Revealing the complexity of health determinants in resource-poor settings

Antimicrobial resistance in Escherichia coli from Swedish piglets with diarrhoea and associations with potential risk factors

BACKGROUND

Antibiotic treatments of diarrhoea in suckling piglets and in pigs after weaning are common worldwide and contribute to antimicrobial resistance (AMR) in Escherichia coli from pigs. In Sweden, during the last decades, resistance to trimethoprim-sulphonamide and ampicillin has increased markedly in E. coli from routine clinical samples from piglets with diarrhoea, hereafter referred to as "clinical submissions". This has occurred despite a comparatively low use of antibiotics in Swedish pig production. However, clinical submissions might be biased towards farms with treatment failures and therefore overestimate occurrence of AMR. To explore the representativeness of data from such samples we compared occurrence of AMR in E. coli from clinical submissions and from concurrent samples collected from piglets with diarrhoea by convenience, referred to as "study samples". We also investigated associations between farm-related potential risk factors and AMR using farm data collected through a questionnaire. Data were evaluated using univariable and multivariable statistical models, as well as a multivariate model.

RESULTS

In all, 158 study samples from 97 herds and questionnaires from 83 herds were analysed. Resistance to streptomycin (37%), trimethoprim-sulphonamide (32%), ampicillin (30%), and tetracycline (18%) were the most frequent traits. Occurrence of AMR in 158 E. coli isolates from study samples was not significantly different from occurrence in 57 isolates from concurrent clinical submissions (P > 0.05). In 70% of herds, more than 10% of the sows were treated with antibiotics in the first week after farrowing, and trimethoprim-sulphonamide was the most common first choice antibiotic. Trimethoprim-sulphonamide resistance was associated with the proportion of sows receiving post-farrowing treatment. Resistance to ampicillin, tetracycline, and streptomycin resistances were indirectly associated with sow treatments, likely via co-resistance to trimethoprim-sulphonamide. There was no significant association between high dose zinc oxide supplementation and AMR (P > 0.05).

CONCLUSIONS

Clinical submissions do not overestimate occurrence of AMR in E. coli from Swedish piglets with diarrhoea and are therefore relevant for AMR monitoring. Even at low treatment rates, post-farrowing treatment of sows increases the risk for AMR in piglets. This applies especially for trimethoprim-sulphonamide resistance, but also for resistance to other antibiotics, and indicates that antibiotic use must be reduced substantially to achieve a reduction of AMR.

Contribution of Structure Learning Algorithms in Social Epidemiology: Application to Real-World Data

Epidemiologists often handle large datasets with numerous variables and are currently seeing a growing wealth of techniques for data analysis, such as machine learning. Critical aspects involve addressing causality, often based on observational data, and dealing with the complex relationships between variables to uncover the overall structure of variable interactions, causal or not. Structure learning (SL) methods aim to automatically or semi-automatically reveal the structure of variables' relationships. The objective of this study is to delineate some of the potential contributions and limitations of structure learning methods when applied to social epidemiology topics and the search for determinants of healthcare system access. We applied SL techniques to a real-world dataset, namely the 2010 wave of the SIRS cohort, which included a sample of 3006 adults from the Paris region, France. Healthcare utilization, encompassing both direct and indirect access to care, was the primary outcome. Candidate determinants included health status, demographic characteristics, and socio-cultural and economic positions. We present two approaches: a non-automated epidemiological method (an initial expert knowledge network and stepwise logistic regression models) and three SL techniques using various algorithms, with and without knowledge constraints. We compared the results based on the presence, direction, and strength of specific links within the produced network. Although the interdependencies and relative strengths identified by both approaches were similar, the SL algorithms detect fewer associations with the outcome than the non-automated method. Relationships between variables were sometimes incorrectly oriented when using a purely data-driven approach. SL algorithms can be valuable in exploratory stages, helping to generate new hypotheses or mining novel databases. However, results should be validated against prior knowledge and supplemented with additional confirmatory analyses.

Unique duck rearing practice in irrigated rice paddy fields driving recurrent H5N1 avian influenza outbreaks in two districts of Kerala, India

Highly pathogenic avian influenza (HPAI) outbreaks have repeatedly occurred in two districts of Kerala state, India, over the last few years. The outbreaks in the wetland areas coincided with the arrival of migratory birds. At the time, the factors responsible for local transmission in ducks were not known. This study aimed to identify the socio-economic factors responsible for spatial variation in the occurrence of HPAI outbreaks in the two districts using Bayesian network modelling (BNM) and Stochastic Partial Differential Equation (SPDE) model. Further, information was collected on the duck rearing practices in rice paddy fields to identify the risk factors for local - spread of the outbreaks. We found that the SPDE model without covariates explained variation in occurrence of outbreaks. The number of rice paddy fields used by the duck farmers was identified as risk factor. We concluded based on BNM and SPDE that the infected migratory birds were the source of infection for the first few duck farms in the wetland areas and subsequent transmission was driven by shifting of ducks from one rice paddy field to other fields. There is a probability of persistent and recurrent infections in the ducks and possible spill over to humans. Hence, it is important to have surveillance in ducks to prevent recurrent outbreaks in the region.

Multivariate variable selection in N-of-1 observational studies via additive Bayesian networks

An N-of-1 observational design characterizes associations among several variables over time in a single individual. Traditional statistical models recommended for experimental N-of-1 trials may not adequately model these observational relationships. We propose an additive Bayesian network using a generalized linear mixed-effects model for the local mean as a novel method for modeling each of these relationships in a data-driven manner. We validate our approach via simulation studies and apply it to a 12-month observational N-of-1 study exploring the impact of stress on daily exercise engagement. We demonstrate the improved performance of the additive Bayesian network to recover the underlying network structure. From the empirical study, we found statistically discernible associations between reports of stress and physical activity on a population level, but these associations may differ at an individual level.

Analysis of the Associations of Measurements of Body Composition and Inflammatory Factors with Cardiovascular Disease and Its Comorbidities in a Community-Based Study

The associations of cardiovascular disease (CVD) with comorbidities and biochemical and body composition measurements are repeatedly described but have not been studied simultaneously. In the present cross-sectional study, information on CVD and comorbidities [type 2 diabetes mellitus (T2DM), hypertension (HTN), and hyperlipidemia (HDL)], body composition, levels of soluble markers, and other measures were collected from 1079 individuals. When we examined the association of each comorbidity and CVD, controlling for other comorbidities, we observed a clear pattern of the comorbidity-related specific associations with tested covariates. For example, T2DM was significantly associated with GDF-15 levels and the leptin/adiponectin (L/A) ratio independently of two other comorbidities; HTN, similarly, was independently associated with extracellular water (ECW) levels, L/A ratio, and age; and HDL was independently related to age only. CVD showed very strong independent associations with each of the comorbidities, being associated most strongly with HTN (OR = 10.89, 6.46-18.38) but also with HDL (2.49, 1.43-4.33) and T2DM (1.93, 1.12-3.33). An additive Bayesian network analysis suggests that all three comorbidities, particularly HTN, GDF-15 levels, and ECW content, likely have a main role in the risk of CVD development. Other factors, L/A ratio, lymphocyte count, and the systemic inflammation response index, are likely indirectly related to CVD, acting through the comorbidities and ECW.

Relationships between Circulating Biomarkers and Body Composition Parameters in Patients with Metabolic Syndrome: A Community-Based Study

Metabolic syndrome (MetS) is a complex disease involving multiple physiological, biochemical, and metabolic abnormalities. The search for reliable biomarkers may help to better elucidate its pathogenesis and develop new preventive and therapeutic strategies. In the present population-based study, we looked for biomarkers of MetS among obesity- and inflammation-related circulating factors and body composition parameters in 1079 individuals (with age range between 18 and 80) belonging to an ethnically homogeneous population. Plasma levels of soluble markers were measured by using ELISA. Body composition parameters were assessed using bioimpedance analysis (BIA). Statistical analysis, including mixed-effects regression, with MetS as a dependent variable, revealed that the most significant independent variables were mainly adipose tissue-related phenotypes, including fat mass/weight (FM/WT) [OR (95% CI)], 2.77 (2.01-3.81); leptin/adiponectin ratio (L/A ratio), 1.50 (1.23-1.83); growth and differentiation factor 15 (GDF-15) levels, 1.32 (1.08-1.62); inflammatory markers, specifically monocyte to high-density lipoprotein cholesterol ratio (MHR), 2.53 (2.00-3.15), and a few others. Additive Bayesian network modeling suggests that age, sex, MHR, and FM/WT are directly associated with MetS and probably affect its manifestation. Additionally, MetS may be causing the GDF-15 and L/A ratio. Our novel findings suggest the existence of complex, age-related, and possibly hierarchical relationships between MetS and factors associated with obesity.

Identifying risk factors of developing type 2 diabetes from an adult population with initial prediabetes using a Bayesian network

Background

It is known that people with prediabetes increase their risk of developing type 2 diabetes (T2D), which constitutes a global public health concern, and it is associated with other diseases such as cardiovascular disease.

Methods

This study aimed to determine those factors with high influence in the development of T2D once prediabetes has been diagnosed, through a Bayesian network (BN), which can help to prevent T2D. Furthermore, the set of features with the strongest influences on T2D can be determined through the Markov blanket. A BN model for T2D was built from a dataset composed of 12 relevant features of the T2D domain, determining the dependencies and conditional independencies from empirical data in a multivariate context. The structure and parameters were learned with the bnlearn package in R language introducing prior knowledge. The Markov blanket was considered to find those features (variables) which increase the risk of T2D.

Results

The BN model established the different relationships among features (variables). Through inference, a high estimated probability value of T2D was obtained when the body mass index (BMI) was instantiated to obesity value, the glycosylated hemoglobin (HbA1c) to more than 6 value, the fatty liver index (FLI) to more than 60 value, physical activity (PA) to no state, and age to 48-62 state. The features increasing T2D in specific states (warning factors) were ranked.

Conclusion

The feasibility of BNs in epidemiological studies is shown, in particular, when data from T2D risk factors are considered. BNs allow us to order the features which influence the most the development of T2D. The proposed BN model might be used as a general tool for prevention, that is, to improve the prognosis.

Genetic and environmental correlational structure among metabolic syndrome endophenotypes

Metabolic syndrome (MetS) is diagnosed by the presence of high scores on three or more metabolic traits, including systolic and diastolic blood pressure (SBP, DBP), glucose and insulin levels, cholesterol and triglyceride (TG) levels, and central obesity. A diagnosis of MetS is associated with increased risk of cardiovascular disease and type 2 diabetes. The components of MetS have long been demonstrated to have substantial genetic components, but their genetic overlap is less well understood. The present paper takes a multi-prong approach to examining the extent of this genetic overlap. This includes the quantitative genetic and additive Bayesian network modeling of the large TwinsUK project and examination of the results of genome-wide association study (GWAS) of UK Biobank data through use of LD score regression and examination of the number of genes and pathways identified in the GWASes which overlap across MetS traits. Results demonstrate a modest genetic overlap, and the genetic correlations obtained from TwinsUK and UK Biobank are nearly identical. However, these correlations imply more genetic dissimilarity than similarity. Furthermore, examination of the extent of overlap in significant GWAS hits, both at the gene and pathway level, again demonstrates only modest but significant genetic overlap. This lends support to the idea that in clinical treatment of MetS, treating each of the components individually may be an important way to address MetS.

Additive Bayesian network analysis of the relationship between bovine respiratory disease and management practices in dairy heifer calves at pre-weaning stage

Background

Nursery farms that accept nursing and growing pre-weaned heifer calves from private dairy farms must work to prevent bovine respiratory disease (BRD). Knowledge of the BRD-associated risk factors related to calf management and calves’ condition will help to develop appropriate neonatal management practices at original farms and to identify calves at higher risk for BRD at nursery farms. In this study, the relationship between BRD and calf management practices (colostrum feeding, dam parity, serum total protein concentration at introduction (TP), body weight at introduction, introduction season, and daily average growth) was investigated using observational data from pre-weaned dairy calves introduced into a nursery farm in Hokkaido, Japan between 2014 and 2018 (n = 3185). Using additive Bayesian network (ABN) analysis, which is a multivariate statistical modelling approach, the direct and indirect associations between these factors were assessed.

Results

Colostrum feeding contributed to an increase in TP (correlation 1.02 [95 % CI, 0.94;1.10]), which was negatively associated with BRD directly (log odds ratio − 0.38 [− 0.46;−0.31]) and indirectly through increasing daily growth (correlation 0.12 [0.09;0.16]). Calves of multiparous dams had higher body weight at introduction (correlation 0.82 [0.74;0.89]), which indirectly reduced BRD risk through the increasing daily growth (correlation 0.17 [0.14;0.21]). Calves introduced during winter had the highest risk for BRD (log odds ratio 0.29 [0.15;0.44]), while those introduced in summer had the lowest risk (log odds ratio − 0.91 [− 1.06;−0.75]). The introduction season was also associated with BRD indirectly through dam parity, body weight at introduction, and daily growth.

Conclusions

The following calf management practices are recommended for preventing BRD in pre-weaned calves at nursery farms: (1) encouraging colostrum feeding to neonatal calves at their original farms; and (2) identifying calves with higher BRD risk, i.e., those without feeding colostrum, born to primiparous cattle, with low body weight at introduction, and/or introduced in winter, and paying intensive attention to the calves for rapid detection of BRD. ABN analysis applied enabled us to understand the complex inter-relationships between BRD incidence and the risk factors, which will help to reduce BRD incidence and to rear healthy calves at nursery farms.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-021-03018-1.

A scoping review on the use of machine learning in research on social determinants of health: Trends and research prospects

BACKGROUND

Machine learning (ML) has spread rapidly from computer science to several disciplines. Given the predictive capacity of ML, it offers new opportunities for health, behavioral, and social scientists. However, it remains unclear how and to what extent ML is being used in studies of social determinants of health (SDH).

METHODS

Using four search engines, we conducted a scoping review of studies that used ML to study SDH (published before May 1, 2020). Two independent reviewers analyzed the relevant studies. For each study, we identified the research questions, Results, data, and algorithms. We synthesized our findings in a narrative report.

RESULTS

Of the initial 8097 hits, we identified 82 relevant studies. The number of publications has risen during the past decade. More than half of the studies (n = 46) used US data. About 80% (n = 66) utilized surveys, and 70% (n = 57) employed ML for common prediction tasks. Although the number of studies in ML and SDH is growing rapidly, only a few studies used ML to improve causal inference, curate data, or identify social bias in predictions (i.e., algorithmic fairness).

CONCLUSIONS

While ML equips researchers with new ways to measure health outcomes and their determinants from non-conventional sources such as text, audio, and image data, most studies still rely on traditional surveys. Although there are no guarantees that ML will lead to better social epidemiological research, the potential for innovation in SDH research is evident as a result of harnessing the predictive power of ML for causality, data curation, or algorithmic fairness.

Causal Datasheet for Datasets: An Evaluation Guide for Real-World Data Analysis and Data Collection Design Using Bayesian Networks

Developing data-driven solutions that address real-world problems requires understanding of these problems' causes and how their interaction affects the outcome-often with only observational data. Causal Bayesian Networks (BN) have been proposed as a powerful method for discovering and representing the causal relationships from observational data as a Directed Acyclic Graph (DAG). BNs could be especially useful for research in global health in Lower and Middle Income Countries, where there is an increasing abundance of observational data that could be harnessed for policy making, program evaluation, and intervention design. However, BNs have not been widely adopted by global health professionals, and in real-world applications, confidence in the results of BNs generally remains inadequate. This is partially due to the inability to validate against some ground truth, as the true DAG is not available. This is especially problematic if a learned DAG conflicts with pre-existing domain doctrine. Here we conceptualize and demonstrate an idea of a "Causal Datasheet" that could approximate and document BN performance expectations for a given dataset, aiming to provide confidence and sample size requirements to practitioners. To generate results for such a Causal Datasheet, a tool was developed which can generate synthetic Bayesian networks and their associated synthetic datasets to mimic real-world datasets. The results given by well-known structure learning algorithms and a novel implementation of the OrderMCMC method using the Quotient Normalized Maximum Likelihood score were recorded. These results were used to populate the Causal Datasheet, and recommendations could be made dependent on whether expected performance met user-defined thresholds. We present our experience in the creation of Causal Datasheets to aid analysis decisions at different stages of the research process. First, one was deployed to help determine the appropriate sample size of a planned study of sexual and reproductive health in Madhya Pradesh, India. Second, a datasheet was created to estimate the performance of an existing maternal health survey we conducted in Uttar Pradesh, India. Third, we validated generated performance estimates and investigated current limitations on the well-known ALARM dataset. Our experience demonstrates the utility of the Causal Datasheet, which can help global health practitioners gain more confidence when applying BNs.

Revealing antibiotic cross-resistance patterns in hospitalized patients through Bayesian network modelling

OBJECTIVES

Microbial resistance exhibits dependency patterns between different antibiotics, termed cross-resistance and collateral sensitivity. These patterns differ between experimental and clinical settings. It is unclear whether the differences result from biological reasons or from confounding, biasing results found in clinical settings. We set out to elucidate the underlying dependency patterns between resistance to different antibiotics from clinical data, while accounting for patient characteristics and previous antibiotic usage.

METHODS

Additive Bayesian network modelling was employed to simultaneously estimate relationships between variables in a dataset of bacterial cultures derived from hospitalized patients and tested for resistance to multiple antibiotics. Data contained resistance results, patient demographics and previous antibiotic usage, for five bacterial species: Escherichia coli (n = 1054), Klebsiella pneumoniae (n = 664), Pseudomonas aeruginosa (n = 571), CoNS (n = 495) and Proteus mirabilis (n = 415).

RESULTS

All links between resistance to the various antibiotics were positive. Multiple direct links between resistance of antibiotics from different classes were observed across bacterial species. For example, resistance to gentamicin in E. coli was directly linked with resistance to ciprofloxacin (OR = 8.39, 95% credible interval 5.58-13.30) and sulfamethoxazole/trimethoprim (OR = 2.95, 95% credible interval 1.97-4.51). In addition, resistance to various antibiotics was directly linked with previous antibiotic usage.

CONCLUSIONS

Robust relationships among resistance to antibiotics belonging to different classes, as well as resistance being linked to having taken antibiotics of a different class, exist even when taking into account multiple covariate dependencies. These relationships could help inform choices of antibiotic treatment in clinical settings.

Applications of artificial intelligence in drug development using real-world data

The US Food and Drug Administration (FDA) has been actively promoting the use of real-world data (RWD) in drug development. RWD can generate important real-world evidence reflecting the real-world clinical environment where the treatments are used. Meanwhile, artificial intelligence (AI), especially machine- and deep-learning (ML/DL) methods, have been increasingly used across many stages of the drug development process. Advancements in AI have also provided new strategies to analyze large, multidimensional RWD. Thus, we conducted a rapid review of articles from the past 20 years, to provide an overview of the drug development studies that use both AI and RWD. We found that the most popular applications were adverse event detection, trial recruitment, and drug repurposing. Here, we also discuss current research gaps and future opportunities.

A Bayesian network approach for the identification of relationships between drivers of chlordecone bioaccumulation in plants

Plants were sampled from four different types of chlordecone-contaminated land in Guadeloupe (West Indies). The objective was to investigate the importance of biological and agri-environmental parameters in the ability of plants to bioaccumulate chlordecone. Among the plant traits studied, only the growth habit significantly affected chlordecone transfer, since prostrate plants concentrated more chlordecone than erect plants. In addition, intensification of land use has led to a significant increase in the amount of chlordecone absorbed by plants. The use of Bayesian networks uncovers some hypothesis and identifies paths for reflection and possible studies to identify and quantify relationships that explain our data. Graphical abstract.

Biosecurity risk factors for highly pathogenic avian influenza (H5N8) virus infection in duck farms, France

Highly pathogenic avian influenza (HPAI) subtype H5N8 outbreaks occurred in poultry farms in France in 2016-2017, resulting in significant economic losses and disruption to the poultry industry. Current evidence on associations between actual on-farm biosecurity risk factors and H5N8 occurrence is limited. Therefore, a retrospective matched case-control study was undertaken to investigate the inter-relationships between on-farm biosecurity practices and H5N8 infection status to provide new insights regarding promising targets for intervention. Data were collected on 133 case and 133 control duck farms (i.e. the most affected species) located in one area of the country that was mostly affected by the disease. Data were analysed using Additive Bayesian Networks which offer a rich modelling framework by graphically illustrating the dependencies between variables. Factors indirectly and directly positively associated with farm infection were inadequate management of vehicle movements (odds ratio [OR] 9.3, 95% credible interval [CI] 4.0-22.8) and inadequate delimitation of farm and units (OR 3.0, 95% CI 1.6-5.8), respectively. Inadequate disposal of dead birds was instead negatively associated with the outcome (OR 0.1, 95% CI 0.0-0.3). The findings highlight that reinforcing farm access control systems and reducing the number of visitors are key biosecurity measures to control farm vulnerability to H5N8 infection and could help setting priorities in biosecurity practices to prevent outbreaks' re-occurrence.

Australian beef producers' knowledge and attitudes relating to hydatid disease are associated with their control practices

Despite available control strategies, hydatid disease in beef cattle has been shown to have a wider geographic range and higher prevalence than previously recognised in Australia. The aim of the current study was to determine whether producer knowledge and attitudes are associated with farm management practices that could influence transmission among domestic dogs, wildlife, livestock and humans. Between June and August 2019, a cross-sectional study was conducted among beef producers throughout Australia (N = 62). Producers were asked to complete an online survey to obtain information on their knowledge about hydatid disease, their attitudes towards the disease and their farm management practices that could affect transmission. Descriptive statistics were conducted to investigate potential predictors for practices that might influence transmission of the parasite. A Bayesian network (BN) model was then constructed to evaluate the interrelationships between variables. The results show that most respondents (87 %; 54/62) had heard of hydatid disease. However, only 61 % of respondents knew how hydatid disease is transmitted (38/62) and only half knew how to prevent transmission (52 %; 32/62). Of respondents that knew that hydatid disease could affect humans (44/62), many did not think their family was at risk (46 %, 20/44) because they dewormed their dogs and prevented their dogs' access to offal. However, most respondents who owned dogs did not deworm their dogs frequently enough to prevent patency of Echinococcus granulosus infection (86 %; 49/57). Almost all respondents (94 %; 58/62) said they would take action if they found out their cattle were infected. BN analysis revealed that implementation of practices that could reduce the risk of hydatid disease transmission were associated with producers' knowledge and attitudes. In the model, practices were most influenced by attitudes (percentage change in variance = 42 %). All respondents in the "hydatid prevention" practices group were in the "good" knowledge group and the "less concerned" attitudes group. In comparison, most of the respondents in the "standard husbandry" practices group were in the "poor" knowledge group and the "more concerned" attitudes group. In summary, the results indicate that greater knowledge of hydatid disease among beef producers is associated with practices that reduce hydatid risk and attitudes of less concern about hydatid impact on properties. Therefore, increasing producer knowledge is warranted to encourage adoption and improvement of hydatid prevention practices and would be well received by beef producers.

Bayesian Network Modeling Applied to Feline Calicivirus Infection Among Cats in Switzerland

Bayesian network (BN) modeling is a rich and flexible analytical framework capable of elucidating complex veterinary epidemiological data. It is a graphical modeling technique that enables the visual presentation of multi-dimensional results while retaining statistical rigor in population-level inference. Using previously published case study data about feline calicivirus (FCV) and other respiratory pathogens in cats in Switzerland, a full BN modeling analysis is presented. The analysis shows that reducing the group size and vaccinating animals are the two actionable factors directly associated with FCV status and are primary targets to control FCV infection. The presence of gingivostomatitis and Mycoplasma felis is also associated with FCV status, but signs of upper respiratory tract disease (URTD) are not. FCV data is particularly well-suited to a network modeling approach, as both multiple pathogens and multiple clinical signs per pathogen are involved, along with multiple potentially interrelated risk factors. BN modeling is a holistic approach—all variables of interest may be mutually interdependent—which may help to address issues, such as confounding and collinear factors, as well as to disentangle directly vs. indirectly related variables. We introduce the BN methodology as an alternative to the classical uni- and multivariable regression approaches commonly used for risk factor analyses. We advise and guide researchers about how to use BNs as an exploratory data tool and demonstrate the limitations and practical issues. We present a step-by-step case study using FCV data along with all code necessary to reproduce our analyses in the open-source R environment. We compare and contrast the findings of the current case study using BN modeling with previous results that used classical regression techniques, and we highlight new potential insights. Finally, we discuss advanced methods, such as Bayesian model averaging, a common way of accounting for model uncertainty in a Bayesian network context.

Additive Bayesian networks for antimicrobial resistance and potential risk factors in non-typhoidal Salmonella isolates from layer hens in Uganda

Background

Multi-drug resistant bacteria are seen increasingly and there are gaps in our understanding of the complexity of antimicrobial resistance, partially due to a lack of appropriate statistical tools. This hampers efficient treatment, precludes determining appropriate intervention points and renders prevention very difficult.

Methods

We re-analysed data from a previous study using additive Bayesian networks. The data contained information on resistances against seven antimicrobials and seven potential risk factors from 86 non-typhoidal Salmonella isolates from laying hens in 46 farms in Uganda.

Results

The final graph contained 22 links between risk factors and antimicrobial resistances. Solely ampicillin resistance was linked to the vaccinating person and disposal of dead birds. Systematic associations between ampicillin and sulfamethoxazole/trimethoprim and chloramphenicol, which was also linked to sulfamethoxazole/trimethoprim were detected. Sulfamethoxazole/trimethoprim was also directly linked to ciprofloxacin and trimethoprim. Trimethoprim was linked to sulfonamide and ciprofloxacin, which was also linked to sulfonamide. Tetracycline was solely linked to ciprofloxacin.

Conclusions

Although the results needs to be interpreted with caution due to a small data set, additive Bayesian network analysis allowed a description of a number of associations between the risk factors and antimicrobial resistances investigated.

Understanding the complex relationships underlying hot flashes: a Bayesian network approach

OBJECTIVE

The mechanism underlying hot flashes is not well-understood, primarily because of complex relationships between and among hot flashes and their risk factors.

METHODS

We explored those relationships using a Bayesian network approach based on a 2006 to 2015 cohort study of hot flashes among 776 female residents, 45 to 54 years old, in the Baltimore area. Bayesian networks were fit for each outcome (current hot flashes, hot flashes before the end of the study, hot flash severity, hot flash frequency, and age at first hot flashes) separately and together with a list of risk factors (estrogen, progesterone, testosterone, body mass index and obesity, race, income level, education level, smoking history, drinking history, and activity level). Each fitting was conducted separately on all women and only perimenopausal women, at enrollment and 4 years after enrollment.

RESULTS

Hormone levels, almost always interrelated, were the most common variable linked to hot flashes; hormone levels were sometimes related to body mass index, but were not directly related to any other risk factors. Smoking was also frequently associated with increased likelihood of severe symptoms, but not through an antiestrogenic pathway. The age at first hot flashes was related only to race. All other factors were either not related to outcomes or were mediated entirely by race, hormone levels, or smoking.

CONCLUSIONS

These models can serve as a guide for design of studies into the causal network underlying hot flashes.

Additive Bayesian Network analysis of associations between antimicrobial consumption, biosecurity, vaccination and productivity in Danish sow herds

In modern livestock farming, there is an increasing understanding that antimicrobial (AM) consumption must be kept low - preferably without compromising animal welfare or productivity. This requires an understanding of the relationship between AM use, productivity, biosecurity, vaccination and herd demographics. To obtain this, we undertook an Additive Bayesian Network analysis using data from 2014 to 2015, covering 157 Danish sow herds with weaners. In general, productivity and biosecurity were high, and AM consumption low. No association was found between prescribed AM and productivity. Other variables, such as biosecurity and enrolment in the Danish Specific Pathogen Free (SPF) system, had stronger associations with sow productivity than AM consumption. In the weaner unit, an association between AM consumption and certain vaccination practices was found, suggesting that vaccines might be used to control preexisting problems. The results reveal that most Danish sow producers are able to maintain productivity while using low amounts of AMs. This conclusion must be interpreted within the context of Danish pig farming i.e. generally high biosecurity and many years of official restrictions aiming at reducing AM consumption.

Determinants of Spatial Heterogeneity of Functional Illiteracy among School-Aged Children in the Philippines: An Ecological Study

Functional literacy is one of the targets of the Sustainable Development Goals (SDGs) of the United Nations. Functional literacy indicators are likely to vary between locations given the geographical variability of its major determinants. This property poses a challenge to decisions around efficient allocation of population services and resources to mitigate the impact of functional literacy in populations most in need. Using functional literacy indicators of 11,313 school-aged children collected in 2008 during the nationwide survey, the current study examined the association between functional literacy and geographical disparities in socioeconomic status (SES), water supply, sanitation and hygiene, household education stimuli, and environmental variables in all three regions of the Philippines (Luzon, the Visayas, and Mindanao). Three nested fixed-effects multinomial regression models were built to determine associations between functional literacy and a wide array of variables. Our results showed the general prevalence rate of functional illiteracy as being 4.7%, with the highest prevalence rate in the Visayas, followed by Mindanao and Luzon (7.5%, 6.9%, and 3.0%, respectively. Our results indicated that in Luzon prevalence of functional illiteracy was explained by variation in household education stimuli scores, sources of drinking water, and type of toilet facility. In Mindanao and the Visayas prevalence of functional illiteracy was primarily explained by geographical variation in SES, and natural environmental conditions. Our study highlights region-specific determinants of functional literacy and the need for geographically targeted, integrated interventions.

Interpreting mutual adjustment for multiple indicators of socioeconomic position without committing mutual adjustment fallacies

Research into the effects of Socioeconomic Position (SEP) on health will sometimes compare effects from multiple, different measures of SEP in "mutually adjusted" regression models. Interpreting each effect estimate from such models equivalently as the "independent" effect of each measure may be misleading, a mutual adjustment (or Table 2) fallacy. We use directed acyclic graphs (DAGs) to explain how interpretation of such models rests on assumptions about the causal relationships between those various SEP measures. We use an example DAG whereby education leads to occupation and both determine income, and explain implications for the interpretation of mutually adjusted coefficients for these three SEP indicators. Under this DAG, the mutually adjusted coefficient for education will represent the direct effect of education, not mediated via occupation or income. The coefficient for occupation represents the direct effect of occupation, not mediated via income, or confounded by education. The coefficient for income represents the effect of income, after adjusting for confounding by education and occupation. Direct comparisons of mutually adjusted coefficients are not comparing like with like. A theoretical understanding of how SEP measures relate to each other can influence conclusions as to which measures of SEP are most important. Additionally, in some situations adjustment for confounding from more distal SEP measures (like education and occupation) may be sufficient to block unmeasured socioeconomic confounding, allowing for greater causal confidence in adjusted effect estimates for more proximal measures of SEP (like income).

Spatial epidemiological approaches to inform leptospirosis surveillance and control: A systematic review and critical appraisal of methods

Leptospirosis is a global zoonotic disease that the transmission is driven by complex geographical and temporal variation in demographics, animal hosts and socioecological factors. This results in complex challenges for the identification of high-risk areas. Spatial and temporal epidemiological tools could be used to support leptospirosis control programs, but the adequacy of its application has not been evaluated. We searched literature in six databases including PubMed, Web of Science, EMBASE, Scopus, SciELO and Zoological Record to systematically review and critically assess the use of spatial and temporal analytical tools for leptospirosis and to provide general framework for its application in future studies. We reviewed 115 articles published between 1930 and October 2018 from 41 different countries. Of these, 65 (56.52%) articles were on human leptospirosis, 39 (33.91%) on animal leptospirosis and 11 (9.5%) used data from both human and animal leptospirosis. Spatial analytical (n = 106) tools were used to describe the distribution of incidence/prevalence at various geographical scales (96.5%) and to explored spatial patterns to detect clustering and hot spots (33%). A total of 51 studies modelled the relationships of various variables on the risk of human (n = 31), animal (n = 17) and both human and animal infection (n = 3). Among those modelling studies, few studies had generated spatially structured models and predictive maps of human (n = 2/31) and animal leptospirosis (n = 1/17). In addition, nine studies applied time-series analytical tools to predict leptospirosis incidence. Spatial and temporal analytical tools have been greatly utilized to improve our understanding on leptospirosis epidemiology. Yet the quality of the epidemiological data, the selection of covariates and spatial analytical techniques should be carefully considered in future studies to improve usefulness of evidence as tools to support leptospirosis control. A general framework for the application of spatial analytical tools for leptospirosis was proposed.

Biosecurity aspects of cattle production in Western Uganda, and associations with seroprevalence of brucellosis, salmonellosis and bovine viral diarrhoea

BACKGROUND

Many low-income countries have a human population with a high number of cattle owners depending on their livestock for food and income. Infectious diseases threaten the health and production of cattle, affecting both the farmers and their families as well as other actors in often informal value chains. Many infectious diseases can be prevented by good biosecurity. The objectives of this study were to describe herd management and biosecurity routines with potential impact on the prevalence of infectious diseases, and to estimate the burden of infectious diseases in Ugandan cattle herds, using the seroprevalence of three model infections.

RESULTS

Farmer interviews (n = 144) showed that biosecurity measures are rarely practised. Visitors' hand-wash was used by 14%, cleaning of boots or feet by 4 and 79% put new cattle directly into the herd. During the 12 months preceding the interviews, 51% of farmers had cattle that died and 31% had noticed abortions among their cows. Interestingly, 72% were satisfied with the health status of their cattle during the same time period. The prevalence (95% CI) of farms with at least one seropositive animal was 16.7% (11.0;23.8), 23.6% (16.9;31.4), and 53.4% (45.0;61.8) for brucella, salmonella and BVD, respectively. A poisson regression model suggested that having employees looking after the cattle, sharing pasture with other herds, and a higher number of dead cattle were associated with a herd being positive to an increasing number of the diseases. An additive bayesian network model with biosecurity variables and a variable for the number of diseases the herd was positive to resulted in three separate directed acyclic graphs which illustrate how herd characteristics can be grouped together. This model associated the smallest herd size with herds positive to a decreasing number of diseases and having fewer employees.

CONCLUSION

There is potential for improvement of biosecurity practices in Ugandan cattle production. Salmonella, brucella and BVD were prevalent in cattle herds in the study area and these infections are, to some extent, associated with farm management practices.

Comparison between generalized linear modelling and additive Bayesian network; identification of factors associated with the incidence of antibodies against Leptospira interrogans sv Pomona in meat workers in New Zealand

BACKGROUND

Additive Bayesian Network (ABN) is a graphical model which extends Generalized Linear Modelling (GLM) to multiple dependent variables. The present study compares results from GLM with those from ABN analysis used to identify factors associated with Leptospira interrogans sv Pomona (Pomona) infection by exploring the advantages and disadvantages of these two methodologies, to corroborate inferences informing health and safety measures at abattoirs in New Zealand (NZ).

METHODOLOGY AND FINDINGS

In a cohort study in four sheep slaughtering abattoirs in NZ, sera were collected twice a year from 384 meat workers and tested by Microscopic Agglutination with a 91% sensitivity and 94% specificity for Pomona. The study primarily addressed the effect of work position, personal protective equipment (PPE) and non-work related exposures such as hunting on a new infection with Pomona. Significantly associated with Pomona were "Work position" and two "Abattoirs" (GLM), and "Work position" (ABN). The odds of Pomona infection (OR, [95% CI]) was highest at stunning and hide removal (ABN 41.0, [6.9-1044.2]; GLM 57.0, [6.9-473.3]), followed by removal of intestines, bladder, and kidneys (ABN 30.7, [4.9-788.4]; GLM 33.8, [4.2-271.1]). Wearing a facemask, glasses or gloves (PPE) did not result as a protective factor in GLM or ABN.

CONCLUSIONS/SIGNIFICANCE

The odds of Pomona infection was highest at stunning and hide removal. PPE did not show any indication of being protective in GLM or ABN. In ABN all relationships between variables are modelled; hence it has an advantage over GLM due to its capacity to capture the natural complexity of data more effectively.

Challenges in developing methods for quantifying the effects of weather and climate on water-associated diseases: A systematic review

Infectious diseases attributable to unsafe water supply, sanitation and hygiene (e.g. Cholera, Leptospirosis, Giardiasis) remain an important cause of morbidity and mortality, especially in low-income countries. Climate and weather factors are known to affect the transmission and distribution of infectious diseases and statistical and mathematical modelling are continuously developing to investigate the impact of weather and climate on water-associated diseases. There have been little critical analyses of the methodological approaches. Our objective is to review and summarize statistical and modelling methods used to investigate the effects of weather and climate on infectious diseases associated with water, in order to identify limitations and knowledge gaps in developing of new methods. We conducted a systematic review of English-language papers published from 2000 to 2015. Search terms included concepts related to water-associated diseases, weather and climate, statistical, epidemiological and modelling methods. We found 102 full text papers that met our criteria and were included in the analysis. The most commonly used methods were grouped in two clusters: process-based models (PBM) and time series and spatial epidemiology (TS-SE). In general, PBM methods were employed when the bio-physical mechanism of the pathogen under study was relatively well known (e.g. Vibrio cholerae); TS-SE tended to be used when the specific environmental mechanisms were unclear (e.g. Campylobacter). Important data and methodological challenges emerged, with implications for surveillance and control of water-associated infections. The most common limitations comprised: non-inclusion of key factors (e.g. biological mechanism, demographic heterogeneity, human behavior), reporting bias, poor data quality, and collinearity in exposures. Furthermore, the methods often did not distinguish among the multiple sources of time-lags (e.g. patient physiology, reporting bias, healthcare access) between environmental drivers/exposures and disease detection. Key areas of future research include: disentangling the complex effects of weather/climate on each exposure-health outcome pathway (e.g. person-to-person vs environment-to-person), and linking weather data to individual cases longitudinally.

A questionnaire study of associations between potential risk factors and salmonella status in Swedish dairy herds

In this study associations between potential risk factors and salmonella status in Swedish dairy herds were investigated. A case-control study design was used, including existing as well as new cases. Herds were assigned a salmonella status on the basis of antibody analysis of bulk milk samples. Information on potential risk factors was collected from registry data and from farmers via a questionnaire. Univariable and multivariable logistic regression analyses were used to investigate associations between salmonella status and potential risk factors. In addition, multivariate analysis with Additive Bayesian Network (ABN) modelling was performed to improve understanding of the complex relationship between all the variables. Because of the difficulty in identifying associations between potential risk factors and infections with low prevalence and a large regional variation, exposure of potential risk factors in the high-prevalence region (Öland) were compared to exposure in other regions in Sweden. In total 483 of 996 (48%) farmers responded to the questionnaire, 69 herds had test-positive bulk milk samples. The strongest association with salmonella status was 'presence of salmonella test-positive herds <5 km' (OR 4.3, 95% CI 2.0-9.4). Associations with salmonella status were also seen between 'feeding calves residue milk only' (OR 2.4, 95% CI 1.2-4.6), 'certified organic herds' (OR 2.5, 95% CI 1.2-4.9) and 'frequently seeing signs of rodents' (OR 0.4, 95% CI 0.13-0.97). The ABN model showed associations between Öland and four of the variables: salmonella status, presence of test-positive herds <5km, shared pastures and providing protective clothing for visitors. The latter is probably a reflection of increased disease awareness in Öland. The ABN model showed associations between herd size and housing as well as several management procedures. This provides an explanation why herd size frequently has been identified as a risk factor for salmonella by other studies. The study confirms the importance of local transmission routes for salmonella, but does not identify specific components in this local spread. Therefore, it supports the use of a broad biosecurity approach in the prevention of salmonella. In Öland, some potential risk factors are more common than in other parts of Sweden. Theoretically these could contribute to the spread of salmonella, but this was not confirmed in the present study. The study also highlights the difficulty in identifying associations between potential risk factors and infections with low prevalence and large regional variation.

A Systems Science Approach to Understanding Polytrauma and Blast-Related Injury: Bayesian Network Model of Data From a Survey of the Florida National Guard

We sought to further define the epidemiology of the complex, multiple injuries collectively known as polytrauma/blast-related injury (PT/BRI). Using a systems science approach, we performed Bayesian network modeling to find the most accurate representation of the complex system of PT/BRI and identify key variables for understanding the subsequent effects of blast exposure in a sample of Florida National Guard members (1,443 deployed to Operation Enduring Freedom/Operation Iraqi Freedom and 1,655 not deployed) who completed an online survey during the period from 2009 to 2010. We found that postdeployment symptoms reported as present at the time of the survey were largely independent of deployment per se. Blast exposure, not mild traumatic brain injury (TBI), acted as the primary military deployment-related driver of PT/BRI symptoms. Blast exposure was indirectly linked to mild TBI via other deployment-related traumas and was a significant risk for a high level of posttraumatic stress disorder (PTSD) arousal symptoms. PTSD arousal symptoms and tinnitus were directly dependent upon blast exposure, with both acting as bridge symptoms to other postdeployment mental health and physical symptoms, respectively. Neurobehavioral or postconcussion-like symptoms had no significant dependence relationship with mild TBI, but they were synergistic with blast exposure in influencing PTSD arousal symptoms. A replication of this analysis using a larger PT/BRI database is warranted.

Multivariate Analysis of the Determinants of the End-Product Quality of Manure-Based Composts and Vermicomposts Using Bayesian Network Modelling

Previous studies indicated that the quality of tropical composts is poorer than that of composts produced in temperate regions. The aim of this study was to test the type of manure, the use of co-composting with green waste, and the stabilization method for their ability to improve compost quality in the tropics. We produced 68 composts and vermicomposts that were analysed for their C, lignin and NPK contents throughout the composting process. Bayesian networks were used to assess the mechanisms controlling compost quality. The concentration effect, for C and lignin, and the initial blend quality, for NPK content, were the main factors affecting compost quality. Cattle manure composts presented the highest C and lignin contents, and poultry litter composts exhibited the highest NPK content. Co-composting improved quality by enhancing the concentration effect, which reduced the impact of C and nutrient losses. Vermicomposting did not improve compost quality; co-composting without earthworms thus appears to be a suitable stabilization method under the conditions of this study because it produced high quality composts and is easier to implement.

Bayesian network modeling: A case study of an epidemiologic system analysis of cardiovascular risk

An extensive, in-depth study of cardiovascular risk factors (CVRF) seems to be of crucial importance in the research of cardiovascular disease (CVD) in order to prevent (or reduce) the chance of developing or dying from CVD. The main focus of data analysis is on the use of models able to discover and understand the relationships between different CVRF. In this paper a report on applying Bayesian network (BN) modeling to discover the relationships among thirteen relevant epidemiological features of heart age domain in order to analyze cardiovascular lost years (CVLY), cardiovascular risk score (CVRS), and metabolic syndrome (MetS) is presented. Furthermore, the induced BN was used to make inference taking into account three reasoning patterns: causal reasoning, evidential reasoning, and intercausal reasoning. Application of BN tools has led to discovery of several direct and indirect relationships between different CVRF. The BN analysis showed several interesting results, among them: CVLY was highly influenced by smoking being the group of men the one with highest risk in CVLY; MetS was highly influence by physical activity (PA) being again the group of men the one with highest risk in MetS, and smoking did not show any influence. BNs produce an intuitive, transparent, graphical representation of the relationships between different CVRF. The ability of BNs to predict new scenarios when hypothetical information is introduced makes BN modeling an Artificial Intelligence (AI) tool of special interest in epidemiological studies. As CVD is multifactorial the use of BNs seems to be an adequate modeling tool.

Discovering complex interrelationships between socioeconomic status and health in Europe: A case study applying Bayesian Networks

Studies assume that socioeconomic status determines individuals' states of health, but how does health determine socioeconomic status? And how does this association vary depending on contextual differences? To answer this question, our study uses an additive Bayesian Networks model to explain the interrelationships between health and socioeconomic determinants using complex and messy data. This model has been used to find the most probable structure in a network to describe the interdependence of these factors in five European welfare state regimes. The advantage of this study is that it offers a specific picture to describe the complex interrelationship between socioeconomic determinants and health, producing a network that is controlled by socio-demographic factors such as gender and age. The present work provides a general framework to describe and understand the complex association between socioeconomic determinants and health.

Antibiotic cross-resistance in the lab and resistance co-occurrence in the clinic: Discrepancies and implications in E.coli

INTRODUCTION

Antibiotic resistance is an important public health issue, and vast resources are invested in researching new ways to fight it. Recent experimental works have shown that resistance to some antibiotics can result in increased susceptibility to others, namely induce cross-sensitivity. This phenomenon could be utilized to increase efficiency of antibiotic treatment strategies that minimize resistance. However, as conditions in experimental settings and in the clinic may differ substantially, the implications of cross-sensitivity for clinical settings are not guaranteed and should be examined.

METHODS

In this work we analyzed data of Escherichia coli isolates from patients' blood, sampled in Rabin Medical Center, Israel, to examine co-occurrence of resistance to antibiotics in the clinic. We compared the co-occurrence patterns with cross-sensitivity patterns observed in the lab.

RESULTS

Our data showed only positively associated occurrence of resistance, even with antibiotics that were shown to induce cross-sensitivity in laboratory conditions. We used a mathematical model to examine the potential effects of cross-sensitivity versus co-occurrence on the spread of drug resistance.

CONCLUSIONS

We conclude that resistance frequencies in the clinic can have a substantial effect on the success of treatment strategies, and should be considered alongside experimental evidence of cross-sensitivity.

Attitudes of Austrian veterinarians towards euthanasia in small animal practice: impacts of age and gender on views on euthanasia

Background

Euthanasia of pets has been described by veterinarians as “the best and the worst” of the profession. The most commonly mentioned ethical dilemmas veterinarians face in small animal practice are: limited treatment options due to financial constraints, euthanizing of healthy animals and owners wishing to continue treatment of terminally ill animals. The aim of the study was to gain insight into the attitudes of Austrian veterinarians towards euthanasia of small animals. This included assessing their agreement with euthanasia in exemplified case scenarios, potentially predicted by demographic variables (e.g. gender, age, working in small animal practice, employment, working in a team, numbers of performed euthanasia). Further describing the veterinarians’ agreement with a number of different normative and descriptive statements, including coping strategies. A questionnaire with nine euthanasia scenarios, 26 normative and descriptive statements, and demographic data were sent to all members of the Austrian Chamber of Veterinary Surgeons (n = 2478).

Results

In total, 486 veterinarians answered sufficiently completely to enable analyses. Responses were first explored descriptively before being formally analysed using linear regression and additive Bayesian networks – a multivariate regression methodology – in order to identify joint relationships between the demographic variables, the statements and each of the nine euthanasia scenarios. Mutual dependencies between the demographic variables were found, i.e. female compared to male veterinarians worked mostly in small animal practice, and working mostly in small animal practice was linked to performing more euthanasia per month.

Conclusions

Gender and age were found to be associated with views on euthanasia: female veterinarians and veterinarians having worked for less years were more likely to disagree with euthanasia in at least some of the convenience euthanasia scenarios. The number of veterinarians working together was found to be the variable with the highest number of links to other variables, demographic as well as ethical statements. This highlights the role of a team potentially providing support in stressful situations. The results are useful for a better understanding of coping strategies for veterinarians with moral stress due to euthanasia of small animals.

Electronic supplementary material

The online version of this article (doi:10.1186/s12917-016-0649-0) contains supplementary material, which is available to authorized users.

Methods of analysis of enteropathogen infection in the MAL-ED Cohort Study

Studies of diarrheal etiology in low- and middle-income countries have typically focused on children presenting with severe symptoms to health centers and thus are best equipped to describe the pathogens capable of leading to severe diarrheal disease. The Etiology, Risk Factors and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development (MAL-ED) cohort study was designed to evaluate, via intensive community surveillance, the hypothesis that repeated exposure to enteropathogens has a detrimental effect on growth, vaccine response, and cognitive development, which are the primary outcome measures for this study. In the setting of multiple outcomes of interest, a longitudinal cohort design was chosen. Because many or even the majority of enteric infections are asymptomatic, the collection of asymptomatic surveillance stools was a critical element. However, capturing diarrheal stools additionally allowed for the determination of the principle causes of diarrhea at the community level as well as for a comparison between those enteropathogens associated with diarrhea and those that are associated with poor growth, diminished vaccine response, and impaired cognitive development. Here, we discuss the analytical methods proposed for the MAL-ED study to determine the principal causes of diarrhea at the community level and describe the complex interplay between recurrent exposure to enteropathogens and these critical long-term outcomes.

Unraveling antimicrobial resistance genes and phenotype patterns among Enterococcus faecalis isolated from retail chicken products in Japan

Multidrug-resistant enterococci are considered crucial drivers for the dissemination of antimicrobial resistance determinants within and beyond a genus. These organisms may pass numerous resistance determinants to other harmful pathogens, whose multiple resistances would cause adverse consequences. Therefore, an understanding of the coexistence epidemiology of resistance genes is critical, but such information remains limited. In this study, our first objective was to determine the prevalence of principal resistance phenotypes and genes among Enterococcus faecalis isolated from retail chicken domestic products collected throughout Japan. Subsequent analysis of these data by using an additive Bayesian network (ABN) model revealed the co-appearance patterns of resistance genes and identified the associations between resistance genes and phenotypes. The common phenotypes observed among E. faecalis isolated from the domestic products were the resistances to oxytetracycline (58.4%), dihydrostreptomycin (50.4%), and erythromycin (37.2%), and the gene tet(L) was detected in 46.0% of the isolates. The ABN model identified statistically significant associations between tet(L) and erm(B), tet(L) and ant(6)-Ia, ant(6)-Ia and aph(3’)-IIIa, and aph(3’)-IIIa and erm(B), which indicated that a multiple-resistance profile of tetracycline, erythromycin, streptomycin, and kanamycin is systematic rather than random. Conversely, the presence of tet(O) was only negatively associated with that of erm(B) and tet(M), which suggested that in the presence of tet(O), the aforementioned multiple resistance is unlikely to be observed. Such heterogeneity in linkages among genes that confer the same phenotypic resistance highlights the importance of incorporating genetic information when investigating the risk factors for the spread of resistance. The epidemiological factors that underlie the persistence of systematic multiple-resistance patterns warrant further investigations with appropriate adjustments for ecological and bacteriological factors.

Statins and Primary Prevention of Cardiovascular Disease in Women

Objectives. A systematic review of randomized clinical trials and meta-analyses evaluating the efficacy, tolerability, and safety of statins in preventing cardiovascular disease (CVD) in women without cardiovascular disease. Background. Several meta-analyses have been performed evaluating statins in CVD primary prevention trials involving women. This review is an update incorporating the results of recent CVD primary prevention trials in women and the recent concerns of statins and new-onset diabetes. Method. PubMed database was searched for primary prevention trials and meta-analyses. The key terms “statins, cardiovascular disease, primary prevention in women” were used. Search was limited to all English publications published up to October 2012. Results. Statin use led to a trend towards reduction in cardiovascular mortality and morbidity in women. No significant increased risk in adverse events was observed. The slight increased incidence of diabetes is outweighed by the greater cardiovascular benefit derived from statin use. Conclusions. The data support the use of statins for primary prevention of CVD in women at higher risk of CVD. The lack of statistical significance in prior randomized controlled trials and meta-analyses is attributable to the lower numbers of women enrolled in these trials and the lower CVD risk of women in the trials resulting in the inadequate powering of these studies. Higher risk women who may benefit from CVD primary prevention with statins may be identified using validated tools such as the Reynolds scoring system, the 2011 American Heart Association risk algorithm for women, and the forthcoming National Heart, Lung, and Blood Institute risk equations.

Human health impacts of antibiotic use in agriculture: A push for improved causal inference

Resistant bacterial infections in humans continue to pose a significant challenge globally. Antibiotic use in agriculture contributes to this problem, but failing to appreciate the relative importance of diverse potential causes represents a significant barrier to effective intervention. Standard epidemiologic methods alone are often insufficient to accurately describe the relationships between agricultural antibiotic use and resistance. The integration of diverse methodologies from multiple disciplines will be essential, including causal network modeling and population dynamics approaches. Because intuition can be a poor guide in directing investigative efforts of these non-linear and interconnected systems, integration of modeling efforts with empirical epidemiology and microbiology in an iterative process may result in more valuable information than either in isolation.

Dynamics of the force of infection: insights from Echinococcus multilocularis infection in foxes

Characterizing the force of infection (FOI) is an essential part of planning cost effective control strategies for zoonotic diseases. Echinococcus multilocularis is the causative agent of alveolar echinococcosis in humans, a serious disease with a high fatality rate and an increasing global spread. Red foxes are high prevalence hosts of E. multilocularis. Through a mathematical modelling approach, using field data collected from in and around the city of Zurich, Switzerland, we find compelling evidence that the FOI is periodic with highly variable amplitude, and, while this amplitude is similar across habitat types, the mean FOI differs markedly between urban and periurban habitats suggesting a considerable risk differential. The FOI, during an annual cycle, ranges from (0.1,0.8) insults (95% CI) in urban habitat in the summer to (9.4, 9.7) (95% CI) in periurban (rural) habitat in winter. Such large temporal and spatial variations in FOI suggest that control strategies are optimal when tailored to local FOI dynamics.

Human alveolar echinococcosis (AE) is caused by the fox tapeworm E. multilocularis and has a high fatality rate if untreated. The frequency of the tapeworm in foxes can be reduced through the regular distribution of anthelmintic baits and thus decrease the risk of zoonotic transmission. Here, we estimate the force of infection to foxes using a mathematical model and data from necropsied foxes. The results suggest that the frequency of anthelmintic baiting of foxes can be optimised to local variations in transmission that depend upon season and type of fox habitat.

Applying Bayesian network modelling to understand the links between on-farm biosecurity practice during the 2007 equine influenza outbreak and horse managers' perceptions of a subsequent outbreak

Australia experienced its first ever outbreak of equine influenza in August 2007. Horses on 9359 premises were infected over a period of 5 months before the disease was successfully eradicated through the combination of horse movement controls, on-farm biosecurity and vaccination. In a previous premises-level case-control study of the 2007 equine influenza outbreak in Australia, the protective effect of several variables representing on-farm biosecurity practices were identified. Separately, factors associated with horse managers' perceptions of the effectiveness of biosecurity measures have been identified. In this analysis we applied additive Bayesian network modelling to describe the complex web of associations linking variables representing on-farm human behaviours during the 2007 equine influenza outbreak (compliance or lack thereof with advised personal biosecurity measures) and horse managers' perceptions of the effectiveness of such measures in the event of a subsequent outbreak. Heuristic structure discovery enabled identification of a robust statistical model for 31 variables representing biosecurity practices and perceptions of the owners and managers of 148 premises. The Bayesian graphical network model we present statistically describes the associations linking horse managers' on-farm biosecurity practices during an at-risk period in the 2007 outbreak and their perceptions of whether such measures will be effective in a future outbreak. Practice of barrier infection control measures were associated with a heightened perception of preparedness, whereas horse managers that considered their on-farm biosecurity to be more stringent during the outbreak period than normal practices had a heightened perception of the effectiveness of other measures such as controlling access to the premises. Past performance in an outbreak setting may indeed be a reliable predictor of future perceptions, and should be considered when targeting infection control guidance to horse owners and managers.

Improving epidemiologic data analyses through multivariate regression modelling

: Regression modelling is one of the most widely utilized approaches in epidemiological analyses. It provides a method of identifying statistical associations, from which potential causal associations relevant to disease control may then be investigated. Multivariable regression - a single dependent variable (outcome, usually disease) with multiple independent variables (predictors) - has long been the standard model. Generalizing multivariable regression to multivariate regression - all variables potentially statistically dependent - offers a far richer modelling framework. Through a series of simple illustrative examples we compare and contrast these approaches. The technical methodology used to implement multivariate regression is well established - Bayesian network structure discovery - and while a relative newcomer to the epidemiological literature has a long history in computing science. Applications of multivariate analysis in epidemiological studies can provide a greater understanding of disease processes at the population level, leading to the design of better disease control and prevention programs.

Using Bayesian networks to explore the role of weather as a potential determinant of disease in pigs

Many pathogens are sensitive to climatic variables and this is reflected in their seasonality of occurrence and transmission. The identification of environmental conditions that influence disease occurrence can be subtle, particularly considering their complex interdependencies in addition to those relationships between climate and disease. Statistical treatment of environmental variables is often dependent on their correlations and thus descriptions of climate are often restricted to means rather than accounting for the more precise aspects (including mean, maximum, minimum, variability). Here we utilize a novel multivariate statistical modelling approach, additive Bayesian network (ABN) analyses, to identify the inter-linkages of different weather variables to better capture short-term environmental conditions that are important drivers of disease. We present a case study that explores weather as a driver of disease in livestock systems. We utilize quality assurance health scheme data on ten major diseases of pigs from 875 finishing pig herds distributed across the United Kingdom over 7 years (2005-2011). We examine the relationship between the occurrence of these pathologies and contemporary weather conditions measured by local meteorological stations. All ten pathologies were associated with at least 2 other pathologies (maximum 6). Three pathologies were associated directly with temperature variables: papular dermatitis, enzootic pneumonia and milk spots. Latitude was strongly associated with multiple pathologies, though associations with longitude were eliminated when clustering for repeated observations of farms was assessed. The identification of relationships between climatic factors and different (potentially related) diseases offers a more comprehensive insight into the complex role of seasonal drivers and herd health status than traditional analytical methods.

Untangling the complex inter-relationships between horse managers' perceptions of effectiveness of biosecurity practices using Bayesian graphical modelling

On-farm biosecurity practices have been promoted in many animal industries to protect animal populations from infections. Current approaches based on regression modelling techniques for assessing biosecurity perceptions and practices are limited for analysis of the interrelationships between multivariate data. A suitable approach, which does not require background knowledge of relationships, is provided by Bayesian network modelling. Here we apply such an approach to explore the complex interrelationships between the variables representing horse managers' perceptions of effectiveness of on-farm biosecurity practices. The dataset was derived from interviews conducted with 200 horse managers in Australia after the 2007 equine influenza outbreak. Using established computationally intensive techniques, an optimal graphical statistical model was identified whose structure was objectively determined, directly from the observed data. This methodology is directly analogous to multivariate regression (i.e. multiple response variables). First, an optimal model structure was identified using an exact (exhaustive) search algorithm, followed by pruning the selected model for over-fitting by the parametric bootstrapping approach. Perceptions about effectiveness of movement restrictions and access control were linked but were generally segregated from the perceptions about effectiveness of personal and equipment hygiene. Horse managers believing in the effectiveness of complying with movement restrictions in stopping equine influenza spread onto their premises were also more likely to believe in the effectiveness of reducing their own contact with other horses and curtailing professional visits. Similarly, the variables representing the effectiveness of disinfecting vehicles, using a disinfectant footbath, changing into clean clothes on arrival at the premises and washing hands before contact with managed horses were clustered together. In contrast, horse managers believing in the effectiveness of disinfecting vehicles (hygiene measure) were less likely to believe in the effectiveness of controlling who has access to managed horses (access control). The findings of this analysis provide new insights into the relationships between perceptions of effectiveness of different biosecurity measures. Different extension education strategies might be required for horse managers believing more strongly in the effectiveness of access control or hygiene measures.

Understanding the associations between on-farm biosecurity practice and equine influenza infection during the 2007 outbreak in Australia

In a previous premises-level case-control study of the 2007 equine influenza outbreak in Australia, the protective effect of several variables representing on-farm biosecurity practices was identified. However, using logistic regression it was not possible to definitively identify individual effects and associations between each of the personal biosecurity measures implemented by horse premises owners and managers in the face of the outbreak. In this study we apply Bayesian network modelling to identify the complex web of associations between these variables, horse premises infection status and other premises-level covariates. We focussed this analysis primarily on the inter-relationship between the nine variables representing on-farm personal biosecurity measures (of people residing on the premises and those visiting), and all other variables from the final logistic regression model of our previous analysis. Exact structure discovery was used to identify the globally optimal model from across the landscape of all directed acyclic graphs possible for our dataset. Bootstrapping was used to adjust the model for over-fitting. Our final Bayesian graphic network model included 18 variables linked by 23 arcs, each arc analogous to a single multivariable generalised linear model, combined in a probabilistically coherent way. Amongst the personal biosecurity measures, having a footbath in place, certain practices of visitors (hand-washing, changing clothes and shoes) in contact with the horses, and the regularity of horse handling were statistically associated with premises infection status. The results of this in-depth analysis provide new insight into the complex web of direct and indirect associations between risk factors and horse premises infection status during the first 7 weeks of the 2007 equine influenza outbreak in Australia. In future outbreaks, unnecessary contact and handling of horses should be avoided, especially by those coming from off the premises. Prior to any such contact, persons handling horses should use a footbath (if present), change their clothes and shoes, and wash their hands.