Subscribe
×

Preference: Infectious Diseases and Parasitology

434

Antimicrobial use in Australian hospitals: how much and how appropriate?

Antimicrobial agents play a central role in modern health care, especially in the hospital setting. Many of the modern advances in health care such as intensive care, neonatal care, cancer chemotherapy, complex surgery and prosthetic joint replacement depend on the ongoing effectiveness of antimicrobials. However, antimicrobial resistance (AMR) has emerged in the most common and important pathogens, and as resistances have accumulated, multiresistant strains have arisen. The impact of resistance and multiresistance is being felt worldwide.1

In Australian hospitals, some multiresistant organisms were epidemic and have now become endemic in many facilities, particularly those delivering tertiary and quaternary care. Much of this resistance is driven by hospital antimicrobial use. Because antimicrobials are necessary for providing safe care in hospitals, questions arise about what factors in antimicrobial use may be adjusted in order to deliver both low levels of resistance and low prevalence of multiresistant organisms. Factors to consider include relationships between volumes of antimicrobials used and AMR; patterns of use and AMR; and appropriateness of prescribing.

The Australian Commission on Safety and Quality in Health Care (the Commission) has produced a range of resources regarding antimicrobial stewardship, now considered to be the most effective tool in promoting rational antimicrobial use. Antimicrobial stewardship is a term that describes a suite of activities designed to maximise the rational use of antimicrobials and minimise the selection pressure for AMR. In 2011, the Commission released a blueprint for antimicrobial stewardship in Australian hospitals.2 In 2012, the Commission released the National Safety and Quality Health Service Standards (NSQHSS) for hospital accreditation. Within Standard 3: Preventing and controlling healthcare associated infections3 there is a requirement for hospitals to develop and implement an antimicrobial stewardship program, and regularly review its effectiveness (Standard 3.14). Subsequently, the Commission has developed a clinical care standard for antimicrobial stewardship, designed to provide a basic set of nine standards that prescribers of antimicrobials should follow.4

In 2013, the Commission also commenced the task of coordinating the development of a national surveillance system for antimicrobial resistance and antimicrobial use — the AURA (Antimicrobial Use and Resistance in Australia) project (http://www.safetyandquality.gov.au/national-priorities/amr-and-au-surveillance-project). Two key elements of the system are passive and targeted surveillance of antimicrobial use in hospitals. In this context, passive surveillance refers to the collation of data primarily collected for other purposes (such as pharmacy dispensing data), while targeted surveillance refers to data gathered for a specific purpose — for example, data on appropriateness of prescribing and compliance with guidelines.

The AURA project was in a good position to develop both of these types of surveillance by building upon existing initiatives. The National Antimicrobial Utilisation Surveillance Program (NAUSP), established in 2004 and funded by the Australian Government Department of Health, forms the basis of the passive antimicrobial surveillance. The program built on a South Australian initiative to incorporate data from other Australian states. Targeted surveillance is based on the National Antimicrobial Prescribing Survey (NAPS), a 2011 initiative of clinicians at Royal Melbourne Hospital to establish a national auditing tool for hospital stewardship teams to use in assessing appropriateness of prescribing in their own hospitals using published methodology.5 This has matured into an online tool that was rolled out nationally in 2013. Although both programs are built on voluntary participation (there are currently no mechanisms available to mandate participation), by 2014 NAUSP had 129 participants and NAPS had 248 participants.

Data sources and representativeness

The data presented here are from the 2014 NAUSP report6 and the 2013 and 2014 NAPS reports.7,8 All these results must be interpreted with caution, given the voluntary nature of participation in the programs.

We examined the representativeness of participation in both programs in 2014 using the data provided in the Australian Institute of Health and Welfare (AIHW) publication on hospital peer groups in November 2015, to which we added two hospital sites that had provided data in 2014 but did not appear on the AIHW list.9 Of the 748 public hospitals in Australia, the NAUSP program collected data from 14.8%, and the NAPS program collected data from 26.3%. The percentages from the 497 private hospitals were 3.6% and 10.3% respectively. The Northern Territory did not provide usable data to NAUSP in 2014. In addition, NAUSP did not collect data from children’s hospitals (see below). Participation was greatest from principal referral hospitals, and public and private acute hospitals in groups A and B of the AIHW peer group list; NAUSP collected data from 61.0% of 195 hospitals in these categories, while NAPS also collected 61.0%; of these 45.1% were in common. The majority of participants were from remoteness groups “major city” and “inner regional” (93.0% for NAUSP and 76.0% for NAPS), reflecting the availability of resources in larger/hospitals and centres. Overall, 92 hospitals contributed data to both programs in 2014. A more detailed analysis of representativeness is given in Supplementary Tables S1–S3 in the Appendix at mja.com.au. Information on the types of data collected in each program, including their limitations, is given in the volumes of use and appropriateness sections below. The imbalances created by voluntary participation are found in these tables.

Volumes of antibacterial use, including variations and trends

NAUSP collects data from public and private hospitals, excluding paediatric hospitals and paediatric wards. These exclusions are because there is no internationally agreed measure equivalent to that used for adults, namely defined daily doses (DDD) per 1000 occupied bed days (OBD). It reports on use of antibacterials, but not antifungals or antiviral agents, dispensed to adult, acute care inpatient wards and emergency departments. Data from psychiatric services, rehabilitation, dialysis and day surgery units are not included in order to align with international definitions of somatic care, and outpatient data are also not included. The data also exclude most topical formulations, as well as antimycobacterials (except rifampicin) and antimicrobials prepared as infuser packs. NAUSP undertakes data validation at the time of receipt, including outlier detection. All suspect data are queried with the relevant participant, and corrections made if errors are confirmed.

Across all Australian hospitals participating in calendar year 2014, the total aggregate use rate of systemic antimicrobials was 936 DDD/1000 OBD, ranging from a low of 330 to a high of 2040 DDD/1000 OBD — more than a sixfold difference. The overall rate compares with 597 DDD/1000 OBD in Sweden,10 747 DDD/1000 OBD in the Netherlands,11 and 944 DDD/1000 OBD in Denmark.12 These are the only countries currently publishing comparable data.

There were noticeable variations between states (Box 1), the highest rates being in Tasmania. The reasons for such variations are not clear. While there is likely to be less antimicrobial use in small states and territories due to lack of availability of highly complex interventions, this would not seem to account for the pattern observed. Higher use in Tas may have been related to statewide adoption of combination narrow spectrum antimicrobials for urinary and intra-abdominal infections (eg, amoxicillin + gentamicin + metronidazole = 10–12 DDD compared with ceftriaxone + metronidazole = 2 DDD). Imbalances in representativeness between states (see Supplementary Table S2 in the Appendix at mja.com.au) are also likely to have contributed to the variation, although even greater variation has been observed nationally between hospitals in the same peer group.13

There is also some variation in use related to hospital type. Using the new peer group classification of public hospitals promulgated by the AIHW,9 aggregate rates of usage in principal referral hospitals in 2014 were 927 DDD/1000 OBD (range, 544–1511), 985 DDD/1000 OBD in large public acute hospitals (range, 451–2050), and 872 DDD/1000 OBD in medium public acute hospitals (range, 504–1345). This variation was different to what might have been expected, given the assumption that principal referral hospitals would have the most complex patients and procedures and therefore require the highest antimicrobial use. This may also be a reflection of the presence of more mature stewardship programs in principal referral institutions.

Box 2 shows usage rates of the various classes of antimicrobials over the ten years to 2014. There has been a steady decline in total usage nationally since 2010. Some of this can be attributed to the increasing numbers of smaller institutions joining the program, but it is thought that at least some part of the reduction can be attributed to an increasing focus on antimicrobial stewardship. During the period from January to December 2014, the aggregate antibacterial usage rate for all participants (n = 129) was 936 DDD/1000 OBD (Box 2). Compared with a rate of 961 DDD/1000 OBD for the previous year (n = 113), this represents a 2.6% decrease. Excluding the new participants, the decrease was still 1.6%.

The most commonly prescribed antibacterials in Australia are shown in Box 3 as a percentage of all use or prescriptions. The differences in data collection methods between NAUSP and NAPS mean that there are differences in many of the percentages: NAUSP captures all use throughout the year and provides output as a percentage of total DDD/1000 OBD. NAPS output is cross-sectional data at a point in time or over a short time interval. Hence, agents such as cefazolin, which is prescribed frequently but often for very short durations, features prominently in NAPS, but less so in NAUSP. Of note is the significant proportion of use/prescriptions for ceftriaxone (4.5% in NAUSP and 9.1% in NAPS), an agent that many antimicrobial stewardship teams would consider should be reserved for specific indications and should only be prescribed with stewardship team approval.

Appropriateness of antimicrobial use, including variations

The principal aim of the NAPS tool is to provide data to antimicrobial stewardship teams on the rates of appropriate use and compliance to guidelines in their own hospitals. The stewardship teams are able to use the data to identify local issues and prioritise their stewardship activities. The utility of the online collection and storage of data allows individual teams to compare themselves with their peers, and provides the opportunity to gain better insight into national and regional rates. Participation is voluntary, but has been significantly boosted by the release of Standard 3.14 in the Commission’s hospital accreditation standards. Participation resource requirements are significantly lower than for NAUSP, because NAPS surveys are conducted annually, largely as a point prevalence survey to coincide with antimicrobial awareness week in November. The survey methods have been developed to support hospitals of all sizes and complexity, so that point prevalence surveys (single and repeated), period prevalence surveys, and random patient selection surveys are all available to hospitals. Benchmarking is only available for point prevalence and random selection surveys. The NAPS tool has been developed using published methods5 modelled on those developed and applied in Europe,14 and which are now being applied internationally.15,16

To ensure data validity, participants are required to follow a structured assessment guide. In addition, training is provided to all participants and phone advice is offered by the NAPS team for participants who have uncertainties or questions about specific prescriptions. To reduce the impact of lack of expertise in sites that do not have direct access to expert advice from infectious diseases physicians or microbiologists, remote support is provided by skilled antimicrobial stewardship personnel.17 The validity of this approach has been reinforced by the consistency of findings between the years 2013 and 2014 (Box 4). In 2014, 248 hospitals participated, including 197 public hospitals (26.3% of all public hospitals in Australia) and 51 private hospitals (10.3%). This was a 64% increase on the number of participants in the previous year. The most popular method, a whole-of-hospital point prevalence survey on a single day, was conducted by 42.3% of participants, followed by a whole-of-hospital repeated point prevalence survey (with the time between surveys varying from consecutive days to several months) by 28.6%. Together, these two methods accounted for 70.9% of hospitals but 79.9% of all prescriptions reviewed. The remaining 29.1% of hospitals chose to survey selected wards or specialties, collected a random sample, selected antimicrobials or indications, or used other methods. The total number of prescriptions reviewed by all hospitals was 19 994. The variation in data collection methods means that the data presented here should be considered at best indicative.

Box 4 shows the overall findings for the key indicators of appropriateness and compliance with guidelines, comparing results from 2014 with those found in 2013. For these indicators, all participants were provided with standard recommendations for assessing appropriateness, and the Therapeutic guidelines: antibiotic18 or locally developed guidelines/protocols were used to judge compliance. Best practice rates for some of these indicators were included. Among the results were:

  • Only 74% of prescriptions had an indication documented in the medical record (best practice, >95%). If prophylaxis indications were excluded, that figure rose to 79.2%.

  • The prescription was considered inappropriate in 24% of cases where a satisfactory assessment of the prescription could be made.

  • Non-compliance with guidelines was detected in 26% of cases where compliance was assessable.

In 2014, there were some differences between states in rates of appropriate prescribing, with the Australian Capital Territory having the lowest rate (68.2%) and Queensland the highest (79.2%). Similarly, rates of compliance with guidelines had differences across states, with ACT having the lowest 65.2% and Queensland the highest (72.5%). There was little variation in rates of appropriate prescribing across hospital peer groups, but there was a tendency for compliance with guidelines to be lower in smaller hospitals. Remoteness had little impact on either appropriateness or compliance with guidelines. Rates for both were slightly lower in private hospitals (75.1% and 65.2% respectively) compared with public hospitals (77.2% and 70.6% respectively).8

A notable finding was the frequency with which surgical antimicrobial prophylaxis was administered for more than 24 hours. For most surgical procedures where antimicrobial prophylaxis is warranted, a single dose of the agent at the time of skin or mucosal incision, or less than 1 hour before the procedure, is known to be adequate.18 In 2013, more than 40% of surgical prophylaxis courses lasted longer than 24 hours, falling to 36% in 2014. The findings of the 2014 survey also identified that surgical prophylaxis was not indicated at least 20% of the times that it was prescribed.

Antimicrobial agents that are often used as directed therapy, for instance flucloxacillin for staphylococcal infection, or antimicrobials prescribed as part of treatment protocols for cancer chemotherapy, tended to have high rates of appropriate use in 2014. For prescriptions that could be adequately assessed, five antimicrobials (cefalexin, amoxicillin–clavulanate, azithromycin, cefazolin and ceftriaxone) all had rates of inappropriate use above 30%, with that for cefalexin exceeding 40%.

Apart from surgical prophylaxis as discussed above, the other notable condition for which antimicrobials were frequently prescribed inappropriately in 2014 was infective exacerbations of chronic obstructive pulmonary disease, with a rate of appropriate prescribing of only 62%. For the latter condition, appropriate prescribing involves the use of oral amoxicillin or doxycycline, and not broader spectrum or intravenous antibacterials.

Possible areas for action and intervention

Despite the voluntary nature of participation in the NAUSP and NAPS, a picture of antimicrobial practices in Australian hospitals is now emerging from these datasets. Before the emergence of these programs, there was no information about antimicrobial use across Australia. The datasets indicate that there are major opportunities for improvement for antimicrobial stewardship programs, and provide strong evidence for their introduction in hospitals where stewardship systems have not yet become established,. Stewardship programs should focus on the implementation of the Antimicrobial Stewardship Clinical Care Standard,4 and also ensure that prescribers have access to treatment guidelines, especially the national Therapeutic guidelines: antibiotic.18

The most important benefits of NAUSP and NAPS are that they provide stewardship programs with the capacity to benchmark and identify issues peculiar to their own institution. This greatly assists in directing efforts to the areas of greatest need at the local level. Nevertheless, NAUSP has identified high rates of use across Australia (57% higher than Sweden for example), albeit with a trend downward. NAPS is the only published antimicrobial prescribing survey that includes appropriateness of use as a composite measure of prescribing quality. This is important because clinical guidelines cannot cover all aspects of empirical and directed therapy. Recent reports show that the number of hospitals participating in NAPS well exceeds those participating in point prevalence surveys in other countries.15,16 Further, NAPS has identified major areas for intervention, such as the appropriate use of surgical prophylaxis, and prescribing for infective exacerbations of chronic obstructive pulmonary disease. The Commission believes that surgical prophylaxis is worthy of a national approach and has contacted several peak bodies such as the Royal Australasian College of Surgeons to drive improvements in this area.

The adoption of the NSQHSS has provided a significant boost to participation in NAUSP and NAPS. Stewardship programs are now able to compare their performance with overall use and prescribing practices. The findings of these two programs are already generating benefits for the participants. However, participation is still voluntary. Ultimately, all hospitals will be required to generate data of this kind for accreditation against Standard 3 of the NSQHSS. Participation in these national programs will be a simple method for them to achieve this.

The voluntary nature of participation in these programs has necessarily created imbalances in participation between hospital types and jurisdictions. Participation requires time and effort from hospital pharmacists, including stewardship pharmacists, infectious diseases physicians, microbiologists and often infection control staff and other nursing staff. Hospitals vary greatly in their size, throughput and resources (see separation rate variation in Supplementary Table S1 in the Appendix at mja.com.au). In some circumstances, particularly in outer regional and remote areas, pharmacy services are delivered offsite, and the hospital does not have access to the data required for NAUSP participation. Nevertheless, the Commission will continue to promote participation in both programs for all hospitals because the principles of antimicrobial stewardship apply everywhere that antimicrobials are used.

Both NAUSP and NAPS have identified specific issues and gaps in their data collection. For NAUSP, the single most important issue is the measure used — DDD/1000 OBD. While this measure is suitable for international comparisons, there are a number of instances where the internationally agreed DDD does not concur with the most commonly prescribed daily dose in Australia. Another important issue is the lack of DDDs for children, leading to underreporting of total volumes of use in the paediatric population. Moreover, generating totals for antimicrobials does not reveal how many patients are exposed to antimicrobials during their stay. This is likely to be a more relevant measure of selection pressure for antimicrobial resistance. The NAPS data are also vulnerable to the level of expertise of the participants in assessing appropriateness of use and compliance with guidelines, although, as noted above, strenuous efforts are being made to overcome this problem. Ultimately, all participants will be trained and skilled to undertake these surveys. As the AURA project moves from establishment to maintenance, it will be necessary to fill in the gaps and solve the issues found in both programs, as well as to encourage both ongoing and increased participation.

The building of workforce capacity and competence to undertake antimicrobial stewardship is an essential requirement to meet the objectives of the National Antimicrobial Resistance Strategy.19 The evolution of pharmacy and clinical information systems, and the introduction of electronic medical records and electronic medicines management, will improve data capture and interpretation for NAUSP and NAPS significantly and will enhance the quality of data contributing to national surveillance of antimicrobial use.

Box 1 –
Total hospital antibacterial use by state (all NAUSP hospitals, all antibacterial classes), 2014*

DDD = defined daily doses. NAUSP = National Antimicrobial Utilisation Surveillance Program. OBD = occupied bed days. * Northern Territory did not provide usable data in 2014.

Box 2 –
Total hospital antibacterial use by year (all NAUSP hospitals, all antibacterial classes)

DDD = defined daily doses. NAUSP = National Antimicrobial Utilisation Surveillance Program. OBD = occupied bed days.

Box 3 –
Top ten antimicrobials in 2014 — proportion of total use (NAUSP) or prescriptions (NAPS)

NAPS = National Antimicrobial Prescribing Survey. NAUSP = National Antimicrobial Utilisation Surveillance Program.

Box 4 –
Results of key indicators in 2013 and 2014 for all participating hospitals8

Key indicator

% of total prescriptions

% change from 2013

2013

2014

Indication documented in medical notes (best practice, > 95%)

70.9

74.0

+ 3.1

Appropriateness

Appropriate (optimal and adequate)

70.8 (75.6)*

72.3 (75.9)*

+ 1.5

Inappropriate (suboptimal and inadequate)

22.9 (24.4)*

23.0 (24.1)*

+ 0.1

Not assessable

6.3

4.7

– 1.6

Compliance with guidelines

Compliant with Therapeutic guidelines: antibiotic18 or local guidelines

59.7 (72.2)†

56.2 (73.7)†

– 3.5

Non-compliant

23.0 (27.8)†

24.3 (26.3)†

+ 1.3

Directed therapy‡

na

10.4

na

No guideline available

11.0

4.6

− 6.4

Not assessable

6.3

4.5

− 1.8

Surgical prophylaxis given for > 24 hours (best practice, < 5%)

41.8

35.9§

− 5.9

na = not applicable * Figures without parentheses apply to all prescriptions. Figures in parentheses apply to prescriptions where appropriateness was assessable (12 001 prescriptions in 2013; 18 998 prescriptions in 2014); the denominator excludes antimicrobial prescriptions marked “Not assessable”. † Figures without parentheses apply to all prescriptions. Figures in parentheses apply to prescriptions where compliance was assessable (10 599 prescriptions in 2013; 15 899 prescriptions in 2014); the denominator excludes antimicrobial prescriptions marked “Directed therapy”, “Not available” or “Not assessable”. ‡ Introduced in the 2014 survey as a new classification category. § Where surgical prophylaxis was selected as the indication (2785 prescriptions).

[Correspondence] Effect of besiegement on non-communicable diseases: haemodialysis

We read with interest the Viewpoint by Slim Slama and colleagues1 highlighting the important considerations for delivery of care for non-communicable diseases in emergencies. However, the article does not address the negative effects of besiegement on non-communicable disease care. More than 1 million people are estimated to be living under siege in Syria. Patients who need haemodialysis suffer during both natural and man-made disasters.2,3 Minimum standards of care for patients who need haemodialysis include at least twice-weekly sessions of longer than 4 h and the availability of erythropoiesis stimulating agents and vitamin D supplements.

The use of lung function testing for the diagnosis and management of chronic airways disease

This demonstration data linkage study investigates the use of lung function testing in the diagnosis and management of chronic airways disease. It uses data from the 45 and Up Study linked to Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme (PBS) data. The study shows that: – contrary to recommended guidelines, most study participants (82%) initiated on medications for managing their chronic airways disease did not have lung function testing performed within 12 months, either before or after their initial prescription – among respondents who reported having current asthma, lung function testing was only claimed for about one-quarter (26%) in a recent 3-year period and 12% had only 1 lung function test claim during that time.The evidence from this study will inform initiatives to help improve appropriate prescribing and health outcomes for people with chronic airways disease. This project also provides a demonstration of methods that could also potentially be used to fill evidence gaps associated with other chronic diseases such as diabetes, arthritis and cancer.

Australian law needs a refresher on the science of HIV transmission

Being diagnosed with HIV is a confronting experience.

However the stigma associated with HIV infection – a hangover from its social and medical history – is responsible for an exaggerated perception of transmission risk through sex, and the harms of living with HIV infection.

In our consensus statement published this week in the Medical Journal of Australia, we detail the latest evidence on HIV transmission risk and recent advances in HIV prevention and treatment.

We propose that legal cases relating to HIV transmission should be considered in light of such evidence, and that alternatives to prosecution such as the public health management approach are often appropriate.

HIV infection no longer a death sentence

There have been many advances in HIV diagnosis, prevention and treatment since the identification of the first AIDS cases in the early 1980s.

In the initial days of the AIDS epidemic, patients would, after a number of years, develop serious infections and other illnesses due to their immune deficiency, usually resulting in death. When the first treatments became available, they bought time but often at the cost of serious medication side-effects, and complicated treatment regimens involving many tablets each day.

While it remains a serious infection, HIV is now a disease that can be effectively managed through medical treatment, regular health monitoring and healthy lifestyle. For many people with HIV, treatment involves taking only a single pill each day. Those taking antiviral therapy can expect to live a normal life, in good health, with a life expectancy similar to their HIV-negative counterparts.

These great improvements, familiar to those working in health, are not as well understood in the legal sector.

Prosecutions for HIV infection

Unlike other diseases, HIV has a long and uneasy relationship with criminal law. In the early years of the epidemic, the stigma around HIV, the fact that it was almost always fatal, and unfounded fears about its potential use as a weapon led to the criminalisation of HIV transmission and exposure.

Since 1991, there have been more than 38 criminal prosecutions for HIV transmission or exposure during sex in Australia. Despite the significant improvements in health and longevity of people living with HIV, the rate of criminal prosecutions has not decreased.

The courts have shown an understanding of the effectiveness of condoms: no one who has used condoms has been convicted. However, people continue to be prosecuted, including for “exposing” others to the risk of HIV infection, even in the absence of actual transmission. This occurs despite the relatively low per-act risk of HIV transmission and the fact that for most people the harms of HIV infection are far less serious than they once were.

New approaches to limit HIV transmission

HIV is actually difficult to transmit. Sexual transmission occurs during only about 1% (or less) of penetrative sexual encounters, even when a condom is not used and the HIV-positive person is not on treatment.

HIV prevention messaging in the early days of the epidemic focused on sexual abstinence and condom use. However, prevention messaging is now more nuanced and has expanded to include new ways of reducing HIV transmission risk. A not insignificant number of people at risk of HIV infection choose to have sex without using a condom, which is why developing alternative methods of HIV prevention have been prioritised in recent years. Such research has delivered game-changing results: “treatment as prevention” and “pre-exposure prophylaxis”.

Treatment as prevention refers to the greatly reduced risk of HIV transmission as a result of HIV-positive people taking antiviral treatment, which suppresses replication of the virus in the infected person’s body. When a person with HIV has a very low viral load (unmeasurable levels of HIV in the blood), the risk of sexual transmission becomes very low. In fact, there has never been a documented case of HIV transmission from a person with an undetectable viral load.

Pre-exposure prophylaxis describes the use of antiviral medication by an HIV-negative person as a way of preventing HIV infection. Pre-exposure prophylaxis is a very effective means of preventing HIV transmission, with only isolated cases of transmission identified among people applying this approach. This groundbreaking new strategy is available in Australia via limited pilot programs, and is being evaluated for Pharmaceutical Benefits Scheme listing.

How the law treats people with HIV

Criminal laws relating to HIV transmission and exposure vary from state to state, but a common factor is that people with HIV are expected to take “reasonable precautions” to prevent transmission. Condom use has long been accepted as meeting this threshold.

The evidence now supports acceptance of treatment as prevention (for the positive partner) and/or pre-exposure prophylaxis (for the negative partner) as meeting the same standard. And the limited harms of HIV infection as a consequence of acts involving low to negligible risk of transmission mean HIV cases generally do not belong in criminal courts.

There is an alternative. All states and territories have health protocols for managing allegations of risky behaviour. This public health approach – involving education, case management and, where required, behavioural orders and isolation – is a much more effective way of protecting public health.

As researchers and clinicians, we are intimately aware of the impact an HIV diagnosis can have. We have all supported patients coming to terms with an HIV diagnosis; many of us having had the painful task of delivering that devastating news.

The criminal law has a role to play, particularly should there ever be a case where a person deliberately transmits HIV.

However, with the advances of recent years in both prevention and treatment, authorities need to be more familiar with latest scientific and medical evidence, and consider alternatives to prosecution such as the public health management approach.

Elizabeth Crock, Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine, co-authored the consensus statement.

The ConversationMark Boyd, Professor, Chair of Medicine, University of Adelaide; Andrew Grulich, Professor and Program Head, UNSW Australia; David Cooper, Scientia Professor of Medicine and Director, Kirby Institute, UNSW Australia; David Nolan, Adjunct Professor, Institute for Immunology and Infectious Disease , Murdoch University; Levinia Crooks, Adjunct Associate Professor, La Trobe University; Michelle Giles, Associate Professor, Department of infectious diseases and Dept of Obstetrics and gynaecology, Monash University; Sharon Lewin, Director, The Peter Doherty Institute for Infection and Immunity, The University of Melbourne and Royal Melbourne Hospital and Consultant Physician, Department of Infectious Diseases, Alfred Hospital and Monash University, The Peter Doherty Institute for Infection and Immunity, and Trent Yarwood, Infectious Diseases Physician, Senior Lecturer, James Cook University and, The University of Queensland

This article was originally published on The Conversation. Read the original article.

Latest news

Whooping cough booster faces axe

The Federal Government may axe the whooping cough vaccine booster for first year high school students as it pulls plans for an Australian Schools Vaccination Register.

An immunisation expert group has been asked to review the pertussis vaccine schedule, including the need for a booster currently being administered to children in secondary school.

The Government has announced that the Australian Technical Advisory Group on Immunisation (ATAGI) has been asked to “provide advice on the clinical place and effectiveness of the pertussis vaccine schedule, including the pertussis booster currently given in the first year of high school”.

Currently, it is recommended that infants receive a dose of the diphtheria-tetanus-acellular vaccine at two, four and six months of age, with further boosters at 18 months and four years. An additional booster is given between 12 and 17 years.

The review comes at a time when the number of whooping cough cases is in decline – about 16,000 cases have been notified so far this year, well down from the 22,500 infections reported in 2015.

But the decline has come not long after the country’s largest-ever recorded outbreak of the disease, between 2008 and 2012, including 38,732 notified cases in 2011 alone.

The National Centre for Immunisation Research and Surveillance said whooping cough was a “challenging” disease to control because immunity waned over time, and epidemics occurred every three to four years.

The Centre said declining immunity was a factor in the last major outbreak, during which 4408 people were hospitalised, including 1832 babies. Between 2006 and 2012, 11 died from pertussis, all but one of them infants less than six months of age.

The review of the pertussis vaccination schedule coincides with the decision not to proceed with the creation of the Australian Schools Vaccination Register.

The Health Department said it had discontinued the tender process for the creation of the Register following advice about the review of the pertussis booster vaccine for secondary school students and the end, in 2018, of the catch-up varicella vaccination program for adolescents.

The Register was announced in the 2015-16 Budget as part of the No Jab No Pay policy, and was portrayed as vital in helping to controlling infectious disease outbreaks by identifying areas where vaccination coverage was low.

But Health Minister Sussan Ley said it had now been “put on hold…pending further advice from independent medical experts on the vaccination needs of adolescents”.

The Health Department said it was possible that the Schools Register would only hold data on the human papilloma virus (HPV) if the pertussis booster for adolescents was axed and once the varicella catch-up vaccination program ends.

The Health Department said it was now looking at alternatives to the Schools Register, including the inclusion of such data in the whole-of-life Australian Immunisation Register which began operations on 30 September.

It is also in discussions with the Victorian Cytology Service about continuing the HPV Register in 2017.

Commonwealth Chief Medical Officer Professor Brendan Murphy was keen to assure that these changes would have “no impact on the health of adolescents because the full range of vaccination services are being delivered to the community, and will continue to do so”.

The move to axe the Register has coincided with the release of Government figures showing that almost 200,000 children have had their vaccinations brought up-to-date following the introduction of the No Jab No Pay reforms.

The figures, reported in the Sunday Herald Sun, show that since the reforms were introduced on 1 January, 86,562 families, including 102,993 children, have been denied childcare payments, and $38 million of Family Tax Benefit A benefits have been suspended. Parents of 8896 children are still not meeting vaccination requirements.

But 183,000 children have had their vaccinations brought up-to-date as a result of the program, under which parents face losing Family Tax Benefit A and childcare payments if they let their child’s immunity slip.

Adrian Rollins

Sexual transmission of HIV and the law: an Australian medical consensus statement

All Australian states and territories have criminal laws that may be applied in cases of alleged human immunodeficiency virus transmission and, in some jurisdictions, exposure. None of these laws are HIV specific; they generally relate to causing grievous bodily harm, serious injury or grievous bodily disease, or to endangerment by exposure to the risk of infection.1

There have been at least 38 Australian criminal prosecutions for HIV sexual transmission or exposure since the first known case in 1991. Such cases require that courts, legal practitioners and juries interpret detailed scientific evidence on HIV transmission risk and the medical impact of an HIV diagnosis. Analysis suggests that scientific concepts have been inconsistently applied in Australian trials involving HIV.1,2 In some cases, the risks and impacts of HIV infection may have been overstated.

It is not the intention of this consensus statement to directly critique the evidence in past trials, or to suggest that any miscarriage of justice has occurred. Rather, we hope to inform the criminal justice system to ensure that future allegations are addressed in a scientifically robust way, consistent with the interests of justice. This statement, which draws on a similar consensus statement published in Canada,3 represents our opinion based on a review of the best available medical and scientific evidence. The evidence demonstrates that most sexual encounters entail low to no possibility of HIV transmission. While HIV remains a serious infection, the medical impact of an HIV diagnosis has decreased due to improved treatment; most people recently infected with HIV are able to commence simple treatment that offers them a life expectancy comparable with that of their HIV-negative peers.

HIV transmission risk

HIV is not transmitted readily from one person to another. HIV cannot survive outside the body or be transmitted through air. For transmission to occur, particular bodily fluids from an HIV-positive person (blood, semen, pre-seminal fluid, rectal fluids, vaginal fluids or breastmilk) must enter an HIV-negative person’s body.

For HIV transmission to occur through sex:

  • There must be a sufficient amount of the virus in the HIV-positive person’s bodily fluid.

  • Bodily fluid containing HIV must come into contact with an HIV-negative mucous membrane or damaged tissue. Mucous membranes are located in the foreskin and urethra of the penis; cervix and vagina; anus and rectum; and mouth and throat. In the absence of trauma, oral mucous membranes are much less vulnerable to HIV transmission than anogenital mucous membranes.

  • The virus must overcome the cellular defences of the HIV-negative person’s mucous membrane and the body’s immune response to pathogens in order to establish an infection in target immune cells.

Assessing HIV transmission risk from sexual activity

HIV transmission during sex is not inevitable. In fact, HIV is more difficult to transmit via sexual acts than many other sexually transmitted infections (STIs). Unlike the risk of transmission through HIV-infected blood, which rises to almost 100% through blood transfusion,4 risk of transmission through sexual acts is relatively low.

This consensus statement uses three categories to define the possibility of HIV transmission per sexual act: low, negligible and none (Box). These categories have been developed to inform consideration of HIV transmission risk in a criminal justice context. They have been deliberately chosen to differ from terms used in the public health context (which usually describes activities as ranging from high to low risk), where the messages function to encourage safe sexual practice. This consensus statement is intended for a legal context and is a more precise reflection of our clinical understanding of the likelihood of HIV transmission per event.

The possibility of HIV transmission during sex varies according to a number of intersecting factors.

The amount of HIV in a person’s bodily fluid (viral load)

Soon after being infected, a person’s viral load is high. Then, as the immune system responds, viral load typically decreases. If a person does not commence treatment, their viral load will eventually increase until HIV overwhelms their immune system and they begin to develop HIV-related illnesses.

When treatment is commenced, the viral load usually drops to undetectable levels (less than 50 copies per millilitre of blood) within a few weeks (“undetectable viral load” or “effective treatment”). With continuing adherence to treatment, undetectable viral load is usually maintained for prolonged periods. Having an undetectable viral load dramatically reduces the risk of transmitting HIV.5,6

Some people may be unable to access treatment, choose to defer commencing treatment or be unable to achieve an undetectable viral load on treatment; however, even a low viral load reduces transmission risk.

Type of sexual activity

Some activities carry higher risk than others (see below).

Condom use

HIV cannot pass through intact latex or polyurethane condoms. HIV transmission is possible only where condom slippage or breakage occurs. Epidemiological evidence on the reliability of condoms shows using condoms reduces transmission risk by at least 80%,7 although recent research suggests that estimate may underestimate condom reliability.8 Importantly, condom-related risk operates in conjunction with the established risk of particular behaviours, so correct condom use reduces the risk of HIV transmission to a level that is almost unquantifiable.9

At a population level, studies have shown that even when factoring in possible instances of incorrect use or breakage, consistent condom use dramatically reduces the possibility of HIV transmission.7

Whether the HIV-negative partner is using effective pre-exposure prophylaxis (PrEP)

PrEP is the use of anti-HIV medication by HIV-negative people to prevent HIV infection. Correct use of PrEP by an HIV-negative individual substantially reduces the risk of HIV acquisition.10–12

Whether a person who believes they have recently been exposed to HIV takes post-exposure prophylaxis (PEP)

PEP describes commencement of short term antiretroviral treatment by an HIV-negative person within 72 hours of a risk event. Even when the virus has entered a person’s body, PEP can stop the virus becoming established, significantly reducing the likelihood of the person becoming HIV-positive.

Risk associated with specific sex acts

Oral sex including oral–penile sex (fellatio) and oral–vaginal sex (cunnilingus)

When an HIV-positive partner has a very low or undetectable viral load, or a condom or similar latex barrier is properly used, or the HIV-negative partner is taking effective PrEP, or where oral sex is performed by an HIV-positive person on an HIV-negative person, there is no possibility of HIV transmission from oral intercourse.13

When oral sex is performed by an HIV-negative individual on an HIV-positive individual with a detectable viral load, there is negligible possibility of HIV transmission.13 Risk of transmission increases if the HIV-negative partner has cuts or lesions in their mouth and the HIV-positive partner ejaculates, but the risk remains very low.

Vaginal–penile intercourse

When a condom is used correctly, or the HIV-positive partner has a very low or undetectable viral load, or the HIV-negative partner is taking effective PrEP, there is negligible possibility of transmission from vaginal–penile intercourse.5,14 When two or more of these prevention strategies are simultaneously employed, risk of transmission approaches zero.

When the HIV-positive partner has a detectable viral load and a condom is not used, there is a low possibility of transmitting HIV during vaginal–penile intercourse.15 If no ejaculation occurs inside the HIV-negative partner’s body, transmission risk decreases significantly.

Anal–penile intercourse

When a condom is used correctly, or the HIV-positive partner has a very low or undetectable viral load, or the HIV-negative partner is taking effective PrEP, there is a negligible possibility of transmission through anal–penile intercourse.16 Where two or more of these prevention strategies are simultaneously employed, risk of transmission approaches zero.

When a condom is not used, and the HIV-positive partner has a detectable viral load, anal–penile intercourse poses a low possibility of transmitting HIV.17 The risk is similar whether the receptive partner is male or female. The risk is lower where the HIV-positive partner takes the receptive, rather than insertive, role. The possibility of HIV transmission during anal intercourse decreases when no ejaculation occurs inside the body.

Other risk factors

Presence of an STI, particularly an ulcerative STI, in either partner has been associated with an increased risk of HIV transmission during sexual activity.18 However, the presence of an STI does not increase transmission risk if a condom is used correctly or the HIV-positive person is on effective anti-retroviral therapy or the HIV-negative person is taking PrEP.

Biting and spitting

There is no possibility of HIV transmission from contact with the saliva of an HIV-positive person through spitting or biting. There is a negligible possibility of HIV transmission from spitting or biting where the HIV-positive person’s saliva contains blood, and their blood comes in direct contact with a mucous membrane or open wound, and they have a high viral load. No transmission through biting or spitting has ever been documented in Australia.

Evidence of transmission: phylogenetic analysis as forensic evidence

A small number of Australian trials have admitted expert evidence based on phylogenetic analysis to show an accused caused the complainant’s HIV infection.19 While phylogenetic studies are invaluable as research tools, their probative value as evidence of causation is limited because of the limitations of the methodology. This statement aims to minimise the incorrect application or interpretation of phylogenetic analysis to prove causation in HIV transmission cases.

HIV’s genetic sequence is widely variable at a population level, and changes rapidly over time both in response to the host immune response and in cases of HIV treatment resistance. Consequently, comparisons of genetic sequences of samples of the virus taken from different individuals can estimate the likelihood that these samples are linked by transmission events.20 To test whether any relationship exists, phylogenetic analyses use a statistical approach to assess the relatedness of viral genetic sequences in a process analogous to, but significantly less specific than, DNA profiling. Samples of viral RNA from the complainant and accused (reference samples) are compared to determine the degree of similarity. A database of samples from randomly selected but unrelated individuals (control samples) is employed to estimate the probability of the reference samples being related through a common ancestor.

This assessment is influenced by the number of control samples, which has typically been small in the context of criminal cases.21 The small number of controls significantly limits the reliability of analysis.

Phylogenetic analysis has its most useful application in definitively ruling out a connection between the infections of the complainant and accused, but it cannot determine beyond reasonable doubt that the reference samples are linked. Even where there is a strong correlation (high similarity) between the RNA sequences (irrespective of the size and characteristics of the control group), alternative hypotheses must be discounted, including infection of the accused by the complainant, infection of both complainant and accused by a common source person, or infection of the complainant by a third person infected by the accused.

Understanding the harms of HIV infection

Over the past two decades, HIV antiretroviral drug regimens have become simpler, more tolerable, and much more effective.22–24 Consequently, HIV is now treated as a chronic illness. For many people diagnosed with HIV, effective treatment is achieved by taking a single pill each day. Most people on treatment are able to achieve an undetectable HIV viral load and maintain a healthy immune system, which makes it highly likely that the person will remain healthy for a very long time and ensures they pose a negligible risk of transmitting HIV to sexual partners.

Life expectancy after HIV diagnosis has dramatically increased,24,25 to the point that the life expectancy of a recently diagnosed adult on antiretroviral treatment approaches that of an adult in the general population.22 Given that individuals living with HIV often exhibit demographic, clinical and behavioural characteristics associated with greater morbidity and mortality than the general population,26 any remaining gap in life expectancy may be attributable to these factors.27

Treatment of HIV has improved to the point that HIV-related illnesses are uncommon among diagnosed individuals. Instead, the focus of HIV care has shifted towards the management of non-HIV-related chronic diseases (such as smoking-related cardiovascular disease) which may be exacerbated by HIV and/or its treatment. Australian HIV surveillance programs no longer record the number of deaths from AIDS (acquired immunodeficiency syndrome), as this is no longer considered a useful marker of Australia’s HIV epidemic.28

A better understanding of treatments and prevention has also revolutionised conception and childbirth where one or both partners is or are HIV-positive. Where a couple is hoping to conceive, specific interventions (including antiretroviral treatment and PrEP) allow them to conceive through vaginal sex with negligible risk of HIV transmission from one partner to another.29 HIV per se does not adversely affect pregnancy. If interventions are implemented to minimise HIV transmission risk from mother to child during pregnancy, birth and postnatally, and a woman’s viral load is less than 50 copies per millilitre of blood, HIV transmission risk is reduced to less than 0.09%, including through vaginal birth.30

Although arguably fear surrounding HIV has decreased as public awareness has improved, people with HIV may experience psychosocial disadvantage such as stigma, discrimination and difficulties in interpersonal relationships.31 Such experiences vary by individual, typically decrease over time, and may be responsive to psychological and counselling interventions.32 Unfortunately, media coverage of criminal trials appears to heighten stigma and exacerbate other psychosocial issues for people living with HIV.

Recommendations

Given the limited per act likelihood of HIV transmission during sex and the limited medical harms experienced by most people recently diagnosed with HIV, we recommend that caution be exercised when considering criminal prosecutions, with careful appraisal of current scientific evidence on HIV risk and harms.

HIV science continues to deliver impressive results. During the past decade, a fuller understanding of the effectiveness of HIV antiretroviral treatments as a preventive tool (including treatment of an HIV-infected person, and use of PrEP and PEP) has resulted in a significant decrease in estimates of HIV transmission risk during sexual acts. Similarly, research shows that improved treatments have delivered consistent increases in life expectancy. Given the rapid pace at which science is evolving, reference to risk and harms associated with HIV must reference the most robust and up-to-date evidence.

It has long been recognised that correct use of condoms is an effective means of reducing HIV sexual transmission risk to an acceptably low level. We now know that if an HIV-positive partner is on treatment and maintains an undetectable viral load, or if the HIV-negative partner takes PrEP correctly, risk is reduced to a similar degree. In our opinion, the use of any one of these strategies reduces the risk of transmission to a negligible level and represents taking reasonable precautions to prevent HIV transmission.

In clinical practice, it is extremely unusual to encounter a person who is dismissive of the need to protect others from HIV infection. In the rare instances where this does occur, public health management processes have proven very effective. Public health management guidelines in each state and territory focus on achieving sustained behaviour change through counselling, education and addressing the underlying causes of risk behaviour. Public health officials are well resourced to provide as much support, direction or restriction as required to prevent individuals putting others at risk of HIV infection, including isolation from the community in extreme circumstances.33

Given the effectiveness and lower cost of the public health management approach, and the relatively low per act risk of HIV transmission during sex, we recommend that prosecutorial authorities give consideration to public health management as an alternative to prosecution wherever appropriate.

Box –
Possibility of HIV transmission

Possibility

Conditions for viral transmission present

Reports of transmission (worldwide)

Possibility of transmission per act

Risk of transmission per event

Low

Usually

Most reports linked to these activities

Possible but risk remains low

1.4% (1/70) to 0.04% (1/2500)

Negligible

Very rarely

Isolated reports, difficult to confirm

Highly unlikely, if not impossible in most circumstances

< 0.016% (1/6250) to < 0.0016% (1/62 500)

None

Never

None reported

Not possible

Too low to quantify

News briefs

Can assisted reproduction lower birth defects for older mothers?

Research from the University of Adelaide suggests that babies born to women aged 40 and over from assisted reproduction have fewer birth defects compared with those from women who conceive naturally at the same age. Published in BJOG: An International Journal of Obstetrics and Gynaecology (doi: 10.1111/1471-0528.14365) the research, led by Professor Michael Davies from the Robinson Research Institute, was based on data of all live births recorded in South Australia from 1986-2002, including more than 301 000 naturally conceived births, as well as 2200 births from in vitro fertilisation (IVF) and almost 1400 from intracytoplasmic sperm injection (ICSI). The average prevalence of a birth defect was 5.7% among naturally conceived births, 7.1% for the IVF births, and 9.9% for the ICSI births, across all age groups. In births from assisted reproduction, the prevalence of birth defects ranged from 11.3% at its highest for women less than age 30 using ICSI, down to 3.6% for women aged 40 and older using IVF. For natural conceptions, the corresponding prevalence across age groups was 5.6% in young women, increasing to 8.2% in women aged over 40 years. “There is some aspect of IVF treatment in particular that could be helping older women to redress the maternal age issues we see among natural conception, where we observe a transition at around the age of 35 years toward a steadily increasing risk of birth defects,” Professor Davies said. “We don’t know what that is quite yet – it could be an aspect of hormonal stimulation that helps to reverse the age-related decline in control of ovulation. More research is desperately needed in this area to understand why this is occurring, and whether it could be adapted to both fertile and infertile women in future to prevent birth defects, which continue to be a major cause of death and disability in the first year of life globally.”

Growing association between obesity and cancer

A new literature review by the International Agency for Research on Cancer published in the New England Journal of Medicine (doi: 10.1056/NEJMsr1606602) has concluded that “having lower overall body fat lowers the risk of developing eight tumor types: cancers of the gastric cardia, liver, gallbladder, pancreas, ovaries, and thyroid, in addition to multiple myeloma and meningioma”, according to a commentary published in the Journal of the National Cancer Institute (doi: 10.1093/jnci/djw243). The review brings the list to 13 following the IARC’s 2002 report which found evidence that the risk of being diagnosed with colorectal cancer, adenocarcinoma of the oesophagus, renal cell carcinoma, postmenopausal breast cancer, and uterine cancer was lower among non-overweight people. “The association between obesity and cancer is still not common knowledge. When people think of obesity, they think of diseases such as diabetes and cardiovascular issues,” said one of the IARC authors.

Does Lyme disease exist in Australia?

Despite a number of reports of putative cases and a discussion across several decades,1–19 locally acquired classic Lyme disease has not been identified in Australia. Despite intensive efforts, the bacteria that cause Lyme disease, Borrelia species collectively termed the Borrelia burgdorferi sensu lato (B. burgdorferi s.l.) complex, have not been cultured from any definite locally acquired cases of the disease. Further, Australia does not appear to have a competent tick vector for these species.8,13,18 Finally, bacterial DNA has not been definitively detected in patients for whom acquisition in a country where B. burgdorferi is known to be endemic could be excluded.10,18

The controversy is not restricted to whether B. burgdorferi s.l. and a competent tick vector exist in Australia. We also need to consider whether chronic Lyme disease exists here. This concept does not require the aetiological agent to be metabolically active beyond maintaining a resting metabolism; it need only be present in the patient and viable. Further, the term “chronic Lyme disease” is not consistently defined: it has been applied to patients who present with active, previously untreated B. burgdorferi s.l. infections, to those who have persistent symptoms after being treated for Lyme borreliosis, to people who have had Lyme borreliosis in the past but whose current illness is unrelated to that infection, and to patients without any history of borreliosis. In Australia, substantial numbers of patients without evidence of current or past B. burgdorferi s.l. infection have been labelled with “chronic Lyme” or “Lyme-like disease”, often after bites by Australian ticks. However, even in countries where classic Lyme disease is endemic, the mainstream medical position is that persistence of infection has not been demonstrated in vivo; lingering, non-specific symptoms appear to be post-infectious sequelae unrelated to ongoing active infection.

Internationally, the concept of chronic Lyme disease polarises opinion. In the United States, the key protagonists in the debate are the Infectious Diseases Society of America (IDSA), an association of physicians and medical scientists, and the public advocacy group, the International Lyme and Associated Diseases Society (ILADS).20,21 Consistent with its model of persistent infection, ILADS and practitioners who share its views15,17,20,22 advocate long term treatment with oral antibiotics and sometimes prolonged use of intravenous antibacterial agents and associated complementary therapies, such as probiotics and natural and alternative therapies, for managing the adverse effects of long term antimicrobial administration.

Laboratory diagnosis of Lyme disease remains a significant challenge

Some people believe they have acquired Lyme disease in Australia because the results of screening antibody tests to B. burgdorferi are positive. However, instances in which there was no overseas exposure are all likely to be false positive test results. All diagnostic tests produce both false positive and false negative results; their frequency depends on the specificity and sensitivity of the test and the prevalence of the disease in the population tested. Even a highly specific test will produce some false positives, so that people who have never been exposed to B. burgdorferi can have reactive antibody results.

Diagnosis is further complicated by antigenic variations in the organisms used for developing these assays. Not only are there antigenic differences between B. burgdorferi s.l. species, but many of their genes are differentially expressed in tick and mammal environments. The sensitivity of detection is low early in the infection (less than 50% during the first week), but increases with time. Specific antibodies are usually detectable in people presenting with late manifestations of disease, such as arthritis.21

To improve specificity, a two-tiered serology testing algorithm has been internationally recommended,23,24 including western blot testing with strict interpretative criteria. Second assays are similarly used to maximise specificity when diagnosing human immunodeficiency virus and syphilis infections. Recombinant purified antigens and peptides derived from the bacterial surface lipoprotein VlsE (C6 peptide) have improved the sensitivity of detection of B. burgdorferi tests while maintaining specificity both in screening assays and immunoblots.

The best independent confirmation of any reactive antibody result is demonstrating the microorganism itself. This usually involves culturing the microbe or detecting its genome by polymerase chain reaction (PCR). PCR targeting various gene targets (flaB, 16SrRNA, recA, p66, ospA, 5SrRNA–23SrRNA gene spacer region) can provide highly specific evidence of B. burgdorferi nucleic acid, but the very low organism load means that even the sensitivity of PCR in this context is not great.25 Further, if too many PCR cycles are undertaken, specificity is lost; there is also the possibility of contamination. Culturing the Borrelia spirochaete is difficult. The number of spirochaetes in clinical specimens (skin, blood, cerebrospinal fluid, synovial fluid) is low, and their culture requires special media and prolonged incubation. As a result, antibody tests, despite their problems, remain the primary basis of diagnosis.

If a patient both receives a confirmed reactive antibody result from an appropriately accredited medical testing laboratory and presents with symptoms consistent with the relevant disease, the predictive value of the positive test result increases. However, these conditions do not appear to have been met for any patient diagnosed with possible Lyme disease acquired in Australia. Further, we need to resist requests by patients and others to provide assays that lack clinical validation, including testing urine for B. burgdorferi antigens, lymphocyte transformation tests, and in-house antibody assays that often apply interpretative criteria different to those of commercially approved assays.26

What are the results of antibody testing in Australia?

A large private diagnostic laboratory that conducts about 250 serological tests for Lyme disease each month provided the samples discussed in this article. Referrals came from all Australian states, with most from New South Wales (45% of all tests) and Queensland (27%); women aged 30–50 years were the largest group tested (Box 1, Box 2, Box 3). Over a 23-month period (September 2014 – July 2016), 5372 of 5628 tests (95.5%) in 5395 patients returned negative results (Box 4). Two-tier testing, including an immunoassay followed by an immunoblot, was performed on the 256 samples (4.5%) for which the screening results were equivocal or positive; the western blot results for three-quarters of these samples (177 of 256) were negative, and the screening results were classified as false positive results. Seventy-nine samples (1% of all samples) returned positive results for both the screen immunoassay and an initial immunoblot. Results for a large subset of these patients (29 of 79, or a further 0.5% of all tests) with a low pre-test probability of infection (no relevant symptoms or epidemiological risk factors) were negative on a second immunoblot; the total number of true positive tests was therefore 50 (0.9% of all tests), from a total of 43 patients. The total number of false positive results was 206 of 256 positive screening tests, or 80.5%. These results highlight the fact that commercially available systems use different recombinant antigens and apply different criteria for a positive result. Because of these variations, this second tier of testing should be regarded as an additional test, with a greater emphasis on specificity, supporting the clinical diagnosis rather than confirming it.27

To minimise the risk of a false positive result, tests should be requested only if there is a well founded clinical suspicion of Lyme disease, and not in situations of low pre-test probability.23 A travel history was available for 37 of the 43 patients with true positive results; all had returned from countries in which Lyme disease is endemic. Most Lyme disease acquired overseas but diagnosed in Australia was European in origin (30 of 43, or 70% of cases; Box 5).

Background to the belief that Lyme disease exists in Australia

Since the early 1990s, the Australian medical community, especially specialist microbiologists and infectious diseases physicians, have debated whether an indigenous form of classic Lyme diseases occurs in Australia, especially in areas with high rates of tick bites.2–4,8–12,15,17,18 This interest motivated some of the early tick surveys.2,17 In 1991, B. burgdorferi s.l. could not be confirmed in any of 176 tick species examined.2,17 The findings of more recent surveys have also been negative.16,19

Native and introduced animals have also been investigated. The presence of Borrelia spp. in native fauna was reported by Mackerras1 and Carley and Pope2 during the 1950s and 1960s. A human volunteer was inoculated with B. queenslandica, but without causing disease,2 and it is very unlikely that B. queenslandica can induce an illness like classic Lyme disease.

The most frequently cited Borrelia species identified in introduced fauna are B. theileri (bovine borreliosis) and B. anserina (avian spirochaetosis).28 Neither of these bacteria causes a disease in humans consistent with classic Lyme disease or a chronic debilitating illness that manifests as a constellation of chronic non-specific symptoms. Neither species is part of the B. burgdorferi s.l. complex as currently defined.

Some medical practitioners in Australia became aware of Lyme disease soon after this emerging disease was described in the northeastern United States. Travellers from endemic areas were diagnosed with Lyme disease when serological tests became available. As the controversy in the US about chronic Lyme disease intensified, patients in Australia began presenting with non-specific symptoms that they related to the putative disorder, such as chronic fatigue, cognitive impairment, myalgias, and arthralgias. These patients were often clustered around a small number of general practitioners who, assessing their symptoms as being consistent with chronic Lyme disease, requested laboratory testing. Most tests undertaken in Australian laboratories returned negative results; specimens were then frequently sent to overseas laboratories, often to facilities describing themselves as being specialised for Lyme and associated diseases. Some of these laboratories reported positive results, interpreted by the treating medical practitioner as confirming their clinical diagnosis of chronic Lyme disease.

As the volume of information shared by patients and their supporters, both medical and non-medical, increased in Australia, their numbers also grew significantly. Online sharing of information via social media and digital forums has brought together thousands of Australians who believe they have Lyme disease. There are now patient advocacy groups in most states and territories, the two most prominent being the Lyme Disease Association of Australia and the Karl McManus Foundation.

Given the lack of evidence that Australia has either the aetiological agent or competent vector required for classic Lyme disease, many advocates have adopted the new label, “Lyme disease-like illness”.29 The problem with this term is that it suggests that chronic Lyme disease is a recognised medical diagnosis, whereas its validity remains contentious. Another description used is “multi-systemic infectious diseases syndrome” (MSIDS), despite the fact that it has not been established that the illness denoted by this term is infectious, nor that its constellation of non-specific symptoms is post-infectious. Environmental toxins and psychological bases have not been excluded as explanations. Moreover, many patients are initially diagnosed with neurological disorders, including motor neurone disease, Parkinson disease, multiple sclerosis and Alzheimer disease, and some advocates claim that these chronic neurological conditions are also caused by Lyme borreliosis.30,31

Clinicians and medical scientists in Australia are able to identify and fully characterise unexpected or previously unknown pathogens, as evidenced by the recent report of a Babesia infection.32 Western Australian researchers recently described a novel Borrelia species in Australian ticks,16,19,33 and many in the “Lyme disease” community are interested in this novel bacterium and the possibility that it might cause their illness. It has, however, not yet been shown to be pathogenic. Many patients and their medical practitioners believe that further pathogens play a significant role in “Lyme disease”, and that co-infection by Babesia, Anaplasma, Bartonella and Ehrlichia species are commonplace.17,20 Although these pathogens are carried by ticks, tick-borne infections caused by these microorganisms are not often diagnosed in Australia. To further confound matters, some chronic Lyme disease advocates argue that ticks are not the only vector for infection, with mosquitoes, midges, sand flies, and even leeches cited as potential alternatives.

While there is no compelling evidence that classic Lyme disease exists in Australia, Australians do acquire unusual novel infections32,33 as well as more commonly recognised infections from ticks (eg, rickettsiosis). Some may have illnesses caused by tick-borne bacteria or viruses that are yet to be identified but which may be widely distributed in Australia.

The Senate inquiry

A recent Senate inquiry raised questions about the epidemiology, diagnosis, treatment, and investigation of Lyme disease. The inquiry began on 12 November 2015, and, because of the double dissolution election on 2 July 2016, the committee issued an interim report on 4 May 2016.29 The inquiry was re-adopted by the 45th Parliament of Australia on 13 September 2016, with a reporting date of 30 November 2016.

The stigma felt by patients with a chronic debilitating illness was a major focus of the Senate inquiry. We believe that the most effective way to reduce this stigma is to concentrate appropriately funded research and clinical studies on discovering any bacterial or viral causes for such a chronic debilitating illness. This is best achieved with a One Health approach incorporating disease management elements of human, animal and environmental health care, investigating not only human infections, but also animal reservoirs and vectors that might harbour microorganisms which cause illness in people or animals.

The reported chronic debilitating symptoms overlap to a considerable degree with those of chronic fatigue syndrome (CFS) and related disorders. In most patients diagnosed with CFS, the exact cause is unknown, although Epstein-Barr virus, cytomegalovirus, and other viruses have all been implicated in syndromes characterised by fatigue-like symptoms that can persist for more than 6 months.

If we focus attention on patients with constellations of chronic debilitating symptoms in Australia, diagnostic assessment will initially include a research component (eg, metagenomic analysis, employing next generation sequencing). This will be of great benefit for identifying the many uncharacterised viruses and bacteria that probably occur in ticks, animals and people in Australia, and should inform diagnostic methods that can be included in the scope of accreditation for medical testing laboratories.

What are the problems for patients, the Australian community and doctors?

The dangers of intravenous therapy

It is vital that medical practitioners not do more harm than good with any intervention they undertake. While the questions of whether persistent B. burgdorferi infection occurs and classic Lyme disease exists in Australia can be debated, there is no denying that receiving intravascular antibiotics is associated with the risk of bloodstream infections by bacteria and fungi, in many cases fatal,34–36 including in some people treated for “Lyme disease”.36 The mortality rate associated with Staphylococcus aureus bloodstream infections is 15–35%.35 The major medical cost of prolonged intravenous therapy is thus accompanied by social costs to the patient, who should therefore not receive prolonged intravenous therapy if there is no clear evidence that it will be beneficial.

Antibiotic overuse and misuse

Many people who believe they have Lyme disease or Lyme disease-like illness, as well as some of their medical practitioners, also believe that prolonged antibiotic therapy, including intravenous antibiotics, may cure their disease.20 Evidence from the US and Europe, where classic Lyme disease is endemic, do not confirm this view. In particular, prolonged intravenous antibiotic therapy (longer than one month)37–40 does not seem to significantly improve symptoms.

Antibiotic resistance resulting from the unnecessary and prolonged use of broad spectrum antibiotics (eg, ceftriaxone) is a major problem. Antibiotic resistance not only harms the person receiving the agent (who will often be colonised by more resistant bacteria) but also the broader community: when resistant bacteria develop or multiply in an individual, they can be spread to family members and to the wider public.

Advocates of long term antimicrobial therapy point to acne, tuberculosis and Hansen disease as examples of diseases in which long term antibiotic therapy has not raised the ire of experts concerned with antimicrobial resistance. This, however, is not true, especially with respect to acne.41 Further, advocates of long term antibiotic therapy for “Lyme disease” do not appreciate that generalisations cannot be made when treating infections caused by different genera and species of bacteria.

Other potential hazards of taking antibiotics unnecessarily include their toxicity, potential hypersensitivity reactions and even anaphylaxis (allergy), and predisposition to infection with Clostridium difficile and antibiotic-resistant bacteria.42

Dilemmas for Australian doctors

Australian medical practitioners are faced with a difficult dilemma. Growing numbers of patients, their supporters, some integrative medical practitioners, and politicians are demanding diagnoses and treatment according to the protocols of the “chronic Lyme disease” school of thought. This situation is reminiscent of campaigns by advocacy groups, who mounted pressure on legislators and policy makers in the US and Canada, leading to the introduction of parliamentary bills.43,44 Both chambers of our parliament29,45 have issued reports that include recommendations for the investigation of and research into “chronic Lyme disease”, as well as for developing a case definition of this debilitating illness.

However, evidence of a local aetiological agent and a competent vector that together could cause classic Lyme disease in Australia has yet to be reported. Novel viruses and bacteria have recently been discovered in Australian ticks.16,19,25,32,46 Further research, using next generation sequencing and metagenomics, may identify links between specific tick-borne pathogens and patients who present with this constellation of chronic, non-specific symptoms, and it is only after such a link has been established that effective, evidence-based management protocols can be safely developed.

Conclusion

With the advent of new molecular scientific tools, it would not be astonishing were new viral or bacterial pathogens discovered, in ticks and other vectors, that cause acute or chronic infections or symptoms in humans in Australia. However, there is no convincing evidence that classic Lyme disease occurs here; in particular, there is no evidence that a species from the B. burgdorferi s.l. complex occurs in Australian animals or ticks. Until there is strong evidence from well performed clinical studies that bacteria present in Australia cause a chronic debilitating illness that responds to extended antibiotic therapy, treating patients with “Lyme disease-like illness” with prolonged intravenous or oral antibiotic therapy is both unjustifiable and unethical, and is likely to do much more harm than good. This harm affects both the individual (eg, intravenous sepsis) and the Australian community (increased antimicrobial resistance rates).

While immediate treatment solutions for patients presenting with these symptoms are not readily available, we strongly recommend a multidisciplinary approach to the investigation, diagnosis and management of “chronic Lyme disease”. This is especially important for ensuring that alternative diagnoses are not missed or ignored. Engaging with general practitioners, specialist physicians and microbiologists is critical for helping each patient.

Box 1 –
Lyme antibody test requests by month and year, Australia, September 2014 – July 2016*

Source: Sonic Pathology (Sullivan Nicolaides Pathology, Brisbane). * Total number of requests: 5628; total number of patients: 5395.

Box 2 –
Origin of 5628 Lyme antibody test requests, Australia, September 2014 – July 2016, by state and territory

OS = overseas. * Based on 2015 Australian Bureau of Statistics population data.

Box 3 –
Lyme antibody test requests, Australia, September 2014 – July 2016, by age group and sex of patients*

* Total number of requests: 5628; total number of patients: 5395.

Box 4 –
The results of Lyme disease antibody test requests, Australia, September 2014 – July 2016

In two-tier testing (2TT), the screening assay used was the recombinant VlsE-based Liaison chemiluminescence immunoassay for Borrelia burgdorferi (IgG) (DiaSorin); the second tier test employed was the anti-Borrelia burgdorferi IgG Euroline-RN-AT immunoblot system (Euroimmun). For the second series of immunoblot tests (3TT; performed on a subset of mostly low pre-test probability specimens), an in-house IgG western immunoblot for B. burgdorferi and B. afzelii, or the MarDx IgG EU Lyme (B. afzelii, B. garinii) &plus; VlsE western blot system (Trinity Biotech) was used. 3TT was performed at Pathology West – ICPMR Westmead, Sydney.

Box 5 –
Travel history for 43 patients with positive serological test results for Lyme disease (3-tier testing)

[Review] Impact of air pollution on the burden of chronic respiratory diseases in China: time for urgent action

In China, where air pollution has become a major threat to public health, public awareness of the detrimental effects of air pollution on respiratory health is increasing—particularly in relation to haze days. Air pollutant emission levels in China remain substantially higher than are those in developed countries. Moreover, industry, traffic, and household biomass combustion have become major sources of air pollutant emissions, with substantial spatial and temporal variations. In this Review, we focus on the major constituents of air pollutants and their impacts on chronic respiratory diseases.

[Editorial] WHO’s war on sugar

On Oct 11, World Obesity Day, WHO upped the ante in its fight against sugar. First, it called for governments to introduce subsidies for fruits and vegetables and taxation of unhealthy foods, with a particular target on sugary drinks. The new WHO recommendations are based on global expert opinion and 11 systematic reviews of the effectiveness of fiscal interventions for improving diets and preventing non-communicable diseases (NCDs). The second move saw the removal of sales and provision of sugary drinks from WHO headquarters, including at official functions.