ABSTRACT

The burden of foodborne illness linked to the consumption of contaminated broiler meat is high in the United States. With the increase in popularity of alternative poultry rearing and production systems, it is important to identify the differences in food safety risks presented by alternative systems compared with conventional methods. Although many studies have been conducted that surveyed foodborne pathogen prevalence along the broiler supply chain, a systematic overview of all of the results is lacking. In the current study, a systematic review and meta-analysis were conducted to quantify the differences in prevalence of Salmonella and Campylobacter spp. in farm environment, rehang, prechill, postchill, and retail samples between conventional and alternative production systems. A systematic search of Web of Science and PubMed databases was conducted to identify eligible studies. Studies were then evaluated by inclusion criteria, and the included studies were qualitatively and quantitatively analyzed. In total, 137 trials from 72 studies were used in the final meta-analysis. Meta-analysis models were individually constructed for subgroups that were determined by sample type, pathogen, and production type. All subgroups possessed high amounts of heterogeneity (I2 > 75%). For environmental sample subgroups, Campylobacter prevalence was estimated to be 15.8 and 52.8% for conventional and alternative samples, respectively. Similar prevalence estimates for both production types were observed for Salmonella environmental samples and all retail samples. For conventional samples, Campylobacter and Salmonella prevalence was highest in prechill samples followed by rehang and postchill samples, respectively. The results herein will be useful in future quantitative microbial risk assessments for characterizing the differences in foodborne illness risks presented by different broiler production systems.

HIGHLIGHTS
  • Meta-analysis models were constructed to estimate pathogen prevalence.

  • Prevalence was estimated for broiler farming, processing, and retail samples.

  • Between-study heterogeneity was described by various moderator variables.

  • Significantly different Campylobacter prevalence was found in environmental samples.

  • Minimal differences were observed for alternative and conventional retail samples.

Foodborne pathogens such as Campylobacter spp. and Salmonella have presented a major problem for the food safety of broiler chicken and chicken products in the U.S. supply chain. From 1998 to 2017, there were 298 chicken-related salmonellosis outbreaks in the United States, resulting in 7,881 illnesses, 905 hospitalizations, and 4 deaths (17). Campylobacter spp. are estimated to cause >800,000 domestically acquired foodborne illnesses annually in the United States (94). In addition, poultry products were implicated in 15 campylobacteriosis outbreaks in the United States from 2004 to 2012 (35). Although much of the efforts are put into controlling Salmonella and Campylobacter in fresh poultry, Listeria spp. have been identified as emerging poultry-related pathogens (88).

In recent years, the demand for organic and alternatively produced food products has increased: retail sales of organic food products in the United States increased from $3.6 billion to $21.1 billion from 1997 to 2008 (26). This trend has also impacted the poultry industry, with organic, free-range, antibiotic-free, and pasture poultry operations becoming more desired than conventional operations. As the popularity with these products increases, the need to understand the food safety hazards around these products becomes imperative. Many consumers believe these types of products to be safer than products from conventional methods due to the reduced use of pesticides, antibiotics, and added hormones, but scientific evidence to support this hypothesis is lacking (50, 102, 117).

Currently, most of the effort in quantifying the differences between conventional and alternative poultry production has gone into microbiological surveying at various points along the poultry supply chain. A useful tool in helping to quantify the efforts of similar studies is through systematic review and meta-analysis. Meta-analysis is used to aggregate the results of individual studies, quantify the estimated effect of each study, and provide an overall estimate on the effect and variability of an intervention or treatment or overall prevalence of an outcome (36). Although widely used in the medical literature, meta-analyses have just started gaining more popularity in the food safety literature, with recommendations to use this tool in a food safety context by Gonzales-Barron and Butler (38) and Sargeant et al. (92). Meta-analysis results are important tools in quantitative microbial risk assessments (QMRAs), as they provide an estimate of pathogen prevalence or reduction at certain stages of a product's supply chain, potentially providing a more accurate number than an estimate based on one study or on expert judgment (38).

Recent poultry-related systematic reviews and meta-analyses have been conducted to estimate the prevalence of foodborne pathogens in poultry samples and the effectiveness of various interventions in reducing foodborne pathogen load in poultry (15, 37, 55, 110, 120, 126). The meta-analysis conducted by Young et al. (126) worked to address differences in foodborne pathogen numbers between organic and conventional poultry samples but included studies from around the world. To our knowledge, there are no current meta-analysis studies used to estimate the prevalence of foodborne pathogens in U.S. broiler chicken samples in conventional and alternative poultry systems. In addition, there have been numerous recent surveys published on foodborne pathogen prevalence in conventional and alternative broiler chicken samples, showing the need for an updated systematic review and meta-analysis.

The purpose of the current study was to use a systematic review and meta-analysis approach to quantify the prevalence numbers of problematic foodborne pathogens in broiler chicken farming, processing, and retail samples in the United States. The results of this study will aid in the construction of QMRAs related to the differences between conventional and alternative broiler chicken production systems and their impact on domestic foodborne illness.

MATERIALS AND METHODS

Literature search strategy

A systematic review process was adapted from Sargeant et al. (93) to address a detailed research question: What are the differences in foodborne pathogen prevalence in farm environment, processing, and retail samples from commercial alternative and conventional broiler chicken production systems in the United States? Foodborne pathogens of interest included Campylobacter spp., Salmonella, and Listeria spp. To address this question, a detailed literature search was performed. On 26 September 2019, the Web of Science (www.webofknowledge.com) and PubMed online databases were searched with the following search terms aimed at addressing the aforementioned research question: (“Salmonella” OR “S. enterica” OR “Salmonella enterica” OR “Campylobacter” OR “C. jejuni” OR “Campylobacter jejuni” OR “Listeria” OR “L. monocytogenes” OR “Listeria monocytogenes”) AND (“poultry” OR “chicken”) AND (“incidence” OR “prevalence” OR “isolation” OR “survey” OR “detection” OR “occurrence”) AND (“United States” OR “name of any individual state”). For this step, there were no language or year limitations. With these criteria, the search totaled 2,356 studies. Additional studies (n = 7) were identified by searching review articles and article reference lists by hand. Studies included peer-reviewed journal articles and governmental agency reports. All references were managed with the EndNote citation manager (EndNote X8, Clarivate Analytics, Philadelphia, PA). After import to EndNote, duplicate studies were removed manually.

Inclusion criteria

Abstracts of articles were screened to determine whether they were appropriate to address the proposed research question. For articles to pass this stage, the articles needed to be U.S. prevalence studies of bacterial foodborne pathogens in commercial broiler farm environments, commercial broiler processing samples, or broiler chicken retail samples. Challenge studies or studies in which flocks of broilers were inoculated with pathogens were excluded.

Following abstract screening, full-text articles were obtained for all remaining studies (n = 262) and analyzed for potential inclusion in the final study. Each article was further assessed with the aforementioned screening criteria as well as additional criteria. If it was not reported that the study was conducted in the United States, inference was made based off of the authors' locations and language throughout the text. This meta-analysis is intended to get current prevalence data to best estimate the risk of foodborne illness from consumption of broiler chickens. Thus, the first additional criterion was that articles needed to be published after 1 January 2000. For the second criterion, the studies needed to report the sample size and the prevalence and/or number of positive samples. The third criterion was that the studies needed to involve the foodborne pathogens of interest: Salmonella, Campylobacter spp., and Listeria spp. Studies that involved targeted sampling for pathogen-positive broiler flocks were excluded because these studies could potentially overestimate the prevalence of pathogens. In addition, studies that did not provide species-level results were excluded. Studies that were questionable for inclusion were discussed by both of us until a consensus was reached.

Data extraction

Articles that were deemed eligible through screening were then included in quantitative and qualitative analysis (n = 80). At this step, quantitative and qualitative data were extracted and analyzed. Because of the problematic nature of incorporating study quality scores as factors in meta-analyses, scores were not assigned to individual studies (46, 53). Study quality was determined by the presence of reputable, replicable microbiological methods.

Collected quantitative variables included number of positive samples and sample size. If the number of positive samples was not reported, but the prevalence was reported, the number of positive samples was estimated by multiplying the prevalence by the sample size and rounding accordingly. Studies that contained inconsistent results throughout the text or figures of the study were excluded.

Qualitative data included the pathogen studied, state performed (if presented), type of poultry production (if presented), sample type, and detection method. Salmonella serotype prevalence for included samples was also extracted if presented in the study. Only one Listeria study was found, so it was not included in the final meta-analysis. For type of poultry production, if no production type was specified in the study or if the study stated that conventional systems were sampled, the production or retail system was inferred to be conventional. Otherwise, the system was marked as alternative. Alternative production and retail systems included organic, pasture-raised, antibiotic-free, farmer's market, or free-range systems. For one included study (6), a conventional production system using antibiotic-free birds was surveyed. We both agreed to include this study in the alternative production category. Sample type was categorized as environmental, processing, or retail. Environmental samples included any samples collected from the environment of broiler farms. Sample types included boot sock, air, feces, soil, litter, water, feed, grass, insect traps, and wild bird droppings. Processing samples included carcass samples collected throughout the processing supply chain. Samples were classified in the manner in which they were referred to in the study (e.g., rehang, prechill, postchill). For the majority of processing samples, whole carcass rinses (WCRs) were collected. A small number of studies also included results for whole carcass enrichment (WCE) and neck skin maceration (NSM) sampling methods. Retail samples included ground chicken, WCRs, and rinses of various chicken parts (e.g., breasts, thighs) that were purchased from a retail establishment or obtained at the end of the production line from a processing establishment. If different types of samples in the same category from a study were collected, those samples were combined for prevalence calculations if the same detection method was used on them. For consistency, prevalence numbers needed to be stated for the sample in the way it was purchased. For example, if a broiler chicken carcass was purchased from a retail store and cut into parts, positive numbers of carcasses needed to be reported. Detection methods were also noted for each study. If a study used multiple detection methods on the same samples, the highest prevalence or the number of true positives among the different methods was used as a fail-safe measure. If different methods were used on different samples, the study was considered as separate trials and data points.

After all data were extracted, data were grouped by sample type, pathogen, and production type. To be considered for meta-analysis, each subgroup needed to have at least two independent studies. Data from environmental, rehang, prechill, postchill, and retail samples were used. All data were stored in Excel version 16.28 (Microsoft, Redmond, WA).

Data analysis

All data analysis was performed using R version 3.6.1 (84). Meta-analyses and forest plot generation were conducted using the meta and metafor packages (97, 118).

A generalized linear mixed model approach (27, 107) combined with the logit transformation has been recommended by various studies (98, 122) and was used in the current study to help stabilize variance (33). For each included study, prevalence values were calculated by dividing the sample size by the number of positive samples. Because of the presence of proportions equal to 0 or 1, values were first transformed using the logit transformation (7, 58):  
\(\def\upalpha{\unicode[Times]{x3B1}}\)\(\def\upbeta{\unicode[Times]{x3B2}}\)\(\def\upgamma{\unicode[Times]{x3B3}}\)\(\def\updelta{\unicode[Times]{x3B4}}\)\(\def\upvarepsilon{\unicode[Times]{x3B5}}\)\(\def\upzeta{\unicode[Times]{x3B6}}\)\(\def\upeta{\unicode[Times]{x3B7}}\)\(\def\uptheta{\unicode[Times]{x3B8}}\)\(\def\upiota{\unicode[Times]{x3B9}}\)\(\def\upkappa{\unicode[Times]{x3BA}}\)\(\def\uplambda{\unicode[Times]{x3BB}}\)\(\def\upmu{\unicode[Times]{x3BC}}\)\(\def\upnu{\unicode[Times]{x3BD}}\)\(\def\upxi{\unicode[Times]{x3BE}}\)\(\def\upomicron{\unicode[Times]{x3BF}}\)\(\def\uppi{\unicode[Times]{x3C0}}\)\(\def\uprho{\unicode[Times]{x3C1}}\)\(\def\upsigma{\unicode[Times]{x3C3}}\)\(\def\uptau{\unicode[Times]{x3C4}}\)\(\def\upupsilon{\unicode[Times]{x3C5}}\)\(\def\upphi{\unicode[Times]{x3C6}}\)\(\def\upchi{\unicode[Times]{x3C7}}\)\(\def\uppsy{\unicode[Times]{x3C8}}\)\(\def\upomega{\unicode[Times]{x3C9}}\)\(\def\bialpha{\boldsymbol{\alpha}}\)\(\def\bibeta{\boldsymbol{\beta}}\)\(\def\bigamma{\boldsymbol{\gamma}}\)\(\def\bidelta{\boldsymbol{\delta}}\)\(\def\bivarepsilon{\boldsymbol{\varepsilon}}\)\(\def\bizeta{\boldsymbol{\zeta}}\)\(\def\bieta{\boldsymbol{\eta}}\)\(\def\bitheta{\boldsymbol{\theta}}\)\(\def\biiota{\\boldsymbol{\iota}}\)\(\def\bikappa{\boldsymbol{\kappa}}\)\(\def\bilambda{\boldsymbol{\lambda}}\)\(\def\\bimu{\boldsymbol{\mu}}\)\(\def\binu{\boldsymbol{\nu}}\)\(\def\bixi{\boldsymbol{\xi}}\)\(\def\biomicron{\boldsymbol{\micron}}\)\(\def\bipi{\boldsymbol{\pi}}\)\(\def\birho{\boldsymbol{\rho}}\)\(\def\bisigma{\boldsymbol{\sigma}}\)\(\def\bitau{\boldsymbol{\\tau}}\)\(\def\biupsilon{\boldsymbol{\upsilon}}\)\(\def\biphi{\boldsymbol{\phi}}\)\(\def\bichi{\boldsymbol{\chi}}\)\(\def\bipsy{\boldsymbol{\psy}}\)\(\def\biomega{\boldsymbol{\omega}}\)\(\def\bupalpha{\bf{\alpha}}\)\(\def\bupbeta{\bf{\beta}}\)\(\def\bupgamma{\bf{\gamma}}\)\(\def\bupdelta{\bf{\delta}}\)\(\def\bupvarepsilon{\bf{\varepsilon}}\)\(\def\bupzeta{\bf{\zeta}}\)\(\def\bupeta{\bf{\eta}}\)\(\def\buptheta{\bf{\theta}}\)\(\def\bupiota{\bf{\iota}}\)\(\def\bupkappa{\bf{\kappa}}\)\(\def\\buplambda{\bf{\lambda}}\)\(\def\bupmu{\bf{\mu}}\)\(\def\bupnu{\bf{\nu}}\)\(\def\bupxi{\bf{\xi}}\)\(\def\bupomicron{\bf{\micron}}\)\(\def\buppi{\bf{\pi}}\)\(\def\buprho{\bf{\rho}}\)\(\def\bupsigma{\bf{\sigma}}\)\(\def\buptau{\bf{\tau}}\)\(\def\bupupsilon{\bf{\upsilon}}\)\(\def\bupphi{\bf{\phi}}\)\(\def\bupchi{\bf{\chi}}\)\(\def\buppsy{\bf{\psy}}\)\(\def\bupomega{\bf{\omega}}\)\(\def\bGamma{\bf{\Gamma}}\)\(\def\bDelta{\bf{\Delta}}\)\(\def\bTheta{\bf{\Theta}}\)\(\def\bLambda{\bf{\Lambda}}\)\(\def\bXi{\bf{\Xi}}\)\(\def\bPi{\bf{\Pi}}\)\(\def\bSigma{\bf{\Sigma}}\)\(\def\bPhi{\bf{\Phi}}\)\(\def\bPsi{\bf{\Psi}}\)\(\def\bOmega{\bf{\Omega}}\)\begin{equation}\tag{1}{\rm{logit}}\ p = {\rm{ln}}\left( {{p \over {1 - p}}} \right){\rm }\end{equation}
with variance  
\begin{equation}\tag{2}{\rm{var}}\left( {{\rm{logit}}\ p} \right) = {1 \over {Np}} + {1 \over {\left( {1 - Np} \right)}}{\rm }\end{equation}
where p is the prevalence of pathogen reported in a study and N is the sample size of that study.

Following transformation, data for each pathogen and sample type combination were partitioned based on production type (i.e., conventional or alternative) to allow for subgroup analysis (47). If data were only available for conventional productions systems for a given pathogen–sample type combination, these data were modeled alone. A random intercept logistic regression model was then fitted to each subgroup to estimate the population prevalence and its 95% confidence interval (CI) as well as values to describe the between-study variance (τ2) and heterogeneity (I2) present (13, 48). As suggested by Higgins et al. (48),I2 values of 25, 50, and 75% were considered as low, medium, and high measures of heterogeneity, respectively. In addition, Cochran's Q-test with α = 0.10 was performed to compare the effect sizes yielded by each subgroup meta-analysis model (39, 40). If high amounts of heterogeneity were observed during subgroup analysis, a meta-analysis model with various moderating variables was fitted to the entire population of each pathogen–sample type combination to attempt to describe the between-study heterogeneity (37). The moderating variables included production system, detection method, and year for all sample types and type of chicken sample for retail samples. Within each population, each moderating variable had to have at least two levels to be included. If multiple detection methods or sample types were used in a study, a value of “multiple” was applied to variable. The P values were obtained for all moderating variables, and the amount of between-study variability explained (R2) by the moderating variables was calculated.

RESULTS

Study results

The complete systematic review process is outlined in Figure 1. In total, 137 trials from 72 studies were included in the final meta-analysis models, providing 14,735 environmental samples, 9,200 rehang samples, 1,270 prechill samples, 63,306 postchill samples, and 24,355 retail samples, for a total of 112,866 samples (Tables 1 through 3). Between all sample types, there were in total 82,671 Salmonella samples and 30,195 Campylobacter samples. There was a significantly greater number of conventional samples than alternative samples, with 108,873 and 3,993 total samples, respectively.

FIGURE 1

Flow diagram of the systematic review process.

FIGURE 1

Flow diagram of the systematic review process.

TABLE 1

Summary of broiler farm environmental sampling studies used in meta-analysisa

Summary of broiler farm environmental sampling studies used in meta-analysisa
Summary of broiler farm environmental sampling studies used in meta-analysisa

Table 4 presents the prevalence of various Salmonella serotypes in the different types of samples studied. It has been identified that some serotypes of Salmonella have greater public health risk than others, with Salmonella Typhimurium, Enteritidis, Newport, and Heidelberg being identified as a few of particular concern in the United States (31). In the present study, Salmonella Kentucky was identified as the most common serotype collected from various sample types along the chicken supply chain.

TABLE 2

Summary of broiler processing sampling studies used in meta-analysisa

Summary of broiler processing sampling studies used in meta-analysisa
Summary of broiler processing sampling studies used in meta-analysisa
TABLE 3

Summary of retail broiler sampling studies used in meta-analysisa

Summary of retail broiler sampling studies used in meta-analysisa
Summary of retail broiler sampling studies used in meta-analysisa
TABLE 4

Salmonella serotype prevalence in various sample types collected along the chicken supply chain

Salmonella serotype prevalence in various sample types collected along the chicken supply chain
Salmonella serotype prevalence in various sample types collected along the chicken supply chain

Meta-analysis results of data from environmental sampling studies

Meta-analysis models were constructed for each combination of pathogen and production type from the environmental sampling data (Table 5). For Campylobacter, the predicted population prevalence was significantly different (P < 0.10) for conventional (15.8%) and alternative (52.8%) environmental samples (Fig. 2A and Table 5). The predicted population prevalence for Salmonella samples was 22.9% (95% CI: 14.5 to 34.2%) and 19.9% (95% CI: 7.1 to 44.8%) for conventional and alternative samples, respectively (Fig. 2B). These values were not significantly different. Heterogeneity was very high for all four populations, with I2 ≥ 94.9% for all sample sets. For the combined Campylobacter model, 32.8% of the between-study variability was explained by the type of production system, detection method, and year (Table 6). Production system and year accounted for 15.31% of the between-study variability for the Salmonella population.

TABLE 5

Meta-analysis results for pathogen prevalence at different stages throughout the chicken supply chain subgrouped by type of production system

Meta-analysis results for pathogen prevalence at different stages throughout the chicken supply chain subgrouped by type of production system
Meta-analysis results for pathogen prevalence at different stages throughout the chicken supply chain subgrouped by type of production system
FIGURE 2

Forest plots for the Campylobacter spp. (A) and Salmonella (B) environmental sample subgroups.

FIGURE 2

Forest plots for the Campylobacter spp. (A) and Salmonella (B) environmental sample subgroups.

TABLE 6

Heterogeneity explained by various moderator variables in the combined meta-analysis models for each sample type and pathogen combination

Heterogeneity explained by various moderator variables in the combined meta-analysis models for each sample type and pathogen combination
Heterogeneity explained by various moderator variables in the combined meta-analysis models for each sample type and pathogen combination

Meta-analysis results of data from processing sampling studies

Meta-analysis models were separately created for rehang, prechill, and postchill processing samples to represent three prevalence benchmarks in the broiler chicken production chain (Figs. 3 through 5 and Table 5). If possible, each sample type was evaluated for each combination of pathogen and production type. Among conventional samples, Campylobacter prevalence was highest in prechill samples (97.9%; 95% CI: 6.4 to 100%) followed by rehang (84.9%; 95% CI: 52.4 to 96.6%) and postchill (60.9%; 95% CI: 41.3 to 77.6) samples. The same trend was observed for Salmonella among conventional processing samples, with prevalence values of 68.6 (95% CI: 20.1 to 95.0%), 42.9 (95% CI: 24.3 to 63.8%), and 14.3% (95% CI: 6.3 to 29.2%) for prechill, rehang, and postchill samples, respectively. The only alternative subgroup to have a sufficient number of independent studies to be considered for meta-analysis was the postchill Campylobacter group. An estimated population prevalence of 34.3% (95% CI: 8.4 to 74.8%) was calculated for this group, and this value was not significantly different from the conventional estimated population prevalence. Heterogeneity was high for all study groups, and between-study variance (τ2) was largest for prechill samples. Year was a significant moderator variable for Campylobacter rehang and postchill models and Salmonella rehang and prechill samples. For all but the Salmonella rehang model, it was estimated that prevalence has decreased over time (Table 6).

FIGURE 3

Forest plots for the Campylobacter spp. (A) and Salmonella (B) rehang sample subgroups.

FIGURE 3

Forest plots for the Campylobacter spp. (A) and Salmonella (B) rehang sample subgroups.

FIGURE 4

Forest plots for the Campylobacter spp. (A) and Salmonella (B) prechill sample subgroups.

FIGURE 4

Forest plots for the Campylobacter spp. (A) and Salmonella (B) prechill sample subgroups.

FIGURE 5

Forest plots for the Campylobacter spp. (A) and Salmonella (B) postchill sample subgroups.

FIGURE 5

Forest plots for the Campylobacter spp. (A) and Salmonella (B) postchill sample subgroups.

Meta-analysis results of data from retail sampling studies

Retail study groups contained a greater number of studies than other sample types. Meta-analysis models were constructed for each combination of pathogen and production type (Fig. 6 and Table 5). Predicted Campylobacter prevalence in retail broiler chicken for both production types was similar, with 59.2% (95% CI: 47.6 to 69.8%) and 55.4% (95% CI: 34.5 to 74.6%) for conventional and alternative samples, respectively. A similar trend was observed with Salmonella prevalence. The random-effects model predicted a population prevalence of 19.0% (95% CI: 12.2 to 28.3%) for conventional samples and 23.0% (95% CI: 14.8 to 34.0%) for alternative samples. For both sets of subgroups, there was no significant difference among estimated effect size between alternative and conventional production systems. Again, high heterogeneity (I2 ≥ 91.8%) was observed among each study group. In the combined Campylobacter model, type of production system, detection method, year, and type of chicken sample accounted for 50.14% of the between-study variability. Production system, detection method, year, and type of chicken sample accounted for 45.81% of the between-study variability in the Salmonella combined model, with production system, detection method, and type of chicken all being significant (P < 0.10) moderators.

FIGURE 6

Forest plots for the Campylobacter spp. (A) and Salmonella (B) retail sample subgroups.

FIGURE 6

Forest plots for the Campylobacter spp. (A) and Salmonella (B) retail sample subgroups.

DISCUSSION

The annual burden of foodborne illness caused by the consumption of contaminated poultry in the United States is high. From 1998 to 2012, poultry caused the greatest number of foodborne outbreaks (279), illnesses (9,760), and hospitalizations (565) in the United States when compared with other food products (18). As such, it is vitally important to understand the transmission and prevalence of foodborne pathogens throughout the poultry supply chain and the risks of various interventions and production methods. It is still unclear whether alternative poultry production practices contain more risk than conventional practices. The current meta-analysis attempted to help quantify the differences in Salmonella and Campylobacter prevalence throughout conventional and alternative broiler chicken supply chains, by using data from the food safety literature. Although Listeria has been identified as an emerging foodborne pathogen of concern to the poultry industry, more studies need to be conducted on the presence of this bacteria throughout the poultry supply chain (88).

Pathogen prevalence in the environment of broiler farms

In the current study, random-effect models were generated for each study group of interest, due to the existence of high heterogeneity in each group (24). The high amounts of heterogeneity present in the groups could be due to the varying types of environmental samples that were collected. Contaminated poultry litter, feed, and drinking water have all been identified as potential contamination risk factors for broilers and can result from contaminated feces and soil; small animals, such as rodents and insects; and poor worker hygiene (62, 116, 119). This highlights the importance of trying to characterize the prevalence of foodborne pathogens in a wide variety of environmental samples. To our knowledge, this is the first meta-analysis to estimate pathogen prevalence in the preharvest environment of poultry farms.

The estimated environmental prevalence of Campylobacter was higher for alternative poultry farms than conventional poultry farms (Fig. 2A and Table 5). Further research needs to be conducted to determine the cause of the increased prevalence of Campylobacter in alternative poultry farms. One possible reason could be due to the effect of climate and seasonality on Campylobacter prevalence (99). Alternative poultry production methods often provide more outside access to broilers than conventional methods, where broilers are more likely to be introduced to environments and can become contaminated with environmental pathogens.

Estimated Salmonella prevalence numbers were similar for both conventional and alternative production types (Fig. 2B and Table 5). Two included studies compared the prevalence of Salmonella among environmental samples from conventional and alternative farms. Alali et al. (1) isolated Salmonella from 28.8 and 4.33% of feces, feed, and water samples collected from conventional and organic broiler farms, respectively. Siemon et al. (100) found a similar trend with Salmonella isolated from 29.8 and 16.2% of feces samples collected from conventional and pasture poultry farms, respectively.

Pathogen prevalence in broiler processing samples

During broiler processing, interventions are put into place to attempt to control for bacterial pathogen contamination; thus, it is important to understand the difference in pathogen prevalence at various steps in the processing chain (10). As part of a systematic review, Guerin et al. (41) found that Campylobacter prevalence on broiler carcasses was high throughout the entire processing chain but generally decreased after chilling. These results are similar to those of the current study. Campylobacter prevalence was high in conventional rehang and prechill samples, with 84.9 and 97.9% positive samples, respectively, but prevalence lowered after chilling to 60.9% (Table 5). It is important to note that the prechill model only contained two studies, and one of these studies found Campylobacter in all collected samples (103). Similar trends have been noted for Salmonella prevalence in the processing supply chain. Rivera-Perez et al. (86) found that the chilling of carcasses effectively reduced Salmonella numbers in broiler samples. In the presented models, estimated Salmonella prevalence reduced from 68.6 to 14.3% (Table 5).

Sufficient processing carcass data for alternative production systems were only available for the postchill Campylobacter study group. The group consisted of three trials from two studies (Table 2). Campylobacter prevalence was lower for alternative samples than conventional samples, but more studies need to be conducted to allow for direct comparison.

Pathogen prevalence in retail broiler meat

Retail samples of broiler meat are important because they are the types of samples that are available for consumers for direct use. Although raw chicken is cooked before consumption, it is important for pathogen prevalence to be controlled for to prevent cross-contamination of surfaces and other food items in the kitchen, as well as potential risks due to undercooking of the meat (61).

No discernible difference was observed between pathogen prevalence of conventional and alternative retail meat samples. This refutes consumers' belief that alternatively produced chicken is safer than conventionally produced chicken (117). Various studies directly compared products from the different production groups. For example, Lestari et al. (56) did not identify significant differences in Salmonella prevalence between conventional and organic retail broiler carcasses. Cui et al. (23) found that Campylobacter prevalence on organic and conventional retail carcasses was similar, but Salmonella prevalence was slightly higher in organic samples. Mollenkopf et al. (64) found that retail chicken breast samples were routinely contaminated with bacterial pathogens but that production status did not seem to play a role, which agrees with the results in the current study. Although more studies were included in the conventional random-effects models for retail samples, a significant amount of research has been conducted on pathogen prevalence in alternative retail broiler meat allowing for better comparison of the two methods (Tables 3 and 5).

Use of meta-analysis in future QMRAs

One useful method for risk estimation is through the use of QMRA (121). QMRAs use qualitative and quantitative data and use probability distributions to estimate the foodborne illness risk of the pathogen of interest. Depending on the question at hand, QMRAs can include details on the entire farm-to-fork continuum or focus on a certain area of interest within the continuum (115). QMRAs depend on the quality of the data at hand and sometimes have to rely on expert elicitation to fill in gaps of knowledge (38). As such, systematic methods, such as meta-analysis, are important tools in QMRAs. Meta-analysis provides data-driven estimates that can be used to estimate the prevalence of foodborne pathogens at stages along the continuum and the effects that interventions have on pathogen prevalence.

In recent years, QMRAs have been conducted to estimate the annual burden of salmonellosis and campylobacteriosis due to consumption of poultry (19, 70, 82), but comprehensive QMRAs analyzing the differences in risk between conventional and alternative poultry production practices are still lacking. Rosenquist et al. (87) found that the risk of Campylobacter infection was 1.7 times higher in organically produced broiler meat than conventionally produced broiler meat, but this study was limited to Danish-produced meat. Similar studies analyzing the effects of conventional and alternative poultry production methods on foodborne pathogen risk in the United States need to be conducted. The meta-analysis provided in the current study should aid in the production of future QMRAs addressing these needs.

Normal distributions can be used to fit the results generated by meta-analysis when there are low amounts of heterogeneity present (28). As discussed above, all study groups contained high levels of between-study heterogeneity (Table 5). As such, it is suggested that a beta-PERT distribution is used to describe the information found in the meta-analysis, by using both ends of the 95% CI as the minimum and maximum parameters and the observed mean as the most probable value (16, 28).

Potential limitations and advantages

While considering the results of meta-analysis, it is important to take note of the inherent limitations associated with the method. One of these limitations is the potential existence of publication bias, or the tendency to publish studies that contain “positive” results (104). Many methods exist for evaluating the potential for publication bias in a meta-analysis, including funnel plot evaluation and statistical tests such as Egger's regression and Begg's adjusted rank correlation tests (8, 29). Although these evaluation tools are recommended in many cases, it has been shown that the results can be misleading in meta-analyses with fewer than 10 studies or high amounts of between-study heterogeneity (106). It is suggested that it is very difficult to evaluate the true results of statistically significant publication bias tests, especially in the presence of high heterogeneity. In the current meta-analysis, only 6 of 15 evaluated subgroups contained >10 studies, but high levels of between-study heterogeneity existed (Table 5). As such, funnel plots were not constructed for any of the study groups. To address this concern, as high levels of heterogeneity were expected, the systematic review portion of the analysis did not exclude unpublished or nonpeer-reviewed studies. For example, two governmental reports (113, 114) were included in the final meta-analysis for various study groups. In addition, because included studies were meant to survey foodborne pathogen prevalence throughout the broiler chicken supply chain, it was not anticipated that the lack of pathogen-positive samples would cause reports not to be published. In fact, prevalence numbers ranged from 0 to 100% in included studies (Tables 1 through 3).

Some potential limitations also exist for the current meta-analysis. The first is that many of the included studies did not include information on whether collected samples came from conventional or alternative production chains. We believed that the best way to handle these instances was to include these studies in the conventional study category, because this was the most likely event. Conversely, studies included in the alternative production category clearly stated the production type of the sample population. This should give the most accurate estimation of each group's pathogen prevalence without having to discard any studies with valuable data. In addition, random-effects models were generated for both conventional and alternative studies combined for each pathogen–sample type combination, when available. Combined random-effects models were provided for Salmonella and Campylobacter environmental and retail study groups as well as for the Campylobacter postchill study group. It is also important to take into consideration that some study groups contained as few as two studies. These results should be taken with caution but should provide a better estimate of population behavior than a single study would provide, which is especially important when developing a risk assessment model (38, 108).

In conclusion, this study provided random-effects meta-analysis models as a means to estimate the prevalence of Campylobacter spp. and Salmonella at various points throughout the broiler chicken supply chain. The results will be of great use in the construction of future QMRAs and as a means for characterizing the differences in risk between conventional and alternative broiler chicken production methods.

REFERENCES

REFERENCES
1.
Alali,
W. Q.,
Thakur
S.,
Berghaus
R. D.,
Martin
M. P.,
and
Gebreyes
W. A.
2010
.
Prevalence and distribution of Salmonella in organic and conventional broiler poultry farms
.
Foodborne Pathog. Dis
.
7
:
1363
1371
.
2.
Bailey,
J. S.,
and
Cosby
D. E.
2003
.
Detection of Salmonella from chicken rinses and chicken hot dogs with the automated BAX PCR system
.
J. Food Prot
.
66
:
2138
2140
.
3.
Bailey,
J. S.,
and
Cosby
D. E.
2005
.
Salmonella prevalence in free-range and certified organic chickens
.
J. Food Prot
.
68
:
2451
2453
.
4.
Bailey,
J. S.,
Cox
N. A.,
Craven
S. E.,
and
Cosby
D. E.
2002
.
Serotype tracking of Salmonella through integrated broiler chicken operations
.
J. Food Prot
.
65
:
742
745
.
5.
Bailey,
J. S.,
Stern
N. J.,
Fedorka-Cray
P.,
Craven
S. E.,
Cox
N. A.,
Cosby
D. E.,
Ladely
S.,
and
Musgrove
M. T.
2001
.
Sources and movement of Salmonella through integrated poultry operations: a multistate epidemiological investigation
.
J. Food Prot
.
64
:
1690
1697
.
6.
Bailey,
M. A.,
Taylor
R. M.,
Brar
J. S.,
Corkran
S. C.,
Velasquez
C.,
Novoa Rama
E.,
Oliver
H. F.,
and
Singh
M.
2019
.
Prevalence and antimicrobial resistance of Campylobacter from antibiotic-free broilers during organic and conventional processing
.
Poult. Sci
.
98
:
1447
1454
.
7.
Barendregt,
J. J.,
Doi
S. A.,
Lee
Y. Y.,
Norman
R. E.,
and
Vos
T.
2013
.
Meta-analysis of prevalence
.
J. Epidemiol. Community Health
67
:
974
978
.
8.
Begg,
C. B.,
and
Berlin
J. A.
1988
.
Publication bias: a problem in interpreting medical data
.
J. R. Stat. Soc. A Stat
.
151
:
419
445
.
9.
Berghaus,
R. D.,
Thayer
S. G.,
Law
B. F.,
Mild
R. M.,
Hofacre
C. L.,
and
Singer
R. S.
2013
.
Enumeration of Salmonella and Campylobacter spp. in environmental farm samples and processing plant carcass rinses from commercial broiler chicken flocks
.
Appl. Environ. Microbiol
.
79
:
4106
4114
.
10.
Berrang,
M. E.,
Bailey
J. S.,
Altekruse
S. F.,
Patel
B.,
Shaw
W. K.,
Meinersmann
R. J.,
and
Fedorka-Cray
P. J.
2007
.
Prevalence and numbers of Campylobacter on broiler carcasses collected at rehang and postchill in 20 U.S. processing plants
.
J. Food Prot
.
70
:
1556
1560
.
11.
Berrang,
M. E.,
Bailey
J. S.,
Altekruse
S. F.,
and
Shaw
W. K.
2008
.
Presence and numbers of Campylobacter, Escherichia coli, and Salmonella determined in broiler carcass rinses from United States processing plants in the hazard analysis and critical control point-based inspection models project
.
J. Appl. Poult. Res
.
17
:
354
360
.
12.
Berrang,
M. E.,
Meinersmann
R. J.,
Cox
N. A.,
and
Thompson
T. M.
2018
.
Multilocus sequence subtypes of Campylobacter detected on the surface and from internal tissues of retail chicken livers
.
J. Food Prot
.
81
:
1535
1539
.
13.
Borenstein,
M.,
Hedges
L.,
and
Rothstein
H.
2007
.
Meta-analysis: fixed effect vs. random effects
.
Available at: https://www.meta-analysis.com/downloads/M-a_f_e_v_r_e_sv.pdf. Accessed 6 November 2019.
14.
Brichta-Harhay,
D. M.,
Arthur
T. M.,
and
Koohmaraie
M.
2008
.
Enumeration of Salmonella from poultry carcass rinses via direct plating methods
.
Lett. Appl. Microbiol
.
46
:
186
191
.
15.
Bucher,
O.,
Farrar
A. M.,
Totton
S. C.,
Wilkins
W.,
Waddell
L. A.,
Wilhelm
B. J.,
McEwen
S. A.,
Fazil
A.,
and
Rajić
A.
2012
.
A systematic review-meta-analysis of chilling interventions and a meta-regression of various processing interventions for Salmonella contamination of chicken
.
Prev. Vet. Med
.
103
:
1
15
.
16.
Buczinski,
S.,
and
Vandeweerd
J.
2016
.
Diagnostic accuracy of refractometry for assessing bovine colostrum quality: a systematic review and meta-analysis
.
J. Dairy Sci
.
99
:
7381
7394
.
17.
Centers for Disease Control and Prevention
.
2020
.
National Outbreak Reporting System (NORS), outbreaks per state, Salmonella, chicken
.
Available at: https://wwwn.cdc.gov/norsdashboard/. Accessed 17 January 2020.
18.
Chai,
S.,
Cole
D.,
Nisler
A.,
and
Mahon
B. E.
2017
.
Poultry: the most common food in outbreaks with known pathogens, United States, 1998–2012
.
Epidemiol. Infect
.
145
:
316
325
.
19.
Chapman,
B.,
Otten
A.,
Fazil
A.,
Ernst
N.,
and
Smith
B. A.
2016
.
A review of quantitative microbial risk assessment and consumer process models for Campylobacter in broiler chickens
.
Microb. Risk Anal.
2
:
3
15
.
20.
Cox,
N. A.,
Buhr
R. J.,
Smith
D. P.,
Cason
J. A.,
Rigsby
L. L.,
Bourassa
D. V.,
Fedorka-Cray
P. J.,
and
Cosby
D. E.
2014
.
Sampling naturally contaminated broiler carcasses for Salmonella by three different methods
.
J. Food Prot
.
77
:
493
495
.
21.
Cox,
N. A.,
Richardson
L. J.,
Cason
J. A.,
Buhr
R. J.,
Vizzier-Thaxton
Y.,
Smith
D. P.,
Fedorka-Cray
P. J.,
Romanenghi
C. P.,
Pereira
L. V. B.,
and
Doyle
M. P.
2010
.
Comparison of neck skin excision and whole carcass rinse sampling methods for microbiological evaluation of broiler carcasses before and after immersion chilling
.
J. Food Prot
.
73
:
976
980
.
22.
Craven,
S. E.,
Stern
N. J.,
Line
E.,
Bailey
J. S.,
Cox
N. A.,
and
Fedorka-Cray
P.
2000
.
Determination of the incidence of Salmonella spp., Campylobacter jejuni and Clostridium perfringens in wild birds near broiler chicken houses by sampling intestinal droppings
.
Avian Dis
.
44
:
715
720
.
23.
Cui,
S.,
Ge
B.,
Zheng
J.,
and
Meng
J.
2005
.
Prevalence and antimicrobial resistance of Campylobacter spp. and Salmonella serovars in organic chickens from Maryland retail stores
.
Appl. Environ. Microbiol
.
71
:
4108
4111
.
24.
DerSimonian,
R.,
and
Kacker
R.
2007
.
Random-effects model for meta-analysis of clinical trials: an update
.
Contemp. Clin. Trials
28
:
105
114
.
25.
Dickins,
M. A.,
Franklin
S.,
Stefanova
R.,
Schutze
G. E.,
Eisenach
K. D.,
Wesley
I.,
and
Cave
M. D.
2002
.
Diversity of Campylobacter isolates from retail poultry carcasses and from humans as demonstrated by pulsed-field gel electrophoresis
.
J. Food Prot
.
65
:
957
962
.
26.
Dimitri,
C.,
and
Oberholtzer
L.
2009
.
Marketing US organic foods: recent trends from farms to consumers
.
Available at: www.ers.usda.gov/Publications/EIB58/. Accessed 13 February 2019.
27.
Dobson,
A. J.,
and
Barnett
A. G.
2008
.
An introduction to generalized linear models
.
Chapman and Hall/CRC
,
New York
.
28.
Dogan,
O. B.,
Clarke
J.,
Mattos
F.,
and
Wang
B.
2019
.
A quantitative microbial risk assessment model of Campylobacter in broiler chickens: evaluating processing interventions
.
Food Control
100
:
97
110
.
29.
Egger,
M.,
Smith
G. D.,
Schneider
M.,
and
Minder
C.
1997
.
Bias in meta-analysis detected by a simple, graphical test
.
BMJ
315
:
629
634
.
30.
Erickson,
A. K.,
Murray
D. L.,
Ruesch
L. A.,
Thomas
M.,
Lau
Z.,
and
Scaria
J.
2018
.
Genotypic and phenotypic characterization of Salmonella isolated from fresh ground meats obtained from retail grocery stores in the Brookings, South Dakota, area
.
J. Food Prot
.
81
:
1526
1534
.
31.
Foley,
S. L.,
Nayak
R.,
Hanning
I. B.,
Johnson
T. J.,
Han
J.,
and
Ricke
S. C.
2011
.
Population dynamics of Salmonella enterica serotypes in commercial egg and poultry production
.
Appl. Environ. Microbiol
.
77
:
4273
4279
.
32.
Fratamico,
P. M.
2003
.
Comparison of culture, polymerase chain reaction (PCR), TaqMan Salmonella, and Transia Card Salmonella assays for detection of Salmonella spp. in naturally-contaminated ground chicken, ground turkey, and ground beef
.
Mol. Cell. Probes
17
:
215
221
.
33.
Freeman,
M. F.,
and
Tukey
J. W.
1950
.
Transformations related to the angular and the square root
.
Ann. Math. Stat
.
21
:
607
611
.
34.
Gad,
A. H.,
Abo-Shama
U. H.,
Harclerode
K. K.,
and
Fakhr
M. K.
2018
.
Prevalence, serotyping, molecular typing, and antimicrobial resistance of Salmonella isolated from conventional and organic retail ground poultry
.
Front. Microbiol
.
9
:
2653
.
35.
Geissler,
A. L.,
Bustos Carrillo
F.,
Swanson
K.,
Patrick
M. E.,
Fullerton
K. E.,
Bennett
C.,
Barrett
K.,
and
Mahon
B. E.
2017
.
Increasing Campylobacter infections, outbreaks, and antimicrobial resistance in the United States, 2004–2012
.
Clin. Infect. Dis
.
65
:
1624
1631
.
36.
Glass,
G. V.
1976
.
Primary, secondary, and meta-analysis of research
.
Educ. Res
.
5
:
3
8
.
37.
Goncalves-Tenorio,
A.,
Silva
B.,
Rodrigues
V.,
Cadavez
V.,
and
Gonzales-Barron
U.
2018
.
Prevalence of pathogens in poultry meat: a meta-analysis of European published surveys
.
Foods
7
:
69
.
38.
Gonzales-Barron,
U.,
and
Butler
F.
2011
.
The use of meta-analytical tools in risk assessment for food safety
.
Food Microbiol
.
28
:
823
827
.
39.
Gonzales-Barron,
U.,
Gonçalves-Tenório
A.,
Rodrigues
V.,
and
Cadavez
V.
2017
.
Foodborne pathogens in raw milk and cheese of sheep and goat origin: a meta-analysis approach
.
Curr. Opin. Food Sci
.
18
:
7
13
.
40.
Gonzales-Barron,
U.,
Thébault
A.,
Kooh
P.,
Watier
L.,
Sanaa
M.,
and
Cadavez
V.
2019
.
Strategy for systematic review of observational studies and meta-analysis modelling of risk factors for sporadic foodborne diseases
.
Microb. Risk Anal.
41.
Guerin,
M.,
Sir
C.,
Sargeant
J.,
Waddell
L.,
O'Connor
A.,
Wills
R.,
Bailey
R.,
and
Byrd
J.
2010
.
The change in prevalence of Campylobacter on chicken carcasses during processing: a systematic review
.
Poult. Sci
.
89
:
1070
1084
.
42.
Guran,
H. S.,
Mann
D.,
and
Alali
W. Q.
2017
.
Salmonella prevalence associated with chicken parts with and without skin from retail establishments in Atlanta metropolitan area, Georgia
.
Food Control
73
:
462
467
.
43.
Han,
F.,
Lestari
S. I.,
Pu
S.,
and
Ge
B.
2009
.
Prevalence and antimicrobial resistance among Campylobacter spp. in Louisiana retail chickens after the enrofloxacin ban
.
Foodborne Pathog. Dis
.
6
:
163
171
.
44.
Hanning,
I.,
Biswas
D.,
Herrera
P.,
Roesler
M.,
and
Ricke
S. C.
2010
.
Prevalence and characterization of Campylobacter jejuni isolated from pasture flock poultry
.
J. Food Sci
.
75
:
M496
M502
.
45.
Hardy,
B.,
Crilly
N.,
Pendleton
S.,
Andino
A.,
Wallis
A.,
Zhang
N.,
and
Hanning
I.
2013
.
Impact of rearing conditions on the microbiological quality of raw retail poultry meat
.
J. Food Sci
.
78
:
M1232
M1235
.
46.
Herbison,
P.,
Hay-Smith
J.,
and
Gillespie
W. J.
2006
.
Adjustment of meta-analyses on the basis of quality scores should be abandoned
.
J. Clin. Epidemiol
.
59
:
1249
1256
.
47.
Higgins,
J. P.,
Andreatti Filho
R. L.,
Higgins
S. E.,
Wolfenden
A. D.,
Tellez
G.,
and
Hargis
B. M.
2008
.
Evaluation of Salmonella-lytic properties of bacteriophages isolated from commercial broiler houses
.
Avian Dis
.
52
:
139
142
.
48.
Higgins,
J.,
Thomas
J.,
Chandler
J.,
Cumpston
M.,
Li
T.,
Page
M.,
and
Welch
V.
2019
.
Cochrane handbook for systematic reviews of interventions, 2nd ed
.
John Wiley & Sons
,
Chichester, UK
.
49.
Higgins,
J. P.,
Thompson
S. G.,
Deeks
J. J.,
and
Altman
D. G.
2003
.
Measuring inconsistency in meta-analyses
.
BMJ
327
:
557
560
.
50.
Hughner,
R. S.,
McDonagh
P.,
Prothero
A.,
Shultz
C. J.,
and
Stanton
J.
2007
.
Who are organic food consumers? A compilation and review of why people purchase organic food
.
J. Consumer Behav
.
6
:
94
110
.
51.
Jain,
S.,
and
Chen
J.
2006
.
Antibiotic resistance profiles and cell surface components of Salmonellae
.
J. Food Prot
.
69
:
1017
1023
.
52.
Jung,
Y.,
Porto-Fett
A. C. S.,
Shoyer
B. A.,
Henry
E.,
Shane
L. E.,
Osoria
M.,
and
Luchansky
J. B.
2019
.
Prevalence, levels, and viability of Salmonella in and on raw chicken livers
.
J. Food Prot
.
82
:
834
843
.
53.
Jüni,
P.,
Witschi
A.,
Bloch
R.,
and
Egger
M.
1999
.
The hazards of scoring the quality of clinical trials for meta-analysis
.
JAMA (J. Am. Med. Assoc.)
282
:
1054
1060
.
54.
Kegode,
R. B.,
Doetkott
D. K.,
Khaitsa
M. L.,
and
Wesley
I. V.
2008
.
Occurrence of Campylobacter species, Salmonella species and generic Escherichia coli in meat products from retail outlets in the Fargo metropolitan area
.
J. Food Saf
.
28
:
111
125
.
55.
Kerr,
A. K.,
Farrar
A. M.,
Waddell
L. A.,
Wilkins
W.,
Wilhelm
B. J.,
Bucher
O.,
Wills
R. W.,
Bailey
R. H.,
Varga
C.,
and
McEwen
S. A.
2013
.
A systematic review-meta-analysis and meta-regression on the effect of selected competitive exclusion products on Salmonella spp. prevalence and concentration in broiler chickens
.
Prev. Vet. Med
.
111
:
112
125
.
56.
Lestari,
S. I.,
Han
F. F.,
Fei
W.,
and
Ge
B. L.
2009
.
Prevalence and antimicrobial resistance of Salmonella serovars in conventional and organic chickens from Louisiana retail stores
.
J. Food Prot
.
72
:
1165
1172
.
57.
Line,
J. E.,
Stern
N. J.,
Lattuada
C. P.,
and
Benson
S. T.
2001
.
Comparison of methods for recovery and enumeration of Campylobacter from freshly processed broilers
.
J. Food Prot
.
64
:
982
986
.
58.
Lipsey,
M. W.,
and
Wilson
D. B.
2001
.
Practical meta-analysis
.
Sage Publications, Inc
.,
Thousand Oaks, CA
.
59.
Liu,
L.,
Hussain
S. K.,
Miller
R. S.,
and
Oyarzabal
O. A.
2009
.
Efficacy of mini VIDAS for the detection of Campylobacter spp. from retail broiler meat enriched in Bolton broth, with or without the supplementation of blood
.
J. Food Prot
.
72
:
2428
2432
.
60.
Luangtongkum,
T.,
Morishita
T. Y.,
Martin
L.,
Choi
I.,
Sahin
O.,
and
Zhang
Q.
2008
.
Prevalence of tetracycline-resistant Campylobacter in organic broilers during a production cycle
.
Avian Dis
.
52
:
487
490
.
61.
Luber,
P.
2009
.
Cross-contamination versus undercooking of poultry meat or eggs—which risks need to be managed first?
Int. J. Food Microbiol
.
134
:
21
28
.
62.
Maciorowski,
K.,
Jones
F.,
Pillai
S.,
and
Ricke
S.
2004
.
Incidence, sources, and control of food-borne Salmonella spp. in poultry feeds
.
Worlds Poult. Sci. J
.
60
:
446
457
.
63.
Melendez,
S. N.,
Hanning
I.,
Han
J.,
Nayak
R.,
Clement
A. R.,
Wooming
A.,
Hererra
P.,
Jones
F. T.,
Foley
S. L.,
and
Ricke
S. C.
2010
.
Salmonella enterica isolates from pasture-raised poultry exhibit antimicrobial resistance and class I integrons
.
J. Appl. Microbiol
.
109
:
1957
1966
.
64.
Mollenkopf,
D. F.,
Cenera
J. K.,
Bryant
E. M.,
King
C. A.,
Kashoma
I.,
Kumar
A.,
Funk
J. A.,
Rajashekara
G.,
and
Wittum
T. E.
2014
.
Organic or antibiotic-free labeling does not impact the recovery of enteric pathogens and antimicrobial-resistant Escherichia coli from fresh retail chicken
.
Foodborne Pathog. Dis
.
11
:
920
929
.
65.
Mollenkopf,
D. F.,
De Wolf
B.,
Feicht
S. M.,
Cenera
J. K.,
King
C. A.,
van Balen
J. C.,
and
Wittum
T. E.
2018
.
Salmonella spp. and extended-spectrum cephalosporin-resistant Escherichia coli frequently contaminate broiler chicken transport cages of an organic production company
.
Foodborne Pathog. Dis
.
15
:
583
588
.
66.
Musgrove,
M. T.,
Cox
N. A.,
Berrang
M. E.,
and
Harrison
M. A.
2003
.
Comparison of weep and carcass rinses for recovery of Campylobacter from retail broiler carcasses
.
J. Food Prot
.
66
:
1720
1723
.
67.
Myint,
M. S.,
Johnson
Y. J.,
Tablante
N. L.,
and
Heckert
R. A.
2006
.
The effect of pre-enrichment protocol on the sensitivity and specificity of PCR for detection of naturally contaminated Salmonella in raw poultry compared to conventional culture
.
Food Microbiol
.
23
:
599
604
.
68.
Nannapaneni,
R.,
Hanning
I.,
Wiggins
K. C.,
Story
R. P.,
Ricke
S. C.,
and
Johnson
M. G.
2009
.
Ciprofloxacin-resistant Campylobacter persists in raw retail chicken after the fluoroquinolone ban
.
Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess
.
26
:
1348
1353
.
69.
Nannapaneni,
R.,
Story
R.,
Wiggins
K. C.,
and
Johnson
M. G.
2005
.
Concurrent quantitation of total Campylobacter and total ciprofloxacin-resistant campylobacter loads in rinses from retail raw chicken carcasses from 2001 to 2003 by direct plating at 42 degrees C
.
Appl. Environ. Microbiol
.
71
:
4510
4515
.
70.
Nauta,
M.,
Hill
A.,
Rosenquist
H.,
Brynestad
S.,
Fetsch
A.,
van der Logt
P.,
Fazil
A.,
Christensen
B.,
Katsma
E.,
and
Borck
B.
2009
.
A comparison of risk assessments on Campylobacter in broiler meat
.
Int. J. Food Microbiol
.
129
:
107
123
.
71.
Noormohamed,
A.,
and
Fakhr
M. K.
2012
.
Incidence and antimicrobial resistance profiling of Campylobacter in retail chicken livers and gizzards
.
Foodborne Pathog. Dis
.
9
:
617
624
.
72.
Noormohamed,
A.,
and
Fakhr
M. K.
2014
.
Prevalence and antimicrobial susceptibility of Campylobacter spp. in Oklahoma conventional and organic retail poultry
.
Open Microbiol. J
.
8
:
130
137
.
73.
Oscar,
T. P.
2013
.
Initial contamination of chicken parts with Salmonella at retail and cross-contamination of cooked chicken with Salmonella from raw chicken during meal preparation
.
J. Food Prot
.
76
:
33
39
.
74.
Oscar,
T. P.
2014
.
Use of enrichment real-time PCR To enumerate Salmonella on chicken parts
.
J. Food Prot
.
77
:
1086
1092
.
75.
Oscar,
T. P.
2019
.
Process risk model for Salmonella and ground chicken
.
J. Appl. Microbiol
.
127
:
1236
1245
.
76.
Oscar,
T. P.,
Rutto
G. K.,
Ludwig
J. B.,
and
Parveen
S.
2010
.
Qualitative map of Salmonella contamination on young chicken carcasses
.
J. Food Prot
.
73
:
1596
1603
.
77.
Oyarzabal,
O. A.,
Macklin
K. S.,
Barbaree
J. M.,
and
Miller
R. S.
2005
.
Evaluation of agar plates for direct enumeration of Campylobacter spp. from poultry carcass rinses
.
Appl. Environ. Microbiol
.
71
:
3351
3354
.
78.
Oyarzabal,
O. A.,
Williams
A.,
Zhou
P.,
and
Samadpour
M.
2013
.
Improved protocol for isolation of Campylobacter spp. from retail broiler meat and use of pulsed field gel electrophoresis for the typing of isolates
.
J. Microbiol. Methods
95
:
76
83
.
79.
Parveen,
S.,
Taabodi
M.,
Schwarz
J. G.,
Oscar
T. P.,
Harter-Dennis
J.,
and
White
D. G.
2007
.
Prevalence and antimicrobial resistance of Salmonella recovered from processed poultry
.
J. Food Prot
.
70
:
2466
2472
.
80.
Price,
L. B.,
Johnson
E.,
Vailes
R.,
and
Silbergeld
E.
2005
.
Fluoroquinolone-resistant Campylobacter isolates from conventional and antibiotic-free chicken products
.
Environ. Health Perspect
.
113
:
557
560
.
81.
Price,
L. B.,
Lackey
L. G.,
Vailes
R.,
and
Silbergeld
E.
2007
.
The persistence of fluoroquinolone-resistant Campylobacter in poultry production
.
Environ. Health Perspect
.
115
:
1035
1039
.
82.
Rajan,
K.,
Shi
Z.,
and
Ricke
S. C.
2017
.
Current aspects of Salmonella contamination in the US poultry production chain and the potential application of risk strategies in understanding emerging hazards
.
Crit. Rev. Microbiol
.
43
:
370
392
.
83.
Ramirez-Hernandez,
A.,
Bugarel
M.,
Kumar
S.,
Thippareddi
H.,
Brashears
M. M.,
and
Sanchez-Plata
M. X.
2019
.
Phenotypic and genotypic characterization of antimicrobial resistance in Salmonella strains isolated from chicken carcasses and parts collected at different stages during processing
.
J. Food Prot
.
82
:
1793
1801
.
84.
R Core Team
.
2019
.
R: A language and environment for statistical computing
.
R Foundation for Statistical Computing
,
Vienna
.
85.
Richardson,
L. J.,
Cox
N. A.,
Bailey
J. S.,
Berrang
M. E.,
Cox
J. M.,
Buhr
R. J.,
Fedorka-Cray
P. J.,
and
Harrison
M. A.
2009
.
Evaluation of TECRA broth, Bolton broth, and direct plating for recovery of Campylobacter spp. from broiler carcass rinsates from commercial processing plants
.
J. Food Prot
.
72
:
972
977
.
86.
Rivera-Perez,
W.,
Barquero-Calvo
E.,
and
Zamora-Sanabria
R.
2014
.
Salmonella contamination risk points in broiler carcasses during slaughter line processing
.
J. Food Prot
.
77
:
2031
2034
.
87.
Rosenquist,
H.,
Boysen
L.,
Krogh
A. L.,
Jensen
A. N.,
and
Nauta
M.
2013
.
Campylobacter contamination and the relative risk of illness from organic broiler meat in comparison with conventional broiler meat
.
Int. J. Food Microbiol
.
162
:
226
230
.
88.
Rothrock,
M. J.,
Jr.,
Davis
M. L.,
Locatelli
A.,
Bodie
A.,
McIntosh
T. G.,
Donaldson
J. R.,
and
Ricke
S. C.
2017
.
Listeria occurrence in poultry flocks: detection and potential implications
.
Front. Vet. Sci
.
4
:
125
.
89.
Roy,
P.,
Dhillon
A. S.,
Lauerman
L. H.,
Schaberg
D. M.,
Bandli
D.,
and
Johnson
S.
2002
.
Results of Salmonella isolation from poultry products, poultry, poultry environment, and other characteristics
.
Avian Dis
.
46
:
17
24
.
90.
Salaheen,
S.,
Peng
M.,
and
Biswas
D.
2016
.
Ecological dynamics of Campylobacter in integrated mixed crop-livestock farms and its prevalence and survival ability in post-harvest products
.
Zoonoses Public Health
63
:
641
650
.
91.
Sanchez,
M. X.,
Fluckey
W. M.,
Brashears
M. M.,
and
McKee
S. R.
2002
.
Microbial profile and antibiotic susceptibility of Campylobacter spp. and Salmonella spp. in broilers processed in air-chilled and immersion-chilled environments
.
J. Food Prot
.
65
:
948
956
.
92.
Sargeant,
J. M.,
Amezcua
M. D. R.,
Rajic
A.,
and
Waddell
L.
2005
.
A guide to conducting systematic reviews in agri-food public health
.
Available at: http://publications.gc.ca/Collection/HP5-9-2005E.pdf. Accessed 13 October 2019.
93.
Sargeant,
J. M.,
Rajic
A.,
Read
S.,
and
Ohlsson
A.
2006
.
The process of systematic review and its application in agri-food public-health
.
Prev. Vet. Med
.
75
:
141
151
.
94.
Scallan,
E.,
Hoekstra
R. M.,
Angulo
F. J.,
Tauxe
R. V.,
Widdowson
M.-A.,
Roy
S. L.,
Jones
J. L.,
and
Griffin
P. M.
2011
.
Foodborne illness acquired in the United States—major pathogens
.
Emerg. Infect. Dis
.
17
:
7
.
95.
Scheinberg,
J.,
Doores
S.,
and
Cutter
C. N.
2013
.
A microbiological comparison of poultry products obtained from farmer's markets and supermarkets in Pennsylvania
J. Food Saf.
33
:
259
264
.
96.
Schroeder,
M. W.,
Eifert
J. D.,
Ponder
M. A.,
and
Schmale
D. G.
2014
.
Association of Campylobacter spp. levels between chicken grow-out environmental samples and processed carcasses
.
Poult. Sci
.
93
:
734
741
.
97.
Schwarzer,
G.
2007
.
meta: an R package for meta-analysis
.
R News
7
:
40
45
.
98.
Schwarzer,
G.,
Chemaitelly
H.,
Abu-Raddad
L. J.,
and
Rücker
G.
2019
.
Seriously misleading results using inverse of Freeman-Tukey double arcsine transformation in meta-analysis of single proportions
.
Res. Synth. Methods
10
:
476
483
.
99.
Sibanda,
N.,
McKenna
A.,
Richmond
A.,
Ricke
S. C.,
Callaway
T.,
Stratakos
A. C.,
Gundogdu
O.,
and
Corcionivoschi
N.
2018
.
A review of the effect of management practices on Campylobacter prevalence in poultry farms
.
Front. Microbiol
.
9
:
2002
.
100.
Siemon,
C. E.,
Bahnson
P. B.,
and
Gebreyes
W. A.
2007
.
Comparative investigations of prevalence and antimicrobial resistance of Salmonella between pasture and conventionally reared poultry
.
Avian Dis
.
51
:
112
117
.
101.
Simmons,
M.,
Fletcher
D. L.,
Cason
J. A.,
and
Berrang
M. E.
2003
.
Recovery of Salmonella from retail broilers by a whole-carcass enrichment procedure
.
J. Food Prot
.
66
:
446
450
.
102.
Sofos,
J. N.
2008
.
Challenges to meat safety in the 21st century
.
Meat Sci
.
78
:
3
13
.
103.
Son,
I.,
Englen
M. D.,
Berrang
M. E.,
Fedorka-Cray
P. J.,
and
Harrison
M. A.
2007
.
Prevalence of Arcobacter and Campylobacter on broiler carcasses during processing
.
Int. J. Food Microbiol
.
113
:
16
22
.
104.
Spector,
T. D.,
and
Thompson
S. G.
1991
.
The potential and limitations of meta-analysis
.
J. Epidemiol. Community Health
45
:
89
.
105.
Stern,
N. J.,
and
Pretanik
S.
2006
.
Counts of Campylobacter spp. on U.S. broiler carcasses
.
J. Food Prot
.
69
:
1034
1039
.
106.
Sterne,
J. A.,
Sutton
A. J.,
Ioannidis
J. P.,
Terrin
N.,
Jones
D. R.,
Lau
J.,
Carpenter
J.,
Rücker
G.,
Harbord
R. M.,
and
Schmid
C. H.
2011
.
Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials
.
BMJ
343
:
d4002.
107.
Stijnen,
T.,
Hamza
T. H.,
and
Özdemir
P.
2010
.
Random effects meta-analysis of event outcome in the framework of the generalized linear mixed model with applications in sparse data
.
Stat. Med
.
29
:
3046
3067
.
108.
Sutton,
A. J.,
Abrams
K. R.,
and
Jones
D. R.
2001
.
An illustrated guide to the methods of meta-analysis
.
J. Eval. Clin. Pract
.
7
:
135
148
.
109.
Thakur,
S.,
Brake
J.,
Keelara
S.,
Zou
M.,
and
Susick
E.
2013
.
Farm and environmental distribution of Campylobacter and Salmonella in broiler flocks
.
Res. Vet. Sci
.
94
:
33
42
.
110.
Totton,
S. C.,
Farrar
A. M.,
Wilkins
W.,
Bucher
O.,
Waddell
L. A.,
Wilhelm
B. J.,
McEwen
S. A.,
and
Rajić
A.
2012
.
A systematic review and meta-analysis of the effectiveness of biosecurity and vaccination in reducing Salmonella spp. in broiler chickens
.
Food Res. Int
.
45
:
617
627
.
111.
Trimble,
L. M.,
Alali
W. Q.,
Gibson
K. E.,
Ricke
S. C.,
Crandall
P.,
Jaroni
D.,
and
Berrang
M.
2013
.
Salmonella and Campylobacter prevalence and concentration on pasture-raised broilers processed on-farm, in a Mobile Processing Unit, and at small USDA-inspected facilities
.
Food Control
34
:
177
182
.
112.
Trimble,
L. M.,
Alali
W. Q.,
Gibson
K. E.,
Ricke
S. C.,
Crandall
P.,
Jaroni
D.,
Berrang
M.,
and
Habteselassie
M. Y.
2013
.
Prevalence and concentration of Salmonella and Campylobacter in the processing environment of small-scale pastured broiler farms
.
Poult. Sci
.
92
:
3060
3066
.
113.
U.S. Department of Agriculture
.
2009
.
The nationwide microbiological baseline data collection program: young chicken survey
.
114.
U.S. Department of Agriculture
.
2012
.
The nationwide microbiological baseline data collection program: raw chicken parts survey
.
115.
van Gerwen,
S. J.,
and
Zwietering
M. H.
1998
.
Growth and inactivation models to be used in quantitative risk assessments
.
J. Food Prot
.
61
:
1541
1549
.
116.
van Immerseel,
F.,
De Zutter
L.,
Houf
K.,
Pasmans
F.,
Haesebrouck
F.,
and
Ducatelle
R.
2009
.
Strategies to control Salmonella in the broiler production chain
.
Worlds Poult. Sci. J
.
65
:
367
392
.
117.
van Loo,
E. J.,
Caputo
V.,
Nayga,
R. M.
Jr.,
Meullenet
J.-F.,
and
Ricke
S. C.
2011
.
Consumers' willingness to pay for organic chicken breast: evidence from choice experiment
.
Food Qual. Prefer
.
22
:
603
613
.
118.
Viechtbauer,
W.
2010
.
Conducting meta-analyses in R with the metafor package
.
J. Stat. Softw
.
36
:
1
48
.
119.
Volkova,
V. V.,
Bailey
R. H.,
and
Wills
R. W.
2009
.
Salmonella in broiler litter and properties of soil at farm location
.
PLoS One
4
:
e6403
.
120.
Voss-Rech,
D.,
Potter
L.,
Vaz
C. S. L.,
Pereira
D. I. B.,
Sangioni
L. A.,
Vargas
A. C.,
and
de Avila Botton
S.
2017
.
Antimicrobial resistance in nontyphoidal Salmonella isolated from human and poultry-related samples in Brazil: 20-year meta-analysis
.
Foodborne Pathog. Dis
.
14
:
116
124
.
121.
Walls,
I.,
and
Scott
V. N.
1997
.
Use of predictive microbiology in microbial food safety risk assessment
.
Int. J. Food Microbiol
.
36
:
97
102
.
122.
Warton,
D. I.,
and
Hui
F. K.
2011
.
The arcsine is asinine: the analysis of proportions in ecology
.
Ecology
92
:
3
10
.
123.
White,
D. G.,
Zhao
S.,
Sudler
R.,
Ayers
S.,
Friedman
S.,
Chen
S.,
McDermott
P. F.,
McDermott
S.,
Wagner
D. D.,
and
Meng
J.
2001
.
The isolation of antibiotic-resistant Salmonella from retail ground meats
.
N. Engl. J. Med
.
345
:
1147
1154
.
124.
White,
P. L.,
Naugle
A. L.,
Jackson
C. R.,
Fedorka-Cray
P. J.,
Rose
B. E.,
Pritchard
K. M.,
Levine
P.,
Saini
P. K.,
Schroeder
C. M.,
Dreyfuss
M. S.,
Tan
R.,
Holt
K. G.,
Harman
J.,
and
Buchanan
S.
2007
.
Salmonella Enteritidis in meat, poultry, and pasteurized egg products regulated by the US food safety and inspection service, 1998 through 2003
.
J. Food Prot
.
70
:
582
591
.
125.
Williams,
A.,
and
Oyarzabal
O. A.
2012
.
Prevalence of Campylobacter spp. in skinless, boneless retail broiler meat from 2005 through 2011 in Alabama, USA
.
BMC Microbiol
.
12
:
184
.
126.
Young,
I.,
Rajić
A.,
Wilhelm
B.,
Waddell
L.,
Parker
S.,
and
McEwen
S.
2009
.
Comparison of the prevalence of bacterial enteropathogens, potentially zoonotic bacteria and bacterial resistance to antimicrobials in organic and conventional poultry, swine and beef production: a systematic review and meta-analysis
.
Epidemiol. Infect
.
137
:
1217
1232
.
127.
Zhang,
J. Y.,
Massow
A.,
Stanley
M.,
Papariella
M.,
Chen
X.,
Kraft
B.,
and
Ebner
P.
2011
.
Contamination rates and antimicrobial resistance in Enterococcus spp., Escherichia coli, and Salmonella isolated from “no antibiotics added”-labeled chicken products
.
Foodborne Pathog. Dis
.
8
:
1147
1152
.
128.
Zhao,
C.,
Ge
B.,
De Villena
J.,
Sudler
R.,
Yeh
E.,
Zhao
S.,
White
D. G.,
Wagner
D.,
and
Meng
J.
2001
.
Prevalence of Campylobacter spp., Escherichia coli, and Salmonella serovars in retail chicken, turkey, pork, and beef from the Greater Washington, D.C., area
.
Appl. Environ. Microbiol
.
67
:
5431
5436
.
129.
Zhao,
S.,
McDermott
P. F.,
Friedman
S.,
Abbott
J.,
Ayers
S.,
Glenn
A.,
Hall-Robinson
E.,
Hubert
S. K.,
Harbottle
H.,
and
Walker
R. D.
2006
.
Antimicrobial resistance and genetic relatedness among Salmonella from retail foods of animal origin: NARMS retail meat surveillance
.
Foodborne Pathog. Dis
.
3
:
106
117
.
130.
Zhao,
S.,
Young
S. R.,
Tong
E.,
Abbott
J. W.,
Womack
N.,
Friedman
S. L.,
and
McDermott
P. F.
2010
.
Antimicrobial resistance of Campylobacter isolates from retail meat in the United States between 2002 and 2007
.
Appl. Environ. Microbiol
.
76
:
7949
7956
.