Volume 182 Issue 2

A national rollout of faecal occult-blood screening, federally funded, is the best approach Bowel cancer is Australia’s commonest internal cancer.1 There is indisputable evidence that population screening with faecal occult-blood testing (FOBT), allowing early detection of cancer and detection and removal of the precursor adenomatous polyp, could save close to 2000 lives each year.2 The federal government is to be commended on its orderly approach to the issue through the Bowel Cancer Screening Pilot Programme (www.cancerscreening.gov.au/bowel/). It is important that this commitment to colorectal cancer screening continues, given also its clear cost-effectiveness.3-6 But who pays? There is no Medicare rebate for almost all screening in Australia, and definitely not for bowel cancer. But what about other methods of screening? Arguments centre on whether we need evidence from meta-analyses of multiple randomised-controlled trials (RCTs) of screening showing mortality reduction before recommending a particular method, or whether less rigorous proof will suffice. Colonoscopy has the highest level of sensitivity and specificity for detection of colorectal neoplasia, but there are no RCTs of screening using colonoscopy, let alone any meta-analyses. Because the FOBT trials have shown a link between a favourable shift in staging and mortality reduction in populations invited to participate in screening, the standard of proof for considering any screening program to be effective can now be the less stringent demonstration of a favourable shift in staging compared with controls. But, without controlled trials, even that information is unavailable for colonoscopy. Colonoscopy is not without risk, with rates of postpolypectomy transfusion requirement, perforation and death being 1 : 500, 1 : 1000, and 1 : 10 000, respectively.7 These complication figures may be too high when applied to the more robust screening population, but they nevertheless underpin the need to be careful before advocating an invasive screening procedure for a healthy population. Flexible sigmoidoscopy is another option, supported by level III evidence and some cost-effectiveness calculations. Controlled trials of flexible sigmoidoscopy are ongoing. So what about computed tomography (CT) colonography (virtual colonoscopy)? Great expectations have been generated by the lack of need for sedation, a low complication rate, short examination time, safety (apart from radiation8), and potential (not yet actual) avoidance of the need for bowel preparation. In the best centres, the sensitivity and specificity for neoplasia detection is equal to conventional colonoscopy, and it is cheaper.9,10 However, once it is positioned beyond its dedicated pioneers, the performance becomes less certain.11 The Royal Australian and New Zealand College of Radiologists has reservations about its widespread implementation for screening (Clinical Associate Professor Richard Mendelson, Colorectal Cancer Reference Group Member, RANZCR, personal communication). For best results, the hardware needs to be advanced (eg, 16-detector spiral scanners), the software optimised (providing three-dimensional endoluminal “fly through” views), scans obtained in supine and prone position, and the radiologists skilled and experienced.12 Conventional colonoscopy is needed to confirm and remove detected lesions (in an Australian study, 27% of participants needed colonoscopy13). Managing the small polyps found by CT colonography, which many would consider an incidental finding of minimal risk, inflates the necessity for colonoscopy both immediately and at follow-up. The chance that two bowel preparations will be required is unpalatable, partly explaining the lack of preference for virtual over actual colonoscopy.14 Having both procedures after one bowel preparation is possible but difficult to organise. Finally, there is not level I, II or even III evidence for cancer mortality reduction or stage shift with CT colonography. Determining an individual’s best screening strategy involves assessing familial and personal risk factors, and age-specific risk for colorectal cancer for the 5- to 10-year period over which colonoscopy affords protection against the risks of screening. FOBT is advocated in average-risk people aged 50–75 years, based on RCT evidence, validating taxpayer funding. Whether the uncertainty (with respect to risks versus benefits) of more invasive screening is acceptable becomes an individual choice. The “What would you do, Doc?” question, which is often personality rather than evidence driven, may tip the balance. In the United States, colorectal cancer screening guidelines emphasise “choice”15 — “Just do something”. Participation in population screening is very important, but whether offering “choice” improves rather than paralyses participation is uncertain. But who pays? There is no Medicare rebate for almost all screening in Australia, and definitely not for bowel cancer. Indeed, the Medicare-rebatable FOBT strategy is inappropriate for screening for colorectal cancer. It specifies a guaiac and an immunochemical test, a combination that has unknown performance characteristics in screening. For average-risk people, there is certainly no Medicare rebate for any of the more invasive endoscopic or CT screening techniques. All this information needs to be considered in the informed-consent process. So what would you do, Doc? A well organised (rather than once-only) screening program is important. A national rollout of faecal occult blood screening, federally funded, is the commendable approach being tested in the pilot program. But beyond or outside that? I would choose one of the two tests (Bayer “Magstream” or Enterix “!nform”) used in the national pilot program — both have adequate performance characteristics for bowel cancer screening — favouring perhaps the Australian “!nform” test because of its associated program of implementation, and ready applicability in general practice.16 Virtual colonoscopy? Not yet, and certainly not until I have identified a neighbourhood CT facility with performance characteristics equal to the best published to date.8 Self-funded colonoscopy? No, and certainly not before my risk for colorectal cancer death over 5–10 years overtakes my risk of serious complications from colonoscopy by an order of magnitude — that is, at age 55–60 years.2 And, for any colonoscopy required in the screening pathway, I would choose a colonoscopist with a good performance “score card” and a licensed centre.

Finlay A Macrae MD, FRACP, FRCP

The medical community has important roles in reducing harm from alcohol and other drugs A contemporary vision of how to prevent harmful alcohol and drug use is emerging, at a time when a new approach is vitally needed. In 1998 (the most recent year for which mortality data on all recreational substances are available), substance use killed some 23 000 Australians.1 Licit drug use accounted for 96% of these deaths, with tobacco the leading cause. In the same year, drug use cost the Australian community $34.7 billion, representing almost 2% of GDP for alcohol, 1.71% for tobacco and 1.76% for illicit drugs.2 Rates of tobacco and alcohol use have increased over the past decade among adolescents and young adults. Although the problem is of large scale,3 there have been major recent advances in the understanding of how to prevent much harmful drug use. One such advance is the “developmental pathways” approach, emphasised in Australian mental health4 and crime prevention5 strategies. This approach draws on life-course development research, community epidemiology and preventive intervention trials.6,7 Studies have demonstrated that from early in life similar developmental, social-risk and protective factors lead to a range of problem and risk behaviours in adolescence and young adulthood, including problematic substance use.8 Attention to these underlying factors is an essential element in preventing problematic substance use. Hence, we need to consider how these forms of prevention can be integrated into Australia’s existing harm-minimisation framework. Recent evidence also warrants an increased acknowledgement of the significant and influential role regulation and legislation play in prevention,9 including the symbolic role of law in reinforcing social norms against harmful drug use.10 The challenge is to integrate this new knowledge while accepting that there have also been clear advances through the use of harm-reduction strategies for people who are unable or unwilling to abstain from risky drug use .11 In recognition of the need for these different approaches to prevention to be integrated into national drug policy, the Australian Department of Health and Ageing commissioned a major review of Australian and international literature. The review was recently published as a monograph, The prevention of substance use, risk and harm in Australia.12,13 As part of the focus of integrating different prevention approaches, 159 preventive interventions were reviewed. The highest level of evidence for effectiveness was found for eight interventions (Box). What can the medical profession do to assist?Interventions for families and adolescentsA number of effective interventions for families and adolescents are implemented predominantly by healthcare professionals: Antenatal and postnatal home visiting by nurses to support high-risk parents in effectively meeting the child’s basic needs and to encourage healthy bonding; Early identification of fetuses or infants at risk of manifesting the effects of drug exposure, including early intervention to encourage reduction of harmful maternal drug use, particularly smoking; Child development support for families with problems associated with alcohol and drug use; Assistance for parents and families in developing skills and gaining support to enhance healthy child development and prevent substance use beginning at an early age or occurring regularly during adolescence; Identification of training and evaluation strategies to improve the preventive screening and health promotion offered to adolescents by primary healthcare professionals. Interventions within the general communityAs well as specific medical interventions, the medical profession plays an important role in supporting the development of evidence-based alcohol and other drug policy.14,15 Evidence attests to the value of interventions in the general community which prevent the sale of tobacco to minors,16 encourage responsible alcohol marketing and distribution,17 integrate treatment and harm reduction services18 and reduce the availability of illicit drugs.19 Behavioural risk factors for a variety of health issues can be managed in general practice using initiatives such the Smoking, Nutrition, Alcohol and Physical activity (SNAP) Framework to address cardiovascular health.20 Brief interventions by general practitioners appear effective for reducing both smoking and early-stage alcohol problems.21 Despite this, GP uptake of brief interventions has been poor, and many GPs fail to detect individuals at risk of developing alcohol and other drug problems.22 Professional support for GPs can improve rates of screening and brief interventions. Practice nurses should also be considered as alternative service delivery agents.22 There is a solid research base to show that treatment for a range of drug and alcohol problems is effective and can improve mental and physical health and social functioning. Treatment is an essential aspect of prevention, having population-level effects on levels of crime and disorder. Treatment of families minimises the intergenerational transmission of substance-use problems. However, most treatment programs engage only a small proportion of the people with drug and alcohol dependence. Including advice from a GP, only one in three people with an alcohol problem will receive any kind of treatment from a healthcare professional in a 12-month period.23 ConclusionAn integrated vision of prevention brings together action from many areas, including health, with a common goal of creating healthy social environments. Healthcare providers play a critical role and are encouraged to see the provision of services to drug- and alcohol-dependent individuals as a core responsibility.24,25 The evidence supports an increase in the capacity of mainstream healthcare providers to provide brief and early interventions and treatment. Further funding for drug-dependency services, training for healthcare practitioners in managing drug-dependent patients, improved access for GPs to specialist support, and the recognition of medical practice in the drug-dependency field as a legitimate medical specialty have all been recommended.24 GPs are ideally placed to identify children at risk of developing psychosocial problems because of their family backgrounds, particularly where adults and children present with problems associated with substance use by parents. Early identification of these children and appropriate treatment or referral of both the child and the parent may help to prevent the intergenerational transfer of alcohol and drug problems within families. Finally, advocacy by the medical and healthcare professions for effective non-medical interventions such as taxation, law enforcement and harm reduction is vital to ensure their wider and more effective application. Eight interventions with the highest level of evidence to prevent harms associated with substance use Tobacco taxation to create and maintain price disincentives Enforcement of environmental tobacco smoke regulations Alcohol taxation based on alcohol content of drinks Random breath testing of drivers Brief interventions by primary healthcare providers in relation to alcohol and tobacco use* Treatment for dependent alcohol and other drug use* Needle and syringe distribution programs Hepatitis B vaccination* * There is a role for the medical profession in these three interventions.

Wendy M Loxley BA(Hons), M.Psych, PhD · John W Toumbourou BA(Hons), MA, PhD · Timothy R Stockwell MA(Oxon), MSc, PhD

It is truly a bold concept — to recruit 1% of the UK population into a massive cohort study — but, 1000 days into the project, not a single participant had been enrolled Over 5 years ago leading minds within the Wellcome Trust and the Medical Research Council (MRC) realised that charting the entire genome of one human would do nothing to improve health. They argued that the real meaning and significance of genes could only be unravelled by also studying their owners’ behaviours and environments to determine which combinations conferred resilience and which resulted in disease. And so the idea of the UK Biobank was born. UK Biobank . . . seems baffled by its own complexity, stranded in a bunkum of companies, contracts and consortia that makes decoding the double helix look like child’s play It is truly a bold concept — to enrol 1% of the United Kingdom’s population, or around 3% of people in the target age-group 45–69 years, into a massive cohort study in which genetic material is held for all participants. With 500 000 participants, the UK Biobank cohort stands to be 50% larger than the set of men originally screened for the Multiple Risk Factor Intervention Trial in the United States, twice the size of the cohort for the combined US Nurses Health Studies, 12 times bigger than that of the study of smoking in British doctors, and 100 times larger than that of the famous Framingham Heart Study. Indeed, the driving force behind UK Biobank is a desire to have adequate statistical power to study gene–environment interactions for individual types of cancer, bearing in mind that, in the UK, all types of breast cancer combined account for about 4% of all deaths in women, and all types of lung cancer collectively cause about 7% of deaths in men. Despite this, there is not yet any commitment to go beyond “risk factorology” to answer questions of public health importance. For example, Biobank would be an invaluable opportunity to explore the relationship between individual characteristics and either the contextual influences on health, such as the physical and social nature of one’s neighbourhood, or the impact of health and social policy programs. Between them, Wellcome, the MRC and the UK Department of Health have made £61 million available to UK Biobank. But, 1000 days into the project, not a single participant had been enrolled, even into a pilot study. Instead, time and energy has been consumed in: creating UK Biobank Limited as a private company, limited by guarantee, and as a registered charity; having consortia of universities (mainly their medical schools) bid to be regional collaborating centres, and then drafting and redrafting contracts for them to provide “research services” to the private company; recruiting a Chief Executive Officer, a Chief Scientific Officer, a Chief Operating Officer, a Director of Operations (Laboratory), a Director of Clinical Operations, a Head of Communications, and a Chief Information Officer; commissioning a market research company to report on attitudes of people in later middle age to inform a Communications and Consultation Strategy; developing an Ethics and Governance Framework, compliance with which is to be overseen by an Ethics and Governance Council, whose members will be recommended to the funders by an appointments committee; writing an Intellectual Property and Access Policy; and asking a marketing agency to develop a logo and brand. There have also been protracted debates about which behavioural, lifestyle and environmental factors to document, and whether the collaborating centres that enrol participants should have any preferential access to the data and samples held by UK Biobank, let alone a share in their formal ownership. As yet, there has been very little consideration as to how endpoints of interest will be identified and validated. Sir Humphrey Appleby would be proud! On the subject of public image, perhaps it is the picture of a happy granny riding a moped without a helmet, shown in the original documents with general information about the project, that crystallises a suspicion that people with world-class expertise in the “molecules of life” do not necessarily have a strong grasp of public health and the skills required to enrol and follow up very large numbers of intact, free-living humans. Also interesting is the stark contrast between UK Biobank and Australia’s Risk Factor Prevalence Study. The latter was initiated by a non-government organisation, the National Heart Foundation (NHF), which succeeded in completing three large, population-based surveys of risk factors for cardiovascular disease, mainly in Australia’s capital cities, during the 1980s.1 Enrolment involved completion of a questionnaire, a brief physical examination, collection of a blood sample and sometimes a dietary survey, directly equivalent to what participation in UK Biobank is likely to involve. The survey centres had their costs covered, as is proposed for the UK project, but, unlike UK Biobank, their leaders made up the main committee overseeing the study and directed the principal analyses and their publication, initially under the aegis of the NHF alone, and in the last survey in collaboration with the Australian Institute of Health and Welfare. Each survey took about a year to plan, a year to complete and two further years to publish. Covering slightly under 0.1% of Australia’s population, the Risk Factor Prevalence Study was an order of magnitude less ambitious than UK Biobank, but it constitutes a model of energy, trust, efficiency and goodwill that the Brits are struggling to emulate. It was not always thus, for Richard Doll and Richard Peto (and colleagues) recently published the 50-year results from their study of British doctors, a project whose outstanding achievements include a follow-up that is 99% complete.2 UK Biobank, by contrast, seems baffled by its own complexity, stranded in a bunkum of companies, contracts and consortia that makes decoding the double helix look like child’s play.

Konrad Jamrozik DPhil, FAFPHM, MFPH · David P Weller MPH, PhD, FRACGP, FAFPHM · Richard F Heller MD, FRCP, FRACP, FAFPHM

Objective: To compare attitudes of Australian physicians and paediatricians about treatment and randomised controlled trial (RCT) participation.Design and participants: A cross-sectional survey using the validated “Physician Orientation Profile” (POP), with 250 physicians and 250 paediatricians surveyed.Outcome measures: Five indices — primary allegiance, decision making under uncertainty, professional activities, perceived rewards, and peer-group influence — with scores for each participant ranging along a continuum from clinician-oriented to research-oriented and expressed as a number between 0 and 1.Results: Overall response rate was 60%, with 135 physicians (54%) and 165 paediatricians (66%) responding. Paediatricians and physicians were similar in their attitudes to RCT participation, being generally clinician-oriented rather than research-oriented and less inclined to participate in RCTs when there is uncertainty about the best treatment. Most assign limited time to research, with 26.9% not currently involved in research and 31.5% having no experience of RCT participation. Doctors perceive few rewards and little peer-group influence regarding trial participation. Independent predictors of favourable attitudes to trial participation (based on POP scores) were the presence of allocated research time (0.37 for no allocated research time v 0.61 for > 70% research time; P < 0.0001), previous experience enrolling a patient in an RCT (0.40 for no experience v 0.46 for experience; P < 0.0001), and articles published in the past 12 months (0.40 for no publications v 0.55 for > 3 publications; P < 0.0001).Conclusions: This study highlights the minor importance of research for most Australian physicians. Research plays only a small role in their professional activities, and the importance of research participation is not recognised. They are clinician-oriented in their attitudes to RCT participation. To encourage greater involvement in trials among physicians in Australia, clinical research needs to be restructured in a primarily clinically oriented setting with dedicated research time.

Patrina H Y Caldwell FRACP, PhD · Jonathan C Craig FRACP, PhD · Phyllis N Butow MClinPsych, PhD

Objective: To determine the extent to which the Framingham function predicts the risk of coronary heart disease (CHD) in Aboriginal people.Design and setting: Cohort study in an Aboriginal community in the Northern Territory.Participants: 687 Aboriginal people aged 20–74 years were followed up from a baseline examination in 1992–1995 through to 31 December 2003.Main outcome measure: First CHD events were identified through hospital and death records during the follow-up period.Methods: An original Framingham function was used to predict CHD risk according to the duration of follow-up and the values of traditional risk factors, which included age, sex, total cholesterol level, high-density lipoprotein (HDL) cholesterol level, blood pressure, the presence of diabetes, and smoking status. The predicted CHD incidence using the Framingham function was 4.4 per 1000 person-years, while the observed incidence was 11.0 (95% CI, 8.7–13.9) per 1000 person-years. The observed number of CHD events (68) was 2.5 times the number predicted (27) using the Framingham function. The observed incidence was about four and three times the predicted incidence for age groups < 35 and 35–44 years, respectively, and about twice the predicted incidence for those over 45 years of age. The Framingham function was a particularly unreliable predictor for women, especially younger women, in whom the observed CHD rate was 30 times the predicted rate.Conclusions: The Framingham function substantially underestimates the actual risk of CHD observed in Aboriginal people in a remote community, especially for women and younger adults. This implies that traditional risk factors have different degrees of impact and/or that other factors are contributing to risk. A population-specific risk function is needed.

Zhiqiang Wang PhD, MSc, MB · Wendy E Hoy MB BS, BScMed, FRACP

Objective: To identify and explore behavioural characteristics of registrars that interns find helpful in their working relationships and workplace learning.Design, setting and participants: Semistructured interviews with 18 interns at Nepean Hospital, Penrith, NSW, at the end of their first working year as doctors. The survey was conducted between December 2003 and February 2004.Main outcome measure: Desirable and undesirable behavioural characteristics in registrars, as reported by interns.Results: Overall, interns’ opinions of registrars were positive. Desirable characteristics in registrars included approachability, availability, good communication skills, and a willingness to teach. Undesirable characteristics included an unwillingness to listen, unreasonably high expectations, a condescending attitude, apathy and rudeness.Conclusion: The behavioural characteristics of registrars that interns find helpful are identifiable, and there is significant room for improvement in the quality of clinical mentoring by registrars. The next step is to facilitate regular feedback from interns on registrars’ performance, and to develop ways to encourage desirable behaviours in registrars while actively discouraging undesirable behaviours.

Christopher S Lack BA, BM · John A Cartmill MB BS, BSc(Med), FRACS

We report an outbreak of a “rash” syndrome in patients attending methadone clinics in New South Wales. It presents with a pruritic, exanthematous or purpuric rash involving the trunk, limbs, palms and soles, which develops over a week and proceeds in most patients to desquamation (mainly of palms and soles) persisting for 3–4 weeks. Mucosae are not involved, and patients are generally systemically well. To date, the rash has affected 22% of 316 patients attending one methadone clinic in western Sydney, as well as patients in clinics elsewhere in Sydney and rural NSW. The aetiology is as yet unknown. We report an outbreak of a “rash” syndrome in patients attending a number of methadone clinics across New South Wales during October and November 2004. The syndrome first came to our attention when, over a week, two patients presented to a methadone clinic in western Sydney and three to the Westmead Hospital emergency department with a distinctive rash. Subsequent enquiries and patient surveillance revealed that 70 of 316 patients (22%) at the methadone clinic had developed a similar “rash” syndrome in October and November. All were prescribed methadone syrup. Clusters of patients have also been increasingly reported at other methadone clinics across metropolitan Sydney and some regional and rural areas in NSW. To date, informal communication with interstate methadone clinics has identified small numbers of patients with the “rash” syndrome outside NSW. In the first western Sydney case reliably identified by history, symptoms developed in August 2004. The principal features of the “rash” syndrome are a pruritic, exanthematous or purpuric rash that typically develops over 2 to 4 days on the hands, feet, trunk and lower limbs and persists for up to 7 days. It is usually followed by a desquamative phase that particularly involves the hands and feet and lasts up to 3 to 4 weeks. Some patients develop only the desquamative phase. The condition appears relatively benign, with few, if any, systemic symptoms, although the palms and soles of the feet can become painful with pressure after desquamation. In several patients, the rapidly developing purpuric nature of the presenting rash raised initial concern about meningococcal disease or a systemic vasculitic syndrome sufficient to warrant referral for specialist assessment. We describe four illustrative cases. Clinical recordsPatient 1A man aged in his 30s presented to a hospital emergency department with a 3-day history of a petechial and purpuric rash. He was an intravenous drug user who had been in a methadone treatment program for 7 years. He intermittently injected his oral methadone intravenously, most recently 24 hours before onset of the rash. This initially involved the lower limbs, but spread over 24 hours to affect the buttocks, lower back and abdomen. Associated but relatively mild symptoms included malaise and nausea for a week before rash onset, followed by sore throat, myalgia, ankle arthralgia, abdominal and chest pain and vomiting. At presentation, the patient was afebrile. Blood pressure was 120/60 mmHg, and pulse 70 bpm. A sparse petechial and purpuric rash was present on lower limbs, feet, buttocks and lower abdomen. There was no pedal oedema, joint effusion or tenderness. There were no abnormalities on respiratory and cardiovascular examination, no clinical evidence of endocarditis, no lymphadenopathy, and mucosae were normal. The right upper abdominal quadrant was tender, but the liver and spleen were not enlarged, and no renal masses were palpable. Investigations were uninformative (Box 1). Inpatient progress was unremarkable, and, 10 days after presentation, all symptoms had resolved, despite ongoing oral and intravenous methadone use. Patient 2A middle-aged man presented with a 2-day history of an erythematous, pruritic rash over his trunk and limbs which was now beginning to desquamate. He was also an intravenous drug user in a methadone treatment program. In addition to taking prescribed oral methadone, he intermittently injected both methadone and stimulants, such as amphetamine, intravenously. There was no history of fever, oropharyngeal, genital, eye or systemic symptoms. On examination, he was afebrile, looked well and had a generalised exanthem, with erythema and significant desquamation of soles and palms (Box 2, A and B). There were no oral, mucosal or eye signs, and no lymphadenopathy or hepatosplenomegaly. Results of investigations were unremarkable (Box 1). He was treated for 2 days with oral prednisolone and an antihistamine, and then discharged. The rash settled over a week, although he continued to have desquamation of the soles and palms 2 weeks later. Patient 3A young man who was an intravenous drug user in a methadone treatment program presented to the same hospital with a 2-day history of a purpuric lower-limb rash. In addition to taking prescribed oral methadone, he intermittently injected both heroin and methadone intravenously. Five days before presentation, he developed bilateral calf pain and generalised myalgia. He was initially seen at another hospital, where he was treated with broad-spectrum intravenous antibiotics for presumed sepsis. He discharged himself after 24 hours and was admitted to our hospital about 12 hours later because of his concern about the rash. On admission, he was afebrile, with blood pressure of 115/65 mmHg and pulse of 75 bpm. He had a prominent purpuric rash involving both lower limbs (Box 2C), with sparse lesions on both forearms. Mucosae were normal, and there was no meningism, lymphadenopathy, hepatosplenomegaly, joint swelling or tenderness, no abnormalities on respiratory and cardiac examination, and no stigmata of endocarditis. Results of investigations were once again unremarkable (Box 1). The patient remained well despite the rash and was discharged from hospital 24 hours after admission. Patient 4A young woman who was an intravenous drug user in a methadone treatment program presented to the methadone clinic with a 4-day history of an erythematous, pruritic rash over her trunk, limbs and hands. Other than oral methadone, she was taking no drugs and was otherwise well. Examination revealed an extensive exanthem over her trunk, hands and legs. She had no fever, and blood pressure was normal. She was reviewed a week later and still had an extensive generalised erythematous exanthem, as well as finger and palm desquamation (Box 2, D and E). Results of investigations were unremarkable (Box 1). DiscussionThe aetiology of this “rash” syndrome is yet to be elucidated. Currently, it appears to be restricted to people using methadone syrup, with no reports of rash in over 100 patients in western Sydney prescribed buprenorphine for treatment of opioid dependence, nor among non-methadone-using family members of patients with the rash, nor among healthcare workers in contact with these patients. To date, all patients with the “rash” syndrome who have been assessed for hepatitis C exposure are seropositive, but not all are viraemic. Some patients with the “rash” syndrome smoke cannabis and intermittently inject methadone or other drugs. However, these characteristics are not universal among affected patients, nor more frequent than in unaffected patients on the methadone program, among whom they are also common. Similarly, the use of prescription or complementary medicines does not seem to be associated with the “rash” syndrome. Similar rashes and associated desquamation are common in staphylococcal and streptococcal toxin-induced illnesses, such as toxic shock syndrome and scalded skin syndrome,1-3 and in some viral illnesses, such as parvovirus infection and measles.4 However, the patients in the current outbreak did not give a history of bacterial or viral illness, and family members not taking methadone do not appear to have developed the syndrome. HIV antibody testing has been performed in some affected patients and has been negative. Throat swabs taken in some patients have grown only normal respiratory flora. Markers of streptococcal infection, such as antideoxyribonuclease B antibodies and anti-streptolysin O titre, are positive in some but not all patients. Skin biopsy performed in a number of patients has failed to help define the aetiology of the rash. Histological examination often shows focal and mild spongiosis with superficial perivascular chronic inflammation, while direct immunofluorescence examination shows deposition of IgM and complement 3 in dermal capillaries. These findings are consistent with an immunological reaction in the skin, but do not clarify whether it is the primary cause of the rash or a secondary phenomenon. A hypersensitivity reaction to a contaminant in the methadone syrup could present with such a picture. The fact that, to date, all the patients identified in western Sydney had been taking methadone syrup from a single manufacturer raises the possibility of batch contamination; however, batches are distributed nationally, so more widespread involvement would probably be expected if this was the basis of the syndrome. In addition, examination of the methadone syrup has failed to detect any contamination. The possibility of alternative sources of contamination, such as methadone storage or delivery devices, remains to be explored. We believe it is important for physicians to be aware of this newly emerging syndrome, both to assist with more accurate delineation of its epidemiology and pathogenesis, and to permit more effective investigation and treatment of affected patients. State public health units and the Therapeutic Goods Administration are investigating this outbreak to try to determine the cause of this new syndrome. 1 Results of investigations in four patients with rash Investigations Reference range Patient 1 Patient 2 Patient 3 Patient 4 Full blood count and film Normal; platelet aggregates on film Normal apart from WBC 10.8 x 109/L; occasional reactive lymphocytes Normal Normal, apart from Hb 107 g/L Haemoglobin (Hb) (g/L) 115–161 White blood cell count (WBC) (x 109/L) 3.7–9.5 ESR (mm/h) 0–15 4 11 5 38 C-reactive protein (mg/L) 0–11 20 27 22 15 Liver function tests Normal Abnormal Abnormal Normal γ-Glutamyltransferase (U/L) 8–43 47 48 Alanine aminotransferase (U/L) 10–47 88 157 Aspartate aminotransferase (U/L) 12–45 104 154 ANA, ANCA, ENAs, rheumatoid factor, complement C3 and C4 Normal nd Normal nd Cryoglobulins Absent nd Detected nd Prothrombin time (s) 11–18 Normal Normal Normal nd APTT (s) 25–36 Normal 39 Normal nd Hepatitis C virus IgG-positive; undetectable viral load (< 600 IU/mL) IgG-positive; viral load not assessed IgG-positive; viral load > 850 000 IU/mL IgG-positive; refused viral load assay HIV antibody Negative nd nd nd Urinalysis Trace protein (39 mg/24 h); no casts/red cells Normal Normal nd Blood culture Negative Negative Negative nd Throat swab Nd Normal flora nd nd Electrocardiogram Normal nd nd nd Chest x-ray Normal Normal Normal nd Echocardiogram Transthoracic normal; transoesophageal not tolerated by patient nd nd nd ESR = erythrocyte sedimentation rate. nd = not done. ANA = antinuclear antibody. ANCA = antineutrophil cytoplasmic antibody. ENAs = extractable nuclear antigen antibodies. APTT = activated partial thromboplastin time. 2 Features of the rash in four patients ↑A. Generalised exanthem on trunk and limbs in Patient 2. ↑B. Palm desquamation in Patient 2. ↑C. Purpuric rash involving lower limbs, with areas of confluence on the lower calf in Patient 3. ↓D. Extensive erythematous exanthem over trunk and limbs in Patient 4. ↑E. Palm desquamation in Patient 4.

Jon N Currie FRACP, FAChAM · Jimmy Chien BMed · Lisa Snell RN · Margaret Cluff RN · Karen Scrivener RN · Lucinda Wallman PhD, FRACP, FRCPA · Elizabeth M Benson FRACP, FRCPA

A man who had been taking prescribed methadone for many years presented with a desquamating rash (predominantly affecting the hands and feet) complicated by cellulitis of the right leg. There have now been multiple reports of a similar rash among methadone users in Sydney. The cause remains unknown. We report a man attending a methadone program in south-east Sydney who presented with a distinctive rash complicated by lower-leg cellulitis. This case adds to the widespread reports of a similar rash in methadone users around Sydney. The cause is under investigation by the New South Wales Department of Health. Clinical record In October 2004, a middle-aged man was referred to a hospital in south-east Sydney with a 14-day history of a painful, swollen, erythematous right lower leg and fever. Two days before onset of the leg symptoms, he had noticed a non-pruritic rash which started on his legs and feet and spread to abdomen and arms; it was accompanied by swelling, and then desquamation of the hands and feet. He had been diagnosed with cellulitis of the right lower leg 7 days before presentation and was prescribed oral flucloxacillin, but his condition did not improve significantly. Duplex ultrasound examination of the right leg 2 days before presentation excluded deep vein thrombosis. He was admitted to hospital for intravenous antibiotic treatment. The patient was a former intravenous heroin user and had been in a methadone program at a local pharmacy for the previous 8 years. He denied recreational drug use or injecting or sharing the oral methadone. His past medical history included previous right-leg deep vein thrombosis, chronic hepatitis B and C infection, gastro-oesophageal reflux, as well as melanoma excision several years before, and cholecystectomy a month previously. He had been taking griseofulvin for about 3 months for onychomycosis. He was taking methadone syrup (160 mg daily), griseofulvin (500 mg daily) and, when required, diazepam (5 mg three times daily), codeine/paracetamol (30/500 mg three times daily) and oxycodone (40 mg twice daily). He had recently started using, when required, cyproheptadine (4 mg at night), hyoscine (20 mg four times daily) and triamcinolone acetonide (0.02% cream topically) for the rash, and metoclopramide (10 mg three times daily) for mild nausea. On physical examination, the patient was haemodynamically stable and looked well. His temperature was 37.2°C. There was erythema, tenderness and warmth below the right knee, consistent with cellulitis. There was desquamation of the skin of both his lower legs, soles of feet (Box, a) and palms and fingers (Box, b), as well as non-pitting oedema of the feet and hands. There was hyperkeratosis of the soles, with xerosis and an erythematous maculopapular rash of the arms. The skin condition did not appear typical of disorders causing hyperkeratosis and desquamation, which usually do not present concurrently. He had no mucosal ulceration. Differential diagnoses included the early phase of an exfoliative erythroderma, psoriasis triggered by infection, early pityriasis rubra pilaris, bacterial toxin-mediated exfoliation, viral exanthem, sarcoidosis and syphilis. Rash in a patient who used oral methadone A: Hyperkeratosis and desquamation of the soles of the feet. B: Desquamation of the palms. C: Biopsy specimen from the maculopapular rash on the left arm, showing a slightly thickened epidermis with focal parakeratosis (P), and mild perivascular lymphocytic infiltrate (I). (Original magnification, × 20; haematoxylin–eosin stain.) Measurement of serum electrolyte, urea and creatinine levels and liver function tests all gave normal results. A full blood count revealed haemoglobin level of 125 g/L (reference range [RR], 130–180 g/L) and slight eosinophilia (0.48 × 10 9/L; RR, 0.04–0.44 × 10 9/L), but total white cell count was within the reference range (6.2 × 109/L; RR, 3.5–11.0 × 10 9 /L). Erythrocyte sedimentation rate was 10 mm/h (RR, 1–10 mm/h) and C-reactive protein level was 24 mg/L (RR, < 3 mg/L). Blood cultures showed no growth. A test for Treponema pallidum antibody was negative. Chest x-ray was normal, and HIV antibody test negative. Skin biopsy of the red macules on his left arm (Box, c) and the right-leg biopsy revealed non-specific histological changes, suggestive of chronic dermatitis. Periodic acid–Schiff staining for fungi was negative. Empirical treatment was begun with mometasone cream (0.1%), calcipotriol ointment (0.005%) and sorbolene cream twice daily to affected areas, and coal tar (5%) and salicylic acid (5%) in sorbolene base cream at night to the feet. Intravenous cephazolin (1 g three times daily) and oral clindamycin (300 mg four times daily) were begun for the cellulitis. The patient was discharged on Day 4 with the above topical preparations and oral cephalexin (500 mg three times daily), after both the rash and right-leg cellulitis abated significantly. He failed to attend a follow-up appointment, but reported by telephone that the rash had gradually resolved over several weeks. Discussion This case raises the alert to a possible adverse reaction to a methadone preparation. The cause may be methadone itself, another component of the preparation, or a contaminant. Previously reported cutaneous reactions to methadone include angioedema, facial oedema, flushing, pruritus, purpura, rash and urticaria.1 To our knowledge, no cases have been reported of a desquamating rash associated with methadone or other opioids. The cause of the rash in our patient did not appear infectious. He had no clinical evidence of staphylococcal toxic shock syndrome; he remained clinically well and did not develop the diffuse confluent erythema typical of this syndrome. Nor was the rash typical of a viral exanthem, in which a widespread morbilliform eruption predominates, without confluent erythema (as occurred on the lower legs), hyperkeratosis or desquamation. A reaction to a medication other than methadone seems less likely, as there had been no recent change. We are aware of other patients with a similar rash, oedema and desquamation of the hands and feet, all taking methadone: six patients in a methadone program at the same pharmacy as our patient, 20 from a local methadone clinic, and others at other methadone centres in Sydney (Mary Anne Ford, Registered Nurse, Bayside Clinic, Sydney, NSW, personal communication). We know of no patient with a similar rash who is not using methadone, and we believe all affected patients were taking the same brand and formulation of the drug. Most cases have been mild, and close contacts have not been affected. All patients appear to have continued using methadone from their usual dispensing clinic, and the rash has resolved over several weeks. The cause of the reaction remains unknown, and is possibly even an illicit drug available on the street. However, as all affected patients appear to have been taking methadone, this is perhaps the most likely agent. Marijuana has been reported, albeit rarely, to cause allergic reactions, and occasionally becomes contaminated with biological or chemical substances that might cause a reaction.2 -4 However, in that situation, one would expect cases to be more widely distributed outside the methadone-using population. In addition, at least one client with this rash from the local methadone clinic had a negative urine test for cannabinoids (Mary Anne Ford, as above, personal communication). Further investigation of these cases is required to determine the aetiology. Variables to be considered include the brand and batch of methadone used, storage, dose, mixing of batches in the dispensing pump, and other solutions included in the preparation to increase palatability. Pharmacists often mix batches of methadone and are not required to record the batch number dispensed to each patient, which makes tracing difficult. Also important to consider are any other prescription or non-prescription medications taken by patients, illicit drugs used, and sharing of methadone between patients from different methadone clinics. Many more cases may have been unreported and unrecognised, as most affected patients have had relatively mild and self-limiting symptoms. An investigation is now under way by the New South Wales Department of Health. Methadone clinics, pharmacists, dermatologists, general practitioners and emergency medicine staff need to be aware of the possibility of these reactions.

Natalie Kordjian BPharm, MB BS · Annabelle D Donaldson MB ChB · Steven A Krilis PhD, FRACP · Dedee F Murrell MA, BM BCh, FAAD(USA)

In most cases, childhood leukaemia has a fetal origin, but multiple molecular events are required after birth for pre-leukaemic cells to progress to leukaemia. Cure rates for acute lymphoblastic leukaemia (ALL) now approach 80%. A high level of minimal residual disease detected by polymerase chain reaction in patients with ALL in remission has profound prognostic importance and is the focus of a major Australian study attempting to prevent relapse in these children. Greater awareness of the late effects of chemotherapy has led to changes in the treatment protocols for ALL, with improvement in neurocognitive outcomes and reduced rates of second malignancies. Pharmacogenetics is a new field of research that aims to enhance treatment efficacy by assessing the individual’s metabolism of and response to chemotherapeutic agents. Targeted therapies currently being developed show some promise of being able to further improve cure rates. Adolescents with ALL have a better prognosis if treated with paediatric rather than adult protocols.

David S Ziegler MB BS · Glenn M Marshall MB BS, FRACP · Luciano Dalla Pozza MB BS, FRACP · Keith D Waters MB BS, FRACP

Given fundamental differences in the recommendations in guidelines from major national and international committees, we cannot rely on them unquestioningly. Different antihypertensive agents are known to have differing effects according to age and race. Exchanging (rather than following guideline recommendations of adding to) an ineffective first-line antihypertensive drug can result in control of hypertension with monotherapy. Conclusions about a preferable first-line antihypertensive agent are limited by trial protocols with varying drug doses and questionable drug combinations. Guidelines are often based on meta-analyses of drugs of a particular class, which could ignore important differences between drugs within a class. Trials of 3–5 years cannot determine the long-term effects of drugs which patients often take for decades.

Roger E Peverill PhD, FRACP

QuestionCan a regimen of chewing sucrose-free gum after meals reduce the incidence of caries and remineralise white spot lesions (demineralised and non-cavitated incipient caries) in a population with a moderate incidence of caries? Trial details Design: A cluster-randomised controlled trial with two arms. Setting: Elementary schools in Budapest, Hungary, where water was not fluoridated at the time of the study. Participants: 583 schoolchildren aged 8–13 years (mean age, 9.6 years). Participants and their parents or guardians were required to sign an informed consent form. Non-participation rates and profiles were not reported, and inclusion and exclusion criteria were not specified. Interventions: The intervention group chewed sucrose-free gum (65% polyols [sorbitol and mannitol], 30% gum base and 5% sweeteners and flavours) for 20 minutes after meals, three times a day; the control group did not chew gum. Two meals a day were available at the schools, facilitating supervision of gum chewing. Other chewing sessions were unsupervised. After baseline clinical examination, classes within grade levels were randomly assigned to either arm, provided that each grade level had more than one class with sufficient participants available. No significant differences in baseline caries scores were detected between the intervention and control groups. No modifications were made to the oral hygiene and dietary practices of participants. Main outcome measures: Participants were examined at 1 and 2 years by a single blinded examiner, using a mouth mirror, explorer and transillumination to aid diagnosis of interproximal caries. Drying of teeth and radiographs were not used. Clinical examination results were expressed as World Health Organization (WHO) DMFS (decayed, missing and filled surfaces) scores or Radicke scores. WHO DMFS scores include a category for incipient carious lesions, and are commonly used caries experience indices, quantifying decayed, missing and filled surfaces. Main results: 1-year and 2-year scores for DMFS increment were adjusted for by baseline DMFS scores. Using the Radicke DMFS scores, reductions in caries increment for the intervention group were in the order of 43.6% (P = 0.008) and 38.7% (P = 0.018) at 1 and 2 years, respectively. Using the WHO DMFS scores (inclusive of incipient lesions), reductions in caries increment for the intervention group were in the order of 41.7% (P = 0.028) and 33.1% (P = 0.008) at 1 and 2 years, respectively. Conclusions: The authors concluded that chewing sucrose-free gum after meals provided a positive anti-caries effect. CommentaryRationale for the trialTo confirm previous trials reporting positive anti-caries effects of sucrose-free gum chewing.1-5 Whereas most previous trials were conducted in populations with a high incidence of caries, the authors sought to investigate within an industrialised population with moderate caries incidence. Trial methodsAs no intervention was elected for the control group, participant blinding was not feasible. Thus, the authors did not offer alternative explanations for the positive result. The intervention may have initiated different preventive oral hygiene and dietary practices between groups. Also, the intervention group may have been less likely to seek confectionery, and hence a less cariogenic diet, compared with the control group. Participant flow was poorly reported, obscuring potential sources of bias. Follow-up was excellent, with 93.8% presenting for 2-year clinical examination. However, this follow-up rate was not broken down between intervention and control groups. It is feasible that this rate may have been lower in the more procedurally demanding intervention group, and those not presenting for follow-up may represent participants with less concern for their oral health. Reporting of withdrawal after randomisation (of particular interest to the intervention group) was also poor. Mechanisms for capturing adverse events were not described. No adverse events were reported, except for one that was not related to the chewing gum, but resulted in withdrawal. There is concern that adverse events not related to the chewing gum were not adequately monitored. Compliance, especially for out-of-school chewing, is a challenge in a trial such as this. School chewing compliance was reported as unproblematic, but the method of monitoring was not described. Out-of-school chewing compliance was assessed by gum wrapper return (93% of students returned more than 90% of wrappers). This does not accurately reflect if gum was chewed and by whom. The authors recognised that gum chewing could not be accurately assessed in the control group. However, prohibition of gum chewing in schools was described. New informationThis study suggests a positive anti-caries effect of chewing sucrose-free chewing gum after meals within an industrialised population with moderate caries incidence. To confirm such an effect, the role of bias would require greater attention. Implications for clinical practiceThe authors advise that sucrose-free gum-chewing be considered on an individual and organisational level. From the perspective of a cost–benefit analysis, this is premature. The gum-chewing regimen described is expensive and needs to be considered against other preventive measures, such as fluoride gels and mouth rinses, before the authors’ recommendations can be embraced. The evidence for the superior efficacy of chewing gum sweetened with xylitol over that sweetened with sorbitol is not unanimous.3-5 Budapest may benefit more from fluoridation of its water supply.

Claudine E Tsao BDSc · Michael V Morgan BDSc, MDSc, PhD

Good study design involves minimising all possible sources of bias. Two important sources of bias arise through failure to mask (ie, conceal), first, the randomisation process and, second, the treatments after randomisation. Allocation concealment is the term used to describe the procedure for protecting the randomisation process so that the treatment to be allocated is not known before the patient is entered into the study. Blinding relates to the masking of the treatments after randomisation — from the patient, the investigator or the outcomes assessor. Without exception, allocation concealment is achievable in all randomised clinical trials. In contrast, it is not always possible to blind people to study treatments received. The CONSORT statement strongly encourages detailed reporting of the allocation concealment process and the measures taken to preserve blinding (Box 1).1 Allocation concealmentFailure to conceal the process of random allocation will potentially result in a non-randomised trial, while successful allocation concealment will reduce selection bias. No matter what method is chosen to randomly allocate patients (eg, by simple random numbers, permuted blocks or minimisation), if the investigator or clinician (or the patient) is able to identify the impending treatment allocation and is able to influence the enrolment (or selection) of participating patients, the value of randomisation is compromised. Selection bias may have been introduced, whereby the treatment assignment is no longer truly random and an imbalance in prognostic factors between treatment groups occurs. If this arises, assessment of the treatment comparison is compromised. Failed concealment from the investigator or clinicianSingle-centre trials in which randomisation is conducted on site and the randomisation method is known to the participating investigators are at high risk of this problem. If an investigator has influence over the number of patients enrolling in a trial, or the order in which they are randomised, he or she has the potential to introduce selection bias into the study by directing particular patients into preferred treatment groups while excluding others. However, even if the investigator knows the allocation sequence, the problem will not arise if all consecutive eligible patients are being enrolled in order of presentation, as is often the case in trials in emergency departments, intensive care units or any acute care setting. Ideally, a successful allocation concealment process is ensured when all investigators are ignorant of future treatment allocations and have no control over the order of patients randomised into the trial. Failed concealment from the patientPatients may change their willingness to participate in a trial if they know or suspect which treatment they will receive. If patients are aware of their allocation before randomisation, they may be influenced to withdraw before randomisation or wait until their preferred treatment is available before entering the study. Baseline imbalances as a direct result of concealment violation may not be evident across the whole trial, although patients from particular sites or patients from particular investigators may show consistent differences between treatment groups in various baseline prognostic factors. In practice, the method of allocation concealment is often not reported or is poorly described.2-7 Any trial report should provide enough detail to describe the quality of both allocation concealment and blinding strategies. The use of sequentially numbered, opaque, sealed envelopes, pharmacy-controlled allocations, coded identical containers or kits, and central randomisation systems (telephone or web based) are considered adequate concealment methods and are often implemented to protect the randomisation.8 Allocation lists that can be prepared in advance or may be guessed because of a known pattern (eg, permuted blocks of fixed size) are more susceptible to deciphering. No strategy is entirely tamper-proof, although remote systems are generally more secure.9 Blinding (masking)Blinding describes the status of the patient and/or the clinician-investigator after randomisation: Single blind: Either the patient or clinician (usually the patient) remains unaware of the treatment assignment. Double blind: Both the patient and investigator are unaware of the allocated treatment. Open label: All parties are aware of treatment being received after randomisation. Triple blind: The patient, the investigator and either those who adjudicate the study outcomes (the outcomes assessment committee) or those who monitor the study safety (the safety and data monitoring committee) are unaware of the allocated treatment. A blinded safety committee will see the data from the treatment groups in coded form (ie, labelled as X and Y), and so are also blinded to treatment allocation. Unblinding: The disclosure, planned or unintended, of the allocation of one, a group, or all of the participants. Failure or inability to blind people to the treatment assignment in a trial potentially introduces important biases. These include reporting bias (by either the patient or investigator), assessment bias (where assessment is part of an investigator’s role), and concomitant treatment bias by either the patient or investigator; all such biases contribute to differences between groups other than those resulting from the allocated study treatment. In open-label trials, all of these biases might occur. It is recommended that the study’s blinding be explicitly detailed when trial design and results are reported.1,10 In particular, authors should report how blinding was maintained for patients, investigators or clinicians and outcomes assessment committees. Consequences of treatment knowledge after randomisationIf patients are not blinded to their allocated treatment, then knowledge of treatment may influence their responses to the intervention and their reporting. Commonly, patients assume that the new intervention will be more beneficial than the control or standard treatment. Compared with clinical event outcomes, patient-rated outcomes (for example, quality of life, pain and discomfort) are particularly sensitive to patients’ knowledge of the intervention to which they have been allocated. Similarly, if investigators are aware of the patients’ study treatment, their knowledge may influence, first, their management of the patient and, second, their classification of responses and events. For example, in the Aspirin Myocardial Infarction Study,11 men and women who experienced a myocardial infarction were randomised to receive 1 g aspirin a day or matching placebo. If they or their doctors had been aware that they were taking aspirin, they might have ascribed all their gastrointestinal symptoms to aspirin. Symptoms suggesting peptic ulcer, gastritis or erosion of gastric mucosa occurred in 23.7% of the aspirin group but also 14.9% of the placebo group. Because the study was well blinded, it was possible to attribute only 8.8% of the symptoms to the aspirin treatment. Generally, drug trials involving oral, topical or intravenous administration can be set up as a double-blind study by the use of a matching placebo. Many surgical trials comparing intraoperative techniques with an identical external incision can be designed as single-blind trials, but comparisons between medical and surgical interventions are inevitably unblinded — that is, open-label trials. When a treatment has a distinctive side effect which is likely to be expressed in most patients (such as toxic effects from chemotherapy), the use of a placebo may be futile. To blind or not to blind?A double-blind design is desirable for any trial. When there is an active comparator in a drug evaluation trial, blinding can be ensured by the use of a so-called double-dummy design. This involves giving all patients two formulations: one group receives the active treatment plus a placebo of the alternative treatment, and the other group receives the opposite combination (see Box 2). To successfully mask the treatment, a placebo treatment should be identical in appearance (size, colour, weight, feel, odour, etc) and route of administration, and should be tested to make certain that its benign nature cannot be detected. Patients’ or investigators’ preconceptions about the value of the treatment may affect a trial’s results. For example, in a trial of the effect of vitamin C on symptoms of colds, volunteers took vitamin C or a placebo for 9 months, with an increase in the dose at the onset of a cold.12 Because of the differences in taste between the vitamin and the placebo, some of the participants became aware of their treatment. In this group, the vitamin C had a reported benefit, but vitamin treatment did not appear to help those who remained blinded. The breakdown of the blinding led to inconclusive results, illustrating the importance of ensuring that blinding is done with care. Masking data and intermediate outcomes during the studyWhen routine collection of clinical data during a trial (eg, blood tests) may potentially unblind investigators, providing investigators with summary information should be considered. For example, in a long-term cardiovascular trial assessing the effect of cholesterol-lowering treatment on the incidence of cardiovascular events, knowing individual lipid profiles during follow-up may unblind the clinician and patient. If the blood tests are performed by a central laboratory, it may be possible to provide a summary report simply stating that a patient’s values do or do not fall within a prespecified range, without compromising the patient’s safety, masking the individual data and preserving the blinded status of the investigator. Blinded outcome assessmentBlinded assessment of outcomes is possible in most trials, regardless of whether the clinician or the patient is aware of the treatment allocation. Clinical outcomes should be assessed by people who are not aware of the patient’s treatment allocation (and preferably not involved in the patient’s clinical management) so that all patients are assessed identically. In this way, outcomes will not be influenced by beliefs about the study treatments being compared. In general, blinding becomes less important for reducing observer or information bias as the outcomes become less subjective and more objective (that is, bias is more prevalent with a subjective outcome such as degree of pain or extent of depression, but is eliminated with an outcome such as death from any cause). For outcomes such as biochemical or pathology result markers, assessment by calibrated, accurate instruments is often sufficient to ensure no bias is present. If patients are required to provide their own assessment of an outcome, it is recommended that more than one item of the same type be reported to allow reliability to be measured. Effect of allocation concealment and blinding on the interpretation of trial resultsAn open-label trial with successful allocation concealment and blinded assessment may provide more reliable and more valid results than a double-blind trial with unsuccessful allocation concealment or compromised outcome assessment. Blinding does not guarantee an absence of bias, although empirical evidence does endorse a reduction of bias when adequate blinding strategies have been implemented. Allocation concealment (before randomisation) is thought to have a stronger influence on the reduction of bias than blinding (after randomisation).3,5 A checklist for successful allocation concealment and blinding is provided in Box 3. When and how to unblind participants and investigators during the trialProcedures should be established at the start of any randomised controlled trial for possible unblinding of the investigator and patient to an individual patient’s data during the trial. In most cases, unblinding would be carried out when the patient’s safety is at risk and this knowledge is required for emergency treatment. Often, however, if simply ceasing study treatment is a viable option for the patient’s care, it should not be necessary for unblinding to occur. Whenever possible, the chief investigator’s agreement should be sought before requests for individual unblinding are made. The process of unblinding should be such that only the data for one patient should be disclosed at any one time, and an audit trail maintained of all requests and their justifications. 1 CONSORT checklist of items to report when reporting a randomised trial1 Section and topic Item no. Descriptor 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? Blinding (masking) 11 Whether or not participants, those administering the interventions and those assessing the outcomes were blinded to group assignment. If done, how the success of blinding was evaluated. 2 Double-blind double-dummy trial design Randomised allocation Patient receives Active treatment 1: Tablet Active treatment 2: Capsule 3 Checklist for successful allocation concealment and blinding Allocation concealment Investigator-clinicians are unaware of exact details of how the chosen randomisation method is being implemented (eg, ignorant of block sizes used in a permuted block randomisation scheme). All staff responsible for providing allocations have adequate training. An audit trail is maintained to ensure the integrity of the allocation process.13 If possible, a centralised or remote randomisation service is used (either by telephone, fax, email, the internet, a coordinating centre or site pharmacy). Remote randomisation can mean telephoning or faxing another person or clinic department to receive the next allocation. Blinding Appropriateness of blinding of patients and investigators (double-blind or single-blind) is ascertained. The implementation process for blinding is planned, including deciding whether a placebo treatment is to be used, and ensuring the process maintains the relevant parties’ ignorance of the treatment after randomisation. Evaluation of outcomes ensures objective assessment of all patients. Unblinding Ceasing study medication, if this is an option, is preferable to unblinding. A method of unblinding is available quickly in a genuine emergency. Specific methods If envelopes are used, there is an audit trail recording each envelope opened (date and time) and envelopes are numbered, tamper-proof and opaque so that the contents remain concealed unless the envelope is destroyed or damaged.13 If kits or containers are used, they are identical (weight, size, appearance), numbered and tamper-proof. The allocation process is reproducible.

Peta M Forder BSc, MPH · Val J Gebski BA, MStat · Anthony C Keech FRACP, MScEpid

Philip Kessly died on 18 August 2004 in Perth, Western Australia, after a long illness. His career had been one conspicuous by his dedication to his community, family, and the art and science of medicine. The eleventh of 12 children, Phil was born in London on 22 February 1922 to Jewish migrants who had fled pogroms in their native Ukraine. He was educated in London, Merseyside and Manchester. In 1937, he took an apprenticeship in pharmacy, which he completed in 1940. During the war, Phil enlisted voluntarily in the British army, where he worked on radar. His service included a period in India, from where he returned in 1946. After demobilisation, he went to Edinburgh to study medicine, his great passion in life. It was there that he met Lilian, who later became his wife. Phil migrated with his family to Australia in 1956, where he joined a practice in Mt Hawthorn, WA. He later established a very successful practice in Yokine, in suburban Perth. He continued in general practice until 1986, when ill health forced him to seek out a quieter practice in Como, where he continued to work until shortly before his death. In addition to running a very busy general practice, he involved himself in other aspects of medicine. He was a clinical assistant in dermatology at the Royal Perth Hospital (1960–1972) and a lecturer in the Department of Social Work at the University of Western Australia (1977–1979). He realised the importance of passing on his skills, and was a foundation member of the Department of General Practice at the University of Western Australia. Phil was also actively involved in many political aspects of Australian medicine. He served on the WA Branch Council of the Australian Medical Association (AMA) (1972–1982) and was President in 1975. He was elected a Fellow of the AMA in 1979. He saw the need to recognise general practice as a special discipline, and was instrumental in establishing the WA Faculty of the Royal Australian College of General Practitioners, serving as its Provost from 1976 to 1978. His regular appearance at Beatty Park Aquatic Centre in white robe, snorkel and mask will be sadly missed, as will his sharp intellect, his encyclopaedic knowledge and his commitment and compassion in medical practice. Australian medicine was fortunate to have had the contribution and enthusiasm of such a gifted and exceptional man.

Peter M Winterton BA, DRACOG, FRACGP

Trevor Anderson was born in Camberwell, Victoria, on 20 July 1941. The son of an army officer, he attended Melbourne High School before studying medicine at the University of Melbourne (1961–1966). In 1964, he joined the army, and, after completing his residency, was assigned to the Royal Australian Regiment as a Medical Officer in 1968. On 21 July 1969, in South Vietnam, Captain Anderson flew in to assist with treatment and evacuation after a landmine had killed and injured members of a platoon. As this process ended, another mine exploded, killing a corporal and causing further casualties. Trevor was blinded and wounded in the legs and abdomen. After rehabilitation, Trevor trained in psychiatry at Royal Park Hospital, Melbourne, and the Parkville Psychiatric Unit, studying with the help of his wife Janice. In 1976, he became Psychiatrist in Charge at the Elizabeth Street Clinic, remaining for 28 years as the clinic moved to become first the Ellery Clinic, in Carlton, then the Waratah Clinic, in Moonee Ponds. There he developed community psychiatry, trained young psychiatrists and inspired medical students. He also pioneered psycho-oncology, starting a liaison service to the Peter McCallum Cancer Institute. Trevor had a profound respect for individual rights. He never lost sight of the fact that our services are there to serve patients. His opening gambit to patients, “How can I help you?”, represented the ethical core of his practice philosophy. He was described by one colleague as “magic with difficult people”. Trevor served on numerous Department of Veterans’ Affairs committees and won the 1991 Returned Servicemen’s League Anzac of the Year award. He was President of the Victorian Blinded Soldiers’ Association, served on the board of the Royal Victorian Institute for the Blind, and, as president from 1999, oversaw an amalgamation with Vision Australia and the Royal Blind Society of NSW. He was a member of the Victorian Psychological Council and an adviser to the telephone counselling service Lifeline. No portrait of Trevor would be complete without mention of his extraordinary family. Married in 1968 to Janice Biggs, the home shared with their four children was an epicentre of activity and love. Despite his blindness, he sailed, canoed, skied, and rode a tandem bicycle on numerous excursions. After facing pancreatic cancer with characteristic courage, Trevor died on 29 October 2004. He is survived by Janice and his children Penelope, Hamish, Virginia and Emily.

David Ames BA, MD, FRANZCP

To the Editor: A recent editorial in the British Medical Journal reported advice from the UK Committee on Safety of Medicines that most types of selective serotonin-reuptake inhibitors (SSRIs) should not be used in the treatment of major depression in children.1 The editorial sparked interest in the Australian media, resulting in articles in large metropolitan newspapers with titles such as “Army of kids on antidepressants”.2 General practitioners were targeted as the cause of reported “over-prescribing”. Unfortunately, while the media drew data from the national BEACH program (Bettering the Evaluation and Care of Health; a continuing study of general practice activity3), the data presented were inflated: a “child” was defined as someone aged under 20 years (while the UK advice related to children under 18 years), and national figures were extrapolated from the upper confidence limit. Reliable estimates of GP prescribing of antidepressants to children in Australia are needed. We derived age-specific rates of antidepressants prescribed per encounter in Australian general practice for the period April 2001 to March 2004 from the BEACH data (Box 1). The data showed that children were prescribed antidepressants far less often than adults. Those aged under 12 years were rarely prescribed antidepressants. Most of those prescribed were tricyclics, which are more commonly used in management of enuresis than of depression. The media’s inclusion of 18–19-year-olds as “children” greatly increased the reported rate.2 The prescribing rate of antidepressants in children aged under 18 years was 0.47 per 100 encounters (5 per 1000 encounters), but was six times higher for 18–19-year-olds (2.82 per 100 encounters). Most antidepressants prescribed for 12–17-year-olds were SSRIs. Fluoxetine is the only SSRI currently approved for use in children in the UK.4 In Australia, caution is advised when prescribing any antidepressant to children, but venlafaxine and the SSRI paroxetine are specifically advised against.5 Nevertheless, venlafaxine and paroxetine were more often prescribed (accounting for 10% and 8%, respectively, of total antidepressants for children) than fluoxetine (5%). However, GPs provided concomitant counselling at almost 20% of contacts with children aged under 12 years where an antidepressant was prescribed, and at 40% with 12–18-year-olds (Box 2). GPs were also more likely to refer the children to a specialist than when prescribing antidepressants for adults. We do not know how many of these children have been referred to a specialist at a previous encounter, nor how often antidepressant medication is initiated by a specialist. However, it will be interesting to see whether the new advice reduces the current level of prescribing of antidepressants (SSRIs in particular) in children. 1 Antidepressant prescribing in Australian general practice, April 2001 to March 2004 Age-specific rate per 100 encounters (95% CI) Variable (ATC group)† <12 years (n = 31 869) 12–17 years (n = 11 576) 18–19 years (n = 5823) ≥ 20 years (n = 247 231) All antidepressants 0.11 (0.07–0.14) 1.48 (1.18–1.77) 2.82 (2.35–3.28) 4.18 (4.05–4.31) SSRIs (N06AB) 0.03 (0.01–0.05) 1.08 (0.82–1.34) 1.84 (1.49–2.19) 2.38 (2.29–2.46) Fluoxetine (N06AB03) 0.003 (–)* 0.08 (0.00–0.13) 0.07 (0.00–0.14) 0.28 (0.25–0.30) Paroxetine (N06AB05) 0.003 (–)* 0.13 (0.06–0.20) 0.22 (0.10–0.34) 0.48 (0.45–0.52) Other SSRIs 0.02 (0.01–0.04) 0.87 (0.63–1.12) 1.55 (1.22–1.87) 1.62 (1.55–1.69) Tricyclics (N06AA) 0.07 (0.04–0.10) 0.14 (0.07–0.21) 0.22 (0.09–0.35) 0.92 (0.87–0.97) Other antidepressants 0.006 (–)* 0.26 (0.16–0.36) 0.76 (0.49–1.02) 0.89 (0.83–0.94) Venlafaxine (N06AX16) 0 0.17 (0.09–0.25) 0.52 (0.29–0.74) 0.47 (0.43–0.51) * Insufficient observations for calculating 95% confidence intervals. † Drug group according to the World Health Organization Anatomic Therapeutic Chemical (ATC) classification. SSRIs = selective serotonin reuptake inhibitors. 2 Concomitant management provided at encounters where an antidepressant was prescribed in Australian general practice, April 2001 to March 2004 Concomitant management (% of encounters where at least one antidepressant was prescribed [95% CI]) Management <12 years (n = 34) 12–17 years (n = 171) 18–19 years (n = 164) ≥ 20 years (n = 10 137) Counselling 17.6% (4.8%–30.5%) 40.4% (32.8%–47.9%) 44.1% (36.1%–52.1%) 30.4% (29.0%–31.8%) Referral to specialist 5.9% (–)* 6.4% (2.7%–10.2%) 6.8% (2.7%–10.9%) 2.7% (2.4%–3.1%) * Insufficient observations for calculating 95% confidence intervals.

Christopher M Harrison BPsych(Hons), MSocHlth · Helena C Britt BA, PhD

To the Editor: I report a patient with splenic infarction associated with group A streptococcal sepsis that occurred post partum. Although spontaneous splenic infarcts have been associated with many types of infections, to my knowledge this is the first published report of an association with this organism. A 29-year-old woman had an unremarkable term labour and vaginal delivery of her third child. On Day 2, she felt feverish, but no abnormalities were detected on clinical examination or pelvic ultrasound examination. Over the next 24 hours, she developed abdominal pain and sweats, and appeared flushed. On Day 3, her temperature was 37.6°C, and she developed nausea and diarrhoea. Empirical treatment was begun with intravenous ampicillin and metronidazole. She developed hypotension (blood pressure, 90/60 mmHg) and an erythematous rash of the legs and diffuse erythema of the trunk, anterior thighs and face. Relevant results of laboratory investigations are summarised in Box 1. On Day 4, a vaginal swab was taken, and antibiotic therapy changed to ticarcillin–clavulanate and clindamycin on the basis of presumed toxic shock syndrome. The next day, the patient developed oedema of the hands and feet, a sore throat and sore ankles. Group A streptococcus grew from the vaginal swab. Blood cultures showed no growth, but the samples had been taken after antibiotic therapy was begun. Over the next few days, the patient’s condition improved, but on Day 9 again deteriorated, with recurrence of low-grade fever and the development of sharp, retrosternal chest pain. Computed tomography (CT) of the chest with a pulmonary angiogram revealed a small right lower-lobe opacity, suggestive of a pulmonary infarct. The CT scan also revealed multiple splenic infarcts (Box 2). Screening for thrombophilia gave normal results. She was treated initially with intravenous heparin, followed by oral warfarin for 3 months. Her clinical recovery was slow but complete. This patient had probable toxic shock syndrome caused by group A streptococcus.1 She had the non-specific features of toxic shock syndrome2 (fever, nausea, diarrhoea, rash, abnormal hepatic and renal function) and disproportionate abdominal pain as the initial symptom. The only criterion lacking for “definite” toxic shock syndrome was the isolation of group A streptococcus from a normally sterile site (it was isolated only from the vagina). Puerperal toxic shock syndrome caused by group A streptococcus is well reported,1,2 with mortality of 25%–50%.2 The patient’s clinical course was complicated by multiple splenic infarcts and a possible pulmonary infarct, thought to have developed in situ with no identifiable prothrombotic diathesis. Splenic infarction is not common and is usually associated with a haematological or rheumatological disorder.3,4 Spontaneous splenic infarcts have been associated with infections, but there is only one report of these infarcts in association with toxic shock syndrome, in that case caused by Staphylococcus aureus.5 The infarcts have been postulated to be caused by circulating endotoxin.5 1 Abnormal laboratory results Investigation Result RR White cell count (x 109/L) 11.07* 3.50–11.00 Platelet count (cells x 109/L) 63 150–450 Prothrombin time (s) 15.3 12–5 ESR (mm/h) 22 0–12 C-reactive protein (mg/L) 56 < 3 INR 1.2 0.8–1.1 Alkaline phosphatase (U/L) 298 38–126 γ-Glutamyltransferase (U/L) 95 0–30 Albumin (g/L) 18 33–48 Creatinine (μmol/L) 111 60–110 * 63% band forms. ESR = erythrocyte sedimentation rate. INR = international normalised ratio. RR = reference range. 2 Computed tomography of the abdomen Scan shows one of multiple splenic infarcts — in the lateral third of the spleen.

Adrienne Torda

To the Editor: Cokis and Faris describe an intraoperative complication detected by transoesophageal echocardiography (TOE).1 Their letter is interesting in that it describes a rare complication during aortic valve surgery, and it is provocative in that it is critical of the Department of Health and Ageing decision not to rebate TOE (except in valve repair or replacement) to anaesthetists. A rare complication is not an argument for routine monitoring. Justification for monitoring requires detailed analysis of complication rates. The number needed to monitor for this and other complications is not known. Cokis and Faris do not discuss the rate of complications from TOE, which could be similar to that of the rare complication they describe. That there is a link between efficacy and the likelihood of a Medicare rebate is yet to be shown, and the authors themselves allude to this. TOE can be performed without a rebate. This is good for patients and also for a healthcare service which is strapped for funds. There are arguably other pressing needs for Medicare funds in the healthcare system. Presumably, both doctors were remunerated for their presence at the operation. Eligibility for a Medicare rebate can be a “perverse incentive” leading to overservicing. I have seen this with monitoring with TOE. Procedures have a clinical and financial cost as well as perceived benefit. I have seen other diagnoses missed or misinterpreted because of routine use of TOE, and it has occasionally led to prolonged intensive care unit stays and other complications. None of my arguments should deny TOE a place as a useful monitoring tool. It may become as routine during cardiac surgery as central venous pressure and arterial pressure monitoring is now. Whether that happens should not depend on whether TOE is eligible for a Medicare rebate. The use of TOE during surgery should depend on whether there is evidence of a meaningful benefit, and it is well to remember that the routine use of any procedure is hard to justify and can sometimes be dangerous. Early in Australian cardiac surgery, it was argued that the rebate for coronary bypass surgery should be related to the number of grafts. This argument was rightly not accepted. Similarly, the rebates for cardiac anaesthesia should not be related to the number of monitors used. Cokis and Faris should be commended on their excellent care of the patient. However, their argument for a rebate is not compelling.

John W Stokes FANZCA, FJFICM

In reply: Stokes raises a number of relevant issues, but we would like to make the following points. The case we reported occurred in a teaching hospital and neither of us undertakes routine transoesophageal echocardiography (TOE) in a private capacity. While a Medicare rebate is not directly relevant to the clinical usefulness of a medical procedure, the Medicare Benefits Schedule functions as a surrogate marker for clinical legitimacy. Stokes quotes anecdotes of occasional misuse or overuse of TOE. We agree that single cases neither justify nor give cause to reject a particular kind of monitoring. However, case reports, although lacking a denominator, are a start. The main point of our letter was, in fact, to report the complication of surgery and the vital role played by TOE in achieving a good outcome. Nevertheless, many of us who routinely use TOE consider that its advantages over other kinds of monitoring regularly benefit patients. We agree there is little “hard” evidence to support this, but detailed risk–benefit analysis for many of our routine monitoring devices is similarly non-existent. The Swan–Ganz catheter is a classic example. We suspect that if a group of cardiac anaesthetists and surgeons was asked to review the usefulness of TOE in routine cardiac surgery, the decision of the Department of Health and Ageing might be different.

Chris Cokis MB BS, FANZCA · John Faris MB ChB, DAvMed, FFOM, FANZCA

To the Editor: de Crespigny and Savulescu1 criticise legal and media attention given to the abortion of a 32-week fetus with suspected dwarfism in a case in which the expectant mother had become suicidal. At the same time, they appeal to populism to support legislative change in favour of easier access to late-term abortion. This paradox raises an intriguing point about public opinion and medical ethics. How can members of the public develop opinion unless they are told what is going on? The case was made known to the public (although the patient’s name and face were not shown) because a newspaper editor considered it to be of great interest to many people. The public rightly has an interest in this tragic case, as it relates to at least three controversial themes of significance to public and social health — disability, suicide and abortion. Unfortunately, in recent decades, balanced and informed debate about abortion has been lacking. Instead, the issue has been portrayed simplistically in terms of a woman’s “right to choose”, with little research into the desperate and often coercive circumstances and the harmful consequences of those choices. That late-term abortions are subject to unclear and complex state laws is testament to the fact that abortion is far from a closed case for the public, healthcare professionals and politicians. de Crespigny and Savulescu acknowledge that the case raises “profound and divisive ethical issues”. Yet, they later assure readers that the case “appears to be ethical”, adding to their overall implication that, if only Australia’s laws were clear, uniform and liberal, the media and the public would not have had any business in the matter. Many aspects of the doctor–patient relationship are regulated by legislation — an “intrusion” that is accepted in jurisdictions in which democracy and state-funded healthcare coexist, and necessarily so, to protect both doctor and patient. If legislative change is to occur, public consultation must be sought, and the public must be given more, not less, information about abortion.

Selena R Ewing BHSc

To the Editor: The article by de Crespigny and Savulescu is certainly thought provoking and timely. The harms they cite as caused by an uncertain legal environment are lamentable, although the outcomes were probably the result of multiple factors in addition to the justice system. To bring order, reason, compassion and justice to a clinical problem as complex as termination of pregnancy — especially late termination — requires a framework for decision-making. This should operate at the hospital level, at the national level among the professionals involved, and, as de Crespigny and Savulescu contend, in the national legal system. Most hospitals have now developed consistent guidelines to assist clinicians and patients in decisions regarding pregnancy termination (in the past, there was significant intrahospital diversity and uncertainty). More recently, clinicians involved with late termination of pregnancy for fetal abnormality in eight centres in six states and the Australian Capital Territory met in Melbourne to develop a consistent national set of guidelines. There was adequate consensus to produce a document that will soon be submitted for publication for wider community comment. Hopefully, this will facilitate better outcomes for all and perhaps even provide a stimulus for review and consistency of abortion laws.

Leslie Reti SM, FRCOG, FRANZCOG

In reply: Reti is correct that the outcomes of the late abortion case probably resulted from multiple factors in addition to the justice system. A pivotal one was the decision to divulge confidential patient information before there had been a thorough internal review. It is pleasing that hospitals are developing consistent guidelines, although, as we indicated in our article, these guidelines should not include responsibility for clinical decision-making by committee. The responsibility for clinical decision-making should reside with the doctor, and committees should have a purely advisory role. Consistent national professional guidelines are needed. These could be a stimulus for law reform. Without reform, it is only a matter of time before a single complaint about a case leads to a similar succession of adverse outcomes. Ewing writes that we “criticise legal and media attention given to the abortion”. We have no criticism of the media attention and would not presume to criticise the legal processes. Our criticism was of the “decision to expose the events to legal and media scrutiny”. That is, we criticised the decision to expose the case and those involved before a thorough internal review had been conducted. We support transparency and believe that secrecy in relation to medical procedures is contrary to public interest. We agree with Ewing that public consultation must be sought before legislative change. But one thing is clear — abortion law reform is essential. It is unacceptable that, in some cases, such as the late-abortion case we described, doctors may be charged with an indictable offence whether they agree to perform the abortion or not — under the law on abortion or child destruction if they agree to abortion, or under the law of homicide by negligence if they refuse abortion and the patient subsequently commits suicide.

Lachlan J de Crespigny MD, FRANZCOG · Julian Savulescu MB BS, BMedSci

To the Editor: In a recent article, de Costa and Robson1 suggest that Australia’s high rates of caesarean surgery — currently among the highest in the Western world — may be beneficial, and causally related to our low perinatal mortality rate. In support, they cite a single article that reports the outcomes from three large hospitals in Dublin between 1979 and 2000.2 In these hospitals, as in most of the Western world, caesarean rates increased and perinatal mortality rates declined over this 21-year period. The authors of the article ascribe a causal relationship, but admit that “. . . it was not possible to allow for the confounding effect of time”.2 The time factor also confounds the interpretation of Australian data. Furthermore, results from an earlier Dublin study “. . . do not support the contention that the expansion in cesarean birth rates has contributed significantly to reduced perinatal mortality in recent years,”3 and there are many other articles with similar conclusions.4 Moreover, de Costa and Robson do not acknowledge the significant morbidity associated with caesarean surgery, nor the risks to mother and baby in subsequent pregnancies. A recent large retrospective cohort study in Scotland found that women whose first baby had been born by caesarean section had twice the risk of unexplained stillbirth at term in the subsequent pregnancy.5 There are also well documented increased risks of placental pathology (placenta praevia, accreta and percreta) in this group. Such problems are likely to increase in Australia in proportion to the increase in caesarean rate. I note also that King et al, who discuss maternal mortality in the same issue of the Journal, specifically mention the contribution of previous caesarean surgery to severe obstetric haemorrhage and emergency hysterectomy.6 They report that maternal death from amniotic fluid embolism occurred in association with induction in five of seven cases. Australian rates of induction and augmentation are among the highest in the Western world. Finally, as regards onus of proof, I agree with the statements by Enkin et al7 that “. . . the only justification for practices that restrict a woman’s autonomy, her freedom of choice, and her access to her baby, would be clear evidence that these restrictive practices do more good than harm; and second, that any interference with the natural processes of pregnancy and childbirth should also be shown to do more good than harm”, and “. . . the onus of proof rests on those who advocate any intervention that interferes with either of these principles”.

Sarah J Buckley MB ChB, DipObst

In reply: Buckley makes some important points regarding caesarean section, but overlooks the fact that our brief was to explore childbirth options from the viewpoint of the baby, not the mother. The Irish study of Matthews (a paediatrician) and colleagues took as its outcome measure deaths during pregnancy or within one week of birth of normally formed infants weighing > 2.5 kg.1 This was done because two of the main contributors to crude perinatal mortality rates (in Australia and other developed countries) are lethal abnormalities and very low birthweight — neither of which is likely to be improved by increasing caesarean section rates. More than 400 000 births over 22 years were studied retrospectively. While the time factor is acknowledged, the authors clearly show that as caesarean section rates have risen mortality among these normally formed babies of normal weight has fallen. They state that “. . . the caesarean section rate is an important part of the overall package of care delivered”, a “package” that includes the advances in antenatal surveillance and neonatal care of the past 22 years, as well as wider indications for caesarean section. The authors invite other centres to publish “similar matching caesarean section and mortality rates . . . to see whether some hospitals are capable of delivering packages of care that include low caesarean section rates (? < 15%) and perinatal mortality rates of < 1.5/1000 for normally formed babies of normal birthweight”. To date, none have done so, but there have been reports from other large maternity hospitals of similar findings to those of Matthews et al. One of these adds that “[our] low incidence of intrapartum hypoxic ischaemic encephalopathy (1.3/1000 births) . . . suggests that a policy of more liberal caesarean section may benefit babies in ways other than simply avoiding death”.2,3 In other words, this very large and careful study, and others resulting from it, strongly support the view that current caesarean section rates are good for babies.

Caroline M de Costa FRANZCOG, FRCOG

To the Editor: The article by de Costa and Robson1 is a timely reminder that the ideology and politics surrounding maternity services could have an adverse impact on Australia’s excellent record as one of the safest countries in the world in which to be born.2 de Costa and Robson highlighted continuity of care as the attribute of antenatal supervision and birthing that women value most highly, and they quote evidence of the safe care provided by a midwife or general practitioner in a “low-tech” environment. This type of care is currently provided by a diminishing number of GP obstetricians and midwives in small obstetric units throughout rural Australia, where continuity of carer ensures the continuity of care that leads to maternal satisfaction and good health outcomes. Data show a lower rate of adverse events in small rural hospitals compared with urban hospitals. Studies in diverse environments suggest communication breakdowns and handovers between multiple carers are major risk factors.3,4 These points of vulnerability are minimised in the close environment of a small rural hospital. National and international data demonstrate the safety of small rural maternity services,5 and yet rural obstetric units continue to be closed at an alarming and accelerating rate. The proponents of “de-medicalising” birth and improving maternal satisfaction through continuity of care are focused on perceived problems in the delivery of obstetric care in large urban hospitals. The evidence presented by de Costa and Robson confirms that women are most satisfied with care by a midwife and GP in a “low-tech” environment. While this option may now be unavailable in many urban areas, it is generally the model that exists in rural areas. Unfortunately, the politics of change is resulting in the application of urban- and ideology-based processes to rural maternity units, where they are often inappropriate and can lead to reduced support for rural procedural obstetricians. This is likely to result in the eventual closure of the maternity units — a situation in which women, their babies, local healthcare professionals and their communities will all lose out in the end. For rural communities, the risk in local maternity services is not to the standard of care, but to the continued existence of their services. Transferring alternative urban models of maternity care to country hospitals may be superficially attractive to budget-focused health authorities or ideologues, but it is rural people and their babies who will have to live with the consequences.

Graham M Slaney · Susan M Stratigos

CorrectionRe: “Prescribing of amino acid formula”, by Andrew S Kemp in the 15 November 2004 issue of the Journal (Med J Aust 2004; 181: 574-575). The author’s position and address were omitted. Dr Kemp is Professor of Paediatric Allergy, The Children’s Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145. andrewk5ATchw.edu.au The html and pdf versions of his letter published in the eMJA were corrected on 21 December 2004.

Andrew S Kemp

What comes first? The cannabis or the psychosis? European researchers have challenged the commonly held belief that people may start using cannabis because they are predisposed to psychosis. They surveyed 2500 young Germans, aged 14 to 24 years, once in 1995 and again, four years later, in 1999. A predisposition to psychosis, determined from a self-report symptom checklist, did not predict future cannabis use four years later. However, exposure to cannabis did increase the risk of psychotic symptoms later in life, especially in those with a predisposition to psychosis. Moreover, there was a dose-response relationship — the greater the cannabis use at baseline, the greater the risk of developing psychotic symptoms, independent of a predisposition to psychosis. BMJ, doi:10.1136/bmj.38267.664086.63 (published 1 Dec 2004) Doctors in films Patients' expectations of doctors may be adversely affected by the current negative movie portrayals of doctors, says a US author. Flores, an associate professor of paediatrics, epidemiology and health policy, reviewed over 130 films depicting doctors, made over eight decades. Until the 1960s, doctors were mostly seen as compassionate and idealistic individuals, but since then movie doctors have been more likely to be portrayed as greedy, egotistical, uncaring and unethical. Flores does not suggest we counter this by joining forces with Hollywood to produce a film that shows doctors not as gods or demons but as people just trying to do the best they can. However, he does put The Doctor at the top of his list of movies that may be useful in medical education. Made in 1991 and starring William Hurt, Flores says the movie is about a hotshot surgeon with a detached demeanour who learns about empathy and compassion when he contracts cancer. Arch Dis Child 2004; 89: 1084-1088 Look into the future When it comes to postoperative care, some patients may be just as happy to see their surgeon on screen — via a robotic telerounding visit (see figure) — as in person. Ellison and colleagues compared this procedure with standard ward round-type visits and a web-based video-conferencing system (involving laptop computers) in a randomised trial involving 85 patients who had undergone elective, short-stay, minimally invasive urological surgery. The "robot" arm of the study was assessed just as favourably by patients, and rated particularly well with regard to physician availability. Safety was not assessed. J Am Coll Surg 2004; 199: 523-530 Slimming slumber Some welcome news for our patients and the one or two of us who are overweight and not keen on exercising — chronic sleep curtailment may be a previously unrecognised risk factor for obesity. In a small, crossover study involving a dozen healthy young men, two nights of sleep restriction (four hours of sleep) led to increased hunger and appetite compared with two nights of sleep extension (10 hours). There were corresponding changes in circulating hormone levels — a drop in leptin (anorexigenic) and an increase in ghrelin (orexigenic). Ann Intern Med 2004; 141: 846-850 Stop the super-spreaders Ever wished there were more ways to slow down, if not stop, the rapid spread of airborne diseases like influenza and SARS? A small, international research study may show us the way forward. Edwards and colleagues explored their study hypothesis — that it might be possible to reduce the amount of aerosols we exhale by altering lung airway surface properties — by administering nebulised isotonic saline to 11 healthy adult volunteers. They found that the number of particles exhaled varied dramatically among the subjects. About half the group were "high-producers", and in these individuals, but not the others, nebulised saline led to a decrease in the number of particles exhaled. Proc Natl Acad Sci U S A 2004; 101: 17383-17388 All’s well that ends well In 1993, Australian researchers reported in The Lancet the results of their randomised controlled trial involving about 3 000 pregnant mothers. They found that a routine policy of multiple scans in pregnancy (from 18 weeks' gestation onwards) did not improve pregnancy outcomes over and above a single scan (at 18 weeks). However, they were concerned about an unexplained growth restriction in newborns whose mothers had been allocated to the repeated ultrasound group, and have followed the children involved in the study ever since. Now, they are pleased to report that, after 8 years of follow-up, there has been no adverse effect on long-term growth and development. However, the researchers caution that there is much more to learn before we can rest assured that ultrasound scanning in pregnancy is safe and without risk — today’s equipment is increasingly powerful and, moreover, scans may be conducted much earlier in pregnancy nowadays. Lancet 2004; 364: 2038-2044 Dr Ann Gregory, MJA

Ann Gregory