Analysis of correlated discrete observations: background, examples and solutions

Efficacy of bacterial vaccines to prevent respiratory disease in swine: a systematic review and network meta-analysis

A systematic review and network meta-analysis (MA) was conducted to address the question, 'What is the efficacy of bacterial vaccines to prevent respiratory disease in swine?' Four electronic databases and the grey literature were searched to identify clinical trials in healthy swine where at least one intervention arm was a commercially available vaccine for one or more bacterial pathogens associated with respiratory disease in swine, including Mycoplasma hyopneumoniae, Actinobacillus pleuropneumonia, Actinobacillus suis, Bordetella bronchiseptica, Pasteurella multocida, Stretococcus suis, Haemophils parasuis, and Mycoplasma hyorhinis. To be eligible, trials had to measure at least one of the following outcomes: incidence of clinical morbidity, mortality, lung lesions, or total antibiotic use. There were 179 eligible trials identified in 146 publications. Network MA was undertaken for morbidity, mortality, and the presence or absence of non-specific lung lesions. However, there was not a sufficient body of research evaluating the same interventions and outcomes to allow a meaningful synthesis of the comparative efficacy of the vaccines. To build this body of research, additional rigor in trial design and analysis, and detailed reporting of trial methods and results are warranted.

The efficacy of litter management strategies to prevent morbidity and mortality in broiler chickens: a systematic review and network meta-analysis

A systematic review and network meta-analysis (NMA) were conducted to address the question, 'What is the efficacy of litter management strategies to reduce morbidity, mortality, condemnation at slaughter, or total antibiotic use in broilers?' Eligible studies were clinical trials published in English evaluating the efficacy of litter management in broilers on morbidity, condemnations at slaughter, mortality, or total antibiotic use. Multiple databases and two conference proceedings were searched for relevant literature. After relevance screening and data extraction, there were 50 trials evaluating litter type, 22 trials evaluating litter additives, 10 trials comparing fresh to re-used litter, and six trials evaluating floor type. NMAs were conducted for mortality (61 trials) and for the presence or absence of footpad lesions (15 trials). There were no differences in mortality among the litter types, floor types, or additives. For footpad lesions, peat moss appeared beneficial compared to straw, based on a small number of comparisons. In a pairwise meta-analysis, there was no association between fresh versus used litter on the risk of mortality, although there was considerable heterogeneity among studies (I2 = 66%). There was poor reporting of key design features in many studies, and analyses rarely accounted for non-independence of observations within flocks.

A systematic review of the efficacy of antibiotics for the prevention of swine respiratory disease

Prevention and control of respiratory disease is a major contributor to antibiotic use in swine. A systematic review was conducted to address the question, 'What is the comparative efficacy of antimicrobials for the prevention of swine respiratory disease?' Eligible studies were controlled trials published in English evaluating prophylactic antibiotics in swine, where clinical morbidity, mortality, or total antibiotic use was assessed. Four databases and the gray literature were searched for relevant articles. Two reviewers working independently screened titles and abstracts for eligibility followed by full-text articles, and then extracted data and evaluated risk of bias for eligible trials. There were 44 eligible trials from 36 publications. Clinical morbidity was evaluated in eight trials where antibiotics were used in nursery pigs and 10 trials where antibiotics were used in grower pigs. Mortality was measured in 22 trials in nursery pigs and 12 trials in grower pigs. There was heterogeneity in the antibiotic interventions and comparisons published in the literature; thus, there was insufficient evidence to allow quantification of the efficacy, or relative efficacy, of antibiotic interventions. Concerns related to statistical non-independence and quality of reporting were noted in the included trials.

Reprint of "Epidemiology of brucellosis, Q Fever and Rift Valley Fever at the human and livestock interface in northern Côte d'Ivoire"

Northern Côte d'Ivoire is the main livestock breeding zone and has the highest livestock cross-border movements in Côte d'Ivoire. The aim of this study was to provide updated epidemiological data on three neglected zoonotic diseases, namely brucellosis, Q Fever and Rift Valley Fever (RVF). We conducted three-stage cross-sectional cluster surveys in livestock and humans between 2012 and 2014 in a random selection of 63 villages and a sample of 633 cattle, 622 small ruminants and 88 people. We administered questionnaires to capture risk factors and performed serological tests including the Rose Bengal Plate Test (RBPT), Brucella spp. indirect and competitive ELISAs, Coxiella burnetii indirect ELISA and RVF competitive ELISA. The human seroprevalence for Brucella spp. was 5.3%. RBPT-positive small ruminants tested negative by the indirect ELISA. The seroprevalence of Brucella spp. in cattle adjusted for clustering was 4.6%. Cattle aged 5-8 years had higher odds of seropositivity (OR=3.5) than those aged ≤4years. The seropositivity in cattle was associated with having joint hygromas (OR=9), sharing the pastures with small ruminants (OR=5.8) and contact with pastoralist herds (OR=11.3). The seroprevalence of Q Fever was 13.9% in cattle, 9.4% in sheep and 12.4% in goats. The seroprevalence of RVF was 3.9% in cattle, 2.4% in sheep and 0% in goats. Seropositive ewes had greater odds (OR=4.7) of abortion than seronegative ones. In cattle, a shorter distance between the night pens and nearest permanent water bodies was a protective factor (OR=0.1). The study showed that the exposure to the three zoonoses is rather low in northern Côte d'Ivoire. Within a One Health approach, cost-benefit and cost-effectiveness of control measures should be assessed for an integrated control.

Survey for Listeria monocytogenes in and on Ready-to-Eat Foods from Retail Establishments in the United States (2010 through 2013): Assessing Potential Changes of Pathogen Prevalence and Levels in a Decade

A multiyear interagency Listeria monocytogenes Market Basket Survey was undertaken for selected refrigerated ready-to-eat foods purchased at retail in four FoodNet sites in the United States. Food samples from 16 food categories in six broad groups (seafood, produce, dairy, meat, eggs, and combination foods) were collected weekly at large national chain supermarkets and independent grocery stores in California, Maryland, Connecticut, and Georgia for 100 weeks between December 2010 and March 2013. Of the 27,389 total samples, 116 samples tested positive by the BAX PCR system for L. monocytogenes , and the pathogen was isolated and confirmed for 102 samples. Among the 16 food categories, the proportion of positive samples (i.e., without considering clustering effects) based on recovery of a viable isolate of L. monocytogenes ranged from 0.00% (95% confidence interval: 0.00, 0.18) for the category of soft-ripened and semisoft cheese to 1.07% (0.63, 1.68) for raw cut vegetables. Among the 571 samples that tested positive for Listeria-like organisms, the proportion of positive samples ranged from 0.79% (0.45, 1.28) for soft-ripened and semisoft cheese to 4.76% (2.80, 7.51) for fresh crab meat or sushi. Across all 16 categories, L. monocytogenes contamination was significantly associated with the four states (P < 0.05) but not with the packaging location (prepackaged by the manufacturer versus made and/or packaged in the store), the type of store (national chain versus independent), or the season. Among the 102 samples positive for L. monocytogenes , levels ranged from <0.036 most probable number per g to 6.1 log CFU/g. For delicatessen (deli) meats, smoked seafood, seafood salads, soft-ripened and semisoft cheeses, and deli-type salads without meat, the percentage of positive samples was significantly lower (P < 0.001) in this survey than that reported a decade ago based on comparable surveys in the United States. Use of mixed logistic regression models to address clustering effects with regard to the stores revealed that L. monocytogenes prevalence ranged from 0.11% (0.03, 0.34) for sprouts (prepackaged) to 1.01% (0.58, 1.74) for raw cut vegetables (prepackaged).

Epidemiology of brucellosis, Q Fever and Rift Valley Fever at the human and livestock interface in northern Côte d'Ivoire

Northern Côte d'Ivoire is the main livestock breeding zone and has the highest livestock cross-border movements in Côte d'Ivoire. The aim of this study was to provide updated epidemiological data on three neglected zoonotic diseases, namely brucellosis, Q Fever and Rift Valley Fever (RVF). We conducted three-stage cross-sectional cluster surveys in livestock and humans between 2012 and 2014 in a random selection of 63 villages and a sample of 633 cattle, 622 small ruminants and 88 people. We administered questionnaires to capture risk factors and performed serological tests including the Rose Bengal Plate Test (RBPT), Brucella spp. indirect and competitive ELISAs, Coxiella burnetii indirect ELISA and RVF competitive ELISA. The human seroprevalence for Brucella spp. was 5.3%. RBPT-positive small ruminants tested negative by the indirect ELISA. The seroprevalence of Brucella spp. in cattle adjusted for clustering was 4.6%. Cattle aged 5-8 years had higher odds of seropositivity (OR=3.5) than those aged ≤4years. The seropositivity in cattle was associated with having joint hygromas (OR=9), sharing the pastures with small ruminants (OR=5.8) and contact with pastoralist herds (OR=11.3). The seroprevalence of Q Fever was 13.9% in cattle, 9.4% in sheep and 12.4% in goats. The seroprevalence of RVF was 3.9% in cattle, 2.4% in sheep and 0% in goats. Seropositive ewes had greater odds (OR=4.7) of abortion than seronegative ones. In cattle, a shorter distance between the night pens and nearest permanent water bodies was a protective factor (OR=0.1). The study showed that the exposure to the three zoonoses is rather low in northern Côte d'Ivoire. Within a One Health approach, cost-benefit and cost-effectiveness of control measures should be assessed for an integrated control.

Explanation and Elaboration Document for the STROBE-Vet Statement: Strengthening the Reporting of Observational Studies in Epidemiology-Veterinary Extension

The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement was first published in 2007 and again in 2014. The purpose of the original STROBE was to provide guidance for authors, reviewers, and editors to improve the comprehensiveness of reporting; however, STROBE has a unique focus on observational studies. Although much of the guidance provided by the original STROBE document is directly applicable, it was deemed useful to map those statements to veterinary concepts, provide veterinary examples, and highlight unique aspects of reporting in veterinary observational studies. Here, we present the examples and explanations for the checklist items included in the STROBE-Vet statement. Thus, this is a companion document to the STROBE-Vet statement methods and process document (JVIM_14575 "Methods and Processes of Developing the Strengthening the Reporting of Observational Studies in Epidemiology-Veterinary (STROBE-Vet) Statement" undergoing proofing), which describes the checklist and how it was developed.

High lifetime and reproductive performance of sows on southern European Union commercial farms can be predicted by high numbers of pigs born alive in parity one

Our objectives were 1) to compare reproductive performance across parity and lifetime performance in sow groups categorized by the number of pigs born alive (PBA) in parity 1 and 2) to examine the factors associated with more PBA in parity 1. We analyzed 476,816 parity records and 109,373 lifetime records of sows entered into 125 herds from 2008 to 2010. Sows were categorized into 4 groups based on the 10th, 50th, and 90th percentiles of PBA in parity 1 as follows: 7 pigs or fewer, 8 to 11 pigs, 12 to 14 pigs, and 15 pigs or more. Generalized linear models were applied to the data. For reproductive performance across parity, sows that had 15 or more PBA in parity 1 had 0.5 to 1.8 more PBA in any subsequent parity than the other 3 PBA groups ( P< 0.05). In addition, they had 2.8 to 5.4% higher farrowing rates in parities 1 through 3 than sows that had 7 or fewer PBA (P < 0.05). However, there were no differences between the sow PBA groups for weaning-to-first-mating interval in any parity (P ≥ 0.37). For lifetime performance, sows that had 15 or more PBA in parity 1 had 4.4 to 26.1 more lifetime PBA than sows that had 14 or fewer PBA (P < 0.05). Also, for sows that had 14 or fewer PBA in parity 1, those that were first mated at 229 d old (25th percentile) or earlier had 2.9 to 3.3 more lifetime PBA than those first mated at 278 d old (75th percentile) or later (P < 0.05). Factors associated with fewer PBA in parity 1 were summer mating and lower age of gilts at first mating (AFM; P < 0.05) but not reservice occurrences (P = 0.34). Additionally, there was a 2-way interaction between mated month groups and AFM for PBA in parity 1 (P < 0.05); PBA in parity 1 sows mated from July to December increased nonlinearly by 0.3 to 0.4 pigs when AFM increased from 200 to 310 d old (P < 0.05). However, the same rise in AFM had no significant effect on the PBA of sows mated between January and June (P ≥ 0.17). In conclusion, high PBA in parity 1 can be used to predict that a sow will have high reproductive performance and lifetime performance. Also, the data indicate that the upper limit of AFM for mating between July and December should be 278 d old.

Making the most of clustered data in laboratory animal research using multi-level models

In the following review article, I address the fitting of multi-level models for the analysis of hierarchical data in laboratory animal medicine. Using an example of paternal dietary effects on the weight of offspring in a mouse model, this review outlines the reasons and benefits of using a multi-level modeling approach. To start, the concept of clustered/autocorrelated data is introduced, and the implications of ignoring the effects of clustered data on measures of association/model coefficients and their statistical significance are discussed. The limitations of other methods compared with multi-level modeling for analyzing clustered data are addressed in terms of statistical power, control of potential confounding effects associated with group membership, proper estimation of associations and their statistical significance, and adjusting for multiple levels of clustering. In addition, the benefits of being able to estimate variance partition coefficients and intra-class correlation coefficients from multi-level models is described, and the concepts of more complex correlation structures and various methods for fitting multi-level models are introduced. The current state of learning materials including textbooks, websites, and software for the nonstatistician is outlined to describe the accessibility of multi-level modeling approaches for laboratory animal researchers.

Risk factors associated with enteric septicemia of catfish on Mississippi commercial catfish farms

A gram-negative bacterium, Edwardsiella ictaluri, is the cause of enteric septicemia of catfish (ESC), which is one of the most prevalent bacterial diseases in farm-raised catfish. The objective of this study was to identify risk factors associated with ESC mortalities and are reported by farm personnel. To identify risk factors a catfish management database was developed. The odds ratios (OR) of the final multivariable logistic regression model were: (1) volume of the pond (OR, 0.56), (2) interval from harvest until a mortality event (OR, 1.49), (3) interval from stocking until a mortality event (OR, 0.52), (4) nitrite measured within 14 d of a mortality (OR, 3.49), (5) total ammonia measured within 14 d of a mortality (OR, 20.48), and (6) sum of feed fed for 14 d prior to the disease outbreak (OR, 1.02), all of which were significantly (P ≤ 0.05) associated with ESC occurrence. This study showed that some commonly recorded production variables were associated with ESC outbreaks and if monitored could help identify "at risk" ponds prior to disease outbreaks.

Pond-level risk factors associated with columnaris disease on Mississippi commercial catfish farms

A large commercial catfish enterprise encompassing over 500 food fish ponds from five farms covering multiple counties in the Mississippi Delta was included in this analysis of columnaris risk factors. A gram-negative bacterium, Flavobacterium columnare, is the cause of columnaris disease and is considered the second-most prevalent bacterial disease in farm-raised catfish. The objective of this study was to determine if pond-level risk factors reported by farm personnel were associated with columnaris disease mortalities. To identify risk factors affecting susceptibility of farm-raised channel catfish Ictalurus punctatus to columnaris disease, a Catfish Management database was developed. Logistic regression was used to model the relationships between probability of columnaris in ponds and risk factors examined. Generalized linear mixed models incorporating hierarchically structured random effects of ponds and one or more fixed-effects risk factors were fitted. In the screening process, each risk factor was evaluated in the basic model as a single fixed-effects factor, and if associated with the outcome (P ≤ 0.20), was retained for development of multivariable models. Two multivariable logistic regression models were constructed from data collected at the pond level by producers. The first was constructed from data in which water quality was not considered. Pond depth and reduced feed consumption for a 14-d period prior to disease outbreaks measured on a per hectare basis were significantly (P ≤ 0.05) associated with columnaris disease. The second, in which water quality variables were also considered, pond depth, reduced feed consumption, shorter intervals from stocking to disease outbreaks, and total ammonia nitrogen were significantly (P ≤ 0.05) associated with columnaris occurrence. This study showed some commonly recorded production variables were associated with columnaris disease outbreaks and, if monitored, could help identify "at risk" ponds before disease outbreaks occur.

Case control study to investigate risk factors for bovine neonatal pancytopenia (BNP) in young calves in southern Germany

A case control study on farm level was conducted at the Clinic for Ruminants, LMU Munich, to identify possible risk factors associated with the observed increase in numbers of calves showing clinical signs of Bovine Neonatal Pancytopenia (BNP) since 2006 in southern Germany. Interviews were conducted between August 2008 and June 2010.

The characteristics of 56 dairy farms with at least one confirmed case of BNP (thrombocytopenia and leukocytopenia and/or typical findings in post-mortem examination and bone marrow histology) were compared with those of two sets of 50 control dairy farms each, with no history of BNP. The first set of 50 control farms was selected randomly from veterinary practices which had never observed a BNP case on the farms they serviced. The second set of 50 control farms was matched by the veterinary practices which had provided case farms. Two separate analyses were conducted: (1) case farms (n = 56) vs. randomly selected control farms (n = 50) and (2) case farms (n = 56) vs. a matched set of control farms (n = 50). All variables with p < 0.2 in the univariable analysis were included in stepwise logistic regression models. In the first analysis, only the use of PregSure^® BVD vaccine was positively associated with BNP with an odds ratio of 1292 (95% CI: 114–14707). In the second analysis, conditional logistic regression models did not converge, therefore non-conditional logistic regression models were conducted. In the non-conditional analysis five variables remained in the model, three of which were negatively associated with BNP: the use of vitamin E and selenium, the frequent use of mastitis tubes, and the use of stem growth regulators in grain production. The use of prophylactic measures (such as control of parasites or vaccination of calves against respiratory disease) was positively associated with BNP with an odds ratio of 14.3 as well as the use of PregSure^® BVD vaccine with an odds ratio of 426 (95% CI: 20–9095).

An empirical comparison of methods for analyzing correlated data from a discrete choice survey to elicit patient preference for colorectal cancer screening

Background

A discrete choice experiment (DCE) is a preference survey which asks participants to make a choice among product portfolios comparing the key product characteristics by performing several choice tasks. Analyzing DCE data needs to account for within-participant correlation because choices from the same participant are likely to be similar. In this study, we empirically compared some commonly-used statistical methods for analyzing DCE data while accounting for within-participant correlation based on a survey of patient preference for colorectal cancer (CRC) screening tests conducted in Hamilton, Ontario, Canada in 2002.

Methods

A two-stage DCE design was used to investigate the impact of six attributes on participants' preferences for CRC screening test and willingness to undertake the test. We compared six models for clustered binary outcomes (logistic and probit regressions using cluster-robust standard error (SE), random-effects and generalized estimating equation approaches) and three models for clustered nominal outcomes (multinomial logistic and probit regressions with cluster-robust SE and random-effects multinomial logistic model). We also fitted a bivariate probit model with cluster-robust SE treating the choices from two stages as two correlated binary outcomes. The rank of relative importance between attributes and the estimates of β coefficient within attributes were used to assess the model robustness.

Results

In total 468 participants with each completing 10 choices were analyzed. Similar results were reported for the rank of relative importance and β coefficients across models for stage-one data on evaluating participants' preferences for the test. The six attributes ranked from high to low as follows: cost, specificity, process, sensitivity, preparation and pain. However, the results differed across models for stage-two data on evaluating participants' willingness to undertake the tests. Little within-patient correlation (ICC ≈ 0) was found in stage-one data, but substantial within-patient correlation existed (ICC = 0.659) in stage-two data.

Conclusions

When small clustering effect presented in DCE data, results remained robust across statistical models. However, results varied when larger clustering effect presented. Therefore, it is important to assess the robustness of the estimates via sensitivity analysis using different models for analyzing clustered data from DCE studies.

Mastitis therapy and antimicrobial susceptibility: a multispecies review with a focus on antibiotic treatment of mastitis in dairy cattle

Mastitis occurs in numerous species. Antimicrobial agents are used for treatment of infectious mastitis in dairy cattle, other livestock, companion animals, and humans. Mastitis is an economically important disease of dairy cattle and most mastitis research has focused on epidemiology and control of bovine mastitis. Antibiotic treatment of clinical and subclinical mastitis in dairy cattle is an established component of mastitis control programs. Research on the treatment of clinical and subclinical mastitis in other dairy species such as sheep and goats has been less frequent, although the general principles of mastitis therapy in small ruminants are similar to those of dairy cattle. Research on treatment of clinical mastitis in humans is limited and as for other species empirical treatment of mastitis appears to be common. While antimicrobial susceptibility testing is recommended to direct treatment decisions in many clinical settings, the use of susceptibility testing for antibiotic selection for mastitis treatments of dairy cattle has been challenged in a number of publications. The principle objective of this review is to summarize the literature evaluating the question, "Does antimicrobial susceptibility predict treatment outcome for intramammary infections caused by common bacterial pathogens?" This review also addresses current issues related to antimicrobial use and treatment decisions for mastitis in dairy cattle. Information on treatment of mastitis in other species, including humans, is included although research appears to be limited. Issues related to study design, gaps in current knowledge and opportunities for future research are identified for bovine mastitis therapy.

Correlated time to event data: Modeling repeated clinical mastitis data from dairy cattle in New York State

Mastitis is the most prevalent production disease in dairy herds worldwide and is considered to be the most economically important disease of dairy cattle. Modeling the risk of cows contracting mastitis is therefore of great interest for both targeting prevention programs and evaluating treatment protocols. Clinical mastitis (CM) is a disease of recurrent nature, thus correlation between the subsequent events within one cow may be present. This would violate the assumption behind most statistical time-to-event models. In the case of time to event models, the semi-parametric Cox regression models have become the default tool in modeling the time to an event. Limited methods are currently available to evaluate marginal and random (frailty) effects to account for multiple correlation sources. The objective of this study was to explore the implications of using several Cox or related semi-parametric or parametric models to estimate the hazard for CM in the presence of correlation between events. We evaluated the Andersen-Gill model which uses robust standard errors to account for the correlation, the Conditional Anderson-Gill model that uses stratification to account for event dependence, the Frailty model that introduces a random term to account for unobserved (cow level) heterogeneity, and a related generalized linear mixed model that uses Poisson regression to allow multi-level modeling of time-to-event data. We analyzed data on the occurrence of CM from five dairy farms in New York State. Data were from 8206 cows with 721, 275, 119, and 57 first, second, third, and fourth occurrences of CM, respectively, in the same lactation. The analysis of our sample dataset demonstrated that both cow- and farm-level correlation are present in the case of CM. The Conditional Frailty model was able to model one source of correlation in a random effect and one in a fixed effect. Poisson modeling allowed for simultaneous estimation of within cow correlation and within herd correlation.

Risk factors associated with Salmonella status of broiler flocks delivered to grow-out farms

In a prospective field observational study in the southeastern USA, we sampled gastrointestinal (GI) tracts from chicks of 65 broiler flocks delivered to conventional grow-out farms for rearing. The flocks were hatched at seven broiler hatcheries. The mean within-flock prevalence of Salmonella-positive samples was 6.5% and ranged from 0% to 86.7%. Of the 65 flocks studied, 25 (38.5%) had at least one Salmonella-positive sample. Accounting for confounding variability among the hatcheries and broiler companies, we tested whether the probability of detecting Salmonella in GI tracts of the chicks delivered was associated with certain characteristics of parent breeder flocks; hatchery production volume; hatchery ventilation system; hatchery egg-room conditions; egg incubation, candling, hatching, eggshell and bird separation, and bird-processing procedures; management of hatchery-to-farm transportation; day of week of hatch; weather conditions during transportation; or season of the hatch. Two risk factor models were adopted. The first model indicated that a greater number of parent flocks, manual separation of eggshell and bird, and a greater amount of fluff and feces on tray liners used during hatchery-to-farm transportation at delivery were associated with increased probability of detecting Salmonella in chick GI tracts, whereas a greater number of birds in the delivery vehicle was associated with decreased probability. The second model indicated that broiler flocks hatched on Tuesdays versus either Mondays or Thursdays (with no hatches on Wednesdays, Fridays or week-ends), increased average hatchability of the eggs from the parent flocks, and greater amounts of fluff and feces on the transport tray liners at delivery were all associated with increased probability of detecting Salmonella in chick GI tracts. The results of this study suggest potential management decisions to lessen Salmonella contamination of broilers supplied by commercial hatcheries and areas for further research.

Spatial and temporal clustering of calcium oxalate and magnesium ammonium phosphate uroliths in dogs living in Ontario, Canada between 1998 and 2006

Using the spatial scan statistic with a Bernoulli model, in a comparison of the two most common canine uroliths, calcium oxalate (CaOx) and magnesium ammonium phosphate (struvite) we determined whether there was evidence of spatial and/or temporal clustering of each urolith type based on canine submissions from Ontario to the Canadian Veterinary Urolith Centre (CVUC) between 1998 and 2006. During this period, there were 10,478 canine submissions, excluding cases that were identified as recurrent. We were able to georeference approximately 93% of these incident cases. After adjusting for spatial and temporal distributions of dogs based on the demographic risk factors of age, sex, and breed-type, statistically significant spatial and temporal clusters were present for both CaOx and struvite urolith types. A purely temporal struvite cluster occurred between February 10, 1998 and December 20, 2001, whereas, a purely temporal CaOx cluster occurred between September 2, 2005 and December 21, 2006. Hypotheses to explain the spatial clustering of uroliths include variation in the spatial distribution of water hardness, diet-type, access to veterinary care, and the use of surgical versus medical therapies to treat these uroliths. Based on the cluster locations, water hardness was unlikely to explain the spatial difference between the two cluster types whereas variables related to human population density were more consistent with our findings; the CaOx cluster occurred in the highest population density area of Ontario, and the struvite cluster occurred in the lowest population density area of southern Ontario. The temporal struvite cluster at the beginning and CaOx cluster at the end of the study period reflect a similar trend away from struvite towards CaOx urolithiasis among both canines and humans in the developed countries of North America and Europe.

Salmonella in broiler litter and properties of soil at farm location

Contamination of litter in a broiler grow-out house with Salmonella prior to placement of a new flock has been shown to be a precursor of the flock's Salmonella contamination further down the production continuum. In the southern USA, broiler grow-out houses are primarily built on dirt pad foundations that are placed directly on top of the native soil surface. Broiler litter is placed directly on the dirt pad. Multiple grow-out flocks are reared on a single litter batch, and the litter is kept in the houses during downtime between flocks. The effects of environmental determinants on conditions in broiler litter, hence Salmonella ecology within it, has received limited attention. In a field study that included broiler farms in the states of Alabama, Mississippi and Texas we assessed Salmonella in broiler litter at the end of downtime between flocks, i.e. at the time of placement of a new flock for rearing. Here we utilized these results and the U.S. General Soil Map (STATSGO) data to test if properties of soil at farm location impacted the probability of Salmonella detection in the litter. The significance of soil properties as risk factors was tested in multilevel regression models after accounting for possible confounding differences among the farms, the participating broiler complexes and companies, and the farms' geographical positioning. Significant associations were observed between infiltration and drainage capabilities of soil at farm location and probability of Salmonella detection in the litter.

Identification of risk factors for the prevalence and persistence of Salmonella in Belgian broiler chicken flocks

According to the European Food Safety Authority, salmonellosis is still one of the main causes of infectious foodborne gastroenteritis in humans. Broilers are an important source of salmonellosis after eggs and pork. Between 1987 and 1999 the trend of human salmonellosis incidence in Belgium increased constantly. However, from 2000 until 2005 a decrease in human cases was observed, probably following the sanitary measures implemented in the poultry breeder and laying sector. In order to decrease human infections it is essential to tackle the problem at the farm level to minimize cross-contamination from farm to fork. This paper seeks to answer two questions: (i) given the Salmonella status of the farm at a certain occasion (equal to the sampling time of the flock), what are the risk factors that the farm will be Salmonella positive at a following occasion? And (ii) what are the risk factors for a farm to be persistently positive for two consecutive flocks? We used surveillance data on 6824 broiler flocks studied for Salmonella infectivity from 2005 to 2006 in Belgium. The farms were tested regularly (3 weeks before slaughter of each broiler flock) for the presence of Salmonella based on multiple faecal samples per flock on a farm yielding clustered data. Generalized estimating equations, alternating logistic regression models, and random-intercept logistic regression models were employed to analyse these correlated binary data. Our results indicated that there are many factors that influence Salmonella risk in broiler flocks, and that they interact. Accounting for interactions between risk factors leads to an improved determination of those risk factors that increase infection with Salmonella. For the conditional analysis, the risk factors found to increase the risk of Salmonella infection on a farm at a current occasion given the previous Salmonella status included: Salmonella infection of day-old chicks (of the current flock); a previously infected flock even though the farm was equipped with a hygiene place to change clothes prior to entering the broiler house; having temporary workmen when there was a separation between birds of different species; and separating birds of different species in the Walloon region relative to the Flanders region. Sanitary measures such as a cleaning and disinfecting procedure conducted by an external cleaning firm, applying the all-in all-out procedure, and hand washing decreased the risk despite their interaction with other factors. From the joint analysis, the most important factors identified for increased risk for persistent Salmonella on a farm involved the interaction between having temporary workmen when there were poultry or farmers in contact with foreign poultry or persons, and the interaction between having temporary workmen when there were poultry or farmers in contact with external poultry or persons.

A four year longitudinal sero-epidemiological study of bovine herpesvirus type-1 (BHV-1) in adult cattle in 107 unvaccinated herds in south west England

Background

Bovine herpesvirus type-1 (BHV-1) is an important pathogen of cattle that presents with a variety of clinical signs, including the upper respiratory tract infection infectious bovine rhinotracheitis (IBR). A seroepidemiological study of BHV-1 antibodies was conducted in England from 2002 – 2004: 29,782 blood samples were taken from 15,736 cattle from 114 herds which were visited on up to three occasions. Antibody concentration was measured using a commercial ELISA. Farm management information was collected using an interview questionnaire, and herd size and cattle movements were obtained from the cattle tuberculosis testing database and the British Cattle Movement Service. Hierarchical statistical models were used to investigate associations between cattle and herd variables and the continuous outcome percentage positive (PP) values from the ELISA test in unvaccinated herds.

Results

There were 7 vaccinated herds, all with at least one seropositive bovine. In unvaccinated herds 83.2% had at least one BHV-1 seropositive bovine, and the mean cattle and herd BHV-1 seroprevalence were 42.5% and 43.1% respectively. There were positive associations between PP value, age, herd size, presence of dairy cattle. Adult cattle in herds with grower cattle had lower PP values than those in herds without grower cattle. Purchased cattle had significantly lower PP values than homebred cattle, whereas cattle in herds that were totally restocked after the foot-and-mouth epidemic in 2001 had significantly higher PP values than those in continuously stocked herds. Samples taken in spring and summer had significantly lower PP values than those taken in winter, whereas those taken in autumn had significantly higher PP values than those taken in winter. The risks estimated from a logistic regression model with a binary outcome (seropositive yes/no) were similar.

Conclusion

The prevalence of BHV-1 seropositivity in cattle and herds has increased since the 1970s. Although the study population prevalence of BHV-1 was temporally stable during study period, the associations between serological status and cattle age, herd size, herd type, presence of young stock and restocked versus continuously stocked herds indicate that there is heterogeneity between herds and so potential for further spread of BHV-1 within and between herds.

A multi-level approach for investigating socio-economic and agricultural risk factors associated with rates of reported cases of Escherichia coli O157 in humans in Alberta, Canada

Using negative binomial and multi-level Poisson models, the authors determined the statistical significance of agricultural and socio-economic risk factors for rates of reported disease associated with Escherichia coli O157 in census subdivisions (CSDs) in Alberta, Canada, 2000-2002. Variables relating to population stability, aboriginal composition of the CSDs, and the economic relationship between CSDs and urban centres were significant risk factors. The percentage of individuals living in low-income households was not a statistically significant risk factor for rates of disease. The statistical significance of cattle density, recorded at a higher geographical level, depended on the method used to correct for overdispersion, the number of levels included in the multi-level models, and the choice of using all reported cases or only sporadic cases. Our results highlight the importance of local socio-economic risk factors in determining rates of disease associated with E. coli O157, but their relationship with individual risk factors requires further evaluation.

The effect of season on somatic cell count and the incidence of clinical mastitis

Bulk milk somatic cell count (BMSCC), individual cow somatic cell count (ICSCC), and incidence rate of clinical mastitis (IRCM) are all udder health parameters. So far, no studies have been reported on the effect of season on BMSCC, IRCM, and ICSCC in the same herds and period over multiple years. The objectives of this study were to determine the seasonal pattern over a 4-yr period of 1) BMSCC, 2) elevated ICSCC, 3) IRCM, and 4) pathogen-specific IRCM. Bulk milk somatic cell count, ICSCC, and pathogen-specific clinical mastitis data were recorded in 300 Dutch dairy farms. For the analyses of BMSCC, ICSCC, and IRCM, a mixed, a transitional, and a discrete time survival analysis model were used, respectively. Sine and cosine were included in the models to investigate seasonal patterns in the data. For all parameters, a seasonal effect was present. Bulk milk somatic cell count peaked in August to September in all 4 years. The probability of cows getting or maintaining a high ICSCC was highest in August and May, respectively. Older and late-lactation cows were more likely to develop or maintain a high ICSCC. Incidence rate of clinical mastitis was highest in December to January, except for Streptococcus uberis IRCM, which was highest in August. Totally confined herds had a higher Escherichia coli IRCM in summer than in winter. Compared with the major mastitis pathogens, the seasonal differences in IRCM were smaller for the minor pathogens. Distinguishing between Strep. uberis, Streptococcus dysgalactiae, Streptococcus agalactiae, and other streptococci is essential when identifying Streptococcus spp. because each of them has a unique epidemiology. Streptococcus uberis IRCM seemed to be associated with being on pasture, whereas E. coli IRCM was more housing-related.

Epidemiological characteristics of reported sporadic and outbreak cases of E. coli O157 in people from Alberta, Canada (2000-2002): methodological challenges of comparing clustered to unclustered data

Using multivariable models, we compared whether there were significant differences between reported outbreak and sporadic cases in terms of their sex, age, and mode and site of disease transmission. We also determined the potential role of administrative, temporal, and spatial factors within these models. We compared a variety of approaches to account for clustering of cases in outbreaks including weighted logistic regression, random effects models, general estimating equations, robust variance estimates, and the random selection of one case from each outbreak. Age and mode of transmission were the only epidemiologically and statistically significant covariates in our final models using the above approaches. Weighing observations in a logistic regression model by the inverse of their outbreak size appeared to be a relatively robust and valid means for modelling these data. Some analytical techniques, designed to account for clustering, had difficulty converging or producing realistic measures of association.

Subclinical and Clinical Mastitis in Heifers Following the Use of a Teat Sealant Precalving

This study investigated the effect in heifers of infusion of a bismuth subnitrate teat-canal sealant and bacterial intramammary infection (IMI) precalving on prevalence of postcalving IMI and incidence of clinical mastitis in the first 2 wk postcalving. Glands (n = 1,020) from heifers (n = 255) in 5 seasonally calving, pasture-fed dairy herds were randomly assigned within heifer to 1 of 4 treatment groups (no treatment; mammary gland secretion collection; infusion of a teat sealant; or sample collection with infusion of teat sealant). Heifers within a herd were enrolled on one calendar day, 31 d on average before the planned start of the seasonal calving period. Duplicate milk samples were collected from each gland within 4 d after calving for bacterial culture. Herd owners collected duplicate milk samples, before treatment, for bacterial culture from glands they defined as having clinical mastitis. The gland prevalence of IMI precalving was 15.5% and did not differ between herds. Bacteria isolated precalving included coagulase-negative staphylococci (76.9% of all bacteriologically positive samples), Streptococcus uberis (14.1%), Staphylococcus aureus (5.1%), Corynebacterium spp. (3.8%), and others (0.1%). The presence of an IMI precalving increased the risk of an IMI postcalving 3.6-fold and the risk of clinical mastitis 4-fold, relative to no IMI precalving. Infusion of the teat sealant reduced the risk of postcalving IMI due to Strep. uberis by 84%, and of clinical mastitis by 68%. Sampling the glands precalving had no effect on postcalving IMI or on clinical mastitis incidence. Use of an internal teat canal sealant in heifers precalving may be a useful tool for reducing the risk of subclinical and clinical mastitis in heifers.

Invited Review: The Role of Cow, Pathogen, and Treatment Regimen in the Therapeutic Success of Bovine Staphylococcus aureus Mastitis

Staphylococcus aureus is an important cause of udder infections in dairy herds. Both lactational and dry cow therapy are part of Staph. aureus control programs. Reported cure rates for Staph. aureus mastitis vary considerably. The probability of cure depends on cow, pathogen, and treatment factors. Cure rates decrease with increasing age of the cow, increasing somatic cell count, increasing duration of infection, increasing bacterial colony counts in milk before treatment, and increasing number of quarters infected. Staphylococcus aureus mastitis in hind quarters has a low cure rate compared with front quarters. Antimicrobial treatment of intramammary infections with penicillin-resistant Staph. aureus strains results in a lower cure rate for treatment with either beta-lactam or non-beta-lactam antibiotics. Other strain-specific factors may affect the probability of cure but routine diagnostic methods for use in bacteriology laboratories or veterinary practices are not yet available. The most important treatment factor affecting cure is treatment duration. Increased duration of treatment is associated with increased chance of cure. Economically, extended treatment is not always justified, even when indirect effects of treatment such as prevention of contagious transmission are taken into consideration. Usefulness of treatment trials could be improved by standardization of case definitions, consideration of host and strain factors, and sufficient statistical power. Treatment of young animals with penicillin-sensitive Staph. aureus infections is often justified based on bacteriological cure and economic outcome, whereas treatment of older animals, chronic infections, or penicillin-resistant isolates should be discouraged.

Interdependence and distribution of subclinical mastitis and intramammary infection among udder quarters in dairy cattle

The objective of the present study was to determine the distribution of subclincial mastitis and intramammary infection (IMI) across udder quarters and to quantify, if any, the level of interdependence between quarters. Subclinical mastitis was defined as an udder quarter somatic cell count of > or =250,000. Intramammary infection was defined as the presence of pathogens in an udder quarter. The data used in the analyses consisted 6577 test-day records from 2034 lactations with observations on subclinical mastitis and 8428 test-day records from 2575 lactations with observations on IMI; each test-day record consisted data on all four udder quarters. Records were from animals on three research herds in southern Ireland. Observed distributions of IMI were compared to expected distributions, estimated using a binomial probability distribution, assuming independence among quarters. Generally, the observed frequencies deviated significantly from the expected frequencies, indicating some level of interdependence between quarters. Intraclass correlations were estimated from a mixed model including cow and test-day record nested within cow as random effects; fixed effects in the model included herd, year of calving, month of calving, parity, days in milk at sampling and interactions. Intraclass correlations within test-day record ranged from 0.14 to 0.50 for subclinical mastitis and from 0.02 to 0.40 for IMI. Thus, substantial interdependence between quarters exists which should be accounted for when analysing data from experiments across udder quarters. Failure to do so may result in incorrect inferences from statistical tests.

Intra-class correlation attributable to claw trimmers scoring common hind-claw disorders in Dutch dairy herds

Data of a cross-sectional study on prevalence of seven common bovine-claw disorders were used to calculate the intra-class correlation coefficients (ICCs) for claw trimmers. Fifteen professional claw trimmers were trained in diagnosing claw disorders uniformly. During regular trimming of all dairy cows in a herd, they recorded the presence (yes/no) of hind-claw disorders for digital dermatitis (DD), chronic laminitis (CL), sole ulcer (SU), white-line disease (WLD) and interdigital hyperplasia (HYP). For the claw disorders interdigital dermatitis/heel-horn erosion (IDHE) and sole haemorrhage (SH), claw trimmers also recorded (using graduated scores 1-3) the extent of the lesion. Complete information of 21,153 animals from 361 herds in The Netherlands was recorded between July 2002 and December 2003. To assure objective recording of claw disorders, ideally there would not be any correlation between observations. However, quantified ICCs from claw-trimmers observations ranged from 4.9% for DD to 38.0% for CL, while ICCs for herd ranged from 7.9% for SU to 26.7% for DD. Changes in the cut-off values for the diagnosis of IDHE and SH did not result in any relevant changes of the ICCs from claw-trimmers observations and these remained larger than the ICC explained by the herd. Based on these results, we conclude that for estimating measures of frequencies for different claw disorders, uniformity in recording certain diagnoses becomes an issue to consider when involving more than one observer.

Health and welfare of dairy cows in different husbandry systems in Switzerland

Our objective was to compare health and welfare of dairy cows kept in three types of husbandry systems: (1) tie stalls with regular exercise in summer but minimal outdoor access during winter (the reference level for analyses); (2) tie stalls with regular exercise in an exercise yard or pasture throughout the year; (3) loose-housing with regular access to an outdoor exercise yard or pasture. A total of 134 farms were visited (two to) three times in two years. Cows were examined for lameness, skin alterations at the hock joints, scars or injuries at the teats, and skin injuries at other locations. Lying and standing-up behavior were also evaluated. Farmers were requested to record the incidence of medical treatments for the whole observation period. A multivariable logistic-regression analysis was performed for each indicator of health and welfare with husbandry type, aspects of the housing system, farm characteristics, and management routines as the predictor variables. For welfare indicators recorded on individual animals, regression was performed correcting for clustering of observations within herds by Generalized Estimation Equation. Risk factors for the incidence of medical treatments were analyzed in a negative-binomial regression model. The odds for lameness were reduced for tie stalls with regular exercise throughout the year (OR=0.7). The prevalence of alterations at the hock joints was lower in loose-housing with regular outdoor exercise (OR=0.4). Teat injuries were less frequent in loose-housing with regular outdoor exercise (OR=0.1) and tie stalls with regular exercise (OR=0.4). Farms with loose-housing and regular outdoor exercise had a lower incidence of medical treatments (IR=0.6) than reference level farms.

The impact of badger removal on the control of tuberculosis in cattle herds in Ireland

In Ireland, the herd prevalence of bovine tuberculosis has remained stable for several decades, and in common with several other countries, progress towards eradication has stalled. There is evidence in support of the potential role of infected badgers (Meles meles, a protected species) in bovine tuberculosis in Ireland and Britain. However, this evidence on its own has not been sufficient to prove disease causation. Field trials are likely to offer the best opportunity to define this role. Building on the earlier East Offaly project, our objectives were to assess the impact of badger removal on the control of tuberculosis in cattle herds in Ireland. The study was conducted from September 1997 to August 2002 in matched removal and reference areas (average area of 245.1km(2)) in four counties: Cork, Donegal, Kilkenny and Monaghan. Badger removal was intensive and proactive throughout the study period in the removal areas, but reactive (in response to severe tuberculosis outbreaks in cattle) in the reference areas. Removal intensity in the removal and reference areas during the first 2 years of the study averaged 0.57 and 0.07 badgers/km(2)/year, respectively. The outcome of interest was restriction of cattle herds due to confirmed tuberculosis, where tuberculous lesions were detected in one or more animals. Data were analysed using logistic regression (modelling the probability of a confirmed herd restriction) and survival analysis (modelling time to a confirmed herd restriction). During the study period, there was a significant difference between the removal and reference areas in all four counties in both the probability of and the time to a confirmed herd restriction due to tuberculosis. In the final year of the study, the odds of a confirmed herd restriction in the removal (as compared to the reference areas) were 0.25 in Cork, 0.04 in Donegal, 0.26 in Kilkenny and 0.43 in Monaghan. Further, the hazard ratios (removal over reference) ranged from 0.4 to 0.04 (a 60-96% decrease in the rate at which herds were becoming the subject of a confirmed restriction).

Estimation of infection prevalence from correlated binomial samples

Infection prevalence in a population often is estimated from grouped binary data expressed as proportions. The groups can be families, herds, flocks, farms, etc. The observed number of cases generally is assumed to have a Binomial distribution and the estimate of prevalence is then the sample proportion of cases. However, the individual binary observations might not be independent--leading to overdispersion. The goal of this paper was to demonstrate random-effects models for the estimation of infection prevalence from data which are correlated and in particular, to illustrate a nonparametric random-effects model for this purpose. The nonparametric approach is a relatively recent addition to the random-effects class of models and does not appear to have been discussed previously in the veterinary epidemiology literature. The assumptions for a logistic-regression model with a nonparametric random effect were outlined. In a demonstration of the method on data relating to Salmonella infection in Irish pig herds, the nonparametric method resulted in the classification of herds into a small number of distinct prevalence groups (i.e. low, medium and high prevalence) and also estimated the relative frequency of each prevalence category in the population. We compared the estimates from a logistic model with a nonparametric distribution for the random effects with four alternative models: a logistic-regression model with no random effects, a marginal model using a generalised estimating equation (GEE) and two methods of fitting a Normally distributed random effect (the GLIMMIX macro and the NLMIXED procedure both in SAS). Parameter estimates from random-effects models are not readily interpretable in terms of prevalences. Therefore, we outlined two methods for calculating population-averaged estimates of prevalence from random-effects models: one using numerical integration and the other using Monte Carlo simulation.