Volume 187 Issue 7

Their value in modern medical education needs to be reaffirmed Arguably the greatest claim to fame of the renowned English surgeon John Hunter was not his outstanding contribution to anatomy and surgery but the remarkable collection that now forms the Hunterian Museum in London.1 Hunter’s collection is testimony to his passion for science and his aptitude for self-directed enquiry, independent study and life-long learning. These attributes are now cherished cornerstones of modern medical education. Yet it is the adoption of these educational principles in modern medical curricula that is contributing to the demise of the time-honoured medical museum. No doubt Hunter would be saddened and dismayed at the plight that has befallen some of our wonderful collections of pathology specimens, medical and surgical artefacts and memorabilia. Most of these collections, which are housed in medical schools and teaching hospitals, were for a long time the exclusive province of medical students. Unfortunately, these repositories of medical history and the manifestations of disease are increasingly neglected, closed or under threat of closure.2 There are many reasons for this turn of events, including dramatic changes in medical education and medical practice over the past few decades, as well as financial problems common to all health care delivery systems. Medical museums (which include museums of pathology, anatomy and nursing) are perceived to be expensive facilities that do not have a well defined role in modern medical education and training. This is particularly so in graduate-entry medical courses and highly integrated courses, where it is often hard to identify the anatomy and pathology components of the curriculum sufficiently distinctly to be able to link them to museum-related study activities. Increasingly, medical students are sent to rural hospitals and practices remote from the central medical school (and museum) to gain clinical experience and encourage rural practice. In these environments, there is little opportunity to actually see “diseased tissue” as displayed in museums, or to pursue museum-related study activities. Reduced funding for medical education in a number of countries, including Australia,3 has diminished the survival prospects of medical museums. Additionally, the acquisition of new specimens, particularly specimens of human disease, has become a major problem. This is in part due to major advances in surgical techniques and the striking decline in the number of autopsies,4 so that acquisition of pathology specimens suitable for presentation and display has become near impossible. This has occurred in parallel with changes in the pattern of disease in developed societies, which have had an impact on the acquisition of specimens of infectious diseases (eg, tuberculosis, osteomyelitis and meningitis). Surgical specimens and donated bodies for anatomy dissection have been used by some medical schools to overcome the shortage of appropriate disease specimens. Unfortunately, changes to legislation on human tissue in various countries have made it so onerous and time-consuming to comply with the legislation that many pathology laboratories avoid retaining tissue.5 With the diminution in the museum’s traditional role in teaching medical students, museums that have survived and thrived have had to reinvent themselves. For example, the Museum of Human Disease at the University of New South Wales (UNSW), Sydney, has diversified its activities and no longer just provides specimens for use in tutorials and practical classes. Satellite museums have been established at the major teaching hospitals, so that students on secondment to remote teaching hospitals still have access to pathology specimens for study and for use in examinations. The museum at UNSW and several other university museums have also made their collections available online and/or in compact disk format. Students can now have the option of either visiting the museum or accessing its collection of specimens via the Internet. Several museums, such as those at the University of Western Australia, the University of Melbourne and UNSW, also have community education programs for high-school students and other interest groups. These have been enormously popular. For example, the Museum of Human Disease at UNSW now receives over 20 000 visits each year from high-school students. This has not only lifted the public profile of these museums and their medical schools, but has also provided additional funds to help support other museum activities. To survive in the increasingly difficult environment of medical education, it is important that museums become integrated into the new medical curricula. The value of pathology museums as a teaching resource should be recognised and emphasised. This is a particularly important issue, because some people see museums as a relic of the past that is dispensable. How can we promote the survival of these museums? Medical practitioners can help by supporting and promoting medical museums, as well as by doing volunteer work for, donating to and attending these wonderful repositories of our history. All those involved in medical education should stress the value of careful observation of disease and disease processes, on display in our museums, as a core learning activity. Fostering the public’s interest in disease by making medical museums more accessible will also contribute to the survival of this threatened species. Students and volunteers in the Museum of Human Disease, University of New South Wales.

Denis Wakefield MD, FRACP, FRCPA

Empathy should be sought and supported as a desirable, career-long trait in doctors Many observers have argued that medical practice is increasingly dehumanised, dominated by impersonal technologies and economic imperatives.1 A solution that is sometimes offered is to cultivate empathy in doctors.2 Many doctors may be sceptical, wondering whether empathy is too ill-defined to make a difference in the pressured arena of clinical work. Perhaps it can be safely left to essay-writing medical humanists or to the nursing staff . . . Such scepticism is unwarranted. Empathy can be defined as the capacity to take the perspectives of others, to be sensitive to their inner experience and to engage with them compassionately, rather than simply sharing their emotions (sympathy). In the medical context, empathy can be defined as “appreciation of the patient’s emotions and expression of that awareness to the patient”.3 More importantly, an emerging body of research indicates that empathy is associated with a variety of positive clinical outcomes. Much of the evidence linking doctors’ empathy to patients’ outcomes is indirect. Large bodies of research show that positive outcomes are associated with the quality of the doctor–patient relationship and doctor–patient communication, and that empathy enhances both. Direct links between empathy and clinical outcomes have also been established. More empathic medical students received higher ratings of clinical competence and performed better on history-taking and standardised physical examinations.4,5 More empathic medical students and doctors received higher patient satisfaction ratings.5 Patients judge empathy to be very important in consultations,6 and show better treatment adherence and greater enablement with more empathic doctors.7 When doctors report a loss of empathy they subsequently show an increase in their rate of major medical errors.8 Doctors’ communication skills are associated with a variety of positive outcomes for patients9 and with reduced risk of malpractice claims, and patients judge their doctors’ empathy on the basis of such skills (eg, being reassuring, showing understanding, explaining procedures, not ignoring their concerns).10 If empathy promotes positive clinical outcomes, then we should want doctors to be empathic. This aim could be achieved by promoting empathy in medical training, and by using empathy as a selection criterion for entry to medical training. These two approaches reflect the reality that empathy is a disposition or quality, but also one that is malleable. Many ways of promoting empathy have been proposed, ranging from training in communication skills, to education in the medical humanities, through to complete overhaul of the medical curriculum. A review of intervention studies suggests that communication skills workshops have the greatest impact on medical students’ empathy.3 Assessing empathy as a criterion in medical school admission is more controversial, and there is as yet little direct evidence that it would produce better doctors. Nevertheless, several studies make an indirect case for supplementing traditional selection procedures, which are based heavily on educational attainment and cognitive ability, with measures of empathic traits. First, measures of empathy are generally unrelated to indices of attainment and ability, but capture non-redundant information that is relevant for selection. For example, one study found that empathy correlated highly with selection interview scores,11 but not with academic grades or a standardised aptitude test. Second, psychometric measures of empathy have been shown to predict real-world empathic behaviour, and should therefore be able to identify applicants who are especially likely to interact empathically with patients. Finally, including empathy assessment in the selection process is likely to increase access to medical training for less socially privileged applicants. Such students tend to fare less well on traditional criteria, but perform equally well or better on assessments of empathic traits.12 The possibility that broadened selection criteria might tend to favour female applicants, given evidence of sex differences,2 must also be entertained. Even if doctors’ empathy can be enhanced by medical education or selection, it is at least as important to prevent the erosion of empathy over the course of professional training and practice. There is now considerable evidence that levels of empathy tend to decline during medical school and beyond.13 This decline has been attributed to the growing reliance on technology, and to economic and time constraints imposed by health care systems. Others identify psychological dynamics beneath the loss of empathy, such as creeping cynicism, a self-protective disengagement from people’s suffering and a sense of hopelessness in the face of therapeutic failure. Research indicates that the loss of empathy is linked to burnout — notably, feelings of emotional fatigue and a tendency to depersonalise patients. Trainees who make major medical errors subsequently experience a decline in empathy.8 Similarly, doctors who feel less in control of patient outcomes tend to engage in styles of closed, impersonal and inattentive communication that lead patients to see them as unempathic.10 There is a positive side to all of this. Empathy can be lost, but it can also be gained. If doctors lose empathy because of burnout, unhappiness and a loss of a sense of control over their work, then remedying these conditions should indirectly increase empathy. Indeed, doctors who report higher levels of wellbeing and a greater sense of personal accomplishment also report higher levels of empathy,14 and increases in wellbeing among residents are accompanied by increases in empathy.15 If empathic doctors are indeed better doctors, then improving doctors’ wellbeing should benefit patients.

Nick Haslam PhD

Objective: To estimate hospital inpatient costs by age, time to death and cause of death among older people in the last year of life.Design and setting: Cross-sectional analytical study of deaths and hospitalisations in New South Wales from linked population databases.Participants: 70 384 people aged 65 years and over who died in 2002 and 2003.Main outcome measures: Hospital costs in the year before death.Results: Care of people aged 65 years and over in their last year of life accounted for 8.9% of all hospital inpatient costs. Hospital costs fell with age, with people aged 95 years or over incurring less than half the average costs per person of those who died aged 65–74 years ($7028 versus $17 927). Average inpatient costs increased greatly in the 6 months before death, from $646 per person in the sixth month to $5545 in the last month before death. Cardiovascular diseases (43.1% of deaths) were associated with an average of $11 069 in inpatient costs, while cancer (25.0% of deaths) accounted for $16 853. The highest average costs in the last year of life were for people who died of genitourinary system diseases ($18 948), and the highest average costs in the last month of life were for people who died of injuries ($8913).Conclusion: Population ageing is likely to result in a shift of the economic burden of end-of-life care from the hospital sector to the long-term care sector, with consequences for the supply, organisation and funding of both sectors.

Katina Kardamanidis RN, MPH · Kim Lim RN, BEc, MAppStat · Cristalyn Da Cunha BSc, MHA · Lee K Taylor MB BS, MPH, FAFPHM · Louisa R Jorm BVSc, MSc(Epid), PhD

Objective: To determine the risk of suicide and drug overdose death among recently released prisoners.Design, setting and participants: Retrospective cohort study of 85 203 adult offenders who had spent some time in full-time custody in prisons in New South Wales between 1 January 1988 and 31 December 2002.Main outcome measures: Association between time after release and risk of suicide and overdose death.Results: Of 844 suicides (795 men, 49 women), 724 (86%) occurred after release. Men had a higher rate of suicide than women both in prison (129 v 56 per 100 000 person-years) and after release (135 v 82 per 100 000 person-years). The suicide rate in men in the 2 weeks after release was 3.87 (95% CI, 2.26–6.65) times higher than the rate after 6 months. Male prisoners admitted to the prison psychiatric hospital had a threefold higher risk than non-admitted men both in prison and after release. No suicides among women were observed in the 2 weeks after release. No increased risk of suicide was observed among Aboriginal Australians in the first 2 weeks after release. Of 1674 deaths due to overdose, 1627 (97%) occurred after release. Drug-related mortality in men was 9.30 (95% CI, 7.80–11.10) times higher, and in women was 6.42 (95% CI, 3.88–10.62) times higher, in the 2 weeks after release than after 6 months.Conclusions: Prisoners are at a heightened risk of suicide and overdose death in the immediate post-release period. After 6 months post-release, the suicide rate approaches the rate observed in custody.

Azar Kariminia MSc · Matthew G Law PhD · Tony G Butler PhD · Michael H Levy MD · Simon P Corben MSc · John M Kaldor PhD · Luke Grant MSc

Objective: To determine the risk of occurrence and recurrence of postpartum haemorrhage (excessive bleeding after childbirth) among women having at least two consecutive pregnancies.Design and setting: Population-based study using longitudinally linked hospital discharge and birth records from New South Wales for the period 1 January 1994 to 31 December 2002.Participants: All 125 295 women having at least a first and second pregnancy resulting in a singleton birth at > 400g or ≥ 20 weeks’ gestation in the study period.Main outcome measures: Risk of occurrence of postpartum haemorrhage (PPH) in any pregnancy, and of recurrence of PPH in subsequent (second and third) pregnancies.Results: 5.8% of women (7327/125 295) had a PPH in their first pregnancy, and 4.5% (5318/117 968) had a first PPH in their second pregnancy. Among the 23 095 women who had three pregnancies in the study period, 4.4% (908/20 839) had a first PPH in their third pregnancy. The risk of recurrence in a second consecutive pregnancy was 14.8% (1082/7327), and in a third consecutive pregnancy (after two previous PPHs) was 21.7% (43/198); even with an intervening pregnancy with no PPH (ie, PPH in the first and third pregnancies only), the risk for the third pregnancy was 10.2% (111/1085).Conclusions: These consistently elevated risks of recurrence highlight the need for women with a history of PPH to have active management of the third stage of labour and to give birth in a hospital that has onsite blood cross-match facilities.

Jane B Ford BA(Hons), PhD · Christine L Roberts MPH, DipObs, DrPH · Jane C Bell BDS, MAppEpid, MPH · Charles S Algert BA, BSc, MPH · Jonathan M Morris MB ChB, FRANZCOG, PhD

Objective: To determine the prevalence of health problems in New South Wales train drivers and the impact of the new national health-assessment standard on train drivers’ fitness for work.Design, setting and participants: Retrospective audit of files of all RailCorp train drivers (743) and train driver recruits (283) who were assessed under the new national standard for health assessment of rail safety workers between February 2004 and February 2005.Main outcome measures: Smoking status; prevalence of hypertension, heart disease, diabetes and obstructive sleep apnoea; alcohol use disorders; body mass index (BMI); total cholesterol level; fasting blood glucose level; cardiac risk score; fitness status.Results: 25.2% of drivers and 27.9% of recruits were smokers; 43.8% of drivers and 21.9% of recruits were hypertensive; 34.6% of drivers and 31.4% of recruits had high total cholesterol levels (> 5.5 mmol/L). Median BMI values were 29 kg/m2 (range, 18–59 kg/m2) for drivers and 28 kg/m2 (range, 19–55 kg/m2) for recruits. The prevalence of obesity (BMI ≥ 30.0 kg/m2) was higher in both male drivers and recruits compared with the general male population. At initial assessment, 65.1% of drivers and 88.0% of recruits were certified as unconditionally fit for work; 12.4% of drivers and 7.1% of recruits were assessed as temporarily unfit; and 22.5% of drivers and 4.6% of recruits were considered fit subject to review (after periods ranging from 3 to 12 months). Two per cent of drivers and 2.5% of recruits were subsequently deemed to be permanently unfit, the most common reasons being heart conditions, psychiatric disorders, orthopaedic problems, colour vision impairment and sleep apnoea.Conclusions: Cardiovascular risk factors and cardiovascular disease are the most significant health issues affecting train drivers’ fitness for work. With the more stringent health assessment and regular review required by the new standard, most drivers can continue with their duties, with the added benefits of improved personal health and greater safety to the rail network and the public.

Reem Mina MB BS, GradDipOccEnvHealth, FAFOM · Armand Casolin MB BS, MSciTech(OccMed), FAFOM

An 80-year-old woman presented with abdominal pain, nausea and diarrhoea. She had undergone a laparoscopic cholecystectomy 10 years previously. The surgery was complicated by an umbilical hernia at the umbilical port site, which was repaired surgically 5 years later. The patient reported intermittent drainage of clear fluid from the umbilicus after the hernia repair. She had right lower quadrant tenderness without guarding or a palpable mass. Laboratory investigations showed a serum lipase concentration of 393 U/L (reference range, 114–286 U/L) with normal white blood cell count. Abdominal computed tomography revealed a 2.8 × 2.4 cm mass in the right lower quadrant, with a calcific rim and foci of calcium within the lesion (Figure, A). The mass appeared to be outside the small bowel, suggesting a mucocele or mucoid tumour. Laparoscopy revealed a 2.5 cm solid mass containing a gallstone (Figure, B and C). The drainage from the umbilicus continued after surgery and was apparently unrelated to the spilled gallstone. Gallstone spillage is more common during laparoscopic than open cholecystectomy. It occurs in 5%–40% of laparoscopic cholecystectomies, but complications are very rare,1 with a reported incidence of 0.08%–0.3% of cases.2 The reason that only a fraction of patients develop complications is uncertain.1 Figure A: Computed tomography scan of the abdomen showing a 2.8 × 2.4 cm mass (arrow) external to the small bowel. B, C: The gallstone recovered during exploratory laparoscopy.

Edgard Wehbe · Reggie J Voboril · Elisha J Brumfield

The federal Privacy Act 1988 (Cwlth) has recently been amended to permit the disclosure of genetic information to an at-risk relative when there is a serious (although not necessarily imminent) threat to that person’s life, health, or safety. This represents a significant exception to the statutory obligations to maintain the privacy of a patient’s health information. However, its scope of operation is limited in that it applies only to doctors and other health professionals working in the private sector, and does not cover those working in State public hospitals or for Commonwealth Government agencies.

Margaret F A Otlowski PhD

Medical education faces global challenges because of the changing health care needs of an ageing and more demanding society, and the consequent requirement for increased health care workforce capacity and different workforce models. In the United Kingdom, education reform has spanned the medical, nursing and allied health professions, and has introduced new health professions with specific roles within a new, team-based model of comprehensive health care. In medical education, the UK reforms span undergraduate, prevocational, vocational and continuing education, with the aim of providing a framework for faster, more flexible career development that can adapt to future changes in workforce need. While some reforms are controversial, most appear sensible and are supported by most observers. The Modernising Medical Careers process suffered implementation difficulties in 2007: the national, web-based application scheme for vocational training posts could not cope with such a large process, disrupting both the recruitment of an appropriate workforce for hospitals and the career progression of many recent UK medical graduates. The main problem appears to have been in management of change — too much was attempted too quickly on too large a scale — resulting in a backlash against any significant change. There may be lessons for Australia and New Zealand, which face similar challenges and are considering broadly similar changes.

Richard B Hays MD, FRACGP, FACRRM, MRCGP

The four main classes of antifungal drugs are the polyenes, azoles, allylamines and echinocandins. Clinically useful “older” agents include topical azole formulations (for superficial yeast and dermatophyte infections), first-generation triazoles (fluconazole and itraconazole, for a range of superficial and invasive fungal infections), amphotericin B formulations (for a broad range of invasive fungal infections) and terbinafine (for dermatophyte infections). Clinically important “newer” agents include members of the echinocandin class (eg, caspofungin) and second-generation triazoles (eg, voriconazole and posaconazole). Voriconazole and posaconazole have broad-spectrum activity against yeasts and moulds, including Aspergillus species. Posaconazole is the only azole drug with activity against zygomycete fungi. Caspofungin and the other echinocandins are effective in treating Candida and Aspergillus infections. The azoles are relatively safe, but clinicians should be aware of drug–drug interactions and adverse effects, including visual disturbances (with voriconazole), elevations in liver transaminase levels, and skin rashes. Caspofungin has minimal adverse effects. Combination antifungal therapy may be appropriate in selected patients with invasive fungal infections, but is empiric and driven by individual physician practice.

Sharon C A Chen PhD, FRACP, FRCPA · Tania C Sorrell MD, MB BS, FRACP

Since the National Mental Health Policy was implemented in 1992, dissonance between mental health professionals and the general public on the success of the deinstitutionalisation components of the strategy has grown. Many of the premises on which the deinstitutionalisation components were based are false, and this has led to many problems in the system. Community psychiatry is not fundamentally flawed, and what has been learned in the past 15 years can be used to build on the foundations that have been laid. Better cooperation between state and federal governments is needed to effect real change.

Bruce S Singh PhD, FRACP, FRANZCP · David J Castle MD, MRCPsych, FRANZCP

A 61-year-old woman diagnosed with multifocal, poorly differentiated epithelial mesothelioma in September 2002 went into sustained spontaneous remission within months. She was completely disease-free within 6 months, and remained so 5 years later. This case demonstrates that this tumour may, very rarely, regress spontaneously, with no recurrence for many years. A greater knowledge of the underlying immune mechanisms would aid future management of this and other tumours. Clinical recordIn early September 2002, a 61-year-old woman was referred to our centre from the emergency room of a local private hospital. She had presented with a 2-day history of intermittent, sudden onset, severe right-sided lateral pleuritic chest pain lasting a few minutes. In the private hospital emergency room, chest radiography and computed tomography (CT) had shown pleural masses, which were subsequently found to be poorly differentiated epithelial mesothelioma. The patient reported that, over the previous week, she had felt weak and lethargic, but was otherwise well. She had no previous serious illnesses, but had recently started taking iron supplements for anaemia and occasionally took non-steroidal anti-inflammatory agents for osteoarthritis. She was a non-drinker, and had ceased smoking 5 months previously. (She started smoking at the age of 16 years and had been smoking 40 cigarettes a day.) She was a widow with three adult children, and had emigrated from the United Kingdom in 1969. Her husband died at the age of 39 years of a myocardial infarction. Her father died of carcinoma of the oesophagus, and her mother of “old age”. Her four siblings and three adult children were all well. Although she had no history of asbestos exposure from any of her husband’s occupations, she could have been exposed to asbestos during two periods of her life. From the age of 15–22 years in the UK, she worked as a machinist in a factory where asbestos lagging was used for the steam pipes of steam presses and central heating. Then, in 1984, her son worked for a year for a company making asbestos gaskets, and throughout this period she washed his work overalls. On examination, she looked well and was not in pain. She weighed 74 kg and was of normal build, but had slight conjunctival pallor. An electrocardiogram showed sinus rhythm; her blood pressure was 140/60 mmHg, and her jugular venous pressure was not elevated. There was no cyanosis or clubbing. Examination of the chest, cardiovascular system, breast, abdomen and peripheries showed no abnormalities. InvestigationsA chest radiograph taken before referral showed a pleural density measuring 10 cm × 2 cm overlying the posterior aspect of the right lower lobe, with no pleural plaques or other stigmata of asbestos exposure. A CT pulmonary angiogram performed the same day to exclude pulmonary embolism showed three pleural masses in the right side of the chest — the first corresponding to the opacity visible on the chest radiograph in the right costovertebral gutter at the level of the tracheal bifurcation, the second having a diameter of 6 cm and located in the right cardiophrenic angle (not of fatty attenuation), and the third in the right posteromedial costophrenic recess, just above the diaphragm (Box 1A and Box 1B). The lungs, mediastinum and upper abdomen (including the pancreas and para-aortic nodes) were normal. Laboratory tests showed her haemoglobin level was 104 g/L (reference range [RR], 115–160 g/L), with a normochromic normocytic anaemia; she had thrombocytosis (555 × 109/L [RR, 150–450 × 109/L]) and leukocytosis (11.2 × 109/L [RR, 4.0–11.0 × 109/L]), with mild neutrophilia (8.2 × 109/L [RR, 2.0–7.5 × 109/L]). Her erythrocyte sedimentation rate was markedly elevated at 110 mm/h (RR, 1–30 mm/h), as was her serum C-reactive protein level (294 mg/L [RR, 0–6 mg/L]). She had mildly elevated concentrations of liver enzymes (alanine aminotransferase, 98 U/L [RR, 0–45 U/L]; aspartate aminotransferase, 70 U/L [RR, 0–41 U/L]; and lactate dehydrogenase, 268 U/L [RR, 80–250 U/L]), with normal serum bilirubin and alkaline phosphatase levels. Her serum iron level was low (2 μmol/L [RR, 10–33 μmol/L]). A core biopsy (20 mm × 1 mm) of one of the right pleural masses showed morphological and immunohistochemical features of a poorly differentiated epithelial mesothelioma. A pathology report by an experienced pathologist with a special interest in pulmonary and pleural pathology read: There is a proliferation of poorly cohesive large cells many of which had vesicular nuclei, prominent nucleoli and abundant eosinophilic cytoplasm. Occasional binucleate and multinucleate forms are present and there is a small amount of associated collagenous stroma with a mild chronic inflammatory cell infiltrate [Box 2A]. There is no evidence of mucin production, and immunoperoxidase stains for a variety of keratins are strongly positive, along with positive staining for calretinin [Box 2B] and cytokeratin 5/6, both markers of mesothelial differentiation [Box 2C]. Stains for LCA and S100 protein are negative. The pathology results were later reviewed by another pathologist with considerable experience of mesothelioma, who drew the same conclusion. ManagementThe patient was told the diagnosis and referred to an oncologist with a special interest in mesothelioma in another tertiary hospital where clinical trials of drugs for the treatment of mesothelioma were in progress. She was offered chemotherapy and entry in a thalidomide trial, but, by the time she was entered, the tumour was already showing signs of spontaneous regression. The patient opted for no treatment, as she felt well. By 30 December 2002, a repeat CT scan of the chest showed a decrease in the size of the large, right-sided pleural mass in the costovertebral gutter from 17 mm × 9 mm (CT chest scan, 29 November 2002) to 12 mm × 6 mm. The second pleural mass was now so small it was difficult to see. The third mass was not visible, and there were two small intrapulmonary nodules — one in the right middle lobe and the other in the left lower lobe. By March 2003, the first and second pleural masses were even smaller (Box 1C) and, by June 2003, they had disappeared (Box 1D). A CT chest scan in June 2004 was normal except for the two tiny intrapulmonary nodules that had not changed in size and were probably granulomas. The patient was last reviewed in June 2007 and was in good health, with no evidence of tumour, and was scheduled for next review in 6 months. DiscussionThere have been a few reports of spontaneous regression of malignant mesothelioma, but prolonged, disease-free periods are rare. Our report appears to be the first to describe a patient in Australia with a poorly differentiated, multifocal epithelial mesothelioma that regressed spontaneously, with the disease remaining in remission for 5 years. There has been one case report of a patient with malignant mesothelioma of the pleura that regressed spontaneously, but after 6 years there was a single recurrence, which was resected surgically, and the patient was followed up for a total of 12 years.1 This raises the question in such cases of the duration of follow-up. In the latter case, a prominent host response to tumour was seen in both the primary tumour and the recurrence. In another case, a patient had a spontaneous remission of a malignant peritoneal mesothelioma, and had high spiking fevers when the tumour recurred.2 A report from Western Australia described a woman whose tumour regressed spontaneously but who eventually died 20 months later.3 It was noted that the tumour tissue was infiltrated with mononuclear cells, and as the tumour recurred some malignant mesothelioma antigens disappeared. Several aspects of our case should be noted. Histopathology The histopathological findings for our patient’s tumour were re-examined by another pathologist. Not all three lesions were biopsied, as it was felt highly probable that the pathological findings for all three would be identical. Author’s experience I have considerable experience in the management of mesothelioma and benign asbestos-related conditions and currently see about 600 patients with this condition a year. I also act as an expert witness for the courts. Mesothelioma is a relatively common condition in Australia at present and, as a result, our thoracic physicians, oncologists and pathologists have considerable experience in this area. Course of the disease The patient had markedly elevated inflammatory markers and is likely to have had the tumour for several months before diagnosis. It is suspected that her natural killer cells and cell-mediated immunity accounted for the regression of the tumour. Evidence of the beginning of spontaneous regression was unusually rapid, occurring within months of diagnosis. Exposure Our patient’s exposure to asbestos seems to have been relatively mild and incidental, as often occurs with women who develop mesothelioma. Self-treatment The patient did not use any unusual therapies, such as alternative medicines, diets and faith healing, after the diagnosis. This case highlights the possibility that spontaneous regression of mesothelioma may occur occasionally. Spontaneous tumour regression therefore should be seen as part of the spectrum of the natural history of mesothelioma and other tumours. A detailed study of the immunity of such individuals “after the event” is unlikely to reveal any particular abnormality but, in hindsight, it would have been interesting to have performed detailed immunological studies during the initial regression period. The role of mesothelin-related serum proteins needs further evaluation. The understanding of this process is likely to be pivotal in the improved treatment of this usually lethal condition. 1 Computed tomography (CT) scans of the chest at referral (September 2002) and 6 and 9 months later A, B: CT pulmonary angiogram at referral (September 2002) showing (A) the pleural-based mass in the right costovertebral gutter at the level of the tracheal bifurcation (arrow) and (B) the two lower pleural-based masses — one in the right cardiophrenic angle and the other in the right posteromedial costophrenic recess (arrows). C: CT scan (March 2003) showing that the pleural-based mass in the right costovertebral gutter had all but disappeared (arrow). D: CT scan (June 2003) which appeared normal apart from a small stable nodule in the left lower lobe which was probably a granuloma (arrow). 2 Histological examination of a core biopsy specimen from one of the pleural masses (Box 1) A: Histological section of the core biopsy of the pleural mass in Box 1A showing sheet-like proliferation of pleomorphic epithelioid cells with abundant eosinophilic cytoplasm, representing a poorly differentiated epithelioid malignant mesothelioma. B: Core biopsy showing a positive result on immunoperoxidase staining for calretinin. C: Core biopsy showing a positive result on immunoperoxidase staining for cytokeratin 5/6.

Roger K A Allen FRACP, FCCP, PhD

Clinical records Patient 1 A 27-year-old man presented to an infectious diseases outpatient clinic in May 2006 with a 3-month history of nocturnal dry cough, paroxysmal dyspnoea, malaise and unintentional weight loss of 5 kg. He had previously presented to a general practitioner and was treated with two courses of antibacterial drugs without a decrease in symptoms. He had no significant past history of chronic respiratory illness or tuberculosis, was a non-smoker and had been working in Australia for 5 years. He was born in India and had recently returned there on a holiday to visit family and friends. Physical examination, including respiratory system examination, was unremarkable. Full blood examination revealed marked eosinophilia (28.8 × 109/L; reference range [RR], 0.0–0.5 × 109/L). A chest x-ray and computed tomography (CT) showed a diffuse, bilateral fine micronodular pattern throughout both lung fields (Figures A and B). The patient had a raised serum IgE concentration of 24 020 kU/L (RR, 0–120 kU/L), and was positive for filarial IgG by enzyme immunoassay, but negative for filaria on a midnight blood smear. Strongyloides serological tests were also positive, but treatment with two doses of ivermectin did not resolve the nocturnal cough. Three stool specimens were examined for helminths, all of which were negative. The patient was given antifilarial treatment with diethylcarbamazine (150 mg three times daily) for 14 days. Symptoms decreased rapidly, and the eosinophil count was nearly normal (2.7 × 109/L) by 4 weeks. Tropical pulmonary eosinophilia was diagnosed on the basis of the clinical syndrome, positive serological results for filaria, exclusion of other parasitic infections and successful clinical response to a trial of antifilarial treatment. Patient 2 A 25-year-old woman from Sri Lanka presented to the Royal Melbourne Hospital emergency department in September 1994 with a 3-day history of productive cough, pleuritic chest pain and increasing exertional dyspnoea. She had arrived in Australia 6 months previously. Empirical treatment with salbutamol and doxycycline prescribed by her GP had not decreased the symptoms. Examination revealed scattered bilateral expiratory wheezes. A full blood examination revealed an eosinophil count of 21.5 × 109/L, and the initial chest x-ray showed diffuse pulmonary infiltrates, which were confirmed on CT. A bronchoscopy was performed to investigate these pulmonary lesions, which showed pus cells but no visible parasites. An open lung biopsy was then performed, before knowledge of relevant serological results, which revealed an eosinophilic infiltration of the alveolar spaces, suggestive of an eosinophilic pneumonia (Figures C and D). Results of other investigations included three negative stool specimens, a strongly positive serological result for filaria, negative blood film for microfilariae, and a raised serum IgE concentration (28 400 kU/L). Serological tests for schistosomiasis and strongyloides were also both positive at low titres. Given the marked eosinophilia, widespread chest infiltrates and strongly positive filarial serological result, the most likely diagnosis was thought to be tropical pulmonary eosinophilia. The patient was treated with diethylcarbamazine (100 mg three times daily) for 21 days. After some initial nausea, her symptoms decreased, and the eosinophil count was resolving (0.7 × 109/L) by 8 weeks after treatment. Patient 3 A 30-year-old man from India presented to an infectious diseases outpatient clinic via migrant screening in October 2006 with an abnormal appearance on chest x-ray, which showed fine reticulonodular opacities throughout both lung fields. The patient was born near Calcutta and had arrived in Australia 3 months previously. He was a non-smoker and reported a history of non-productive cough over several days. Physical examination was unremarkable including the respiratory system examination. A full blood examination revealed eosinophilia (13.0 × 109/L) and a positive serological result for filarial IgG. Results of other investigations included a raised serum IgE concentration (> 5000 kU/L), a positive serology result for strongyloides and negative serology results for schistosomiasis and toxocara, a negative immunochromatography result for Wuchereria bancrofti, and three negative stool specimens to particularly exclude strongyloides. A blood film did not show microfilariae. Pulmonary function tests showed moderate restriction (forced vital capacity, 3.0 L, or 67% of reference range) without obstruction and normal gas transfer. The patient was initially treated for strongyloides infection with ivermectin, but the eosinophilia persisted. Diethylcarbamazine (150 mg three times daily) was given for 14 days. Within 2 weeks, the eosinophil count had dropped to 1.0 × 109/L. The patient was clinically well at follow-up 3 months later. A: Patient 1 — chest x-ray showed diffuse fine nodules. B: Patient 1 — computed tomography showed a widespread, bilateral fine micronodular pattern. C, D: Patient 2 — low and high magnification (× 200 and × 400) views of a lung biopsy specimen showed eosinophilic infiltration of alveolar spaces (haematoxylin and eosin stain). Tropical pulmonary eosinophilia is a rare but well recognised syndrome characterised by pulmonary interstitial infiltrates and marked peripheral eosinophilia. We report three cases of this syndrome presenting with cough in immigrants to Australia, to highlight awareness of this treatable infectious disease. This condition is more widely recognised and promptly diagnosed in filariasis-endemic regions, such as the Indian subcontinent, Africa, Asia and South America. In non-endemic countries, patients are commonly thought to have bronchial asthma.1,2 Chronic symptoms may delay the diagnosis by up to 5 years.1 Early recognition and treatment with the antifilarial drug, diethylcarbamazine, is important, as delay before treatment may lead to progressive interstitial fibrosis and irreversible impairment.3 Lessons from practice Tropical pulmonary eosinophilia should be considered in patients who have lived in filaria-endemic countries, such as the Indian subcontinent, and present with respiratory symptoms and hypereosinophilia. The most common misdiagnosis is asthma, with overlapping symptoms of chronic cough, paroxysmal dyspnoea and wheeze. Early diagnosis and treatment with diethylcarbamazine (DEC) may prevent progressive pulmonary disease. The condition of marked eosinophilia with pulmonary involvement was first termed tropical pulmonary eosinophilia in 1950.4 The syndrome is caused by a distinct hypersensitivity immunological reaction to microfilariae of W. bancrofti and Brugia malayi.3,5 However, only a small percentage (< 0.5%)6 of the 130 million people globally who are infected with filariasis apparently develop this reaction. The clearance of rapidly opsonised microfilariae from the bloodstream results in a hypersensitive immunological process and abnormal recruitment of eosinophils, as reflected by extremely high IgE levels of over 1000 kU/L.3,7 The typical patient is a young adult man from the Indian subcontinent.5 The diagnostic criteria for tropical pulmonary eosinophilia7 include: history supportive of exposure to lymphatic filariasis; peripheral eosinophilia count (> 3 × 109/L); elevated serum IgE levels (> 1000 kU/L); increased titres of antifilarial antibodies; peripheral blood negative for microfilariae; and clinical response to diethylcarbamazine. High antifilarial IgG titres to microfilariae often result in cross reactivity with other non-filarial helminth antigens,8,9 such as strongyloides and schistosoma antigens, as demonstrated in our reported cases. It is important to exclude other parasitic infections before tropical pulmonary eosinophilia is diagnosed, by serological tests, examination of stool specimens in a laboratory experienced in parasitic infections, or a trial of antihelminth medication. Other parasitic infections, such as the zoonotic filariae, dirofilariasis, ascariasis, strongyloides, visceral larva migrans and hookworm disease, may also be confused with tropical pulmonary eosinophilia because of overlapping clinical features, serological profile and response to diethylcarbamazine3,7,9,10 (Box 1). Radiological findings are non-specific, with normal appearance on chest x-ray in up to 20%.5 Although lung biopsy was performed in Patient 2, it is not part of the routine diagnostic work-up of tropical pulmonary eosinophilia. No universal treatment guidelines have been established for tropical pulmonary eosinophilia.1,7 The antifilarial diethylcarbamazine (6 mg/kg/day for 21 days6) remains the main therapeutic agent and is generally well tolerated. Reported side effects include headache, fever, pruritis and gastrointestinal upset.11 The eosinophil count often falls dramatically within 7–10 days of starting treatment.3 Diethylcarbamazine is available only through the Special Access Scheme of the Therapeutic Goods Administration. Symptoms persist after treatment in up to 25% of patients.5 The role of adjunctive therapy with corticosteroids in preventing long-term fibrosis has not been studied. Our three cases demonstrate the variable clinical presentations and symptom duration of tropical pulmonary eosinophilia. With increased travel and migration of patients from filaria-endemic areas, physicians need to remain aware of tropical pulmonary infections presenting with cough, dyspnoea and variable systemic symptoms, as delayed recognition of this uncommon clinical entity may increase morbidity. 1 Characteristic features of parasitic infections with pulmonary symptoms and eosinophilia Condition Parasite Respiratory symptoms Geographical distribution Laboratory diagnosis Treatment Tropical pulmonary eosinophilia Wuchereria bancrofti, Brugia malayi Nocturnal cough, wheeze, dyspnoea Tropical and subtropical areas, especially India and Sri Lanka Serology, blood film, IgE levels Diethyl- carbamazine Strongyloides Strongyloides stercoralis Loeffler’s-like syndrome,* hyperinfection syndrome Tropical and subtropical, including northern Australia Serology, stool Ivermectin Schistosomiasis Schistosoma mansoni, Schistosoma haematobium, Schistosoma japonicum Katayama fever, pulmonary hypertension, cor pulmonale Asia, Africa, South America Serology, stool and urine Praziquantel Ascariasis Ascaris lumbricoides Loeffler’s-like syndrome* Asia, Africa, South America Stool Albendazole Visceral larva migrans Toxocara canis, Toxocara cati Eosinophilic pneumonia, wheeze, dyspnoea Worldwide Serology Albendazole Dirofilariasis Dirofilaria immitis Pulmonary lesion Tropical and subtropical, including Australia Serology None Hookworm disease Ancylostoma duodenale Loeffler’s-like syndrome Tropical and subtropical areas Stool Albendazole * Loeffler’s-like syndrome: transient pulmonary infiltrates and eosinophilia from transpulmonary passage of helminth larvae.

Michelle K Yong MB BS · Caroline L Marshall FRACP · Damon P Eisen FRACP, MD

To the Editor: The Journal’s unbalanced supplement on spirituality and health fails to satisfy your policy on sponsored supplements, cites at least one fraudulent study, and contains much poor science and non-science. I discuss here only a fraction of the supplement’s flaws. Firstly, dissenting voices were not cited or discussed in the supplement. An objective appraisal of the field would have included sceptical viewpoints such as those of Paul,1 who demonstrated lower levels of societal dysfunction in highly secular democracies than in more religious societies such as that of the United States. It is remarkable that the supplement article by Williams and Sternthal2 ignored Paul’s study. Secondly, Jantos and Kiat3 cite Cha and Wirth’s debunked Columbia University study into the relationship between intercessory prayer and fertility rates for in-vitro fertilisation treatment. Flamm, a Californian professor of obstetrics and gynaecology, demolished this article,4,5 and his rebuttal was reported widely.6 That this citation survived the peer review process suggests either that the reviewers did not know their field well enough, or that they deliberately allowed unqualified citation of a fraudulent study. Either way, they failed in their role as reviewers. Thirdly, Jantos and Kiat state that scientific investigation of prayer may not be possible, adding that scientists “must” accept that “some aspects of prayer . . . may go beyond the reach of science”. (Yet prayer’s putative physical effects must be measurable!) They also regard bible stories of Jesus’ healings as scientifically valid observations, stating that “All were examples of healing by supernatural means” — an unsupported, unscientific statement of belief that has no place in a peer-reviewed scientific journal. Similarly, Eckersley7 states that “we are spiritual beings, psychically connected to our world”. It is extraordinary and lamentable that statements such as these survived the editorial process. Finally, potential authorial conflict of interest is not disclosed. Koenig8 is the Co-Director of the Center for Spirituality, Theology and Health at Duke University Medical Center, a significant role that is not noted in his author details. Despite Jantos and Kiat’s assertions about the limitations of science, Koenig’s Center supports many studies of prayer. The well funded, US-based push to research the interface between religion and science, especially medical science, jeopardises scientific integrity. Its apotheosis, the Templeton Prize, is — at $US1.5 million — the world’s richest academic prize.9 The Journal’s supplement is best perceived as being a part of this agenda. It is an indictment on the MJA as a scientific journal that it was published.

Chris O Jackson

To the Editor: I take issue with the presentation of evidence by Jantos and Kiat1 — firstly, on the effect of intercessory prayer on health, and secondly, on prayer as a supernatural intervention. The minor positive findings of the Byrd study2 on prayer for coronary care patients, given so much column space, have proved non-reproducible.3 When meta-analysis is applied to the review by Astin et al4 of randomised trials of “distant healing”, the quoted “inconsistent” results of prayer become most definitely non-significant.5 The largest and most robust trial of prayer, by Benson et al (involving 1802 subjects),6 showing no positive effect of prayer on recovery after heart surgery, is mentioned but somewhat dismissed by Jantos and Kiat. As for the Cha and Wirth study on prayer and in-vitro fertilisation7 cited by the authors, simple investigation reveals it to have been an embarrassing fraud. The article was subsequently removed from the journal that published it, and one of the authors went to jail. In the section entitled “Plausible mechanisms by which prayer delivers health benefits”, the paragraph on “supernatural intervention” includes bible quotations on healing miracles presented as “evidence”. This may constitute sectarian theological material, but it is not medical science. Analogous to the financial interests of authors, religious groups have their own vested interest in the outcome and interpretation of medical studies involving religious issues. This is due to the intrinsic nature of religious faith, whereby a point of belief constitutes an absolute truth to the believer, irrespective of any other data, but seems implausible to non-believers. I would suggest that, for the benefit of a secular readership, in articles concerning religion and medicine in the Journal, the Editor should require the authors’ religious position to be stated under “competing interests”.

Jon Clarke

To the Editor: In your recent supplement on spirituality and health, I note that none of the authors cited competing interests. However, several authors gave their affiliations as Loma Linda University, an institution owned and operated by the Seventh-day Adventist Church, a fundamentalist Christian sect with strong evangelical and millenarian beliefs. I also note the financial support given to the publication by the same church. Does the above not constitute “competing interests”?

Ross B Holland

To the Editor: It is an embarrassment to your Journal that an article such as Jantos and Kiat’s “Prayer as medicine: how much have we learned?”1 should have been allowed publication. The article neglects to apply scientific rigour to the topic of prayer research in failing to effectively review the most significant and largest studies on the efficacy of prayer.2-5 The results of the largest study of third-party prayer, which suggested such prayer was ineffective in reducing complications following heart surgery, were noticeably absent.2 The article by Jantos and Kiat1 begins by suggesting that a spiritual search for meaning and hope is integral to human existence. This may be true for some, but certainly not all — which means that it can not be “integral” to human existence. Their abstract asserts the efficacy of prayer, without showing a causal relationship between prayer and improved outcomes anywhere in the article. Articles discussing the efficacy of prayer should include, if not an original study, a meta-analysis and interpretation of existing studies. The article also outlines “plausible mechanisms by which prayer delivers health benefits”, one of which includes the claim that it could in fact involve “supernatural intervention”. In a scientific publication, such a suggestion deserves thorough scientific evidence. Instead, all the authors provide is an anecdotal story and a bible passage. The article also uses the phrase “critics of prayer research”, presumably to describe people who are critical of the efficacy of prayer. A distinction is important, because critics of the efficacy of prayer are not necessarily critical of the research. In fact, critics would most likely encourage research so that they can, if evidence warrants, show how ineffective prayer is. Furthermore, the statement “prayer may not be transparent to scientific investigation and may go beyond the reach of science” begs the question: what, if prayer is beyond the realms of science, is this article doing in a scientific journal? Your publication has lent undue credibility to a gratuitous article without scientific substance.

Lahn D Straney

To the Editor: I am seeking some clarification on the recent MJA supplement article by Jantos and Kiat.1 One of the mechanisms suggested by the authors for the alleged beneficial effect of prayer is that it is “a channel for supernatural intervention”. Can I just clarify with you, given the vagueness of the statement and the religious overtone of the paragraphs that followed it, whether the authors were implying that there is a personal, caring God who performs supernatural interventions for people who pray — ie, that the beneficial effects are a direct result of such interventions. Or are they saying that the belief in the existence of such a being is itself the plausible mechanism — in which case, it should be more appropriately classified as a “placebo” effect (ie, the second mechanism listed in their article).

Tom Huang

To the Editor: I read with interest the recent MJA supplement on spirituality and health. It is interesting to observe the relationship between religious affiliation and life expectancy at birth over the past 100 years using population data supplied by the Australian Bureau of Statistics (Box).1,2 The life expectancy data shown here is for males, but the graph is almost identical for females. While there are many variables in this relationship, they are akin to those chiefly neglected in most of the articles in the supplement. It is easy to see from the graph that, as religious affiliation within the community has declined, life expectancy (a gross surrogate measure of health) has increased. The correlation is very good. Religious affiliation versus life expectancy at birth in Australia1,2* R2 = correlation coefficient. * Lines on graph are lines of best fit.

Robert F Grace

In reply: The spirituality and health supplement was a compendium on religion and spirituality in clinical practice, based on recent presentations at the National Spirituality and Health Conference. The authors sought to highlight clinically relevant research exploring associations between religiosity and health. Clarke and Jackson express concerns about the quality of the literature on intercessory prayer. Yet all the studies on intercessory prayer cited in our article on prayer and medicine1 belong to references included in the 2007 Cochrane database systematic review on the subject.2 As stated by Clarke, the pioneering work of Byrd was given prominent mention, by being the first of several studies on intercessory prayer employing a prospective, randomised, double-blind protocol. Clarke and Jackson question the ethics of not declaring authors’ religious affiliations. This factor (along with other personal factors not listed as “competing interests”, such as race, sex and age) should not influence the clarity, objectivity, and validity of peer-reviewed scientific publications, nor impair the authors’ objectivity, integrity and performance as clinicians. Their concern about the legitimacy of citing biblical verse is also unfounded. The medical literature is replete with biblical citations. A recent publication on the topic of biblical origins of placebo3 is just one example from the MEDLINE database. Straney expresses concern regarding the clinical relevance of prayer. More than 88% of the world’s populations believe in the supernatural,4 and prayer, being one of the oldest and most widely practised spiritual rituals, is frequently practised by patients when they have health problems.1 Knowing that patients commonly resort to prayer as a means of coping and finding meaning in life obliges clinicians to have some insight into how this practice affects patients’ health. Courses in spirituality and medicine are now offered in medical schools in the United States.5 In examining the relationship between belief in supernatural agents and mortality, Norenzayan and Hansen6 concluded: “our findings support the idea that belief in the supernatural agency is a core response to the human awareness of mortality” (original authors’ emphasis). Their conclusion supports the view that the spiritual search for meaning and hope in life is integral to human existence. Such exploration is virtually universal, albeit to varying degrees of depth, length and frequency. The study by Benson et al that Straney alludes to as being absent from our article on prayer as medicine1 was in fact cited (reference 12) and discussed in the body of the text. Our article reviewed the plausible mechanisms by which prayer may benefit individuals who engage in the practice. The postulation of a mechanism of action does not imply an explanation of aetiology. For example, a postulation that the resolution of an infection occurs through bacterial mitotic inhibition would neither prove nor disprove that a particular antibiotic is the agent of healing. Thus, Huang’s contention that “belief in the existence of such a being [(God)] is itself the plausible mechanism . . . [and] should be . . . classified as a ‘placebo’ effect” is a naturalistic fallacy. Grace cited data on religious affiliation and life expectancy among Australians. However, these data have not been subject to systematic and appropriate analysis for possible association. In contrast, religiosity was positively correlated with longevity among 10 000 Israelis over a 23-year period.7 Similar results were obtained in a 28-year follow-up of 5000 Californians.8 Religiosity was also shown to be an independent risk factor in a prospective study of over 300 elderly patients having cardiac surgery:9 compared with patients who reported preoperatively that they derived strength and comfort from their religious beliefs, those who didn’t were found to have a threefold greater risk of perioperative and 6-month mortality. Furthermore, a study of 34 000 Seventh-day Adventists in California10 revealed that they outlived other white male and female Californians by 7.3 and 4.4 years, respectively, giving them probably the highest life expectancy of any formally described population. Improved health practices and social connection are probably factors that provide a causal link between religious affiliation and health or longevity. In an era of spiralling health care costs, the big picture is the potential public health implications of religiosity as a low-cost health measure. Individuals or bodies responsible for health care policy, budgeting and delivery should welcome further exploratory research into the preventive and therapeutic effects and cost-effectiveness of religious practice in health promotion within appropriate community settings.

Hosen Kiat · Marek Jantos

In reply: The MJA’s spirituality and health supplement has reawakened the long-standing tension between supporters of science and supporters of spirituality, and I welcome their conflicting and unaccommodating views. The barrage of letters received reflects a perceived incongruity between spirituality and evidence-based medicine, with its requirement for evidence that is controlled, measured, counted and analysed by statistical methods.1 But the task of physicians has always been to understand not only the disease but also the patient. And for some patients, religion and spirituality are important — or, indeed, central — to their lives and health. The purpose of the supplement was to explore this area. Jackson is perturbed that, in publishing the supplement, the Journal has sold its scientific soul. However, one of the goals of the Journal is to provide a scholarly forum for continuing education and informed debate on standards of clinical practice, ethics, and social, legal and other issues related to health care in Australia.2 While it must be acknowledged that religion and spirituality involve areas of knowledge not governed by the scientific method, it would be intellectually dishonest to refuse to consider any role for spirituality in health care. Jackson also alleges that the supplement fails to satisfy the Journal’s policy on sponsored supplements, but proffers no evidence. In fact, the supplement complies completely with our policy for such publications.3 Despite the ongoing conflict between science and spirituality, exploration of the latter is increasingly considered for inclusion in modern medical curricula.

Martin B Van Der Weyden MD, FRACP, FRCPA

Women with diabetes Improvements in mortality trends for people with diabetes in the United States appear to be limited to men, with the disease continuing to greatly increase the risk of death for women. Using data from the National Health and Nutrition Examination Surveys spanning the period from 1971 to 2000, researchers analysed mortality data in self-reported diabetics and compared the results with the non-diabetic population. Men with diabetes showed a 43% relative reduction in mortality rate, a result similar to that for non-diabetic men. Mortality rates did not decrease among women with diabetes. Moreover, the difference in mortality rates between diabetic and non-diabetic women doubled. Limitations of the study include the reliance on self-reporting of diabetes, changes in diagnostic criteria over time, and the relatively small samples in the sex-specific analysis. Despite these problems, the authors surmise that the observed trend in mortality is significant, and postulate that the result may be secondary to differences in the pathophysiology, diagnosis, and treatment of cardiovascular disease and diabetic complications in women. Ann Intern Med 2007; 147: 149-155 Calcium for healthy bones Calcium and vitamin D supplements appear to reduce bone loss and lower the risk of fractures, according to Australian researchers. In a meta-analysis of 29 randomised controlled trials involving over 63 000 individuals, outcome measures of bone mineral density and fractures were analysed. Results showed that calcium supplementation, alone or in combination with vitamin D, appears to be effective in prevention of osteoporotic fractures. Treatment over an average duration of 3.5 years was associated with a 12% risk reduction in fractures of all types and a reduced rate of bone loss of 1.19% in the spine. Daily doses of calcium over 1200 mg and of vitamin D over 800 IU were associated with a greater risk reduction. The improvement in risk reduction was greater for those older than 70 years and for participants whose daily calcium intake was previously low. The authors comment that although results of previous trials have been inconclusive, the large number of patients included in this meta-analysis provides adequate statistical power to detect a treatment effect. Lancet 2007; 370: 657-666 Selenium supplements Despite findings from animal models suggesting that selenium supplementation may improve glucose metabolism, it does not appear to be protective against type 2 diabetes in human subjects. Using secondary analysis of a randomised, double-blind, placebo-controlled trial, researchers aimed to determine whether selenium supplementation had a protective effect against diabetes over an average follow-up period of over 7 years. The trial, involving over 1000 people, was initially designed to assess the effects of selenium intake on the incidence of skin cancer. The development of diabetes was a secondary outcome of the original trial. Not only did selenium supplementation of 200 μg daily fail to show a benefit in reducing type 2 diabetes, the cumulative incidence of diabetes was higher in those receiving selenium compared with patients given a placebo. The authors point out limitations of their study, including the use of a secondary end point, the self-reporting of diabetes, and a lack of information on possible unmeasured confounding risk factors. Despite these problems, the researchers conclude that selenium supplementation does not seem to prevent type 2 diabetes, and could actually increase the risk of developing the condition. Ann Intern Med 2007; 147: 217-223 AF: to screen or not to screen Detection and treatment of atrial fibrillation (AF) leads to a dramatic reduction in negative sequelae such as stroke. British researchers aimed to determine the most effective screening method for AF by conducting a cluster randomised controlled trial involving over 14 500 participants.1 Primary care practices were randomly assigned to non-screening and screening arms. Patients over 65 years of age seen at screening practices were randomly allocated to either systematic screening (invitation for electrocardiography [ECG]) or opportunistic screening (pulse-taking by GP and ECG if pulse was irregular). Using newly identified AF as the main outcome measure, the detection rate of AF was assessed for each method. Active screening was found to detect significantly more cases than no screening. There were no differences in the rate of detection of AF between the two screening methods. The authors conclude that the greater cost, labour intensiveness, and intrusiveness of systematic screening are not justified and that opportunistic screening is the preferred method. Separate research from the same UK-based group focuses on the accuracy of diagnosing AF on ECG by GPs and interpretative diagnostic software.2 The results show that primary care practitioners have trouble detecting AF on ECG and that the addition of interpretative software does not necessarily address this problem. They conclude that strategies for community screening and diagnosis of AF must include appropriate training in reading the electrocardiogram. 1 BMJ 2007; 335: 3832 BMJ 2007; 335: 380

Tanya Grassi

Med J Aust 2007; 187 (7 Suppl).