Letters

To the Editor: As Clay-Williams and colleagues,1 we have also experienced frustration at the time-consuming, expensive2 and protracted processes required as a prerequisite before undertaking research involving identified patient data across Australia. Ironically, in the case of clinical quality registries, we collect data designed to measure quality of care so that patients may benefit from improved clinical processes. Yet, the system conspires to delay and undermine these efforts. There are lessons we have learned in developing clinical quality registries that deal with some governance issues identified by Clay-Williams and colleagues. They describe developing a legal agreement, which 21% of hospitals refused to accept. The Southern Eastern Border States (SEBS) Committee was developed with representatives of health departments from Victoria, New South Wales, Queensland and South Australia to streamline and prevent duplication of legal agreements.3 We worked with a SEBS Committee representative to develop special clauses and conditions to include in the Medicines Australia Clinical Trials Research Agreement schedule, subsequently endorsed by the SEBS Committee. Participating sites, both public and private, recognise and routinely accept this agreement. This same committee, or one with a similar construct, could determine nationally whether research is low risk to address the confused and contradictory advice given by ethics committees. We agree that standardisation of forms and processes is required — participating sites should not be permitted to introduce their own forms, causing confusion to researchers and introducing additional delays. A national information portal and repository for protocol amendments which governance officers can access to review or approve changes would be welcomed. However, while Clay-Williams and colleagues recommend that there should be no requirement for principal investigators to be employees of the participating site, we believe that for clinical quality registries, having a local principal investigator provides a vital link between researchers and local staff. As previously suggested,4 limits should apply on an acceptable time period to provide a definitive determination on study conduct within an institution, as occurs in Europe with clinical trials.5 It is unacceptable that this process lasts more than 60 days. As researchers usually have no capacity to have an impact on institutional authorisation processes, greater accountability should vest with governance offices. The current costly and dysfunctional system is stifling research in Australia.

Sue M Evans · John R Zalcberg · Ri Scarborough

In reply

Jeffrey Braithwaite · Robyn Clay-Williams · Natalie Taylor

To the Editor: In their recent article, Shulruf and colleagues1 concluded that prior academic achievement constituted the most effective means of predicting timely graduation. This outcome measure overlooks a more important graduate attribute: professionalism — the values and skills that the profession and society expects of doctors. This attribute is identifiable at admission, and it is possible to test for and select for this.2 When considering the desired outcome of satisfactory performance at junior medical officer level, the authors state that the “outcome of subsequent workplace performance, while important, is moderated by influences beyond the undergraduate environment”.1 This statement contradicts extensive literature suggesting otherwise. An erosion of vicarious empathy during medical education programs is well documented and is evident across other health care professions.3 Medical school education fails to consistently foster the development of advanced moral reasoning in medical students, with particular problems developing during the period of clinical immersion, when the influence of the “hidden curriculum” becomes evident to students.3 The above have been linked to the experience of burnout in students, which can manifest as professionalism lapses in both pre-clinical and clinical rotations.4 Papadakis and colleagues5 suggested that disciplinary action by a medical board was strongly associated with prior unprofessional behaviour in medical school. Most complaints against doctors are due to conduct, not competence. Many organisations have developed guidelines to ensure medical students adhere to professional standards. Academic and intellectual qualities alone cannot predict the ideal candidate for admission to medical school, and facets of professionalism such as moral orientation, resilience and self-control are acknowledged to contribute to one’s efficacy as a doctor. Furthermore, prioritising timely completion may promote students not seeking help, perpetuating poor performance in an effort to ensure timely completion. This perpetuates a workplace culture where people do not feel able to seek help, with significant repercussions as seen in the recent suicides of young doctors. Timely completion may indeed be predicted by prior academic success; however, attitudinal and behavioural factors are highly relevant at selection and throughout subsequent careers. An overemphasis on prior academic achievements may de-emphasise student characteristics associated with the development of professionalism.

Mark H Arnold · Jennifer Smith-Merry · Andrew S Lane

To the Editor: We are concerned about the emergence of antibiotic-resistant gonorrhoea in Australia. On 17 April 2018, the Commonwealth’s Chief Medical Officer Brendan Murphy released a statement about two cases of multidrug-resistant gonorrhoea that had recently been detected in Australia1 — at least one of these patients acquired the infection in Southeast Asia. These particular cases were similar to the one recently reported in the United Kingdom,2 where the patient was reported3 as having high level resistance to azithromycin and to ceftriaxone — the cornerstone of treatment. Also notable in that case was treatment failure using spectinomycin, with ongoing detection of the bacterium on throat swab. Treatment with intravenous ertapenem was successful. It should be of concern that gonorrhoea may only have the option of intravenous treatment, but the real problem here is that we may be on the precipice of untreatable gonorrhoea. With almost 750 000 short term resident returns every month,4 and over 200 000 of these being returns from Southeast Asia, the likelihood of repeated introductions is real. In the current context of rising gonorrhoea rates in Australia,5 further importation, transmission and spread of these resistant organisms will add substantial challenges to the disease control, especially in men who have sex with men and in Indigenous Australians. Such spread will incur significant health and health care costs for the sexual and reproductive health of Australians. There is an urgent need for all treating doctors to ensure that swabs for culture are taken for all symptomatic patients, as well as for those with an initial positive polymerase chain reaction result; for travellers to be aware of the risks of having unprotected sex; and for any multidrug-resistant patients and contacts to be referred for expert advice to ensure testing and treatment.

Brett Sutton · Mihaela Ivan

To the Editor:I read with interest the article by Brukner and colleagues1 on traumatic cricket-related fatalities in Australia, which describes two autopsy-confirmed deaths due to subarachnoid haemorrhage following vertebral artery dissection, with a further 11 deaths suspected to be secondary to this condition. Two recent articles described a total of 230 cases of carotid or vertebral artery dissection temporally related to 45 different sports or recreational activities.2,3 The majority of episodes of arterial dissection were related to non-contact sports, including jogging, walking, swimming, golf, basketball, tennis and scuba diving. The mean age of patients was 35 years. The mechanism of non-traumatic dissection is thought to relate to shearing stress on the arterial wall with sudden neck rotation. Thus arterial dissection in golfers affected the right side in 11 of 14 patients (79%), and involved the posterior circulation in 12 of 14 patients (86%).4 Controversy regarding the association between neck manipulation and arterial dissection persists, although a retrospective case–control study found an odds ratio of 12.8 for prior neck manual therapy in individuals aged 55 years or less presenting with craniocervical arterial dissection.5 Arterial dissection may also occur spontaneously, the risk being increased in the setting of systemic lupus erythematosus, other connective tissue disorders, migraine and in the postpartum period. It is important that health professionals recognise that arterial dissection may occur spontaneously or as a result of non-contact sports and activities, and that persons of any age presenting with symptoms suggestive of anterior or posterior circulation ischaemia require urgent review and neuroimaging.

Adam Morton

To the Editor:A recent article by O’Brien and colleagues1 highlights the worsening epidemic of Buruli ulcer in Australia. The steep rise in both the incidence and severity of the disease is associated with estimated health care costs of over $2.5 million per year in Victoria.1 The increase in Buruli ulcer cases in Australia parallels the increase in non-tuberculous mycobacterial (NTM) infections, especially lymphadenitis and Buruli ulcer, reported worldwide. Although this rise might be partly attributable to improved awareness and diagnostic methods, it might also be related to the discontinuation of universal bacillus Calmette–Guérin (BCG) vaccination in settings where the rate of tuberculosis has declined. Routine vaccination with BCG through the school program was discontinued in Victoria in the mid-1980s. The live-attenuated strain of Mycobacterium bovis contained in BCG vaccine shares epitopes with NTM, which makes cross-protection plausible. Our recently published meta-analysis indicates that BCG vaccination has a protective effect against NTM.2 In particular, two randomised controlled trials provide strong evidence for protection against Mycobacterium ulcerans.3,4 However, immunity might only be short lived, as the highest protection was observed in the first year after vaccination. Nevertheless, studies also report that compared with BCG-naive individuals, those who have received the BCG vaccine have smaller skin lesions,4 a shorter duration to healing5 and protection against severe forms of Buruli ulcer with multiple skin lesions.6 Buruli ulcer is a serious condition, which, despite prolonged antibiotic treatment and surgical intervention, can lead to complications such as osteomyelitis and other crippling sequelae. In light of the worsening epidemic, the protective effect of BCG vaccination should not be overlooked.

Petra Zimmermann · Adam Finn · Nigel Curtis

To the Editor:Salmonella infection manifests as enteritis and enteric fever, predominantly acquired overseas. When required, therapy with azithromycin, ciprofloxacin or ceftriaxone is recommended by the Therapeutic guidelines: antibiotic;1 however, reduced susceptibility to fluoroquinolones in Asia limits the use of ciprofloxacin unless susceptibility is confirmed.2 The Australian Bureau of Statistics recorded a 546% increase in short term departures to Indonesia over 10 years, with 1.2 million nationally in 2016.3 A 31-year-old man was taking long term azithromycin 250 mg daily to prevent bronchiolitis obliterans syndrome after a bilateral lung transplant several years earlier for cystic fibrosis. Three weeks after returning from Bali, he was admitted with fatigue, fever, diarrhoea and abdominal pain. He had acute kidney injury. His C-reactive protein level was 190 mg/L (reference interval [RI], < 5 mg/L) and procalcitonin concentration was 3.5 μg/L (RI, < 0.05 μg/L). A single set of blood cultures was negative. Stool culture isolated Salmonella enterica serovar Paratyphi B var Java, sensitive to ceftriaxone and ciprofloxacin, with a raised azithromycin minimum inhibitory concentration (MIC) of 64 mg/L by ETEST (bioMérieux); an MIC > 16 mg/L indicates non-wild-type4 and is associated with treatment failures. Owing to his immunocompromised state, the patient received a 14-day course of ciprofloxacin (MIC, 0.016 mg/L). Our review of 2015–2017 data from the Western Australian public pathology provider revealed that two of 31 typhoidal Salmonella isolates (Salmonella Paratyphi A and Salmonella Typhi bacteraemia, each acquired in India) and one of 15 non-typhoidal Salmonella isolates (S. typhimurium, no clinical details provided) had an azithromycin MIC > 16 mg/L. Ceftriaxone resistance was low at 0% (0/117) of typhoidal Salmonella and 0.4% (7/1648) of non-typhoidal Salmonella; ciprofloxacin resistance was higher at 48.0% (47/98) of typhoidal Salmonella and 6.8% (94/1384) of non-typhoidal Salmonella. By comparison, azithromycin MIC > 16 mg/L was found in 16.1% of typhoidal Salmonella from travellers returning to the Netherlands,5 and in 1.3% of non-typhoidal Salmonella in the United States.4 Azithromycin use while travelling probably selected for resistant Salmonella infection in this case. However, our data show that azithromycin susceptibility cannot be assumed in Salmonella infections; testing should therefore occur in serious cases, along with ongoing surveillance for evolving resistance.

Alan J Rogers · Gar-hing A Lee · Peter Boan

To the Editor:Based on the Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for mood disorders,1 Malhi and colleagues have produced guideline summaries for major depression2 and bipolar disorder.3 The major depression summary is likely to be used as a stand-alone guideline by primary care physicians treating depression.2 Bipolar disorder often presents as recurrent depression.1,3 It is therefore worrying that the major depression summary omits discussion of bipolar disorder. Identifying bipolar disorder is important because the treatment is different from that required for other types of depression.2,3 The bipolar disorder summary looks to the early detection of mania, as bipolar depression cannot be reliably distinguished from major depression.3 However, there are clues to a bipolar diagnosis during the preceding depressions. Bipolar disorder might be suspected in a pervasive depressive episode which does not make sense psychologically.4,5 A family history may also be suggestive.1 Bipolar disorder should be fully integrated into the management of major depression. After diagnosing major depression, ask routinely: “Could this be an episode of bipolar disorder?”. While the question cannot be answered definitively, it most definitely warrants the asking. Suspecting bipolar disorder can provide the patient and family with some explanation and can involve them in decisions about treatment. Suspecting bipolar disorder earlier could lead to a better outcome. It may allow earlier bipolar treatment and avoid exacerbating the condition with antidepressant or psychotherapy monotherapy. At least, we could warn patients of the risk of inducing mania or cycling with antidepressants.1,3 Omitting bipolar disorder from the guidelines for major depression means that such a warning may not be considered. The major depression summary2 recommends lithium, atypical antipsychotics and electroconvulsive therapy for treatment-resistant depression. Some psychiatrists believe that these are only effective in treating melancholic depression. For the more common treatment-resistant non-melancholic depression, stronger psychotherapies such as dialectical behaviour therapy and acceptance and commitment therapy may be more efficacious than stronger biological therapies. A final thought:Depression’s a broad diagnosis and mostly a kind of neurosis. But try to enable the bipolar label and really improve the prognosis.

Norman Zimmerman

In reply

Gin Malhi

To the Editor:The Editorial by Harvey and Diug1 on the value of food fortification as a public health intervention was prompted by demonstration of the effectiveness of mandatory iodine fortification in reducing iodine deficiency.2 Mandatory fortification of wheat flour for bread making was introduced in Australia at the same time to prevent neural tube defects. Harvey and Diug state that the two are conceptually different, as the former addresses a population iodine deficiency, whereas folic acid fortification is to compensate a presumed genetic defect that cannot be individually recognised, thus raising ethical questions about exposing the many for the benefit of the few. However, this is the case in almost all public health interventions. Using an example of Harvey and Diug, we expose the whole population to the mandatory fortification of flour with thiamine to prevent Wernicke–Korsakoff syndrome, a condition largely confined to people with a chronic alcohol problem. The concern they raise about mandatory folic acid fortification1 is exposure to unmetabolised folic acid, proposed as possibly increasing adverse effects, but which have not been clearly or conclusively shown. They refer to an Irish study reporting that seven of the 68 children in the study had detectable levels of unmetabolised folic acid in their blood.3 Ireland does not have mandatory folic acid fortification; the main sources of folic acid were voluntarily fortified breakfast cereals and fortified milk — products that are fortified with relatively high levels of folic acid. In Australia, breakfast cereals and other food products are also permitted to be fortified voluntarily — breakfast cereals often contain around 200–300 μg of added folic acid per 100 g (or about 100 μg per serve). This compares with mandatory fortification of flour of 200–300 μg per 100 g flour, or about 40 μg folic acid per slice of bread. While either source could lead to circulating unmetabolised folic acid, the higher doses in voluntarily fortified products are more likely to do so. With the introduction of mandatory fortification, there was a reduction in neural tube defects.4 Importantly, there has been a 68% reduction in previously higher rates of neural tube defects in Indigenous people.5 Mandatory fortification provides a more equitable, consistent and cheaper source of folic acid to the population than voluntary fortification. Let’s leave mandatory folic acid fortification preventing neural tube defects in our population.

Carol Bower · Fiona J Stanley · Mike Daube

To the Editor:Tick-borne infectious diseases, including rickettsial infections, acquired in Australia or after international travel remain a diagnostic challenge.1 A 68-year-old man presented with umbilical pain 10 days after returning from a 2-month camping trip through the south-west of the United States (ie, Texas, New Mexico, Arizona, Colorado and Utah). On examination, a live tick was detected and removed from the patient’s umbilicus (Box). The umbilical pain resolved after tick removal. There was no development of fevers, constitutional symptoms, or rash to suggest a tick-borne illness. Laboratory investigations were unremarkable. He was educated about the signs and symptoms of tick-borne illnesses and prescribed a single dose of doxycycline 200 mg for prophylaxis due to his high risk exposure. The tick was identified as Dermacentor andersoni (Rocky Mountain wood tick), which is endemic to North America and not known to occur in Australia.2 D. andersoni adult ticks are principal vectors of Rickettsia rickettsii (the cause of Rocky Mountain spotted fever), and are associated with transmission of other pathogens to humans, including Colorado tick fever virus and Francisella tularensis (the cause of tularemia).2,3 Although isolated from D. andersoni ticks, transmission of Coxiella burnetii (the cause of Q fever) is uncommon. D. andersoni is not known to transmit Lyme disease.2 Analysis of the tick for rickettsial DNA was positive. No Borrelia DNA was detected. Rickettsia was isolated in cell culture and identified as Rickettsia peacockii based on sequencing of the 17kDa, OmpB, gltA and Sca4 genes. R. peacockii is a member of the spotted fever group of rickettsiae.3,4 The presence of R. peacockii in ticks is correlated with reduced prevalence of R. rickettsii.2,3 R. peacockii is closely related to R. rickettsii, and deletion or mutation of genes, possibly resulting in loss of virulence in R. peacockii, have been identified.3 R. peacockii is not known to be a pathogen of humans or other animals.3,4 Rickettsial serology 10 weeks after the tick bite showed detectable antibodies (titre, 1/256), predominantly to the spotted fever group of Rickettsia, compatible with exposure to R. peacockii identified in the tick. Unfortunately, definitive seroconversion or a rising antibody titre was not able to be demonstrated as no earlier sera were available for parallel testing. Pre-existing antibodies from a distant rickettsial exposure from his tick-prone lifestyle (history of extensive international camping trips) cannot be excluded. The patient remains asymptomatic 9 months later and is still an avid traveller. Tick-borne rickettsial infections in Australia include Queensland tick typhus (Rickettsia australis), Flinders Island spotted fever and Australian spotted fever (Rickettsia honei) and Q fever transmitted by ticks including Ixodes spp., Amblyomma triguttatum and Bothriocroton hydrosauri.1 With increasing international travel, recognition of tick-borne rickettsial diseases is becoming more important. Dermacentor ticks have been detected on livestock exported from North America into Europe.5 This case shows the ability of human ectoparasites, and their potentially pathogenic bacteria, to bypass stringent Australian quarantine controls. Further studies of Australian and imported tick-borne infections are required to increase understanding of these emerging infectious diseases. Box – Dermacentor andersoni removed from the patient’s umbilicus

Sadid F Khan · Mythili Tadepalli · John Stenos · Stephen R Graves · Tony M Korman

To the Editor:We congratulate Cheney and colleagues1 for throwing light on the contributions of overweight and obesity on adverse birth outcomes by analysing data from a teaching hospital in central Sydney.1 Around 16% of the women presenting between 2010 and 2014 were overweight, while 7% were obese. Furthermore, despite obesity being an important risk factor for adverse pregnancy outcomes, their study showed a lack of recording of body mass index (BMI) in patients’ records. Adverse pregnancy outcomes are more common among Indigenous Australian women than non-Indigenous women;2 obesity levels are high in pre-conception and in pregnancy, and the subsequent adverse impact on increased metabolic health in offspring is likely contributing to early onset of diabetes and chronic disease in Indigenous Australians. Hence, we want to extend the debate to report on what is happening in primary health care (PHC) settings for Indigenous women. We have analysed continuous quality improvement data from audits of adherence to evidence-based guidelines for maternal care in 65 Indigenous PHC centres (1091 patient records) across Australia during 2012–2014.3 The majority of women at most PHC centres had the first trimester weight recorded (mean, 90%; range, 60–100%), but there was wide variation in recording of BMI (mean, ∼ 60%; range, 0–100%) (Box). This indicates that most barriers to BMI recording are more to do with clinicians’ understanding of the value of and ability to calculate BMI than around women’s willingness to be weighed. For women with an abnormal BMI (mean, ∼ 30%; range, 0–100%), there was wide variation in documented BMI management plans (mean, ∼ 40%; range, 0–100%). Dealing with these generally low levels of recording and wide variation in recording between PHC centres is a vital early step in limiting the contribution of obesity to adverse pregnancy outcomes and improving long term health outcomes for the mother and baby. Women attending PHCs that had participated in continuous quality improvement activities were more likely to receive recommended pregnancy care related to screening and brief interventions for modifiable lifestyle-related risk factors, such as obesity.4,5 These findings support the incorporation of continuous quality improvement activities into the delivery of maternal care. Box – Record of scheduled maternal care services received by Indigenous women at Indigenous primary health care centres, 2012–2014* BMI = body mass index. * More information on how to interpret box plots is available in Gibson-Helm et al,3 page 21.

Jodie Bailie · Jacqueline A Boyle · Ross S Bailie

To the Editor:With the release of the 2017 Australia’s Health Tracker by Socio-economic Status1 comes the latest evidence of an unjust health divide in Australia. With few exceptions, people who are more socially disadvantaged have worse health than their advantaged counterparts — a paradigm known as the social gradient of health.1 Despite increased need, socially disadvantaged individuals are less likely to access health care compared with advantaged people.2 It was refreshing, therefore, to read that the federal Shadow Minister for Health and Medicare Catherine King has placed health inequities front and centre of the Australian Labor Party health strategy, highlighting the importance of equitable prevention and public health and pledging to reduce the inequities in access to health care.3 Only 1.3% of the total Australian health expenditure is dedicated to the prevention of non-communicable diseases, much of which is directed at behavioural factors, including physical inactivity, poor diet and smoking1 — all of which are known to influence non-communicable diseases, such as diabetes, cardiovascular disease and osteoporosis.4 However, focusing primarily on behaviour ignores the wider context of health problems: the social, economic and cultural factors influencing that behaviour.5 The mounting costs of living and insecure employment increase vulnerability to non-communicable diseases. Improving the availability of affordable housing, quality employment and educational opportunities may better serve individuals at the economic margins, thereby addressing the stark inequities observed between advantaged and disadvantaged people. Dealing with these inequities would achieve discrete sectoral and health goals, providing cost-effective intersectoral cobenefits. What then is the role, for example, of the employment or educational sectors in driving the policies that address the socio-economic conditions which ultimately contribute to a healthy Australia? What is role for the health sector? The Department of Health could play a stewardship role, supporting and enabling other government agencies to manage their own sectoral goals while positively influencing health outcomes. Dealing with health inequities requires an intersectoral approach, a long term view, courageous leadership at the highest executive level, and support by an empowered public sector based on principles of fairness. Australia must act on social determinants of health and address this major problem for the health of our society.

Sharon L Brennan-Olsen · Sharon Friel

To the Editor:We read with interest the recent position statement on the management of non-neurogenic overactive bladder (OAB) published in the Journal.1 This considered statement reflects recent guidelines from national and international urological societies, including the American Urological Association, the European Association of Urology, the International Continence Society, the Urological Society of Australia and New Zealand and the UroGynaecological Society of Australasia. In recognition of the often repeated phrase in urological research “the bladder is an unreliable witness”, there has been an increasing emphasis on addressing factors beyond the bladder when managing patients with OAB and lower urinary tract symptoms (LUTS).2 The authors of the MJA statement deal with this issue more than most by encouraging their readers to be mindful of known OAB risk factors and comorbid conditions. However, one condition in particular, obstructive sleep apnea (OSA), continues to be overlooked in the diagnosis and management of OAB and LUTS. Our group has previously shown a substantial burden of undiagnosed OSA in a large sample of community-based men aged 40 years and over, with 53% of men without prior diagnosis shown to have some degree of OSA, of which 14% and 12% were found to have moderate and severe OSA, respectively.3 We also demonstrated an independent association between nocturia — the most frequent and bothersome of the OAB symptoms2 — and the presence of moderate to severe OSA, with such men showing a 52% adjusted risk increase for the presence of nocturia.3 Recent systematic reviews of continuous positive airway pressure in the treatment of nocturia indicated a mean standardised difference of −2.28 episodes per night (95% confidence interval, −2.42 to −2.15) between treatment groups in patients with nocturia.4 This difference is comparable to the efficacy observed with frontline treatments for this condition. Nocturia and other OAB symptoms are also known to be sentinel markers of cardiovascular disease,5 providing further motivation for dealing with known risk factors, such as OSA, in the management of OAB. We believe that it would be helpful to include recommendations to assess for the presence of OSA in patients presenting with nocturia, either alone or in combination with other OAB-related symptoms.

Sean A Martin · Nicholas R Brook · Gary A Wittert

In reply

Eric Chung

To the Editor: I write to comment on the Perspective by Wolfenden and colleagues1 on e-cigarette use by young people. In the article, the authors make regulatory recommendations for e-cigarettes based on the assertion that vaping causes young people to take up smoking — the so-called gateway theory. However, while many studies have found that adolescents who try vaping are more likely to try smoking, there is no evidence of cause and effect. The longitudinal studies described only show an association and are unable to demonstrate causality.2 An alternative explanation is common liability; that is, that young people who are more attracted to experimentation are more likely to try both products. In fact, international reviews suggest that vaping is diverting adolescents away from smoking and reducing smoking rates.3 As vaping rates have been increasing in young people, smoking rates are rapidly declining — a finding inconsistent with a gateway effect. Real world studies show that most vaping by young people is experimental and short lived and only a minority use nicotine. Regular vaping among teens is almost exclusively confined to those who already smoke. A recent analysis of five national surveys in the United Kingdom of 60 000 adolescents found that regular vaping by never-smoking 11–16-year-olds was 0.1–0.5%.4 Perhaps most important of all, Wolfenden and colleagues1 dismiss the substantial public health benefit that could result from vaping. Population studies have indicated that millions of smokers have quit using e-cigarettes and modelling studies have found a substantial net public health benefit, even using pessimistic scenarios.5 Banning wider access to e-cigarettes on the basis of an unproven risk to adolescents would prevent access to a life-saving quitting aid for millions of smokers. A better solution is to employ strategies to minimise youth access and make vaping available for adult smokers who are otherwise unable to quit smoking with conventional therapies.

Colin P Mendelsohn

To the Editor: Wolfenden and colleagues1 fail to justify the ban on the sale of nicotine-containing e-cigarettes in Australia and do not address the supporting arguments for allowing smokers to access these products.2,3 The concern that e-cigarettes may recruit young smokers justifies a regulatory response but not a ban on sales of nicotine-containing e-cigarettes to adults,3 who are thereby prevented from using these products. The authors cite a meta-analysis of studies that showed an association between the past 30-day use of e-cigarettes and combustible cigarettes;4 however, they ignore the weaknesses in the studies. First, most studies defined an adolescent e-cigarette user or cigarette smoker as one who had used either product in the past 30 days. These studies do not show that adolescents who used e-cigarettes were more likely to become daily smokers than their peers who did not use them. Second, the studies had substantial subject attrition (more than 30% in four out of nine studies) and they often poorly controlled for a shared liability to use drugs. The association was weakest in the largest and best controlled study.5 In addition, Wolfenden and colleagues make no mention of the steep decline in youth smoking rates that has occurred in both the United Kingdom and the United Sates while vaping has increased among young people.5,6 Nor do they mention that the UK’s smoking prevalence is now the same as Australia’s, in the absence of plain packaging or high tobacco tax.6 The authors’ claim that a relaxation of the ban on e-cigarette sales in Australia “has the potential for unintended harm” ignores the direct harms caused by the current ban; for example, the unregulated use of e-cigarettes, a black market, and the criminalisation of e-cigarettes users.6 Finally, even if the evidence for a gateway effect were accepted, it would, at most, justify a tighter regulation of e-cigarettes; it does not justify a prohibition on adult e-cigarette use. If it did, we would be morally obliged to prohibit the sale of cigarettes to smokers because on their argument this would be the most effective way of preventing adolescents from smoking cigarettes.

Wayne D Hall

In reply

Sze Lin Yoong · Emily Stockings · Luke Wolfenden

To the Editor:There are some populations for whom unequal health cannot be overcome and a preoccupation with preserving health detrimentally affects their quality of life.1 Imagine you live in residential aged care, your health is irreversibly poor and you have days left to live. You want to go for a walk in the sunshine, but are prevented from doing so in case you fall and harm yourself. Your health cannot improve; yet, you are denied the opportunity to enjoy life. Quality of life is influenced by health, as well as social support, autonomy and spiritual fulfilment.2 Society encourages younger people to take risks to fulfil their potential and enjoy life; however, we prevent older people from doing the same. By denying this population the “dignity of risk”,3 we are denying them an opportunity to optimise their quality of life. This optimisation requires recognising that focusing on inequality in health will not always overcome an unequal quality of life, especially for vulnerable older people. Most residents of aged care facilities have multiple comorbidities contributing to poorer health, but this need not equate to a suboptimal quality of life. The goal should be to enable older people to enjoy their life by supporting them to fulfil their potential. People with a similar stage of dementia who live at home report a higher quality of life than those in residential aged care.4 Living at home provides greater independence, illustrating the importance of considering a multifaceted quality of life. If professionals focus only on improving health and not all factors contributing to quality of life, then inadvertently, neither health nor quality of life will improve. Enabling individuals to make choices, even those with a risk of harm, is a pathway to improving quality of life for everyone, including people with irreversible poor health.

Alice L Holmes · Marta H Woolford · Joseph E Ibrahim

To the Editor:A 49-year-old man, with a history of occasional red eye self-treated with over-the-counter naphazoline eye drops, accidentally self-administered a drop of “fibreglass resin catalyst”, containing concentrated (approximately 33%) methyl ethyl ketone peroxide (MEKP), to his left eye. The MEKP bottle was purchased at a local hardware store. The patient had previously placed the MEKP bottle on his bedside table and mistook it for his regular eye drops the following morning. Upon instillation, he felt immediate pain and irrigated his eye with water at home before presenting to the emergency department (ED). After appropriate ED irrigation and treatment, he was managed in the hospital’s ophthalmology department. After 13 days, his cornea had healed and his vision had returned to normal. However, there was residual evidence of limbal ischaemia — a poor prognostic sign in ocular chemical injuries.1,2 Ocular chemical injuries are a major source of preventable morbidity and blindness. Strongly oxidative compounds, such as MEKP, especially at high concentrations, cause rapid oxidative damage to corneal epithelium leading to cell death.3 MEKP in particular has been associated with long term symptoms and ocular surface dysfunction in patients who delay primary irrigation.4 The injury reported herein was particularly deleterious as it involved the direct administration of MEKP to the eye, because the patient mistook it for his regular eye drops (Box). This confusion was due to the similar packaging of MEKP to many common eye drops used in ophthalmology. Similar injuries have been reported previously in the literature4 in patients who mistook MEKP for their regular eye drops. Often, patients’ underlying vision is poor, making it difficult for them to differentiate the products without careful inspection.4,5 Packaging MEKP in this medically familiar format poses a significant danger to the general public. We urge manufacturers to reconsider the presentation of this product by obvious colouring of the entire bottle (eg, bright orange with a black lid) so it is impossible to mistake MEKP for regular eye drops. Such a change could significantly reduce the number of preventable vision-threatening injuries sustained in this manner. Box – Package similarity between naphazoline eye drops (left) and methyl ethyl ketone peroxide (right) bottles

Richard T Parker · Dominic P McCall · Chameen Samarawickrama

To the Editor: Carey and colleagues are to be applauded for drawing attention to the potential impacts of extreme hot weather on the health of vulnerable groups in the community, particularly older people.1 However, in addition to the public health effects they covered, it is important to highlight the health impacts of high ambient temperature on Australian workers. The working population, ranging widely in age, is subject to many of the factors described in the article that influence vulnerability to heat illness, including chronic diseases and medications used to manage these. In addition, workers are often exposed to other heat stress factors, such as requirements to wear restrictive, impermeable personal protective clothing or to operate heavy machinery. Occupational circumstances may limit opportunity for acclimatisation, rest breaks, maintenance of adequate hydration or access to assistance if heat illness arises. A growing body of evidence shows associations between high ambient temperature and the risk of work-related injury.2,3 Importantly, this increased risk is not just observed among outdoor workers. Studies set in Adelaide2 and Melbourne3 have identified young workers (< 25 years of age) and male workers as being at increased risk of injury in association with increasing ambient daytime temperatures. Associations have also been observed between daily maximum temperature and risk of injury for workers engaged in occupations with the heaviest physical strength requirements (such as plumbers, paramedics and bricklayers).3 Moreover, a relationship has been observed between overnight temperatures and risk of injury for younger workers and female workers.3 In spite of climate projections for more frequent hot days and more severe heatwaves, there is currently a lack of specific and consistent policy and guidance targeting the work-related health impacts of hot weather.4 It is important that the health of workers and factors that influence their vulnerability to injury and illness are included in heat health discussions and are considered in the development of adaptive strategies.

Judith A McInnes · Peter Smith · Malcolm R Sim

In reply

Marion G Carey · Mark P Monaghan

To the Editor:Ahern and colleagues1 explore the potential benefits and pitfalls of benchmarked reporting in the Australian context. As a binational registry of patients on renal replacement therapy in Australia and New Zealand, the Australia and New Zealand Dialysis and Transplant Registry has been producing and distributing centre-specific performance reports to renal units for over 20 years; these share many of the challenges faced by clinician-level reporting. In the past few years, this has extended to provision of an abridged version of the report on our website, containing unit-specific risk-adjusted outcome data for each dialysis and transplant unit (http://www.anzdata.org.au/v1/hospitalreport.html). The authors highlight the challenges of low case numbers resulting in statistical models that are underpowered to detect poor performance and require long observation periods that will limit timely detection of outliers and opportunities for remedial action. Co-opting statistical techniques used for quality control in other industries may present an opportunity to address these issues in the health care sector. Cumulative sum control charts2 provide a method for sequentially monitoring cumulative performance over time, which may permit early detection of poor performance and account for varying activity levels by including the number of procedures performed, rather than just a fixed time frame. Similarly, Bayesian approaches that involve updating prior probability distributions within a dynamic model may address these concerns3 and offer the conceptual advantage of explicitly testing not just the statistical difference from average but the likelihood of performance falling into a defined poor performance category. Finally, there are systems that use differing criteria for smaller and larger units.4 Ahern and colleagues discuss the potential consequences of poor performance, but omit any reference to exactly who should oversee this process. We assert that the relevant specialty or subspecialty body has a crucial role in overseeing the interpretation of reports. The detection of an outlier is dependent on the nature of the boundaries set for acceptable performance, and the vulnerability of the statistical adjustment model to bias and unmeasured confounders. Such interpretation requires detailed knowledge of the relevant field, an appreciation of the variation between centres, and substantial epidemiological knowledge.

Matthew P Sypek · Matthew D Jose · Stephen P McDonald

In reply

Susannah Ahern · Sue M Evans · Ingrid Hopper · Arul Earnest