Plant species affects establishment of Escherichia coli O157:H7 gfp+ on leafy vegetables.

AIMS: Greenhouse trials were conducted with different cultivars of baby leaf spinach, rocket and Swiss chard and inoculation of Escherichia coli O157:H7 gfp+, to determine whether plant species and cultivar have an impact on the establishment of this strain. METHODS AND RESULTS: Three cultivars each of spinach, rocket and Swiss chard were spray inoculated with E. coli O157:H7 gfp+ at doses of log 7 CFU per ml. Due to the different lengths of growing period spinach and Swiss chard were spray inoculated three times and rocket five times, with final inoculation performed 3 days prior to harvest. After a growing period of 26-33 days, E. coli O157:H7 gfp+ was recovered from the leaf surface in mean populations between log 1 and 6 CFU per gram. The lowest occurrence of E. coli O157:H7 gfp+ was found on rocket leaves and the highest on spinach. There was no significant difference in the establishment of E. coli O157:H7 gfp+ between cultivars, but there were differences between plant species. Indigenous phyllosphere bacteria were pure cultured and identified with 16S rRNA gene sequencing. CONCLUSIONS: Despite the same high inoculation dose of E. coli O157:H7 gfp+ on leaves, the establishment rate differed between plant species. However, plant cultivar did not affect establishment. Pantoea agglomerans dominated the identified bacterial isolates. SIGNIFICANCE AND IMPACT OF THE STUDY: As previous studies are inconclusive on choice of model plant species and cultivar, we studied whether plant species or cultivar determines the fate of E. coli O157:H7 gfp+ on leafy vegetables. The findings indicate that plant species is a key determinant in the establishment of E. coli O157:H7 gfp+.

Novel foods: a risk profile for the house cricket (Acheta domesticus).

Novel foods could represent a sustainable alternative to traditional farming and conventional foodstuffs. Starting in 2018, Regulation (EU) 2283/2015 entered into force, laying down provisions for the approval of novel foods in Europe, including insects. This Approved Regulation establishes the requirements that enable Food Business Operators to bring new foods into the EU market, while ensuring high levels of food safety for European consumers. The present risk profile tackles the hazards for one of the most promising novel food insects, the house cricket (Acheta domesticus). The risk profile envisages a closed A. domesticus crickets rearing system, under Hazard Analysis and Critical Control Points (HACCP) and good farming practices (GFP), in contrast with open cricket farms. The methodology used involves screening the literature and identifying possible hazards, followed by adding relevant inclusion criteria for the evidence obtained. These criteria include animal health and food safety aspects, for the entire lifespan of crickets, based on the farm to fork One Health principle. When data were scarce, comparative evidence from close relatives of the Orthoptera genus was used (e.g. grasshoppers, locusts and other cricket species). Nevertheless, significant data gaps in animal health and food safety are present. Even if HACCP-type systems are implemented, the risk profile identifies the following considerable concerns: (1) high total aerobic bacterial counts; (2) survival of spore-forming bacteria following thermal processing; (3) allergenicity of insects and insect-derived products; and (4) the bioaccumulation of heavy metals (e.g. cadmium). Other hazards like parasites, fungi, viruses, prions, antimicrobial resistance and toxins are ranked as low risk. For some hazards, a need for additional evidence is highlighted.

Toxoplasma gondii seroprevalence in wild boars (Sus scrofa) in Sweden and evaluation of ELISA test performance.

Toxoplasma gondii is a zoonotic protozoan parasite, infecting a wide range of warm-blooded animals. The Swedish wild boar population is expanding and increased hunting provides its meat to a growing group of consumers. We performed a spatio-temporal investigation of T. gondii seroprevalence in Swedish wild boars. An ELISA was set up and evaluated against a commercial direct agglutination test, using Bayesian latent class analysis. The ELISA sensitivity and specificity were estimated to 79% and 85%, respectively. Of 1327 serum samples, 50% were positive. Thirty-four per cent of young wild boars and 55% of adults were positive (P < 0.001). The total seroprevalence ranged from 72% in 2005 to 38% in 2011 (P < 0.001), suggesting a declining trend. The highest seroprevalence, 65%, was recorded in South Sweden. In other regions it varied from 29% in Stockholm to 46% in East Middle Sweden.

Risk factors associated with the presence of Campylobacter-positive broiler flocks in Sweden.

Approximately 40% of Swedish broiler producers deliver Campylobacter-negative broilers in 90-100% of their flocks, showing that it is possible to produce Campylobacter-free broilers in Sweden. This study investigated risk factors associated with the presence of Campylobacter-positive flocks at Swedish broiler producers. A total of 37 producers, with 90 broiler houses including 144 compartments, participated in the study. An on-farm interview was performed to collect information on potential risk factors for Campylobacter infection in broilers, with questions about farmcharacteristics such as the environment around the farm and broiler houses, design of the broiler houses and on-farm management practices. Negative binomial regression models were used to assess the statistical significance of risk factors associated with the within-farm number of Campylobacter-positive flocks (the outcome). Campylobacter is transmitted in broiler flocks due to multiple factors and several potential sources, illustrating the complexity of Campylobacter epidemiology in broilers. Factors significantly associated with increased proportion of Campylobacter-positive flocks were the presence of other livestock on the farm, or the presence of cattle, swine, poultry or fur animals within 1km of the farm. Poor or average general tidiness were associated with increased proportion of Campylobacter-positive flocks, but decreased if split slaughter was seldom or never applied or if farm workers changed footwear twice or three times instead of once before entering the broiler house.

Validation of a national disease recording system for dairy cattle against veterinary practice records.

In Sweden, morbidity in dairy cattle is monitored through a national disease recording system. This system gives valuable information for research as well as advisory work and genetic evaluation. Our main objective was to evaluate the completeness in the disease recording system. Farm copies of veterinary records (n=851) from 112 herds, from March 2003 to April 2004, were compared with the information registered in the recording system. The evaluation of completeness was performed at two stages: (i) in the raw data transferred from the Swedish Board of Agriculture (SBA) to the Swedish Dairy Association (for records, cases and diagnostic events) and (ii) in the dairy disease database (DDD) at the Swedish Dairy Association (for diagnostic events). The evaluation was stratified by record type: manual and computerized records from state-employed veterinarians and private veterinarians, respectively. The completeness was high both for records (95-100%) and cases (90-99%) except manual records from private veterinarians (76% for records and 74% for cases). The overall completeness for diagnostic events was 75% in the DDD, with significant differences between record types. For all record types other than manual records from private veterinarians, the majority of diagnostic events lost disappeared after registration in the raw data from the SBA. The reasons for loss found suggest that there is potential for improvement. A multilevel logistic regression analysis showed that the completeness of diagnostic events in the DDD depended on region, diagnosis and veterinary employment. The random effect of veterinarian accounted for 35% of the modeled variation. Future studies are needed to assess how the differential misclassification affect estimates based on the data, and how to account for it.

Comparison between dairy cow disease incidence in data registered by farmers and in data from a disease-recording system based on veterinary reporting.

Sweden has a national disease-recording system based on veterinary reporting. From this system, all cattle-disease records are transferred to the dairy industry cattle database (DDD) where they are used for several purposes including research and dairy-health statistics. Our objective was to evaluate the completeness of this data source by comparing it with disease data registered by dairy farmers. The proportion of veterinary-treated disease events was estimated, by diagnosis. Disease incidence in the DDD was compared, by diagnosis and age, with disease data registered by the farmers. Comparison was made, by diagnosis, for (i) all disease events and (ii) those reported as veterinary-treated. Disease events, defined as "observed deviations in health, from the normal" were recorded by the farmers during January, April, July and October 2004. For the diagnoses calving problems, peripartum disorders, puerperal paresis and retained placenta, incidence proportions (IP) with 95% confidence intervals (CIs) were estimated. For all other disease problems, incidence rates (IR) were used. In total, 177 farmers reported at least 1 month and 148 reported all 4 months. Fifty-four percent of all disease events in the farmers' data were reported as veterinary-treated. For several of the most common diagnoses, the IRs and IPs for all events were significantly higher in farmers' data than in the DDD. Examples are, in cows: clinical mastitis, cough, gastro-intestinal disorders and lameness in hoof and limb; and in young stock: cough and gastro-intestinal disorders. For veterinary-treated events only, significant differences with higher IR in the farmers' data were found in young stock for sporadic cough and sporadic gastro-intestinal disorders. The diagnosis "other disorders" had significantly more events in the DDD than in farmers' data, i.e. veterinarians tended to choose more unspecific diagnoses than the farmers. This result indicates that the true completeness is likely to be higher than our estimate. We conclude that for the time period studied there was differential under-reporting associated with the diagnosis, the age of the animal and whether the herd was served by a state-employed or private veterinarian.

Phenotypic and molecular characterization of Brachyspira spp. isolated from laying hens in different housing systems.

Several species of intestinal spirochaetes, Brachyspira (B.) alvinipulli, B. intermedia and B. pilosicoli, may cause reduced egg production and faecal staining of eggshells in chickens. The aim of this study was to characterize potentially pathogenic and presumably non-pathogenic Brachyspira spp. from commercial laying hens. Selective culture, phenotyping, PCR and 16S rRNA gene sequencing were used and clinical data were collected. Phenotypic profiles were obtained for 489 isolates and 351 isolates obtained after subculture, and 30 isolates were selected for molecular characterization. Seven isolates were positive by a B. intermedia-specific PCR based on the nox gene, and two were positive in a B. hyodysenteriae-specific 23S rRNA gene based PCR. By comparative phylogenetic analysis in combination with PCR and phenotyping, seven isolates were identified as B. intermedia, eight isolates as B. innocens, five as B. murdochii, and three isolates each as B. alvinipulli and "B. pulli". The remaining four isolates could not be assigned to any presently recognized species. Co-infection with several species or genetic variants of Brachyspira spp. were detected in some flocks and samples, suggesting a high level of diversity. Organic flocks with access to outdoor areas were at higher risk (RR=2.3; 95% CI 1.5-3.6) for being colonized than chickens in other housing systems. No significant differences between colonized and non-colonized flocks were found regarding clinical parameters, i.e. mortality, egg production, faecally contaminated eggshells, and wet litter. Our results show that a combination of traditional laboratory diagnostics, molecular tests and phylogeny is needed for identification of Brachyspira sp. from chickens.

Summary of the Swedish Campylobacter program in broilers, 2001 through 2005.

A Campylobacter monitoring program in broiler chickens was carried out in Sweden from 2001 through 2005. The objective was to reduce the occurrence of Campylobacter in the food chain through preventive measures, starting with primary production. The program involved collecting samples from all broiler flocks at slaughter and occasional additional times. The annual incidence of Campylobacter-positive slaughter batches progressively decreased from 20% in 2002 to 13% in 2005. Most of the positive batches had a high within-flock prevalence of Campylobacter. However, about 18% of the positive batches had a low-within-flock prevalence; Campylobacter spp. were isolated from at most 50% of the cloacal samples. The incidence of batches contaminated at slaughter ranged between 6 and 9% during the study period. During all 5 years, a seasonal peak of incidence was observed in the summertime. In an additional study, quantitative analyses were performed on neck skin samples and carcass rinse samples. Those results were compared with the positive and negative findings of the cloacal, cecum, and neck skin samples at slaughter. When Campylobacter was found in the cecum, there was a higher level of Campylobacter in the quantitative analyses. Those batches where Campylobacter already had been found on the farm had a higher concentration of Campylobacter than those batches in which Campylobacter was found only at slaughter. During the study period, about one-third of producers seldom delivered Campylobacter-positive batches (< 10% positive batches per year). Thus, it is possible to produce Campylobacter-free broilers in Sweden.

Correlations between Campylobacter spp. prevalence in the environment and broiler flocks.

AIMS: To investigate (i) possible correlations between the presence of Campylobacter spp. in the surroundings of broiler farms and their incidence in flocks, and (ii) possible associations between weather conditions and the occurrence of Campylobacter spp. METHODS AND RESULTS: Farms were selected according to previous results from the Swedish Campylobacter programme. Samples were collected in and around broiler houses during the rearing period from 131 flocks on 31 farms, including sock samples from the ground outside, from the floor in the broiler houses and anterooms, and samples from insects, water, feed and ventilation shafts. CONCLUSIONS: As expected, there was a difference in Campylobacter isolation rates for different categories of farms regarding samples taken in the houses. However, there were no differences regarding the presence of Campylobacter spp. in the environment between producers that often deliver Campylobacter-positive slaughter batches and those that rarely deliver positive batches. Campylobacter spp. were more frequently found in the surroundings on rainy days when compared with sunny days. SIGNIFICANCE AND IMPACT OF THE STUDY: Physical barriers between outside and inside the houses appeared to be important for preventing Campylobacter spp. in the environment to be transferred into the broiler houses.

Feed-borne outbreak of Salmonella cubana in Swedish pig farms: risk factors and factors affecting the restriction period in infected farms.

In 2003, a feed-borne outbreak of Salmonella Cubana occurred in Sweden as a result of contamination in a feed plant. Salmonella Cubana was detected in 49 out of 77 pig farms having received possibly contaminated feed. In this study, potential risk factors for farms being salmonella positive were examined, and a survival analysis was performed to investigate risk factors affecting the restriction period for salmonella positive farms. The median restriction time for all 49 farms was 17 weeks. An increased risk for farms being salmonella infected (positive in feed and/or faeces) was seen for fattening farms and farms feeding soy. The survival analysis showed that herds with a low level of infection and farms with a high hygiene level had shorter restriction times. This study is unique as it investigates a real outbreak of feed-borne salmonella, where the source of infection was reliably identified, the period of exposure could be defined and data were collected from all exposed farms.

Postchill campylobacter prevalence on broiler carcasses in relation to slaughter group colonization level and chilling system.

Data from an ongoing national surveillance program of Campylobacter prevalence in broiler slaughter groups were related to results from a 1-year baseline study of broiler carcasses postchill. The goals were to establish the relation between Campylobacter prevalence in slaughter groups and on carcasses and to determine the effect of various chilling systems on Campylobacter prevalence. Pooled cloacal and neck skin samples from the surveillance program were analyzed after enrichment. Carcass rinse samples from the baseline study were analyzed after enrichment and by direct plating. Data from both studies were available for 614 carcasses. Direct-plating analyses indicated that the percentages of carcasses positive for Campylobacter jejuni and other Campylobacter spp. in slaughter groups with negative cloacal samples were 2 and 10%, respectively, whereas enrichment analyses indicated prevalences of 2% in both cases. Campylobacter prevalence in slaughter groups with a high degree of intestinal colonization (more than half of the pooled cloacal samples positive) was significantly higher than in slaughter groups with a low degree of colonization (76 to 85% and 30 to 50%, respectively, depending on Campylobacter spp. and analytical method). The prevalence of Campylobacter-positive carcasses postchill was at the same level as the prevalence of carcasses that originated from slaughter groups with positive neck skin samples at four of the six slaughterhouses. Only at one slaughterhouse, with an air-chilling system, was the postchill prevalence (13%) lower than that expected from slaughter group data (23%). The postchill prevalence (43%) was higher than that expected from slaughter group data (33%) at one slaughterhouse with immersion chilling.

Transmission of Campylobacter spp. to chickens during transport to slaughter.

AIMS: To determine the prevalence of Campylobacter-contaminated transport crates and to determine whether contaminated crates represent a risk for contamination of chickens during transport to slaughter. METHODS AND RESULTS: Samples were collected from cleaned transport crates before they were dispatched to the farms. Chicken groups were sampled within 24 h before transport to slaughter and at the slaughterhouse. Campylobacter spp. were isolated from 69 of 122 (57%) sampled batches of transport crates. Twenty-six slaughter groups, negative at farm level, were transported in batches of crates from which Campylobacter spp. had been isolated. In 11 (42%) of these 26 slaughter groups, Campylobacter spp. were found in samples taken at slaughter. The corresponding figure for at-farm-negative slaughter groups transported in negative crates was four (15%) testing positive at slaughterhouse of 27 slaughter groups [relative risk (RR) = 2.9, 95% CI 1.1-7.3]. In four of 11 slaughter groups, genetic subtyping by pulsed-field gel electrophoresis was able to support the hypothesis of contamination from crates to chickens during transport to slaughter. CONCLUSIONS: Despite washing and disinfection, crates were frequently contaminated with Campylobacter and it could have contaminated chickens during transport to slaughter. SIGNIFICANCE AND IMPACT OF THE STUDY: Campylobacter-positive crates are a risk factor for chickens testing campylobacter-positive at slaughter.

Risk factors for hazard of release from salmonella-control restriction on Swedish cattle farms from 1993 to 2002.

In Sweden, only a few cattle farms are infected with salmonella each year and this can be attributed to the Swedish salmonella control programme. All findings of salmonella in animals, feed and food of animal origin are notifiable and restrictions are always put on infected herds until they have been cleaned up from the infection. However, there has been concern about increasing costs for clean-up of salmonella-infected farms as well as increasing length of the restriction periods. Our aim was to investigate potential risk factors associated with the length of restriction periods on Swedish cattle farms between 1993 and late 2002. All 112 cattle farms that were notified to the Swedish Board of Agriculture as infected with salmonella during the study period, were included in this longitudinal and retrospective study. The putative risk factors were analysed using the proportional-hazards model. There was a lower hazard for release from salmonella control restrictions after the European Union (EU) accession in 1995, and/or change of testing from one to two negative herd tests for release of restrictions (hazard ratio (HR)=0.56, 95% confidence interval (CI)=0.38, 0.84), for every additional number of 100 cows (HR=0.83; CI=0.7, 0.97), if rodents and/or wild birds were abundant (HR=0.5, CI=0.27, 0.98) and if there was more than one farm site in the company (HR=0.47, CI=0.28, 0.81).

Prevalence of verotoxin-producing Escherichia coli (VTEC) 0157 in Swedish dairy herds.

A prevalence study of verotoxin-producing Escherichia coli O157 (VTEC O157) was performed in 371 randomly selected dairy herds distributed throughout Sweden. Faecal and manure samples were collected and analysed by immunomagnetic separation and culturing. Data were recorded for each herd regarding herd size, age of sampled animals and whether, in addition to cattle, the farm kept other animals. VTEC O157 was isolated from 33 (8.9%) of the 371 investigated herds. The prevalence was higher (23.3%) in Halland county than in the rest of Sweden (P > 0.01). Halland was also the county in Sweden that during the study period had the highest incidence of human VTEC O157 cases. VTEC O157 was not detected on any farm in northern Sweden. Identified risk factors, in the multivariate analyses, for herds being VTEC O157 positive were herd size, geographical localization, presence of pigs on the farm and median age of sampled animals.

Surveillance programme for Campylobacter species in Swedish broilers, July 2001 to June 2002.

A total of 20,599 samples from 4133 slaughter groups representing about 71 million broilers slaughtered between July 1,2001 and June 30, 2002 were tested for Campylobacter species. From each slaughter group, 40 cloacal samples and 10 samples of neck skin were taken from the carcases on the slaughter line and the samples were pooled, 10 to a pool. Campylobacter species were detected in cloacal samples from 17 per cent of the slaughter groups and in neck skin samples from 22 per cent of the groups, and the monthly prevalence figures ranged from 6 to 32 per cent. In most of the infected groups all four pooled cloacal samples were positive, but 18 per cent of the infected groups had only one or two positive samples. More than 90 per cent of the Campylobacter-positive slaughter groups were delivered by half the producers, and one in four of the producers delivered no Campylobacter-positive slaughter groups.

Survey of Campylobacter species, VTEC O157 and Salmonella species in Swedish wildlife.

Samples collected from 791 wild animals (Canada geese, roe deer, hares, moose, wild boar and gulls) shot during hunting were examined for verocytotoxin-producing Escherichia coli (VTEC) O157, and thermophilic Campylobacter and Salmonella species. With the exception of one positive isolate from a wild boar, VTEC O157 was not isolated from any of the animals. Salmonella species were isolated only from the gulls, of which 4 per cent were estimated to be positive. Thermophilic Campylobacter species were commonly isolated from all the species except deer.

Bulk milk testing for antibody seroprevalences to BVDV and BHV-1 in a rural region of Peru.

Bulk milk from 60 herds of dairy cattle in a rural region in the central highlands of Peru was tested for antibodies to bovine viral-diarrhoea virus (BVDV) and bovine herpesvirus type 1 (BHV-1). None of the herds had been vaccinated against BVDV or BHV-1. Commercially available indirect ELISA-kits were used for antibody detection. True prevalences of BVDV and BHV-1 antibody-positive herds were 96 and 51%, respectively. A relatively low proportion of strongly positive herds suggests, however, a low prevalence of active BVDV infection. BVDV optical densities (ODs) in bulk milk increased with herd size--indicating a higher within-herd prevalence in the larger herds (probably, in part a consequence of a higher rate of animal movement into these herds). For BHV-1, this pattern was not found; a relatively high proportion of the herds was free from BHV-1 infection in each size category. This could indicate a low rate of reactivation of latent BHV-1 infection.

Animal- and herd-level risk factors for leptospiral seropositivity among sows in the Mekong delta, Vietnam.

In 1998, a total of 424 sows had sera collected in the Mekong delta in Vietnam. Of these, 283 sows were from 151 small-scale family farms in 19 villages, and 141 from seven large-scale state farms. The sera were subjected to the microscopic agglutination test (MAT) for antibodies to 13 Leptospira serovars. The overall leptospiral seroprevalence for titres > or =1:100 and > or =1:400, was 73 and 29%, respectively, and was higher (P=0.001) at small- than at large-scale farms. The highest seroprevalence was recorded for Leptospira interrogans serovar (sv) bratislava (52%). At small-scale farms, higher prevalences were found to serovars L. interrogans sv icterohaemorrhagiae (P=0.04) and L. interrogans sv pomona (P=0.02). Epidemiological information (at the individual-animal and herd-levels) was collected with a questionnaire. The data were analysed using logistic multiple regression. At the animal-level, sows seropositive for L. interrogans sv australis and sv autumnalis had less direct contact with sows in neighbouring pens (odds ratio (OR)=0.3 and 0.4, respectively) and sows seronegative for L. interrogans sv bratislava were of lower age (OR=0.1 for seropositivity). Also, sows seropositive for L. interrogans sv icterohaemorrhagiae had higher odds (OR=5.8) if they had not been born on the farm (had been introduced to it as gilts). Herds seropositive for sv javanica showed association with farms not taking measures to control the local rodent population (OR=7.8). Serovar pomona was also linked to the use of artificial insemination (AI), as opposed to natural-breeding services (OR=11.2). These results indicate that housing and management could affect the seroprevalence of Leptospira infection in pigs.

Persistence of verocytotoxin-producing Escherichia coli O157:H7 in calves kept on pasture and in calves kept indoors during the summer months in a Swedish dairy herd.

In 1997, a Swedish dairy farm was implicated in a human case of verocytotoxigenic Escherichia coli (VTEC) infection. The bacterium was found in a faecal sample from the human case and in faecal samples from cattle on the farm. Subtyping with pulsed field gel electrophoresis (PFGE) showed that the isolates were identical. The farm was further studied to assess the occurrence and the epidemiology of the agent at the farm level. The objective of this part of the study presented here was to examine the persistence of VTEC O157:H7 in calves that were kept on pasture and indoors, respectively, during the summer. Twelve calves in the herd, with one positive faecal sample each of VTEC O157:H7 in April 1999, were followed by faecal sampling during the summer months. Six calves were kept indoors and six were kept on pasture. Faecal samples from each calf were collected once a month on five occasions from April to September. Bacterial examination was performed with immunomagnetic separation (IMS) and cultivation on CT-SMAC. PCR was used to test for the presence of genes encoding for verocytotoxin (VT), intimin (eaeA), enterohemorrhagic E. coli-hemolysin (EHEC-Hly) and the flagellar antigen H7. PFGE was used for genotyping the isolates. The faecal samples from the calves kept on pasture were negative during the whole period. It is possible that the faecal samples had bacterial counts lower than the detection limits for our procedure, or that the faecal samples were free from the bacteria at the time of sampling. This suggests that calves on pasture may be less exposed to the bacteria or that they clear themselves. In the pen group, there were between one and six culture positive individuals per sampling occasion. One of the calves that was housed indoors was positive in faecal culture on four consecutive samplings.

Cross-sectional study of the seroprevalence to Borrelia burgdorferi sensu lato and granulocytic Ehrlichia spp. and demographic, clinical and tick-exposure factors in Swedish horses.

A cross-sectional study of the seroprevalence to Borrelia burgdorferi sensu lato and granulocytic Ehrlichia spp. in Swedish horses was conducted to evaluate associations with demographic, clinical and tick-exposure factors. From September 1997-1998, blood samples from 2018 horses were collected from the animals presented to veterinary clinics affiliated with the Swedish Horserace Totalizator Board (regardless of the primary cause for consultation). Standardized questionnaires with information both from owners and attending veterinarians accompanied each blood sample. The apparent seroprevalences to B. burgdorferi s. l. and granulocytic Ehrlichia spp. were 16.8 and 16.7%, respectively. The northern region had the lowest seroprevalences. Four logistic models were developed (controlling for demographic variables). In the disease model of seropositivity to B. burgdorferi s. l., age, breed, geographic region, the serologic titer to granulocytic Ehrlichia spp., season and the diagnosis coffin-joint arthritis were significant. In the tick-exposure model of B. burgdorferi s. l., pasture access the previous year and gender were significant. Age, racing activity, geographic region, season and the serologic titer to B. burgdorferi s. l. were associated with positivity to granulocytic Ehrlichia spp. In the tick-exposure model of granulocytic Ehrlichia spp., pasture access was a risk factor. An interaction between racing activity and geographic region showed that the risk of positive serologic reactions to Ehrlichia spp. was increased in the horse population in the south and middle of Sweden, but only among horses not used for racing. Except for the positive association between coffin-joint arthritis and serologic reactions to B. burgdorferi s. l., there were no significant associations in the multivariable models between non-specific or specific clinical sign or disease with seropositivity to either of these agents.