Investigation of biosecurity risks associated with the feed delivery: A pilot study

Abstract

This study explored potential biosecurity issues related to the delivery of feed to commercial farms. A pilot study was conducted to collect information about the day-to-day feed delivery, including biosecurity concerns at the level of the feed truck, the driver, and the farm. In addition, a reusable rubber boot was tested in an effort to increase the proportion of farms at which truck drivers wore clean footwear, and to explore an alternative to the standard plastic disposable boots that may be unsafe in winter conditions. Most farms did well in terms of proper dead-stock management and keeping the farm lane and feed bin areas clean. The provision of reusable rubber boots significantly increased the proportion of deliveries in which the driver wore clean footwear.

Introduction

Biosecurity is becoming increasingly important in the swine industry (1). Protocols related specifically to the delivery of feed are of interest because movement of trucks and drivers among farms increases the potential for the spread of pathogens. There are no mandatory biosecurity measures for feed delivery to livestock farms once the feed has left the mill. However, there are industry recommendations that enhance biosecurity during pig transportation and several of these recommendations can be applied to feed delivery. The Transport Quality Assurance Handbook stresses that everyone involved in pig production must have a biosecurity plan in order to reduce the risk of introducing and spreading disease; including movement of pigs, people, vehicles, and equipment (2). Ideally, there will be a line of separation between the area used by the feed-truck driver and that used by farm personnel (2). Protective clothing including boots, coveralls, and gloves should be used. Organic material can be carried on a variety of fomites such as boots, clothing, tires, and the undercarriage of vehicles and must be removed prior to disinfection (2). The feed truck should be cleaned, washed, disinfected, and dried frequently; drying is critical to killing viruses and bacteria (2,3).

Contaminated fomites such as boots, hands, and vehicles may be involved in transmission of pathogens, including Salmonella (4–6), porcine reproductive and respiratory syndrome virus (PRRSV) (7–11), transmissible gastroenteritis virus (TGEV) (12,13), Brachyspira hyodysenteriae (14), and Lawsonia intracellularis (15). Basic sanitation protocols, including changing clothing and footwear, and washing hands or showering, can limit introduction of some pathogens, including PRRSV (8,10), TGEV (12), and Escherichia coli (16). Salmonella has been isolated from feed trucks and feed samples (4,17). A PRRSV-positive herd status is associated with failure to clean trucks between herd visits (18).

This study summarizes information about biosecurity risks related to feed delivery to farms in Ontario collected by truck drivers from 3 feed companies (19). The objectives were to determine the prevalence of farm- and truck-level biosecurity risks during feed delivery, and to test the implementation of reusable and washable overshoes and gloves for feed-truck drivers. The functional feasibility and the cost of using the boots were also described.

Materials and methods

This study received approval from the University of Guelph Research Ethics Board (REB #12SE014). Forty feed-truck drivers from 3 Ontario companies participated for 6-week during January through March, 2013. The feed companies were conveniently selected based on the fact that they were large companies serving the swine industry and their managers volunteered to participate. The focus of the study was on feed delivered to swine farms but data were collected on other types of farms when a driver delivered to a swine farm and another type of farm on 1 day.

Phase 1 of the study, not included in this manuscript, used participatory epidemiology to gather information (19). Focus groups and key-informant interviews were held with 21 feed-company personnel from 8 companies and 15 swine producers. Participants outlined the current protocols that affect the risk of pathogen spread through feed delivery and identified management changes at farm, truck, and feed-company levels that might further reduce that risk. Participants were encouraged to list all possible management changes that could potentially reduce the risk of disease spread. The list of over 80 recommendations is beyond the scope of this manuscript and is detailed in Dewey et al (19). Examples of recommendations for the feed companies were to enhance pest control, maintain a list of biosecurity and disease information for each farm, have 1 truck used only for high-health farms, provide dedicated feed truck wash bays, and avoid the return of skids and extra feed. Dispatchers for feed trucks were encouraged to sequence delivery of feed to high health and high biosecurity farms before lower health and lower biosecurity farms.

Examples of recommendations for drivers included regularly washing and disinfecting the truck’s floor mats, steering wheel, and exterior of the truck and washing the truck between farm visits. Other suggestions were following each farm’s biosecurity protocol; wearing cleaned and disinfected boots in the yard; not entering the barn unless absolutely necessary, and when necessary, wearing barn-specific coveralls and boots. Farm management recommendations included building a new driveway dedicated to the delivery of feed, maintaining a clean laneway free of manure, locating feed bins away from exhaust fans, manure pits and compost piles, providing farm-specific blow pipes and also providing a container near the feed bins for the mill order. Other farm recommendations were ensuring dead-stock are properly contained to reduce runoff, reducing the number of feed deliveries, providing clean boots and coveralls for feed-company personnel entering the barn or, better still, providing a location for bagged feed other than the barn.

After the focus groups and interviews were complete, participants rated the recommendations according to economic feasibility, logistic feasibility, and effectiveness for disease control. Highly rated factors that could be implemented or measured over a 6-week period were considered for inclusion in this pilot study.

The pilot study is described herein. The research team, with input from feed-company managers, designed log sheets to be filled out by feed-truck drivers during farm feed deliveries. Managers provided a list of drivers, and each driver was provided with a letter of information and voluntarily decided whether they would participate. Of the 45 drivers contacted, 40 agreed to participate, signed a consent form, and were compensated with $10 gift cards each week.

Log sheets included information about the date and the order of farms visited, type of truck used and when it was last washed; feed type being delivered (bags and/or bulk); farm type; cleanliness of the farm lane and feed-bin area; dead-stock management; location of bagged feed delivery; where the mill order was left; whether the driver used clean boots and/or gloves, or entered the barn. Drivers were asked if they entered the barn and if so, what part of the barn (office, feed room, where there were or were not animals). If they delivered bagged feed they recorded if the feed was deposited outside a building, in the barn office, feed room in the barn, or a building that was not the barn such as a shed, garage, or out building. Thirdly, they were asked what type of boots they wore in the yard and when they entered a building.

Feed-truck drivers in each company were randomly assigned to a treatment or control group. The control-group drivers (n = 18) were asked to only fill out log sheets at each farm. They were not provided with gloves or new types of boots and were asked to follow their normal routine. Drivers in the treatment group (n = 22) filled out log sheets and were asked not to enter the barn unless absolutely necessary. Although all 3 companies provided disposable, plastic boots to all drivers, 1 aspect of this study was to investigate the use of other boots that were considered less slippery in winter conditions. Three footwear options for drivers in the treatment group were sourced: i) a disposable plastic boot with a treaded bottom (ITSI Select TREADER; Insemination Technics & Supplies International, Princeton, Ontario; cost $0.91/pair); ii) a reusable overshoe made of stretchy rubber (Rubber Overshoes; Country Boy Equipment and Supplies, Mildmay, Ontario; cost $6.50/pair), and iii) a reusable overshoe made of sturdier rubber (Black Overshoe; ONGUARD Industries, Havre de Grace, Maryland, USA; cost $15.75/pair). Each feed company was provided with 1 of the 3 footwear options, and drivers in the treatment group were asked to wear a clean pair of the boots at each farm unless they considered it a safety risk due to ice and snow. Additionally, treatment-group drivers at one company were supplied with cotton gloves (Country Boy Equipment and Supplies; cost $0.60/pair). One feed company with its own laundering facilities hired a staff member to wash the reusable boots, and 1 company used a commercial laundering business that charged $1.75 to wash and dry each pair of reusable boots. The third company used disposable plastic boots with a treaded bottom.

Completed log sheets were mailed in self-addressed pre-stamped envelopes to the researchers, so drivers’ responses were confidential. After 6-week, all drivers were invited to complete an exit survey, which asked i) if there was anything they started doing during the study that they would continue to do; ii) if they had suggestions about how to change the log sheets; iii) what they found most difficult about participating; and iv) whether the drivers in the treatment group would continue to use the boots provided.

Statistical analysis

The farm visits were described by type of farm, animal species, types of feed delivered, and by biosecurity factors listed on the log sheets. These factors included cleanliness of the farm lane, farmyard, and feed bin area and where the bagged feed and mill order were left. The outcomes of interest for the randomized trial were the proportion of farm visits during which drivers wore clean boots and clean gloves in the treatment and control groups. The association between wearing clean boots in the yard and selected factors was determined using univariate mixed logistic regression. The outcome of interest was whether the driver wore clean boots in the yard, and the selected factors included the following binary variables: treatment versus control group, delivery of bagged feed, driver entered the barn, driver wore clean boots if they entered a building. Driver was included as a random effect, and the intra-class correlation coefficient (ICC) was calculated for each univariate analysis (Stata/IC 11.2; StataCorp, College Station, Texas, USA). For all statistical analyses, “Rubber” and “Disposable” boots were considered “Clean” while the driver’s own boots were considered “Unclean.” Farm boots worn only inside a building were also considered in the “Clean” category for entry to a building. For gloves, drivers indicated either clean or disposable for the “Clean” category and unclean or none for the “Unclean” category.

Multiple correspondence analysis (MCA) (SPSS 18.0; SPSS, Chicago, Illinois, USA) is a multivariate technique used to visualize relationships within a set of categorical variables (20). It indicates the total variance in the dataset, and how much influence is exerted by each variable. It is used to analyze survey data — each variable corresponds to a survey question, and categories of each variable correspond to the possible responses (21). While there is no outcome variable; the procedure identifies patterns among responses to variables. Multiple correspondence analysis was used to investigate relationships between farm-level variables. Relationships between different categories of the variables are represented as points in a two-dimensional space. Farm-level variables that often occur together are plotted closely together, and farm-level variables that rarely occur together are plotted further apart. The plot illustrates which biosecurity measures typically occur together on individual farms. Additionally, variable categories located close to the axis of the plot represent the most common responses. Discrimination measures provide insight into the influence exerted by each variable (22).

Variables included in the MCA solution were: feed type delivered (bulk/bags/both); barn entry (didn’t enter/entered the barn/entered the office/entered an area with animals); farm lane clean (yes/no), feed bin area clean (yes/no/not applicable); and dead-stock management adequate (yes/no). Farm type (swine/poultry/ruminant/mixed) was included as a supplementary variable, which was not used in the solution but aids in interpretation of the result.

Results

Descriptive information

Descriptive information from 2202 farm visits made by 40 Ontario feed-truck drivers is in Table 1. On average, drivers submitted information about deliveries to 3 farms per day (standard deviation = 1.59), with a minimum of 1 and a maximum of 10. Over half of the feed deliveries were to swine farms (58.7%), with the remaining visits being to poultry (26.0%), ruminant (13.9%), or mixed (1.4%) farms. The ruminant category included 301, 3, and 2 visits to cattle, sheep, and goat farms, respectively. The mixed category included 18, 8, and 2 visits to swine and cattle, poultry and cattle, swine and poultry farms, respectively, 1 to swine and another unspecified species, and 1 to swine and horse farms. One farm had only horses. All farms are included in the results.

Table 1.

Farm and feed truck data describing 2202 Ontario farm visits by truck drivers from 3 feed companies, over a 6-week period in the winter of 2013

Variable	Categories	Overall (%) n = 2202	Swine (%) n = 1291	Poultry (%) n = 570	Ruminant (%) n = 306	Mixed (%) n = 35
Biosecuritya	Driver needs permission to enter the barn	44.1	52.3	31.6	33.3	35.5
	Biosecurity sign on gate	21.4	14.7	46.1	4.2	9.7
	Biosecurity sign on barn	48.4	52.5	45.4	38.2	32.3
Farm laneb	Clean	82.4	83.1	84.8	75.5	80.7
	Mud	11.1	10.2	8.8	19.0	12.9
	Puddles	9.2	8.3	6.9	17.0	6.5
	Manure	1.9	0.7	2.1	6.5	0
	Dead-stock	0.9	0.8	1.2	0.7	0
	Snowc	8.6	7.9	9.2	8.8	16.1
Feed type delivered	Bulk only	84.3	83.9	98.9	58.5	83.9
	Bags only	6.1	4	0.2	26.1	3.2
	Bulk & bags	9.6	12.1	0.9	15.4	12.9
Feed bin areab	Cleand	80.8	78.6	90.8	68.8	75.9
	Spilled feedd	15.9	18.2	5.7	27.7	24.1
	Mudd	5.1	5	2.3	13.0	0
	Snowc,d	2.1	2.4	2.3	0	3.5
	Not applicable	5.6	3.6	0.2	24.5	3.3
Bagged feed drop off Overall (n = 345): Swine (n = 207), Poultry (n = 6), Ruminant (n = 127), Mixed (n = 5)	Barn office	4.4	2.9	0	7.1	0
	Loading chute	13.6	22.7	0	0	0
	Feed room	39.7	32.9	16.7	52.8	20
	Shed/garage	23.2	18.4	33.3	30.7	20
	Truck-to-truck	0.9	1.5	0	0	0
	Other	18.3	21.8	50.0	9.5	60.0
Barn entry	Didn’t enter	92.8	96.0	98.1	71.0	77.4
	Entered office	0.9	0.4	0.7	3.3	0
	Entered barn	4.7	3.3	1.1	16.2	19.4
	Entered an area with animals	1.6	0.4	0.2	9.6	3.2
Dead-stockb	None	91.3	89.4	91.9	98.0	90.3
	Where I walked	1.1	1.1	1.6	0	0
	Where I drove	2.0	2.3	1.9	0.3	3.2
	Visible	6.3	7.6	6.0	1.6	6.5
Location where mill order was left	At the feed bin	24.4	25.6	29.2	10.6	22.6
	Barn	7.2	3.8	2.0	30.1	19.4
	Barn office	2.8	1.7	4.4	5.0	0
	Door jamb	9.0	10.0	1.2	19.9	6.5
	Handed to customer	2.2	1.3	2.5	4.6	9.7
	Shed/garage	1.0	0.8	0.4	3.0	0
	Mailbox on the barn	45.3	52.3	45.0	16.2	38.7
	Regular mailbox	6.8	3.5	14.5	6.3	3.2
	Other	1.5	1.1	0.9	4.3	0

^aRepresents situations where the driver was aware that he/she needed permission to enter the facility, or where the driver indicated that they noticed biosecurity signs on the gate and/or barn. As a result, these categories may be under-represented.

^bThese categories are not mutually exclusive; as a result, some response patterns add up to > 100%.

^cThe option “snow” was not provided on the log sheets, but several of the drivers wrote it in. As a result, this category is likely under-represented.

^dThese values are presented as the proportion of farms where a feed bin was visited and was therefore applicable.

Most farms had clean laneways (80.7%) and feed bin areas (75.9%) and few farms (9.7%) had visible dead-stock (Table 1).

Types of trucks by percent of farm visits included blower, 65.4%; auger, 23%; box, 6.1%; premix or pickup, 3.5%; and auger/blower combination truck, 2%. Drivers washed the steering wheel 49.3%, floor mats 76.6%, tires 36.2%, and the exterior of the truck 32.1% percent of the time, within 24 h of the farm visits. For the truck exteriors not washed within 24 h of the visit, 50.2% were washed within 2 to 3 d, 44.2% within 4 to 7 d and 5.8% within 8 to 12 d.

Treatment and control groups

The use of gloves and footwear by drivers is presented in Table 2. Clean gloves were worn on 29% of farm visits (632/2179), disposables on 1.1% (24/2179), unclean gloves on 66.3% (1445/2179), and no gloves on 3.6% (78/2179). Drivers in the treatment group wore a clean pair of gloves significantly more often (64.8% of farm visits) than control-group drivers (18.6% of visits) (P < 0.001). Rubber boots were worn on 21.1% of farm visits (465/2200), disposable boots on 6.6% of visits (145/2200), and the driver’s own boots on 72.3% of visits (1590/2200) (Table 2). After controlling for individual driver as a random effect, drivers in the treatment group tended to wear clean boots in the farmyard (45% of visits) more often than did the control group drivers (9.7% of visits) (OR = 1.2; P = 0.06). Drivers were more likely to wear clean boots in the farmyard if they delivered bagged feed (OR = 1.04; P = 0.015), or if they wore a clean pair of boots in the building (OR = 1.8; P = < 0.001) (Table 3). On average, 62% of the variation in wearing boots in the yard was clustered within the driver.

Table 2.

Gloves and boots worn by feed truck drivers as part of a pilot study, over a 6-week period in Ontario (winter 2013)

	Treatment groupa (n = 489) (%)	Control group (n = 1690) (%)
Gloves
Clean	64.8	18.6
Disposable	1.0	1.1
Unclean	20.9	79.5
None	13.3	0.8
	Treatment groupb (n = 1121)	Control group (n = 1079)
Boots worn in yard
Clean bootsc	45.0	9.7
Personal boots	55.0	90.3

^aRepresents log sheets received from drivers in the treatment group from the 1 feed company that was provided with washable cotton gloves.

^bRepresents all drivers in the treatment group.

^cClean boots worn in the yard included both reusable, cleaned, rubber overshoe and plastic disposable boots with or without a treaded bottom, but did not include farm-owned boots used only in the barn. Reusable rubber overboots and disposable plastic boots were worn by drivers in the treatment group on 41.5% and 3.6% of farm visits respectively. Drivers in the control group only had access to disposable plastic boots. Clean boots in the barn included farm-owned boots plus reusable, cleaned rubber overshoe and plastic disposable boots.

Table 3.

Factors associated with feed truck drivers wearing clean boots in the farmyard, after controlling for the driver as a random variable, over a 6-week period in Ontario (winter 2013)

Variable	Odds ratio	Coefficient	Confidence interval	P-value	Intra-class correlation coefficienta
Belonged to the treatment group	1.3	0.236	−0.015, 0.487	0.065	0.694
Delivered bagged feed	1.04	0.044	0.008, 0.080	0.015	0.710
Entered the barn	NA	−0.006	−0.053, 0.040	0.795	0.709
Wore clean boots when entering building	1.8	0.576	0.477, 0.674	< 0.001	0.370

^aIntra-class correlation coefficient represents the variation in the wearing of boots that is clustered within the driver.

NA — Not available.

Drivers entered a building that did not house animals on 11% (238/2191) of farm visits and drivers entered the barn on 7.2% (158/2191) of farm visits. Among the visits to the barn, drivers entered the office 12% of the time (19/158), an area with animals 22.8% of the time (36/158), or another area of the barn 65.2% of the time (103/158). Drivers in the treatment group entered the barn more often (8.7% of visits) than drivers in the control group (5.7% of visits) (P = 0.007). Treatment-group drivers delivered more bagged products (18.4% of farm visits) than control-group drivers (12.9% of visits) (P < 0.001). However, among those visits where the driver entered the barn, there was not a significant difference in the proportion of visits delivering bagged feed between drivers in the treatment and control groups. Drivers entering the barn or another building while delivering feed were asked to record the type of footwear worn. These drivers wore rubber boots 4.2% (10/238), disposable boots 5.9% (14/238), farm boots 22.3% (53/238), and their own boots 67.6% (161/238) of the time.

Exit surveys

Twenty-one exit surveys were received from feed-truck drivers. Twelve were drivers who had tried new footwear. Five (of 6) drivers said the stretchy rubber boots did not easily fit over their own winter boots. The remaining drivers said they were easy to get on and off, and that he/she would continue to wear them. Of the 3 drivers who used the sturdy rubber overshoe, 1 said they didn’t fit over his own winter boots; two said they did fit and they would continue to wear them. All 3 drivers who used the treaded disposable plastic boots said they were unsafe and they would not continue to use them.

Several drivers made written or unrecorded oral comments or suggestions about the study itself, indicating they enjoyed participating and found the process interesting. A few drivers commented that the same rules need to be followed by all — including employees, sales people, and fuel-truck drivers. The drivers suggested the following changes to the study: collect information about farm-to-farm traffic (fuel and hydro trucks, sales people, repair people, livestock and dead-stock trucks); conduct the study in other seasons; and collect information about exhaust fans blowing onto the truck and driver during feed delivery, overall appearance of the farm, and washing blow pipes.

Multiple correspondence analysis of farm-level variables

The MCA solution is presented in Figure 1. Farms clustered in 3 separate groups based on the biosecurity decisions on the farm and by delivery of feed (P < 0.10). Group A included most farms, represented positive biosecurity and involved farms with swine, poultry, and mixed species. These farms had delivery of only bulk feed; clean farm lane; clean feed bin area; good dead-stock management; and the driver did not enter the barn. Group B farms had poor biosecurity including delivery of bulk and bagged feed; unclean farm lane and feed bin areas; and inadequate dead-stock management. Group C farms, with poor biosecurity had delivery of bagged feed where the driver entered the barn office and the area with animals.

Figure 1.

Clustering of farms by biosecurity measures specifically related to the delivery of feed. The farm-level variables, measured during 2202 feed deliveries in Ontario over a 6-week period (winter 2013), were analysed using multiple correspondence analysis. The farms clustered in 3 sections based on these biosecurity indicators. Section A represents farms clustered by positive biosecurity decisions including those with clean farm lane and feed bin areas, dead-stock not visible, driver did not enter the barn, farm received only bulk feed and swine, poultry or mixed farms by species. Section B represents farms in 1 category of farms with poor biosecurity including those that received both bulk and bagged feed, the farm lane and feed bin area were not clean, and there was inadequate dead-stock management. Section C represents a second group with poor biosecurity including farms that only received bagged feed and did not own feed bins, the driver delivered feed into the area of the barn with animals and entered the barn office.

In total, 76.2% of the variability in the biosecurity group was explained by the factors in the model. Feed type delivered and whether the feed bin areas and laneways were clean were the most influential factors (discrimination measures of 0.85, 0.84, and 0.69, respectively). The driver entering the barn, farm type, and adequate dead-stock management were less influential with discrimination measures of 0.42, 0.17, and 0.16, respectively.

Discussion

Data were collected by 40 feed truck drivers from 3 companies during 2202 farm visits in Ontario. Most producers had proper dead-stock management and maintained clean yards and feed bin areas. Feed-truck drivers cleaned their truck’s floor mats and steering wheels regularly and rarely entered the barn. The objective of this study was to determine biosecurity factors related to feed delivery to swine farm; therefore, most data were collected about swine farms. However, feed trucks frequently travel to multiple farm types within the same route. Therefore, deliveries to poultry, ruminant, and mixed species farms were recorded by some drivers. The breakdown of farms by animal species is not representative of feed company business as a whole; swine farms are over-represented, and farms with other species are under-represented.

When considering the feed delivery among farms, biosecurity concerns related to poor dead-stock management, manure in the farm lane, and spilled feed. Dead-stock is a biosecurity risk because pathogens may persist in run-off from, or in dead-stock itself (23,24). A variety of pathogens can survive in manure, including E. coli, Brachyspira hyodysenteriae, Salmonella, Cryptosporidium, and Giardia (25–27). Additionally, spilled feed might attract pathogen-carrying pests; pathogens that infect swine have been isolated from wild birds (5,14) and rodents (5,28–30). While these issues pose serious threats to biosecurity, this study revealed that they occurred infrequently. Dead-stock management was appropriate (> 90%), the farm lane and the feed bin area were clean (> 80%), and dead-stock or manure were rarely on the lane (< 3%) during the farm visits. These biosecurity risks considered to be a concern of drivers are infrequent. Drivers said driving or walking through manure or run-off from dead-stock, increased their risk of transmitting pathogens to the next farm.

Feed-truck drivers are asked to enter the barn when delivering bagged feed. Bacterial and viral pathogens can be spread on contaminated fomites such as boots, hands, and clothing. The odds of Salmonella seropositivity are lower in herds in which farmers wash their hands, and where there is a dedicated room for changing clothing and footwear (31). One study found feed-company employees entered the facilities on 24.4% of farms (32). In another study, feed-delivery personnel were allowed access on 28% of farms in a medium-density pig-producing region, and on 13% of farms in a high-density pig-producing region (33). The drivers in this study entered the barn on < 10% of farm visits. Where only bulk feed was delivered, the driver entered the barn less often than when bags were being delivered (2% versus 35% respectively). This information supports the general consensus that bagged-feed delivery presents a higher biosecurity risk, primarily because the driver is required to enter the barn. Half of the bagged feed deliveries were inside the barn. A previous Ontario study found 24% of bags were delivered inside the barn, concluding that this could be a potential route of entry for PRRSV (18).

Disposable plastic boots are available for feed-truck drivers to wear at each farm but they are slippery, causing a safety concern especially during winter. Drivers often wear their own boots from farm to farm. Although disposable plastic boots with treaded bottoms were expected to be safer than those without, none of the drivers in the study considered them safe. Without clean boots, pathogens that spread more easily in winter, TGEV (13) and PRRSV (7,9), may be carried by the driver from farm to farm. Treatment-group drivers tended to wear clean footwear at more farm visits than control-group drivers. On average 62% of the variation in wearing boots in the yard was due to individual driver decision. This represents individual driver choice and likely reflects safety and fit of clean footwear. This study found that washable rubber overshoes are a suitable alternative to the standard disposable plastic boot. However, the overshoes must easily fit over typical winter boots, as small size was the biggest challenge for the drivers. Drivers were more likely to wear clean boots in the farmyard if they also wore clean boots when they entered a building. This may indicate a personal choice by drivers to wear clean boots during all stages of the farm visit. The drivers entered the barn on only 3.3% and 1.1% of swine and poultry farm visits, whereas they entered the barn on 16.2% and 19.4% of ruminant and mixed-animal farms. Treatment-group drivers wore clean gloves at significantly more farm visits than did control-group drivers. Drivers require an insulated, washable glove for winter months. Use of clean footwear and gloves are expected to decrease risk of pathogen spread. Feed companies will need to determine if purchasing and cleaning reusable overboots and gloves is economically justified.

Footbaths cannot replace the use of clean boots. They are only effective if the boots and bath are free of organic matter. If not, they can increase the spread of infectious organisms (34).

The MCA illustrated that most farms did well in all areas of biosecurity (22,35). Farms with a clean lane tended to also have a clean feed bin area, good dead-stock management, and were likely to be swine and poultry farms (P < 0.10). Farms with poor biosecurity failed in multiple areas. Bagged-feed delivery was closely associated with the driver entering the barn and going into the area with animals. This illustrates the biosecurity risk of bagged-feed delivery. Farms with inadequate dead-stock management also had an unclean feed bin area. While most farms had good biosecurity surrounding feed delivery, the poor biosecurity farms pose a risk for pathogen spread.

This study was subject to some limitations. The log sheets did not distinguish between barn offices inside or outside the main barn. The rubber boots were too small for many drivers, preventing these drivers from using clean boots.

The study was done only in winter, and the findings may differ by season. Only the drivers in the treatment group were given clean gloves and footwear. Therefore differences in the use of clean gloves and footwear may be a measure of compliance. However all drivers had access to disposable plastic boots that could be used as clean boots. The purpose of this portion of the study was to determine if usage of clean footwear and gloves could be improved over the baseline, represented by the control group. This investigation has proved useful in that respect.

This 6-week pilot study has provided a baseline regarding potential feed-delivery biosecurity risks and increased awareness about biosecurity among the participants. Most farms managed dead-stock well and kept the farm lane and feed-bin areas clean. Drivers often wore their personal boots in the farmyard because disposable plastic boots are unsafe. Drivers provided with reusable rubber overshoes tended to wear clean footwear more often than drivers who only had access to plastic disposable boots. These findings are encouraging and suggest that a system for using washable rubber boots is logistically feasible. However, feed companies need to determine if the cost associated with washing these boots between farm visits can be justified. These pilot study data provide an overview from which further biosecurity research can continue.