Volume 177 Issue 10

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

When asked about features of women's health services that would best meet their needs, specific groups of Aboriginal and Torres Strait Islander women, despite their diversity, have given very similar responses.1-3 They want women's healthcare that takes a holistic rather than a narrow "single-disease" or biomedical approach; services that are accessible, flexible and supportive; and providers they can trust, who are respectful and who can communicate well. For many Aboriginal and Torres Strait Islander women, having access to a female provider is critical to their acceptance of women's healthcare services. The higher cervical cancer incidence and mortality for Aboriginal and Torres Strait Islander women compared with other women, and the available evidence about screening effectiveness, provide a strong imperative for healthcare providers and funders to listen carefully and respond to what women say they want.4 The article by Coory and colleagues in this issue of the Journal (page 544) quantifies and compares women's participation in cervical screening by analysing data from the Queensland Health Pap Smear Registry.5 Participation for women living in rural and remote Aboriginal and Torres Strait Islander communities in Queensland was generally lower than for women living in other areas. Proportions of women in these communities who had had a Pap smear over a two-year period ranged from 19% to 63%. These results suggest women's needs for women's health services are being better met in some communities than others. In interpreting their analysis, Coory et al used residence in a community where most people were Aboriginal and/or Torres Strait Islander as a proxy for Indigenous status. We believe this is a resourceful and reasonably valid way around Indigenous status not being identified on the Pap smear register. However, one limitation is that we can learn nothing about Aboriginal and Torres Strait Islander women living in other localities (ie, the majority of Aboriginal and Torres Strait Islander women in both Queensland and Australia more generally). It is important that the needs of these women are not neglected because of the lack of quantitative data with which to measure them. We commend the researchers for acknowledging the sensitivities of identifying data from individual Aboriginal and Torres Strait Islander communities in their research. However, rather than only obtaining permission to do so from a government department, we believe consulting directly with members of the communities concerned at an early stage of the project may have been beneficial. Although such a practice is uncommon in this type of research, and may be challenging and more time-consuming, it may also create or strengthen trust, links and understanding, which could be useful when implementing and evaluating subsequent interventions. Coory et al suggest that the higher cervical screening participation rates in some communities are an indication of what is achievable, and express support for a strategy of strengthening primary health care. We agree with these conclusions, but disagree that an intervention study where communities are randomised would be an ideal next step. Although randomised-community intervention trials have been implemented in other settings,6 for Aboriginal and Torres Strait Islander communities the barriers to delivery of women's health services are likely to be highly location-specific and the means to overcome them not amenable to random allocation. We believe any available resources would be better spent on (i) exploring in more detail the factors contributing to high and low levels of participation, and (ii) responding actively to identified issues in communities with lower levels of participation. Barriers to Aboriginal and Torres Strait Islander women accessing women's cancer screening services, and ways of responding to them, have been reviewed — most recently in the context of considering how to support the roles of general practitioners.1,4 We would like to highlight the need to also support the roles of Aboriginal Health Workers (AHWs). Because of their key role in providing primary health care for Aboriginal and Torres Strait Islander people, the need for improved clarity, recognition and support of AHW roles has been identified as a national priority.7 We have worked with many female AHWs who have had personal experience of the impact of cervical cancer on Aboriginal and Torres Strait Islander women and their communities, and are keen to be involved in women's health education and promotion activities. Some AHWs also want to provide women's clinical care, including taking Pap smears. Some of the specific areas needing attention are the provision of better training for AHWs in women's health, and issues of accreditation, legal cover and quality assurance for those wanting to take Pap smears. Finally, we urge caution about evaluating cervical screening programs solely on the basis of participation rates. Recent commentaries have begun to question a primary aim for screening programs of maximising participation, arguing that this may lead to the positive effects of screening being overstated, and the limitations and possible negative effects of screening and its sequelae being ignored or downplayed.8,9 These commentators acknowledge that providing more balanced information about screening may have a negative impact on participation rates, but stress the importance of individuals being informed about screening and being able to choose for themselves whether or not to participate.8 Qualitative research conducted with women in one rural Aboriginal community with high rates of participation in cervical screening found that many of the women had little understanding of cervical screening or its implications.10 For programs successful in terms of participation, questions may remain about the extent to which women are making an informed choice about screening. In many localities, providers' attempts to consistently give adequate information to Aboriginal and Torres Strait Islander women can be constrained by many factors, including lack of time, and language and cultural differences. These barriers, combined with a high level of concern about cervical cancer and evaluation criteria based mainly on participation rates, may lead to an emphasis on persuading women to have a Pap smear rather than on providing information and an opportunity for informed choice. We strongly advocate that evaluators of cervical screening programs take into account not only participation rates, but also Aboriginal and Torres Strait Islander women's views about available health services and their understanding of screening-related issues.

Jennifer M Hunt MB BS MPH FAFPHM Public Health · Lynore K Geia BN, RM, MPH

Glaucoma is the most common neurodegenerative disease of the optic nerve, with a prevalence of about 3%.1 This means that about 150 000 Australians, about 75% of whom are aged over 70, have glaucoma. This number will double over the next 30 years as our population ages.2 After macular degeneration, glaucoma is the second most common cause of irreversible blindness in our community,3 and the commonest cause of preventable blindness. The basis for the most common form of glaucoma is multifactorial. Genetic linkage analysis has isolated several putative genes for open-angle glaucoma, but these account for only a small percentage of cases. Risk factors for open-angle glaucoma identified in Australian cross-sectional analyses include age and intraocular pressure, family history of glaucoma,4 myopia, systemic hypertension and diabetes.5 In glaucoma there is a relatively slow loss of retinal ganglion-cell axons. Early loss is usually in the mid-peripheral visual field. The disease becomes symptomatic at a relatively late stage when central vision is affected and the visual acuity declines, or extensive loss of peripheral vision leads to problems with mobility. However, because progression of the visual field loss is relatively slow, glaucoma is responsible for a relatively small number of the new cases of visual acuity impairment detected annually.6 Underpinning the treatment of glaucoma is a reduction in intraocular pressure. Recent reports7,8 provide evidence for setting a target intraocular pressure level for each patient, depending on the assessed risk of progressive visual damage, such as extent and rate of prior damage, proximity of the visual damage to the point of fixation (most sensitive central vision), likely number of years of life remaining for the patient, family history, and the level of intraocular pressure at which damage has occurred. Usually, topical drug therapy is used first. Newer drugs such as prostaglandin F2α agonists (latanoprost, travoprost), prostamides (bimatoprost), topical carbonic anhydrase inhibitors (dorzolamide, brinzolamide), β-blockers (timolol, laevobunolol, betaxolol) and α2-agonists (brimonidine, apraclonidine) have tended to replace pilocarpine and adrenalin-related compounds. Since drugs are absorbed directly from the nasal mucosa into the venous circulation with hepatic by-pass, topical agents mimic intravenous drugs — their safety margin can be widened by simple eyelid closure and digital occlusion of the tear duct for at least two minutes after instillation. Non-compliance and difficulties with instillation techniques remain major challenges for the long-term treatment of this incurable and asymptomatic condition. Failure to achieve target intraocular pressures by medical means usually leads to laser procedures. Laser trabeculoplasty techniques offer a 75% chance of helpful intraocular pressure reduction, with a 50% chance of continuing benefit for up to five years. Lasers are also used to achieve peripheral iridectomies for angle-closure glaucoma, and to inhibit aqueous inflow by ciliary-body destruction in blind, painful eyes. If medical and laser treatments fail, incisional surgery is performed to create an alternative pathway for the aqueous humor onto the scleral surface. Augmented with antifibrotic agents (5-fluorouracil, mitomycin-C), a long-term success rate of up to 90% can be achieved. For patients with exaggerated healing responses (glaucoma secondary to uveitis or rubeosis), intraocular pressure reduction can be achieved with plastic tubes draining into plastic reservoirs (there are three types of implants — Molteno [Molteno Ophthalmic, Dunedin], Baerveldt [Pharmacia, Kalamazoo, Mich, USA], and Ahmed [New World Medical, Rancho Cucamonga, Calif, USA]). Beyond intraocular pressure reduction lies the hope of neuroprotection — an attempt to prevent initiation or progression of intracellular processes resulting in retinal ganglion-cell apoptosis (induced cell suicide). A large international multicentre prospective randomised clinical trial of the NMDA [N-methyl-d-aspartate]-receptor antagonist memantine is under way, with results expected in 2006. Our conventional approach of looking for visual field defects, or for their progression, with white-on-white automated perimetry is still the main method of monitoring for glaucoma stability, but a large proportion of the nerve fibres can be lost before an initial defect is seen. (In white-on-white automated perimetry, a white light target is projected into a white background bowl — the patient responds when the light is just seen and its intensity is varied until the threshold for seeing has been crossed, which determines the sensitivity of the retina at that point.) There are now several new techniques for the detection of glaucoma which are specifically designed to detect change at earlier stages of the disease. Psychophysical tests are available which target smaller subpopulations of ganglion cells, such as frequency-doubled perimetry, and short-wavelength automated perimetry (blue target/yellow background). Multifocal objective perimetry, recently developed at the Save Sight Institute,9 records a multifocal visual evoked potential and removes the need for patients' subjective responses. Optic disc and nerve fibre imaging techniques using scanning laser ophthalmoscopes (eg, Heidelberg retina tomograph [HRT, Heidelberg Engineering, Heidelberg, Germany]; GDx Access [Laser Diagnostic Technologies, Jackson, Fl, USA]) or optical coherence tomography can provide objective measures of structural change. Glaucoma blindness is largely preventable. While the visual damage is not reversible, it can usually be arrested. To achieve this, early diagnosis of this otherwise progressive, asymptomatic process is essential. Ninety per cent of the Australian population visits a general practitioner annually, yet 50% of patients with glaucoma identified in population surveys are undiagnosed and untreated.3 Clinically, the first changes occur at the optic disc and it is vital that clinicians look for the characteristic sign of optic disc cupping. Every GP should view a patient's optic disc with an ophthalmoscope from time to time, especially if one or more risk factors are present. Anyone with significant optic disc cupping, or asymmetry between the optic discs of the two eyes, should be referred for investigation.

Ivan Goldberg FRANZCO, FRACS · Stuart L Graham FRANZCO, FRACS · Paul R Healey FRANZCO, BMedSc, MMed

Patients undergoing surgery need good advice not only about whether a particular elective procedure is truly necessary and likely to be of benefit, but also about the nature and chances of an adverse outcome from the surgery itself. Those with heart failure, coronary artery disease, diabetes or emphysema are more likely to suffer serious complications or death after major surgery. Epidural anaesthesia and analgesia may be a preferable technique in such patients,1 as epidural block can attenuate the neurohumoral stress response to surgery,2 potentially improving postoperative cardiorespiratory function and reducing complications. Many small randomised controlled trials (RCTs) have supported this conjecture, but, because most serious complications after anaesthesia and surgery are rare, none has had sufficient power to demonstrate whether epidural block significantly improves postoperative outcome. A solution to this problem is to combine the results of all available trials in a meta-analysis. Applying this approach to data from 141 RCTs involving 9559 patients, Rodgers et al showed that the use of epidural or spinal block (with or without general anaesthesia) resulted in a significant 30% reduction in mortality after surgery.3 Outcomes causing major morbidity (major morbidity endpoints) such as thromboembolism and pneumonia were also reduced. Another recent systematic review found that epidural block reduces postoperative myocardial infarction.4 Although these findings are supportive, they were based exclusively on small RCTs, and meta-analyses sometimes give conflicting results when compared with large RCTs.5 In 1984, Yusuf et al explained how large, simple randomised trials can reliably detect moderate effects on important but uncommon outcomes such as death or major morbidity after surgery.6 Two of us are part of an Australian group that has recently published the results of a large multicentre RCT of epidural block in 888 high-risk patients undergoing major abdominal surgery (the MASTER trial).7 Patients were randomly allocated to receive general anaesthesia with or without epidural block. The epidural block was established before the commencement of surgery (epidural anaesthesia) and epidural analgesia was continued for three days after surgery. All other care was left to the discretion of the anaesthetist and surgical team: most patients were managed in general surgical wards after surgery, although some required high-dependency or intensive care. Thus, our trial was a test of effectiveness in routine practice and its results can be generalised. There was no significant difference in mortality at 30 days or in overall morbidity — 57% of epidural and 61% of control group patients had at least one morbidity endpoint (sepsis, respiratory failure, myocardial infarction, heart failure, renal failure, gastrointestinal [bleeding or need for parenteral nutrition], hepatitis, or haematological [anaemia, leukopenia, or thrombocytopenia]) or died (P = 0.30). Mortality at 30 days was low in both groups (epidural, 5.1%; control, 4.3%). Of the eight morbidity endpoints studied, only one, respiratory failure, occurred less frequently in patients managed with epidural block (epidural, 23%; control, 30%; P = 0.03). Another large RCT published recently showed similar results.8 Thus, there is no evidence that epidural block improves outcome in most patients undergoing major abdominal surgery with general anaesthesia, other than for respiratory complications. Nevertheless, in the MASTER trial, pain scores over the first three days after surgery were significantly lower in the epidural group. This difference occurred despite most participants in the control group receiving multimodal analgesia.1 This demonstrates some benefit from epidural block: a reduction in pain may assist deep breathing and coughing after surgery, and this may help prevent atelectasis and pneumonia. The MASTER trial, and other recent data,3,8 provide some evidence to support this conjecture. What are the risks of epidural block? One should take into account the risk of an epidural haematoma when an epidural or spinal needle is placed, particularly in a patient receiving anticoagulation therapy (eg, perioperative thromboprophylaxis). The risk of epidural haematoma or abscess is very low, with estimates varying widely from about 1 : 1700 to 1 : 200 000.9,10 Concern regarding damage to the spinal cord (leading to paraplegia) or nerves must be weighed up against the benefits of improved postoperative analgesia. Evidence-based practice is dependent on good quality research, and the best evidence comes from large trials.6 The MASTER trial provides reliable information for doctors and their patients on which to base their decisions regarding the best methods of anaesthesia and analgesia after surgery. Epidural block seems to provide additional benefit for patients at increased risk of postoperative respiratory complications. When considering use of epidural block, doctors and patients also need to weigh up the benefits of improved pain relief against the (rare) risk of paraplegia and other nerve injury. Our experience has been that some patients will choose postoperative epidural analgesia after being given this information.

Paul S Myles MPH, MD, FCARCSI, FANZCA · Ian Power FRCA, MD, FFPMANZCA, FANZCA · Konrad Jamrozik DPhil, FAFPHM, MFPHM

Objectives: To examine trends in the licit consumption of the psychostimulants dexamphetamine and methylphenidate in Australia and nine other countries from 1994 to 2000 and in each State and Territory of Australia from 1984 to 2000.Design: Annual rates of consumption of psychostimulants were compared using Poisson regression models. All drug consumption was standardised to defined daily doses per 1000 population per day.Main outcome measures: Rates of consumption of each psychostimulant in each country and in each Australian State and Territory.Results: For the 10 countries from 1994 to 2000, total psychostimulant consumption increased by an average 12% per year, with the highest increase from 1998 to 2000. Australia and New Zealand ranked third in total psychostimulant use after the United States and Canada. Australia consumed significantly more than the United Kingdom, Sweden, Spain, the Netherlands, France or Denmark. In Australia, from 1984 to 2000, the rate of consumption of licit psychostimulants increased by 26% per year, with an 8.46-fold increase from 1994 to 2000. Western Australia ranked first, with nearly twice the consumption rate of total psychostimulants as New South Wales, which ranked second. Methylphenidate is the main psychostimulant consumed in the US and Canada, and dexamphetamine in Australia.Conclusions: The consumption of psychostimulants in Australia is high internationally and varies significantly between States and Territories. The results imply varied jurisdictional prescribing determinants and supply processes throughout Australia, which may require new national prescribing standards and access to online patient data for prescribers and dispensers.

Constantine G Berbatis BSc, MSc · V Bruce Sunderland PhD, FPS · Max Bulsara BSc(Hons), MSc

Objective: To investigate the extent of participation in cervical cancer screening among women who live in discrete rural and remote Indigenous communities in Queensland.Design: Descriptive analysis of data from the Queensland Health Pap Smear Registry for the period March 1999 to February 2001.Subjects: Women aged 20–69 years who had given their address of usual residence as one of 13 discrete rural and remote Indigenous communities in Queensland.Main outcome measures: Proportion of women who participated in cervical screening over a two-year period ("biennial participation percentage") and variation in participation across the 13 communities.Results: Overall, the biennial participation percentage in the Indigenous communities was 41.1%. This was 30% lower (risk ratio, 0.70; 95% CI, 0.67–0.72) than that for the rest of Queensland. There was statistically significant variation among communities, with biennial participation percentage ranging from 19.9% to 63.5%.Conclusions: The variation in participation across the communities suggests that the problem of low participation among Indigenous women is not intractable. Achieving participation rates similar to the highest rates found in our study would be of major benefit to Indigenous women.

Michael D Coory MB BS, PhD · Jennifer M Muller MEnvCommHealth, GradDipHealthProm · Nathan A M Dunn BSc(Hons) · Patricia S Fagan MB BS, FAFPHM

Objective: To determine the incidence and appropriateness of use of allogenic packed red blood cell (RBC) transfusion in Australian and New Zealand intensive care practice.Setting: Intensive care units of 18 Australian and New Zealand hospitals: March 2001.Design: Prospective, observational, multicentre study.Methods: All admissions to participating intensive care units were screened and all patients who received a transfusion of RBC were enrolled. The indications for transfusion were recorded and compared with Australian National Health and Medical Research Council guidelines. Transfusions conforming to these guidelines were deemed appropriate.Main outcome measures: RBC transfusion in intensive care and transfusion appropriateness.Results: 1808 admissions to intensive care units were screened: 357 (19.8%) admissions (350 patients) received an RBC transfusion while in intensive care. Overall, 1464 RBC units were administered in intensive care on 576 transfusion days. The most common indications for transfusion were acute bleeding (60.1%; 880/1464) and diminished physiological reserve (28.9%; 423/1464). The rate of inappropriate transfusion was 3.0% (44/1464). Diminished physiological reserve with haemogloblin level ≥ 100 g/L was the indication in 50% (22/44) of inappropriate transfusions; no indication was provided for 31% (15/44).Conclusion: The rate of inappropriate transfusion in Australian and New Zealand intensive care units in 2001 was remarkably low.

and the ANZICS Clinical Trials Group*

Fentanyl is a synthetic opioid with a potency about 200 times that of morphine. It has been misused since it was first introduced, mostly by medical and paramedical personnel because of limited access. Illicit manufacture of fentanyl and its analogues ("China White") occurs sporadically and has been associated with regional fatal epidemics in the United States.1 With the introduction of a transdermal delivery system for fentanyl (Durogesic, Janssen-Cilag) for managing chronic and cancer pain, there is a widening pool of individuals with access. Prescribed transdermal fentanyl patches can be sold or stolen. Clinical recordA 35-year-old woman with a history of intravenous drug use was brought by ambulance to the emergency department after an intravenous overdose of the contents of a transdermal fentanyl patch. The ambulance had been called to a private home where there were two people unconscious, a man and a woman. Both appeared to have had acute narcotic overdoses. It was later confirmed that they had shared (and injected intravenously) the contents of a transdermal fentanyl patch (5 mg) found at the scene. Both patients were rapidly assessed by the ambulance officers, and the initial resuscitation concentrated on the male patient, who, at first assessment, appeared to be in a more critical state. He was unrousable and was reported to have Cheyne–Stokes respiration. His blood sugar level was checked (10.5 mmol/L) and he was given 1.2 mg naloxone intravenously. He recovered consciousness within five minutes and subsequently absconded from the scene while the second patient was being treated. In the interim, the female patient had suffered a cardiorespiratory arrest. Cardiopulmonary resuscitation was commenced, with the assistance of police officers who were also in attendance. According to ambulance records, her initial rhythm was electromechanical dissociation, which subsequently deteriorated into ventricular fibrillation. A direct current countershock (200 J energy) was applied. The patient went into asystole. She was intubated and intermittent positive pressure ventilation with 100% oxygen was started. Naloxone 1.6 mg, adrenalin 10 mg (total dose) and atropine 2 mg were administered intravenously. Subsequently, she developed a narrow complex tachycardia with a rate of 130 beats/minute and had a palpable cardiac output. The total time spent at the scene was 40 minutes, and transport time to hospital took 5 minutes. On arrival at the emergency department she was unconscious, with a Glasgow Coma Score of 3. Her pupils were dilated and non-reactive to light. She was making occasional attempts at respiration and was ventilated as above with 100% oxygen. Her heart rate was 120 beats/minute in sinus rhythm, systolic blood pressure 55 mmHg and oxygen saturation 97%. One litre of Haemaccel and a noradrenalin infusion were administered, resulting in an initial improvement in systolic blood pressure to 95 mmHg. It was evident that she had vomited at the scene, and clinical signs were consistent with aspiration, which was later confirmed on chest x-ray. Laboratory results for arterial blood, serum and urine are shown in Box 1. She was transferred to the intensive care unit, where she subsequently developed diabetes insipidus, abnormal liver function, disseminated intravascular coagulation and had ongoing haemodynamic instability. The next day, cerebral computed tomography (CT) scan showed changes in the basal ganglia and mild generalised cerebral swelling consistent with severe hypoxia (Box 2A). A CT scan of her abdomen showed generalised changes in the bowel wall and mesentery consistent with bowel necrosis (Box 2B). Surgical intervention was considered to offer little in view of the severe neurological damage and multiorgan failure. Following extensive discussion with family members, inotropic support was withdrawn and she died soon afterwards. DiscussionSurprisingly, this is the first report of death caused by intravenous misuse of a fentanyl patch, although there have been fatalities after oral ingestion.2 Non-fatal inhalational misuse has been reported.3 In one death (a funeral home employee who obtained a patch from a deceased person in a nursing home), the route of administration was never determined.4 Finally, there is a report of respiratory arrest from intentional intravenous misuse of transdermal fentanyl in a patient with chronic pain.5 Transdermal fentanyl is used in the management of both acute and chronic pain.6 In Australia, it is listed on the Pharmaceutical Benefits Scheme for pain caused by malignant neoplasia, but it has increasingly found a role in the management of chronic non-malignant pain. Durogesic patches are available in four sizes, the smallest designed to deliver 25 μg/h of fentanyl (total 2.5 mg fentanyl), the largest 100 μg/h (total 10 mg fentanyl) (Box 3). Extracting the contents of a patch can lead to a large variation in the dose administered, making a dangerous habit perilous. The pharmacokinetics of fentanyl only add to the danger. Fentanyl is highly lipid-soluble and has a large volume of distribution (60–300 L). A 2 μg/kg dose leads to high immediate serum concentrations (up to 11 ng/mL) that fall rapidly to less than 1 ng/mL after one hour.7 The offset of action is mainly by redistribution. An effective analgesic concentration of fentanyl ranges from 0.3–0.7 ng/mL.8 Serum concentrations in health professionals who died after fentanyl misuse have ranged from 0.1–5 ng/mL,1 in keeping with the average concentration of 3 ng/mL found in a larger series of deaths from fentanyl misuse.9 The serum fentanyl concentration in our case accords with concentrations given in other published cases. The absence of detectable opioid in the urine was consistent with a recent overdose. Alterations to the formulation or presentation of the patch might make it harder to misuse the drug intravenously, although previous experience with temazepam in several countries has shown that such changes are not necessarily the answer. In response to the ease of misuse of temazepam, the pharmaceutical industry produced a gel-filled formulation that would be as "resistant" as possible to injecting. However, it has since been shown that the gel-filled preparation is readily injectable,10 and results in more medical complications, including superficial thrombophlebitis, abscesses and deep venous thrombosis. There are several case reports of ischaemic necrosis of digits resulting from intra-arterial injection of temazepam.11 To date, all reported deaths from misuse of fentanyl patches have been the result of other than intravenous misuse. Prescribers need to be alert to the potential for misuse and the consequent risk of fatality. They need to impress on patients the importance of secure storage of patches. Stricter regulation of fentanyl patches could be enforced by accounting for patches after they have been dispensed (eg, ensuring that, on the death of a patient issued with patches, unused medications are returned). A more draconian measure would be to require that used patches be returned before more can be issued. However, this would not prevent the theft of patches, and would put the onus of returning used patches on patients already suffering from debilitating pain. It is clearly necessary to continue to strictly control the availability and prescribing of these patches, which potentially are a highly sought after product on the black market. However, it is also important to avoid widespread public warning of the risk, which might serve only to advertise a novel activity for risk-takers. 1: Laboratory results in a patient after intravenous misuse of transdermal fentanyl Arterial blood (reference range) pH 6.88 (7.36–7.44) Pco2 62 (36–44) mmHg Po2 170 (85–100 mmHg on room air) Bicarbonate 11.3 (22–26) mmol/L Base excess – 22.6 mmol/L Haemoglobin 114 g/L Serum Paracetamol Negative Salicylate Negative Benzodiazepines Negative Tricyclic antidepressants Negative Fentanyl 2.0 ng/mL Ethanol 0.16 g/L Urine Cannabinoids Negative Opioids Negative Amphetamine Negative 2: Computed tomography (CT) scans A: Brain — arrows indicate areas of low attenuation in basal ganglia. B: Abdomen — arrows show changes consistent with bowel necrosis. 3: Transdermal fentanyl patch Fentanyl, dissolved in ethanol (a flux enhancer) and gelled with hydroxyethyl cellulose, is held in a drug reservoir between a backing layer and a rate-controlling membrane on an adhesive base.

Mark D Reeves MB BS, FANZCA · Corinne J Ginifer MB BS, FACEM

Current controversy about diagnosis and treatment of attention deficit hyperactivity disorder (ADHD) reflects the divergence between developmental and non-developmental approaches. While there is growing evidence for biological vulnerabilities associated with ADHD, we believe that environmental factors, including early problems in parental attachment, are also important in determining the type and timing of deficit that a child develops, the risk to academic and social performance and eventual outcome. We warn against labelling children with ADHD simply because they fulfil the cross-sectional diagnostic symptom criteria of the Diagnostic and statistical manual of mental disorders — 4th edition (DSM-IV). We advocate an integrated biopsychosocial approach to diagnosis and management with a thorough developmental assessment to identify developmental factors, such as deficits in early attachment, contributing to the presentation.

George Halasz MRCPsych, FRANZCP · Alasdair L A Vance MD, FRANZCP

Chlamydia pneumoniae has been detected in atherosclerotic plaques, while seropositivity to this organism confers a slightly increased risk of coronary events. However, no aetiological link has been established; a major difficulty when investigating this link is the lack of a gold standard for diagnosing chronic vessel infection. The outcomes of case–control studies and prospective trials of macrolides in treatment and prevention of cardiovascular disease have been ambiguous but suggest a short-term preventive effect. Whether this is due to the antimicrobial or anti-inflammatory activity of the macrolides is unknown. Larger and longer prospective trials currently under way may provide better insight into the association of C. pneumoniae with cardiovascular disease. At present, there is no justification for treating cardiovascular disease with antibiotics.

Mikkel M Larsen BSc · Birgitte Moern MD, PhD · Paul L Andersen MD, DMSc · Lars J Ostergaard PhD, DMSc · Andrew Fuller FRACP

QuestionShould patients with resectable rectal cancer receive preoperative pelvic radiotherapy? Trial details Design: Multicentre international randomised controlled trial. Setting: 108 hospitals, with a predominance of hospitals in Holland. Patients: 1805 eligible patients with histologically proven adenocarcinoma of the rectum. Patients with fixed tumours were excluded. All patients had either a low anterior resection or an abdominoperineal resection using the standardised surgical approach of total mesorectal excision (TME). Intervention: After stratification for treatment centre and anticipated type of operation, 897 patients were assigned to receive, or not receive, preoperative short course pelvic radiotherapy (25 Gy in five daily fractions). Main outcome measures: Overall survival and local recurrence at 2 years; overall and distant recurrence at 2 years; and postoperative mortality and morbidity (operative blood loss and perineal complications). Overall survival was analysed on an intention-to-treat basis and included all eligible patients. The rate of local recurrence was calculated for patients in whom macroscopically complete resection had been achieved. The rate of distant recurrence was calculated for patients who did not have distant disease at the time of surgery. Main results: Median follow-up in surviving eligible patients without local recurrence was 24.9 months (range, 1.1–56 months). There was no difference in overall survival between the two groups (82.0% radiotherapy and surgery v 81.8% surgery alone; P = 0.84). However, local recurrence was significantly lower in patients who had received radiotherapy compared with those who had not (2.4% v 8.2%; P < 0.001). There was no difference in the rate of distant recurrence (P = 0.87). Median operative blood loss was marginally increased in the combined arm (1000 v 900 mL; P < 0.001), as were perineal complications after abdominoperineal resection (26% v 18%; P = 0.05). Conclusion: A short course of preoperative pelvic radiotherapy reduces pelvic recurrence after total mesorectal excision. Despite this, overall survival was not improved. CommentaryRationale for the trialThe primary treatment for rectal cancer is surgery. The rationale for also irradiating the pelvis is to prevent the suffering associated with local recurrence, even if survival is not improved. Pelvic recurrence after surgery occurs in 20%–50% of patients with poor pathological features, such as transmural spread and nodal involvement.1 Adjuvant treatment may be delivered either before or after surgery and may also include chemotherapy. The risk of serious late bowel complications after postoperative adjuvant treatment is reported to be higher when compared with preoperative treatment.2 It is this finding that has, in part, prompted interest in delivering radiotherapy before surgery. The optimal scheduling for preoperative radiotherapy remains unclear. Radiation oncologists in Europe favour a short preoperative approach (25 Gy in five daily fractions), while those in North America and Australia favour a more protracted preoperative regimen (45–50.4 Gy in 25–28 daily fractions, with infusional chemotherapy) given to a more select subgroup of patients with unfavourable clinical features such as tethering and fixation,3 with the aim of "downstaging" the cancer and enhancing operability. While the approach to pelvic radiotherapy may differ, many surgeons advocate that surgery alone, using a total mesorectal excision (TME) technique, could negate the need for pelvic radiotherapy as a result of relatively low pelvic recurrence rates (< 10%).4 Trial methodsThis was a large and well-constructed multicentre trial with an adequate sample size and an equal balance of important variables, especially disease stage. The random allocation was simple, and the surgery was standardised by training and supervision of surgeons. Pathological findings were independently reviewed. Importantly, the authors detailed all patients deemed ineligible after randomisation with an acceptable figure of 3% in both arms. All protocol violations were listed. A limited subgroup analysis failed to show a differing treatment effect based on TNM stage or tumour location. The results (overall survival and recurrence) are presented at two years, which is appropriate considering the median follow-up time was only 24.9 months and the number of patients reaching three and four years of follow-up limited at the time of analysis. Short-term toxicity data are presented without any significant difference between the two groups, except for perineal complications. In a subsequent report, the authors showed no difference in postoperative mortality (4% v 3.3%; P = 0.49).5 This is an important finding, as previous studies in which a suboptimal radiotherapy technique was used showed an increased mortality rate. New informationThis is one of the largest studies examining the potential benefits of a short course of preoperative pelvic radiotherapy. Unlike a previous Swedish study,6 this trial failed to show any overall survival benefit (at least at two years), despite a significant reduction in local recurrence from 8.2% to 2.4%. This reduction was seen despite standardised TME. The question remains whether the morbidity (acute and late) of radiotherapy justifies the reduction in local recurrence and its associated morbidity (intractable pain, bleeding, discharge). Longer-term toxicity data are not yet available from this trial. Nor are data on the late effect of bowel function and quality of life from the addition of radiotherapy. Despite what many would consider an excellent pelvic control rate with TME alone, the local recurrence rate after surgery only in those diagnosed with tumour–node–metastases (TNM) stage III disease was 15%. Pelvic radiotherapy reduced this to 4.3% (P < 0.001). Those with TNM stage II disease also benefited from radiotherapy, although not as much in absolute terms (5.7% v 1%; P = 0.01). Notably, this study included over 500 patients with TNM stage I disease, so it is not surprising that the local recurrence rate in both arms was less than 1%. Patients with TNM stage I disease are not routinely given adjuvant treatment (see TNM classification of malignant tumours7). Implications for clinical practiceThe optimal approach to adjuvant treatment in rectal cancer remains unclear. The National Health and Medical Research Council (NHMRC) guidelines on the prevention, early detection and management of colorectal cancer currently recommend that any adjuvant treatment should include both chemotherapy and radiotherapy,8 as this combination has been consistently shown to give a survival advantage.9 To date, there are insufficient data to recommend preoperative radiotherapy alone as the standard approach. The NHMRC guidelines recommend combined treatment be given to patients with high-risk rectal cancer (transmural spread and/or nodal involvement). The Trans Tasman Radiation Oncology Group (TROG) is currently comparing long-course preoperative radiotherapy (50.4 Gy in 28 daily fractions) and chemotherapy (infusional fluorouracil) with short-course radiotherapy (25 Gy in five daily fractions), and early surgery in patients with T3 resectable rectal cancer. TROG is also comparing combination preoperative treatment (50.4 Gy and infusional fluorouracil) with surgery plus optional combined adjuvant treatment. The results from this trial will be important in clarifying many unanswered questions. The Dutch Colorectal Cancer Trial raises the not uncommon scenario of a small but potentially clinically meaningful benefit (ie, in this case, decreased pelvic recurrence after total mesorectal excision) from a treatment. But who should benefit? Those at higher risk will obviously benefit the most in absolute terms and perhaps be more willing to accept the toxicity associated with treatment. Ultimately, it is the patient who must decide if the benefits outweigh the inconvenience and potential toxicities of treatment.

Michael J Veness MB BS, MMed(ClinEpi),FRANZCR

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

Clinical record In 2001, an outbreak of Bordetella pertussis infection occurred in a hospital special care nursery. The index case, Parent A, had a two- to three-week history of non-productive cough without paroxysms, whooping or vomiting. She was admitted to hospital in labour and gave birth to Baby A on Day 0 in early January 2001. The baby was born four weeks prematurely and remained in the special care nursery for 15 days. The mother spent several hours there each day handling her infant and occasionally assisted with the care of infants in adjoining cots. Serum taken 19 days after she developed cough was negative for Bordetella pertussis IgA and IgM and equivocal for IgG (Figure) by enzyme-linked immunosorbent assay (ELISA). On Day 15, Baby B (in the cot adjacent to Baby A) developed a cough and respiratory distress. Initial investigations did not reveal a cause. A nasopharyngeal aspirate collected on Day 23 was positive for B. pertussis by direct fluorescent antigen testing. On Day 18, a nurse who had cared for Baby A developed a cough. Her serum was positive for B. pertussis IgA on Day 31 by ELISA; she commenced roxithromycin therapy. On Day 31, serum collected from Parent A was positive for B. pertussis IgA. On Day 39, Baby A developed a respiratory illness despite a seven-day prophylactic course of erythromycin. A nasopharyngeal aspirate was positive for B. pertussis by polymerase chain reaction (PCR) on Day 43. Both received erythromycin. On Day 41, a nasopharyngeal aspirate collected during contact tracing from Baby C (in the nursery from Day 0 to 6) grew B. pertussis. She had not received chemoprophylaxis and had a mild cough. She was treated with erythromycin. Contact tracing Contact tracing of infants, parents and staff was undertaken. The incubation period of pertussis is six to 21 days, and is generally less than 10 days.1 The parents of all 19 babies who had been in cots adjoining Baby A were notified, and erythromycin prophylaxis was recommended. Eleven babies received prophylaxis. Parents of other babies in the nursery were contacted and advised to notify the hospital if they or their babies developed a cough over the following 21 days. No further cases were identified. Fifty-three staff with close contact with the cases were screened using direct fluorescent antigen testing of nasopharyngeal aspirates and serum IgA testing. Thirty-three received antibiotic prophylaxis. No staff had illness consistent with acute pertussis or developed infection, as shown by nasopharyngeal and serum IgA tests. There were 10 339 notifications of pertussis in 1999– 2000 in Australia.2 People aged 15 years and over comprised 62% of these notifications.2 Studies indicate that 12% to 32% of cases of prolonged cough (over two weeks' duration) in adolescents and adults are due to B. pertussis.3 Recent articles highlight reasons for the apparent shift in disease burden.3-6 Although adolescents and adults are the primary reservoir of the disease and the main source of infection in infants and unvaccinated children, pertussis often goes unsuspected in these age groups.3,4,7 Pertussis in adults is difficult to detect clinically. Delays and errors in diagnosis occur because the presentation is commonly mild or atypical, and may be asymptomatic.7 This may be a result of immunological recall. As past vaccination or illness does not confer lifelong immunity to pertussis, reinfection is common.8 Detection in adults is further impeded by the poor sensitivity of culture and serological tests,1 particularly if performed several weeks after onset of cough.6 Polymerase chain reaction of nasopharyngeal swabs or aspirates improves detection.6 Deaths from pertussis average one per year in Australia, and almost all involve neonates and unvaccinated infants.2 These groups acquire pertussis in a range of settings, including hospitals;9,10 parents, visitors and healthcare staff are a source of nosocomial outbreaks.7 During the events described here, there was no concurrent epidemic of pertussis in the community (Dr B McCall, Director, Brisbane Southside Public Health Centre, personal communication). Hospitalised infants exposed to pertussis are at high risk of morbidity and mortality, as maternal antibodies provide inadequate protection, while infants less than four weeks old are too young to commence vaccination.11 While erythromycin is the prophylaxis of choice,11 it has been associated with hypertrophic pyloric stenosis in neonates.12 Preventing nosocomial transmissionB. pertussis is spread via mucosal contact with infectious respiratory droplets and secretions.13 We suggest the following standard and transmission-based infection control precautions to protect hospitalised infants: limit access of staff and visitors to nurseries; provide adequate space between open cots (at least one metre); discourage parents from contact with infants other than their own; and perform hand antisepsis before entering the nursery and before and after patient contact. Surgical masks are chiefly designed to protect the wearer from exposure to blood and body fluids; it is suggested that they may also reduce the opportunity for onward transmission of infections spread via large droplets.13 Other important measures include educating staff and visitors to report coughs, diagnosing their aetiology, and providing treatment and prophylaxis. A diagnosis of pertussis can be confirmed by nasopharyngeal aspirate (culture, direct fluorescent antigen testing or PCR) or serological testing (IgA).7 Erythromycin is the treatment and prophylaxis of choice at all ages.11 Targeted booster vaccination of parents and healthcare workers in contact with infants has been recommended, although not universally adopted.3-6 An acellular pertussis vaccine is available for this purpose, but its efficacy and duration of protection remain to be determined.3-5 Australia's pertussis immunisation schedule is restricted to infants and young children.11 Infection control measures are important because of the difficulty in prohibiting all people with coughs from entering hospital nurseries, and the cost implications of vaccinating all adults in contact with infants. Lessons from practice Pertussis should be considered in adults and older children with a cough lasting over two weeks regardless of a past history of pertussis vaccination or infection. If pertussis is suspected, a nasopharyngeal aspirate or swab should be tested by polymerase chain reaction. Serological testing may also be useful when symptoms have been established for longer than two weeks. Adolescents and adults are the primary reservoir of the disease and the main source of infection in infants and unvaccinated children. Hospitalised infants are at high risk for morbidity and mortality from pertussis. Infection control measures are paramount, and targeted vaccination of adults should be considered.

Natalie M Spearing BSN, MBA · Robert L Horvath FRACP · Joseph G McCormack FRCP, MD, FRACP

In October 2000, a 10-year-old Aboriginal boy from the Darwin region of the Northern Territory was referred to hospital with a 24-hour history of abdominal pain, initially generalised, but then localising to the right iliac fossa. The pain was accompanied by occasional vomiting, but no fever or diarrhoea was noted. At laparotomy, a gangrenous, unruptured appendix was removed. Postoperatively, the patient made a good recovery. Neither he nor any family members had travelled outside the Northern Territory. Histological sections of the surgical specimen showed changes typical of acute suppurative appendicitis. Closer examination, however, revealed numerous round-to-oval structures resembling trophozoites (see Box). When the possibility of invasive amoebiasis was raised, staining of the section with Entamoeba histolytica-specific sera confirmed the diagnosis. E. histolytica serology was negative. DiscussionE. histolytica is a protozoan parasite of humans that causes infectious colitis and amoebic liver abscess. It has recently been recognised that invasive disease is caused exclusively by the species E. histolytica, while the morphologically identical species E. dispar exists in the colonic lumen as a harmless saprophyte.1 Patients with invasive amoebiasis living in Australia and other developed countries generally acquire the infection outside their country in a region where the pathogenic species is known to be endemic. While cases of amoebiasis in Australians who have not travelled overseas have been reported,2-6 all except one pre-date the recognition of pathogenic and non-pathogenic species, and in none was the primary source identified. Isolated appendicitis due to E. histolytica is a rare condition. Even in regions where the organism is endemic, it causes less than 1% of cases of appendicitis.7 The diagnostic method of choice for amoebic colitis is testing for parasite antigen in stools.1 In this case, for the first time, the species-specific antibody normally used in the stool antigen test was adapted for immunohistochemistry and confirmed E. histolytica infection. Possible alternative methods for confirming the diagnosis include the polymerase chain reaction (PCR) or detection of E. histolytica-specific antibodies or parasite antigen in patient serum.1 However, PCR is not readily available outside research laboratories. The sensitivity of the E. histolytica antibody test in intestinal disease is 52%, a likely explanation for the negative serological test. It is curious that no outbreaks of the readily identified and much more common manifestations of this infection, namely amoebic colitis and amoebic liver abscess, have been recently reported in northern Australia. The potential public health significance of a case of proven autochthonous infection in an Australian population is reinforced by its high transmissibility in settings where hygiene may be suboptimal.8 Furthermore, there is a well documented possibility of prolonged latency between infection and the development of invasive disease.1 Both these factors support the merit of selective screening and appropriate treatment of at-risk contacts if invasive disease is detected. The use of new diagnostic techniques, such as the adapted staining method we used, will enable a more accurate determination of the endemicity of E. histolytica in Australia. Entamoeba histolytica in an inflamed appendix A: Periodic acid–Schiff stain, showing trophozoites (arrowed) within an inflammatory infiltrate. B: Immunoperoxidase stain of trophozoites (arrowed) using E. histolytica-specific sera. Note the surrounding necrotic tissue.

James S McCarthy FRACP MD · David Peacock MB BS · Kevin P Trown FRCPA · Patrick Bade FRACS · William A Petri Jr MD · Bart J Currie FRACP

To the Editor: Regarding the position statement about COX-2 inhibitors, we agree that openness about all potential conflicts of interest is the least we should expect from guideline developers, but this is not enough.1 Fifteen (65%) of the 23 members of the Australian COX-2 Specific Inhibitor Prescribing Group (including all eight of the rheumatologists involved) declared current financial links with Pfizer and Merck, Sharp and Dohme, the two drug companies marketing COX-2 inhibitors in Australia.2 The Prescribing Group can be viewed at best as a tight collaboration between some healthcare professionals and drug companies. At worst the statement published in the Journal can be seen as the "happy end" of a successful marketing campaign, which began some years ago with the enrolment of the most influential Australian rheumatologists to the advisory boards of the drug companies.3 Members of the group disregarded the "industry" bias on the basis that "some form of bias is the inevitable consequence of knowledge and involvement". However, numerous studies have shown that industry-sponsored drug information is characterised by an overemphasis on the benefits of drugs and a minimisation of the risks.4 Full trial results with celecoxib are available on the United States Food and Drug Administration website and had been consulted by the Prescribing Group.5 These data show that celecoxib is not better than diclofenac (P = 0.414) or ibuprofen (P = 0.64) in terms of ulcer complications, the prespecified primary outcome of the trials. There was also no significant difference between celecoxib and diclofenac for the combined outcome of complicated and benign ulcers (P = 0.296). It has been shown that the results previously presented in JAMA for celecoxib were flawed and had been manipulated.6,7 The wide distribution of the JAMA article by the drug company as part of intensive marketing campaigns contributed to huge sales for celecoxib. Sales of celecoxib between August 2000 and June 2002 cost Australian taxpayers more than $288 m through the Pharmaceutical Benefits Scheme (PBS), more than five times the cost for all other NSAIDs during the same time frame.8 There is some evidence to show that the PBS blow-out observed after the launch of COX-2 inhibitors is at least partly due to their use outside their approved indications (osteoarthritis and rheumatoid arthritis).9 The Prescribing Group did not give any indication for the use of COX-2 inhibitors and did not consider the cost issue, arguing that "there is little pertinent cost-effectiveness literature in the public domain". The position statement appeared to be an evidence-based review of the safety of COX-2 inhibitors involving eminent rheumatologists, active members of the PHARM committee and National Prescribing Service staff. We believe, however, that the statement promotes misinformation from the pharmaceutical industry. We invite readers to look at sources of drug information that are truly independent of drug companies, such as the Australian Medicines Handbook (www.amh.net.au), Australian Prescriber (www.australianprescriber.com) and Therapeutic Guidelines (www.tg.com.au).

Agnes I Vitry · Eve Hurley

To the Editor: We agree that the "what", "how" and "who" of guideline development all deserve equal, explicit and systematic attention.1 A fundamental task for architects of consensus guidelines is to get the "what" right first. Agreement about the importance of the topic and the objective of the exercise is crucial to its ultimate success. Edmonds and colleagues state that "formulation of precise indications for the use of NSAIDs [non-steroidal anti-inflammatory drugs] rather than CSIs [COX-2-specific inhibitors] (or vice versa) would generate interminable controversy".2 The foundation for this assertion is not clear and the authors do not present data about the level of agreement on this by the experts initially assembled. The NSW Therapeutic Assessment Group (NSW TAG) believes that providing timely, independent and evidence-based guidance to clinicians about the place in therapy for such new drugs is extremely important. The membership of NSW TAG identified this as a priority soon after the marketing of celecoxib in Australia, and agreed unanimously to develop evidence-based recommendations on indications for the use of this drug. Our consensus development process involved a wide variety of experts in therapeutics and was successfully completed without generating "interminable controversy".3 We wonder whether our different experiences may be partly related to a difference in the initial level of consensus on the importance of the chosen topic. The "how" of the process followed by Edmonds et al is not described in sufficient detail to enable systematic evaluation of its validity. How systematic was the search for evidence or the process for inclusion or exclusion of studies? What was the level of evidence on which final recommendations were based? Importantly, high quality guideline development processes require a "balance of healthcare disciplines in the guideline development group".4 Getting the right "who" is a prerequisite for getting the "how" right. Edmonds et al state that membership was arbitrary, with predominant representation from rheumatologists and relevant pharmaceutical companies. Given the problems associated with physician–industry interactions,5 it has been suggested that authors with significant conflicts of interest should be excluded from participating in guideline development.6 The rationale for arbitrary selection of members and inclusion of members from the pharmaceutical industry is not explicitly stated. These issues may have contributed to the difficulties the group experienced, and may detract from the validity of their recommendations. Future trips down the "road to consensus" should run more smoothly after careful consideration of the "what", "how" and "who" at the outset — no "ifs and buts" about it.

Madlen Gazarian · Karen I Kaye

In reply: Both Vitry and Hurley and Gazarian and Kaye would have had our consensus group address different or broader issues than safe prescribing and use of COX-2-specific inhibitors (CSIs). Indications for use, leakage and cost effectiveness are important issues, but our goal, clearly stated in our article,1 was different and, we believe, important: if a clinician has decided to use a CSI, what considerations are needed to prescribe the drug safely? Disagreements in reaching consensus were not, as suggested by Gazarian and Kaye, due to confusion about the aim of the exercise, but to differences in interpreting evidence and expressing conclusions in simple and direct terms. It would have been easy to avoid these problems by limiting participants to a small group of like-minded colleagues, but we chose to involve a broad range of people who may represent a more realistic spectrum of attitudes and approaches. We find Vitry and Hurley gratuitously pejorative in their description of the participants in this exercise. With the exception of two rheumatologists with epidemiological expertise (who did not sign off on the position statement2), all the rheumatologists involved were members of one or both advisory boards. They were a relevant group precisely because this role should involve a responsibility to provide sound advice to the industry paying for it, and equally to the profession, both in the interests of good patient care. "Current financial links" is not the way such a consultancy is usually described. They call the exercise "at best a tight collaboration between some healthcare professionals and drug companies" and "at worst . . . as the 'happy end' of a successful marketing campaign". Given that one of the two pharmaceutical companies involved declined to sign off on the statement, as did two rheumatologists who were advisory board members for the other company, this is a curious outcome of "tight collaboration". With respect to the relative safety of selective versus non-selective COX inhibitors, our considerations were based on data available from peer-reviewed studies published to the end of May 2001 and available on the United States Food and Drug Administration website, as indicated in the position statement2 and the accompanying article.1 A number of the references quoted by Vitry and Hurley became available after May 2001. Renewed scrutiny and analysis of existing datasets is interesting, but the results are best used to decide whether unresolved issues are of sufficient importance to justify further studies, and how these could be designed to deliver evidence that will convince us all, one way or the other. We made the point at the conclusion of the position statement that this is an evolving field and that conclusions may well change with emerging data.3 We consider the statements made in the considerations article1 represent a fair expression of our assessment of the data available to us. Not everyone in the group agreed. In publishing the position statement with the list of participants who endorsed it and those who did not, and by adding an article on the process we adopted, we hoped to highlight the fact that there are controversies and uncertainties about aspects of CSIs which require careful consideration in clinical use and further high quality data to resolve currently unresolvable issues.

John P Edmonds · Richard O Day · James V Bertouch

To the Editor: Advocating implementation of evidence-based clinical practice guidelines is one aspect of the current drive to provide quality healthcare across different centres. Quality theory demands that outcomes are continuously sought and that practices are modified accordingly — the "quality loop". Therefore, Scott and Harper are to be applauded for their pursuit of improved outcomes, not just improved processes, in studying guideline-discordant care in acute myocardial infarction.1 I believe that this type of study, which objectively demonstrates the role of practice guidelines in "real world" practice, is very important. However, as a geriatrician, my patient population is unlikely to intersect with populations enrolled in large cardiology trials (eg, those for thrombolysis in myocardial infarction).2,3 Comorbidities, such as renal impairment, cognitive impairment and poor functional status at baseline, were not explicit exclusion criteria, but, when present, would have reduced an individual's chance of being enrolled. These types of comorbidities are likely to be associated with a reluctance on the part of patients and physicians to pursue life-prolonging interventions. They are also likely to be associated with poorer outcomes, whatever the intervention. Therefore, I believe that these non-cardiac comorbidities are potential confounders for study designs, such as that of Scott and Harper.1 Older age per se has been well studied in the cardiology literature on management of myocardial infarction. However, in the literature on adherence to guidelines, few studies have attempted to fully identify the non-cardiac-related characteristics of those receiving guideline-discordant care. Krumholz et al reported that altered mental state is one factor, and that, of a large "real-world" cohort aged 65 or more, only 8% were considered ideal candidates for thrombolytic therapy.4 Quality healthcare involves multiple dimensions, including both personal and process factors. Practice guidelines are valuable tools to reduce practice variation, but we need to continue to evaluate whether they can be applied as broadly as may be advocated. Surely, evidence-based guidelines can only be confidently applied to situations for which an evidence base exists. It will be important to test the application of guidelines in many settings, with attention to potential confounders, and, in particular, to outcome measures.

Kristen J Pearson

In reply: We thank Pearson for her kind comments and agree the design of our study1 prevented identification of all patient factors that may, quite reasonably, impact on clinicians' decisions to administer specific treatments to older patients with acute myocardial infarction (AMI). These factors may also have precluded such patients from enrolment in clinical trials, the results of which underpin recommendations within clinical practice guidelines. On the other hand, we know advancing age is an independent predictor of increased mortality after AMI, with several possible causes: age-related reductions in protective mechanisms (such as myocardial preconditioning),2 presence of cardiac and non-cardiac comorbidities unaffected by treatments for AMI,3 and — the focus of our study — underuse of effective therapies in the absence of discernible contraindications.4,5 While cognitive impairment, renal dysfunction and poor functional status may dissuade patients and/or clinicians from pursuing "aggressive" management, we have no evidence that these factors, singly or in combination, necessarily attenuate the benefits of specific interventions for AMI in patients at high baseline risk of cardiac death.6 We also adjusted mortality comparisons between concordant- and discordant-care groups for multiple measures of illness severity at presentation which predict a poor prognosis. Nevertheless, we support calls for more randomised trials of treatments for AMI and other conditions in older patients with liberal, "real-world" inclusion criteria in determining absolute risks and benefits of intervention in the presence of multiple comorbidities and impaired function.

Ian A Scott · Catherine M Harper

To the Editor: Several studies have documented the high incidence of adverse events arising during hospital admission. The potential for discontinuity of care and poor communication is significant when patients are admitted to and discharged from hospitals, hence the Australian Pharmaceutical Advisory Council (APAC) has established guidelines to ensure continuity in the quality use of medicines.1 A study reported in 2001 by Mant et al found very low compliance with a minimum dataset based on the APAC guidelines.2 These authors subsequently held workshops to identify problems, develop action plans and refine these strategies. However, the follow-up report, published recently in the Journal, reported little change in adherence to the minimum dataset.3 Why are providers failing to follow the APAC guidelines? Certainly, one cannot assume that the formulation and dissemination of guidelines will necessarily lead to their implementation.4 To be effective, users must be aware of guidelines and convinced that they will add value to the way in which they work. Guidelines need to be credible and should make sense in the "real world". Given the attitudinal barriers of some groups to the uptake of guidelines, multiple strategies are required to ensure their effective implementation. Among these is the involvement of key stakeholders in guideline development. Who are the key stakeholders for ensuring continuity of care regarding therapeutics between hospital and the community? While Mant and colleagues report workshops involving general practitioners and hospital staff, their reports do not identify which hospital staff were involved.2,3 Were clerical, pharmacy and junior medical staff included? These staff could make a critical difference in adherence to the minimum dataset. Furthermore, are these staff even aware of the APAC guidelines? The APAC guidelines use the definition of discharge planning established by the Council on the Ageing (Victoria). This describes people, hospitals and community-based services working together — but the guidelines and associated minimum dataset place little importance on the patient. Patients' knowledge of their medications is discounted. Despite being mentioned in principles 4 and 6 of the APAC guidelines, patient knowledge of medication changes and satisfaction with the communication regarding medications is not considered in the minimum dataset.1 Strategies involving consumers should be explored as a mechanism for improving information exchange between hospitals and GPs. Similarly, an enhanced role for pharmacists warrants further consideration.5 Certainly, further critique of the APAC guidelines and exploration of reasons for their poor uptake is important to ensure optimal patient outcomes.

Michael Jefford · Joanne L Clancy · Sharon M Butler

To the Editor: It is refreshing to see quality initiatives like that of Mant et al,1 which examine issues of continuity across different territories. Use of a simple audit tool (minimum dataset) and methodology has worked well to illuminate what misleadingly appears to be a simple problem (ie, the two-way exchange of information between the hospital and general practitioner in relation to medication). Unfortunately, like many problems that appear straightforward and easily fixable, the reality is that this issue is far more complex to address. A lack of clinical governance has been made apparent in both the public hospitals and the Divisions of General Practice that participated. Clinical governance demands that organisations be accountable for standards and performance in relation to clinical care,2 and this is integral to addressing problems both internally and across the continuum of care. Mant et al demonstrated that many hospitals had policies and strategies to accomplish the exchange of medication information,1 indicating that these procedures were thought to be achievable with current resources. Before this study, knowledge among staff of implementation was scant and confused, and there was no system of review to reveal existing problems. When problems were exposed and changes made, staff were not upskilled to incorporate the change into their practice (eg, junior doctors not completing the new discharge referral form). Similarly, the Divisions of General Practice did not resolve issues surrounding the production and distribution of business cards that they had agreed to undertake. This study has determined a means to measure performance and has intervened to analyse problems and yield some improvements. However, if the organisations involved do not commit to develop clinical governance, the path towards continued improvement will be extremely slow (if at all) and the findings of this project made irrelevant.

Catherine L Wilson

To the Editor: Mant et al1 explore an important aspect of the quality use of medicines in their study on the continuity of medicines from hospital to community. Their study on compliance with an agreed minimum dataset for patient medication information exchange between hospitals and general practitioners provides a useful perspective of an approach to systems change. I wish to point out a number of limitations that may have affected their results and make some suggestions to improve the quality use of medicines. GPs were audited on whether they provided medication information to hospitals. Many GPs work part-time. There is the possibility that the medical practice was contacted by a hospital employee, who obtained the information from a doctor other than the patient's usual GP. The audit covered discharge summaries received by the GP by fax. Although faxing discharge summaries is convenient, there are potential problems with this method. There are the possibilities of dialling a wrong number, and faxed discharge summaries (particularly handwritten ones) may be difficult to read, which could also result in medication errors. In addition, a discharge summary may have been posted to the practice instead of faxed, which would under-report the true percentage of GPs who received the information. It is not uncommon for patients to have multiple GPs.2 However, it is my experience that only one GP is documented in the patient's medical file. This issue could have influenced the results of the GP audit and would be a further factor complicating the continuity of medicines from hospital to the community. The authors mention the introduction of GP liaison officers to facilitate the notification of GPs about patient admissions and the rationale for medication changes. They do not report any other measures that they plan to introduce to improve their results. Given that systems problems have multifaceted answers, further expansion on what other steps could be taken would have been a useful addition to their article. I suggest that it would have been appropriate to include a broader range of key stakeholders in the workshops, such as community pharmacists and patients. In addition, a computerised hospital prescribing system could be integrated with an on-line evidence-based clinical guide to prescribing to assist in optimal medication selection. This could also be used to generate a discharge medication list that was automatically sent to the patients' GPs. Such an approach would reduce errors and improve outcomes.3,4

Peter W New

In reply: We were pleased at the number of letters received exploring issues raised by our recent article.1 Jefford and colleagues question whether staff are aware of the Australian Pharmaceutical Advisory Council guidelines.2 We expect to answer this question during the consultancy that the NSW Therapeutic Assessment Group is conducting for the Commonwealth Department of Health and Ageing. In this consultancy, we are evaluating the implementation and effectiveness of these guidelines. After its completion in October this year, we will have an implementation map of activities being undertaken relevant to the guidelines in Australia. Recommendations formalised at a national workshop will be made to the Department of Health and Ageing regarding future implementation, including alternative models and workable solutions. All three letters raise questions about the key stakeholders for continuity of care. We invited a range of clinical and administrative staff, including senior and junior medical staff, nurse clinicians, clinical pharmacists and general practitioners from the relevant Divisions. However, participation of junior medical staff was limited. As Jefford et al comment, consumers and community pharmacists warrant further consideration in strategies for improving information exchange between hospitals and GPs. We also agree that patients' knowledge of their medications is often deficient: with limited resources, we were only able to address this through the GPs who interviewed their patients following discharge. Wilson rightly observes that continued improvement is dependent on organisations' developing clinical governance — accountability is a key driver for change. Change in practice is usually incremental, however, as we found in our study. Sustained change requires ongoing effort and support. New comments on practical problems with faxing discharge summaries. In our study, GPs identified faxing as preferable because of problems experienced with summaries posted or delivered by patients. New also refers to the problem of patients having multiple GPs. It is up to the patient to advise the hospital appropriately, and this is a matter for consumer education. His suggestion of computerised hospital prescribing is an ideal we all hope will come to fruition sooner rather than later. In the meantime, we believe we have demonstrated the quality improvement process and its limitations and the value of GP audit in prompting that process.

Andrea Mant · Karen I Kaye · Linda Kehoe · Wendy C Rotem

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

Breathing easy New research into air pollution confirms that it is associated with cardiorespiratory death and shows that, at least in one urban environment, reducing air pollution will save lives. In a 5000-person subset of the Netherlands Cohort Study on Diet and Cancer, exposure to traffic-related air pollution (regional and urban background concentrations of black smoke and nitrogen dioxide, and proximity of residence to a major road) was correlated with mortality between 1986 and 1994.1 Cardiorespiratory death was associated with living near a major road (RR, 1.95), but not with background pollution levels. A second study, in Dublin, Ireland, compared pollution levels and daily death rates in the two six-year periods before and after the sale and burning of coal was banned in the city.2 The ban was instituted in 1990 and led to an immediate and permanent reduction in average monthly particulate concentrations. Average black smoke concentrations declined by 70% in the second period. After adjustment for population changes, weather, respiratory epidemics and secular changes in death rates in the rest of Ireland, overall non-trauma death rates decreased by 5.7%, respiratory deaths by 15.5% and cardiovascular deaths by 10.3%. This amounted to about 116 fewer respiratory deaths and 243 fewer cardiovascular deaths per year after the ban. 1. Lancet 2002; 360: 1203-1209 2. Lancet 2002; 360: 1210-1214 Statin myopathy Muscular pain is common in older people, and so is statin therapy. Inevitably, many patients will have both, but might they be linked more often than we have suspected? A preliminary report from researchers working on the IMPOSTER (Is Mypoathy Part of Statin Therapy?) trial has found evidence of statin-associated myopathy in four patients with normal serum creatine kinase levels.1 The four were among the first 20 patients recruited for an ongoing double-blind crossover trial designed to investigate this issue. All had developed muscle pain while taking statins, which resolved during a two-week washout period and recurred within 48 hours of blind rechallenge. Muscle biopsies while on statins revealed evidence of mitochondrial dysfunction, including increased lipid stores, fibres that did not stain for cytochrome oxidase activity, and ragged red fibres, abnormalities which reversed in the three patients who had repeat biopsies after ceasing statin therapy. An accompanying editorial 2 stressed that the findings were clearly preliminary, but that, given the millions of people who are prescribed lifelong statin therapy (with clear cardiovascular benefit), further study of low-grade myopathy in patients taking the drugs is justified. 1. Ann Intern Med 2002; 137: 581-585 2. Ann Intern Med 2002; 137: 617-618 All due care According to research conducted in the United Kingdom, it is not a good idea to discharge patients who present with deliberate self-poisoning home from hospital without a formal psychosocial assessment. A prospective study conducted in six hospitals in northwest England identified 604 people who deliberately poisoned themselves over an eight-week period, of whom 246 received a psychosocial assessment before discharge. Only 10% of these, compared with 18% of the 358 patients who did not receive an assessment, presented to hospital with self-poisoning again over the next 12 weeks. After adjustment for known risk factors for repeated self-poisoning, such as previous self-poisoning, psychiatric history and substance dependence, patients who had not been assessed were still more than twice as likely to present with another episode as those who had been assessed. BMJ 2002; 325: 866-867 Baubles and bacteria In these days of imaginative body piercing, it is interesting to note the results of a British study, which suggests that the practice of removing all jewellery before involvement in surgical procedures may, like many time-held rituals, do more harm than good! Volunteers with finger rings, pierced ears or nose piercings (or a combination of these) were swabbed on the jewellery surface, immediately behind the jewellery after its removal, and on an adjacent area of skin (20 examples of each type of jewellery). After culture the lowest bacterial counts were found on the adjacent (control) areas of skin, with intermediate counts on the jewellery itself, and the highest counts on the skin underneath the jewellery. The researchers suggested that nose-rings and earrings should be worn in theatre by staff who usually do so, and covered with masks and hats. J Hosp Infect 2002; 52: 68-70