SciELO - Scientific Electronic Library Online

 
vol.44 número3Contenido de alimento y metabolismo ceco-cólico en el tracto digestivo de poblaciones silvestres de iguana negra (Ctenosaura pectinata) en Morelos, MéxicoCalidad de la leche cruda en unidades de producción familiar del sur de Ciudad de México índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

Compartir


Archivos de medicina veterinaria

versión impresa ISSN 0301-732X

Arch. med. vet. vol.44 no.3 Valdivia  2012

http://dx.doi.org/10.4067/S0301-732X2012000300004 

Arch Med Vet 44, 225-235 (2012)

ORIGINAL ARTICLE

 

Knowledge and attitudes towards food safety and reported use of good production practices among a sample of cattle producers in Santa Fe, Argentina #

Conocimiento y actitudes hacia seguridad alimentaria y buenas prácticas de producción adoptadas en productores ganaderos de Santa Fe, Argentina

 

I Young a,b*, A Rajić a,b, E Perez c, J Sanchez d, A Larriestra e, LA Perez f, O Drivet f, M Monteverde f, SA McEwen b

a Laboratory for Foodborne Zoonoses, Public Health Agency of Canada, Guelph, Ontario, Canada.
b Department of Population Medicine, University of Guelph, Guelph, Ontario, Canada.
c Area of Health Surveillance and Disease Prevention and Control, Pan-American Health Organization/World Health Organization, Representation in Panama, Panama.
d Department of Health Management, University of Prince Edward Island, Charlottetown, PE, Canada.
e Departamento de Patología Animal, Universidad Nacional de Río Cuarto, Córdoba, Argentina.
f Colegio de Médicos Veterinarios de Santa Fe, Santa Fe, Argentina.
# Funded by the Health Canada and Pan-American Health Organization Canadian Biennial Work Plan Fund.
* iyoung@uoguelph.ca.


RESUMEN

La implementación de buenas prácticas de producción (BPP) en establecimientos pecuarios y la traducción efectiva de conocimiento sobre inocuidad de los alimentos a productores agropecuarios se recomienda para un suministro de alimentos más seguros. Un estudio piloto fue realizado durante 2009-2010 para evaluar el conocimiento y las actitudes hacia la inocuidad de los alimentos y el uso de BPP entre 930 ganaderos en Santa Fe, Argentina. El porcentaje de respuestas de las encuestas fue 31,8% (n = 296). Varios encuestados indicaron que raramente o nunca aislaban al ganado enfermo (25,8%), mantenían registros de enfermedades (32,5%) o del uso de antibióticos (43,3%), y aseguraban que los visitantes y los empleados del establecimiento se lavaran las manos (79,2% y 31,2%) y usaran ropa de protección (79,0% y 31,3%). La mediana de las respuestas de las 13 BPP encuestadas fue evaluada en un modelo ordinal de regresión logística. Capacitación previa sobre inocuidad de los alimentos (OR = 2,59), discusión frecuente (OR = 5,89) o siempre (OR = 6,33) de inocuidad de los alimentos con el veterinario, y ser productor de leche comparado a un productor de cría (OR = 3,86) fueron asociados con un mejor uso de BPP. Un 40% de la variación total en el uso de BPP se debió a factores relacionados a veterinarios, indicando que deberían tener un papel importante en la educación de los productores ganaderos sobre inocuidad de los alimentos. Estos resultados iniciales se deben utilizar para apoyar la toma de decisiones sobre la inocuidad de los alimentos en la producción de ganado en Santa Fe.

Palabras clave: conocimiento, actitudes, prácticas, productor ganadero.


SUMMARY

On-farm implementation of good production practices (GPP) and effective translation of food safety knowledge to food-animal producers are recommended to achieve a safer food supply. A pilot study was conducted during 2009-2010 to assess the knowledge and attitudes towards food safety and reported use of GPp among a sample of 930 cattle producers in Santa Fe, Argentina. A response percentage of 31.8% (n = 296) was obtained. Several respondents indicated that they rarely or never isolate sick cattle (25.8%), keep records of diseases (32.5%) or antimicrobial use (43.3%), and ensure that farm visitors and employees, respectively, wash their hands (79.2% and 31.2%) and wear protective clothing (79.0% and 31.3%). Respondents' median use of 13 GPp was calculated and evaluated in a multivariable ordinal regression model. Previous training in food safety (OR = 2.59), often (OR= 5.89) or always (OR= 6.33) discussing food safety with the veterinarian, and being a dairy producer compared to cow-calf producer (OR= 3.86) were associated with a higher median use of GPP. Approximately 40% of the total variation in respondents' median use of GPp was attributable to veterinarians, indicating that they should have an important role in the education of cattle producers about food safety in Santa Fe. These preliminary results should be used to inform future research and decision-making about food safety and GPp in cattle production in Argentina.

Key words: knowledge, attitudes, practices, cattle producer.


 

INTRODUCTION

Foodborne and gastrointestinal disease cause significant morbidity and mortality in the human population and result in an important economic burden worldwide (Kosek et al 2003, Scallan et al 2011). For example, the mean annual incidence of human gastrointestinal illness in a region of Santa Fe, Argentina, was estimated to range from 0.46 to 1.68 episodes per person-year (Thomas et al 2010). Argentina also has a high incidence rate of haemolytic uremic syndrome, a life-threatening complication of Shiga toxin-producing Escherichia coli (STEC) infection (Rivas et al 2006). Beef and dairy cattle are important sources of STEC and other foodborne pathogens such as Brucella spp., Salmonella spp. and Listeria monocytogenes (Samartino 2002, Fossler et al 2005, Vilar et al 2007, Fernandez et al 2009). STEC has been isolated with various frequencies from cattle feces, the cattle farm environment and retail beef products in Argentina (Meichtri et al 2004, Fernandez et al 2009, Etcheverria et al 2010, Tanaro et al 2010), which is a concern given that the consumption of undercooked beef is an important source of sporadic STEC infections (Rivas et al 2008).

The implementation of on-farm good production practices (GPP) by food-animal producers can help to minimize the risks of pathogen contamination throughout the food chain by using a farm-to-fork approach (OIE Animal Production Food Safety Working Group 2006). "GPP" are defined as a collection of on-farm management and biosecurity practices that can be implemented by producers to prevent the risk of contamination or infection of their animals with microbial, chemical or physical food safety hazards (Young 2010). To ensure successful implementation of GPP, producers should understand the importance of implementing GPp and they should have knowledge of the food safety hazards that GPp are designed to control. Previous surveys have investigated cattle producers' knowledge and attitudes towards food safety and use of GPp in the United States of America (USA), Canada and the United Kingdom (Hoe and Ruegg 2006, Gunn et al 2008, Young et al 2010a, Young et al 2010b). However, information on these factors has not been previously investigated in Argentina, a leading cattle-producing country without formal on-farm food safety programs for cattle production. This information could be used to develop and measure the effectiveness of such programs. In addition, there is a need to investigate the role of veterinarians in the use of GPp among producers in Argentina, because previous studies have indicated that producers view veterinarians as trusted and important sources of information about food safety and GPp (Gunn et al 2008, Young et al 2010a). An exploratory, pilot study was conducted to determine preliminary information about the knowledge and attitudes towards food safety and use of GPp among a sample of cattle producers in Santa Fe, Argentina. A secondary objective was to explore whether the reported use of GPp among producers in this study was associated with the frequency of veterinarian visits to the farm and frequency of discussing food safety and GPp with the veterinarian. The Province of Santa Fe was selected as the site for this study because it is the second largest producer of beef cattle, with over 6,900,000 head of cattle in 2009, and it is one of the largest milk-producing provinces in Argentina (Instituto Nacional de Tecnología Agropecuaria 2010, Taverna 2010).

MATERIAL AND METHODS

QUESTIONNAIRE DEVELOPMENT

We developed a 12-page questionnaire based on similar questionnaires that were administered to producers in North America (Hoe and Ruegg 2006, Young et al 2010a, Young et al 2010b). The questionnaire was developed in English and translated to Spanish, and then it was reviewed and discussed with 12 local veterinarians of the Colegio de Médicos Veterinarios de Santa Fe (CMVSF) during a knowledge-exchange workshop about food safety and GPp in Santa Fe, Argentina (September, 2009). A more detailed review, pre-test and finalization of the questionnaire was conducted during the second day of the workshop, which included the study authors and three selected veterinarians. Unfortunately, we were not able to pre-test the questionnaire with a sample of producers due to limited resources and logistical constraints.

The final questionnaire consisted of three broad sections: demographics, management practices, and food safety. The 'demographics' section contained 11 multiple choice and yes/no questions about the farm operator's age (n = 1 question), type of animals produced (n = 1), organic status (n = 1), use of antimicrobials and feed supplements (n = 3), interactions with the veterinarian (n = 2), and previous training and interest in learning more about food safety (n = 3). This section also included one open-ended question about the number of animals produced. The 'management practices' section consisted of seven multiple choice and yes/no questions about pest control (n = 1 question), dead cattle disposal (n = 1), purchase of replacement cattle (n = 4) and on-farm slaughter and processing (n = 1). It also contained two questions about the use of 13 different GPp on a five-point scale (from "never" to "always"). The 'food safety' section consisted of three multiple choice questions about antimicrobial resistance (AMR) (n = 2) and knowledge of foodborne pathogens (n = 1), and one question about stakeholder knowledge of food safety measured on a five-point scale (from "not" to "very" knowledgeable). Additional questions (n = 12) were asked in the questionnaire but are not reported here for brevity reasons. A copy of the questionnaire is available as supplementary material.

QUESTIONNAIRE ADMINISTRATION

The target population for this pilot study was cattle producer clients of veterinarians from the CMVSF. Veterinarians must be registered with the CMVSF in order to practice in Santa Fe. The target population was not intended to represent all producers in Santa Fe, but to obtain preliminary baseline information about the range of knowledge, attitudes and use of GPp in the province. A total of 930 questionnaires were distributed to a convenience sample of 58 of 600 (9.7%) large-animal veterinarians from the CMVSF. Each veterinarian received approximately 5-10 questionnaires by mail or email from September to November, 2009, to distribute to producers. Veterinarians were contacted by study authors from the CMVSF by telephone before being sent questionnaires to explain the study purpose and give instructions on administration. Veterinarians administered the questionnaires during their regularly scheduled client farm visits. They were given instructions not to complete the questionnaires on behalf of the producers, but they were not blinded to the results. They introduced the study to producers and, among those who agreed to participate, provided a questionnaire for completion during the farm visit. Veterinarians collected the completed questionnaires and returned them in-person to the CMVSF office in Santa Fe. Follow-up telephone calls were made with veterinarians in February, 2010, to remind them to complete and return the questionnaires. This study received ethics approval from the University of Guelph Research Ethics Board (protocol #09AU020).

STATISTICAL ANALYSIS

Questionnaires were entered into an Access database (Microsoft Corporation, Redmond, WA). Descriptive tabulations and summaries were conducted for each variable with missing values excluded. The proportion of cattle farms stratified by herd size (total number of cattle) for beef cattle respondents was compared to the provincial average in 2009 using a chi-square goodness-of-fit test (INTA 2010). Dairy respondents' herd size (number of milking cows) was compared to the provincial average for 2005 using a one-sample t-test (Gobierno de Santa Fe 2006). Descriptive analyses were performed in Stata 10.0 (Stata Corporation, College Station, TX).

A multilevel ordinal regression model was used to identify predictors associated with respondents' median use of GPP. The model was estimated with the GLLAMM command in Stata 10.0 using adaptive quadrature (Rabe-Hesketh and Skrondal 2008). The model outcome was an index variable representing respondents' median use of 13 different GPp measured on a five-point scale (never, rarely, sometimes, often, always). For example, if a respondent indicated that they "rarely", "sometimes", and "often" use six, one, and six of the 13 GPP, respectively, their median use of GPp would be "sometimes" because it is the middle value. Two levels of variation were included in the model: level one was the producer (i.e. respondent) and level two was the producer's veterinarian. Ordinal regression models assume that the outcome represents categories of an underlying continuous latent variable with a logistic distribution (Rabe-Hesketh and Skrondal 2008). Therefore, if σvet represents the variance due to the producer's veterinarian, the proportion of the total variance in the outcome attributable to veterinarians (ρ) was calculated as follows (Rabe-Hesketh and Skrondal 2008):

Eight variables were pre-selected as predictors of interest and included in the model-building process: cattle producer type (cow-calf, feedlot, dairy, mixed production); herd size (total number of cattle); farm operator age (< 40, 40-49, 50-59 and ≥ 60 years); production of other food animals (yes/no); organic status (yes/no); previous completion of a course or seminar about GPp and food safety (yes/no); frequency of veterinarian visits to the farm (≤ 4, 5-8 or > 8 times/year); and frequency of discussing food safety and GPp with the veterinarian (never, rarely, sometimes, often or always). The latter two variables were our primary predictors of interest, while the other six variables were included because they were considered to be potential confounding variables based on causal reasoning and previous research (Hoe and Ruegg 2006, Young et al 2010a, Young et al 2010b). Spearman correlations and chi-square tests were used to investigate collinearity and associations between each pair of predictors. The predictors were screened in univariable ordinal regression models and were entered into a multivariable model if P ≤ 0.20. A manual backwards-selection process was used to achieve the final model. Significance was assessed using likelihood-ratio tests and predictors were retained if P ≤ 0.05. All variables were re-evaluated for significance and assessed for evidence of confounding (changes of > 20% in the coefficients of other predictors) in the final model. Two-way interactions were investigated between all predictors in the final model.

The ordinal regression model estimates only one coefficient for each predictor, which assumes that the coefficients do not depend on the outcome level. This is referred to as the proportional-odds assumption, and it was assessed by comparing the final model to a model with an estimated coefficient for each level of the outcome variable using a likelihood-ratio test. P > 0.05 was selected to indicate that this assumption is not violated (Dohoo et al 2003). The influence of veterinarians on the model was investigated by examining predicted probabilities stratified by veterinarian.

RESULTS

DEMOGRAPHICS

A total of 313 questionnaires were returned, 17 of which were removed from the dataset because of incomplete responses (n = 11) or because respondents were not cattle producers (n = 6), leading to a final response of 31.8% (296/930). A median of three questionnaires was returned per veterinarian (SD = = 5, range = 1-26). Of the 296 respondents, 58.8% were cow-calf producers, 23.0% were dairy producers, 7.1% were feedlot producers and 11.2% produced multiple cattle types. The median herd size was 249 (SD = 548, 25-75th percentile = 124-550) among beef cattle producers and 173 (SD = 152, 25-75th percentile = 100-247) among dairy cattle producers. The beef-cattle producers in this survey had a larger farm size compared to the provincial average (P < 0.001; table 2), while dairy producers had a similar number of milking cows compared to the provincial average (P = 0.875; mean = 190 vs. 187, respectively).

FARM CHARACTERISTICS, KNOWLEDGE AND ATTITUDES TOWARDS FOOD SAFETY, AND USE OF GPP

Respondents' farm characteristics and attitudes towards food safety are shown in table 1. Their rating of the level of knowledge about food safety and GPp among different stakeholders is shown in figure 1, and their knowledge of different foodborne pathogens is shown in figure 2. Respondents' use of general farm-management practices is shown in table 3 and their use of GPp measured on a five-point scale is shown in table 4.

Table 1. Respondents' farm characteristics and attitudes towards food safety, Santa Fe, Argentina.
Características y actitudes hacia la inocuidad de los alimentos entre entrevistados, Santa Fe, Argentina.
 

Table 2. Comparison of herd size strata among respondents and the provincial average for beef cattle producers, Santa Fe, Argentina.
Comparación de estrato tamaño de hato entre los entrevistados y promedio provincial para los productores ganaderos, Santa Fe, Argentina.
 
 
a Survey respondents were significantly different than the provincial average (P < 0.001).

 
Figure 1. Stakeholder knowledge about GPp and food safety as rated by respondents, Santa Fe, Argentina.
El conocimiento de interesados acerca de buenas prácticas de producción y seguridad de los alimentos tan tasado por los entrevistados, Santa Fe, Argentina.

 
Figure 2. Respondents' knowledge of whether foodborne pathogens can be transmitted from contaminated food to humans and cause disease, Santa Fe, Argentina.
El conocimiento de entrevistados de ya sea los agentes patógenos transmitidos por los alimentos pueden transmitirse de los alimentos contaminados a los seres humanos y causar enfermedad, Santa Fe, Argentina.

Table 3. Respondents' reported use of general farm-management practices, Santa Fe, Argentina.
El uso informado de las prácticas de manejo generales entre entrevistados, Santa Fe, Argentina.
 

Table 4. Respondents' reported use of GPP, Santa Fe, Argentina.
El uso informado de buenas prácticas de producción entre entrevistados, Santa Fe, Argentina.
 
 
Calculated from the median of the above 13 GPp variables and used as the outcome in the multilevel ordinal regression model.

 

MULTILEVEL ORDINAL REGRESSION

Results from the final multivariable ordinal regression model are shown in table 5. The likelihood ratio test indicated that the proportional-odds assumption was valid (P = 0.279). No significant interactions were identified. Respondents that were dairy producers were more likely to report a higher median use of GPp compared to cow-calf producers, while feedlot and mixed cattle producers were not significantly different. Respondents that often or always discussed GPp and food safety with their veterinarian and that had previously taken a course or seminar about these topics were also more likely to report a higher median use of GPp (table 5). The odds ratios in table 5 indicate the odds of respondents having a median use of GPp above any given level compared to being at or below that level. For example, respondents that previously completed a course or seminar about GPp and food safety were 2.59 times more likely to have a median use of GPp of "sometimes" compared to "rarely or never", "often" compared to "sometimes, rarely or never", and "always" compared to any other response. Veterinarians contributed to 40.5% of the total variation in respondents' median use of GPp (σvet = 2.24). The model was recalculated with only "cattle producer type" included to examine its influence on the variation attributable to veterinarians, but only small changes were noted (43.0%). To highlight the influence of different veterinarians on producer's use of GPP, figure 3 shows the predicted probabilities of the model stratified by veterinarian identification number. The veterinarian with the lowest predicted probabilities (ID = 38) had four producer responses, all of whom had never completed training in food safety, rarely (n = 3) or sometimes (n = 1) discussed food safety with the veterinarian and had a median use of GPp of rarely (n = 1) or never (n = 3). The veterinarian with the most producer responses (ID = 9, n = 26) had very high predicted probabilities.

Table 5. Final multivariable ordinal regression model of predictors associated with respondents' median use of GPP, Santa Fe, Argentina.
Modelo final de regresión ordinal multivariable de los factores predictivos asociados con la mediana de las respuestas de buenas prácticas de producción entre entrevistados, Santa Fe, Argentina.

 
 
a Model characteristics: N = 247; variance attributable to veterinarians, 2.24; likelihood ratio test of the proportional-odds assumption, P = 0.279. OR, odds ratio; SE, standard error; CI, confidence interval.

 
Figure 3. Box plot of predicted probabilities from the final multivariable ordinal regression model stratified by veterinarian. The predicted probabilities represent the probability of having a median use of GPp of "sometimes, often or always" vs. "rarely or never". Veterinarians with only one producer response (n = 18) were excluded from this graph.
El diagrama de bloques de probabilidades predichas del modelo final de regresión ordinal multivariable estratificado por el veterinario. Las probabilidades predichas representan la probabilidad de tener un uso mediano de buenas prácticas de producción de "a veces, a menudo o siempre" versus "raramente o nunca". Los veterinarios con sólo una respuesta productora (n = 18) fueron excluidos de este gráfico.

DISCUSSION

The response percentage (31.8%) was lower than in similar surveys of dairy and beef cattle producers in the USA (Hoe and Ruegg 2006, Brandt et al 2008), but higher than in a recent Canadian survey (20.9%) (Young et al 2010a). Due to the low response, selection bias is a potential limitation of this study. For example, respondents to the questionnaire could have had a higher use of GPP, a greater knowledge of food safety or a stronger relationship with their veterinarian compared to non-respondents. In addition, veterinarians that participated in this study might have had a stronger relationship with their clients and might have been stronger advocates and educators about food safety and GPp to their clients compared to veterinarians that did not participate. Unfortunately, due to logistical constraints we could not determine the characteristics of producers and veterinarians that chose not to participate. Therefore, we cannot determine the true extent and impact of potential non-responses biases on the study findings, and the results should be interpreted with caution.

The target population for this study was not intended to represent all producers in Santa Fe, and the comparisons of respondents' herd size to the provincial average were conducted for illustration purposes only. However, they indicate that beef cattle producer respondents may have had a larger herd size than the provincial average. In addition, dairy producers might have been over-represented in this sample, as the approximate proportion of cattle producers in Santa Fe that produce beef and dairy cattle is 84% and 16%, respectively (Gobierno de Santa Fe 2006, INTA 2010). Therefore, the results of this study might be less applicable to producers with smaller herds. Despite these limitations, we believe that the results still provide useful information that can be used to inform larger surveys of cattle producers in Santa Fe and for targeting areas for future education and knowledge transfer.

Most respondents (73.3%) indicated that their veterinarian visited their farms more than four times a year, and nearly 75% indicated that they sometimes, often or always talk about GPp and food safety with their veterinarian. These results suggest that cattle producers in this study have regular interactions with their veterinarian. In addition, most respondents viewed veterinarians as the most knowledgeable stakeholder and preferred source to learn about food safety and GPP, which corresponds with results from previous research (Vanbaale et al 2003, Gunn et al 2008, Young et al 2010a). Veterinarians in Santa Fe should continue to discuss food safety and GPp issues regularly with their producer clients.

Respondents' knowledge of Brucella spp., Trichinella spp. and E. coli 0157 as foodborne pathogens is not surprising given that they cause endemic disease in humans in Argentina. In addition, these responses might be reflective of Argentina's brucellosis control program in cattle and public education about E. coli 0157 (Samartino 2002). In contrast, many respondents were not familiar with Listeria spp. (62.1%) or Salmonella spp. (35.5%). Cattle are an important reservoir for these pathogens, which they can shed in their faeces or milk with or without showing clinical signs of illness (Oliver et al 2005). Neither listeriosis nor non-typhoidal salmonellosis are nationally notifiable diseases in Argentina, so most human cases of these diseases are likely not detected. Future education with cattle producers in Santa Fe should highlight the importance of Listeria and Salmonella as foodborne pathogens.

Most respondents believed that AMR is affecting their treatment of sick cattle and that AMR in humans is linked to antimicrobial use in food animals. These results correspond to high concern about AMR expressed by dairy producers in North America (Raymond et al 2006, Young et al 2010a). Almost all respondents indicated that they use antimicrobials to treat sick cattle, although no data were collected about the amount of antimicrobials typically used for this purpose. The importance of judicious antimicrobial use should be promoted to producers and veterinarians in Santa Fe to minimize the potential for AMR selection pressure in pathogens associated with cattle. The use of other feed supplements was rarely reported among respondents (< 2.5%), although evidence suggests that some feed and water additives, such as probiotics and certain organic acids, may be effective in reducing cattle shedding of pathogens such as STEC (Sargeant et al 2007). The use of feed supplements other than antimicrobials should be considered as part of a comprehensive on-farm food safety strategy for cattle production in Santa Fe.

More than 20% of respondents indicated that they don't take any preventive measures when adding replacement cattle to their herd. This is a concern given that the introduction of purchased cattle into a herd is an identified risk factor for herd infection with infectious disease agents such as STEC (Schouten et al 2004). Cattle producers in Santa Fe should be informed about the potential risks of introducing replacement cattle into their herd without proper screening for pathogens or animal quarantine. Additionally, most respondents (> 80%) indicated that they leave their dead cattle in a cemetery area of the farm, which could lead to the possible transmission of infectious disease agents between wild scavenger animals and cattle. On-farm processing of meat was reported by 25% of respondents. However, the questionnaire did not distinguish between on-farm meat processing for personal use or sale to the public. Veterinary and food safety officials in Santa Fe should monitor on-farm meat processing by cattle producers to ensure that establishments that sell processed meat products to the public follow appropriate food safety standards and requirements.

Some areas for improvement were identified in respondents' reported use of GPP, such as isolating sick animals in an area separate from healthy animals. This practice is important, particularly in dairy cattle production, because sick cattle are more likely to shed pathogens such as Salmonella and could be a source of infection for other cattle in the herd (Fossler et al 2005). Respondents' record-keeping practices could also be improved to help ensure appropriate monitoring of disease occurrences and judicious antimicrobial use on the farm. Restricting farm entry and ensuring that protective clothing is worn by farm visitors and employees was another gap in respondents' use of GPP. These practices are important to prevent the introduction of infectious disease agents such as STEC and foot-and-mouth disease virus into the herd, particularly if visitors or employees have recently visited another farm (van Schaik et al 2002, Ellis-Iversen et al 2011). However, these practices might be less practical or feasible for smaller farms with limited visitor contact.

The ordinal regression model results indicated that respondents who often or always discussed food safety or GPp with their veterinarian had a higher median use of GPP. This result supports our hypothesis that veterinarian extension and knowledge exchange with cattle producers in this study is associated with producer use of GPP. In addition, respondents that had taken an educational course or seminar about food safety and GPp had a higher median use of GPP. Other studies have also reported that the completion of educational courses is associated with more positive attitudes towards food safety and more frequent use of GPp (Moore and Payne 2007, Young et al 2010a, Young et al 2010b), underscoring the importance of providing training opportunities to cattle producers. Respondents' who were dairy producers had a higher median use of GPp compared to cow-calf producers. This difference is likely reflective of the nature of dairy production, which requires additional control measures to maintain herd health and achieve safe milk production. Future promotion of the use of GPp in Santa Fe should primarily target cow-calf producers and their veterinarians, who might be less familiar with GPp than those in the dairy industry. The veterinarian with the most producer responses (n = 26) had very high predicted probabilities, and it is possible that this veterinarian's responses were biased towards a high reported use of GPP. However, we kept this veterinarian in the model because a sensitivity analysis showed that the removal of his responses did not unduly affect the regression coefficients (i.e. changes of < 20%).

Veterinarians contributed to > 40% of the variation in the ordinal regression model outcome, which provides additional support that veterinarians in this study appear to serve an important role in producers' use of GPP. These results highlight the need for veterinarians in Santa Fe to expand upon their traditional role in animal health and production and be key mediators and educators about food safety knowledge among producers. Part of the veterinarian's role should be to disseminate relevant information to producers and guide them to improve their knowledge of food safety and use of GPP. Additional training and resources, including financial incentives, will be required to engage veterinarians effectively in these activities (Gunn et al 2008). The remaining model variation was due to producer, farm and herd level factors, and future research is necessary to understand how these factors contribute to producers' use of GPP.

One of the limitations of this study is that the questionnaire was not pre-tested on the target population of cattle producers. However, we developed, revised and pre-tested the questionnaire with local veterinarians, and we believe that their input and feedback helped to ensure that the questionnaire was user-friendly and appropriately designed for producers. Another limitation is that producers could have over-reported their use of GPp to provide a more socially-desirable response (e.g. a higher frequency of using GPP). However, it is also possible that the veterinarians' presence could have deterred producers from over-stating their responses. Producers also might not have had any motive to exaggerate their responses given that the questionnaire was anonymous. Although some specificity might have been lost in using an index variable as the outcome in the ordinal regression model, we believe that the results provide a useful overall summary of factors associated with respondents' use of GPP. However, it should be noted that some important confounding factors might have been absent from the model, such as financial (e.g. income) and personnel (e.g. number of farm employees) variables, which were not measured in the current questionnaire due to logistical and sensitivity reasons, but should be considered in future surveys.

It can be concluded that this pilot study identified the knowledge and attitudes towards food safety and reported use of GPp among a sample of cattle producers in Santa Fe, Argentina. Future research about food safety and GPp among producers in Santa Fe should be prioritized based on these results. Veterinarians should be engaged as key educators and promoters of food safety and GPp among cattle producers in this region. Veterinarians and producers should work together to develop on-farm food safety programs for beef and dairy cattle production in Santa Fe to improve food safety for these commodities.

ACKNOWLEDGMENTS

We thank the CMVSF for supporting this study and managing the questionnaire distribution and data entry. We thank the veterinarians and producers that distributed and completed questionnaires, attended the knowledge-exchange workshops and provided feedback on the questionnaire. We thank the Public Health Agency of Canada, the Representation of the Pan-American Health Organization in Argentina (Dr. Celso Rodriguez), and the Health Canada and Pan-American Health Organization's Canadian Biennial Work Plan Fund for providing financial support for this study.

 

REFERENCES

Brandt AW, MW Sanderson, BD DeGroot, DU Thomson, LC Hollis. 2008. Biocontainment, biosecurity, and security practices in beef feedyards. J Am Vet Med Assoc 232, 262-269.         [ Links ]

Dohoo I, W Martin, H Stryhn. 2003. Veterinary epidemiologic research. Atlantic Veterinary College, Charlottetown, PEI, Canada.         [ Links ]

Ellis-Iversen J, RP Smith, JC Gibbens, CE Sharpe, M Dominguez, AJC Cook. 2011. Risk factors for transmission of foot-and-mouth disease during an outbreak in southern England in 2007. Vet Rec 168, 128.         [ Links ]

Etcheverria AI, NL Padola, ME Sanz, R Polifroni, A Kruger, J Passucci, EM Rodríguez, AL Taraborelli, M Ballerio, AE Parma. 2010. Occurrence of shiga toxin-producing E. coli (STEC) on carcasses and retail beef cuts in the marketing chain of beef in Argentina. Meat Sci 86, 418-421.         [ Links ]

Fernandez D, EM Rodríguez, GH Arroyo, NL Padola, AE Parma. 2009. Seasonal variation of shiga toxin-encoding genes (stx) and detection of E. coli O157 in dairy cattle from Argentina. J Appl Microbiol 106, 1260-1267.         [ Links ]

Fossler CP, SJ Wells, JB Kaneene, PL Ruegg, LD Warnick, LE Eberly, SM Godden, LW Halbert, AM Campbell, CA Bolin, AM Zwald. 2005. Cattle and environmental sample-level factors associated with the presence of Salmonella in a multi-state study of conventional and organic dairy farms. Prev Vet Med 67, 39-53.         [ Links ]

Gunn GJ, C Heffernan, M Hall, A McLeod, M Hovi. 2008. Measuring and comparing constraints to improved biosecurity amongst GB farmers, veterinarians and the auxiliary industries. Prev Vet Med 84, 310-323.         [ Links ]

Hoe FG, PL Ruegg. 2006. Opinions and practices of Wisconsin dairy producers about biosecurity and animal well-being. J Dairy Sci 89, 2297-2308.         [ Links ]

Instituto Nacional de Tecnología Agropecuaria. 2010. Análisis de la actividad ganadera bovina de carne por estratos de productores y composición del stock: Años 2008 y 2009. Observatorio Estratégico.         [ Links ]

Kosek M, C Bern, RL Guerrant. 2003. The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bull World Health Organ 81, 197-204.         [ Links ]

Meichtri L, E Miliwebsky, A Gioffre, I Chinen, A Baschkier, G Chillemi, BE Guth, MO Masana, A Cataldi, HR Rodríguez, M Rivas. 2004. Shiga toxin-producing Escherichia coli in healthy young beef steers from Argentina: Prevalence and virulence properties. Int J Food Microbiol 96, 189-198.         [ Links ]

Moore DA, M Payne. 2007. An evaluation of dairy producer emergency preparedness and farm security education. J Dairy Sci 90, 2052-2057.         [ Links ]

OIE Animal Production Food Safety Working Group. 2006. Guide to good farming practices for animal production food safety. Rev Sci Tech 25, 823-836.         [ Links ]

Oliver SP, BM Jayarao, RA Almeida. 2005. Foodborne pathogens in milk and the dairy farm environment: Food safety and public health implications. Foodborne Pathog Dis 2, 115-129.         [ Links ]

Rabe-Hesketh S, A Skrondal. 2008. Multilevel and longitudinal modeling using Stata. 2nd ed. College Station, Stata Press, Texas, USA.         [ Links ]

Raymond MJ, RD Wohrle, DR Call. 2006. Assessment and promotion of judicious antibiotic use on dairy farms in Washington state. J Dairy Sci 89, 3228-3240.         [ Links ]

Rivas M, E Miliwebsky, I Chinen, CD Roldan, L Balbi, B Garcia, G Fiorilli, S Sosa-Estani, J Kincaid, J Rangel, PM Griffin. 2006. Characterization and epidemiologic subtyping of shiga toxin-producing Escherichia coli strains isolated from haemolytic uremic syndrome and diarrhea cases in Argentina. Foodborne Pathog Dis 3, 88-96.         [ Links ]

Rivas M, S Sosa-Estani, J Rangel, MG Caletti, P Valles, CD Roldan, L Balbi, MC Marsano de Mollar, D Amoedo, E Miliwebsky, I Chinen, RM Hoekstra, P Mead, PM Griffin. 2008. Risk factors for sporadic shiga toxin-producing Escherichia coli infections in children, Argentina. Emerg Infect Dis 14, 763-771.         [ Links ]

Samartino LE. 2002. Brucellosis in Argentina. Vet Microbiol 90, 71-80.         [ Links ]

Sargeant JM, MR Amezcua, A Rajić, L Waddell. 2007. Pre-harvest interventions to reduce the shedding of E. coli O157 in the faeces of weaned domestic ruminants: A systematic review. Zoonoses Public Health 54, 260-277.         [ Links ]

Scallan E, RM Hoekstra, FJ Angulo, RV Tauxe, MA Widdowson, SL Roy, JL Jones, PM Griffin. 2011. Foodborne illness acquired in the United States-major pathogens. Emerg Infect Dis 17, 7-15.         [ Links ]

Schouten JM, M Bouwknegt, AW van de Giessen, K Frankena, MC De Jong, EA Graat. 2004. Prevalence estimation and risk factors for Escherichia coli O157 on Dutch dairy farms. Prev Vet Med 64, 49-61.         [ Links ]

Tanaro JD, GA Leotta, LH Lound, L Galli, MC Piaggio, CC Carbonari, S Araujo, M Rivas. 2010. Escherichia coli O157 in bovine feces and surface water streams in a beef cattle farm of Argentina. Foodborne Pathog Dis 7, 475-477.         [ Links ]

Taverna M. 2010. Documento base: Programa nacional leches. Instituto Nacional de Tecnología Agropecuaria, Rafaela, Argentina.         [ Links ]

Thomas MK, E Perez, SE Majowicz, R Reid-Smith, S Albil, M Monteverde, SA McEwen. 2010. Burden of acute gastrointestinal illness in Galvez, Argentina, 2007. J Health Popul Nutr 28, 149-158.         [ Links ]

van Schaik G, YH Schukken, M Nielen, AA Dijkhuizen, HW Barkema, G Benedictus. 2002. Probability of and risk factors for introduction of infectious diseases into Dutch SPF dairy farms: A cohort study. Prev Vet Med 54, 279-289.         [ Links ]

Vanbaale MJ, JC Galland, DR Hyatt, GA Milliken. 2003. A survey of dairy producer practices and attitudes pertaining to dairy market beef food safety. Food Prot Trends 23, 466-473.         [ Links ]

Vilar MJ, E Yus, ML Sanjuán, FJ Diéguez, JL Rodríguez-Otero. 2007. Prevalence of and risk factors for Listeria species on dairy farms. J Dairy Sci 90, 5083-5088.         [ Links ]

Young I. 2010. A mixed-methods approach to evaluate producer knowledge, attitudes and practices towards food safety. PhD Dissertation, University of Guelph, Guelph, Canada.         [ Links ]

Young I, S Hendrick, S Parker, A Rajić, JT McClure, J Sanchez, SA McEwen. 2010. Knowledge and attitudes towards food safety among Canadian dairy producers. Prev Vet Med 94, 65-76.         [ Links ]

Young I, A Rajić, S Hendrick, S Parker, J Sanchez, JT McClure, SA McEwen. 2010. Attitudes towards the Canadian quality milk program and use of good production practices among Canadian dairy producers. Prev Vet Med 94, 43-53.         [ Links ]

 


Accepted: 17.05.2012

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons