Modelling African swine fever introduction in diverse Australian feral pig populations

African swine fever (ASF) is a viral disease that affects domestic and feral pigs. While not currently present in Australia, ASF outbreaks have been reported nearby in Indonesia, Timor-Leste, and Papua New Guinea. Feral pigs are found in all Australian states and territories and are distributed in a variety of habitats. To investigate the impacts of an ASF introduction event in Australia, we used a stochastic network-based metapopulation feral pig model to simulate ASF outbreaks in different regions of Australia. Outbreak intensity and persistence in feral pig populations was governed by local pig recruitment rates, population size, carcass decay period, and, if applicable, metapopulation topology. In Northern Australia, the carcass decay period was too short for prolonged persistence, while endemic transmission could possibly occur in cooler southern areas. Populations in Macquarie Marshes in New South Wales and in Namadgi National Park in the Australian Capital Territory had the highest rates of persistence. The regions had different modes of transmission that led to long-term persistence. Endemic Macquarie Marshes simulations were characterised by rapid transmission caused by high population density that required a fragmented metapopulation to act as a bottleneck to slow transmission. Endemic simulations in Namadgi, with low density and relatively slow transmission, relied on large, well-connected populations coupled with long carcass decay times. Despite the potential for endemic transmission, both settings required potentially unlikely population sizes and dynamics for prolonged disease survival.

Source attribution of campylobacteriosis in Australia, 2017-2019

Campylobacter jejuni and Campylobacter coli infections are the leading cause of foodborne gastroenteritis in high-income countries. Campylobacter colonizes a variety of warm-blooded hosts that are reservoirs for human campylobacteriosis. The proportions of Australian cases attributable to different animal reservoirs are unknown but can be estimated by comparing the frequency of different sequence types in cases and reservoirs. Campylobacter isolates were obtained from notified human cases and raw meat and offal from the major livestock in Australia between 2017 and 2019. Isolates were typed using multi-locus sequence genotyping. We used Bayesian source attribution models including the asymmetric island model, the modified Hald model, and their generalizations. Some models included an "unsampled" source to estimate the proportion of cases attributable to wild, feral, or domestic animal reservoirs not sampled in our study. Model fits were compared using the Watanabe-Akaike information criterion. We included 612 food and 710 human case isolates. The best fitting models attributed >80% of Campylobacter cases to chickens, with a greater proportion of C. coli (>84%) than C. jejuni (>77%). The best fitting model that included an unsampled source attributed 14% (95% credible interval [CrI]: 0.3%-32%) to the unsampled source and only 2% to ruminants (95% CrI: 0.3%-12%) and 2% to pigs (95% CrI: 0.2%-11%) The best fitting model that did not include an unsampled source attributed 12% to ruminants (95% CrI: 1.3%-33%) and 6% to pigs (95% CrI: 1.1%-19%). Chickens were the leading source of human Campylobacter infections in Australia in 2017-2019 and should remain the focus of interventions to reduce burden.

The Cost of Foodborne Illness and Its Sequelae in Australia Circa 2019

Foodborne illnesses cause a significant health burden, with Campylobacter and norovirus the most common causes of illness and Salmonella a common cause of hospitalization and occasional cause of death. Estimating the cost of illness can assist in quantifying this health burden, with pathogen-specific costs informing prioritization of interventions. We used a simulation-based approach to cost foodborne disease in Australia, capturing the cost of premature mortality, direct costs of nonfatal illness (including health care costs, medications, and tests), indirect costs of illness due to lost productivity, and costs associated with pain and suffering. In Australia circa 2019, the cost in Australian Dollars (AUD) of foodborne illness and its sequelae was 2.44 billion (90% uncertainty interval 1.65-3.68) each year, with the highest pathogen-specific costs for Campylobacter, non-typhoidal Salmonella, non-Shiga toxin-producing pathogenic Escherichia coli, and norovirus. The highest cost per case was for Listeria monocytogenes (AUD 776,000). Lost productivity was the largest component cost for foodborne illness due to all causes and for most individual pathogens; the exceptions were pathogens causing more severe illness such as Salmonella and L. monocytogenes, where premature mortality was the largest component cost. Foodborne illness results in a substantial cost to Australia; interventions to improve food safety across industry, retail, and consumers are needed to maintain public health safety.

Lymphatic filariasis endgame strategies: Using GEOFIL to model mass drug administration and targeted surveillance and treatment strategies in American Samoa

American Samoa underwent seven rounds of mass drug administration (MDA) for lymphatic filariasis (LF) from 2000-2006, but subsequent surveys found evidence of ongoing transmission. American Samoa has since undergone further rounds of MDA in 2018, 2019, and 2021; however, recent surveys indicate that transmission is still ongoing. GEOFIL, a spatially-explicit agent-based LF model, was used to compare the effectiveness of territory-wide triple-drug MDA (3D-MDA) with targeted surveillance and treatment strategies. Both approaches relied on treatment with ivermectin, diethylcarbamazine, and albendazole. We simulated three levels of whole population coverage for 3D-MDA: 65%, 73%, and 85%, while the targeted strategies relied on surveillance in schools, workplaces, and households, followed by targeted treatment. In the household-based strategies, we simulated 1-5 teams travelling village-to-village and offering antigen (Ag) testing to randomly selected households in each village. If an Ag-positive person was identified, treatment was offered to members of all households within 100m-1km of the positive case. All simulated interventions were finished by 2027 and their effectiveness was judged by their 'control probability'-the proportion of simulations in which microfilariae prevalence decreased between 2030 and 2035. Without future intervention, we predict Ag prevalence will rebound. With 3D-MDA, a 90% control probability required an estimated ≥ 4 further rounds with 65% coverage, ≥ 3 rounds with 73% coverage, or ≥ 2 rounds with 85% coverage. While household-based strategies were substantially more testing-intensive than 3D-MDA, they could offer comparable control probabilities with substantially fewer treatments; e.g. three teams aiming to test 50% of households and offering treatment to a 500m radius had approximately the same control probability as three rounds of 73% 3D-MDA, but used < 40% the number of treatments. School- and workplace-based interventions proved ineffective. Regardless of strategy, reducing Ag prevalence below the 1% target threshold recommended by the World Health Organization was a poor indicator of the interruption of LF transmission, highlighting the need to review blanket elimination targets.

Source attribution of salmonellosis by time and geography in New South Wales, Australia

BACKGROUND

Salmonella is a major cause of zoonotic illness around the world, arising from direct or indirect contact with a range of animal reservoirs. In the Australian state of New South Wales (NSW), salmonellosis is believed to be primarily foodborne, but the relative contribution of animal reservoirs is unknown.

METHODS

The analysis included 4543 serotyped isolates from animal reservoirs and 30,073 serotyped isolates from domestically acquired human cases in NSW between January 2008 and August 2019. We used a Bayesian source attribution methodology to estimate the proportion of foodborne Salmonella infections attributable to broiler chickens, layer chickens, ruminants, pigs, and an unknown or unsampled source. Additional analyses included covariates for four time periods and five levels of rurality.

RESULTS

A single serotype, S. Typhimurium, accounted for 65-75% of included cases during 2008-2014 but < 50% during 2017-2019. Attribution to layer chickens was highest during 2008-2010 (48.7%, 95% CrI 24.2-70.3%) but halved by 2017-2019 (23.1%, 95% CrI 5.7-38.9%) and was lower in the rural and remote populations than in the majority urban population. The proportion of cases attributed to the unsampled source was 11.3% (95% CrI 1.2%-22.1%) overall, but higher in rural and remote populations. The proportion of cases attributed to pork increased from approximately 20% in 2009-2016 to approximately 40% in 2017-2019, coinciding with a rise in cases due to Salmonella ser. 4,5,12:i:-.

CONCLUSION

Layer chickens were likely the primary reservoir of domestically acquired Salmonella infections in NSW circa 2010, but attribution to the source declined contemporaneously with increased vaccination of layer flocks and tighter food safety regulations for the handling of eggs.

Has the effectiveness of Australia's travel bans against China on the importation of COVID-19 been overestimated?

In our re-analysis with adjusted assumptions, we found that a previous study substantially oversestimated the effectiveness of Australia’s complete travel ban against China in the introduction of SARS-CoV-2 into Australia.

Molecular characterization of Campylobacter spp. recovered from beef, chicken, lamb and pork products at retail in Australia

Australian rates of campylobacteriosis are among the highest in developed countries, yet only limited work has been done to characterize Campylobacter spp. in Australian retail products. We performed whole genome sequencing (WGS) on 331 C. coli and 285 C. jejuni from retail chicken meat, as well as beef, chicken, lamb and pork offal (organs). Campylobacter isolates were highly diverse, with 113 sequence types (STs) including 38 novel STs, identified from 616 isolates. Genomic analysis suggests very low levels (2.3-15.3%) of resistance to aminoglycoside, beta-lactam, fluoroquinolone, macrolide and tetracycline antibiotics. A majority (>90%) of isolates (52/56) possessing the fluoroquinolone resistance-associated T86I mutation in the gyrA gene belonged to ST860, ST2083 or ST7323. The 44 pork offal isolates were highly diverse, representing 33 STs (11 novel STs) and harboured genes associated with resistance to aminoglycosides, lincosamides and macrolides not generally found in isolates from other sources. Prevalence of multidrug resistant genotypes was very low (<5%), but ten-fold higher in C. coli than C. jejuni. This study highlights that Campylobacter spp. from retail products in Australia are highly genotypically diverse and important differences in antimicrobial resistance exist between Campylobacter species and animal sources.

Some simple rules for estimating reproduction numbers in the presence of reservoir exposure or imported cases

For many diseases, the basic reproduction number (R0) is a threshold parameter for disease extinction or survival in isolated populations. However no human population is fully isolated from other human or animal populations. We use compartmental models to derive simple rules for the basic reproduction number in populations where an endemic disease is sustained by a combination of local transmission within the population and exposure from some other source: either a reservoir exposure or imported cases. We introduce the idea of a reservoir-driven or importation-driven disease: diseases that would become extinct in the population of interest without reservoir exposure or imported cases (since R0<1), but nevertheless may be sufficiently transmissible that many or most infections are acquired from humans in that population. We show that in the simplest case, R0<1 if and only if the proportion of infections acquired from the external source exceeds the disease prevalence and explore how population heterogeneity and the interactions of multiple strains affect this rule. We apply these rules in two case studies of Clostridium difficile infection and colonisation: C. difficile in the hospital setting accounting for imported cases, and C. difficile in the general human population accounting for exposure to animal reservoirs. We demonstrate that even the hospital-adapted, highly-transmissible NAP1/RT027 strain of C. difficile had a reproduction number <1 in a landmark study of hospitalised patients and therefore was sustained by colonised and infected admissions to the study hospital. We argue that C. difficile should be considered reservoir-driven if as little as 13.0% of transmission can be attributed to animal reservoirs.

Seasonality and community interventions in a mathematical model of Clostridium difficile transmission

BACKGROUND

Clostridium difficile infection (CDI) is the leading cause of antibiotic-associated diarrhoea with peak incidence in late winter or early autumn. Although CDI is commonly associated with hospitals, community transmission is important.

AIM

To explore potential drivers of CDI seasonality and the effect of community-based interventions to reduce transmission.

METHODS

A mechanistic compartmental model of C. difficile transmission in a hospital and surrounding community was used to determine the effect of reducing transmission or antibiotic prescriptions in these settings. The model was extended to allow for seasonal antibiotic prescriptions and seasonal transmission.

FINDINGS

Modelling antibiotic seasonality reproduced the seasonality of CDI, including approximate magnitude (13.9-15.1% above annual mean) and timing of peaks (0.7-1.0 months after peak antibiotics). Halving seasonal excess prescriptions reduced the incidence of CDI by 6-18%. Seasonal transmission produced larger seasonal peaks in the prevalence of community colonization (14.8-22.1% above mean) than seasonal antibiotic prescriptions (0.2-1.7% above mean). Reducing transmission from symptomatic or hospitalized patients had little effect on community-acquired CDI, but reducing transmission in the community by ≥7% or transmission from infants by ≥30% eliminated the pathogen. Reducing antibiotic prescription rates led to approximately proportional reductions in infections, but limited reductions in the prevalence of colonization.

CONCLUSION

Seasonal variation in antibiotic prescription rates can account for the observed magnitude and timing of C. difficile seasonality. Even complete prevention of transmission from hospitalized patients or symptomatic patients cannot eliminate the pathogen, but interventions to reduce transmission from community residents or infants could have a large impact on both hospital- and community-acquired infections.

Clostridium difficile classification overestimates hospital-acquired infections

BACKGROUND

Clostridium difficile infections occur frequently among hospitalized patients, with some infections acquired in hospital and others in the community. International guidelines classify cases as hospital-acquired if symptom onset occurs more than two days after admission. This classification informs surveillance and infection control, but has not been verified by empirical or modelling studies.

AIM

To assess current classification of C. difficile acquisition using a simulation model as a reference standard.

METHODS

C. difficile transmission was simulated in a range of hospital scenarios. The sensitivity, specificity and precision of classifications that use cut-offs ranging from 0.25 h to 40 days were calculated. The optimal cut-off that correctly estimated the proportion of cases that were hospital acquired and the balanced cut-off that had equal sensitivity and specificity were identified.

FINDINGS

The recommended two-day cut-off overestimated the incidence of hospital-acquired cases in all scenarios and by >100% in the base scenario. The two-day cut-off had good sensitivity (96%) but poor specificity (48%) and precision (52%) to identify cases acquired during the current hospitalization. A five-day cut-off was balanced, and a six-day cut-off was optimal in the base scenario. The optimal and balanced cut-offs were more than two days for nearly all scenarios considered (ranges: four to nine days and two to eight days, respectively).

CONCLUSION

Current guidelines for classifying C. difficile infections overestimate the proportion of cases acquired in hospital in all model scenarios. To reduce misclassification bias, an infection should be classified as being acquired prior to admission if symptoms begin within five days of admission.