Characterizing an outbreak of vancomycin-resistant enterococci using hidden Markov models

Bayesian Calibration to Address the Challenge of Antimicrobial Resistance: A Review

Antimicrobial resistance (AMR) emerges when disease-causing microorganisms develop the ability to withstand the effects of antimicrobial therapy. This phenomenon is often fueled by the human-to-human transmission of pathogens and the overuse of antibiotics. Over the past 50 years, increased computational power has facilitated the application of Bayesian inference algorithms. In this comprehensive review, the basic theory of Markov Chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC) methods are explained. These inference algorithms are instrumental in calibrating complex statistical models to the vast amounts of AMR-related data. Popular statistical models include hierarchical and mixture models as well as discrete and stochastic epidemiological compartmental and agent based models. Studies encompassed multi-drug resistance, economic implications of vaccines, and modeling AMR in vitro as well as within specific populations. We describe how combining these topics in a coherent framework can result in an effective antimicrobial stewardship. We also outline recent advancements in the methodology of Bayesian inference algorithms and provide insights into their prospective applicability for modeling AMR in the future.

Check your assumptions: Further scrutiny of basic model frameworks of antimicrobial resistance

Since the mid-1990s, growing concerns over antimicrobial resistant (AMR) organisms has led to an increase in the use of mathematical models to explore the inter-host transmission of such infections. Previous work reviewing such models categorised them into generic frameworks based on their underlying assumptions. These assumptions dictated the coexistence between AMR and antimicrobial sensitive strains. We add to this work performing stability analyses of the frameworks, along with simulating them deterministically and stochastically. Stability analyses found that many of these assumptions lead to models having the same equilibria, but showed differences in the equilibria's stability between models. Deterministic simulations reveal that assuming replacement of one infecting strain by another leads to an unusual antimicrobial treatment threshold. Increasing beyond this threshold causes a discontinuous increase in disease burden. The cost of AMR to pathogen fitness (lowered transmission) dictates both the threshold of treatment that causes the discontinuous increase in disease burden and the size of that increase. It was also shown that Superinfection states can be biased against resident strains and so favour coexistence of both strains. Stochastic simulations demonstrated that differing scenario starting conditions can guide models to converge upon equilibria that they may not have under deterministic simulation. These findings highlight the importance of checking assumptions when modelling AMR and strain competition more widely.

A mathematical model and inference method for bacterial colonization in hospital units applied to active surveillance data for carbapenem-resistant enterobacteriaceae

Widespread use of antibiotics has resulted in an increase in antimicrobial-resistant microorganisms. Although not all bacterial contact results in infection, patients can become asymptomatically colonized, increasing the risk of infection and pathogen transmission. Consequently, many institutions have begun active surveillance, but in non-research settings, the resulting data are often incomplete and may include non-random testing, making conventional epidemiological analysis problematic. We describe a mathematical model and inference method for in-hospital bacterial colonization and transmission of carbapenem-resistant Enterobacteriaceae that is tailored for analysis of active surveillance data with incomplete observations. The model and inference method make use of the full detailed state of the hospital unit, which takes into account the colonization status of each individual in the unit and not only the number of colonized patients at any given time. The inference method computes the exact likelihood of all possible histories consistent with partial observations (despite the exponential increase in possible states that can make likelihood calculation intractable for large hospital units), includes techniques to improve computational efficiency, is tested by computer simulation, and is applied to active surveillance data from a 13-bed rehabilitation unit in New York City. The inference method for exact likelihood calculation is applicable to other Markov models incorporating incomplete observations. The parameters that we identify are the patient-patient transmission rate, pre-existing colonization probability, and prior-to-new-patient transmission probability. Besides identifying the parameters, we predict the effects on the total prevalence (0.07 of the total colonized patient-days) of changing the parameters and estimate the increase in total prevalence attributable to patient-patient transmission (0.02) above the baseline pre-existing colonization (0.05). Simulations with a colonized versus uncolonized long-stay patient had 44% higher total prevalence, suggesting that the long-stay patient may have been a reservoir of transmission. High-priority interventions may include isolation of incoming colonized patients and repeated screening of long-stay patients.

Extended models for nosocomial infection: parameter estimation and model selection

We consider extensions to previous models for patient level nosocomial infection in several ways, provide a specification of the likelihoods for these new models, specify new update steps required for stochastic integration, and provide programs that implement these methods to obtain parameter estimates and model choice statistics. Previous susceptible-infected models are extended to allow for a latent period between initial exposure to the pathogen and the patient becoming themselves infectious, and the possibility of decolonization. We allow for multiple facilities, such as acute care hospitals or long-term care facilities and nursing homes, and for multiple units or wards within a facility. Patient transfers between units and facilities are tracked and accounted for in the models so that direct importation of a colonized individual from one facility or unit to another might be inferred. We allow for constant transmission rates, rates that depend on the number of colonized individuals in a unit or facility, or rates that depend on the proportion of colonized individuals. Statistical analysis is done in a Bayesian framework using Markov chain Monte Carlo methods to obtain a sample of parameter values from their joint posterior distribution. Cross validation, deviance information criterion and widely applicable information criterion approaches to model choice fit very naturally into this framework and we have implemented all three. We illustrate our methods by considering model selection issues and parameter estimation for data on methicilin-resistant Staphylococcus aureus surveillance tests over 1 year at a Veterans Administration hospital comprising seven wards.

Mathematical modelling of vancomycin-resistant enterococci transmission during passive surveillance and active surveillance with contact isolation highlights the need to identify and address the source of acquisition

Background

Clinical studies and mathematical simulation suggest that active surveillance with contact isolation is associated with reduced vancomycin-resistant enterococci (VRE) prevalence compared to passive surveillance. Models using pre- and post-intervention data that account for the imperfect observation and serial dependence of VRE transmission events can better estimate the effectiveness of active surveillance and subsequent contact isolation; however, such analyses have not been performed.

Methods

A mathematical model was fitted to surveillance data collected pre- and post-implementation of active surveillance with contact isolation in the haematology-oncology ward. We developed a Hidden Markov Model to describe undetected and observed VRE colonisation/infection status based on the detection activities in the ward. Bayesian inference was used to estimate transmission rates. The effectiveness of active surveillance was assumed to be via increased detection and subsequent contact isolation of VRE positive patients.

Results

We estimated that 31% (95% credible interval: 0.33–85%) of the VRE transmissions were due to cross-transmission between patients. The ratio of transmission rates from patients with contact isolation versus those without contact isolation was 0.33 (95% credible interval: 0.050–1.22).

Conclusions

The majority of the VRE acquisitions in the haematology-oncology ward was estimated to be due to background rates of VRE, rather than within ward patient to patient acquisition. The credible interval for cross-transmission was wide which results in a large degree of uncertainty in the estimates. Factors that could account for background VRE acquisition include endogenous acquisition from antibiotic selection pressure and VRE in the environment. Contact isolation was not significantly associated with reduced VRE transmission in settings where the majority of VRE acquisition was due to background acquisition, emphasising the need to identify and address the source of acquisition. As the credible interval for the ratio of VRE transmission in contact isolated versus non-contact isolated patients crossed 1, there is a probability that the transmission rate in contact isolation was not lower. Our finding highlights the need to optimise infection control measures other than active surveillance for VRE and subsequent contact isolation to reduce VRE transmission. Such measures could include antimicrobial stewardship, environmental cleaning, and hand hygiene.

Electronic supplementary material

The online version of this article (10.1186/s12879-018-3388-y) contains supplementary material, which is available to authorized users.

Quantifying where human acquisition of antibiotic resistance occurs: a mathematical modelling study

BACKGROUND

Antibiotic-resistant bacteria (ARB) are selected by the use of antibiotics. The rational design of interventions to reduce levels of antibiotic resistance requires a greater understanding of how and where ARB are acquired. Our aim was to determine whether acquisition of ARB occurs more often in the community or hospital setting.

METHODS

We used a mathematical model of the natural history of ARB to estimate how many ARB were acquired in each of these two environments, as well as to determine key parameters for further investigation. To do this, we explored a range of realistic parameter combinations and considered a case study of parameters for an important subset of resistant strains in England.

RESULTS

If we consider all people with ARB in the total population (community and hospital), the majority, under most clinically derived parameter combinations, acquired their resistance in the community, despite higher levels of antibiotic use and transmission of ARB in the hospital. However, if we focus on just the hospital population, under most parameter combinations a greater proportion of this population acquired ARB in the hospital.

CONCLUSIONS

It is likely that the majority of ARB are being acquired in the community, suggesting that efforts to reduce overall ARB carriage should focus on reducing antibiotic usage and transmission in the community setting. However, our framework highlights the need for better pathogen-specific data on antibiotic exposure, ARB clearance and transmission parameters, as well as the link between carriage of ARB and health impact. This is important to determine whether interventions should target total ARB carriage or hospital-acquired ARB carriage, as the latter often dominated in hospital populations.

Decoding and modelling of time series count data using Poisson hidden Markov model and Markov ordinal logistic regression models

Hidden Markov models are stochastic models in which the observations are assumed to follow a mixture distribution, but the parameters of the components are governed by a Markov chain which is unobservable. The issues related to the estimation of Poisson-hidden Markov models in which the observations are coming from mixture of Poisson distributions and the parameters of the component Poisson distributions are governed by an m-state Markov chain with an unknown transition probability matrix are explained here. These methods were applied to the data on Vibrio cholerae counts reported every month for 11-year span at Christian Medical College, Vellore, India. Using Viterbi algorithm, the best estimate of the state sequence was obtained and hence the transition probability matrix. The mean passage time between the states were estimated. The 95% confidence interval for the mean passage time was estimated via Monte Carlo simulation. The three hidden states of the estimated Markov chain are labelled as 'Low', 'Moderate' and 'High' with the mean counts of 1.4, 6.6 and 20.2 and the estimated average duration of stay of 3, 3 and 4 months, respectively. Environmental risk factors were studied using Markov ordinal logistic regression analysis. No significant association was found between disease severity levels and climate components.

Evaluating hospital infection control measures for antimicrobial-resistant pathogens using stochastic transmission models: Application to vancomycin-resistant enterococci in intensive care units

Nosocomial pathogens such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococci (VRE) are the cause of significant morbidity and mortality among hospital patients. It is important to be able to assess the efficacy of control measures using data on patient outcomes. In this paper, we describe methods for analysing such data using patient-level stochastic models which seek to describe the underlying unobserved process of transmission. The methods are applied to detailed longitudinal patient-level data on vancomycin-resistant Enterococci from a study in a US hospital with eight intensive care units (ICUs). The data comprise admission and discharge dates, dates and results of screening tests, and dates during which precautionary measures were in place for each patient during the study period. Results include estimates of the efficacy of the control measures, the proportion of unobserved patients colonized with vancomycin-resistant Enterococci, and the proportion of patients colonized on admission.

A Dynamic Transmission Model to Evaluate the Effectiveness of Infection Control Strategies

BACKGROUND

The advancement of knowledge about control of antibiotic resistance depends on the rigorous evaluation of alternative intervention strategies. The STAR*ICU trial examined the effects of active surveillance and expanded barrier precautions on acquisition of methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE) in intensive care units. We report a reanalyses of the STAR*ICU trial using a Bayesian transmission modeling framework.

METHODS

The data included admission and discharge times and surveillance test times and results. Markov chain Monte Carlo stochastic integration was used to estimate the transmission rate, importation, false negativity, and clearance separately for MRSA and VRE. The primary outcome was the intervention effect, which when less than (or greater than) zero, indicated a decreased (or increased) transmission rate attributable to the intervention.

RESULTS

The transmission rate increased in both arms from pre- to postintervention (by 20% and 26% for MRSA and VRE). The estimated intervention effect was 0.00 (95% confidence interval [CI], -0.57 to 0.56) for MRSA and 0.05 (95% CI, -0.39 to 0.48) for VRE. Compared with MRSA, VRE had a higher transmission rate (preintervention, 0.0069 vs 0.0039; postintervention, 0.0087 vs 0.0046), higher importation probability (0.22 vs 0.17), and a lower clearance rate per colonized patient-day (0.016 vs 0.035).

CONCLUSIONS

Transmission rates in the 2 treatment arms were statistically indistinguishable from the pre- to postintervention phase, consistent with the original analysis of the STAR*ICU trial. Our statistical framework was able to disentangle transmission from importation and account for imperfect testing. Epidemiological differences between VRE and MRSA were revealed.

Population Dynamics of Patients with Bacterial Resistance in Hospital Environment

During the past decades, the increase of antibiotic resistance has become a major concern worldwide. The researchers found that superbugs with new type of resistance genes (NDM-1) have two aspects of transmission characteristics; the first is that the antibiotic resistance genes can horizontally transfer among bacteria, and the other is that the superbugs can spread between humans through direct contact. Based on these two transmission mechanisms, we study the dynamics of population in hospital environment where superbugs exist. In this paper, we build three mathematic models to illustrate the dynamics of patients with bacterial resistance in hospital environment. The models are analyzed using stability theory of differential equations. Positive equilibrium points of the system are investigated and their stability analysis is carried out. Moreover, the numerical simulation of the proposed model is also performed which supports the theoretical findings.

Modeling the impact of interventions against Acinetobacter baumannii transmission in intensive care units

The efficacy of infection control interventions against Acinetobacter baumannii remains unclear, despite such information being critical for effective prevention of the transmission of this pathogen. Mathematical modeling offers an alternative to clinical trials, which may be prohibitively expensive, unfeasible or unethical, in predicting the impact of interventions. Furthermore, it allows the ability to ask key “what if” questions to evaluate which interventions have the most impact. We constructed a transmission dynamic model to quantify the effects of interventions on reducing A. baumannii prevalence and the basic reproduction ratio (R₀) in intensive care units (ICUs). We distinguished between colonization and infection, and incorporated antibiotic exposure and transmission from free-living bacteria in the environment. Under the assumptions and parameterization in our model, 25% and 18% of patients are colonized and infected with A. baumannii, respectively; and R₀ is 1.4. Improved compliance with hand hygiene (≥87%), enhanced environmental cleaning, reduced length of ICU stay of colonized patients (≤ 10 days), shorter durations of antibiotic treatment of A. baumannii (≤6 days), and isolation of infected patients combined with cleaning of isolation rooms are effective, reducing R₀ to below unity. In contrast, expediting the recovery of the intestinal microbiota (e.g. use of probiotics) is not effective. This study represents a biologically realistic model of the transmission dynamics of A. baumannii, and the most comprehensive analysis of the effectiveness of interventions against this pathogen. Our study provides important data for designing effective infection control interventions.

Characterising the Transmission Dynamics of Acinetobacter baumannii in Intensive Care Units Using Hidden Markov Models

Little is known about the transmission dynamics of Acinetobacter baumannii in hospitals, despite such information being critical for designing effective infection control measures. In the absence of comprehensive epidemiological data, mathematical modelling is an attractive approach to understanding transmission process. The statistical challenge in estimating transmission parameters from infection data arises from the fact that most patients are colonised asymptomatically and therefore the transmission process is not fully observed. Hidden Markov models (HMMs) can overcome this problem. We developed a continuous-time structured HMM to characterise the transmission dynamics, and to quantify the relative importance of different acquisition sources of A. baumannii in intensive care units (ICUs) in three hospitals in Melbourne, Australia. The hidden states were the total number of patients colonised with A. baumannii (both detected and undetected). The model input was monthly incidence data of the number of detected colonised patients (observations). A Bayesian framework with Markov chain Monte Carlo algorithm was used for parameter estimations. We estimated that 96-98% of acquisition in Hospital 1 and 3 was due to cross-transmission between patients; whereas most colonisation in Hospital 2 was due to other sources (sporadic acquisition). On average, it takes 20 and 31 days for each susceptible individual in Hospital 1 and Hospital 3 to become colonised as a result of cross-transmission, respectively; whereas it takes 17 days to observe one new colonisation from sporadic acquisition in Hospital 2. The basic reproduction ratio (R0) for Hospital 1, 2 and 3 was 1.5, 0.02 and 1.6, respectively. Our study is the first to characterise the transmission dynamics of A. baumannii using mathematical modelling. We showed that HMMs can be applied to sparse hospital infection data to estimate transmission parameters despite unobserved events and imperfect detection of the organism. Our results highlight the need to optimise infection control in ICUs.

A warning threshold for monitoring tuberculosis surveillance data: an alternative to hidden Markov model

OBJECTIVES

Although hidden Markov model (HMM) is known as a powerful tool for the detection of epidemics based on the historical data, the frequent use of such a model poses some limitation especially when decision-making is required for new observations. This study was aimed to address a warning threshold for monitoring the weekly incidences of tuberculosis as an alternative to HMM.

METHODS

We extracted the weekly counts of newly diagnosed patients with sputum smear-positive pulmonary TB from 2005 to 2011 nationwide. To detect unexpected incidences of the disease, two approaches: Serfling and HMM, were applied in presence/absence of linear, seasonal and autoregressive components. Models were subsequently evaluated in terms of goodness of fit, and their results were compared in detection of the disease phases. Then, multiple hypothetical thresholds were constructed based on the estimate of models and the optimal one was revealed through ROC curve analysis.

RESULTS

Findings from both adjusted R-square (R~2) and Bayesian information criterion (BIC) presented a higher goodness of fit for periodic autoregressive HMM (BIC = -1323.6; R~2=0.74) than other models. According to ROC analysis, better values for both Youden's index and area under curve (0. 96 and 0. 98 respectively) were obtained by the threshold based on the estimate of periodic autoregressive model.

CONCLUSIONS

As the optimal threshold presented in this study is simple in concept and has no limitation in practice, especially for monitoring new observations, we would recommend such a threshold to be used for monitoring of TB incidence data in the surveillance system.

Optimizing hospital infection control: the role of mathematical modeling

Multidrug-resistant bacteria are major causes of nosocomial infections and are associated with considerable morbidity, mortality, and healthcare costs. Preventive strategies have therefore become increasingly important. Mathematical modeling has been widely used to understand the transmission dynamics of nosocomial infections and the quantitative effects of infection control measures. This review will explore the principles of mathematical modeling used in nosocomial infections and discuss the effectiveness of infection control measures investigated using mathematical modeling.

Nosocomial transmission of C. difficile in English hospitals from patients with symptomatic infection

Background

Recent evidence suggests that less than one-quarter of patients with symptomatic nosocomial Clostridium difficile infections (CDI) are linked to other in-patients. However, this evidence was limited to one geographic area. We aimed to investigate the level of symptomatic CDI transmission in hospitals located across England from 2008 to 2012.

Methods

A generalized additive mixed-effects Poisson model was fitted to English hospital-surveillance data. After adjusting for seasonal fluctuations and between-hospital variation in reported CDI over time, possible clustering (transmission between symptomatic in-patients) of CDI cases was identified. We hypothesised that a temporal proximity would be reflected in the degree of correlation between in-hospital CDI cases per week. This correlation was modelled through a latent autoregressive structure of order 1 (AR(1)).

Findings

Forty-six hospitals (33 general, seven specialist, and six teaching hospitals) located in all English regions met our criteria. In total, 12,717 CDI cases were identified; seventy-five per cent of these occurred >48 hours after admission. There were slight increases in reports during winter months. We found a low, but statistically significant, correlation between successive weekly CDI case incidences (phi = 0.029, 95%CI: 0.009–0.049). This correlation was five times stronger in a subgroup analysis restricted to teaching hospitals (phi = 0.104, 95%CI: 0.048–0.159).

Conclusions

The results suggest that symptomatic patient-to-patient transmission has been a source of CDI-acquisition in English hospitals in recent years, and that this might be a more important transmission route in teaching hospitals. Nonetheless, the weak correlation indicates that, in line with recent evidence, symptomatic cases might not be the primary source of nosocomial CDI in England.

A modeling framework for the evolution and spread of antibiotic resistance: literature review and model categorization

Antibiotic-resistant infections complicate treatment and increase morbidity and mortality. Mathematical modeling has played an integral role in improving our understanding of antibiotic resistance. In these models, parameter sensitivity is often assessed, while model structure sensitivity is not. To examine the implications of this, we first reviewed the literature on antibiotic-resistance modeling published between 1993 and 2011. We then classified each article's model structure into one or more of 6 categories based on the assumptions made in those articles regarding within-host and population-level competition between antibiotic-sensitive and antibiotic-resistant strains. Each model category has different dynamic implications with respect to how antibiotic use affects resistance prevalence, and therefore each may produce different conclusions about optimal treatment protocols that minimize resistance. Thus, even if all parameter values are correctly estimated, inferences may be incorrect because of the incorrect selection of model structure. Our framework provides insight into model selection.

Modelling the transmission of healthcare associated infections: a systematic review

Background

Dynamic transmission models are increasingly being used to improve our understanding of the epidemiology of healthcare-associated infections (HCAI). However, there has been no recent comprehensive review of this emerging field. This paper summarises how mathematical models have informed the field of HCAI and how methods have developed over time.

Methods

MEDLINE, EMBASE, Scopus, CINAHL plus and Global Health databases were systematically searched for dynamic mathematical models of HCAI transmission and/or the dynamics of antimicrobial resistance in healthcare settings.

Results

In total, 96 papers met the eligibility criteria. The main research themes considered were evaluation of infection control effectiveness (64%), variability in transmission routes (7%), the impact of movement patterns between healthcare institutes (5%), the development of antimicrobial resistance (3%), and strain competitiveness or co-colonisation with different strains (3%). Methicillin-resistant Staphylococcus aureus was the most commonly modelled HCAI (34%), followed by vancomycin resistant enterococci (16%). Other common HCAIs, e.g. Clostridum difficile, were rarely investigated (3%). Very few models have been published on HCAI from low or middle-income countries.

The first HCAI model has looked at antimicrobial resistance in hospital settings using compartmental deterministic approaches. Stochastic models (which include the role of chance in the transmission process) are becoming increasingly common. Model calibration (inference of unknown parameters by fitting models to data) and sensitivity analysis are comparatively uncommon, occurring in 35% and 36% of studies respectively, but their application is increasing. Only 5% of models compared their predictions to external data.

Conclusions

Transmission models have been used to understand complex systems and to predict the impact of control policies. Methods have generally improved, with an increased use of stochastic models, and more advanced methods for formal model fitting and sensitivity analyses. Insights gained from these models could be broadened to a wider range of pathogens and settings. Improvements in the availability of data and statistical methods could enhance the predictive ability of models.

Empirical mode decomposition and k-nearest embedding vectors for timely analyses of antibiotic resistance trends

Background

Antibiotic resistance is a major worldwide public health concern. In clinical settings, timely antibiotic resistance information is key for care providers as it allows appropriate targeted treatment or improved empirical treatment when the specific results of the patient are not yet available.

Objective

To improve antibiotic resistance trend analysis algorithms by building a novel, fully data-driven forecasting method from the combination of trend extraction and machine learning models for enhanced biosurveillance systems.

Methods

We investigate a robust model for extraction and forecasting of antibiotic resistance trends using a decade of microbiology data. Our method consists of breaking down the resistance time series into independent oscillatory components via the empirical mode decomposition technique. The resulting waveforms describing intrinsic resistance trends serve as the input for the forecasting algorithm. The algorithm applies the delay coordinate embedding theorem together with the k-nearest neighbor framework to project mappings from past events into the future dimension and estimate the resistance levels.

Results

The algorithms that decompose the resistance time series and filter out high frequency components showed statistically significant performance improvements in comparison with a benchmark random walk model. We present further qualitative use-cases of antibiotic resistance trend extraction, where empirical mode decomposition was applied to highlight the specificities of the resistance trends.

Conclusion

The decomposition of the raw signal was found not only to yield valuable insight into the resistance evolution, but also to produce novel models of resistance forecasters with boosted prediction performance, which could be utilized as a complementary method in the analysis of antibiotic resistance trends.

Limits of patient isolation measures to control extended-spectrum beta-lactamase-producing Enterobacteriaceae: model-based analysis of clinical data in a pediatric ward

Background

Extended-spectrum beta-lactamase–producing Enterobacteriaceae (ESBL-E) are a growing concern in hospitals and the community. How to control the nosocomial ESBL-E transmission is a matter of debate. Contact isolation of patients has been recommended but evidence supporting it in non-outbreak settings has been inconclusive.

Methods

We used stochastic transmission models to analyze retrospective observational data from a two-phase intervention in a pediatric ward, successively implementing single-room isolation and patient cohorting in an isolation ward, combined with active ESBL-E screening.

Results

For both periods, model estimates suggested reduced transmission from isolated/cohorted patients. However, most of the incidence originated from sporadic sources (i.e. independent of cross-transmission), unaffected by the isolation measures. When sporadic sources are high, our model predicted that even substantial efforts to prevent transmission from carriers would have limited impact on ESBL-E rates.

Conclusions

Our results provide evidence that, considering the importance of sporadic acquisition, e.g. endogenous selection of resistant strains following antibiotic treatment, contact-isolation measures alone might not suffice to control ESBL-E. They also support the view that estimating cross-transmission extent is key to predicting the relative success of contact-isolation measures. Mathematical models could prove useful for those estimations and guide decisions concerning the most effective control strategy.

Contribution of mathematical modeling to the fight against bacterial antibiotic resistance

PURPOSE OF REVIEW

Modeling of antibiotic resistance in pathogenic bacteria responsible for human disease has developed considerably over the last decade. Herein, we summarize the main published studies to illustrate the contribution of models for understanding both within-host and population-based phenomena. We then suggest possible topics for future studies.

RECENT FINDINGS

Model building of bacterial resistance has involved epidemiologists, biologists and modelers with two different objectives. First, modeling has helped largely in identifying and understanding the factors and biological phenomena responsible for the emergence and spread of resistant strains. Second, these models have become important decision support tools for medicine and public health.

SUMMARY

Major improvements of models in the coming years should take into account specific pathogen characteristics (resistance mechanisms, multiple colonization phenomena, cooperation and competition among species) and better description of the contacts associated with transmission risk within populations.

Transmission dynamics of methicillin-resistant Staphylococcus aureus in a medical intensive care unit in India

BACKGROUND

Methicillin-resistant Staphylococcus aureus (MRSA) is a global pathogen and an important but seldom investigated cause of morbidity and mortality in lower and middle-income countries where it can place a major burden on limited resources. Quantifying nosocomial transmission in resource-poor settings is difficult because molecular typing methods are prohibitively expensive. Mechanistic statistical models can overcome this problem with minimal cost. We analyse the transmission dynamics of MRSA in a hospital in south India using one such approach and provide conservative estimates of the organism's economic burden.

METHODS AND FINDINGS

Fifty months of MRSA infection data were collected retrospectively from a Medical Intensive Care Unit (MICU) in a tertiary hospital in Vellore, south India. Data were analysed using a previously described structured hidden Markov model. Seventy-two patients developed MRSA infections and, of these, 49 (68%) died in the MICU. We estimated that 4.2% (95%CI 1.0, 19.0) of patients were MRSA-positive when admitted, that there were 0.39 MRSA infections per colonized patient month (0.06, 0.73), and that the ward-level reproduction number for MRSA was 0.42 (0.08, 2.04). Anti-MRSA antibiotic treatment costs alone averaged $124/patient, over three times the monthly income of more than 40% of the Indian population.

CONCLUSIONS

Our analysis of routine data provides the first estimate of the nosocomial transmission potential of MRSA in India. The high levels of transmission estimated underline the need for cost-effective interventions to reduce MRSA transmission in hospital settings in low and middle income countries.

Recent advances in the analysis of behavioural organization and interpretation as indicators of animal welfare

While the incorporation of mathematical and engineering methods has greatly advanced in other areas of the life sciences, they have been under-utilized in the field of animal welfare. Exceptions are beginning to emerge and share a common motivation to quantify 'hidden' aspects in the structure of the behaviour of an individual, or group of animals. Such analyses have the potential to quantify behavioural markers of pain and stress and quantify abnormal behaviour objectively. This review seeks to explore the scope of such analytical methods as behavioural indicators of welfare. We outline four classes of analyses that can be used to quantify aspects of behavioural organization. The underlying principles, possible applications and limitations are described for: fractal analysis, temporal methods, social network analysis, and agent-based modelling and simulation. We hope to encourage further application of analyses of behavioural organization by highlighting potential applications in the assessment of animal welfare, and increasing awareness of the scope for the development of new mathematical methods in this area.

Nosocomial transmission clusters and risk factors inMoraxella catarrhalis

We report an objective examination of nosocomial transmission events derived from long-term (10-year) data from a single medical centre. Cluster analysis, based on the temporal proximity of genetically identical isolates of the respiratory pathogenMoraxella catarrhalis, identified 40 transmission events involving 33 of the 52 genotypes represented by multiple isolates. There was no evidence of highly transmissible or outbreak-prone genotypes. Although most clusters were small (mean size 3·6 isolates) and of short duration (median duration 25 days), clustering accounted for 38·7% of all isolates. Significant risk factors for clustering were multi-bed wards, and winter and spring season, but bacterial antibiotic resistance, manifested as the ability to produce a β-lactamase was not a risk factor. The use of cluster analysis to identify transmission events and its application to long-term data demonstrate an approach to pathogen transmission that should find wide application beyond hospital populations.

An augmented data method for the analysis of nosocomial infection data

The analysis of nosocomial infection data for communicable pathogens is complicated by two facts. First, typical pathogens more commonly cause asymptomatic colonization than overt disease, so transmission can be only imperfectly observed through a sequence of surveillance swabs, which themselves have imperfect sensitivity. Any given set of swab results can therefore be consistent with many different patterns of transmission. Second, data are often highly dependent: the colonization status of one patient affects the risk for others, and, in some wards, repeated admissions are common. Here, the authors present a method for analyzing typical nosocomial infection data consisting of results from arbitrarily timed screening swabs that overcomes these problems and enables simultaneous estimation of transmission and importation parameters, duration of colonization, swab sensitivity, and ward- and patient-level covariates. The method accounts for dependencies by using a mechanistic stochastic transmission model, and it allows for uncertainty in the data by imputing the imperfectly observed colonization status of patients over repeated admissions. The approach uses a Markov chain Monte Carlo algorithm, allowing inference within a Bayesian framework. The method is applied to illustrative data from an interrupted time-series study of vancomycin-resistant enterococci transmission in a hematology ward.

Outbreaks of nosocomial infections: lessons learned and perspectives

PURPOSE OF REVIEW

The review summarizes the results of selected outbreak reports and systematic analyses of nosocomial outbreaks from 2007 and focuses on different aspects of hospital epidemiology and infection control.

RECENT FINDINGS

A single outbreak report is likely be influenced by the local setting. In contrast, a systematic analysis of a large number of similar outbreaks draws a much better picture of the real conditions on the pathogen's reservoirs, on modes of transmission, and on appropriate infection control measures to prevent the spread of the microorganism. Isolation, unit closures, sick leave, cleaning, and diagnostic/therapeutic measures may lead to enormous costs during an outbreak. Thus, cost calculations of outbreaks should be performed to justify future expenses for infection control. Mathematic modeling is a fairly new approach to estimate the risk of pathogen transmission in outbreak settings. Examples are shown to differentiate between epidemic and sporadic infections and to evaluate the influence of infection control interventions.

SUMMARY

Outbreak reports may add some very important information to the understanding of transmission and infection control. There is a need for a more structured publication of nosocomial outbreaks to ensure that no key data are lacking in the article.