Materials and Methods

Data for the present study were obtained from the electronic medical records of all dogs examined for intestinal nematode parasitism at primary care veterinary hospitals owned by Banfield, The Pet Hospital, between January 1, 2003, and December 31, 2006. At the time of the study, all Banfield hospitals used the same proprietary software^a to record and store patient information, with each dog identified by a unique identification number. Methods used by Banfield hospitals during the study period for fecal sample collection, processing, and testing have been described.¹⁶ A commercial system^b with zinc sulfate solution^c (specific gravity, 1.18) was used for fecal flotation testing. When there was an insufficient quantity of feces for flotation testing, the fecal sample was diluted with saline (0.9% NaCl) solution and direct smears were examined microscopically.

Medical record information for dogs that had been examined for intestinal nematode parasitism was transferred to the Purdue University School of Veterinary Medicine and entered into a database.^d The following information was extracted for each dog included in the database: unique identification number, age, weight, sex, neuter status, breed, state of owner residence, month and year of the clinic visit when testing for intestinal parasitism occurred, and whether results of fecal testing were positive for Toxocara spp, Ancylostoma spp, or Trichuris spp. Breeds were grouped according to American Kennel Club breed groups¹⁷ as herding dogs, hounds, nonsporting dogs, sporting dogs, terriers, toy dogs, and working dogs. Mixed-breed dogs were those designated in the database as mixed-breed dogs or as belonging to a breed not recognized by the American Kennel Club. Age was grouped as < 0.5 years, 0.5 to 2.0 years, > 2.0 to 5.0 years, and > 5.0 years; weight was grouped as < 5.0 kg (11 lb), 5.0 to 10.0 kg (11 to 22 lb), > 10.0 to 24.0 kg (22 to 53 lb), and > 24.0 kg (53 lb).

Geographic regions used to map the prevalence of intestinal nematode parasitism were the same as those used by the CDC¹⁸ and consisted of New England, east north central, west north central, mid Atlantic, south Atlantic, east south central, west south central, mountain, and Pacific. State of owner residence was used to assign dogs to geographic regions. If the owner's state of residence was not recorded in the medical record, the state of the Banfield hospital where the dog was examined was used.

Parasite prevalence was calculated as 100 times the number of dogs identified as being infected by the specific parasite of interest divided by the number of dogs tested, and was calculated on the basis of results for the first fecal test performed on each dog. In addition, for dogs in which 2 or more fecal tests were performed, parasite prevalence was calculated on the basis of results for the second fecal test performed; for dogs in which 3 or more fecal tests were performed, parasite prevalence was calculated on the basis of results of the third fecal test performed; and for dogs in which 4 or more fecal tests were performed, parasite prevalence was calculated on the basis of results of the fourth fecal test performed. Observed parasite prevalence was not adjusted for sensitivity or specificity of the fecal flotation test.

Statistical analysis—The F² test was used to test for associations between host and environmental factors and the presence of intestinal parasites. The association between potential risk factors and the presence of intestinal parasites was expressed as the OR and 95% CI. Multivariable logistic regression with stepwise forward selection of variables on the basis of results of the likelihood ratio test was used to calculate ORs and 95% CIs adjusted for age, sex, neuter status, body weight, breed, region of residence, and month of examination. All analyses were performed with standard software,^d and values of P < 0.05 were considered significant. Geographic information system software^e was used to map prevalence of intestinal nematode parasitism by use of geographic region.

Results

The study population consisted of 1,213,061 dogs that underwent 2,785,248 fecal tests during visits between January 1, 2003, and December 31, 2006, to 547 Banfield veterinary hospitals located in 44 states. Results of a second fecal test performed during the study period were recorded for 667,408 dogs, results of a third fecal test performed during the study period were recorded for 397,853 dogs, and results of a fourth fecal test performed during the study period were recorded for 506,936 dogs (Table 1). Of the study dogs, 359,591 (29.64%) were spayed females, 349,814 (28.84%) were neutered males, 281,711 (23.22%) were sexually intact males, and 221,945 (18.30%) were sexually intact females. Toy dogs and sporting dogs were the most common, comprising 269,039 (22.18%) and 221,334 (18.25%), respectively, of the dogs in the study population. In addition, there were 97,871 (8.07%) hounds, 96,352 (7.94%) herding dogs, 96,220 (7.93%) nonsporting dogs, 84,417 (6.96%) terriers, 56,683 (4.67%) mixed-breed dogs, and 163,321 (13.46%) dogs of breeds that were not recognized by the American Kennel Club. The largest proportion of study dogs (240,728 [19.84%]) lived in the east south central region, whereas the smallest proportion (3,037 [0.25%]) lived in the west south central region.

For the study period as a whole, the number of samples tested per month ranged from a low of 83,295 (6.87%) in November to a high of 113,209 (9.33%) in May. There was a steady decrease by year in the prevalence of intestinal nematode parasitism after 2003, despite a significant (P < 0.001) increase in the number of dogs tested per year over the same time period (Table 2).

Results were positive for intestinal nematodes for 106,594 of the 1,213,061 (8.79%) initial fecal tests performed. For most of the dogs with positive fecal test results, only a single parasite genus was identified (88,123 [82.67%]), whereas for 17,902 (16.79%) dogs with positive fecal test results, 2 parasite genera were identified, and for 569 (0.53%) dogs, all 3 parasite genera were identified.

Puppies < 0.5 years old had significantly (P < 0.001) higher prevalences of Toxocara parasitism (10.88%) and Ancylostoma parasitism (7.77%), compared with dogs > 5.0 years old (0.34% and 1.03%, respectively; Table 3). However, puppies < 0.5 years old had a significantly (P < 0.001) lower prevalence of Trichuris parasitism (0.45%), compared with dogs > 5.0 years old (0.64%). Similarly, dogs weighing < 5.0 kg had significantly (P < 0.001) higher prevalences of Toxocara parasitism (7.89%) and Ancylostoma parasitism (5.65%), compared with dogs weighing > 24.0 kg (0.92% and 2.01%, respectively). In contrast, prevalence of Trichuris parasitism in dogs weighing < 5.0 kg was significantly (P < 0.001) lower (0.26%) than prevalence in dogs weighing > 24.0 kg (1.19%).

Prevalences of Toxocara (8.22%), Ancylostoma (7.46%), and Trichuris (0.95%) parasitism in sexually intact male dogs were significantly (P < 0.001) higher than prevalences in neutered male dogs (2.84%, 2.37%, and 0.67%, respectively; Table 3). Similarly, prevalences of Toxocara (8.30%), Ancylostoma (7.42%), and Trichuris (1.02%) parasitism in sexually intact female dogs were significantly (P < 0.001) higher than prevalence in spayed female dogs (2.69%, 2.45%, and 0.72%, respectively).

The highest prevalences of Toxocara (11.56%), Ancylostoma (13.76%), and Trichuris (1.91%) parasitism were observed in dogs in the west south central region (Figures 1–3; Table 3). Dogs in the mountain region had the lowest prevalences of Toxocara (1.89%), Ancylostoma (0.35%), and Trichuris (0.11%) parasitism.

View Full Size

Regional prevalence of Toxocara infection among dogs (n = 1,213,061; only results of the first recorded fecal test were considered) examined at 547 private veterinary hospitals in the United States between January 1, 2003, and December 31, 2006. Prevalence was expressed as the percentage of all dogs with positive results during the study period.

Citation: Journal of the American Veterinary Medical Association 234, 5; 10.2460/javma.234.5.631

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Regional prevalence of Ancylostoma infection among dogs (n = 1,213,061; only results of the first recorded fecal test were considered) examined at 547 private veterinary hospitals in the United States between January 1, 2003, and December 31, 2006. Prevalence was expressed as the percentage of all dogs with positive results during the study period.

Citation: Journal of the American Veterinary Medical Association 234, 5; 10.2460/javma.234.5.631

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Regional prevalence of Trichuris infection among dogs (n = 1,213,061; only results of the first recorded fecal test were considered) examined at 547 private veterinary hospitals in the United States between January 1, 2003, and December 31, 2006. Prevalence was expressed as the percentage of all dogs with positive results during the study period.

Citation: Journal of the American Veterinary Medical Association 234, 5; 10.2460/javma.234.5.631

• Download Figure
• Download figure as PowerPoint slide

Toy dogs had the lowest prevalences of Toxocara (2.24%), Ancylostoma (1.91%), and Trichuris (0.27%) parasitism (Table 3), whereas mixed-breed dogs had the highest prevalences of Toxocara (7.94%) and Ancylostoma (7.79%) parasitism, and herding dogs had the highest prevalence of Trichuris (1.21%) parasitism.

Prevalence of Toxocara parasitism was highest among dogs tested in December (6.19%; Table 3) and lowest among dogs tested in June (4.53%). Prevalence of Ancylostoma parasitism was highest among dogs tested in July (5.45%) and lowest among dogs tested in January (3.94%). Prevalence of Trichuris parasitism was highest among dogs tested in January (0.91%) and lowest among dogs tested in June and July (0.75%).

Factors associated with intestinal nematode parasitism—The final multivariable logistic regression models for risk factors associated with Toxocara, Ancylostoma, and Trichuris parasitism, developed on the basis of results of the first fecal test performed in each dog, included factors for age, weight, sex and neuter status, breed, geographic region, and month of testing (Table 4).

Dogs < 0.5 years old, 0.5 to 2.0 years old, and > 2.0 to 5.0 years old all had higher odds of Toxocara and Ancylostoma parasitism than did dogs > 5.0 years old. Similarly, dogs 0.5 to 2.0 years old and > 2.0 to 5.0 years old had higher odds of Trichuris parasitism than did dogs > 5.0 years old. However, odds of Trichuris parasitism among dogs < 0.5 years old was not significantly different (P = 0.40) from odds of Trichuris parasitism among dogs > 5.0 years old. Dogs weighing < 5.0 kg, 5.0 to 10.0 kg, and > 10.0 to 24.0 kg all had higher. odds of Toxocara and Ancylostoma parasitism than did dogs weighing > 24.0 kg, but dogs weighing < 5.0 kg. and dogs weighing 5.0 to 10.0 kg had lower odds of Trichuris parasitism than did dogs weighing > 24.0 kg.

Sexually intact male dogs were more likely to be infected with Toxocara spp (OR, 1.73), Ancylostoma spp (OR, 1.91), and Trichuris spp (OR, 1.63), and sexually intact female dogs were more likely to be infected with Toxocara spp (OR, 1.65), Ancylostoma spp (OR, 1.88), and Trichuris spp (OR, 1.89) than were spayed female dogs (Table 4). Neutered male dogs were more likely to be infected with Toxocara spp (OR, 1.06) and less likely to be infected with Trichuris spp (OR, 0.96) than were spayed female dogs. Dogs in all regions had higher odds of Toxocara, Ancylostoma, and Trichuris parasitism than did dogs in the mountain region, and dogs in all breed groups had higher odds of Toxocara, Ancylostoma, and Trichuris parasitism than did toy dogs.

Small variations in the odds of intestinal nematode parasitism were identified when dogs were grouped on the basis of month tested (Table 4). In general, the odds of Toxocara and Trichuris infection was highest among dogs tested during January or February, compared with dogs tested during June, and the odds of Ancylostoma infection was lowest among dogs tested during November or December, compared with dogs tested during June.

Discussion

Prevalences of intestinal nematode parasitism in the present study were similar to those reported in a previous study¹⁹ from Finland, but were considerably lower than prevalences reported in previous studies^10,12,20 from the United States. For example, a study¹² that involved flotation testing of 6,458 fecal samples collected from dogs in humane shelters in all 50 states and the District of Columbia reported that the prevalence of roundworm infection was 14.5%, the prevalence of hookworm infection was 19.2%, and the prevalence of whipworm infection was 14.3%. In a study¹⁰ involving 8,077 fecal samples obtained from dogs examined at the University of Pennsylvania Veterinary Hospital between 1984 and 1991 and tested with a zinc sulfate centrifugal flotation method, overall monthly prevalence of ascarid infection was 5.5%, prevalence of hookworm infection was 9.7%, and prevalence of whipworm infection was 9.7%. Finally, a study²⁰ performed at the Oklahoma State University Veterinary Medical Teaching Hospital that involved 1,250 fecal samples obtained each year from 1981 through 1990 reported that the prevalence of Toxocara infection decreased from 8% to 4%, the prevalence of Ancylostoma infection decreased from 39% to 15%, and the prevalence of Trichuris infection decreased from 12% to 9% during the study period. Even though prevalences of intestinal nematode parasitism differ among studies, higher prevalences of roundworm and hookworm infection in younger dogs and a higher prevalence of intestinal nematode parasitism in sexually intact dogs have been common findings.^{10,12,21–23}

Some studies^19,20,23 from the United States and other countries have reported that the prevalence of intestinal nematode parasitism has decreased over time and attributed this reduction to an increase in the use of broad-spectrum anthelmintics. In the United States, many professional and governmental organizations, such as the AVMA, CDC, and Companion Animal Parasite Council, have made efforts to increase awareness among veterinarians and dog owners of the risks of intestinal nematode parasitism and methods for prevention and treatment.

The true prevalence of intestinal nematode parasitism in dogs is likely to be higher than estimated prevalences reported in the present study because sensitivity of the fecal flotation test is < 100% and likely varies with the specific gravity of the solution used. To improve diagnostic accuracy, Banfield hospitals have started to implement fecal flotation coupled with centrifugation for routine examination of fecal samples because flotation with centrifugation is significantly more sensitive than flotation alone^24,25 and has been recommended by the Companion Animal Parasite Council.²⁶ However, underestimation of prevalence in the present study would not necessarily have biased our analyses of potential risk factors.

The finding that the prevalence of intestinal nematode parasitism was higher when results of the first fecal test were considered, compared with prevalences calculated on the basis of the second, third, and fourth fecal tests, was expected. Median age was lower at the time the first fecal test was performed than it was at the time subsequent fecal tests were performed, and younger dogs were more likely to have intestinal parasites. In addition, dogs with positive results for the first fecal test presumably were treated with anthelmintics, making it less likely they would have positive results for subsequent fecal tests.

Sexually intact male and female dogs had significantly higher odds of intestinal parasitism in the present study, compared with spayed female dogs, even after controlling for age, weight, breed, geographic region, and month of examination. Although there was no clear explanation for this finding, it is consistent with results of a previous study.¹³ Gonadectomy may reflect a higher degree of pet stewardship and a greater likelihood that owners would provide preventive care.^12,27 It is also believed that hormonally related behaviors such as roaming in sexually intact dogs may increase the likelihood of contact with parasites in the environment.²² Additionally, pregnancy can result in reactivation of somatic Toxocara canis and Ancylostoma caninum larvae.^12,22,27

The strong association found between young age (< 0.5 years) and a higher prevalence of Toxocara and Ancylostoma infections was expected because of the possibility of transplacental and transmammary infection with these parasites^4,9,28,29 and the likelihood of age-associated immunity.^12,28

Toy dogs had lower odds of intestinal nematode parasitism, compared with dogs in all other breed groups in the present study. Previous studies^12,22 found similar results, suggesting that preventive care or environmental exposures may differ between toy dogs and larger dogs.

The high odds of Ancylostoma infection among dogs in the west south central, east south central, and south Atlantic regions in the present study, compared with odds for dogs in the mountain region, was expected because soil and climate conditions in these regions are optimal for development and persistence of Ancylostoma larvae.^16,30 The more homogenous prevalence of Toxocara infection across various regions of the United States may reflect the ability of Toxocara eggs to remain viable for many years in the environment.^4,16,31 The lower odds of intestinal parasitism in general among dogs living in the mountain region has been reported previously.³²

Month during which fecal testing was performed was not found to be a strong risk factor for intestinal nematode parasitism. The lower odds of Ancylostoma infection among dogs tested in November or December, compared with dogs tested in June, may be attributed to the fact that warm temperatures are optimal for development of larval stages of this parasite.^16,30 Similarly, other studies^21,22 have found seasonal variations in prevalence of Ancylostoma infection in dogs, with peaks during the summer and spring.

Despite the large number of dogs included in the present study, several limitations should be noted. As previously discussed, the use of flotation without centrifugation for parasite detection likely resulted in underestimation of the prevalence of intestinal nematode parasitism. In addition, no attempts were made to identify parasites by species. Although Banfield hospitals that provided data for the present study were widely distributed geographically, it is not known how representative the study findings are for the general pet population in the United States. Nonetheless, the finding of a low prevalence of intestinal parasitism in dogs is somewhat reassuring.

Even though the overall prevalence of intestinal nematode parasitism was low in the present study, because dogs harboring intestinal parasites can shed large numbers of eggs into the environment, a potential risk of zoonotic transmission remains. Therefore, veterinarians should be familiar with recommendations from the AVMA, CDC, and Companion Animal Parasite Council for parasite prevention in puppies and adult dogs. Veterinarians should educate clients about the importance of intestinal nematode parasitism and emphasize the importance of preventive measures to reduce the risks of environmental contamination and zoonotic transmission.