Perspectives

Children may be more susceptible than originally thought and could play a role in community transmission

Zoë Hyde

Lifting lockdown when numbers are low but not zero means that superspreaders may remain, leading to a further wave of the epidemic

David Kault

The response to the largest institutional outbreak of COVID‐19 in health care workers in Australia to date needed to be multidimensional In many countries, high rates of health care workers with coronavirus disease 2019 (COVID‐19) have been associated with inadequate personal protective equipment (PPE), exposure to large numbers of patients with COVID‐19, worker fatigue, and limited access to diagnostic testing.1,2,3 In Australia, during the initial phase of the epidemic, infections in health care workers were largely attributable to international travel, corroborated by genomically distinct severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) lineages.4,5 At the Royal Melbourne Hospital, we observed a marked increase in staff infections during July and August 2020, concurrent with a statewide surge in COVID‐19 cases. To inform future responses in the Australian setting, we present a description of health care worker infections at our institution and the suite of interventions associated with outbreak control. Setting The Royal Melbourne Hospital City Campus is a 550‐bed university‐affiliated tertiary hospital with an additional 150 geriatric and rehabilitation beds at the Royal Park Campus, a large mental health service and four residential aged care facilities, employing about 10 000 staff. Throughout the pandemic, a rapid access COVID‐19 testing clinic was provided for symptomatic staff. To diagnose infection, deep nasal and throat swabs were sampled for nucleic acid testing (reverse transcriptase polymerase chain reaction [RT‐PCR] for SARS‐CoV2). SARS‐CoV‐2 RNA was detected using the Coronavirus Typing assay (AusDiagnostics).5 All positive COVID‐19 tests were notified to the Department of Health and Human Services, with staff members also being notified to the Royal Melbourne Hospital infection prevention services. Infected staff were interviewed by an infection prevention nurse consultant to identify any contacts and enquire about PPE use, work locations in the days before symptoms, nature of their work, characteristics of their patients, and any suspected acquisition events. Contacts with other staff outside work were also explored. Infected staff were required to isolate for 10 days or more after symptom onset and close contacts (≥ 15 minutes of face‐to‐face contact or ≥ 2 hours in a shared space in the 48 hours before symptom onset) were furloughed for 14 days and quarantined, according to state guidelines. Outbreaks (two or more epidemiologically and/or spatially linked staff and/or patients) were managed by a multidisciplinary incident management team. Data regarding health care worker infections were entered into a REDCap 10 (Research Electronic Data Capture) database, a secure web‐based platform, and were analysed using Stata 16. This study was approved by the Melbourne Health Human Research Ethics Committee (QA2020058). Overview Between 1 July and 31 August 2020, 262 cases of COVID‐19 were identified among Royal Melbourne Hospital staff (Box 1 and Box 2). Fifteen individuals (5.7%) required inpatient care and 13 (4.9%) received care by a hospital in the home service. Two were admitted to the intensive care unit (ICU), none requiring mechanical ventilation, with no deaths. Nurses were most commonly affected, followed by support staff (such as food and cleaning services) and doctors (17/21 of these being doctors‐in‐training) (Box 1). The trend in incidence of health care worker infections reflected the prevalence of inpatients with COVID‐19 (Box 3). The ICU had between zero and ten concurrent patients with COVID‐19 over the period (median, 7; interquartile range [IQR], 5.0–8.0), with four ICU staff acquiring COVID‐19. No operating theatre staff and no staff working in affiliated residential aged care facilities were infected. The median turnaround time for health care worker test results (from specimen collection to reporting) was 20.2 hours (IQR, 11.4–29.1 hours). Overt recognised PPE breaches were rarely reported. Contacts with known COVID‐19 cases outside the hospital were infrequent but did occur (eg, health care workers living together). Outbreak linked to geriatric and rehabilitation inpatient wards The Royal Park Campus had the highest number of staff with COVID‐19, making up 40.8% (n = 107) of health care worker infections at the Royal Melbourne Hospital, despite this campus constituting about 10% of the total staff workforce at the hospital (acknowledging that some staff move between sites). Between 12 and 18 July, the Royal Park Campus received a large number of patients from external residential aged care facilities, not affiliated with the Royal Melbourne Hospital, with COVID‐19 outbreaks. These residents were COVID‐19‐positive at admission and were managed with appropriate infection precautions throughout. COVID‐19 cases among staff rapidly escalated across all six wards at the campus after 16 July, peaking on 27 July. The peak number of patients with COVID‐19 at the Royal Park Campus was 60. At the Royal Park Campus there are a variety of buildings constructed from the 1970s to early 2000; most have central air conditioning plants, but one has a local split system. An engineering review of the wards revealed air exchanges met current requirements; however, a more detailed assessment of air movement suggested that some were not as well ventilated as others. Some patients were in single rooms, but many were in multibed spaces. Improved nurse to patient ratios were used to help manage patients. Despite this, because of large numbers of staff furloughs, the remaining staff experienced high workloads. A decision was made on 3 August to close four wards at the Royal Park Campus. Fifteen patients were moved to other health services, while the remaining 45 were moved to single rooms in wards with more modern infrastructure. Outbreaks linked to “hot wards” At the Royal Melbourne Hospital City Campus, most affected staff were working in wards with patients with suspected or confirmed COVID‐19 (“hot wards”) (Box 1). These staff were highly trained in PPE use, PPE was always readily available (ie, gowns, gloves, eye protection, and masks), and use was checked by a PPE “buddy” (usually a colleague) before patient room entry and at doffing. Staff noted that particular behaviour in infected patients appeared to be linked to transmission events (patients shouting, vigorous coughing). The peak combined prevalence of inpatients at the Royal Park and City campuses was 99 on 5 August 2020. As increasing numbers of staff infections were recognised, the density of patients on the COVID‐19 wards was reduced by closing beds in shared rooms and moving each patient to a single room where possible. On 21 July, use of N95 (or P2) masks by all staff at all times on COVID‐19 wards at both campuses was instituted. “Spotters” (supernumerary staff) were deployed to observe PPE donning and doffing, and senior staff ward walk‐arounds and additional cleaning with monitoring were implemented. Staff working on “hot wards” were offered weekly asymptomatic testing to detect any infections early. Outbreaks on “cold wards” On three occasions, clusters occurred outside the designated “hot wards”; that is, in wards not allocated to caring for patients with suspected or confirmed COVID‐19 infection. In some staff, having previously worked at the Royal Park Campus was identified as a potential risk factor. A management plan for these wards was deployed, including closure to new admissions, moving patients to separate rooms (where possible), managing the whole ward using increased precautions, deep cleaning, and voluntary testing of all patients and staff every 3–4 days. Hospital‐wide asymptomatic staff testing was instituted (> 600 staff tested) and whole hospital inpatient testing occurred as a point prevalence activity in late July, with no additional cases identified outside the affected wards. Institutional responses Responses were multifactorial and iterative, with daily review of emerging evidence that informed ongoing decisions. Importantly, a hierarchy of controls was used to manage these outbreaks (Box 4). A proactive approach was used to support infected and furloughed staff wellbeing, with dedicated nursing and medical staff monitoring physical and mental health as well as providing practical supports. This service managed over 680 staff during the outbreak period. Discussion We describe the largest institutional outbreak of SARS‐CoV‐2 health care worker infections reported in Australia to date. Our response was necessarily iterative and pragmatic and advice often pre‐dated formal state and federal recommendations. During these outbreaks, a number of key factors emerged that shaped our responses, extending well beyond a focus on PPE alone. First, the concept of a “critical burden” of infection framed our responses to patient movement and ward closures. Concurrent with large numbers of cases in the hospital and the community, the number of staff who acquired infection rose rapidly. Based on overseas experience,6,7 we hypothesised that large numbers of patients in confined spaces may have created a high density of droplets, aerosols and environmental contamination. This triggered a detailed assessment of ward physical layout, including the possible role of patient placement and air circulation. We elected to use single rooms wherever possible and to physically space infected patients by closing beds on the ward. The intensity of transmission in some wards led to a decision to close wards and move some patients to other health care services. Further, we adopted the use of N95 masks for staff working in areas with large numbers of patients with confirmed or suspected COVID‐19. While use of N95 masks for all COVID‐19 care was not recommended in state or federal guidelines at that time,8,9 this organisational decision was based on our local epidemiology and a need to trial any reasonably available strategy to contain health care worker infections. Second, the availability of rapid and accessible testing for staff was critical to informing real‐time outbreak management, highlighted by international studies.10,11 Rapid availability of data informed our daily incident management meetings and enabled prompt decision making using the best possible information. Finally, the importance of staff communication and wellbeing cannot be understated. Similar to other studies,3,12 many staff reported physical and mental fatigue and stress during these outbreaks. In addition, workforce shortages meant that staff were taking on extra shifts at short notice and working in unfamiliar roles. Accordingly, access to employee support programs was an important element of this response. Box 1 – Demographic characteristics of health care workers with coronavirus disease 2019 (COVID‐19), confirmed by polymerase chain reaction (PCR) testing, at the Royal Melbourne Hospital (1 July – 31 August 2020) Characteristic Number of confirmed cases
(%) Total number of confirmed cases 262 Sex Male 57 (21.8%) Female 205 (78.2%) Median age at diagnosis (IQR), years 32.7 (26.8–44.9) Employee type Nurse 179 (68.3%) Doctor 21 (8.0%) Allied health practitioner 9 (3.4%) Support staff (food services, environmental services) 38 (14.5%) Administrative staff 6 (2.3%) Student 4 (1.5%) Security staff 4 (1.5%) Laboratory staff 1 (0.4%) Location Royal Park Campus (rehabilitation, geriatric rehabilitation) 107 (40.8%) Hot wards* (COVID‐19 wards,† ED, ICU) 57 (21.8%) Cold wards‡ with recognised COVID‐19 outbreaks (3 wards) 20 (7.6%) Cold wards‡ with no outbreaks (1 or 2 unlinked cases; 6 wards) 7 (2.7%) Mental health ward§ 8 (3.1%) Not ward‐based (eg, non‐clinical) 31 (11.8%) Unknown (no campus/ward stated, includes both campuses) 32 (12.2%) ED = emergency department; ICU = intensive care unit; IQR = interquartile range. * Hot wards are wards dedicated to managing patients with confirmed or suspected COVID‐19. † COVID‐19 wards are wards where patients with confirmed or suspected COVID‐19 were managed. ‡ Cold wards are all other wards. § Mental health wards were situated at the City Campus and at other sites. Box 2 – Epidemic curve of health care worker infections at the Royal Melbourne Hospital (1 July – 31 August 2020) RPC = Royal Park Campus. * “Other” includes non‐clinical not ward‐based staff, staff working across several campuses, or ward not known. Mental health wards include off‐site facilities. Box 3 – Prevalence of inpatients with coronavirus disease 2019 (COVID‐19) at both the Royal Melbourne Hospital City Campus and the Royal Park Campus over time (13 July – 31 August 2020)* * Data start on 13 July 2020. Box 4 – Hierarchy of controls used to guide interventions to address health care worker infection with coronavirus disease 2019 (COVID‐19) at Royal Melbourne Hospital Elimination* Public health restrictions to reduce community incidence Testing availability in the community (and for staff) to identify and isolate cases early Rapid turnaround time for test results to identify and isolate cases early Frequent testing of staff and patients in wards with outbreaks for early recognition and management of cases Symptomatic staff furloughed until test results available Furlough asymptomatic staff who are contacts of COVID‐19 cases Work from home policies for staff Telehealth consultations rather than in‐person visits to hospital Visitor restrictions to hospitals (use of phone/iPad to liaise with family) Early discharge of patients not requiring inpatient care, use of hospital in the home services Use of remote meeting technology Engineering controls Attention to ventilation and air circulation in all clinical and non‐clinical areas Availability of negative pressure rooms Physical separation of patient groups (access to single rooms, wards with doors to separate from other wards) Equipment to improve turnaround times for microbiologic testing to enable rapid identification of cases Adequate space for staff to safely don and doff PPE Provision of break rooms with increased space enabling adequate physical separation Physical barriers for public facing non‐clinical staff (eg, perspex barriers) Appropriate cleaning (correct equipment to enable this) Administrative controls Existing policies, procedures and subcommittees (with appropriate governance) in place before the COVID‐19 pandemic regarding infection prevention, PPE, hand hygiene, transmission‐based precautions, cleaning, outbreak management, management of contact tracing, pandemic plan Appropriate governance (Emergency Operations Centre with multidisciplinary representation from all areas) during pandemic Use of national and state guidelines to inform development of hospital COVID‐19 guidelines Regular meetings of key stakeholders to discuss emerging issues Regular communications to staff via email, social media, and remote meetings by hospital executive and managers Policies to encourage physical distancing between staff (staggered breaks, start/stop times, roster redesign) Workflow changes to encourage distancing between staff and patients where possible Use of dedicated “COVID teams” in wards to minimise staff moving between wards Resourcing of staff in “COVID‐19 wards” to ensure manageable workload, improved nurse to patient ratios Bed allocation (avoidance of high density of COVID‐19-positive patients in wards, minimise use of shared rooms) Management of COVID‐19-positive patients in separate wards from COVID‐19‐negative patients Training (baseline and refreshers) and monitoring of PPE use (spotters) for all clinical and non‐clinical staff Increased resourcing of cleaning services and ongoing training in cleaning, using in‐house and not agency staff Monitoring of cleaning (eg, ongoing fluorescent marking programs, spotters) Hand hygiene training and auditing, including development of videos and posters specific to COVID‐19 PPE Universal pandemic precautions (surgical mask and face shields all staff all the time) Masks on patients where possible for source control Use of PPE appropriate to the circumstance (gowns, gloves, surgical masks, N95/P2 masks, eye protection) PPE = personal protective equipment. * Actions to remove or minimise the number of infected people on site.

Kirsty L Buising · Deborah Williamson · Benjamin C Cowie · Jennifer MacLachlan · Elizabeth Orr · Christopher MacIsaac · Eloise Williams · Katherine Bond · Stephen Muhi · James McCarthy · Andrea B Maier · Louis Irving · Denise Heinjus · Cate Kelly · Caroline Marshall

Lessons learnt from COVID-19 outreaks in residential aged care facilities could limit future impacts

Gwendolyn L Gilbert

There is more to female sexual function than circulating testosterone, and symptomatic women require a thorough clinical evaluation The 2019 global consensus position statement on the use of testosterone therapy for women1 aims to provide guidance for clinicians managing women with female sexual dysfunction. The recommendations, graded according to levels of evidence, have been developed by an international taskforce with representatives from a range of organisations and societies, headed by Australian endocrinologist Professor Susan Davis, the current President of the International Menopause Society. The position statement bases many of its recommendations on a systematic review and meta‐analysis of randomised controlled trials.2 The meta‐analysis includes data from 36 randomised controlled trials with 8480 participants and includes studies with a testosterone treatment duration of at least 12 weeks. The primary outcome indicates an improvement in satisfying sexual events, measured as a mean increase of one event over 4 weeks with the use of testosterone. There were also improvements in other parameters associated with sexual function, including sexual desire, arousal and self‐image. There were no cognitive, psychological, wellbeing or musculoskeletal (including bone mineral density) benefits. In doses that approximate physiological levels, the main adverse effects of testosterone therapy are significant increases in acne and hair growth but no difference in alopecia, clitoromegaly or voice change. The position statement provides recommendations covering the assessment of women with female sexual dysfunction, laboratory measurement of testosterone, indications for treatment, and ongoing monitoring once treatment is commenced. It emphasises that the only evidence‐based indication for the use of testosterone in women is the treatment of post‐menopausal women who have been diagnosed with hypoactive sexual dysfunction disorder (HSDD) after formal biopsychosocial assessment. Doses that approximate physiological testosterone concentrations in pre‐menopausal women are recommended. At these doses, testosterone therapy is not associated with serious adverse events. Notably there are no safety data for beyond 24 months of treatment. There are insufficient data regarding the use of testosterone therapy in pre‐menopausal women. The position statement does not comment regarding women with premature ovarian insufficiency and recommends caution in women with a breast cancer diagnosis, reflecting the lack of data.3,4 The classification of female sexual disorders has been a source of controversy and debate. In the Diagnostic and Statistical Manual of Mental Health Disorders, 5th Edition, HSDD and female sexual arousal disorder (FSAD) have been amalgamated and classified as a single entity: female sexual interest/arousal disorder. The writing group for the position statement regards HSDD and FSAD to be distinct conditions, a view shared by experts in the field.5,6 Diagnostic criteria for both conditions are outlined in Box 1.7 It is important to understand this clinical distinction as the position statement does not consider FSAD to be an indication for testosterone therapy. Serum testosterone levels decline during the reproductive years and are significantly reduced in women post‐oophorectomy.8 There was no reported correlation between androgen levels and reported sexual function in one study,9 and there is no serum testosterone cut‐off below which women are more likely to experience HSDD. The authors of the position statement comment that the relationship between endogenous androgen concentrations and sexual function remains uncertain because of issues related to androgen assays in some studies. Notwithstanding these factors, it is recommended that a baseline measurement of testosterone be taken to avoid overtreatment.1 Serum total testosterone rather than free testosterone is the recommended biomarker. The significance of free testosterone in the context of female sexual dysfunction has not been evaluated. Most commercial assays in Australia use immunoassays to measure testosterone, while free testosterone and free androgen index are calculated from total testosterone and sex hormone‐binding globulin. However, immunoassays are considered unreliable, particularly for low levels in the female reference range. Liquid and gas chromatography and mass spectrometry, while not yet in common use, are considered far more accurate and efforts are being made to increase availability.10 The position statement does not comment on whether menopausal hormone therapy should be used concurrently with testosterone therapy. However, a biopsychosocial model of treatment of female sexual dysfunction is recommended, which may include menopausal hormone therapy. Multiple studies have looked at the effect of oestrogen therapy alone, testosterone alone and a combination of the two on female sexual function, with variable outcomes. One of the main criticisms of studies demonstrating no improvement with oestrogen (alone or combined with testosterone) is that low therapeutic doses of oestrogen were used and circulating oestradiol levels were either not measured or were low and did not reach pre‐ovulatory levels consistent with those seen in pre‐menopausal women.11 Only one of eight studies included in the meta‐analysis did not include concomitant menopausal hormone therapy.2 Systemic oestrogen therapy is known to improve hot flushes, urogenital symptoms and mood disturbance, while topical vaginal oestrogen is effective in managing vulvovaginal atrophy.12 Both improve wellbeing in menopausal women and may enhance libido such that other therapies are not required,13 although an improvement in absolute terms has not been described. Our view is that menopausal hormone therapy with oestrogen with or without progestogen should be considered initially in all post‐menopausal women who present with low libido before the initiation of testosterone. A significant issue is the clinical meaningfulness of the finding of a mean increase of one satisfying sexual event per month. Many will argue that such an outcome does not justify the use of testosterone therapy in clinical practice. In context, however, the meaningfulness will depend on the individual woman. For example, for a woman experiencing no satisfying sexual events, an extra one per month would equate to twelve extra events per year and could represent a significant improvement in her quality of life. Consumer involvement in the development of the position statement may have provided insight into this question. In the meta‐analysis, testosterone therapy compared with placebo was found to reduce personal distress, a key component of HSDD, in all studies of post‐menopausal women. Perhaps more relevant to the discussion is whether a satisfying sexual event is adequate as the primary measure of efficacy for treatment. The authors address this in the final part of the position statement. Appropriately, recommendations are made for the design of future clinical trials, including the development of a validated instrument for the screening and diagnosis of HSDD that can also be used as a marker of efficacy of treatment. Australian perspective Sexual dysfunction is common among Australian women. The prevalence of low sexual desire and HSDD was 69.3% and 32.2%, respectively, among a sample of community‐based women aged 40–65 years in one study.14 In older women, the prevalence of HSDD was reported to be one in seven women in a population of women aged 65–79 years living in the community.15 The use of testosterone for therapeutic purposes in women has been controversial. Despite this, clinicians in many countries including Australia have been prescribing testosterone off‐label primarily for low sexual desire in women for several years.16 Various formulations have been used, including subcutaneous pellets, transdermal gels and intramuscular injections that are designed for men, with dosing then adjusted for the female population. In this setting, there is greater potential for treatment to result in supraphysiological testosterone levels. The position statement suggests that ‘‘where approved female preparations are unavailable, off‐label prescribing of an approved male formulation is reasonable”.1 Bio‐identical and compounded products are not recommended. In Australia, a 1% transdermal testosterone cream is available, designed specifically for women and indicated for symptoms caused by testosterone deficiency. Although the product is unlicensed in Australia, it has been available on prescription from pharmacies within Western Australia since 1999 due to an exemption under section 6 of the Therapeutic Goods Act 1989 (Cth). Women prescribed this treatment in other states are required to access it by mail order along with a prescription from their clinician. It was submitted for Therapeutic Goods Administration evaluation and inclusion on the Australian Register of Therapeutic Goods as a registered product in 2019 for the proposed indication of treatment of hypoactive sexual desire dysfunction in post‐menopausal women, and a response is expected by the end of 2020 (Michael Buckley, Medical Director, Lawley Pharmaceuticals, personal communications). Although limited by small sample sizes, pharmacokinetic and clinical studies of the recommended 5–10 mg daily dose of this cream demonstrated total and free testosterone levels within or above the pre‐menopausal range.17,18,19 Its availability in Australia overcomes the need to consider male preparations or compounded and bio‐identical products. The publication of the position statement will increase awareness of female sexual dysfunction in the medical community. Clinicians managing these patients will require education in practical terms about assessing such patients for HSDD and safely prescribing and monitoring testosterone therapy when indicated. Based on the authors’ expertise and experience, we propose an algorithm for the assessment of women presenting with female sexual dysfunction and use of testosterone (Box 2), with women initially undergoing a biopsychosocial assessment.6 We recommend information sheets for clinicians and patients written by local experts be made available on the websites of the Australasian Menopause Society, the Endocrine Society of Australia and the Royal Australian and New Zealand College of Obstetricians and Gynaecologists. Conclusion The position statement is a timely addition to the literature regarding testosterone therapy for women. What is clear is that there is more to female sexual function than circulating testosterone and that symptomatic women require a thorough clinical evaluation, assessing for other factors which may be contributing to their presentation. Testosterone therapy is a small piece of the puzzle in the management of female sexual dysfunction. This position statement provides clarification regarding the indication, adverse effects and knowledge gaps. The availability in Australia of a transdermal testosterone preparation designed for women obviates the need to use male preparations, but further research regarding efficacy and safety is necessary. Future studies should address the safety of long term use of testosterone in women, particularly with respect to cardiovascular disease and breast cancer. Box 1 – Definition of hypoactive sexual desire disorder and female sexual arousal disorder7 Disorder Definition Hypoactive sexual desire disorder Any of the following for a minimum of 6 months: lack of motivation for sexual activity as manifest by reduced or absent spontaneous desire (sexual thoughts or fantasies); or reduced or absent responsive desire to erotic cues and stimulation or inability to maintain desire or interest through sexual activity loss of desire to initiate or participate in sexual activity, including behavioural responses such as avoidance of situations that could lead to sexual activity that is not secondary to sexual pain disorder AND is combined with clinically significant personal distress that includes frustration, grief, guilt, incompetence, loss, sadness, sorrow or worry Female sexual arousal disorder* Female cognitive arousal disorder Distressing difficulty or inability to attain or maintain adequate mental excitement associated with sexual activity as manifest by problems with feeling engaged, or mentally turned on or sexually aroused for a minimum of 6 months Female genital arousal disorder Distressing difficulty or inability to attain or maintain adequate genital response associated with sexual activity for a minimum of 6 months, including: vulvovaginal lubrication engorgement of the genitalia sensitivity of the genitalia associated with sexual activity Disorders related to: vascular injury or dysfunction, or neurological injury or dysfunction * Encompasses female cognitive arousal disorder and female genital arousal disorder Box 2 – Testosterone therapy for post‐menopausal women: proposed algorithm CNS = central nervous system; CVD = cardiovascular disease; HSDD = hypoactive sexual desire disorder; IMS = International Menopause Society; VTE = venous thromboembolism. * Simon et al.6 † Jane and Davis.20

Christina Jang · Jacqueline A Boyle · Amanda Vincent

The involvement of patient advocates should ensure that guidelines are rigorously patient‐focused The fundamental objective of clinical management guidelines for any disease entity is to ensure that the information required to provide evidence‐based management recommendations to patients is readily available to their treating clinicians. It is well established that familiarity with guidelines by clinicians and adherence to them increases the number of patients receiving best‐practice care and improves outcomes.1 However, although management guidelines are intended primarily for clinicians, they must also reflect the patient perspective.2 Patient representation on the Melanoma Guidelines Working Party After identifying the need to produce new Australian guidelines on the management of melanoma, a multidisciplinary working party — under the auspices of Cancer Council Australia and the Melanoma Institute Australia — was established in 2014 to critically assess new evidence and update the previous national guidelines. In addition to clinicians and researchers, two patient advocates (consumer representatives), representing patients with melanoma from around Australia through their affiliations with patient advocacy and support networks, were invited to join the working party. These patient advocates had personal experience of melanoma diagnosis and treatment and extensive prior involvement in melanoma advocacy. This meant that they were able to make important contributions based on their experience as well as providing feedback from the patient networks they represented. Electronic publication Whereas the two previous editions of Australian melanoma guidelines, published in 1999 and 2008 respectively, were printed documents, each taking more than 4 years to compile,3,4 the new guidelines were electronic and were made available on Cancer Council Australia's Wiki platform.5 This publication format allowed individual sections to be published as they were completed and permitted selective updating as new evidence became available. Electronic publication has also meant that the guidelines are more readily accessible to both doctors and patients. They can simply search on their computer, tablet or smartphone for “Australian melanoma guidelines” or go directly to the guidelines website (https://wiki.cancer.org.au/australia/Guidelines:Melanoma).5 The level of evidence supporting each guideline recommendation is clearly documented, informing doctors and assisting patients in their decision‐making process. Patients’ expectations from their treating clinician When patients who are concerned about the possibility of having a primary melanoma consult a general practitioner, dermatologist or surgeon, they are entitled to expect that evidence‐based guidelines will be followed and therefore that the steps below will take place: if there are any suspicious skin lesions, they will be carefully examined; if the doctor suspects that a lesion may be a melanoma, the recommended form of biopsy will be carried out (usually complete excision biopsy with a 2–3 mm margin); if melanoma is diagnosed, the recommended treatment and likely outcome will be clearly explained; the melanoma will be staged correctly, appropriately wide surgical excision will be recommended, and the option of sentinel node biopsy will be discussed for melanomas 1 mm or greater in Breslow thickness or 0.8–1.0 mm in thickness with higher risk pathological features, so that prognosis can be estimated accurately and the eligibility for adjuvant post‐operative systemic therapy can be determined;6,7 and the potential benefits and possible side effects of adjuvant therapy will be discussed. When patients are given a diagnosis of metastatic melanoma in regional lymph nodes, or at a systemic site and are referred to a surgeon, medical oncologist or radiation oncologist, again, they should be able to expect that the following will happen: appropriate surgery will be recommended, and surgery that may be unnecessary (eg, completion lymph node dissection for sentinel node positivity8,9) will not be undertaken without a full discussion of the advantages and disadvantages; and if surgery is not considered appropriate, therapeutic systemic therapy options will be discussed, again with a realistic description of the likely benefits and possible side effects.10 Use of the guidelines While different stakeholders will use guidelines in different ways, they are intended to be useful to both clinicians and patients: to GPs, particularly those who work in skin cancer clinics; to dermatologists, who see many patients when they first present with a primary melanoma but who rarely manage patients with metastatic melanoma; to surgeons who treat patients with both primary and metastatic melanoma; to medical and radiation oncologists who are involved in the care of patients with metastatic disease; and importantly, to patients by providing a reliable source of information. It is hoped that by having access to guidelines based on the best available evidence, both treating clinicians and patients will be better informed, treatment options will be better understood, and patients will receive the most appropriate care. The ongoing involvement of patient advocates in the process of developing and updating the Australian melanoma guidelines should ensure that they are useful and relevant to patients and that the health care outcomes most valued by them are considered, resulting in guidelines that are rigorously patient‐focused, as they should be.

Alison E Button‐Sloan · John F Thompson

A recently developed national transport and cryopreservation service improves equity of access to fertility care for young people across Australia Loss of fertility, which is a well recognised complication of cancer treatment, has a significant impact on quality of life and is ranked as one of the top survivorship concerns.1 Improvements in survival (> 88% for adolescents and young adults) and expansion of fertility preserving options have stimulated rapid evolution of the fertility preservation landscape, aiming to decrease this devastating impact of cancer therapy.2 In addition to the gonadotoxic burden of cancer treatment, societal trends of delayed fertility mean that there are many young people who have not yet completed or even commenced trying for a family when diagnosed with a life‐threatening illness. Fertility discussion, and provision of services to preserve gametes or tissue, is no longer seen as a distraction or a luxury but is acknowledged as a mandatory part of cancer management.3 However, barriers to provision of fertility preservation care remain. These include the lack of education of health care providers about both the long term fertility consequences of cancer treatment and the clinical value of available options. There is also often a lack of clarity about whose role it is to educate patients about these options. Importantly, there can be significant logistic, geographic and economic barriers for patients, especially outside the major centres, such that only 4–50% of young people take up fertility preservation in a timely fashion.4 Established strategies to preserve fertility for the future include medical therapies to protect the primordial follicle pool, vitrification of oocytes and embryos, and ovarian tissue cryopreservation for females. Mature sperm freezing is undertaken in post‐pubertal males, and testicular tissue cryopreservation, while providing the only opportunity for pre‐pubertal boys, is still considered experimental.3 Gonadal tissue cryopreservation Ovarian tissue cryopreservation is the only option for prepubertal girls and may be the only or best option for women at high risk of infertility from cancer treatment, particularly if there are time constraints or safety concerns with other options. Ovarian tissue cryopreservation is no longer considered experimental by peak bodies,3 and there have been over 140 births worldwide following ovarian tissue grafting, including several in young women whose ovarian tissue was cryopreserved as pre‐pubertal children.5 Testicular tissue cryopreservation provides an experimental option for fertility preservation in pre‐pubertal boys at significant risk of azoospermia from gonadotoxic treatments. As boys do not produce mature sperm that can be frozen, a treatment involving testicular biopsy and cryopreservation of spermatogonial stem cells, followed by transplantation into the testis after treatment, is proposed to allow restoration of fertility.6 Recent publications of in vitro sperm maturation and live birth success using the primate animal model provide optimism, such that the joint international consensus statement of peak fertility bodies in 2015 recommended that testicular tissue cryopreservation should be offered for pre‐pubertal boys,3 despite the currently experimental nature of future use, especially as there are no other options. Testicular biopsy can be safely performed and often coordinated concomitantly with other medically necessary procedures without delaying the start of treatment.7 While cryopreservation of eggs, sperm and embryos is a routine part of assisted reproductive laboratory activity, the technique for cryopreservation of ovarian and testicular tissue is biophysically different, and very few centres nationally and internationally have established tissue cryopreservation laboratories with validated, published protocols and clinical success after thawing.8 Due to distance challenges within Australia and lack of resources to meet the needs of these patients, particularly those who reside in rural and remote areas, ovarian and testicular tissue cryopreservation is not accessible to over 70% of people who would benefit from this opportunity.4 Establishment of the National Ovarian and Testicular Transport and Cryopreservation Service To provide equity of access to fertility care, the Fertility Preservation Service at the Royal Women's Hospital in Melbourne has developed the National Ovarian and Testicular Transport and Cryopreservation Service, allowing collaboration between local units and specialised centres, with professional and patient education as part of the program. There are several successful international models for collaborative care with published protocols and data to support transportation and storage of gonadal tissue or gametes in a specialised centre with expertise, health and safety regulations.9 The live birth rates with and without overnight transportation are comparable.9 The Fertility Preservation Service, a partnership between the Royal Women's Hospital and Melbourne IVF, was established 30 years ago. It is the largest service of its kind in Australia, with clinical expertise in counselling, cryopreservation, testing, storage and transport procedures. It sees about 300 patients per year, managed by a multidisciplinary team of fertility specialists, nurses, research scientists, research managers, laboratory staff, counsellors and administrators. There have been increasing referrals each year, reflecting the increased demand. Eighty percent of these are cancer related, with serious medical conditions and gender dysphoria forming the remainder. The Fertility Preservation Service has built extremely strong relationships with cancer centres, both in Victoria and nationally, and collaborates closely with other fertility preservation units, including the Fertility and Research Centre in NSW and the Royal Children's Hospital Melbourne. It supports data collection for the Australasian Oncofertility Registry.10 Based on a recent Fertility Society of Australia survey,11 we believe that Victoria, possibly because of both the dedicated fertility preservation centre and the well developed collaborative relationships, has the highest rate of patients referred for fertility consultation. The service stores gonadal tissue from over 1000 patients. Ovarian tissue grafting has been performed in 40 patients, with five children born and a live birth rate of 20% per embryo transferred. There are also mature eggs and embryos which have been cryopreserved from ovarian tissue stimulation. Testicular tissue has been cryopreserved for 163 patients. The establishment of a centralised national tissue retrieval and transport program allows gonadal tissue harvesting to take place in a local centre with subsequent transportation to the central laboratory for processing, cryopreservation and storage. Communication with the National Ovarian and Testicular Transport and Cryopreservation Service team occurs via a paging service which is checked daily by a dedicated nurse, with treating clinician and, when appropriate, patient follow‐up by teleconference. Subsequently, ovarian tissue grafting may be performed at the Royal Women's Hospital or the tissue can be transported back to the local centres. The service also provides follow‐up and psychological support for patients, and educational resources to assist with all aspects of fertility preservation, both for patients and health practitioners. Fertility preservation is a mandatory part of cancer care; the National Ovarian and Testicular Transport and Cryopreservation Service program will improve equity of access to fertility preservation for young women and men around Australia.

Genia Rozen · Stephanie Sii · Franca Agresta · Debra Gook · Catharyn Stern

Cancer immunotherapy during the COVID‐19 pandemic presents management challenges from immune‐related toxicities, requiring careful patient selection The coronavirus disease 2019 (COVID‐19) pandemic has led to fundamental re‐evaluation of the benefits versus risks of treatment in oncology. Immunotherapy has had an expanding presence in oncology, becoming a primary systemic treatment option in diseases such as melanoma, lung, urothelial, renal, and head and neck cancers. Immune checkpoint inhibitor (ICI) therapy, namely anti‐programmed cell death protein 1 (anti‐PD‐1), anti‐programmed cell death ligand 1 (anti‐PD‐L1) and anti‐cytotoxic T‐lymphocyte‐associated protein 4 (anti‐CTLA‐4) antibodies, halt the negative regulatory checks of T lymphocytes, thus activating the immune response against tumours. Patients with cancer receiving these treatments are faced with a unique set of treatment‐related toxicities driven by an autoimmune mechanism. An association between immune‐related adverse events (irAEs) and severe COVID‐19 has been raised during the current outbreak. In particular, an overlap in the physiological insult from immunotherapy‐mediated pneumonitis and severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2)‐related interstitial pneumonia is hypothesised.1 Both conditions may present with lung parenchymal changes, and their coexistence may potentially aggravate the underlying interstitial inflammatory infiltrate and diffuse alveolar damage, leading to a common final pathway of respiratory failure. Pre‐existing lung pathology is expected to be a risk factor for COVID‐19 pneumonia, with higher incidence in patients with lung cancer and smokers.2 Whether prior thoracic radiation may have an impact on outcomes from COVID‐19 pneumonia is unknown. Parallels have been drawn between the cytokine storm driving COVID‐19‐associated acute respiratory distress syndrome and cytokine release syndrome as a complication following T cell‐engaging therapies, such as chimeric antigen receptor T cell and CD3‐based bispecific T cell engager therapies. It is known that interleukin (IL)‐6, IL‐10 and interferon (IFN)‐γ are key drivers behind cytokine release syndrome. Elevated circulating IL‐6 levels have been observed in patients with COVID‐19‐associated pneumonia.3 Patients with severe COVID‐19 have significantly higher circulating levels of pro‐inflammatory cytokines, including IL‐1B, IL‐6, IL‐8 and IL‐10, compared with milder cases of COVID‐19;3 and elevated IL‐6 has been shown to be a predictor of mortality risk. Patients with immune‐related toxicity have higher levels of 11 circulating cytokines, such as G‐CSF, GM‐CSF, IFN‐α‐2, IL‐1a, IL‐1B, IL‐2 and IL‐12,4 with some but incomplete overlap with the cytokine milieu seen in severe COVID‐19 cases.3 The outcomes of COVID‐19 in patients with cancer treated with immunotherapy remain under investigation, with some2,5 but not all6 studies suggesting a more severe outcome. In a multicentre study from China involving 105 patients with cancer infected with SARS‐CoV‐2, 6% received anti‐PD‐1 therapy within 40 days of COVID‐19 symptom onset and experienced increased risk of death and critical symptoms.2 Another series of 423 cancer patients with SARS‐CoV‐2 infection from New York City also reported that treatment with ICI therapy within 90 days was a predictor for admission to hospital and for severe respiratory illness, defined as the requirement for high flow oxygen supplementation or mechanical ventilation.5 Of interest, even after exclusion of patients with lung cancer, the ICI group experienced worse outcomes, inferring that the ICI therapy itself conferred inferior COVID‐19 outcomes without the confounding effect of lung cancer, which had been shown as an independent predictor of poor prognosis in COVID‐19. However, an interim analysis of the first 200 patients from the Thoracic Cancers International COVID‐19 Collaboration (TERAVOLT) registry of patients with thoracic malignancies did not observe a worse outcome among the 37% of patients receiving ICI therapy (23% ICI alone and 14% ICI plus chemotherapy), with data collection ongoing.6 Dual checkpoint inhibitor (anti‐CTLA4 with anti‐PD1 antibody) therapy has achieved high response rates in a number of cancer types,7,8 but is associated with greater incidence and severity of treatment‐related toxicity compared with monotherapy.7 This has several implications. Firstly, differentiating between immune‐mediated pneumonitis and COVID‐19‐associated pneumonia can be difficult due to similarities in clinical and radiological features. Earlier in the pandemic, there were concerns that this may cause delays in initiation of corticosteroids, which is the standard management of irAEs. However, emerging evidence for potential benefit of dexamethasone in severe cases of COVID‐199 reduces concerns for its empirical use in cases where immune‐mediated pneumonitis is a differential diagnosis. Secondly, patients with severe irAEs, such as immune‐mediated pneumonitis requiring intensive care support may face a health system already strained by demand from COVID‐19 cases. Finally, severe irAEs require treatment with high dose corticosteroid and, at times, additional immunosuppressive agents, such as infliximab and mycophenolate. To avoid rebound of the irAEs, corticosteroids are weaned over 6–8 weeks, subjecting patients to prolonged immunosuppression that can predispose them to opportunistic and nosocomial infections.10,11 This has the potential to add further burden to the health care system. The impact of cancer immunotherapy on microbial infection in general is not fully understood. A retrospective study of patients with metastatic melanoma receiving immunotherapy (mainly ipilimumab, an anti‐CTLA4 antibody) reported a 7.3% incidence of serious infections due to a variety of bacterial, viral, fungal or parasitic infections requiring hospitalisation or parenteral antimicrobials.10 Nonetheless, the study of this interaction is complex, with the receipt of corticosteroids for irAEs and having diabetes as a comorbidity10,12 associated with an increased risk of infection in patients with cancer receiving ICI therapy. Furthermore, immune checkpoint blockade can reactivate tuberculosis and viral infections. There are case reports of acute tuberculosis developing in patients with cancer receiving immunotherapy, without concurrent corticosteroid therapy.13 At least three of five cases were suspected to represent reactivation of latent tuberculosis, which may be directly mediated through PD‐1 inhibition driving an exaggerated immune response to tuberculosis infection. Another consideration for patients with cancer receiving immunotherapy is influenza vaccination during the COVID‐19 pandemic. While there is currently no vaccine specifically against COVID‐19, many health authorities encourage the uptake of influenza vaccination to reduce the concurrent burden from influenza illnesses, particularly for nations approaching winter facing the seasonal influenza period. Controversy surrounds whether influenza vaccination in patients receiving cancer immunotherapy heightens the risk of irAEs.14 Numerous retrospective series support the safety of inactivated influenza vaccine in recipients of anti‐PD‐1 monotherapy, with no increase in irAEs observed.15 Reassuringly, influenza vaccination had no adverse impact on the anticancer effect of ICI therapy.14,15 However, there may be heightened concerns for influenza vaccination in combination immunotherapy (anti‐PD‐1 with anti‐CTLA‐4) recipients, as they are more prone to irAEs, including rarer, but potentially fatal, complications such as immune‐mediated myocarditis. This potential concern for influenza vaccination in recipients of combination ICI can leave this patient population more vulnerable from influenza infection. For patients taking monotherapy ICI, current evidence supports the safety and efficacy for influenza vaccination. There are guidelines addressing the use of cancer immunotherapy in the COVID‐19 era.16,17 These call for careful considerations on the use of dual checkpoint inhibitor therapy depending on the local prevalence of community transmission and the capacity of the local health service to cope with demand.16 On a practical note, this requires individual patient risk–benefit assessment. Patient factors such as age, smoking and comorbidities (eg, diabetes and chronic obstructive pulmonary disease) may affect their recovery from irAEs and affect the outcomes from concomitant COVID‐19. Tumour factors for consideration include the burden and biology of disease. Combination immunotherapy may be justified, for example, in a young patient with metastatic melanoma with high disease burden and/or intracranial metastases. This is in contrast to a patient with underlying comorbidities who has low volume disease and/or disease characteristics, such as underlying B‐Raf proto‐oncogene (BRAF) V600K mutation or desmoplastic melanoma subtype, associated with higher likelihood of response to single‐agent anti‐PD‐1/anti‐PD‐L1 therapy. Current guidelines recommend ICI monotherapy to be delivered at increased dosing intervals, such as nivolumab four times per week and pembrolizumab six times per week.16 These approved alternate schedules have been shown to maintain therapeutic efficacy, while advantageous in reducing patient attendance at health care facilities, potentially reducing exposure and community transmission of COVID‐19. The timing of immunotherapy cessation in patients is another consideration. A number of trials in metastatic non‐small cell lung cancer had a 2‐year treatment duration for immunotherapy in responding patients.18 Data on metastatic melanoma support that cessation of anti‐PD‐1 after at least 6 months of therapy in patients achieving complete response can be feasible without adversely affecting outcome.19 Selection of patients with cancer suitable to stop immunotherapy may further reduce these patients’ hospital visits and may potentially reduce the chance of acquiring COVID‐19. There are international efforts to collate the clinical experience of COVID‐19 in patients receiving cancer immunotherapy.6 These registries will provide a valuable resource for further areas of research, such as assessing the impact of irAEs on COVID‐19. The data will also improve our understanding of the outcomes in this patient population to aid management decisions and counsel patients. Research on potential biomarkers of disease severity may also assist in patient triage. In this rapidly evolving area, it is helpful for practising clinicians to maintain current knowledge through regularly updated resources (Box). In summary, the increased role of ICI therapy in oncology calls for consideration of the impact of their use during the COVID‐19 pandemic. While these agents are not directly immunosuppressive, as with cytotoxic chemotherapy, ICI‐associated toxicities pose diagnostic and therapeutic challenges for management in the setting of a COVID‐19 outbreak. Overlapping clinical and radiographic features in immune‐mediated pneumonitis and COVID‐19‐associated pneumonia may cause diagnostic difficulties at initial presentation. Severe irAEs requiring corticosteroids and prolonged immunosuppression may predispose patients to opportunistic infections. Furthermore, there is a possibility of worse outcomes in the setting of COVID‐19 with underlying immune‐mediated pneumonitis and damaging inflammatory response from immune checkpoint blockade. Practical measures, namely prolonging treatment interval and careful patient selection for combination ICI therapy, may help minimise harm. Box – Practice points for cancer immunotherapy during the coronavirus disease 2019 (COVID‐19) pandemic Judicious use of combination anti‐CTLA-4 and anti‐PD-1/anti‐PD-L1 immunotherapy in patients requiring high tumour response rate with good organ functional reserve. Combination checkpoint therapy is associated with higher rate for immune‐related toxicities (eg, pneumonitis), which may potentially have an adverse impact on outcomes in patients with COVID‐19 Use of approved dosing schedule with longer duration between treatments (eg, nivolumab every 4 weeks, pembrolizumab every 6 weeks) Individualised assessment for pausing or cessation of immunotherapy in patients with controlled low disease burden Rapid assessment and COVID‐19 testing for patients receiving cancer immunotherapy who have clinical presentations with overlapping features for COVID‐19 and immune‐related adverse events Prevention of co‐infections: seasonal influenza vaccination for patients taking single‐agent immune checkpoint inhibitor (the use in combination checkpoint recipients should be individualised). Pneumocystis jirovecii prophylaxis for patients receiving prolonged corticosteroid therapy for immune‐mediated toxicities Maintain current knowledge through professional journals, dynamic resource links (examples below) and webinars sharing clinical knowledge and experience internationally: ▸ Clinical Oncology Society of Australia (https://www.cosa.org.au/publications/covid‐19-updates/articles/) ▸ American Society of Clinical Oncology (https://www.asco.org/asco-coronavirus‐information) ▸ European Society for Medical Oncology (https://www.esmo.org/covid‐19-and‐cancer/covid‐19-full‐coverage) ▸ Journal of Thoracic Oncology (https://www.jto.org/content/covid19) anti‐CTLA‐4 = anti‐cytotoxic T‐lymphocyte‐associated protein 4; anti‐PD‐1 = anti‐programmed cell death protein 1; anti‐PD‐L1 = anti‐programmed cell death ligand 1.

Yada Kanjanapan · Desmond Yip

Fit testing of respirators is recommended to ensure proper fit for individual health care workers and is required to comply with respirator standards

Adrian Regli · Britta S Ungern‐Sternberg

As we face a second wave of the pandemic, mobility data may assist government public health action

Lucinda Adams · Robert J Adams · Tarun Bastiampillai

The arrival of social media coincides with the point at which the antivaccination movement became globally coordinated

Helen Petousis‐Harris · Lisbeth Alley

Name badges are a simple additional method of communicating doctors’ names to patients and their families, but uptake remains poor Most new relationships begin with an exchange of names and most existing relationships are reinforced using names. Except in health care. Despite campaigns such as #hellomynameis (https://www.hellomynameis.org.uk/), clinicians’ names remain absent from many health care experiences and environments. Patients meet many people during an illness journey, particularly when that takes place in a public hospital. There are nurses rotating through different shifts, the specialist under whom the patient is admitted, a registrar, resident or intern, plus maybe a medical student or two. There are also teams of allied health providers. One study found that 75% of inpatients were unable to name anyone when asked to recall the name of the physician in charge of their care.1 Limited recall of doctors’ names is part of a broader pattern. Only 42% of patients can name their diagnosis at discharge,2 and in a study of older patients, only 18% of patients could recall a single message one hour after the ward round, falling to 9% four hours after the ward round.3 The very basic components of effective health care communication, particularly in hospitals, are lagging. Barriers to effective communication are complex and structural.4 Let's for a moment focus on the most fundamental information transfer in any patient–doctor encounter: names. Patients’ names are documented from the moment of admission, printed on sticky labels, placed on wrist bands, attached to meal trays, printed on patient lists, and displayed and discussed in ward and team meetings. In contrast, doctors’ names usually appear only on faded ID swipe cards attached at the hip, or on crowded lanyards around the neck. This asymmetry in identification is just one symptom of the enormous information gulf separating patients and their doctors. Studies consistently show that the majority of patients believe doctors should wear name badges,5 with a preferred site being the breast pocket.6 In 2019, our hospital introduced voluntary name badges for all interns and residents. Something as simple as a name badge, which nearly every other service‐oriented industry employs without question, required careful navigation in the hospital. What name should appear on the badge? Should surnames be included? Should “Intern” or “Resident” or just “Doctor” appear on the badge? Badges were rolled out to interns and residents at the start of the clinical year with a mixed response. Anecdotally, senior doctors commented favourably on the badges, and nurses and allied health workers found it helpful for learning and remembering the names of doctors rotating through their wards. Mid‐year, we collected data on how many interns and residents were wearing badges. During two compulsory teaching sessions, we quietly counted the number of interns and residents wearing name badges: adherence was a lowly 25%. To determine why three‐quarters of interns and residents were not wearing badges, we circulated a voluntary, anonymous and electronic survey to all 108 interns and residents. Our aim was to identify levers or incentives that we could incorporate into a series of behavioural nudges to improve name badge adherence. Around one‐third (34%) of the cohort took part in the survey, 80% of whom did not wear their name badge. Half of respondents reported that their ID swipe card contained their name and was sufficient. About one‐fifth (22%) did not see a need to wear a name badge and a similar number mentioned that senior doctors not wearing badges discouraged them from wearing one. Not wanting members of the public or patients to know their name was a reason indicated by 16% of respondents. Free text responses mainly centred on forgetting to, or being annoyed by, attaching it each day. In response, we have developed new strategies to increase name badge adherence. For example, a brief lecture will be given on the evidence‐base underpinning good communication, coffee vouchers will be provided to doctors seen wearing their badges, badges will be provided to new interns during orientation, and name badges will soon be rolled out across the hospital for all medical staff. Making name badges available to senior doctors is important as they can influence the cultures within units and teams, and our cohort identified a lack of badges among seniors as a barrier to adherence. An informal poll of intern and resident representatives across New South Wales suggests a similar pattern of poor name badge adherence. Five networks with name badges reported that adoption by junior doctors was low. Four networks did not provide name badges. Only three networks provided name badges and have good adherence among junior doctors. As pressure on hospitals, and our clinical interactions, continues to grow, we must look for ways to support effective communication. Alongside a clear introduction, easy‐to‐read name badges reinforce familiarity and contribute to rapport between patients and our (increasingly) busy workforce.

Benjamin D Bravery · Jovana Stojkov · Jeremy Brown

Australia has relied, for most of its history, on international medical graduates (IMGs) to supplement its workforce. Since 1978, IMGs applying for general registration to practise in Australia have usually needed to pass the examinations of the Australian Medical Examining Council, or since 1986, its successor, the Australian Medical Council (AMC). The AMC provides several pathways to registration by the Australian Health Practitioner Regulation Agency (AHPRA). The route now termed “the standard pathway” consists of a two‐part assessment including a multiple choice question (MCQ) examination followed by a clinical examination. While most IMGs are required to pass both examinations, since 2007, IMGs who qualified in the so‐called competent authority countries (the United Kingdom, Ireland, the United States and Canada) have usually not been required to sit these examinations.1 The examinations have sometimes provoked controversy and political responses in various forms.2,3,4 Partly in reaction to these, but mainly through an internal process of continuous improvement, their formats have been adapted considerably over the 42‐year period. The MCQ examination assesses “basic and applied medical knowledge across a wide range of topics,” and since 2000, its pass mark has been set using item response theory.5,6 The original clinical examination used short cases and viva voces; in 2004, this was replaced by a 16‐station objective structured clinical examination (OSCE). The standard of both examinations is set at that “of newly qualified graduates of Australian medical schools who are about to commence intern training”.6 The last account of the demographic features of candidates attempting the examinations and their performance was provided in 2010.5 Now, a decade later, there have been striking changes in both these parameters, which we document and evaluate in this article. A further aim was to identify some demographic or candidate factors that might influence examination success. Source of data De‐identified information about candidates who took the MCQ and clinical examinations of the Australian Medical Examining Council and AMC, from their inception in 1978 until October 2019, were provided by the Council. It included the country and year of primary medical qualification, gender, year of birth, years of first attempt and success, and number of attempts for each candidate. From this information, we calculated the numbers of candidates, numbers of attempts, the success rate per attempt, and the proportion eventually achieving success each year. To examine the contributions of individual countries, results were aggregated into decades. Countries of training were also consolidated into regions, according to the United Nations geographical regions report, last updated in 1999 (Supporting information, table 1).7 Ethics approval was obtained from the University of Melbourne Human Research Ethics Committee (ID: 1750338.3). Demographic features of candidates Over the 42‐year period, a total of 35 699 candidates from 153 countries sat the MCQ examination, 16 588 (46.7%) of whom were female (Box 1). The median age of all candidates at their first MCQ attempt was 32 years (interquartile range [IQR], 28–37 years; range, 20–73 years). The clinical examination was attempted by 20 494 candidates. Their demographic features were similar to that of the candidates for the MCQ. Box 1 shows the number of candidates for the MCQ and clinical examination for the top ten countries of primary medical qualification at each examination. The data for countries grouped by UN region are provided in the online Supporting information, table 1, and data for candidates from all individual countries (except those with very few candidates) are provided in the online Supporting information, table 2. South Asia was the region contributing most candidates, with just under half the total — predominantly graduates from India, Pakistan and Sri Lanka. Next in order were those from South‐East Asia and North Africa. Candidate performance From a low base until about the year 2000, there was a marked increase in candidates attempting each examination, reaching a peak in 2009 for the MCQ and 4 years later for the clinical examination (Box 2 and Box 3). Although the candidate numbers declined slightly after these peaks, they remained almost fourfold higher than in 2000. The pass rate at each attempt in the MCQ examination fluctuated, with most year‐to‐year variations not reaching statistical significance. However, overall pass rates per attempt increased over time, from a low of 28% in 1987 to a high of 66% in 2018. Some candidates showed great persistence: 86 attempted the examination ten or more times. As with the MCQ examination, the pass rate in the clinical examination increased between the 1980s and the 2000s, reaching a peak of 64% in 2007. However, between 2011 and 2012 it fell by more than 10%, followed by a further decline; and for the past 5 years (excepting 2019 when data were incomplete), it has remained just above 30%. Nevertheless, most candidates who persevered managed to pass after one or two further attempts. As with the MCQ, there were a few who found it much more difficult. Five or more attempts were made by 621 candidates (3.0%), 144 of whom have not yet succeeded. Pass rates by individual country are provided in the Supporting information, table 3. In the MCQ, during the past three decades, women had a higher pass rate per attempt and overall, although the magnitude of the difference (about 3%) was small (Box 4). In the clinical examination since 1990, women had both a higher pass rate and fewer attempts. In the most recent decade, the difference in pass rates was substantial (+12%). Box 5 and Box 6 show the pass rates in the MCQ and clinical examinations, respectively, graphed against candidates’ age and the interval (recency) since their medical graduation. There was a marked decline in success with both increasing age and interval since graduation; this was more marked in the clinical examination. While the number of candidates who were 55 years or older was small (245; 1.2% of total), their pass rate was one‐third that of candidates aged 20–29 years, and only 45% of the older group eventually passed. Commentary Before 2000, the number of IMGs attempting AMC examinations annually was usually less than 300 and never exceeded 600. However, between 2000 and 2018, candidate numbers increased more than threefold to an annual mean of 1003 during a period when the number of all Australians born overseas increased only from 4.5 to 7.3 million.8 Some factors likely to have contributed to the increase in candidates were removal in 1998 of the requirement to be an Australian citizen, and offering the computer‐delivered MCQ examination from 2005 in several centres outside Australia. A further increase in candidates for the MCQ examination resulted from the 2006 decision by the Council of Australian Governments that all IMGs with limited or temporary registration with the individual state medical boards should pass that examination. The peak in attempts at the clinical examination in 2013 followed the establishment of the National Registration and Accreditation Scheme in July 2010 and the requirement that limited registrants (non‐specialists) demonstrate progress towards full registration (including passing the AMC clinical examination where applicable). It is important to note that these data are specific to those sitting the AMC examinations. They give only a partial picture of medical immigration over this period. Firstly, they do not include IMGs who were registered as specialists by the various states, and subsequently by AHPRA on advice from specialist colleges. Secondly, until 1992 the Medical Acts in all Australian states allowed graduates from the UK (and usually Ireland) exemption from the need for further examination. For the next 15 years, generalists from those countries usually had to take the AMC examinations, but from 2007 they were again exempted (along with IMGs from Canada and the US) when the AMC introduced the competent authority pathway. The overall success rate in the MCQ examination increased significantly from the 1980s. The AMC made several changes over that time to increase reliability and fairness. One was altering question types to formats less dependent on English language skill; another was publication of annotated question banks to assist candidates in their preparation.5,6,9 From 2000, the pass mark has been set by criterion‐referenced methodology. A further refinement from 2011 was administering the MCQ examination in computer‐adaptive format, where the difficulty of items is adjusted in real time according to a candidate's performance, considered to increase fairness and precision.10 A factor likely to have contributed to the recent lower pass rate in the clinical examination (Box 3) is the removal of nearly all candidates from the competent authority countries. Up till 2009, UK graduates had the highest pass rate in this examination (Supporting information, table 3), and their removal from the pool would inevitably lower the overall rate. However, the decline since 2010 cannot be fully accounted for by this since competent authority candidates comprised less than 10% of the 2000–2009 total. Thus, other factors affecting the most recent cohorts of candidates (eg, the changing mix of parent countries) are likely to have contributed. Many IMGs must often overcome hurdles less likely to be faced by those from competent authority countries. These include adapting to an unfamiliar health system, developing fluency in English, preparing for the examinations while under time pressure from short‐stay visas, and needing to support themselves with sometimes long hours of work outside the health system.11 It is possible, though, that changes in the format or content of the OSCE have also contributed. The differences between the results for women and men in the MCQ should not be overplayed, since the magnitude was small. Others have found little gender effect in postgraduate written examinations in the UK and the US.12,13 However, the outperformance by women in the clinical examination, particularly in the past decade, is more striking. Those findings have been seen elsewhere. Women perform better than men in Step 2 of the United States Medical Licensing Examination.14 Female overseas‐trained doctors were twice as likely as males to pass the UK Federation of Royal Colleges of Physicians’ Practical Assessment of Clinical Examination Skills (PACES) at their first attempt.15 The PACES examination has many similarities to the AMC OSCE, with communication skills important for both. Female superiority in patient–doctor communication has been documented previously,16 and may partly explain the present findings. That performance in the MCQ deteriorated with both age and time since graduation is not entirely surprising: the examination tests knowledge in all domains of medicine, including some of the basic sciences. The longer since these were studied, the more difficult it might be to pass questions based on them, especially for IMGs who had practised as specialists in their original country. More unexpected was the much lower performance in the clinical examination by older candidates. Clinical experience might have been expected to give them an advantage, but this does not appear to have been generally so. We have been unable to find exactly comparable data from medical licensing examinations in other countries. A UK retrospective analysis observed that international graduates aged more than 37 years actually performed better in a postgraduate paediatric examination.17 However, a US analysis noted a negative correlation between age when first certified by the American Board of Internal Medicine and the American Board of Surgery and subsequent success in maintenance of certification examinations.18 Since 1978, these examinations have played an important role in informing the credentialing of generalist IMGs by state medical boards and now the national board. This article has documented substantial changes over the four decades in the demography of candidates, and some factors that were associated with their success in the examinations. The information will be of interest to health planners, but more particularly to those IMGs who have passed through the process and others who are contemplating it. Many rural health services still struggle to meet their workforce needs and rely heavily on doctors who have migrated to practise medicine here.19 Australia continues to owe a debt to its immigrant doctors. Box 1 – Multiple choice question (MCQ) and clinical examinations: numbers of candidates, top ten countries* Country of training 1978–1989 1990–1999 2000–2009 2010–2019 Total MCQ examination India 351 496 2619 2483 5949 Pakistan 32 113 1007 1838 2990 Sri Lanka 159 246 1005 1394 2804 Egypt 179 356 375 1171 2081 Bangladesh 16 99 777 1107 1999 Iran 32 34 664 1197 1927 Philippines 83 182 646 714 1625 China 4 219 641 745 1609 Myanmar 21 66 485 772 1344 Iraq 8 160 420 602 1190 Total all countries 1864 3859 12 722 17 254 35 699 Clinical examination India 190 392 1059 2074 3715 Sri Lanka 101 194 399 960 1654 Pakistan 13 59 342 1168 1582 Bangladesh 7 53 483 831 1374 Iran 8 27 263 688 986 China 0 109 398 475 982 Egypt 78 296 195 375 944 Myanmar 5 45 175 661 886 Philippines 11 104 198 507 820 Iraq 2 85 303 358 748 Total all countries 897 2588 5806 11 203 20 494 * By total number of candidates. Data are listed by the year each candidate first attempted the examination. Many candidates made multiple attempts. International medical graduates trained in the United Kingdom and Ireland were exempted from the Australian Medical Council examinations by most states until 1992. Since 1997, few candidates from the competent authority countries (UK, Ireland, Canada and the United States) were required to take the examinations. Box 2 – Number of candidates and success rate per attempt in the multiple choice question examinations since 1978, and total number of attempts by candidates each year* * Data for 2019 truncated at October. Box 3 – Number of candidates, success rate per attempt and total attempts in the clinical examinations since 1978, and total number of attempts by candidates each year* * Data for 2019 truncated at October. Box 4 – Influence of gender on examination success* Period Gender N Total attempts Total passes Mean (SD) attempts Pass total (%) Pass/attempt (%) MCQ examination 1978–1989 Female 568 1299 412 2.29 ± 1.84 72.5% 31.7% Male 1142 2431 817 2.13 ± 1.68 71.5% 33.6% 1990–1999 Female 1691 3279 1434 1.95 ± 1.57 84.8% 43.7% Male 2164 4275 1729 1.98 ± 1.71 79.9% 40.4% 2000–2009 Female 5438 8666 4813 1.59 ± 1.14 88.5% 55.5% Male 7287 11846 6192 1.63 ± 1.32 85.0% 52.3% 2010–2019 Female 8891 12238 7378 1.38 ± 0.86 83.0% 60.3% Male 8365 12041 6845 1.35 ± 0.85 81.8% 56.8% Clinical examination 1978–1989 Female 257 503 233 1.96 ± 1.58 90.7% 46.3% Male 543 1085 471 2.00 ± 1.47 86.7% 43.4% 1990–1999 Female 1156 2037 1084 1.76 ± 1.07 93.8% 53.2% Male 1432 2917 1243 2.04 ± 1.35 86.8% 42.6% 2000–2009 Female 2636 3662 2428 1.39 ± 0.82 92.1% 66.3% Male 3170 5036 2772 1.59 ± 1.08 87.4% 55.0% 2010–2019 Female 6150 9802 4535 1.59 ± 1.03 73.7% 46.3% Male 5053 9184 3132 1.82 ± 1.30 62.0% 34.1% * MCQ = multiple choice question; SD = standard deviation. * The Australian Medical Examining Council did not list candidates’ gender in a few instances during the first decade. Box 5 – Australian Medical Council multiple choice question (MCQ) examination, 1978–2019: pass rates versus (A) age and (B) recency (interval since graduation) in the year when candidates first attempted the MCQ (all countries combined)* Spearman rank order correlation: (A) r = −0.964, P < 0.001; (B) r = −0.983, P < 0.001. Box 6 – Australian Medical Council clinical examination, 1978–2019: pass rates versus (A) age and (B) recency in year when the examination was first attempted* * Spearman rank order correlation: (A) r = −0.950, P < 0.001; (B) r = −0.950, P < 0.001.

Neville D Yeomans · Jillian R Sewell · Philip Pigou · Stuart Macintyre

Modelling suggests that elimination could have been achieved if Victoria had gone into stage 4 lockdown immediately from 9 July Victoria is the unlucky state in a lucky country. Australian states and territories, other than New South Wales, have achieved elimination of community transmission of the sudden acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2): 28 days of no locally acquired cases where the source is unknown; twice the maximum incubation period. The situation in NSW is mixed. On one hand, NSW had ongoing case notifications of 10–20 per day in the month to mid‐August 2020, arising largely from imported cases from Victoria. On the other hand, on 16 July there had only been three locally acquired cases of SARS‐CoV‐2 infection with no known source in the preceding 28 days, suggesting NSW was on the cusp of elimination.1 If NSW successfully contains the current outbreak, it may resume its prior trajectory towards the elimination of local transmission, leaving Victoria isolated as the only state with community transmission. As of late August, Queensland is also experiencing community transmission — possibly ending its elimination status (28 days of no locally acquired cases where the source is unknown), subject to investigation of the new cases. It seems unlikely that states and territories that have eliminated local transmission will relinquish their status by freely opening borders and engaging with Victoria (and NSW if community transmission remains). Indeed, on 17 August the Queensland Premier stated: “Let me make it very clear, we will always put Queenslanders first and … we do not have any intentions of opening any borders while there is community transmission active in Victoria and in New South Wales”.2 Australia proceeding with two separate systems (six or seven states and territories having eliminated the virus, one or two not) is a significant concern. There are three general strategic policy responses to the challenge of coronavirus disease 2019 (COVID‐19): elimination, suppression, and mitigation (or herd immunity). No response is free of economic, social and health harms; rather, it is about minimising harm. Society has largely rejected a mitigation response because of concerns about the likely high morbidity and mortality arising from such a response. On 24 July, the Australian Health Protection Principal Committee recommended “that the goal for Australia is to have no community transmission of COVID‐19”,3 and on the same day Prime Minister Scott Morrison accepted and affirmed this recommendation, stating “The goal of that is obviously, and has always been no community transmission”.4 Unfortunately, this first clear statement that Australia's goal is to eliminate community transmission was late in coming, as the Victorian outbreak was already in full swing, with case numbers peaking at a 5‐day average of about 500 per day from 29 July to 5 August, resulting in a stage 4 lockdown in metropolitan Melbourne from 6 pm on 2 August. Elimination strategy We know from New Zealand (population, 5.0 million)5 and Taiwan (23.8 million)6 that elimination of community transmission is achievable in island jurisdictions, with NZ having no community transmission for 102 days until 11 August. The advantage of elimination is that despite international border closures or strict quarantine, citizens can go about life with a near‐normal functioning of their society and economy. Elimination presents challenges. First, there is the extra effort to achieve it, and the fact that aiming to achieve elimination does not guarantee success. Second, having achieved elimination, there is the constant risk of the virus re‐entering due to quarantine breaches (eg, the current outbreak in NZ). How frequently a COVID‐19‐free jurisdiction with tight border controls will retain elimination status is unclear, although we know that NZ lasted 102 days with no community transmission and that Western Australia, Northern Territory, South Australia, Australian Capital Territory, Queensland and Tasmania achieved over 100 days without a locally acquired case with no known source (although the status of Queensland is unclear as of early September). Was elimination achievable with a 6‐week stage 3 lockdown as implemented in Victoria from 9 July, or a more stringent lockdown? Lockdowns are effective for COVID‐19 pandemic control.7,8 Our case for an explicit elimination strategy in Victoria at lockdown commencement in early July was that given Victoria was going into a lockdown for 6 weeks, there was probably only a marginal extra cost of “going hard” with a rigorous public health response that increased the probability of achieving elimination. But was elimination achievable within 6 weeks? We examined four policy scenarios using an agent‐based model, a type of microsimulation of individuals. The model accurately reflects the prior experience of both NZ and Australia ( https://github.com/JTHooker/COVIDModel), and here we adapted it to Victoria (including the case counts up to 14 July; see Supporting Information for details). The four policy approaches, all simulated from 9 July 2020, were: Standard: reflecting the first Australian stage 3 lockdown (calibrated to case numbers as described at https://github.com/JTHooker/COVIDModel), with key parameters including 85% of people observing physical distancing; those observing physical distancing doing so 85% of the time; 30% of adult workers being essential workers; 93% of people asked to isolate doing so; 20% uptake of the COVIDSafe app; but no closure of schools and no mask wearing. Standard with masks at 50%: Standard, plus 50% of people wearing masks in crowded indoor environments. Stringent with masks at 50%: Standard with masks at 50%, plus schools closed and essential workers restricted to 20% of workers. Stringent with masks at 90%: Stringent, with mask use increased to 90% (ie, close to stage 4, which was implemented in Melbourne from 6 pm on 2 August after the 5‐day moving average case numbers increased from 300 to 500 in the first 3 weeks of stage 3). Box 1 shows the percentage likelihood of elimination in Victoria, defined as the date of clearance of infection by the last case, and the date of last acquisition of infection. The model is omniscient about infectious status; in the real world, based on a definition of 28 days of no cases, elimination would occur about 2 weeks after the clearance dates shown in Box 1, A. Under the “standard” policy approach (ie, equivalent to stage 3 without masks), there was no chance that all infected people would have cleared their SARS‐CoV‐2 infection by 19 August (6 weeks after lockdown commenced; Box 1, A). The probabilities for the other three policy approaches achieving elimination 6 weeks after implementation (Box 1, A) were 0% for “standard with masks at 50%”; about 4% for “stringent with masks at 50%”; and 30% for “stringent with masks at 90%”. The probabilities of the last actual infection occurring by 19 August were more encouraging at 0%, 1%, 45% and 90%, respectively (Box 1, B). Of particular note, given that the stage 3 lockdown imposed on 9 July failed because caseloads increased to an average of 500 per day, in our simulations 48% of the 1000 iterations of the “standard” scenario (stage 3, no masks) and 22% of the 1000 iterations for “standard with masks at 50%” had peaks in the first 3 weeks in excess of 400 per day. This is consistent with what eventuated, and further speaks (in hindsight) to the desirability of entering a stage 4 lockdown on 9 July; the “stringent with masks at 90%” scenario had no instances of peak cases greater than 400 per day in the first 3 weeks. Undertaking simulation modelling of SARS‐CoV‐2 policy options is challenging and the uncertainties are still considerable even when using the best estimates available. Nevertheless, our results lend weight to the proposition that elimination was achievable if Victoria had gone into stage 4 lockdown with mandatory wearing of masks immediately from 9 July. A ten‐point plan to maximise the chance of elimination in Victoria Box 2 lists enhancements to the stay‐at‐home orders of the 9 July lockdown. The first and critical point was leadership. As above, we did get a clear statement of an elimination goal from the Chief Health Officers (who comprise the Australian Health Protection Principal Committee membership) and Prime Minister Scott Morrison on 24 July, but with the benefit of hindsight it was perhaps too late. Target‐setting is still not occurring (eg, a target number of cases per day could be set for when we step out of stage 4 under both elimination and suppression strategy options). Moreover, an expert advisory group on elimination was not convened, limiting the capacity for an optimal evidence‐informed policy response. Nevertheless, since the 9 July lockdown, progress with other aspects of the ten‐point plan has been made with the closure of schools, mandatory mask wearing, and commitments to improve contact tracing capacity. Conclusion We argued in the preprint version of this article on 17 July that Melbourne and Victoria should not waste the opportunity that the (then) 6‐week lockdown presented and go hard and early. By learning from the lessons on social and preventive measures to lower SARS‐CoV‐2 transmissibility,7,8,12,14 and specifically the lessons from NZ,3 Taiwan and the six Australian jurisdictions that have achieved elimination, Victoria could have increased its chances of also eliminating community transmission. Our work and that of others who have independently considered the alternatives consistently demonstrates that elimination was possible, and if achieved would have been optimal for health and for the economy in the long term.15,16,17 In this article, we modelled the situation as at mid‐July — we are now updating modelling under the current situation. Authors’ note: This Perspective was submitted to the MJA on 16 July 2020 and published as a preprint on 17 July.9 The revised version, submitted on 23 August, retains the simulation modelling of the original but the uncertainty of inputs was updated to include uncertainty other than stochastic uncertainty. Our aim was rapid modelling to estimate the probability of virus elimination during the planned 6‐week stage 3 lockdown that Victoria had just commenced. The revised version was also published as a preprint on mja.com.au on 4 September, following full peer review and prior to typesetting, pagination and proofreading. Box 1 – Percentage likelihood of elimination of community transmission of SARS‐CoV‐2 infection in Victoria, by date of clearance of last active infection (A) and date of acquisition of last infection (B)* * Across 1000 Monte Carlo simulations in an agent‐based SEIR (susceptible, exposed, infectious, recovered) model. The vertical dashed line is the date 6 weeks after implementation of the lockdown policies. Compared with modelling published in the preprint version of this article,9 the only change here is the inclusion of additional parameter uncertainty in addition to stochastic uncertainty (see Supporting Information), resulting in increased sloping in the curves due to a wider range of potential parameter values (ie, the time distribution to elimination is wider). Box 2 – A ten‐point plan to maximise the chance of successful elimination of community transmission of SARS‐CoV‐2 in Victoria, based on the planned 6‐week lockdown from 9 July 2020 (as published on 17 July 2020)9 Strong and decisive leadership with strategic clarity. An explicit goal of elimination should be articulated, learning from the New Zealand experience (Prime Minister Jacinda Ardern, government ministers and senior officials).10 A clear set of targets for loosening of policies needs to be articulated, so citizens know what is likely to happen and when. Convene an advisory group of experts in the elimination strategy and SARS‐CoV‐2 public health response, reporting weekly to the Victorian Chief Health Officer, with the agenda, papers and minutes made publicly available. Close all schools. Although children do not usually suffer severe illness from SARS‐CoV‐2 infection, the virus still transmits between children and staff in schools.11 Accordingly, schools need to close until such time as the daily rate of SARS‐CoV‐2 infection without a known source falls beneath a target set by the Chief Health Officer. Tighten the definition of essential shops to remain open. Supermarkets and chemists need to remain open. However, department stores and hardware stores should be closed. A staged re‐opening based on set target levels of daily numbers of SARS‐CoV‐2 infection without a known source should then be implemented, so long as mask wearing by both staff and patrons is mandatory, along with hand sanitiser use on entry and exit from stores. Require mask wearing by Melbourne residents in indoor environments where 1.5 m physical distancing cannot be ensured, such as supermarkets and (especially) public transport. While no panacea, the wearing of masks reduces the chance of infected people spreading the virus.12 Tighten the definition of essential workers and work. There is currently a loose definition of who is an essential worker and what is essential work. This needs urgent tightening; for example, as per the NZ definitions used in their level 4 lockdown.13 Require mask wearing by essential workers whenever they are in close contact with people other than those in their immediate household. Ensure financial and other supports to businesses, community and other groups most affected by more stringent stay‐at-home and lockdown requirements, and provide enhancements, targeted where warranted, to programs such as JobKeeper and JobSeeker. Further strengthen contact tracing to ensure the majority of notifications (and their close contacts) are interviewed within 24 hours of the index case notification and placed in isolation if necessary. The use of smart phone and digital adjuncts needs to be improved, be that for initial contact tracing (eg, the COVIDSafe app, or a South Korean‐style use of telecommunications data) or monitoring of adequacy of isolation (eg, text message follow‐up, GPS monitoring, or electronic bracelets). Extend suspension of international arrivals into Victorian quarantine and divert resources. To allow a stronger focus on elimination within Victoria, extend the suspension of international arrivals to Victoria. Quarantine capacity can be redeployed for isolation of Melbourne residents infected with SARS‐CoV‐2 (and potentially high risk close contacts) if they do not have satisfactory home environments for self‐isolation.

Tony Blakely · Jason Thompson · Natalie Carvalho · Laxman Bablani · Nick Wilson · Mark Stevenson

Serological assays for SARS‐CoV‐2 present challenges and opportunities Timely, scalable and accurate diagnostic testing is crucial in the prevention and control of the coronavirus disease 2019 (COVID‐19) pandemic.1 With limited treatment options and no available vaccine, the accurate and timely identification of infectious patients with COVID‐19 is instrumental to the public health outbreak response. Isolation of patients with COVID‐19, contact tracing and quarantine measures, in addition to physical distancing within the community, have proven effective in reducing case numbers.2 Due to the high sensitivity and specificity in symptomatic individuals, detection of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection by reverse transcriptase polymerase chain reaction (RT‐PCR) is the gold standard method for confirming cases of COVID‐19.3 In contrast, serological assays have lower utility in the initial investigation of suspected cases, but are essential in the development and evaluation of therapeutic agents and to inform modelling and public health policy as this pandemic progresses. As part of initial laboratory responses, Chinese investigators released the viral whole genome sequence in early January 2020, which enabled the rapid development of RT‐PCR workflows for the detection of SARS‐CoV‐2.4 However, the unprecedented scale of RT‐PCR diagnostic testing has placed extraordinary demands on health care and laboratory systems, with both challenges relating to supply chains of reagents and to the workforce resource required to support population‐level testing. Since the start of the pandemic, a range of commercially available diagnostic tests has been released, including RT‐PCR assays, point‐of‐care and laboratory‐based serological tests. These tests vary both in analytical performance and in their particular utility in the overall public health response to COVID‐19. Performance aspects of serological tests Following SARS‐CoV‐2 infection, specific antibodies to different components of this virus are generated. Depending on the antigen target used by the assay, detection of these antibodies (IgM, IgA, IgG or total antibody) may indicate exposure (non‐neutralising antibodies) or potential immunity (neutralising antibodies). To date, a range of serological tests for COVID‐19 have been developed, each with particular test characteristics (Box 1). Broadly, these serological tests can be divided into tests that (i) can be performed at the point‐of‐care; (ii) can be performed in routine diagnostic laboratories, and (iii) can only be performed in specialised reference laboratories (Box 1). The majority of point‐of‐care and laboratory‐based assays have incorporated either the nucleocapsid antigen (N) or part of the spike protein (S), often the S1 region or the receptor binding domain (RBD). The RBD has been shown to correlate well with the production of neutralising antibodies,5 while some studies have shown N to be immunodominant, producing an earlier or stronger immune response.6 Most patients with COVID‐19 seroconvert by day 10–14 (~ 80%) following the onset of symptoms, with almost 100% seroconversion by day 20.7 However, comparisons across published studies are challenging due to the different antigens used in assays, differences in the complexity of patient populations, variations in the RT‐PCR assays used as the gold standard for determining the sensitivity of serological assays, and often limited data on the timing of sample collection post‐COVID‐19 symptom onset. Further, it is not clear whether the type and amount of antibody correlate with severity of disease or, more importantly, with immune protection from re‐infection. As noted by the World Health Organization, the Australian Public Health Laboratory Network (PHLN) and the Royal College of Pathologists of Australasia (RCPA), a negative result using a serological test does not rule out SARS‐CoV‐2 infection, particularly in individuals with strong epidemiological risk factors, and both the PHLN and the RCPA note that there is no role for serological point‐of‐care tests (PoCT) in the acute diagnosis of COVID‐19.8,9 Point‐of‐care testing As some of the first COVID‐19 serological assays available, significant publicity accompanied the release of PoCT. PoCT involve detection of anti‐SARS‐CoV‐2 antibodies through binding to immobilised antigens, generally bound to colloidal gold on a test strip (Box 2). The relatively cheap and simple nature of lateral flow assays means that production is suited to scale‐up for increased testing capacity. Post‐market validation studies have demonstrated variable performance characteristics, often inferior to that reported by manufacturers, with sensitivities for IgG reported in the range of 53–100% for samples collected more than 14 days after symptom onset, and specificities of 91.7–100%.10,11 Careful test selection and consideration of the clinical utility before application are therefore critical. The National Pathology Accreditation Advisory Council has existing guidelines on the use of PoCT in Australia.12 These guidelines cover issues such as clinical supervision for performing PoCT, ensuring test quality, staff training and competency for performing PoCT, and appropriate reporting of test results. More recently, this advice has been extended to serological PoCT for COVID‐19, with an emphasis on a robust quality framework to support the implementation and deployment of such tests. Of note, in Australia, the supply of self‐testing kits (eg, testing at home) for many infectious diseases, including COVID‐19, is prohibited under another Therapeutic Goods Administration regulation, the Therapeutic Goods (Excluded Purposes) Specification 2010.13 Laboratory‐based assays A wide variety of laboratory‐based serological assays are now available, most commonly enzyme immunosorbent assays (ELISA) or chemiluminescent immunoassay (CLIA/CMIA) format. Assays may be semi‐automatic or completely automated, lending themselves to large scale testing and reporting. In general, performance characteristics have been more consistent and closer to that reported by manufacturer's compared with PoCT, with IgG sensitivities in the range of 80–100% for samples collected more than 14 days after symptom onset, and specificity commonly falling between 95% and 100%.10,11,14 Use of serological assays Given the time lag from symptom onset to detectable antibody, serological PoCT have no role in the detection of acute COVID‐19. However, there are some settings where serological assays, including PoCT, may have potential utility, including defining antibody prevalence in key populations such as frontline workers and determining the extent of COVID‐19 transmission within the community. For other applications, such as identifying individuals for further evaluation of therapeutic immunoglobulin donation and vaccine development and evaluation, assays that assess neutralising antibody response are likely to be required. Regardless of the type of serological assay used, in order to appropriately deploy serological testing, it is critical to understand the limitations of test performance in the epidemiological context in which tests are used. This is particularly important in a setting such as Australia, where, based on the number of reported cases of COVID‐19 (24 236 cases as of 20 August 2020), there is an estimated COVID‐19 period prevalence of 0.095% (January to August 2020). As such, even with serological tests that are highly sensitive and specific, the majority of positive tests are likely to represent false positive results. When considering the use of serology to inform policies relating to relaxing of physical distancing interventions, the specificity of the assay becomes critical. If most individuals considered immune actually represent false positive results, then the threshold to maintain immunity (if this indeed correlates with antibody detection) within the community will not be achieved. Consideration should therefore be given for confirmation of initial positive results by either retesting on an assay with an alternative target, or retesting with serological gold standard assays, such as microneutralisation or western blot assays.15 Application of serological assays and future research needs Understanding local transmission dynamics and/or exposure risk through serological surveys can inform local health policy at an institutional, state or national level. For example, a recent large serological survey in Spain, including more than 50 000 residents, used both PoCT and a laboratory‐based CLIA to estimate a seroprevalence across the country of 5.0%, following an initial COVID‐19 outbreak in February to April.16 It was estimated that approximately a third of cases were asymptomatic, while health care workers had a higher seropositivity than the community (10.2% v 5.0%). This is in contrast to health care workers in Belgium, where direct contact with patients with COVID‐19 did not increase the odds of being seropositive.17 The degree and duration of immunity following SARS‐CoV‐2 infection is unknown, but if in keeping with other coronaviruses (approximately 40 weeks), immunity is unlikely to be lifelong and may be shorter lived in milder infections.7 Duration of immunity is a critical area for future research, as it is the key component in models estimating the frequency of SARS‐CoV‐2 infection incidence in the coming years (eg, second or third waves, or annual seasonal COVID‐19 activity similar to influenza), and to determine the utility of policies such as “immunity passports”.18 Conclusion The unprecedented demands on laboratories to rapidly upscale testing for COVID‐19 has necessarily led to fast‐tracking of normally stringent regulatory requirements for test approval, both globally and in Australia. Following the recent publication of peer‐reviewed high quality validation data, serological testing is now available in many Australian laboratories. Serological testing will complement the current clinical utility of RT‐PCR for SARS‐CoV‐2 infection diagnosis, highlight local transmission dynamics, and further our understanding of what the future brings for the COVID‐19 pandemic. Box 1 – Main serological assays used to date for the detection of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) Serological assay Detection method Advantages Disadvantages Implications Neutralisation Determines ability of test sera to inhibit live virus replication Gold standard Highly specific Requires PC3 facilities Technically demanding Slow turnaround time Low throughput Only undertaken in specialist laboratories Gold standard for initial validation of other assays and challenging cases Not suited to routine testing Indirect fluorescent antibody (IFA) Whole virus inactivated and fixed to a slide Addition of test sera with fluorescent detection of antibody binding Can be undertaken at PC2 facilities once slides prepared Less technically demanding than neutralisation assays Preparation of slides requires PC3 facilities Less specific than neutralisation Technically demanding Subjective end point Low throughput Not available in routine laboratories Not suited to large‐scale testing Enzyme immunoassay (EIA) Recombinant antigen fixed to solid surface (often 96 well plate), test sera applied and antigen–antibody binding detected by enzyme‐mediated colour change Good sensitivity Less technically demanding than IFA or neutralisation Semi‐automated High throughput Objective end point with machine‐based optical density reading Less specific than neutralisation Initial expertise and time required to determine, test and manufacture suitable recombinant antigen Generally relies on commercial companies to manufacture and distribute test kits Suitable for routine testing Good for screening Lateral flow EIA A particular type of EIA Recombinant antigen present on immunochromatographic paper, test sera applied to test pad, antigen‐antibody binding detected visually by colour change on a membrane Variable sensitivity Least technically demanding Fast turnaround time for individual tests Test on demand May be less sensitive and specific than laboratory‐based assays Limited scalability Subjective end point Data capture less robust Suited to point‐of‐care testing Can be undertaken by non‐laboratory staff Systems for data capture of results need to be implemented PC2 = physical containment level 2; PC3 = physical containment level 3. Box 2 – Schematic of a lateral flow immunoassay for detection of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) IgM and IgG antibodies* *The sample is added to the sample pad, and then travels by capillary motion to the conjugation pad. Anti‐SARS‐CoV‐2 IgM and/or IgG antibodies in the patient sample then bind to the specific SARS‐CoV‐2 antigen. This antigen is bound to colloidal gold, which acts as a colorimetric indicator. The bound antigen‐antibody‐gold complex then travels to the nitrocellulose membrane and bind to specific anti‐human IgM or IgG antibodies, with a resultant colorimetric change. To monitor test validity, excess conjugated colloidal gold binds to antibody on the control line, which allows assessment of whether the fluid has successfully migrated across the test strip. Source: Adapted from Li Z, Yi Y, Luo X, et al. Development and clinical application of a rapid IgM–IgG combined antibody test for SARS‐CoV‐2 infection diagnosis. J Med Virol 2020; https://doi.org/10.1002/jmv.25727. [Epub ahead of print]

Katherine Bond · Eloise Williams · Benjamin P Howden · Deborah A Williamson

We need to burn, but it is not a cure-all and the side effects can be serious

Fay H Johnston

Guidelines need to reflect the mounting evidence for airborne transmission of SARS-CoV-2

C Raina MacIntyre · Michelle Ananda‐Rajah · Mark Nicholls · Ashley L Quigley

Some form of diagnostic error occurs in up to one in seven clinical encounters, and most are preventable Diagnosis consists of eliciting information from history and examination, formulating a differential diagnosis, and selecting a final diagnosis based on the predictive value of specific clinical features and laboratory investigations. A timely and accurate diagnosis is every patient’s expectation. Prevalence, impacts and causes of diagnostic error Diagnostic error comprising missed, wrong or delayed diagnoses (Box 1) affects between 8% and 15% of all hospital admissions in the United States,1,2 with similar rates among patients with common diseases attending outpatient clinics.1 As many as 1.1% of adult hospital admissions will involve diagnostic error that causes harm to patients.3 Nearly a third of all preventable deaths in acute hospitals in the United Kingdom are attributed to diagnostic error.4 In Australia, an estimated 140 000 cases of diagnostic error occur each year, with 21 000 cases of serious harm and 2000–4000 deaths.5 Almost one in two malpractice claims against general practitioners involves diagnostic error.6 More than 80% of diagnostic errors are deemed preventable.7 Cognitive factors in clinician decision making are primary or contributory causes of more than 75% of diagnostic errors, with system errors (eg, missed communication or follow‐up of a laboratory test result) being less frequent.1 Failure to formulate an adequate differential diagnosis8 and overconfidence in incorrect diagnoses9 are major contributors. Clinical culture discourages disclosure of diagnostic errors and they are largely neglected within professional training curricula10 and organisational quality and safety programs.11 Identifying the cognitive causes of diagnostic error which can inform preventive strategies requires an understanding of clinical reasoning (Box 2).12,13,14,15 Intuitive thinking is the preferred reasoning mode, using heuristics (ie, mental shortcuts or rules of thumb) to accelerate the process by limiting the load on short term working memory to no more than seven ideas at a time. While efficient and accurate in many situations, heuristics can be misapplied due to cognitive bias (Box 3). Emotions, fatigue, distractions, peer opinions, and cultural norms can also further impair cognitive fidelity. Strategies to prevent diagnostic error Various preventive strategies have been proposed, the choice of which may vary according to clinician experience, types of clinical scenarios encountered, and the clinical environment. Optimise the clinical interview Taking a good history, including collateral information from relatives and other health professionals, and performing an adequate physical examination are fundamental. In combination, these will yield the correct diagnosis in more than 80% of cases,16 while failure to enact them contributes to 40% of missed diagnoses.5,17 Target education to specific scenarios commonly associated with diagnostic error Knowledge deficits are infrequent (< 5%) causes of diagnostic error among practising clinicians.18 It is not that clinicians are unfamiliar with a diagnosis, they simply fail to consider it when appropriate. Educational interventions to increase overall knowledge do not necessarily improve diagnostic performance.19 More useful is tuition focused on scenarios involving frequently missed or wrongly diagnosed conditions, including vascular events, infections, cancer, and neurological disorders (eg, multiple sclerosis).20 Targeted training, such as how to recognise subarachnoid haemorrhage,21 has prevented some condition‐specific diagnostic errors. Verify past diagnostic labels Between 11% and 40% of listed diagnoses in older patients with Parkinson disease, dementia, heart failure and chronic obstructive pulmonary disease do not satisfy accepted diagnostic criteria.22 Verification of past diagnoses, especially those based solely on subjective judgements and lacking specific diagnostic tests, is needed when clinical trajectories are atypical or appropriate therapies yield no response. Implement strategies for reducing cognitive errors Recent reviews describe various strategies for reducing cognitive errors23,24,25 with varying levels of evidence of efficacy. Lectures, seminars, group discussions, and interactive videos can all improve knowledge of cognitive biases and debiasing strategies, broaden differential diagnosis, and enhance reasoning processes. However, evidence of improved diagnostic accuracy is lacking, suggesting that, despite such educational interventions, clinicians may still not reliably identify when biases are influencing diagnostic decisions. Diagnostic checklists can take various forms: a generic checklist prompting clinicians to optimise their cognitive approach; a differential diagnosis checklist prompting clinicians to consider the correct diagnosis as a possibility; and Only the differential diagnosis checklists show improvements in the completeness of differential diagnosis in simulated or actual cases.27 In one study, a differential diagnosis checklist led to fewer errors overall;28 another similar tool combined with a debiasing checklist increased diagnostic accuracy compared with intuitive reasoning.29 Cognitive forcing strategies, defined loosely as any form of disciplined thinking, require clinicians to consciously slow their thinking and systematically evaluate all potential alternatives and mimics before finalising a diagnosis.30 In some studies,31 but not others,32 this approach improves diagnostic accuracy compared with first impression diagnoses or reasoning without any specific instruction. In one study, instructing participants to reconsider their diagnosis after removing a distracting detail from the case outline greatly improved diagnostic accuracy.33 Similar to cognitive forcing strategies, analytical reasoning involves instructing participants to use a guided, analytical approach (System 2) rather than rapid intuition (System 1). Diagnostic accuracy improves,34,35 more so when dealing with complex cases,36 and in a randomised trial,35 this approach overcame deliberate attempts within test cases to induce cognitive biases. Deliberate practice actively engages clinicians in solving diagnostic conundrums (real or vignette) and verbalising their reasoning (“thinking out loud”) as the case unfolds.37 By comparing participants’ reasoning with those of an expert who has worked through the same case, cognitive errors and knowledge deficits can be identified. Simply seeing more cases, without any attempt at calibration, does not guarantee diagnostic expertise,12 although whether deliberate practice improves diagnostic accuracy remains uncertain. Metacognition involves clinicians thinking about their thinking and reflecting on past diagnoses and appropriate use of heuristics. In some studies, cued and modelled reflection improves diagnostic accuracy compared with a more generic, free‐floating reflection38 or leaving participants to reflect in whatever way they choose.39 Seeking second opinions on one’s diagnoses from one’s clinical peers can increase diagnostic accuracy by as much as a third.40 Seeking the diagnostic opinion of patients, families and other members of the health care team, even if expressed in general terms, can also help detect and prevent errors.41 Following up patients over time, asking patients and colleagues to report errors, and implementing protocols for identifying errors (eg, trigger tools within electronic medical records for identifying unexpected adverse events or unplanned readmissions, or systematic identification of errors within mortality and morbidity meetings) all provide information on final outcomes, thus checking the accuracy of initial diagnoses. Such strategies, combined with reflection on identified errors (“cognitive autopsies”), improve diagnostic performance.42,43 Such feedback is important as clinicians’ self‐assessment of their diagnostic accuracy is unreliable and their level of diagnostic confidence can be insensitive to both accuracy and case difficulty.9 Feedback also tempers over‐reliance on the results of diagnostic tests that are at odds with the overall clinical picture and likelihood of a specific disease.44 High risk clinical environments, in which diagnostic error is more likely to occur, require clinicians to be more vigilant about their reasoning in such circumstances.45 Rushed clinical handovers, heavy caseloads, distractions and interruptions, caring for critically ill or complex multimorbid patients, interactions with uncooperative or non‐communicative patients, and clinician fatigue or personal stressors are some examples.46 Computer‐assisted diagnosis in various forms can improve diagnostic performance. Computed decision support systems that generate differential diagnoses using inputted clinical data yield small improvements in diagnostic accuracy when clinicians revisit their diagnoses following a differential diagnosis generator consultation.47 A digital image library of skin eruptions increased diagnostic accuracy of dermatology residents by 19% in a randomised trial.48 An interactive computed decision support system achieved up to 75% reduction in diagnostic errors relating to vignettes of neurological disorders.49 A web‐based system that facilitated internet crowdsourcing of multiple opinions improved diagnostic accuracy among junior physicians.50 Acknowledging, explaining and sharing diagnostic uncertainty with patients helps to protect clinicians from rushing to ill‐considered diagnoses. Up to 40% of first‐contact primary care consultations involving a diagnostic question do not yield a definite answer.51 In such situations, clinicians may feel pressured to prematurely commit to a diagnosis in order to activate management plans and demonstrate competence. In contrast, patients welcome an open discussion of possible differential diagnoses and a plan and timeline for ongoing review.52 Injudicious ordering of multiple diagnostic tests to reduce uncertainty does not reduce patient anxiety and may cause harm from false positive results.53 Need for more research into diagnostic reasoning While we have sought to shed light on the causes and prevention of diagnostic error, we concede current research has several limitations: enrolment of predominantly novice rather than experienced clinicians; non‐randomised or before and after designs; relatively small samples; short term follow‐up; variable methodological rigour; missing data; and multiple, often unvalidated, measures of error and reasoning style. Primary outcome measures are restricted to improvements in knowledge or skills in vignette studies, although these are deemed reliable proxy measures of real‐world decision making.54 Strengthening the evidence base for error mitigation is one objective of the recently established Australian and New Zealand Affiliate of the US Society to Improve Diagnosis in Medicine. This group aims to improve clinical diagnosis in this country with planned initiatives in practice improvement, research, education, and patient engagement (Supporting information). Conclusion Despite limitations in current research, the scale and harm of diagnostic error obliges clinicians to consider adopting preventive strategies that have reasonable face validity, are easily implementable in workplaces, and target individual decision making. Box 1 – Typology of diagnostic error Diagnostic errors can be of three types: missed diagnosis — the correct diagnosis was never considered; wrong diagnosis — the provisional or working diagnosis is incorrect; delayed diagnosis — sufficient information was available to enable the correct diagnosis, which was eventually made, to be made at an earlier time. The term “overdiagnosis” refers to a separate concept where a diagnosis is correct (eg, a patient has prostate cancer) but the diagnosed condition is not causing symptoms, is of low grade of malignancy, and will not prematurely kill the patient before they die of other diseases. In this scenario, the very act of diagnosing this disease may actually cause harm by invoking needless clinical intervention. It is different to when a diagnosis is actually incorrect, which is the focus of this article. Box 2 – Theories of diagnostic reasoning Proponents of organised (or structured) knowledge emphasise content specificity whereby reasoning proficiency varies from case to case, depending on levels of knowledge of particular clinical scenarios. Clinicians construct multiple illness scripts as mental representations of diagnostic, therapeutic and prognostic attributes of specific conditions.12 These scripts store and, with increasing experience, elaborate knowledge in a readily accessible format for application to new clinical scenarios. This emerging expertise is further developed by deliberate practice under supervision coupled with regular feedback.13 Proponents of cognitive processing (or dual processing theory) describe a rapid, intuitive form of pattern recognition (fast [System 1]) and a more deliberate, analytical approach (slow [System 2]).14 When considering different or even single cases, clinicians oscillate between the two systems according to their level of experience and store of memorised patterns. Expert clinicians spend more time in System 1, novice clinicians more in System 2. Central to System 2 is the hypothetico‐deductive model whereby the initial problem representation, gained from history and containing key clinical features (or cues), triggers a number of possible diagnostic hypotheses. These are ranked in decreasing likelihood and, based on further information from hypothesis‐driven, focused physical examination and selected laboratory investigations, gradually eliminated in arriving at a provisional diagnosis. The two schools of thought are not mutually exclusive and are in fact interdependent. Clearly, more hypotheses may be generated, or more patterns recognised, if the clinician can draw on a larger store of illness scripts that share cues with the problem at hand. Similarly, knowledge becomes more organised more quickly if clinicians consistently and systematically apply analytical thinking to obscure or atypical cases. Approaches to improving diagnostic reasoning vary in their emphasis on expanding organised knowledge, mitigating cognitive bias, or optimising system of care factors according to how much each, in different circumstances, is considered the prime determinant of diagnostic error.15 Box 3 – Common cognitive biases in diagnostic reasoning Bias Definition Example Premature closure Narrow rapid focus on single or a few clinical features in the clinical presentation to support a diagnostic hypothesis without considering other alternatives Patient with rheumatoid arthritis who is receiving immunosuppressive medication presents with shortness of breath, inspiratory crackles on chest auscultation and diffuse fine infiltrates on chest x‐ray. Congestive heart failure is quickly accepted as the diagnosis but subsequent bronchoscopy reveals Pneumocystis pneumonia Anchoring bias Tendency for clinicians to cling to their initial diagnostic hypotheses even as contradictory evidence accumulates Patient with end‐stage renal disease presents with altered mental status and myoclonus of the left arm, which is attributed to uraemia (the anchor). Failure of this syndrome to improve with dialysis (contradictory evidence) is underweighted until clinicians finally accept the eventual diagnosis of status epilepticus Confirmation bias Tendency to selectively search for features that support the initial or favoured diagnostic hypotheses rather than take deliberate note of features that challenge these hypotheses Patient with past history of coeliac disease presents with symptomatic anaemia and low reticulocyte count, which is diagnosed as iron deficiency anaemia. Iron studies showing borderline low serum ferritin are interpreted as confirmatory evidence, while the finding of a widened mediastinum on chest x‐ray is ignored. The patient is later diagnosed as having a thymoma associated with aplastic anaemia Availability bias Tendency to overestimate the probability of a diagnosis based on how easily it is recalled, which is often skewed by recent and memorable, or emotionally laden cases A clinician who has recently seen a patient with myosarcoma who presented with left calf pain then begins to evaluate all subsequent similar presentations for the possibility of the same diagnosis Representativeness bias/base rate neglect Tendency to greatly overestimate the likelihood of a rare diagnosis on the basis of some prototypical features of that disease Patient presenting with pulsatile headache, palpitations, diaphoresis and elevated blood pressure is diagnosed as having a pheochromocytoma (rare disease) whereas anxiety syndrome complicated by severe migraine (common disease) is later verified Framing bias Tendency for a presentation to be framed in a certain way according to past diagnostic labels (diagnostic momentum) or clinical setting (eg, medical v a surgical ward) Patient with long‐standing anorexia nervosa and post‐traumatic stress disorder presents with weight loss, abdominal pain and diarrhoea. Her past history causes the clinician to frame the problem as one related to her mental health, leading to a diagnosis of irritable colon and laxative misuse associated with restrictive feeding. The presence of intermittent rectal bleeding and an elevated erythrocyte sedimentation rate (ESR) are underemphasised. The patient is eventually diagnosed as having Crohn’s disease

Ian A Scott · Carmel Crock

Telehealth offers an opportunity to address limited access to early medical abortion during COVID‐19 and beyond Access to early medical abortion (EMA), using mifepristone followed by misoprostol to end an early pregnancy, remains a challenge in Australia, especially for women from vulnerable groups and those living in rural and regional areas.1 Low numbers of general practitioner providers, lack of peer networks to support the establishment and ongoing provision of EMA services, and stigma are real barriers as is a broader lack of knowledge regarding medical abortion among health professionals.2,3 Many women are also unaware of the availability of EMA and the current gestational limit of 63 days.4 They also face difficulties navigating the health system to find an EMA provider, particularly when they encounter conscientious objections.4,5 Women can also face other barriers such as needing to travel to access services, take time off work or find childcare, and many need to source financial support to meet the costs.5 The current coronavirus disease 2019 (COVID‐19) pandemic has further highlighted existing barriers to accessing EMA services in Australia. During the pandemic, there has been an increase in the demand for abortion because of a rise in unplanned pregnancies and domestic violence.6 Financial insecurity and delays in accessing abortion services, due to travel restrictions or other pandemic‐related stressors, means that women are often presenting for an abortion at a later gestational age.6 In addition, flight restrictions may have curtailed the ability of clinicians to travel to rural areas to provide surgical abortion services. Delivering EMA through telehealth has been shown to be safe, effective and acceptable to women, both internationally and in Australia.7,8,9 Originally championed by Women on Web (www.womenonweb.org), telehealth delivery of EMA was used to provide abortions clandestinely in countries where they were illegal, such as in Ireland prior to decriminalisation.10 It has now, however, been implemented in many countries worldwide, irrespective of whether restrictive or non‐restrictive abortion laws exist, to provide abortion care to women and improve access to women geographically isolated from EMA services.9 Using telehealth to deliver EMA offers an opportunity to address many of the barriers to EMA provision in Australia. It removes the necessity for proximity between the provider and patient, an issue of particular importance for women living in rural and regional areas where there are fewer abortion providers.5,7 The need to travel to appointments far from home, especially when more than one appointment might be required, can result in women moving past the 9‐week gestational limit and preclude them from being able to undergo an EMA.5,7 Not only does the telehealth delivery of EMA reduce the need for patients to travel but it also increases the capacity of existing providers to deliver services to women from a larger geographical area.5,8 The availability of Medicare Benefits Schedule (MBS) telehealth item numbers, introduced as part of the government's response to the pandemic, has meant that, for the first time, telehealth EMA can be delivered through Medicare to eligible patients.11 With these item numbers in place, all EMA providers are able to use telehealth to deliver this service at a potentially reduced cost to women. Before COVID‐19, telehealth item numbers had very restrictive criteria and could only be billed if the patient lived in a very rural area (Modified Monash Model 6 or 7 location), had an existing clinical relationship with a GP telehealth provider (defined as three face‐to‐face consultations in the previous 12 months) and lived at least 15 km by road from the GP.12 These restrictions unfairly excluded many women in metropolitan or regional areas, particularly young women (who comprise the largest demographic using abortion services), as this demographic does not necessarily attend GPs on a regular basis. It is imperative therefore that MBS‐funded telehealth remains implementable by all GPs so that women are not disadvantaged, and that telehealth abortion can remain accessible via Medicare. Recent restrictions to the temporary MBS item numbers for telehealth GP consultations, which came into effect on 20 July 2020 — namely restricting eligibility to only those who have visited the GP or practice in the previous 12 months or those who have been referred by a specialist except for where there is a current lockdown in place13 — will greatly reduce women's access to EMA. Placing restrictions on the eligibility criteria for MBS‐subsidised telehealth services severely affects women's access to GPs who can provide EMA, and discriminates against women who have not recently engaged with a GP due to various forms of disadvantage, such as family violence and unemployment. Exemptions to the restrictions have already been identified for people who are homeless and for children aged less than 12 months. Therefore, a further exemption should also be issued so that registered prescribers of medical abortion are able to use MBS telehealth item numbers for the benefit of Australian women. In addition, other measures are required to optimise the ability of telehealth to improve access to EMA for all Australian women. Firstly, a national hotline or online platform, similar to the 1800 My Options service (www.1800myoptions.org.au) in Victoria, which directs women to local abortion service providers, is required to assist women to identify an appropriate provider. Secondly, as outlined in a consensus statement on EMA developed by a coalition of key stakeholders (ie, the National Health and Medical Research Council's Centre of Research Excellence in Sexual and Reproductive Health for Women in Primary Care [SPHERE]) and clinician experts,14 changes are required to current Therapeutic Goods Administration (TGA) and Pharmaceutical Benefits Scheme (PBS) provisions restricting the prescription of MS‐2 Step (mifepristone and misoprostol) to up to 63 days’ gestation.15 These criteria are outdated and discordant with current evidence demonstrating that EMA up to 70 days’ gestation is comparable in safety and efficacy to 63 days’ gestation or less.16 Guidance from the United States, Canada and the United Kingdom all concur.17,18,19 Increasing gestational limits for prescribing EMA will not only align Australia with international guidance but will also provide a greater window of opportunity for women to access this service. However, this change requires an application to be made to the TGA, and if TGA approval of the extended indication is successful, a subsequent application to the Pharmaceutical Benefits Advisory Committee for subsidy of the extended indication would be required. This is a costly and time‐consuming exercise. Thirdly, modifications are required to EMA protocols, particularly during the COVID‐19 pandemic. Internationally, “no‐touch/no‐test” protocols have been devised and endorsed to minimise the risk of COVID‐19 transmission between patients and providers and circumvent delays created by closed health services (ie, sonography).19,20 In the Australian context, the Royal Australian and New Zealand College of Obstetricians and Gynaecologists has already advised that a clinician may appropriately decide not to administer anti‐D IgG before 10 weeks for the medical management of abortion, particularly when an additional visit may increase exposure of women and staff.21 The SPHERE coalition has additionally recommended that, during the COVID‐19 pandemic, while ultrasound is highly desirable for all women having a telehealth EMA, in situations where obtaining an ultrasound is a significant barrier or poses a significant risk during the COVID‐19 pandemic, EMA may proceed without the necessity of ultrasound assessment.14 However, the consensus statement emphasises that women should be carefully screened for risk factors for ectopic pregnancy. This requires an assessment as to whether an accurate gestational age can be estimated from the woman's history; a discussion regarding the risks of foregoing a pre‐procedure ultrasound as part of the consent process and supported by written information; and a robust follow‐up pathway.14 If the gestation is unable to be accurately identified, or there are red flags for ectopic pregnancy, then an ultrasound assessment must be arranged.14 Finally, abortion has been decriminalised in every state and territory in Australia except South Australia,1 where mifepristone can only be supplied in a hospital setting. This precludes South Australian women from being able to access EMA through community‐based providers such as GPs or via telehealth. The relevant South Australian legislation therefore requires a change.

Danielle Mazza · Seema Deb · Asvini Subasinghe

The advent of CAR T‐cell therapy has seen significant improvements in survival and is a potential cure for patients with advanced haematological malignancies Cancer immunotherapy is a burgeoning field which, in the last decade, has produced unprecedented improvements in outcomes across a variety of advanced malignancies. The eventual translation of decades of research into clinically available immunotherapies stems from the expanded knowledge of the role that the immune system plays in preventing tumour initiation and progression as well as the mechanisms by which tumours learn to evade this immune surveillance.1 Immunotherapies that have reached the clinic include monoclonal antibodies and, more recently, their augmented counterparts including antibody–drug conjugates and bi‐specific T‐cell engagers. Other treatments are immunomodulatory, meaning that they augment endogenous anti‐tumour immune activity. These include immune checkpoint inhibitors such as pembrolizumab, which are prolonging survival in melanoma and several solid organ malignancies as well as relapsed or refractory Hodgkin lymphoma. Cellular immunotherapies offer the potential to overcome immune tolerance and generate immune memory.1 Allogeneic stem cell transplantation (ASCT), a largely unmanipulated form of cellular immunotherapy, acts by completely replacing the recipient’s entire haematopoeitic and immune systems, leveraging differences between the recipient and donor to produce a graft‐versus‐tumour effect, with the potential negative consequence of immune attack on recipient’s normal tissues, known as graft‐versus‐host disease, as well as other serious toxicities. ASCT has been the only curative option for many patients with haematological malignancies. However, it is generally considered a consolidative therapy; that is, the patient’s malignancy must be in or near complete remission in order to be effective. This is not always possible in refractory cases. For others, ASCT may be contraindicated because of age or comorbidities. With advances in genetic manipulation technology, the notion of combining the specificity of a monoclonal antibody with the cytotoxicity and memory of a T‐cell came to fruition in the chimeric antigen receptor (CAR) T‐cell. “Chimeric” here means that the DNA comes from two or more sources; the antigen‐binding domain of the CAR construct is an antibody fragment, tethered to the intracellular signalling domain of the T‐cell receptor, with an additional co‐stimulatory domain acting to improve their expansion and persistence in vivo. The fundamental steps in generating and delivering CAR T‐cell therapy are summarised in Box 1.2 Specific toxicities are characteristic of CAR T‐cell therapy, the two most important being cytokine release syndrome and neurotoxicity. Cytokine release syndrome is an inflammatory state induced by the rapid proliferation of CAR T‐cells and tumour cell death, releasing an array of inflammatory cytokines. The hallmark is a fever, with the potential for hypotension, hypoxia and organ dysfunction. As one of the key cytokines driving the syndrome is interleukin‐6, its blockade using the interleukin‐6 receptor antagonist tocilizumab is now routinely used for more severe grades of cytokine release syndrome. The pathogenesis of neurotoxicity has not been fully elucidated; however, it most often manifests with speech disturbance or aphasia, dysgraphia and attention deficits, with more severe manifestations including altered level of consciousness, seizures and, rarely, cerebral oedema. Fortunately, even patients with severe neurotoxicity who are adequately supported in intensive care settings most often have complete neurological recovery. By far the most successful antigen target of all CAR T‐cell therapies developed to date is the pan‐B‐cell antigen CD19, as it arguably comes closest to the characteristics of an ideal target. CD19 is widely expressed across the full maturation spectrum of B‐cell malignancies, from B‐cell lymphoblastic leukaemia cells to mature B‐cell lymphomas, giving broad applicability. Second, CD19 is only expressed on B‐cells (normal and malignant) and not other tissues. Third, the toxicity resulting from the on‐target, off‐tumour effects, in this case normal B‐cell aplasia, is manageable by immunoglobulin replacement in patients who experience recurrent or severe infections. The decision for health authorities to fund a personalised, genetically engineered treatment is a complex one, taking into account considerations such as cost, efficacy, safety, the level of evidence and the maturity of outcome data, alternative therapies, equity of access, and resource utilisation. The cost of a single product is measured in hundreds of thousands of dollars and the mechanism by which such therapies will be funded is certainly not self‐evident. In the Australian context, the new therapy is evaluated by the Medical Services Advisory Committee, an independent committee that appraises new medical services proposed for public funding, taking into account all the above‐mentioned considerations, and providing advice to government. Moreover, given the high cost and limited, immature data, regulatory bodies worldwide have come to unprecedented outcomes‐based reimbursement agreements with pharmaceutical companies — such as rebates and staged payments according to defined response criteria — in order to mitigate risk. Two CAR T‐cell products targeting CD19 were approved by the United States Food and Drug Administration in 2017 and 2018: tisagenlecleucel and axicabtagene ciloleucel. The landmark studies which led to their approval, and a summary of their key outcomes, are shown in Box 2.3,4,5,6 In Australia, tisagenlecleucel is approved by the Therapeutic Goods Administration for paediatric and young adult patients up to 25 years of age with B‐cell lymphoblastic leukaemia that is refractory, in relapse after transplant or in second or later relapse, as well as adult patients with relapsed or refractory diffuse large B‐cell lymphoma after two or more lines of systemic therapy. In April 2019, a joint state and federal government funding initiative commenced for tisagenlecleucel for the B‐cell lymphoblastic leukaemia indication, and in January 2020, the government announced its funding for diffuse large B‐cell lymphoma. Soon after, the Therapeutic Goods Administration approved axicabtagene ciloleucel in February 2020 and the Medical Services Advisory Committee made a positive recommendation for its public funding for the lymphoma indication. Further, Novartis announced an agreement with an Australian cell and gene therapy manufacturing company for manufacture of tisagenlecleucel for the region to commence in late 2020 (https://www.celltherapies.com.au/kymriah-to-be-manufactured-at-cell-therapies-pty-ltd-marking-australias-first-on-shore-commercial-production-of-car-t-therapy/). In the case of tisagenlecleucel for relapsed or refractory B‐cell lymphoblastic leukaemia, evaluation began with a comparison with best available therapy. In the ELIANA trial4 outcomes compared very favourably with other chemo‐ or immunotherapeutic salvage options such as clofarabine7 and blinatumumab,8 respectively. For example, blinatumomab, a bispecific T‐cell engager, in the paediatric relapsed or refractory setting produced a complete remission rate of 39% within the first two cycles, with a relapse‐free survival at 6 months of 42%, and this therapy is considered to be a bridging therapy to ASCT. Tisagenlecleucel on the other hand can be used as a stand‐alone therapy; however, it is notable that a substantial proportion of responders do relapse, particularly between 6 and 12 months, which raises the question of whether this treatment should also serve as a bridge to ASCT. The available evidence is currently insufficient to confidently provide an answer, and practice therefore varies among treating centres worldwide. However, a major concern is the financial implications of CAR T‐cell therapy as a bridge to ASCT, which itself is a highly resource‐intensive therapy, with some suggestion that the cost‐effectiveness may be unbalanced if this practice were routine. In the case of high grade B‐cell lymphomas, patient outcomes also appear to be superior to other available treatments in the third line setting. In the ZUMA‐1 trial, the recently updated 3‐year overall survival rate of 47% does likely reflect a significant cure fraction.6 In comparison, the SCHOLAR‐1 retrospective study of the outcomes of patients with refractory diffuse large B‐cell lymphoma showed that this pooled patient population only achieved complete remission rates of 7% with conventional therapies and had a median overall survival of 6.3 months.9 One concern is that the outcomes of the CAR T‐cell trials may not be generalisable to the real‐world population where patient selection may not be as strict as in clinical trials. Interestingly, the real‐world data seems to be conflicted on this, with the US experience from the Center for International Blood and Marrow Transplant Research registry being comparable to trial data for both tisagenlecleucel and axicabtagene ciloleucel, while preliminary United Kingdom experience appears to be considerably worse.10,11,12 The cause of this discrepancy is unclear. In terms of future directions, many clinical trials are assessing CAR T‐cells in earlier lines of therapy. For example, two trials are randomising patients in first relapse of large B‐cell lymphoma to receive either CAR T‐cell therapy or standard salvage plus autologous stem cell transplant: ZUMA‐7 (NCT03391466) and BELINDA (NCT03570892). The results of these trials, if favourable, could greatly alter treatment paradigms. Other trials are assessing CAR T‐cells in other B‐cell lymphomas, such as follicular non‐Hodgkin lymphoma (ELARA [NCT03568461]) and mantle cell lymphoma.13 The response to KTE‐X19, an anti‐CD19 CAR T‐cell therapy with a manufacturing process that removes circulating tumour cells, seen in the ZUMA‐2 trial in relapsed or refractory mantle cell lymphoma (overall response rate of 93%) is the highest reported response rate in patients with mantle cell lymphoma who failed previous BTK inhibitor treatment, with a high proportion of durable responses in this very challenging malignancy.13 Strategies to improve the availability and timeliness of CAR T‐cell therapy include the development of third party allogeneic CAR T‐cells, which could produce off‐the‐shelf treatments for many patients. Other alterations to the CAR construct aim to improve characteristics such as persistence and safety, as well as addressing the problem of antigen escape, where the malignancy loses the targeted antigen, potentially through multi‐antigen targeting. Others are combining CAR T‐cells with immunomodulatory therapies such as immune checkpoint inhibition to improve efficacy. Finally, there is great interest in CAR T‐cell therapies for malignancies such as multiple myeloma, acute myeloid leukaemia and T‐cell lymphomas and leukaemias, many of which are at various phases of clinical trials. The furthest advanced are CAR T‐cell therapies targeting B‐cell maturation antigen in multiple myeloma. JNJ‐4528, an investigational B‐cell maturation antigen CAR T‐cell therapy, has recently demonstrated very high response rates in the phase 1b/2 CARTITUDE‐1 study in relapsed or refractory myeloma.14 In the 29‐patient cohort, the overall response rate was 100%, with 69% complete remission, the median time to complete remission being 1 month, and measurable residual disease negativity in all 17 evaluable patients. These are very promising times in cancer immunotherapy and the task ahead for regulatory authorities will be immense as evidence rapidly accumulates for these high cost therapies. In the meantime, we are pleased to add CD19 CAR T‐cell therapy to our armamentarium and await the results of trials across a wide range of haematological and solid organ malignancies. Box 1 – Overview of the processes for manufacture and delivery of a chimeric antigen receptor (CAR) T‐cell product Procurement of T-cells, usually via leukapheresis (1); transduction of the CAR genes via viral vector or non-viral methods (2); ex vivo expansion of the CAR T-cells (3); preconditioning with lymphodepleting chemotherapy (4); and infusion into a patient (5). ◆ Box 2 – Summary of data from pivotal CD19 chimeric antigen receptor (CAR) T‐cell trials Trial name CAR T‐cell product Disease Complete response rate Other response parameters Safety ELIANA4 Tisagenlecleucel Relapsed or refractory paediatric B‐ALL 81% 12‐month OS, 76%; 12‐month EFS, 50% Grade ≥ 3 CRS, 47% Grade ≥ 3 NT, 13% JULIET5 Tisagenlecleucel Relapsed or refractory DLBCL 38% Median OS, 12 months Grade ≥ 3 CRS, 23% Grade ≥ 3 NT, 11% ZUMA‐13,6 Axicabtagene ciloleucel Relapsed or refractory DLBCL 58% 3‐year OS, 47% Grade ≥ 3 CRS, 13% Grade ≥ 3 NT, 28% B‐ALL = B‐cell acute lymphoblastic leukaemia; CRS = cytokine release syndrome; DLBCL = diffuse large B‐cell lymphoma; EFS = event‐free survival; NT = neurotoxicity; OS = overall survival.

Adrian G Selim · Constantine S Tam

Current regulations address product safety, but they do not adequately ensure accuracy of formula preparation and provision Breastmilk is the optimum source of nutrition for most infants born at full term. When breastmilk is unavailable or unsuitable, the only safe and nutritious substitutes are commercial infant formulas.1 Infant formula — predominantly powdered infant formula — makes a major contribution to infant nutrition in Australia, with the 2010 Australian National Infant Feeding Survey reporting that 34% of infants had been introduced formula in their first month of life, 45% by 2 months and 69% by 6 months of age.2 In Australia, infant formula products are regulated under Standard 2.9.1 — Infant Formula Products in the Australia New Zealand Food Standards Code (Std2.9.1IFPANZC).3 All commercially produced infant formula products available in Australia and New Zealand must comply with the composition and safety requirements outlined in the Code. Std2.9.1IFPANZC specifies the mandatory nutrient content for infant formula and follow‐on formula to ensure that nutrition requirements are met. The standard includes labelling requirements, specifically prohibiting various claims, images and symbols. While these regulations address product safety, they do not adequately ensure accuracy of formula preparation and provision. In particular, potential for error remains around formula powder reconstitution, given multiple differing brands with variable scoop to water ratios, and volume of feed for differing ages and body weights. In this article, we discuss the infant formula range available in Australia, the infant formula powder reconstitution variability and the potential impacts, and the variability of manufacturer feeding guides compared with recognised recommendations and potential implications. Formula brands and types In Australia, there are more than ten brands of infant formula from which to choose. Within each brand there are often minor variations, from standard formulas meeting basic Food Standards Australia New Zealand (FSANZ) formula composition guidelines through to manufacturer‐specified gold formulas and condition‐specific formulas (Supporting information, table 1). FSANZ guidelines describe the purpose of infant formula labelling as providing information to caregivers to make informed choices, as well as information about appropriate preparation and safe use of infant formula products. Under FSANZ guidelines, all infant formulas must meet essential nutrient requirements. Specific nutrient content and health claims are prohibited in Clause 3 of Standard 1.2.7.4 Despite regulation, there are often misleading names or ingredient claims on infant formulas which construe a health claim or benefit and create doubt or sway opinion in consumers. For example, “[trade name] constipation”, as a name of a formula may be assumed by a consumer to be a superior formula for babies with constipation. Similarly, a statement of “fish oil to help support brain and eye development” could potentially be interpreted by a consumer as a health claim. There is currently no unbiased, freely available source of information to help parents choose a formula and this is often the first point of confusion. The authors frequently encounter parents swapping formulas in response to their infant's behaviour, believing that another formula may offer benefit. Typical examples in clinical practice are changing from a standard term formula to a colic, antireflux, or casein‐predominant formula when there is irritability or spilling. Powdered formula reconstitution While infant formulas are made in liquid ready‐to‐feed and in powdered forms, the latter is predominantly used in the home. Under Std2.9.1IFPANZC, the labelling of a powdered formula product must include the powder to water reconstitution ratio to achieve the specified nutritional composition, and the weight of powder in one scoop. However, the Standard does not dictate scoop size and, consequently, the scoop to water reconstitution ratio is determined by the manufacturer, although the powder weight to water ratio is relatively constant between manufacturers. In Australia, there is significant variation in reconstitution ratios across brands. Australian infant formula dilution reconstitution ratios are most commonly either one scoop per 30 mL water, per 50 mL water or per 60 mL water. The choice between a smaller or larger ratio is manufacturer‐specific. Explanations company representatives have provided for choosing a smaller scoop to water ratio include being able to make up smaller quantities of formula, greater accuracy, and a reconstitution method that yields rounded number volumes of 100 mL. In contrast, companies with larger scoop to water ratios propose reduced risk of error in sleep‐deprived parents who might lose count of scoops. However, none of these justifications are evidenced‐based. While there is a general expectation that parents use the formula label instructions or community advice, brand changes enhance potential for parental miscalculation of formula concentration. Under Std2.9.1IFPANZC, all powdered infant formula products must carry a warning stating, “Warning — follow instructions exactly. Prepare bottles and teats as directed. Do not change proportions of powder except on medical advice. Incorrect preparation can make your baby very ill”.3 This warning is often not obvious, and in practice, we have observed parents swapping between formulas and either assuming that the scoop to water reconstitution ratio is the same, confusing the ratios between brands, or using the incorrect scoop with a different manufacturer's powder, resulting in incorrect formula concentration. A systematic review of five studies supports this observation, finding that significant errors may be made when reconstituting formulas.5 Incorrect dilution ratio results in a formula strength that is either too dilute, increasing risks of nutrient deficiencies and faltering growth, or too concentrated, risking hypernatraemic dehydration or excessive weight gain. A review of reconstitution recipes of the major brands of standard infant formulas reveals a formula powder to water ratio of 0.142–0.15 g/mL (Supporting information, table 2) and a narrow caloric strength range of 4.8–5.2 Kcal/g. Clearly, a standardised reconstitution recipe is possible. We propose that standardisation of reconstitution ratio of powdered infant formula to water would minimise error and risk while providing clarity for parents and health professionals. Formula feeding guides Infant feeding guidelines for health workers1 state that as a formula is designed to remain at a constant strength, it is the amount of formula that should increase as the infant grows. The guidelines outline approximate formula requirements for infants (Supporting information, table 3), which correlate appropriately with the estimated energy requirements of infants as outlined in the National Health and Medical Research Council (NHMRC) Nutrient reference values for Australia and New Zealand.6 The infant feeding guidelines also note that feeding guidelines on formula packaging recommending certain amounts for various ages are guides only and do not necessarily suit every infant.1 Manufacturers of commercial infant formulas usually include a feeding guideline on the formula packaging that outlines the number and volume of feeds recommended for the corresponding ages. This is not a requirement under the Food Standards Code. There is substantial inconsistency in the feeding guidelines for volume and frequency of feed by age printed on the containers (Box), both from one manufacturer to another and also compared with the NHMRC‐recommended volumes by age and weight. The lack of weight standardisation means that the caregivers of a small infant may overfeed, while a genetically larger infant might be underfed. Greater consistency or standardisation of manufacturer feeding guides that correlate appropriately with the NHMRC feeding guidelines may help reduce both over‐ and underfeeding as well as alleviate parental confusion and anxiety around feeding volumes. While both under‐ and overfeeding may have negative clinical consequences, there are no published data to support adverse outcomes as a common consequence of parental misunderstanding. The absence of published evidence, however, should be considered in the light of anecdotal experience of health professionals within our health service, who report spilling and irritability from overfeeding and parental anxiety when their baby does not achieve volumes stated on the formula tin. Expert opinion concurs with our own clinical experience, as shown by the 2018 guideline on gastroesophageal reflux issued jointly by the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPAGHAN), where the first step in management of an infant presenting with excess vomiting is to ensure that overfeeding is avoided.7 However, there is clearly a need for further study in the area of parental interpretation and use of the feeding guides provided on formula tins to determine the impact of variation in product labelling on health outcomes. Conclusion Infant formula is commonly used with a choice of brands and types of formula. There is limited access to unbiased advice on formula selection for parents, caregivers and health professionals who encounter feeding‐related problems in daily practice. In addition, there is a wide variation in reconstitution ratios of powdered infant formula due to differences in scoop sizes between manufacturers, which may contribute to error in formula concentration. Standardisation of reconstitution ratios is an opportunity to minimise error. Lastly, formula feeding guides provided on formula tins vary between companies and, by not accounting for weight, differ from NHMRC recommendations, which may lead to over‐ or underfeeding. Standardisation of formula feeding guides in line with NHMRC feeding guidelines, with clearer warning statements, may help reduce these risks. The absence of evidence as to the effectiveness and risks of current food and nutrition policy with respect to infant formula feeding is a significant gap in ensuring the safe care of infants both in our community and worldwide. We propose that this area becomes a future focus of public health research and advocacy for child health. Box – Standard infant formulas — manufacturer-suggested feeding volume(mL) and number of feeds per day juxtaposed as mL/kg/day for a 3rd centile female infant, 50th centile female infant and 97th centile male infant (World Health Organization growth data)

Shelley Farrent · Brian Coppin · Scott Morris

The pandemic has highlighted some important challenges in our health care system, but system reform can occur rapidly when needed

Gabriel Elan Blecher · Grant A Blashki · Simon Judkins

Compared with the mitigation and suppression approaches of most Western countries, elimination can minimise direct health effects and offer an early return to social and economic activity

Michael G Baker · Amanda Kvalsvig · Ayesha J Verrall

The challenge now is how to implement artificial intelligence technology safely into clinical practice Artificial intelligence is a branch of computer science that, in broad terms, deals with either decision making or classification. The aim of artificial intelligence is to surpass human cognitive functioning such that automated decisions can be made. Machine learning — an application of artificial intelligence — is commonly used in image recognition. In general, the machine, or algorithm, learns from exposure to a large dataset. Once learning has taken place, the algorithm can be applied to unseen data. The potential advantages of this approach in health care are clear: machines can learn from very large datasets in relatively short time frames and can apply themselves to new data without fatigue or intra‐observer replication error. Machine learning has recently demonstrated remarkable performance in image‐based diagnosis across various medical fields, including ophthalmology, radiology, pathology and dermatology. In dermatology, the primary focus has been on developing machine learning systems that facilitate classification and decision support for skin cancer management. Skin cancer (including melanocytic and keratinocytic malignancy) is the most common cancer in Australia and among Caucasian populations worldwide. Melanoma is responsible for the majority of skin cancer deaths in Australia and has various presentations.1,2 While dermoscopy has improved the accuracy of melanoma diagnosis, significant variability occurs and is largely a function of clinical expertise. Recent studies show that machine learning algorithms have the potential to surpass the diagnostic performance of experts, and the challenge now is how to implement this new technology safely into clinical practice. Although there are a number of machine learning algorithms that could be used in the dermatology setting, convolutional neural networks (CNNs) are the most promising. This is largely because they learn from data without any feature specification, and they are known to exhibit superior performance for image recognition in comparison with other machine learning algorithms.3 The aim of the CNN is to generalise its previously learned knowledge on unseen images beyond the training dataset. There are numerous parameters within a CNN that can be tweaked to maximise algorithm performance. Most of these parameters are adjusted automatically by the algorithm, without user input. Therefore, very little can be known, in principle, about why and how the algorithm reaches any particular decision. Currently, there are efforts underway to reduce the “black box” effect of CNNs. Some commercial software programs coupled to imaging devices will provide the user with comparable lesions to justify the algorithm's output and improve transparency. However, this retrieval system may fail for rare or unseen cases and does not provide a decision‐making process. While the black box phenomenon remains, there are two potentially negative implications for clinical practice: first, clinicians may have difficulty upskilling by following the algorithms’ outputs; and second, there exists the potential for deskilling and underperforming due to an over‐reliance on technology.4,5 The effect of a faulty system has been explored by manipulating a previously trusted algorithm to generate incorrect classifications and found that doctors of all experience levels were susceptible to being misled by the recommendation.5 Algorithm performance is dependent on both the size and quality of the training image dataset and on whether the algorithm is used in situations for which it was intended. Depending on the training set, the device may be limited in its ability to diagnose specific lesions (eg, non‐pigmented), or lesions in certain skin types (eg, darker skin) or sites (eg, scalp or acral). Retrospective image databases used to train algorithms may be associated with bias. In addition, artefacts (eg, hair, dermoscopic gel, air bubbles, rulers, pen markings, reflections) can distract from key features. However, if a CNN is trained on a large enough cohort, it can learn to deal with potential artefacts. Nonetheless, unbiased lesion selection and standardised image capture would invariably improve algorithm performance, and recent advances in three‐dimensional (3D) imaging modalities will enable this.6 Several studies have now shown that CNNs trained on retrospective image data collected at a single time point are capable of classifying skin cancer with sensitivities and specificities equal or superior to that of dermatologists (Box 1),5,7,8,9,11 and clinicians with less experience gain most from AI support under experimental conditions.5 Hypomelanotic and acral melanoma can be more challenging to diagnose clinically,1 and this could potentially present a challenge for automated classification. However, CNNs have achieved greater accuracy for hypopigmented and acral lesions in comparison with human experts, at least in silica.9,11 In addition to clinical images, CNNs have been applied to histopathological images of melanoma and benign naevi with promising results.10 The ground truth for lesion diagnosis The gold standard for melanoma diagnosis is histopathological assessment. However, there exists significant inter‐ and intra‐observer variability in histological diagnostic labels attributed to atypical melanocytic lesions.12 The existence of such variability in diagnoses poses the dilemma of whether the CNN has learnt from the correct set of diagnoses. Consensus diagnoses, if practical, may help overcome this problem. Molecular biomarkers may assist in establishing a diagnosis13 and identifying high risk biology,14 but they require extensive validation before clinical use. Pathologists and clinicians also rely on metadata (age, personal and family history, lesion symptoms, recent change), which may influence diagnostic likelihoods. Importantly, it is possible to incorporate different data types, including metadata, sequential image data coupled with histopathology, to train future CNN algorithms and improve diagnostic discrimination of borderline lesions (Box 2). Use of artificial intelligence for melanoma screening It is well known that the incidence of invasive melanoma in Australia has increased over the past 40 years. In addition, there has been a striking increase in incidence of in situ melanoma over the past decade, from 32 cases per 100 000 population in 2004 to 80 per 100 000 population in 2019, with age‐standardised mortality remaining fairly stable.2 The potential causes for the increase in incidence are complex, and involve a true increase, driven by poor sun exposure practices of individuals born before the SunSmart era, combined with increased awareness, excessive screening, and overdiagnosis. It has recently been estimated that 54% of melanomas (15% of invasive melanomas) are overdiagnosed.15 Artificial intelligence‐assisted targeted screening of high risk individuals is likely to be a more effective strategy to save lives than the current opportunistic approach. With sequential whole‐body image datasets linked to metadata, molecular biomarkers and clinical outcomes, our ability to identify lesions associated with sinister biological potential will improve (Box 2), thereby reducing unnecessary biopsies, minimising overdiagnosis and other potential harms associated with screening. Use of artificial intelligence in clinical practice There are advantages and disadvantages of introducing artificial intelligence at different points in the patient care pathway.16 An artificial intelligence system used as a triaging tool before clinician assessment would enable automated risk stratification of individuals and/or lesions (Box 2). This approach could dramatically improve clinician workload and timely access to specialist care for people requiring urgent attention. Alternatively, artificial intelligence consulted following an examination by the clinician may act as a second opinion to improve diagnostic sensitivity and reduce unnecessary biopsies.5 The latter is more closely aligned with current clinical workflows and therefore likely to be preferred while the field matures. There is potential for over‐reliance on artificial intelligence systems in both scenarios. A secondary support system may provide the clinician with a diagnosis or a management decision. Doctors are more likely to change their minds if they are uncertain of a diagnosis and an algorithm provides a conflicting result.5 It is thus important to consider how an algorithm might convey uncertainty to avoid false guidance. For example, a decision‐support output (eg, excise, monitor or reassure) avoids the diagnostic dilemma of differentiating between melanoma and dysplastic naevi. However, the problem is complex and arguments exist as to why, in many situations, a diagnostic probability output might be more desirable. Safe implementation of new technologies The Therapeutic Goods Administration (TGA) has developed an action plan to improve the processes by which new devices are approved for use in Australia, strengthen monitoring and follow‐up, and provide more information to consumers about the devices they use.17 International collaborations also exist with groups, such as the International Medical Device Regulators Forum, to establish better processes for medical device regulation globally. If software is classified as a medical device (ie, it is intended for diagnosis, prevention, monitoring, treatment or alleviation of disease), it must be registered on the Australian Register of Therapeutic Goods following TGA approval and before distribution within Australia. Consumers and clinicians need to be aware of the intended use of an application or device. There are several smartphone applications available to the general public, with functionality ranging from education to monitoring and tracking to skin lesion classification. Some of these provide skin lesion risk assessment, although they may state that they are not intended to be used as a diagnostic device. There is concern that, if this is not immediately obvious to the consumer, unregistered applications may be used in lieu of seeking medical advice. Unsupervised consumer‐operated diagnostic devices would require careful testing before they can be recommended. Conclusion As clinicians, we need to be aware of the limitations of any diagnostic tool and interpret outputs accordingly. Although the performance of artificial intelligence to date is promising, it remains to be seen how diagnostic devices in dermatology will influence decision making in the clinic and affect patient outcomes. Regardless of the specialty, any new technologies need to be rigorously tested before implementation and monitored after implementation. Ultimately, responsibility for patient care remains with the clinician and, as such, a high level of clinical acumen must be maintained. Nonetheless, artificial intelligence in dermatology is primed to become a powerful tool in skin cancer assessment. Box 1 – Comparison of skin cancer classification tasks by artificial intelligence (AI) systems and dermatologists/pathologists Study AI architecture Images Classification task Training dataset size Test dataset size AI Dermatologists/pathologists Sensitivity Specificity AUC/overall accuracy Sensitivity Specificity AUC/overall accuracy Tschandl5 ResNet34 CNN Clinical (dermoscopic) Benign v malignant v non‐neoplastic skin lesions 10 015 1412 0.81 (0.79–0.83)* 0.92 (0.90–0.93)* 0.73†(0.70–0.76)* 0.80 (0.78–0.83)* 0.80 (0.77‐0.82)* 0.60† (0.57–0.63)*,‡ 0.86 (0.84–0.88§)* 0.88 (0.87–0.90§)* 0.74† (0.71–0.77§)* Esteva7 GoogleNet Inception v3 CNN Clinical (macroscopic, dermoscopic) Benign v malignant v non‐neoplastic skin lesions 129 450 1942 na na 72.1%¶ ± 0.9% na na 66.0%¶ Haenssle8 GoogleNet Inception v4 CNN Clinical (macroscopic, dermoscopic) Benign melanocytic naevi v melanoma > 100 000 100 86.6%** 82.5%** 0.86** 86.6%** 71.3%** 0.79** 88.9%†† 82.5%†† 0.86†† 88.9%†† 75.7%†† 0.82†† Tschandl9 GoogleNet Inception v3 CNN Clinical (macroscopic, dermoscopic) Benign v malignant hypo‐pigmented lesions 13 724 2072 81% 53.5% 0.73 78% 51.3% 0.68 Hekler10 ResNet50 CNN Histopathology Benign naevus v melanoma 595 100 76% 60% na 51.8%‡‡ 66.5%‡‡ na Fujisawa11 GoogleLeNet DCNN Clinical (macroscopic) Benign v malignant skin lesions§§ 4867 1142 96.3% 89.5% 92.4%¶ ± 2.1% na na 85.3%¶ ± 3.7% AUC = area under the curve; na = not applicable. * 95% CI. † Youden statistic. ‡ Clinicians with varied experience and training. § Clinician accuracy with multiclass probabilistic AI support. ¶ Overall accuracy. ** Level I: AI and human readers provided with dermoscopic images only. †† Level II: AI provided with dermoscopic images only, human readers provided with dermoscopic images, macroscopic images and additional clinical information. ‡‡ Pathologist. §§ 52.6% of melanomas in this study were acral. Box 2 – Incorporation of different data types to train future convolutional neural network (CNN) algorithms and improve diagnostic discrimination of borderline lesions AI = artificial intelligence.

Miki Wada · ZongYuan Ge · Stephen J Gilmore · Victoria J Mar