article has confused such patients and may lead to their discontinuing life-saving therapy. It cites Bertouch as stating that 75 mg of aspirin "might be more appropriate". There are no data, either from the Prince of Wales study or any other, to support the contention that 75 mg of aspirin causes less bleeding than 100 mg or 150 mg. The press release describes the research as a "world-first study", and Dr Bolin is cited as stating that "we were unaware that really low-dose aspirin had the same risk". However, as early as 1991, the Swedish Aspirin Low-Dose Trial showed that even 75 mg of aspirin produced more bleeding than placebo (P = 0.04).5 As a result of the Sydney Morning Herald article, patients are asking their doctors to make a judgement on ceasing their aspirin therapy, which might prove fatal, or reducing the dose from 100 or 150 mg to 75 mg, which is not supported by evidence and is not even a dose available in Australia. At a time when it is difficult enough to convince patients to take medication which is of proven benefit, both the press and the research community have a responsibility to the public to avoid recommendations which are not evidence-based and which detract from our efforts to reduce the mortality from Australia's biggest killer — cardiovascular disease.

Constantine N Aroney MD, FRACP

In reply: Aroney's letter raises a number of important issues. The first of these is the question of whether a scientific fact requires the blessing of peer review to become established as such. The corollary of this is whether or not all peer-reviewed facts are necessarily true. The answer to both questions is probably no. The second issue is how to control a media report, irrespective of whether it is based on a peer-reviewed study. The issue which concerns Aroney is an abstract presentation of the association of gastrointestinal bleeding with aspirin, non-steroidal anti-inflammatory drugs (NSAIDs) and cyclo-oxygenase II (COX-II) inhibitors, the conclusion of which was that, while the last two might be important in their own right, concurrent use of aspirin, even in a small dose, was more closely associated with bleeding risk, particularly if there was a past history of peptic ulceration.1 A "meta-analysis" of the media reports, which included both television and radio in addition to the quoted report in the Sydney Morning Herald,2 would have made it clear that the theme of the interviews reaffirmed the relative safety of aspirin in the vast majority of individuals, and highlighted the risk of aspirin use concurrently with NSAIDs and COX-II inhibitors, particularly when there is a history of past ulceration. The fact that the SMH report focused on one aspect of the study was counterbalanced by the others. We do not know how journalistic reporting is controlled. A primary question is whether or not aspirin is an effective agent for the prevention of cardiovascular disease beyond the management of acute myocardial infarction. More recent literature than that quoted by Aroney is now questioning the overall cardioprotective value of aspirin.3 This showed that aspirin given as prophylaxis against cardiovascular disease increased the risk of sudden death in every secondary prevention study and left the overall rate of myocardial infarction unchanged.4 Aspirin consistently failed to reduce overall mortality in every study of long-term prophylaxis after myocardial infarction, and in all but one after stroke.3 Furthermore, Cleland and colleagues have argued that a series of meta-analyses, which most people have accepted as proof of the efficacy of aspirin, are of doubtful validity.4 They questioned whether it is appropriate for the medical community to invest so much time and effort in prescribing aspirin and dealing with the adverse consequences of its long-term ingestion to the neglect of other, better proven and apparently more effective therapies such as angiotensin-converting enzyme inhibitors, β-blockers, and statins. At the very least, it can be said that there is controversy in the cardiovascular literature about the benefits of aspirin. Just as important is the issue of the safety of long-term aspirin use for cardioprotection. A recent multidisciplinary expert statement on NSAIDs concluded that, on current evidence, prophylactic use of aspirin should be reserved for patients with established vascular disease, because in other patients bleeding risks may outweigh cardiovascular benefit.5 A Danish study showed that 100–150 mg of aspirin daily increased the risk of haematemesis by a factor of 2.6, with no difference in the risk between enteric and non-coated product; when combined with an NSAID the risk was increased by a factor of 5.6.6 The authors concluded that the bleeding risk may offset some of the benefits of aspirin. It is no longer appropriate to simply "bury" the adverse gastrointestinal effects of low-dose aspirin in the NSAID side-effect "basket".7 It is apparent to us that dogma should not be so enshrined that it prevents the discussion of issues that might helpfully modify that dogma.

Terry D Bolin · James V Bertouch MD

In reply: The letter from Aroney displays some basic misunderstandings of the role of the media in reporting medical issues. Aroney states that "the press [has] a responsibility . . . to avoid recommendations which are not evidence-based and which detract from our efforts to reduce mortality from . . . cardiovascular disease." A press article does not itself make recommendations when it reports the recommendations of others — an essential distinction. In addition, the press has no responsibility to follow the agenda of the medical profession and its slavish insistence on the dogma of evidence-based medicine. The press owes doctors no more favours than it owes any other sector of the community. The role of the press is to raise and debate issues of public interest in a manner that is balanced and responsible. The news report about aspirin did all of this.1 We agree that publication in a peer-reviewed journal may add scientific credibility to research findings and that this may sometimes make them more newsworthy. But our responsibility is to report medical matters of interest to the community, which means we are not limited to peer-reviewed findings. Any substantial fact, observation or opinion relating to medical practice is fair game for a newspaper's attention. After the findings of Bertouch and colleagues were presented at a conference,2 they entered the public domain, as did their later comments made to us directly. It was entirely proper to report them. The fact that the gastrointestinal bleeding study was conducted by two heads of department at a major Sydney teaching hospital was instrumental to our decision to position and headline the report prominently. These individuals are respected experts in their fields, and they expressed to us serious concern about the degree to which aspirin was implicated in gut haemorrhage. It was that concern which led us to focus on the aspirin findings within the broader study. The news process is always selective and there is no obligation to give equal emphasis to all findings. The magnitude of follow-up by other media confirms the inherent public interest in the topic. It has previously been suggested that medical journalists are behaving irresponsibly when they step outside the strictures of peer review.3,4 Yet it is a basic tenet of journalistic ethics that journalists should be independent.5 Why, then, would we subscribe to the doctrine of evidence-based medicine, with all its flaws? We, and the community, have every right to be sceptical of the tyranny of peer review when the pharmaceutical industry manifestly uses financial muscle to influence what is studied and what is published.6 Even the Medical Journal of Australia accepts anecdotal findings when it can persuade itself the public interest is involved, recently publishing an eyewitness account of conditions inside an immigration detention centre7 and defending this on the basis that "our readership is sophisticated enough to interpret the content of such articles."8 Sydney Morning Herald readers are also sophisticated. On what basis should they have been denied this pertinent information about a widely used medicine — that doctor still knows best?

Julie Robotham BA(Oxon) · Robert Whitehead

To the Editor: We write to express our concern at the publication of the recent supplement "Essential role of fats throughout the lifecycle", adorned by the sponsor's logo.1 It was an interesting counterpoint to an accompanying article in the main journal regarding the need for industry–academia collaborations to "strike a balance".2 While the issue of relationships between industry and doctors is complex, and few of us are truly independent, the publication of such a branded document is disquieting. Directed sponsorship, beyond mere advertising, undermines the credibility of such supplements and the Journal itself, regardless of the authors' expertise, objectivity and the importance of the topic. Unfortunately, we are left with the taste that this is a spread designed to butter us up.

Alex A Padiglione · Catherine E Marshall · Tony M Korman

In reply: "Oh what a feeling" to receive a congratulatory letter! But the euphoria was short lived, as, on closer inspection, congratulations turned to castigation. The offending event was the Journal's publication of an industry-supported supplement,1 and its practice of branding supplements with the logos of their sponsoring bodies. Although sponsorship by government agencies or non-profit health organisations rarely provokes comment, industry sponsorship is another matter. As industry support for research and other health-related activities will inevitably increase in the future, we at the Journal are pleased that Padiglione and colleagues have aired their anxiety. Irrespective of the source of sponsorship, the Journal's policy governing the publication of supplements follows the recommendations of the International Committee of Medical Journal Editors.2 These include that: the journal's editor must take full responsibility for policies, practices and content of supplements, must approve the appointment of the editors of supplements, and must retain the authority to reject articles; and the source of funding should be clearly stated and prominently located in supplements, preferably on each page. To these principles the Journal has added its own requirements.3 These include the need for peer review and that editors of and contributors to supplements declare competing interests and compensations. These were clearly identified on the title page of the offending publication.1 For our readers, the Journal is the bread and its supplements the butter. One can always refuse to taste the butter. But, for those who hanker after a little fat, we aim to ensure, through churning by external peer review and transparent sponsorship of the product, that "butter is better".

Martin B Van Der Weyden

In this issue of the Journal, Loke and colleagues (page 291) present data from an analysis of 174 advertisements for pharmaceuticals appearing in six Australian medical publications.1 The findings are striking enough to be restated. Fewer than 8% of the advertisements contained quantitative data about the outcomes of therapy, and most of these framed the information in relative rather than absolute terms. Only 28% of the therapeutic claims in the advertisements conveyed clinical outcomes in any specific, substantive and unambiguous way. In the United States, pharmaceutical advertising is subject to the Federal Food, Drug, and Cosmetic Act,2 and Loke et al suggest that, in Australia, advertisements for drugs may be less informative than in the US. The pharmaceutical industry has long maintained that drug advertisements are an important vehicle for conveying important information about new drugs to prescribers. Is this how industry believes it should communicate with highly trained healthcare professionals? Should we really be surprised by the results of Loke et al, and, more importantly, should we be concerned? We know that the pharmaceutical industry spends enormous sums on promoting its products (about twice the amount spent on research and development),3 but most data on the effect of advertising on prescribing are unpublished, and have been gathered by advertising companies. The Association of Medical Publishers (AMP), a US-based organisation whose membership includes the publishers of nearly 200 biomedical journals, boasts "advertising in medical publications alone... can generate sales for both new and more-established products" [original emphasis].4 AMP reports a number of studies that have shown a significant increase in market share and retail sales as a result of medical journal advertising, which is reported to provide a return on investment (ROI) of about US$5.00 for every dollar spent, greater than detailing (ROI US$1.72) and direct-to-consumer advertising (ROI US$0.19).5 Most advertisements are for new and expensive drugs, so increased use due to promotion will contribute to the financial pressures on the Pharmaceutical Benefits Scheme (PBS). Does journal advertising also lead to inappropriate practices? Although there is a substantial body of research on the effects of pharmaceutical industry promotion generally, relatively little involves printed advertisements in medical journals. In a landmark study, Avorn and colleagues studied physicians' beliefs about the efficacy of two classes of drugs (propoxyphene analgesics and central/peripheral vasodilators) that were being heavily promoted as effective, despite evidence that they lacked any efficacy and offered no advantages over existing treatments.6 The authors found that, even though doctors reported paying little attention to drug advertisements, most doctors believed that these agents were effective. Do the results reported by Loke and colleagues have other implications? What is their relevance for the development of government policy? The Australian Competition and Consumer Commission (ACCC) is currently examining an application for reauthorisation of the Code of Conduct of the Australian Pharmaceutical Manufacturers' Association (now Medicines Australia). As part of the examination of the relationships between pharmaceutical industry participants, the ACCC is investigating claims in the media about some practices and whether they are in the best interests of the community (Lin Enright, Director, Public Relations, ACCC, personal communication). The ACCC should heed the results reported here. It is only two years since the review of direct-to-consumer advertising of pharmaceutical products in Australia.7 Although the review recommended against direct-to-consumer advertising, the subject is under continuing review, and some within the pharmaceutical industry are still pressing for change, maintaining that such a facility would enable them to provide important educational information about drugs to the public. Similar moves to relax laws relating to direct-to-consumer advertising are also occurring in Europe and Canada.8 The information reported by Loke et al on journal advertisements suggests that direct-to-consumer advertising is likely to be uninformative and promotional rather than educational in nature. Where should we look for guidance on appropriate standards for advertising pharmaceutical products? Medicines Australia polices a voluntary code of conduct that aims to set "standards of conduct for the activities of companies when engaged in the marketing of prescription products".9 This document places more emphasis on what not to do when promoting medicines, rather than offering guidance on how to provide balanced advice to clinicians about the efficacy and safety of medicines. Perhaps we should pay more attention to the advertising standards maintained in other industries. Generally, advertisements for technologically sophisticated products include prominent displays of their specifications, performance and selling price. Is it too much to ask that advertisements for modern drugs provide similar information? In an era of evidence-based medicine this should include data on the absolute effects of therapy, such as the response rates with and without treatment, and the number needed to treat, in order to avoid the ambiguities of relative measures such as the relative risk reduction. It would be best if this information related to comparisons with established therapies, not just placebo. Clinicians should also be told the dispensed price of the drug under the Pharmaceutical Benefits Scheme.

David A Newby BPharm, PhD · David A Henry MRCP, FRCP

Objective: To determine the quality of claims in advertisements published in Australian medical publications, describe how benefits and harms are presented, and examine the level of underpinning evidence.Design and setting: Audit of a consecutive three-month sample of advertisements appearing in six popular Australian medical publications.Main outcome measures: Proportion of advertisements with quantitative information; proportion of claims conveying clinical outcomes; where retrievable, level of underpinning evidence.Results: Of 1504 claims, 855 could be substantiated quantitatively. Of these, 45% were supported by compelling evidence (randomised controlled trials or better). Of 13 claims explicitly reporting quantitative outcomes, none provided the absolute risk reduction or the number needed to treat.Conclusions: Our audit invites greater diligence by pharmaceutical companies in substantiating their claims and greater vigilance among clinicians when reading them.

Tim W Loke MB BS(Hons) · Fong Chee Koh MB BS(Hons) · Jeanette E Ward PhD, FAFPHM

To the Editor: Your editorial comment1 about the fact that the "Death and Dying" issue of the Journal [19 November 2001] contained no obituaries because "the obituary tray was empty" prompts me to make a suggestion. Having, over the years, written obituaries for colleagues, including one for my own father, I have always believed we would do colleagues a great favour if each of us wrote our own obituary and left it in the keeping of a colleague, to be dealt with at the appropriate time. Not only would it ensure that the obituary included services for which the departed would like to be remembered, but it would also save the writer of the obituary a great deal of time in research. PS. Now in my own 91st year, I've not yet written my own obituary.

H Stuart Patterson

Thorough definitions and descriptions of the objectives and outcomes in clinical trials lead to results that are more readily interpretable and more easily generalisable. Indeed, the failure to prospectively define both objectives and outcomes in sufficient detail, and to describe how these are to be measured, has been a major criticism of some randomised trials.1 Items 5 and 6 of the CONSORT statement checklist relate to objectives and outcomes in randomised trials.2 ObjectivesThese should be clearly defined, and stated in a manner that will allow the objectives to be investigated by a quantitative assessment of appropriate outcomes. For example, an objective stating ". . . whether allowing free access to heroin will decrease the crime rate . . ." is too vague without a detailed definition of what constitutes a crime and how and when these are to be measured. Loosely stated objectives may appear in studies with a wide range of scientific rigour, and are likely to give rise to scepticism about the trial results, owing to concerns that the definitions may have been created post hoc, with foreknowledge of the study data. Concerns might also arise about studies whose objectives require myriad tests or assessments (eg, psychological, clinical and psychiatric), potentially confusing the reader as to which, if any, evaluated the primary hypothesis. The supporting evidence, outlined in the background and rationale of the study, should be linked logically to the study objectives.3 The outcomes of studies are more convincing when they apply to a single or small number of clearly defined objectives. The objectives should include: a precise statement of the degree of benefit expected from the intervention, as well as the duration of the benefit; clear statements of the time frame of the study (especially in relation to how quickly benefits might commence); and a definition of the patients for whom the benefit is sought.4 Objectives can be classified as either primary or secondary. Primary objectives provide the focus of the study. Collection and measurement of outcomes affecting the primary objective is critical and, if resources are scarce, these take priority over other aspects of the study. An exception to this is that the collection of safety information is always considered of high priority, regardless of whether safety is the focus of the study. It is crucial to minimise missing data related to the primary objective. Secondary objectives allow for investigating subsidiary questions that, while scientifically important, do not have the same priority of clinical interest in the patient group being studied. In most randomised trials, efficacy of the intervention or its equivalence with standard care is the primary objective, whereas safety (eg, toxicity, side effects) is usually a secondary objective. OutcomesAs with the objectives, the outcomes of the trial require precise description and definition. Standard measurement criteria are essential in order for the results to be accepted by the clinical community. For example, in cancer studies, measuring response by tumour area is a widely accepted practice, whereas measuring response by tumour volumes may be questioned by those not routinely using this criterion. The outcomes chosen should be clinically relevant and, where possible, measured in an objective fashion. If objectivity in measurement of outcomes is not possible, some control, on a subjective assessment, is desirable. For example, blinding assessors to treatment allocation provides a powerful tool for reducing measurement bias.5 The frequency of outcome measurement should be clearly stated, as should strategies to be undertaken if the pooled outcome rate is lower than anticipated (eg, adjustment of study sample size). As with objectives, outcomes can be classified as primary or secondary and the same comments relating to primary objectives also apply to the measurement of primary outcomes. If multiple outcomes (eg, liver function test results, or scores on a battery of psychological scales) are measured, precise statements on which aspect of these will be used to investigate the objectives need to be made a priori. The analysis of multiple outcomes requires specialised statistical methods and these should defined in detail in the study protocol and report. Where it is essential to employ multiple outcomes, a priority list detailing which of the outcomes represents the benefit sought should be determined in advance. Thus, for example, if a series of haematological parameters is being measured, then a single parameter (eg, changes in the platelet count) should be defined as the primary outcome. If an outcome is measured repeatedly (eg, muscle strength), the issue of how these repeated measurements will be used to meet the objectives needs to be clearly stated. Thus, an objective considering "the frequency of heroin use in the 28 days before a six-month assessment (ie, during month 5)" will require different outcomes measurement to an objective considering "the average frequency of heroin use over a six-month period". The latter objective would require specification of how often the frequency of heroin use was to be estimated or recorded over the six months. Missing measurements may also become an issue (such as measurement of heroin use in a self-reporting study) and requires careful thought in the design stage. It may be desirable to specify a secondary, "fallback" objective in case data for the primary outcome prove difficult to collect over the study period (a well conducted pilot study would avoid this risk). Surrogate outcomesIn many instances the use of surrogate, or intermediate, outcomes allows for shorter study durations (because surrogate events accumulate faster), with results on the surrogate outcome being translated to an outcome of major clinical interest. A surrogate outcome is one that is intended to capture the treatment effect on an important clinical endpoint, but does not directly measure the main clinical benefit of the intervention. For example, in cancer studies, tumour response, disease-free survival and time to disease progression have been used as surrogates for survival. The conditions under which an outcome is a "good" surrogate are still the subject of research.6 Nevertheless, many surrogate outcomes (eg, elevated tumour marker levels as indicators of tumour activity, such as prostate specific antigen in prostate cancer) provide strong associations with tumour growth as an important question of interest. Statistical methods exist to examine the degree to which a surrogate is associated with a main outcome,7 although these are still being refined. A statement indicating how and how much of any benefit observed in the surrogate will translate to the main clinical outcome should be provided. In clinical studies, where it is not feasible to have adequate statistical power for a clinical endpoint, a valid surrogate may be used as the primary objective, with the main clinical endpoint becoming a secondary objective. 1: CONSORT checklist of items to include when reporting a trial Selection and topic Item no. Description Objectives 5 Specific objectives and hypotheses Outcomes 6 Clearly defined primary and secondary outcome measures and, when applicable, any methods used to enhance the quality of measurements (eg, multiple observations, training of assessors) 2: Checklist for objectives and outcomes in clinical trials Objectives Are the intervention and control (eg, usual care) described in detail? Has the the target patient population been specified? Has the degree of benefit from the intervention on a particular outcome, and the time frame, been specified? Has the primary outcome, including how and when it is to be measured, been specified? Have any secondary outcomes been pre-specified in similar detail? Outcomes Are the outcomes clinically relevant, objective (wherever feasible) and unambiguous? Can the outcomes be measured for all patients and, where possible, assessed with researchers blinded to the allocated treatment? Is the study explicit in the frequency and duration of outcome measurement? Has the study been specially planned from a statistical viewpoint when multiple outcomes are measured? If the outcome is a surrogate, will it adequately reflect a main outcome, and is there an indication of how much a benefit observed on the surrogate outcome will translate to a benefit on a main outcome?

Val J Gebski BA, MStat · Ian Marschner BSc, PhD · Anthony C Keech FRACP, MSc(Epi)

To the Editor: A whole issue of the Journal devoted to Death and Dying [19 November 2001 issue] and not a single obituary!

Paul B Harris MB BS FRACR

In reply: I am reminded of a line from the American poet Edwin Arlington Robinson: "I shall have more to say when I am dead." Nowadays, the opportunity to be the voice commemorating the life of a colleague through an obituary in the Journal seems to be of low priority for members of Australia's medical profession. The decade 1991–2000 saw the lowest number of obituaries published by the Journal in the five decades since 1951.1 We did not include an obituary in the "Death and Dying" issue because, fortunately or unfortunately, the obituary tray was empty.

Martin B Van Der Weyden

To the Editor: The article by Sultan and O'Sullivan on detention of asylum seekers in Australia in the 3/17 December issue of the Journal1 contains several errors of fact and distortions that I believe you have an obligation to address. While I am not addressing medical issues, this nevertheless goes to the issue of credibility. The article made several claims that are factually wrong and would not have withstood even cursory examination had you sought verification. I will not attempt to deal with all distortions and factual errors, but your readers should be made aware of at least some elements. Detention is not arbitrary. It is humane and is not designed to be punitive. Staff are provided with cultural-awareness training and are expected to interact with detainees in an appropriate manner. Further, the level and range of medical services available to detainees exceed those available in many regional areas of Australia. It is true some people have been detained for some years. However, Sultan and O'Sullivan neglected to mention that these cases are ones where people have been found to have no lawful right to remain in Australia, but have been pursuing all legal avenues or have been refusing to cooperate with departure arrangements. Another distortion and half-truth relates to a hunger strike where "power and water supplies were cut to the cell block . . . affecting uninvolved women and children". Firstly, there are no cells at Villawood and the alleged cell block was actually a recreation room. The water supplies were cut when earth-moving equipment outside the centre accidentally cut the mains supply to the area, affecting surrounding homes and the centre. Water containers were provided for the detainees until the water supplies were restored. What Sultan and O'Sullivan do not tell you about the power supply was that it was cut when detainees ripped wires out of electrical equipment and threatened to use them to electrocute staff. This action obviously also posed a risk to the children who were moving in and out of the room. A small number of protesting detainees were denied access to the visitors' area because of the disruption their actions would cause to the majority of detainees who were not involved in the protest. Head counts have taken place at 2 am, but what you were not told was that these followed escapes and are not routine. The claim that there has been a dearth of educational and resource material is a factual error. It is disappointing that a publication of your standing did not make even cursory enquiries on simple factual issues.

Aamer Sultan MB ChB · Kevin O'Sullivan BSc, PhL, DipClinPsychol · Debra Graves MB BS, MHA, FRACMA

In reply: The Minister for Immigration and Multicultural and Indigenous Affairs, Mr Philip Ruddock, claims that our article contains several errors of fact and distortions. He does not comment on the substantive concerns of the article, but implies that the errors undermine the credibility of our finding — that prolonged detention of asylum seekers appears to cause serious psychological harm. In response, we will address the alleged errors of fact and then the credibility of the observations made in our article. The Minister states that we failed to identify the factors leading to long periods of detention for some asylum seekers. It is clearly stated in the article that "lengthier detention is particularly common for detainees who appeal against adverse decisions about their refugee status". The Minister's response adds nothing of substance to this and fails to identify the other major factor accounting for prolonged detention, namely the difficulties in removing individuals who are stateless or from countries such as Iraq and Afghanistan. The Minister states that nightly head counts at Villawood occur only after escapes. Unfortunately, the Minister has not been accurately briefed on this matter. Nightly head counts have been a regular practice at Villawood. One of us (A S) has documented proof from the minutes of both the Community Reference Committee and the Centre Manager's detainee meetings held at Villawood, which record the concerns of the detainees and the refusal of management to discontinue the practice. Moreover, this practice was openly acknowledged by the operators of Villawood to the Parliamentary Joint Standing Committee which visited the centre in February 2001.1 It is pleasing to note that, after the publication of our article, nightly head counts have been discontinued in the Stage Two section of Villawood. There are conflicting accounts between detainees and management regarding the events surrounding the hunger strike referred to by the Minister. One of us (A S) sent a complaint about this incident to the Commonwealth Ombudsman. In his reply, the Ombudsman quoted a letter from the Department of Immigration and Multicultural Affairs dated 27 October 2000, which states: "I can confirm that power to the recreation room was cut-off for a short period of time to ensure the safety of the detainees participating in the protest action who had been threatening self harm". There is no mention of attempts by detainees to electrocute staff, as suggested by the Minister, and one of us (A S) has spoken to some of the detainees involved in the protest, who have categorically denied this. Many detainees remain highly sceptical of the official explanation offered regarding the lack of water to the block where the hunger strikers were being held. The very existence of the mass hunger strike and the conflicting accounts of the incident all attest to the emotionally charged environment that has at times existed at Villawood, which was the major point of discussing this incident in our article. We made it quite clear that there have been recent improvements in recreational resources at Villawood. However, as stated in our article, there had been a long history of neglect of this issue. This was confirmed by the statement of the Human Rights and Equal Opportunity Commission in 1998 that "the recreation facilities at Villawood are inadequate for the number of detainees being held there".2 Moreover, despite recent improvements, our statement that detainees face long periods of unstructured time remains true. Although we did not discuss the issue of whether mandatory detention was arbitrary, it is worth noting in response to the Minister's claim that the Human Rights and Equal Opportunity Commission concluded that "In some instances, individuals . . . have been held for more than five years. This is arbitrary detention and cannot be justified on any grounds".2 The failure of the Minister to consider the substantive issues we raised, namely the negative psychological effects that long-term detention appears to be having on asylum seekers, was disappointing. As we stated, we attempted to report our observations in what we consider to be an objective and truthful manner. Nothing in the Minister's response suggests that we have failed to do this. Even if the findings in our report were to be entirely dismissed, Steel and Silove (in the same issue of the Journal)1 document compelling evidence of the serious psychological symptoms observed among long-term detainees. In conclusion, we can only offer a continued exhortation for the Minister to attend to the collective weight of evidence indicating a mental health crisis within Australian detention centres.

Philip Ruddock MP · Derrick M Silove · Zachary Steel · George Halasz MRCPsych, FRANZCP · Michael Block · Leon Petchkovsky · Howard Cooper · Martin B Van Der Weyden · Ruth M Armstrong · Helen M Randall

In reply: Several aspects of Mr Ruddock's letter concern us. The tendency to evade the substance of the message and instead to attack the messenger, in this case the MJA, is unwarranted. Fortunately, the Journal's reputation rests secure with its more usual, considered readership, but the accusation that the MJA has acted in a cursory manner is particularly jarring — the media response by the Minister to the publication of the relevant articles occurred within hours, hardly sufficient time to subject them to a considered analysis. The haste of the response is also evident in the substance of the letter, with its focus on largely irrelevant technical aspects rather than on the key issue — the legitimate concerns raised about the impact of detention on the mental health of inmates. Are we to assume from the Minister's silence on this core matter that he acknowledges that the majority of long-term detainees are suffering from severe depression and other disabling emotional disturbances, as described in the article by Sultan and O'Sullivan and supported by other data we cite? The rates of depression reported by Sultan and O'Sullivan represent a roughly eightfold greater prevalence than is found in the general Australian population — if correct, this reflects an epidemic of mental illness among long-term detainees. Until proven otherwise, there is every reason to assume that the mental disorders identified are a direct outgrowth of the conditions of detention. One of the key findings of research in this area is that asylum seekers in detention report similar patterns of abuse and trauma as their refugee counterparts authorised to resettle in Australia. It is paradoxical and contradictory that we provide authorised refugees some of the best rehabilitative services in the world through our national network of services for survivors of torture and trauma, while at the same time creating conditions in detention centres that exacerbate the effects of past trauma in their compatriots. The claim that detention is humane is extraordinary. From a medical perspective, the obvious test of such a claim is in the health outcomes. Instead of the steady improvement in psychosocial status witnessed in authorised refugee populations after resettlement, there appears to be a progressive deterioration in the mental health of similarly traumatised persons in detention. A policy that in its implementation directly undermines the capacity of displaced persons to recover from the effects of past stresses cannot, by any stretch of the imagination, be humane. If detention is not punishment, then what is it? We are told repeatedly that detention is a deterrent aimed at discouraging the arrival of asylum seekers. How then does detention act as a deterrent if not by punishing the few to deter the many? Interestingly, punishment, according to many criminologists, is not a good deterrent. This makes the situation even worse — punishing those whose only "crime" is to seek asylum from persecution for a presumed greater benefit that remains unattainable. The practical failure of the "punishment as deterrence" approach is evident in the recent, desperate policy shift to the costly and unsustainable "Pacific solution". What next? In short, detention is not humane — prolonged detention, in particular, exacts severe costs by undermining the mental health of detainees. The stark reality is that political policy is creating a preventable public health tragedy. If doubts remain about this conclusion, then we should turn to science as the final arbiter. It is now pressing that an independent group of researchers undertake a comprehensive survey of detention centres in Australia and the Pacific to establish once and for all the impact of current policies on the mental health of the detainees.

Philip Ruddock

To the Editor: We wish to acknowledge Dr Aamer Sultan's courage in carrying out his research as a "participant–observer" into the impact of long-term detention on psychological health of asylum seekers1 at the Villawood Detention Centre, Sydney. In response to the article, Dr Louise Newman, Fellow of the Royal Australian and New Zealand College of Psychiatrists, called for an urgent assessment of the mental and physical health of asylum seekers in detention. As well, The Australian2 reported that Dr Newman advised members of the College not to accept appointments in detention centres because of serious concerns about how they were run by Australasian Correctional Management. Dr Sultan inspires the medical profession, at considerable risk to himself, in order to fulfil the obligation of the first aphorism of Hippocrates: "The physician must not only be prepared to do what is right himself, but also to make the patient, the attendants, and externals cooperate."

Philip Ruddock

To the Editor: I am writing as a member of the Australian Medical Association, and hence a subscriber to the MJA, to express my concern over the recent publication of the participant–observer account of psychological disturbances in asylum seekers,1 and the subsequent media reporting of the issue as a consequence of the publication. I would like to say at the outset my views are personal. Like most healthcare professionals, I consider access to basic medical care, including care for those with mental illness, is a human right. However, I do not support the use of a peer-reviewed journal such as the MJA for political purposes. The right of freedom of speech is fundamental in a democratic society; however, if the AMA wishes to push a political issue then such articles should be published not in the MJA but in Australian Medicine, with appropriate recognition that the issue is medicopolitical and not scientific. My concerns relate to the responsibility I consider peer-reviewed journals such as the MJA have in ensuring that articles, which include reports of alleged research, are scientifically valid. As can be seen by the media coverage on this issue, the fact that the "research study" was published in the MJA gave the impression to the general public that it was a valid scientific study and should be taken seriously. I have a number of specific issues of concern in relation to the article. Firstly, there is the fact that the principal author and researcher, in being a detainee with serious concerns over his treatment by the democratically elected government of this country, obviously has considerable potential bias in reporting observational research of this kind. Secondly, there is no attempt to assess the baseline mental health of the detainees before coming to Australia, but an assumption that all the symptoms observed are a result of detention. Thirdly, it would appear that there was no attempt to validate the observational research using more rigorous psychological testing. Finally, the methodology of the survey is not outlined, nor is the sample size discussed from the point of view of statistical validity. Admittedly, there was an acknowledgement of the survey's limitations; however, I do not consider that this article should have been published in the MJA. It does the Editorial Committee no credit and the Journal less so.

Philip Ruddock

In reply: The Hon. Philip Ruddock, the Minister for Immigration and Multicultural and Indigenous Affairs, believes that the article by Sultan and O'Sullivan1 makes "several claims that are factually wrong and would not have withstood even cursory examination". When assessing whether contributions are suitable for publication in the Journal, the editors of the MJA submit them to peer review, which has been defined as "independent assessment of the scientific merit of research by experts having knowledge of the research area equal to that of the performers of the work".2 The article referred to by the Minister posed a dilemma. Those with "equal knowledge of the research area" — representatives of the Department of Immigration and Multicultural Affairs or the Department's provider of security services (Australasian Correctional Management) — would have a patent conflict of interest. The Journal did, however, seek advice as to the accuracy of the article from a person with knowledge of the situation at Villawood, who did not advise as to "errors of fact and distortions", as alleged by the Minister. The discrepancy between the truth according to the Minister and according to Sultan and O'Sullivan may reflect differing interpretations and perceptions of information conveyed by intermediaries. Graves implies that detained asylum seekers' mental health and access to healthcare is a medicopolitical issue and that the forum for this debate should be the Australian Medical Association. So it should be! But the MJA has complete editorial independence from the AMA, and we believe that the mental health of asylum seekers transcends medicopolitics and goes to the very core of the ideals of medicine. Finally, it was very clear from the title of the article, and its authorship, that it represented the perspectives of a detained doctor (one of Graves' concerns) and a former visiting psychologist. However, we believe that our readership is sophisticated enough to interpret the content of such articles. We also believe, contrary to Ruddock and Graves, that the standing of the Journal remains solid, as does its policy of "providing a forum for . . . commentary and informed debate on standards of clinical practice, ethics, social, legal and other issues related to health care in Australia".3 To our knowledge, the information conveyed by Sultan and O'Sullivan's article is the best available data on the mental health of detained asylum seekers in Australia. We look forward to a more rigorous and independent assessment, hopefully initiated by the Minister. The opportunity is his.

Philip Ruddock