Abstract
Porcine reproductive and respiratory syndrome (PRRS) is one of the most economically important diseases affecting the swine industry. The main objective of this study was to assess whether sow farm distance to slaughterhouses and meteorological variables were associated with PRRS outbreaks. This case-control study paired 104 sow farms with or without a reported PRRS outbreak (N = 208) during the same period. Data on the distance to the closest slaughterhouse, swine density, presence of an air filtration system, or a neighboring farm, and weather conditions were collected, and a multivariable conditional logistic regression model was created to investigate the association between variables of interest and the occurrence of a PRRS outbreak.
Swine density, presence of an air filtration system, presence of a neighboring farm, and PRRS herd-level status before the outbreak were associated with the occurrence of PRRS outbreaks. Farms in areas with higher swine density and nearby swine farms had increased odds of reporting an outbreak compared to farms in low swine density areas and farms having no neighbors. Under the conditions of this study, none of the meteorological variables or the distance to the closest slaughterhouse were associated with occurrence of PRRS outbreaks.
Résumé
Enquête sur la distance aux abattoirs et les paramètres météorologiques dans l’apparition d’épidémies de syndrome reproducteur et respiratoire porcin dans les troupeaux reproducteurs de porcs aux États-Unis. Le syndrome reproducteur et respiratoire porcin (SRRP) est l’une des maladies les plus importantes sur le plan économique affectant l’industrie porcine. L’objectif principal de cette étude était d’évaluer si la distance entre les élevages de truies et les abattoirs et les variables météorologiques étaient associées aux épidémies de SRRP. Cette étude cas-témoin a apparié 104 élevages de truies avec ou sans éclosion de SRRP déclarée (N = 208) au cours de la même période. Des données sur la distance à l’abattoir le plus proche, la densité porcine, la présence d’un système de filtration d’air ou d’une ferme voisine et les conditions météorologiques ont été recueillies, et un modèle de régression logistique conditionnelle multivariable a été créé pour étudier l’association entre les variables d’intérêt et l’occurrence d’une épidémie de SRRP.
La densité porcine, la présence d’un système de filtration de l’air, la présence d’une ferme voisine et le statut du troupeau relativement au SRRP avant l’épidémie ont été associés à l’apparition d’épidémies de SRRP. Les fermes situées dans des zones à forte densité porcine et d’autres fermes porcines à proximité avaient plus de chances de signaler une épidémie que les fermes situées dans des zones à faible densité porcine et les fermes sans voisins. Dans les conditions de cette étude, aucune des variables météorologiques ni la distance à l’abattoir le plus proche n’étaient associées à la survenue d’épidémies de SRRP.
(Traduit par Dr Serge Messier)
Introduction
Porcine reproductive and respiratory syndrome (PRRS) is an extremely damaging disease for swine producers across the United States (U.S.), costing an estimated $664 million to the industry annually due to losses in breeding and growing pig herds (1). Consequently, this disease has the highest economic impact on the U.S. swine industry, with the virus continuing to infect approximately 40% of sow herds in the U.S. annually (2). To date, the relative importance of factors that may (e.g., biosecurity practices) or may not (e.g., barn location) be controlled by producers, regarding PRRS prevention, is not clearly understood.
An example of a potential risk factor that is not fully understood is the role of meteorological conditions in the risk of PRRS outbreaks. Laboratory studies have documented stability of the PRRS virus (PRRSV) under various conditions, with temperature being most often investigated (3–7). In an experiment that tested the virus in cell culture using virus titration (TCID50) at 4, 10, 20, and 30°C, half-life was longest (155.5 h) at 4°C (5). In another study, the virus had the longest half-life in low temperatures and low humidity, being most stable at 5.0°C and 17.1% relative humidity (4).
Field studies investigating meteorological conditions in PRRS occurrence are rare. In one study, lower wind velocities paired with the presence of wind gusts increased the likelihood of PRRSV detection by RT-PCR, whereas precipitation, sunlight, and temperature did not have a role (8). A follow-up study included experimental infection of a pig population (~300 pigs) with 3 PRRSV variants and Mycoplasma hyopneumoniae to investigate airborne PRRSV detection at extended distances (9). Selected PRRSV variants included PRRSV 1-8-4, previously reported to travel in air up to 4.7 km (8); and PRRS 1-18-2 and PRRSV 1-26-2, which were new highly virulent variants at the time. In that study, a total of 31 sampling points were selected over an area of 166 km2, and from 114 air samples; PRRSV 1-8-4 was detected out to 9.1 km, whereas the other PRRSV variants remained undetected outside the source location. No vaccine variants were examined in this study. In addition, the authors determined that a unit increase in barometric pressure, minimum temperature, or maximum wind gust increased the odds of detecting aerosol PRRSV at an expanded distance by 46, 80, and 200%, respectively (9). These experiments provided valuable information but were conducted on single operations experimentally inoculated with PRRSV, which may not reflect real-world conditions. Under field conditions, there is evidence that PRRS outbreaks are more prevalent during the fall and winter (10), although this may not be the case in warmer climate zones (11). In addition, there is evidence that herd size and ongoing or previous disease outbreaks such as porcine epidemic diarrhea should be accounted for during the time of sampling (11). However, detailed assessment of the impact of specific weather conditions on the occurrence of PRRS outbreaks for multiple farms across the U.S. and over long intervals is lacking.
Another potential contributor to the occurrence of PRRS outbreaks that is anecdotally discussed, but that has not been formally investigated, is a farm’s proximity to highly commingling areas, e.g., slaughterhouses. A recent study highlighted the role of the cull sow market as a candidate for facilitating disease transmission throughout the U.S. swine industry (12). Facilities where a high degree of commingling occurs may be of concern for PRRSV transmission because they bring in animals from farms that may have diverse PRRSV statuses. The impact of such facilities’ proximity to a farm on the occurrence of PRRS outbreaks has apparently not been published.
The objectives of this study were to investigate if weather parameters and proximity to slaughterhouses were associated with PRRS outbreaks. Our main hypotheses were that lower mean temperatures are associated with higher odds of a farm reporting a PRRS outbreak, and that being close to a slaughterhouse increases the odds of farms reporting a PRRS outbreak.
Materials and methods
Study design and herd enrollment
A total of 208 farms were selected from the Morrison Swine Health Monitoring Project (MSHMP), a volunteer project that collects week-level PRRS incidence data from U.S. breeding herds representing approximately half of the U.S. sow population (13). This case-control study was designed using a primary-based open population with incidence density sampling.
First, a total of 104 sow farms that had reported a PRRS outbreak during any week from 2009–2016 (cases) were randomly selected from the MSHMP data set. However, PRRSV-specific characterization (e.g., strain/variant) was not available as part of this program and was not considered in the study. These 104 “case” farms were matched 1:1 with 104 sow farms from the same type of production system that had not reported an outbreak (controls) during that same week in which the outbreaks occurred. The controls were also randomly selected, but from a smaller pool, given they had to be from the same system as the case farm and had not reported an outbreak that same week. For example, if Farm 1 from Production System X reported a PRRS outbreak during the first week of January of 2010, it was matched with a randomly selected farm also from System X that had not reported a PRRS outbreak during the first week of January of 2010. A production system was defined as a group of swine farms that had a common owner or management structure, which could include animals, transportation, personnel, etc. As per our case and control definition, past PRRSV status of randomly selected farms were not considered during selection. This was deemed appropriate, considering past PRRSV status was one of the variables we were interested in investigating, and considering that farms are always at risk of a new PRRSV outbreak, considering the high diversity of this virus.
Weather data and proximity to slaughterhouses
Data on the distance to the closest slaughterhouse and weather conditions during the week prior to the PRRS outbreak were gathered and calculated in collaboration with the Center for Urban and Regional Analysis (14) at The Ohio State University. A period of 1 wk was deemed reasonable for investigation based on discussions with swine veterinarians and since there were no available data on most likely interval between virus introductions into a herd and reporting of an outbreak for PRRSV outbreak scenarios.
Weather conditions included temperature, wind, mean dew point, mean sea level pressure, mean visibility, average precipitation, and the presence of fog, rain/drizzle, rain/ice pellets, hail, thunder, and/or tornados/funnel clouds. Data were gathered from the closest weather stations to the farm through the Global Historical Climatology Network (15) from the National Oceanic & Atmospheric Administration’s National Centers for Environmental Information (16). These data points were averaged across the week prior to the outbreak date, since the exact dates of virus introduction into the herd were unknown.
Distances between case and control farms and the closest swine or multi-species slaughterhouse was calculated using Open Street Map data and the Open Source Routing Machine. Slaughterhouse locations came from the USDA-APHIS’s list of approved facilities (17). Other distance-related variables included distance to the main road and the presence of any farm within 1 and 3 km were gathered manually using Google Maps by the same person from the research team; as there is no standardized register of pig farms. The type of farm and whether the farm was actively in use was not assessed in this study. All personnel involved with data collection were blinded to farm status (case or control). In addition, variables including the presence of an air filtration system, swine density (pigs/km2) (18), and PRRS status the week before the outbreak (obtained from the MSHMP) were considered as potential confounders. PRRSV farm status definitions are defined (as per MSHMP guidelines) in Table 1.
Table 1.
Porcine reproductive and respiratory syndrome (PRRS) status classification used in this study.
| Status | Description |
|---|---|
| 1 | PRRS positive unstable, from outbreak through shedding. |
| 2 | PRRS positive stable, undergoing elimination procedure. Absence of clinical signs, no detectable viremia for a minimum of 90 d (test at least every 30 d). |
| 2fvia | PRRS positive stable, ongoing field virus exposure in gilts and/or sows. Absence of clinical signs, no detectable viremia for a minimum of 90 d (test at least every 30 d). |
| 2vxb | PRRS positive stable, ongoing exposure with live virus vaccine to gilts and/or sows. Absence of clinical signs, no detectable viremia for a minimum of 90 d (test at least every 30 d). |
| 3 | PRRS provisionally negative. Negative breeding replacements have been introduced and have remained seronegative by ELISA for at least 60 d. |
| 4 | PRRS ELISA-negative. Naïve herd confirmed seronegative by ELISA. If previously Category 3, then herds can be classified as Category 4, 1 y after Category 3 classification, upon confirmation of herd seronegativity by ELISA. |
The status 2fvi corresponds to an abbreviation of “field virus inoculation,” which characterizes the category. This abbreviation is used herein given its use by the source population (MSHMP program) and its widespread adoption in the field (13).
The status 2vx corresponds to an abbreviation of “vaccinated,” which characterizes the category. This abbreviation is used herein given its use by the source population (MSHMP program) and its widespread adoption in the field (19).
Statistical methods
Sample size was calculated based on the presence of a farm within 1 km of our study farms; since data related to distance to slaughterhouses and meteorological variables for PRRSV-positive versus PRRSV-negative farms under field conditions were lacking. We estimated a needed sample size of 104 per group (208 total) considering a conservative expected proportion of farms with another farm within 1 km (exposed) of 50%, assuming an odds ratio (OR) of 2.2, a confidence level of 0.95, and a desired power of 80% (20).
A multivariable conditional logistic regression model was created using STATA 13 (College Station, Texas, USA) to investigate the association between variables of interest and the occurrence of a PRRSV outbreak, while accounting for matched data. First, the assumption of linearity was checked for continuous variables and if this assumption was not met, variables were categorized in the median (e.g., swine density and herd size). Secondly, collinearity between variables was assessed using the Spearman correlation coefficient and a cut-off value of 0.80. Thirdly, univariable conditional logistic regression models were built, and all associations with a P-value < 0.20 were offered to the full model, in a backwards stepwise manner. Variables were deemed as confounders and kept in the final model if they changed the value of any model coefficient by > 20%, regardless of statistical significance. Statistical significance was declared at P < 0.05, whereas a statistical trend was declared at 0.05 ≤ P ≤ 0.10.
Results
Enrolled farms were located in 16 states in the U.S. (Figure 1), with most of the high swine-producing states in the country represented. Overall, 35% of the farms were in Iowa, followed by Minnesota (16.3%), North Carolina (13.0%), Oklahoma (10.6%), Nebraska (8.6%), and Illinois (5.8%), with those 6 states comprising 89.43% of the total study population. The percentage distribution among states was very similar for cases and controls. For cases, 36.5% of farms were in Iowa, followed by Minnesota (17.3%), North Carolina (13.46%), and Oklahoma (12.5%); and for controls, 33.6% of the farms were in Iowa, followed by Minnesota (15.4%), North Carolina (12.5%), and Oklahoma and Nebraska (both at 8.65%). Approximately 18% of the farms were enrolled during each of years 2012 and 2015, 14% for 2011, 12.5% for 2013, 11.5% for 2014, 10.6% for 2010 and 2016, and approximately 4% for 2009. The 208 farms belonged to 19 production systems, with each system contributing between 2 (1 case and 1 control) and 40 (20 cases and 20 controls) farms to the study population. As such, 1 system contributed to approximately 20% of the study population, another system contributed to approximately 12.5% of the study population, 4 systems contributed between 10 and 5% of the study population, and the other 13 systems contributed to < 5% of the study population.
Figure 1.
Map of the U.S. showing swine density (yellow shading; source: (21), slaughterhouses (
), case farms (
), and control farms (
). The map was created using ArcMap 10.7.1.
There were no significant differences regarding predictors of interest including weather-related or distance-related variables between case and control farms. Descriptive statistics for all examined meteorological and distance variables by outcome (case or control status) are shown in Tables 2 and 3. Variables including distance to the main road, Euclidian distance to the closest slaughterhouse, presence of rain or drizzle, and presence of fog had P-values < 0.20 in univariable models. In the univariable models, a 1000-foot (0.3-km) increase in the distance from the main road increased the odds of a farm being a case (OR = 1.32; P = 0.19) and having a high Euclidian distance (higher than the median value) to the closest slaughterhouse decreased the odds of farms being a case (OR = 0.50; P = 0.10). Furthermore, the presence of fog was associated with an increase in the odds of farms being cases (OR = 1.72; P = 0.15), and the presence of rain or drizzle was associated with a decrease in the odds of farms being cases (OR = 0.47; P = 0.10). All weather parameters and distance to slaughterhouse variables were not included in the final model as all P-values were > 0.20 in the univariable analysis.
Table 2.
Descriptive analysis for continuous variables for herds participating in the current study that aimed to investigate predictors for porcine reproductive and respiratory syndrome (PRRS) outbreaks in breeding herds.
| Cases | Controls | |||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|||||||
| Variable | Mean | SDa | Median | IQRb | Mean | SDa | Median | IQRb |
| Average inventory (number of sows) | 3421.8 | 1732.8 | 3197.0 | 1800.0 | 3149.9 | 1608.5 | 3000.0 | 1500.0 |
| Distance to main road (m) | 201.4 | 274.3 | 61.3 | 239.6 | 163.6 | 207.3 | 66.8 | 148.7 |
| Mean temperature (°C) | 5.8 | 11.6 | 5.3 | 17.2 | 5.9 | 11.4 | 5.3 | 16.9 |
| Mean Dew Point (°C) | 0.4 | 10.3 | 0.0 | 17.6 | 0.5 | 10.6 | −1.3 | 12.8 |
| Mean sea level pressure (millibars) | 1017.4 | 3.9 | 1017.8 | 4.1 | 1018.0 | 3.6 | 1017.8 | 5.3 |
| Mean visibility (miles) | 8.7 | 1.2 | 9.1 | 1.3 | 8.7 | 1.2 | 9.0 | 1.4 |
| Mean wind speed (knots) | 7.9 | 2.5 | 7.8 | 3.8 | 7.6 | 2.7 | 7.5 | 3.6 |
| Maximum sustained wind speed (knots) | 15.3 | 3.6 | 15.0 | 5.4 | 14.9 | 3.7 | 14.9 | 4.8 |
| Maximum wind gust (knots) | 23.2 | 4.1 | 22.6 | 3.0 | 22.8 | 4.3 | 22.4 | 6.0 |
| Maximum temperature (°C) | 12.1 | 12.6 | 11.2 | 22.5 | 12.1 | 12.2 | 11.0 | 19.9 |
| Minimum temperature (°C) | 0.4 | 11.1 | 0.8 | 16.1 | 0.6 | 10.9 | 0.2 | 14.7 |
| Total precipitation (mm) | 0.0 | 2.5 | 0.0 | 0.0 | 0.0 | 2.5 | 0.0 | 0.0 |
| Distance to the closest slaughterhouse, Euclidian (km) | 110.4 | 77.4 | 95.8 | 52.5 | 112.3 | 72.7 | 97.9 | 64.2 |
| Distance to the closest slaughterhouse, via roads (km) | 139.5 | 99.0 | 122.1 | 64.2 | 138.8 | 91.0 | 127.5 | 70.8 |
Standard deviation.
Inter-quartile range.
Table 3.
Descriptive analysis for dichotomous variables for herds participating in the current study that aimed to investigate predictors for porcine reproductive and respiratory syndrome (PRRS) outbreaks in breeding herds. All variables are shown as n (%).
| Cases | Controls | |||
|---|---|---|---|---|
|
|
|
|||
| Variable | Yes | No | Yes | No |
| Filtered | 10 (9.7) | 93 (90.3) | 19 (18.4) | 84 (81.6) |
| Main road is a highway | 3 (2.9) | 101 (97.1) | 2 (1.9) | 102 (98.1) |
| Presence of farm within 1 km | 71 (68.3) | 33 (31.7) | 54 (51.9) | 50 (48.1) |
| Presence of farm within 3 km | 91 (87.5) | 13 (12.5) | 88 (84.6) | 16 (15.4) |
| Fog | 47 (45.2) | 57 (54.8) | 39 (37.5) | 65 (62.5) |
| Rain or drizzle | 67 (64.4) | 37 (35.6) | 75 (72.1) | 29 (27.9) |
| Hail | 0 (0.0) | 104 (100.0) | 0 (0.0) | 104 (100.0) |
| Thunder | 13 (12.5) | 91 (87.5) | 14 (13.5) | 90 (86.5) |
| Tornado or funnel cloud | 0 (0.0) | 104 (100.0) | 0 (0.0) | 104 (100.0) |
The final model included swine density, presence of an air filtration system, presence of another farm within 1 km, and PRRS status 1 wk prior to the outbreak (Table 4). In the final model, farms located in regions with a high swine density (> 4800 pigs/km2) had increased odds (OR = 2.48, P = 0.034) of reporting a PRRS outbreak compared to those in regions with a lower swine density. In addition, being within 1 km of another farm increased the odds (OR = 3.72, P = 0.002) of farms reporting a PRRS outbreak. Conversely, having an air filtration system decreased the odds (OR = 0.099, P = 0.003) of reporting a PRRS outbreak. Finally, farms with PRRS statuses of ‘4,’ enzyme-linked immunosorbent assay-(ELISA)-negative (OR = 0.072, P = 0.037) were less likely to report a PRRS outbreak (Table 4).
Table 4.
Results from final multivariable conditional logistic regression model investigating predictors for porcine reproductive and respiratory syndrome virus (PRRSV) outbreaks in breeding herds participating in a voluntary PRRS monitoring project.
| Variable | Category | OR (SE)a | 95% CIb | P-value |
|---|---|---|---|---|
| Pig densityc | 2.48 (1.06) | (1.07, 5.74) | 0.0340 | |
| PRRSV status during previous weekd | 1 | Referent | ||
| 2 | 0.107 (0.146) | (0.00726, 1.57) | 0.102 | |
| 2fvie | 0.921 (1.19) | (0.0734, 11.56) | 0.949 | |
| 2vxf | 0.288 (0.367) | (0.0263, 3.51) | 0.329 | |
| 3 | 0.374 (0.511) | (0.258, 5.53) | 0.472 | |
| 4 | 0.0717 (0.0905) | (0.00605, 0.851) | 0.0370 | |
| Presence of farm within 1 km | No | Referent | ||
| Yes | 3.72 (1.61) | (1.60, 8.69) | 0.002 | |
| Filtered | No | Referent | ||
| Yes | 0.0992 (1.06) | (0.0210, 0.467) | 0.003 |
Odds ratio (standard error).
Confidence interval.
Categorized in the median (46 pigs/km2); reference value is < 4800 pigs/km2.
See Table 1 for details on the classification of PRRSV status during the previous week.
The status 2fvi corresponds to an abbreviation of “field virus inoculation,” which characterizes the category. This abbreviation is used herein given its use by the source population (MSHMP program) and its widespread adoption in the field (13).
The status 2vx corresponds to an abbreviation of “vaccinated,” which characterizes the category. This abbreviation is used herein given its use by the source population (MSHMP program) and its widespread adoption in the field (28).
Discussion
This study aimed to provide field-based information on whether proximity to a high commingling facility (e.g., slaughterhouses) and weather parameters were associated with occurrence of PRRS outbreaks in breeding herds. With this information available, producers will be able to better understand whether these risk factors, usually outside of their control, are important predictors of PRRS outbreaks. Using our study design, there was no evidence that weather conditions during the previous week before reporting an outbreak to MSHMP or close proximity to slaughterhouses, were risk factors for PRRSV occurrence.
The main strengths of our study are that we used a sample size of 208 herds from various areas in the USA representing all major swine-production states, from various production systems, and over a long interval. Altogether, this makes our results relatively generalizable to the U.S. sow herd population. However, our study also had limitations. First, weather data were averaged over the week prior to the reported PRRS outbreak, which could have hidden a single weather event that had a significant impact on the spread of PRRS. For example, a strong gust of wind could have been essentially erased by an entire week of calmer winds. Detailed weather data should be considered and examined in future studies. Furthermore, the authors entirely relied on timely reporting of PRRS outbreaks (or lack thereof ) for both case and control farms, which may lead to misclassification bias of the outcome of interest. However, if this happened, this would be classified as non-differential and likely lead results to be biased towards the null.
Another limitation was that determining farms within 1 and 3 km relied on accuracy from the Google Maps’ platform. Even though this platform has been suggested as a promising tool in other fields of research (20), satellite visualization did not allow for the differentiation of swine-only facilities nor whether the facility had active animal presence. This could cause misclassification bias but considering that there was only 1 individual taking these assessments that were standardized for all participating farms, and that this individual was blinded to farm status (case or control), this would be a case of non-differential bias which would bias the estimate towards the null. These images may not be completely up-to-date, and recently built farms may not have been captured. Another limitation is that smaller non-registered slaughterhouses and cull sow markets were not included in our assessment, due to lack of data availability on their location. This limitation should be considered in future studies and attempts should be made to incorporate these smaller facilities. Furthermore, PRRSV-specific variants were not considered in the study, as this was not a requirement for the volunteer study, and as such was not available to the authors. A final limitation is that the project participants were limited to sow herds participating in the MSHMP program, which could limit the generalizability of the study findings.
The main variables of focus for this project, meteorological data and distances to slaughterhouses, were not included in the final statistical model. This suggests that although these variables, specifically wind speeds, temperature, and humidity have been shown under laboratory and field conditions to alter survivability of PRRSV, they may not be as important in transmission as other factors, under real-world scenarios. It could also be the case that our sample size calculations for this study did not allow for detection of potential smaller effects than anticipated. Our study provides data to guide sample size calculations for future field studies. Moreover, regional seasonality and the specific climate zone of the farm location may also influence the importance of these factors in the field. For instance, seasonal PRRS detection are likely to differ between northern more temperate regions compared to sub-tropical or tropical environments and may also differ annually, as reported (10,11).
The important factors highlighted in our analysis included swine density, the presence of a farm within 1 km, the use of a filtration system, and the PRRS status the week before the outbreak date, specifically being ELISA-negative. The negative effects of being in a swine-dense region or having close neighbors seen in our study were in accordance with previous research (11,23,24). Our study, however, did not allow for the differentiation between potential fomite and aerosol transmission. A recent study (25) reported that swine density had the greatest effect on PRRSV risk for a given farm. Specifically, high swine-dense areas were most influenced by swine density, whereas low swine-dense areas were influenced more by climate and land coverage (25).
When examining the impact of a filtration system in the occurrence of PRRS outbreaks, we detected a protective effect, with filtered farms having reduced odds of reporting a PRRS outbreak compared to non-filtered farms. These findings were in agreement with controlled field studies that an air filtration system can be an effective biosecurity measure to prevent PRRS transmission (26–29). However, even though it would be reasonable that such intervention could reduce the occurrence of PRRS outbreaks due to the decrease of viral particles that may enter a swine facility, it is also possible that general biosecurity practices are increased for such herds. This is an important point that cannot be untangled by a retrospective study design; therefore, further studies are warranted.
Finally, herds that had a PRRS status of ‘4,’ i.e., ELISA-negative, could have either successfully recovered from a previous PRRS outbreak, or had been naïve for a long interval, which would indicate that these herds would have a lower chance of having an outbreak than herds at a farm that is PRRS unstable (‘1’).
In conclusion, none of the meteorological variables during the week before the outbreak or the distance to the closest slaughterhouse were associated with increased risk of PRRS outbreaks in our study. Future research conducted prospectively and capturing distance to more high commingling facilities, as well as short-lived weather events, would be helpful to better understand the role of environment versus other factors on PRRSV introduction in swine farms.
Acknowledgments
Funding, wholly or in part, was provided by the National Pork Checkoff and the U.S. Pork Center of Excellence and the Swine Health Information Center. The authors acknowledge participants of the Morrison Swine Health Monitoring Project (MSHMP), Swine Health Information Center-funded project, for data sharing. Preliminary results were presented by the primary author (Moeller) as an oral presentation at the 2019 Allen D. Leman Conference, St. Paul, MN, 14–17 September, 2019, and as a poster presentation at the 51st American Association of Swine Veterinarians Annual Meeting, Atlanta, Georgia, USA, March 7–10, 2020. CVJ
Footnotes
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
References
- 1.Holtkamp DJ. Assessment of the economic impact of porcine reproductive and respiratory syndrome virus on United States pork producers. J Swine Health Prod. 2013;21:72–84. [Google Scholar]
- 2.Tousignant SJP, Perez A, Morrison R. Comparison between the 2013–2014 and 2009–2012 annual porcine reproductive and respiratory syndrome virus epidemics in a cohort of sow herds in the United States. Can Vet J. 2015;56:1087–1089. [PMC free article] [PubMed] [Google Scholar]
- 3.Van Alstine WG, Kanitz CL, Stevenson GW. Time and temperature survivability of PRRS virus in serum and tissues. J Vet Diagn Invest. 1993;5:621–622. doi: 10.1177/104063879300500421. [DOI] [PubMed] [Google Scholar]
- 4.Hermann J, Hoff S, Muñoz-Zanzi C, et al. Effect of temperature and relative humidity on the stability of infectious porcine reproductive and respiratory syndrome virus in aerosols. Vet Res. 2007;38:81–93. doi: 10.1051/vetres:2006044. [DOI] [PubMed] [Google Scholar]
- 5.Jacobs AC, Hermann JR, Muñoz-Zanzi C, et al. Stability of porcine reproductive and respiratory syndrome virus at ambient temperatures. J Vet Diagn Invest. 2010;22:257–260. doi: 10.1177/104063871002200216. [DOI] [PubMed] [Google Scholar]
- 6.Linhares DC, Torremorell M, Joo HS, Morrison RB. Infectivity of PRRS virus in pig manure at different temperatures. Vet Microbiol. 2012;160:23–28. doi: 10.1016/j.vetmic.2012.05.009. [DOI] [PubMed] [Google Scholar]
- 7.Pileri E, Mateu E. Review on the transmission porcine reproductive and respiratory syndrome virus between pigs and farms and impact on vaccination. Vet Res. 2016;47:108. doi: 10.1186/s13567-016-0391-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dee S, Otake S, Oliveira S, Deen J. Evidence of long distance airborne transport of porcine reproductive and respiratory syndrome virus and Mycoplasma hyopneumoniae. Vet Res. 2009;40:1–13. doi: 10.1051/vetres/2009022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Otake S, Dee S, Corzo C, Oliveira S, Deen J. Long-distance airborne transport of infectious PRRSV and Mycoplasma hyopneumoniae from a swine population infected with multiple viral variants. Vet Microbiol. 2010;145:198–208. doi: 10.1016/j.vetmic.2010.03.028. [DOI] [PubMed] [Google Scholar]
- 10.Tousignant SJP, Perez AM, Lowe JF, Yeske PE, Morrison RB. Temporal and spatial dynamics of porcine reproductive and respiratory syndrome virus infection in the United States. Am J Vet Res. 2015;1:70–76. doi: 10.2460/ajvr.76.1.70. [DOI] [PubMed] [Google Scholar]
- 11.Sanhueza JM, Stevenson MA, Vilalta C, Kikuti M, Corzo CA. Spatial relative risk and factors associated with porcine reproductive and respiratory syndrome outbreaks in United States breeding herds. Prev Vet Med. 2020;183:105128. doi: 10.1016/j.prevetmed.2020.105128. [DOI] [PubMed] [Google Scholar]
- 12.Blair B, Lowe J. Describing the cull sow market network in the US: A pilot project. Prev Vet Med. 2019;162:107–109. doi: 10.1016/j.prevetmed.2018.11.005. [DOI] [PubMed] [Google Scholar]
- 13.MSHMP History. University of Minnesota; [Last accessed March 14, 2022]. [homepage on the Internet]. Available from: https://vetmed.umn.edu/centers-programs/swine-program/outreach-leman-mshmp/mshmp-history. [Google Scholar]
- 14.Center for Urban and Regional Analysis. [Last accessed March 14, 2022]. [homepage on the Internet]. Available from: https://cura.osu.edu/about/unit-report.
- 15.Global Historical Climatology Network (GHCN) [Last accessed March 14, 2022]. [homepage on the Internet]. Available from: www.ncdc.noaa.gov/data-access/land-based-station-data/land-based-datasets/global-historical-climatology-network-ghcn.
- 16.National Centers for Environmental Information (NCEI) [Last accessed March 14, 2022]. [homepage on the Internet]. Available from: www.ncdc.noaa.gov/about.
- 17.List of Plants Approved to Receive Immediate Slaughter Animals. [Last accessed March 14, 2022]. [homepage on the Internet]. Available from: www.aphis.usda.gov/import_export/downloads/slaughter_list.pdf.
- 18.Food and Animal Organization of the United Nations. Global pigs distribution in 2010 (5 minutes of arc). [homepage on the Internet] [Last accessed March 14, 2022]. Available from: https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/33N0JG.
- 19.MSHMP PRRS Prevalence Chart Description [homepage on the Internet] [Last accessed March 14, 2022]. Available from: https://umnswinenews.com/2019/05/24/mshmp-prrs-prevalence-chart-description/
- 20.Epitools Epidemiological Calculators. [homepage on the Internet] [Last accessed March 14, 2022]. Available from: https://epitools.ausvet.com.au/
- 21.Gilbert M, Nicolas G, Cinardi G, et al. Global pigs distribution in 2010 (5 minutes of arc), Harvard Dataverse, V3. 2018 [Google Scholar]
- 22.Vandeviver C. Applying Google maps and Google street view in criminological research. Crime Sci. 2014;3:1–16. [Google Scholar]
- 23.Lambert M-È, Arsenault J, Poljak Z, D’Allaire S. Epidemiological investigations in regard to porcine reproductive and respiratory syndrome (PRRS) in Quebec, Canada. Part 2: Prevalence and risk factors in breeding sites. Prev Vet Med. 2012;104:84–93. doi: 10.1016/j.prevetmed.2011.11.002. [DOI] [PubMed] [Google Scholar]
- 24.Brito B, Dee S, Wayne S, Alvarez J, Perez A. Genetic diversity of PRRS virus collected from air samples in four different regions of concentrated swine production during a high incidence season. Viruses. 2014;6:4424–4436. doi: 10.3390/v6114424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Alkhamis MA, Arruda AG, Vilalta C, Morrison RB, Perez AM. Surveillance of porcine reproductive and respiratory syndrome virus in the United States using risk mapping and species distribution modeling. Prev Vet Med. 2018;150:135–142. doi: 10.1016/j.prevetmed.2017.11.011. [DOI] [PubMed] [Google Scholar]
- 26.Pitkin A, Deen J, Dee S. Use of a production region model to assess the airborne spread of porcine reproductive and respiratory syndrome virus. Vet Microbiol. 2009;136:1–7. doi: 10.1016/j.vetmic.2008.10.013. [DOI] [PubMed] [Google Scholar]
- 27.Dee S, Otake S, Deen J. Use of a production region model to assess the efficacy of various air filtration systems for preventing airborne transmission of porcine reproductive and respiratory syndrome virus and Mycoplasma hyopneumoniae: Results from a 2-year study. Virus Res. 2010;154:177–184. doi: 10.1016/j.virusres.2010.07.022. [DOI] [PubMed] [Google Scholar]
- 28.Alonso C, Davies PR, Polson DD, Dee SA, Lazarus WF. Financial implications of installing air filtration systems to prevent PRRSV infection in large sow herds. Prev Vet Med. 2013a;111:268–277. doi: 10.1016/j.prevetmed.2013.05.001. [DOI] [PubMed] [Google Scholar]
- 29.Alonso C, Murtaugh MP, Dee SA, Davies PR. Epidemiological study of air filtration systems for preventing PRRSV infection in large sow herds. Prev Vet Med. 2013b;112:109–117. doi: 10.1016/j.prevetmed.2013.06.001. [DOI] [PubMed] [Google Scholar]