Letters

To the Editor: Until recently, detection of locally acquired tox gene carrying diphtheria in Australia was rare. Toxigenic diphtheria had almost disappeared from the Australian landscape, with the widespread uptake of the diphtheria toxoid vaccine. Diphtheria is predominantly caused by toxigenic Corynebacterium diphtheriae and can present as both respiratory and cutaneous diphtheria disease. There have been increasing reports internationally of diphtheria outbreaks primarily in vulnerable migrant populations.1 In Australia, there were 46 diphtheria cases between 1999 and 2019 (eight respiratory diphtheria and 38 cutaneous), with C. diphtheriae accounting for 87% of these cases.2 Since 2020, a genomically linked clone of tox gene carrying diphtheria bacteria has spread across North Queensland. Cases described here are from the Queensland Health's Notifiable Conditions Register. The Townsville Hospital and Health Service Human Research and Ethics Committee provided an ethics exemption (EX/2022/HREC/88895) for this study. Of the 29 linked cases identified between 2020 and 2022, all have had epidemiological links to North Queensland and 23 were notified in 2022. Three of these cases were detected interstate. Clinically, three cases presented with classic diphtheria, four with mild respiratory diphtheria, and 22 with cutaneous diphtheria. The median age was 21 years (range, 2–59 years). Of the 29 cases, 34% (10/29) were aged 11–20 years and 45% (13/29) were older than 20 years. Further, 38% of cases (11/29) were fully vaccinated, 48% (14/29) were partially vaccinated and 10% (3/29), including two of the classic diphtheria cases, were unvaccinated for their age according to the National Immunisation Program Schedule. All were acquired in Australia, and 86% of patients (25/29) identified as Aboriginal and/or Torres Strait Islander. Isolates in the cluster are closely genomically linked, with between zero and 20 single nucleotide polymorphism differences found on whole genome sequencing. All cases were managed with penicillin or azithromycin, vaccination, and diphtheria antitoxin administered when deemed appropriate by infectious disease physician assessment. Azithromycin was given in most cases, as higher minimum inhibitory concentrations to penicillin have been observed with this clone. Household contacts were screened for symptoms, administered chemoprophylaxis with penicillin or azithromycin, and offered a diphtheria‐containing vaccine if due. A North Queensland public health working group has recently been established to develop a consensus on further public health management. This diphtheria outbreak, almost exclusively in Aboriginal and Torres Strait Islander communities, highlights the continuing impact of social determinants on disease in vulnerable populations. We recommend clinicians be aware of these cases and consider diphtheria among their differentials in patients from, or who recently travelled to, North Queensland. Classic diphtheria is characterised by a sore throat, fever, and membrane on the back of the throat, which may cause difficulty in breathing and swallowing. Cutaneous diphtheria usually presents as a non‐healing ulcerative lesion.2 The toxoid vaccine protects against the toxin effects rather than the infection itself. Unvaccinated individuals are therefore at highest risk of severe disease, including classic diphtheria, myocarditis, and neuropathies. Vaccination remains imperative and timely vaccinations are essential. The National Immunisation Program Schedule currently recommends a diphtheria‐containing vaccine for children at two, four, six and 18 months, and four years, and adolescents at 11–13 years. A diphtheria‐containing vaccine booster is recommended for adults at 50 years.3 Clinicians across the state and nation should be vigilant for future cases.

Allison Hempenstall · Jay Short · Tonia Marquardt · Valmay Fisher · Janice Johnson

To the Editor: A recent survey found that 65% of Australians choose cremation over burial, and that, in the context of the coronavirus disease 2019 (COVID‐19) pandemic, cremations could be increased fourfold.1 It seems appropriate to look at the various roles of the medical practitioner in approving cremations. The first crematorium was built in England in 1878.2 Medical practitioners were at the forefront of the movement in Australia, and argued that cemeteries were already overcrowded and in disrepair, that a fear of being buried alive was common, and that cremation was a hygienic way to dispose of the dead when it was still believed that cemeteries were a source of disease in the living.2 The first Australian crematorium was established in Adelaide in 1903, after which progress halted due to ongoing debate;3 however, objections were overcome and many were constructed in the interwar years (Box 1). The most cogent reason for rejecting cremation was that it could be employed to hide homicides. The contrary argument was that having a medical practitioner verify the cause of death would both prevent undetected crime and advance medical knowledge.4 To provide for this verification step, each Australian state and territory introduced legislation setting out a process for the approval of cremations. Where the cause of death is suspicious or unclear, permission is withheld and referral made to a coroner. In Queensland, Tasmania and Victoria, any medical practitioner who did not complete the death certificate may issue a cremation permit (Box 2). In the Australian Capital Territory, New South Wales and Western Australia, only government‐appointed medical referees may issue a permit. Some states and territories also require assessment of cremation safety, verifying that there is no implanted battery‐operated device, which may explode and damage the cremator, and that the deceased has not received radioisotopes, which may create a public health risk during cremation. At the Public Health Unit, we manage the appointment of medical referees for the NSW Ministry of Health. As part of the review of the effectiveness of our procedures, in 2020 we developed a short questionnaire to identify reasons why medical referees may refuse permission to cremate and to seek suggestions for improvement to this administrative process. We emailed it to a 10% random sample of NSW medical referees (n = 77) and received 46 responses (60%). Of these, 29 respondents (63%) had served in the role for more than 15 years, nine (20%) for five to 15 years, and five (11%) for less than five years. The number of permits issued in the previous 12 months ranged widely: 25 had signed zero to ten, and at the other extreme, two provided estimates of 780 and 1200 permits. This large variation is most likely related to varying demands on medical referees by local funeral directors. The reasons for rejecting applications included inconsistencies in the forms (ten), incorrect or unclear cause of death (seven), a reportable death (four), uncertain identity of the deceased (one), and a retained battery‐powered device (one). No suggestions for improvement were made, but three respondents suggested we provide some medical referee training — we have undertaken to develop an online education module. Although roles differ between Australian jurisdictions, medical practitioners have an important responsibility for verifying that a body is suitable for cremation, a decision that they must make independently of the practitioner who certifies the cause of death. Box 1 – Woronora Crematorium, Sydney, constructed in 1934* * Courtesy of Woronora Memorial Park. Box 2 – Roles of medical practitioners in approval of cremation under current legislation for each Australian state and territory State or territory Legislation Final approver of cremation Name of permit Comments Who assesses cremation risk Australian Capital Territory Cemeteries and Crematoria Regulation 2003 Medical referee, a medical practitioner appointed by the Director‐General Certificate of medical referee The medical referee must view and be satisfied with the “Certificate of medical attendant” The “Certificate of medical referee” must state that there is no medical reason why the remains should not be cremated New South Wales Public Health Regulation 2022 Medical referee, a medical practitioner appointed by the Secretary Medical referee's cremation permit The medical referee must be independent of the applicant and of the medical practitioner who attended the deceased A medical practitioner must provide cremation risk advice concerning battery‐operated implanted devices and radioactive treatment Northern Territory Cemeteries Act 1952 Crematorium manager Cremation permit A permit can only be issued by the cremation manager if the certificate provided is signed by two medical practitioners stating that death was due to natural causes No reference to assessment of cremation risk Queensland Crematorium Act 2003 Independent doctor Permission to cremate (independent doctor) The independent doctor is a doctor who has not signed the cause of death certificate or the “Cremation risk certificate” Any other medical practitioner may complete the “Cremation risk certificate” South Australia Burial and Cremation Act 2013; Burial and Cremation Regulations 2014 Registrar of Births, Deaths and Marriages Cremation permit Registrar of Births, Deaths and Marriages must consider the “Death from natural causes certificate of second doctor” Second doctor certifies that there is no reason why the body of the deceased should not be cremated Tasmania Burial and Cremation Regulations 2015; Burial and Cremation Act 2019 Medical practitioner Cremation permit The medical practitioner who issues the permit must not be a partner, employee or relative of the medical practitioner who issued the medical certificate in respect of the deceased person The medical practitioner who signs the permit must be satisfied that any implanted medical device has been removed Victoria Cemeteries and Crematoria Act 2003; Cemeteries and Crematoria Regulation 2015 Registered medical practitioner Certificate of registered medical practitioner authorising cremation The registered medical practitioner is someone who is not the medical practitioner who completed the notice in respect to the death of the deceased The funeral director must take note of any battery‐operated implanted device mentioned on the “Medical certificate of cause of death” Western Australia Cremation Act 1929; Cremation Regulations 1954 Medical referee, appointed by the Governor Permit to cremate The medical referee must not issue the certificate of cause of death or be in partnership with the medical practitioner who does so The “Certificate of medical practitioner” covers battery‐operated implanted devices and radioactive treatments

Mark J Ferson · Reannon Johnson · Toni Cains

To the Editor: The coronavirus disease 2019 (COVID‐19) pandemic has changed the doctor–patient relationship. The technological revolution that is telehealth has brought undeniable benefits to patients, including greater access and convenience, and more control and empowerment through self‐care.1 There are high satisfaction ratings for patients utilising digital health care technologies.2 Unsurprisingly, there has been an increased uptake in the use of online prescription, referral and medical certificate services where a pre‐existing doctor–patient relationship is not always present.3 In our experience, there is an increasing digital incursion on the traditional general practice consultation. General practitioners are being pressured by their patients to issue prescriptions4 and referrals outside of scheduled consultations. Presumably to preserve the doctor–patient relationship and potentially business interests, we have seen GPs acquiesce to these patient requests, often with little to no consultation. This is occurring despite previous research showing that consumer demand is a driver of unnecessary tests and treatments.5 This non‐contemporaneous doctor–patient interaction exacerbates the demand for fast, convenient health care delivered how and when patients dictate. In this instance, ensuring consumer satisfaction is potentially at odds with ensuring good and safe patient outcomes. There are risks of engaging with this type of instant medicine. Increased accessibility to, and demands made of, GPs increases their workload burden2 and precipitates cognitive overload. This emerging expectation of doctors to engage in asynchronous interactions with patients, outside of consultations, creates what we term a “digital fourth wall”. Patients can access their GP in an inherently one‐sided encounter, with doctors given little time and opportunity to reciprocate. Our concern is that the impersonal nature of this doctor–patient interaction may beget a diminution of professional responsibility towards the patient and suboptimal treatment and diagnosis. Striving for a mutually acceptable outcome through shared decision‐making processes will ensure that duty of care to the patient, based on GPs’ best clinical judgment, is not compromised. This may mean saying “no” to a patient, despite what we observe to be the natural inclination of many GP colleagues to acquiesce to increasing digital demands. The medico‐legal risk, and responsibility for patient care, ultimately rests with the GP. Despite appearances, novel innovations in technology do not always lead to health improvements.

Michael Tran · Katrina Anderson

To the Editor: In a recent Editorial,1 Grimwood and Chang cited a review of long COVID in children and adolescents,2 and wrote that symptoms are similar for those with and without evidence of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection. But this is an inaccurate description of the review's findings. Cases were more likely to experience persistent symptoms than controls in the majority of studies reviewed.2 The difference in prevalence might be even greater than reported owing to the well documented underdetection of coronavirus disease 2019 (COVID‐19) in children,3 resulting in misclassification of cases as controls. A growing body of evidence indicates children are more affected by COVID‐19 than initially thought. A recent US Centers for Disease Control and Prevention (CDC) analysis of 1.4 million children aged under 12 years and 1.7 million adolescents aged 12–17 years found increased rates of asthma, myocarditis and cardiomyopathy, cardiac dysrhythmias, diabetes, renal failure, venous thromboembolism, and coagulation disorders in children with laboratory‐confirmed COVID‐19 compared with children without COVID‐19. These increased risks (excluding asthma) were also experienced by adolescents with COVID‐19, who were additionally at increased risk of pulmonary embolism.4 Although uncommon or rare, such outcomes suggest children are not spared the cardiovascular and metabolic sequelae of COVID‐19. Recent research using low field magnetic resonance imaging (MRI) revealed persistent pulmonary dysfunction in non‐hospitalised children and adolescents (mean age, 11±3 years) who had recovered from COVID‐19 (n = 29) or had long COVID (n = 25). Despite all children having morphologically normal lungs (except for one recovered child), ventilation and perfusion (V/Q) matching was markedly lower in the recovered group (62±19%) and the long COVID group (60±20%) compared with nine healthy controls (81±6%; mean age, 10±3 years).5 Although the MRI study may be limited by selection bias (ie, children with greater symptomatology being more likely to participate), this and similar research indicate the health impact of paediatric COVID‐19 is greater than generally acknowledged. We do not know what the long term impact of SARS‐CoV‐2 infection might be, but the accumulating data are not encouraging. Reinfection is common and SARS‐CoV‐2 spreads readily in schools in the absence of mitigation measures, such as the use of masks, portable HEPA air cleaners, and improved ventilation. Notably, better ventilation has wider benefits, including improved academic performance. A poorly ventilated classroom can be equivalent to a student skipping breakfast.6 The COVID‐19 pandemic is not over. Ongoing commitment to a public health strategy informed by the precautionary principle is required. This will deliver wide‐ranging social, economic and health benefits.

Zoë Hyde

To the Editor: Lord and colleagues’ article1 provides a much‐needed snapshot of breast cancer distant recurrence and metastatic survival. As one of Australia's leading breast cancer advocacy organisations, Breast Cancer Network Australia (BCNA) has long been calling for reporting of recurrence data. Although we can justifiably celebrate a 5‐year survival rate of 92%,2 the long term data on distant recurrence reported by Lord and colleagues demonstrate the importance of looking beyond 5 years to understand the full burden of disease. In addition, Lord and colleagues’ survival data after distant recurrence is a critical first step in understanding the survival experience of this important but neglected group. In the associated Editorial, Redfern and Martin3 highlight that Australia's cancer registries do not systematically collect or report recurrence. The same applies for stage at diagnosis. Consequently, the number of Australians living with metastatic breast cancer is unknown. This is a fundamental problem. Without information to quantify this group of people we cannot adequately plan or deliver services. The problem extends beyond breast cancer to cancers such as prostate, melanoma, colorectal and lung, where targeted therapies, immunotherapies, and antibody drug conjugates are driving improvements in survival for patients with metastatic disease.4 These patients often have long term, complex supportive care needs yet have little visibility in our health care system. BCNA's 2017 national survey of 10318 people with breast cancer identified higher information and support needs among patients with metastatic compared with non‐metastatic breast cancer.5 These information and support needs were also less likely to be met by services for people with metastatic compared with non‐metastatic breast cancer.5 On 13 October 2022, which was Metastatic Breast Cancer Awareness Day, BCNA launched its Making metastatic breast cancer count Issues Paper to draw much‐needed attention to these issues.6 In the absence of cancer registry data, we applied Australian modelling from 2008 to current breast cancer mortality data.2,7 We estimate that in 2020 there were at least 10553 Australians living with metastatic breast cancer. Data from the United States suggest this number will continue to grow.8 Collection of recurrence and stage at diagnosis data will require national leadership and accountability, including continued investment in Cancer Australia's Stage, Treatment and Recurrence Project. Critically, it will also require input from key stakeholders including data‐management experts, cancer epidemiologists, cancer registries, clinicians and consumers. The time for action is now.

Andrea Smith · Vicki Durston · Sam Mills

To the Editor: We applaud the authors for comparing ambulance offload times with mortality and re‐presentation rates for patients presenting with chest pain.1 However, the model employed by the authors fails to account for the independent impacts of access block and emergency department (ED) overcrowding on poor outcomes and thus risks overemphasising the influence of ramping. There is clear evidence that ED overcrowding and access block are associated with worse patient outcomes, including increased mortality, re‐presentation rates and ambulance offload times.2,3,4,5 This could explain many of the study's findings: ramping is the symptom, ED overcrowding and access block are the disease. Secondly, there are fundamental differences between tertiles 1 and 3 that have not been addressed. The patients in tertile 1 are more likely to have been offloaded straight into the waiting room, a common procedure in most EDs, whereas patients in tertile 3 would not, potentially due to poor mobility, dementia, or being assessed as requiring significant cardiorespiratory monitoring. For this reason, patients in tertile 1 would be expected to have better outcomes than those in tertile 3. The article adjusts for the presence of eight comorbid conditions but not their severity. Furthermore, the incidence of each individual comorbid condition was marginally higher in tertile 3, and although individually not statistically significant, the cumulative impact of multiple comorbid conditions would be higher in tertile 3, which could confound the results. The study's use of the Charlson index is limited, as it only uses the identified eight comorbid conditions. The Charlson index also includes dementia, hemiplegia, heart failure, liver disease, and cancer; these are not measured in the study and would all contribute to offload delays and worsened outcomes. Lastly, without including patients presenting via private transport with chest pain, and in the absence of any measures of ED overcrowding and access block, the analysis of patients presenting with chest pain remains incomplete. That ED overcrowding and access block, evidenced by ambulance offload delays, is associated with worsened outcomes is well known. Unfortunately, focusing on the symptom of ambulance ramping, rather than the disease of ED overcrowding and access block, risks leading to ill‐informed policy decisions and ineffective solutions.

James L Mallows · Bridget Honan · Sierra Beck

To the Editor: We read the article by Wang and colleagues1 with great interest. In their study, they used linked administrative datasets from 2005 to 2018 to investigate sex differences in pre‐hospital management of patients with stroke from a single jurisdiction. Among their findings, they observed that a greater proportion of women (52.4%) than men (47.9%) arrived at hospital by ambulance. We seek to draw attention to a similar, nationally representative study on the factors associated with the use of ambulances and access to evidence‐based care among patients with stroke.2 Our study included patients with first‐ever strokes from the Australian Stroke Clinical Registry (2010–2013) linked with administrative data (emergency, hospital admissions),2 as part of the Stroke123 study3 (including 39 hospitals from New South Wales, Queensland, Victoria and Western Australia). We found that among the 6262 patients with first‐ever stroke, 4737 (76%) arrived by ambulance.2 Interestingly, we also found sex differences in arrival by ambulance before adjustment for other covariates (women, 78.8% v men, 72.9%).2 In the Stroke123 study, we adjusted for factors associated with ambulance arrival.2,3 Some of the factors most strongly associated with ambulance arrival were age, frailty and markers of stroke severity. We found that patients who were older, frailer and had more severe strokes were more likely to arrive by ambulance. Following adjustments for age and stroke severity, the sex differences were no longer statistically significant.2 In other research undertaken by our group using linked ambulance, hospital and Australian Stroke Clinical Registry data, other important factors associated with arrival by ambulance included call‐taker or paramedic identification of stroke, which was less often identified in women.4 In the investigations on sex differences by Wang and colleagues, there was no adjustment for markers of stroke severity. We are curious as to why no adjustment for Glasgow Coma Scale was made, particularly when looking at assessment for stroke by paramedics. It is possible that the sex differences observed would not be as marked after adjusting for stroke severity.

Monique Kilkenny · Amminadab L Eliakundu · Joosup Kim

To the Editor: The metaverse is a virtual environment merging physical and digital realities.1 Once the thing of movies, metaverses are tipped to be worth $800 billion by 2024.2 They have potential to revolutionise digital health care delivery, access, education, and patient outcomes.3 Artificial intelligence (AI)‐enabled health care, with metaverse enhancement, could create virtual hospitals,4 transform clinical workflows and accelerate the diagnosis and treatment of conditions relating to mental health,5 cardiology,6,7 ophthalmology,8 and oral health,9 for example. AI‐enabled technology demonstrated potential during the pandemic by predicting the incidence of coronavirus disease 2019 (COVID‐19)10 and helping to identify sites for vaccine trials.11 Yet in the absence of strong national and global health governance and accountability mechanisms, digital health ecosystems create risk for medical confidentiality and privacy breaches12 resulting in data sharing13 and use/reuse by corporations or governments outside its intended purpose and the bounds of patient consent.14 For example, a National Health Service (NHS) Foundation Trust in the United Kingdom established patient data sharing with the international technology company DeepMind to develop machine learning‐based management tools,15 but British patient data moved to the United States when Google acquired DeepMind.14 Repurposing health‐related biometric and genomic data that cannot be altered is similarly concerning, and can result in racial profiling16 and privacy violations in the absence of appropriate legislation. Metaverses also have potential to accelerate discriminatory practices. Bias may be embedded in data used to train AI models through lack of community representation or participation.17 Structural inequalities may be reflected in health care datasets, creating risk for discriminatory outcomes.18 For instance, racial bias in a health care therapy algorithm resulted in the discriminatory underestimation of health risk for millions of American citizens, precluding access to vital treatments.19 Prioritising protections for underserved populations and their right to access safe virtual health care is crucial,20 especially for Australians at the intersection of inequities driven by poverty, disability, gender, homelessness, Indigeneity, and for the 11% highly excluded from accessing digital technologies.21 We recommend that the Department of Health and the Australian Digital Health Agency partner with the Australian eSafety Commissioner's work examining metaverse deployment,2 to ensure the highest ethical standards are upheld as planning moves forward and regulatory frameworks are developed.

Caitlin Curtis · Claire E Brolan

To the Editor: With the increasing interoperability of electronic medical records across health services, smoking and e‐cigarette use need to be systematically collected on hospital admission, and advice to quit smoking should be automatically included on hospital discharge summaries. Including information on smoking status in the discharge summary, and ultimately on My Health Record, presents an opportunity to address the use of tobacco and e‐cigarette products — the first being Australia's leading cause of preventable death and disease and the second an emerging exposure of increasing concern.1 Evidence from the United States Surgeon General reports that smoking cessation after cancer diagnosis lowers the risk of dying by 30–40%.2 For some patients with cancer, cessation benefits are equal to or exceed the value of state‐of‐the‐art cancer therapies. In addition, the Surgeon General report shows most patients admitted to hospital wish to quit smoking,2 and there are proven, workable but underused interventions to cease smoking. Peak medical bodies such as the Australian National Health and Medical Research Council and the Australian Commission on Safety and Quality in Health Care3 advise that adherence to post‐hospital referral practice guidelines leads to better outcomes, fewer readmissions, and improved patient survival. Australia's National Preventive Health Strategy has a goal of reducing the adult smoking prevalence from 14% to 5% over the next 8years.4 The newly released draft National Tobacco Strategy includes key policy actions to increase the use of cessation services and to support people who use tobacco and e‐cigarettes to quit.5 Around 1.2 million Australian adults are admitted to hospital at least once each year, and it is estimated that about one in five of them would benefit from smoking cessation services.6 Given most hospitalised patients want to quit smoking, providing this recommendation in a discharge summary for appropriate referral to a Quitline (www.quit.org.au) or a general practitioner is a practical solution to provide best practice care to patients, reduce the risk of readmission to hospital, and increase their survival.

Freddy Sitas · Ben Harris‐Roxas · Sarah L White · Fiona A Haigh · Margo L Barr · Mark F Harris

To the Editor: When statutes govern clinical activity, doctors need to know exactly what those legislative provisions mean. Nahm and colleagues1 address this in their article on eligibility for Australia's voluntary assisted dying (VAD) laws.1 However, in our opinion, the authors misinterpreted the relevant provisions, risking reduced access for eligible patients. Generally, a statutory provision should be given its “ordinary and natural meaning”;2 in other words, a plain English interpretation. As the authors note, each of the VAD Acts uses a particular form of words to set eligibility around a terminally ill person's life expectancy. In Victoria, for example, a coordinating medical practitioner must conclude their patient has been “diagnosed with a disease, illness or medical condition that … is expected to cause death within weeks or months, not exceeding 6 months”.3 Nothing in that wording refers to a probabilistic estimation of the percentage chance that the patient will die within 6 months nor any estimation of the best‐case scenario, as Nahm and colleagues suggest. If the Victorian Parliament had wanted this type of estimation, wording reflecting it could have been inserted into the Voluntary Assisted Dying Act 2017 (Vic). Instead, what is needed is that doctors have an expectation, based on the patient's clinical condition, that the illness will result in death within weeks or months, with the proviso that the number of months that the expectation of death will occur within is 6 or fewer. That clinical judgement is the beginning and end of this criterion. Although that judgement might be informed by knowledge about survival times and even by knowledge about doctors’ accuracy judging survival times, doctors need only certify that they expect that the patient's illness will cause death within 6 months. Nahm and colleagues are wrong to conclude that this wording might mean that people eligible for VAD would be “those with an expected survival time of 2 months”. That is not what the legislation says, and it is a mistake to introduce elements that are not there. Such an interpretation could, in practice, convert the 6 months test to 2 months for some patients, with the risk of narrowing access to VAD for patients the Parliament intended to be eligible.

Christopher J Ryan · Ben P White · Cameron L Stewart

To the Editor: Molecular diagnostics with whole genome sequencing (WGS) of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) has been well described as a method to monitor the evolving epidemiology of the coronavirus disease 2019 (COVID‐19) pandemic and coordinate public health responses.1,2 WGS allows for the accurate identification of COVID‐19 variants, which, when combined with accessible diagnostic testing with polymerase chain reaction (PCR) and rapid antigen tests, supports contact tracing and public health action.1,2 Since January 2022, by agreement with the Forensic and Scientific Services reference laboratory, it has been our practice at the Prince Charles Hospital — a tertiary cardiothoracic transplant centre in Brisbane — to sequence all SARS‐CoV‐2 samples of patients requiring hospital admission for COVID‐19. The rapid turnaround times, averaging 36 hours, supported decisions on variant‐specific therapeutics and optimised bed management. The first case of the Omicron BA.2 subvariant detected at our centre was from a sample collected on 4 March 2022 (Box). This was 17days before public announcements of the dominance of the BA.2 strain by the Queensland Chief Health Officer on 21 March 2022.3 With the anticipated increase in local BA.2 cases, we were able to tailor our hospital level response in real time.4 Initially, even though case numbers were low, from an infection prevention and control perspective, we cohorted BA.1 and BA.2 patients separately to mitigate the risk of nosocomial co‐infection. We maintained the workforce and resourcing surge capacity with a second COVID‐19 ward that opened within days of the initial BA.2 case being identified. Furthermore, genome sequencing for COVID‐19 therapeutics resistance mutations led to alterations in clinical management as BA.2 became the dominant local strain, given its lack of susceptibility to sotrovimab. At the hospital and health service level, access to rapid WGS for identification of emerging variants provides the basis for greater infection prevention and control practices, resource and workforce management, and epidemiological monitoring, and carries implications for clinical management and therapeutics. WGS allows for rapid detection and contact tracing of nosocomial outbreaks, supporting identification of chains of transmission within hospitals. This information augments infection prevention and control by identifying gaps in current practice as well as supporting safer health care environmental design that minimises nosocomial transmission. Furthermore, detection of novel variants at the hospital level can help to predict new waves that require allocation of additional staffing and resources. The Australian Government has recognised the evolving role of microbial genomics in public health surveillance but has not yet committed to further development to optimise personalised medicine, hospital care, and therapeutic decision making.5 We strongly advocate for the inclusion of funding for sustainable rapid WGS that provides real‐time results under the Medicare Benefits Schedule. This will expedite Australia's transition to living with COVID‐19 and will allow us to better prepare our health system to manage new and emerging pathogens into the future. Box – Number of patients admitted and sequenced with coronavirus disease 2019 (COVID‐19) at the Prince Charles Hospital from 1 January to 3 April 2022 by genotype* *Bubble size range, 1–9; total number of patients, 256.

Matthew B Eustace · Ambika Sud · Craig Thompson · Sanmarie Schlebusch · Robert L Horvath

To the Editor: As nicotine prescribers, we welcome much needed advice for doctors on prescribing nicotine. However, we disagree with several recommendations and concerns raised in the article by Ween and colleagues.1 First, in our experience, the recommended starting nicotine concentration of 18mg/mL is inappropriate for most new users. The most popular devices for transitioning to vaping (pod vapes) have small batteries and require higher nicotine salt concentrations to effectively relieve cravings and withdrawal symptoms, typically 20–50mg/mL.2 On the other hand, 18mg/mL would be too strong for a smoker with low nicotine dependence using a more powerful vape pen or mod device. The concentration of nicotine required should be personalised for each user based on the level of nicotine dependence, device type and puffing topography.3 Second, the authors’ concerns about the toxicity of nicotine are overstated in our view. Nicotine is a toxic poison in its highly concentrated form, but the low concentrations used for vaping carry minimal risk of serious harm, although the long term impact of inhaled nicotine on lung tissue is not yet known.4 Third, Ween and colleagues raise concerns about the addictiveness of nicotine. However, most smokers who switch to vaping are already nicotine‐dependent. Dependence on vaping is generally less than for smoking5 because, in many cases, peak nicotine levels from vaping are lower and nicotine delivery is slower. In vitro and animal studies suggest other chemicals in smoke may also increase dependence, but human studies are lacking.6,7 Fourth, a blanket “3‐month prescription maximum” and an “agreed abstinence plan” do not recognise the diversity of the needs of smokers. Switching to vaping and then ceasing smoking can take many months or years for some smokers. Many continue to vape long term to avoid relapse to smoking or for perceived benefits. Therefore, a more flexible and personalised approach is needed. Last, Ween and colleagues are correct that unknown harms from flavours may appear over time and these need to be carefully monitored. However, flavours are an integral part of the appeal of vaping. Flavours encourage the uptake of vaping by smokers and are associated with higher quit rates.8,9 A recommendation to avoid flavours risks inadvertently increasing smoking.

Colin P Mendelsohn · Carolyn Beaumont

To the Editor: We read with interest the article by Drewett and colleagues,1 on anaphylaxis presentations to emergency departments in Victoria, and the linked editorial by Lucas and Vale,2 which proposes a comprehensive adverse drug reaction registry to improve patient safety and care. Re‐exposure of a patient to a medicine to which they have had a previously documented allergy or adverse drug reaction (ADR) is a preventable clinical incident. The Australian sentinel events list includes “medication error resulting in serious harm or death” to ensure public accountability and transparency and drive national improvements in patient safety.3 The availability of accurate and up to date allergy/ADR information for prescribers is vital to support the decision on appropriate treatment and to prescribe a medicine. It is also critical that health professionals who administer medicines and pharmacists who dispense medicines have access to accurate allergy/ADR information, as they provide redundancy within the medication management system to prevent unintentional re‐exposure. We recently undertook an audit to assess the accuracy of medication history documentation performed by pharmacists. Our methodology has been published elsewhere.4 As part of this work, we assessed the quality of pharmacist‐documented allergy/ADR information. Allergy/ADR documentation best practice suggests documentation of the medication name, reaction details and reaction date.5 New findings from our audit identified a total of 108 allergies/ADRs across 99 patients. Of the 52 patients with a previous history of allergy/ADR, only 31% (n = 16) had complete and correct allergy/ADR information documented. These results highlight that the current systems do not adequately support clinicians to provide safe and high quality care. Moreover, these systems risk patient harm through inadvertent re‐exposure to a medicine where there is a previously documented allergy/ADR, given that, at the point of prescribing, administration or dispensing, data are not available, disparate, incorrect or incomplete. We believe the national allergy/ADR registry proposed by Lucas and Vale2 should go one step further, and legislation should mandate interoperability between all electronic prescribing, administration and dispensing systems in Australia. A further challenge will be populating accurate historical allergy/ADR information into the registry. It would require high quality documentation to assist informed clinical decision making. An additional challenge is the complex, hybrid paper and electronic prescribing and documentation systems used across health care settings. Even though these may be challenges to overcome, there are potential opportunities through patient engagement, incentive schemes and machine learning. There needs to be a national consumer‐focused education campaign so the public can be active participants in their own care to support optimal medicine choice based on their allergy/ADR history. We believe a national allergy/ADR registry is of key importance to further enhance patient safety and will be a true system change to overcome the barriers we face in day‐to‐day clinical practice.

Martin L Canning · Andrew Munns · Bonnie Tai

To the Editor: Data from the United States, the United Kingdom, and Canada have highlighted rapidly rising medical education and specialty training costs as a key deterrent to diverse and lower socio‐economic medical graduates from pursuing specialty training. Australia is likely to be experiencing similar patterns, but there have been little published data or research in the domain. For example, a study into costs of specialty surgical training in the United Kingdom and Ireland revealed that medical officers can spend up to £71000 on mandatory training costs during their Fellowship,1 on an average junior medical officer salary of £41000 pounds per year.2 This does not include the additional expected costs such as conferences and postgraduate degrees. In the background, mean debt on graduation from medical school has increased by 55%, from £17892 in 2004 to £27655 in 2014.3 A 2015 study predicted that most English medical students will be unable to repay their student loan before reaching the 30‐year write‐off.4 Australia is experiencing similar forces. When university fees were deregulated in 2014, the then Australian Medical Association president Brian Owler5 and the Grattan Institute6 predicted that full‐fee domestic places in Australian medical schools could rise by up to 270% to $70000 per annum. As foretold, a full‐fee domestic student now pays up to $77888 per year tuition, for a total estimated course fee of $335707 (Doctor of Medicine, University of Melbourne, 2022),7 which exceeds the combined HELP (Higher Education Loan Program) loan limit of $156847.8 Unlike overseas, there are no reports on medical specialty training cost trends in Australia. However, archived publicly available data give a general sense that local costs are rising much like those overseas. For example, annual training fee sums range from $483 (Royal Australian College of General Practitioners9) to $5714 (Australasian College of Dermatologists10) between colleges. This fee covers mandatory trainee subscription or membership to the college but does not include all the other mandatory costs such as exam and elevation fees, which vary between training programs. There is undoubtedly considerable difference in fee structures, and such numbers are difficult to compare between colleges. Nevertheless, without transparent data, we are unable to interrogate the concerns that have been identified in overseas studies. The ramifications are potentially vast and varied. New Zealand data show that student loan debt is a common influence on choice of specialty and is a consideration when choosing to stay in the country to practise once trained.11 This may lead to long term consequences for the structure of the medical workforce and appropriate service delivery for patients. As a first step, it is critical that we take action to improve collection and transparency of medical training cost data in Australia.

James Brown

To the Editor: The concepts presented by Coman and colleagues1 parallel international trends within anatomical societies, where attitudes on human remains used for educational purposes are under scrutiny.2 Illuminating issues around legacy collections, and pressing for further clarity, transparency and appropriate cultural and ethical solutions, is important for the proper treatment of these precious resources. It is also necessary given recent events that have raised concerns within the public about how human remains and anatomical collections are treated under the guise of education.3 We strongly support the points made by Coman et al around development of repatriation policies for privately held bones, but we respectfully suggest additional considerations. The suggestion “Medical ethicists assert that in the absence of consent, anatomical specimens … should be destroyed”1 is not congruent with current suggestions and practice.4 Anatomical collections that have no provenance or consent are often used as teaching resources and represent the tangible legacies of histories of this field of science. As such, they serve the additional educational purpose of eliciting discussions on contemporary ethical and professional practice.5 There are also benefits associated with using real bones as opposed to three‐dimensional or plastic copies that do not retain the same weight, detail or nuanced anatomical features. Arguments do exist for disposition of unconsented collections, but these should be considered against the educational value these remains deliver and the social, ethical and cultural concerns around their ongoing use. In proposing consultation across various groups to clarify appropriate practice around “managing the legacy of human bone use in education,“1 we suggest it is necessary to include other specific interest groups to ensure community‐appropriate practices are developed. It is essential that community input is acquired, with indigenous representatives contributing knowledge and perspective. In addition, input from organisations that contribute expertise specific to national and international standards and practice are necessary. This should include local organisations, the Australian Institute of Anatomical Sciences and the Australian and New Zealand Association of Clinical Anatomists, and globally, the International Federation of Associations of Anatomists ethics committee. Such broad input is necessary to ensure guidelines are fit for purpose in a global education community.

Jon Cornwall · Sabine Hildebrandt · Thomas Champney

To the Editor: Zhao and colleagues1 recently published an article highlighting the improved life expectancy for Aboriginal and Torres Strait Islander men in the Northern Territory over the past 20years.1 This is both important and welcomed. It reflects consistent and concerted work of countless individuals and organisations that are contributing to the improved health and wellbeing of Aboriginal and Torres Strait Islander men in the NT, despite limited resources to do so. It makes sense that we are beginning, albeit slowly, to see these inroads. One example of contributing to the positive outcomes for Aboriginal and Torres Strait Islander men’s health in the NT is the evolution of the Darwin Men’s Inter‐Agency Network (DMIAN). DMIAN is a network of men from across government and the non‐government organisation sector to collaboratively advocate for Aboriginal and Torres Strait Islander men in Darwin.2 DMIAN has enabled men’s health researchers to better understand and act on the wants and needs of the Aboriginal and Torres Strait Islander men in the community from the perspective that matters most: their own. As Zhao and colleagues1 point out, there is still a long way to go with improving the life expectancy of Aboriginal and Torres Strait Islander men, which sits 15.4years behind non‐Indigenous men. In addition, as the life expectancy of Aboriginal and Torres Strait Islander men increases, so too does that of non‐Indigenous men.1 So if we are to close the gap, we cannot afford to lose momentum on targeted action, particularly that relating to Aboriginal and Torres Strait Islander male health and wellbeing.3,4 There is still a need for this to be a recognised priority in the NT and nationally, and for primary health care and social services in the NT to be resourced appropriately. In particular, the Aboriginal Community Controlled Health Services and Aboriginal medical services have a key role to play and should be funded to develop, implement and evaluate health and social and emotional wellbeing programs for male clients, as this is severely lacking and is ultimately hampering progress in Aboriginal and Torres Strait Islander male health and wellbeing outcomes.4 While the National Men’s Health Strategy identifies Aboriginal and Torres Strait Islander men as a priority population,5 we also need substantially more investment in research and evaluation to find new innovate solutions.6 We hope the important work being done by individuals in health, justice, education and other social services sectors continues to be enabled to support Aboriginal and Torres Strait Islander men for the benefit of their communities and future generations.

Kootsy Canuto · Karla J Canuto · Jason Bonson · James Smith

To the Editor: We endorse the call of Flavel and colleagues1 for improved data on, and greater attention to inequities in, social determinants of health. People with disability experience poorer health than the general population. Much of this disparity is attributable to entrenched disadvantage in social determinants of health such as employment, housing, and violence.2 Ongoing impacts from the coronavirus disease 2019 (COVID‐19) pandemic may increase disparities in social determinants and health outcomes for people with disability. Yet people with disability remain largely invisible in key data sources.3,4 Disability identifiers are included in several national surveys conducted by the Australian Bureau of Statistics (eg, the Survey of Disability, Ageing and Carers5 and General Social Survey6) and some of the longitudinal datasets mentioned by Flavel and colleagues. However, survey data have limitations: they are based on a population sample, capture data at specific time points, and rely on self‐report information (eg, on income, service use). National surveys typically exclude individuals living in settings such as boarding houses and welfare institutions, where people with disability are over‐represented. Administrative datasets complement survey data. They capture more detailed and often continuous data on individuals, their contact with services and related outcomes. Regrettably, disability status is rarely collected in administrative datasets in Australia. Work is underway on the National Disability Data Asset,7 comprising linked data from multiple state, territory and national administrative sources. This asset will require a robust and consistent basis for identifying individuals with disability, beyond the subset of people identified by their contact with disability‐specific programs or payments. Australia urgently needs a nationally agreed, consistent disability identifier to leverage the opportunity provided by the National Disability Data Asset to identify and monitor disparities in health and social determinants experienced by Australians with disability. Development of such a disability identifier is essential to build a strong evidence base to deliver better outcomes and reduce health inequities. Crucially, people with disability and their representative organisations must be involved as key drivers and decision makers at every stage of development and implementation of the to‐be‐developed national disability identifier.

Nicola Fortune · Jodie Bailie · Gwynnyth Llewellyn

To the Editor: Khatri’s and Stuckey’s1 article, Vertebral fractures after denosumab discontinuation for dental procedures: a consequence of distorted perceptions of risk, sums up the authors’ knowledge and experience of medication‐related osteonecrosis of the jaws (MRONJ) in the latter part of the title. They quote the risk of MRONJ as being very low and equal for denosumab and oral bisphosphonates. This is incorrect. The risk of MRONJ is 0.3%.2 In our study we found the risk following extractions at 1.8%.3 The recent 2022 update of the position paper on MRONJ4 found that the risk with denosumab is an order of magnitude higher than for bisphosphonates. There is no discussion in Khatri’s and Stuckey’s article1 of the effect of MRONJ on patients. A patient with stage 3 or end‐stage MRONJ has months of severe pain and requires jaw resection with or without microvascular reconstruction similar to that required for advanced jaw cancer.4,5 The impact of this is similar to vertebral collapse, both largely avoidable disasters. The current Australian recommendations for dental extractions for patients taking denosumab for osteoporosis are to delay extractions to 6 months after the last injection of denosumab and then to allow 4weeks for initial socket healing before the next injection.6 The risk is greater if the patient has been taking antiresorptives for more than 4years and if they are immunocompromised.4,7 We would agree that education and communication between prescribers, patients and dentists are key. This can only be achieved by close, mutually respectful communication and understanding between medical, dental, oral and maxillofacial surgeons and patients. However, this is easier to say than put into meaningful practice. Most definitive articles on MRONJ are in oral and maxillofacial surgery journals, which are not commonly read by physicians who prescribe antiresorptives. The most likely reason that the patient was taken off denosumab for 5 months was that the dental plan was not only to extract the teeth but to replace them with dental implants. Implants require time for osseointegration.8 Hopefully, this letter helps correct the distorted precepts expressed by Khatri and Stuckey for prescribers of this, otherwise, useful drug.

Alastair Goss