Hotspots of canine leptospirosis in the United States of AmericaAllison M. White a, Carlos Zambrana-Torrelio a,*, Toph Allen a, Melinda K. Rostal a,Andrea K. Wright b, Eileen C. Ball b, Peter Daszak a, William B. Karesh a

a EcoHealth Alliance, 460 West 34th Street, 17th Floor, New York, NY 10001, USAb Zoetis, 100 Campus Drive, Florham Park, NJ 07932, USA

A R T I C L E I N F O

Article history:Accepted 28 February 2017

Keywords:CanineLeptospirosisZoonosisRiskTopographyClimate

A B S T R A C T

Leptospirosis is a widespread zoonotic disease that causes hepatic and renal disease in dogs and humanbeings. The incidence of leptospirosis in dogs in the USA appears to be increasing. This study used 14years of canine leptospirosis testing data across 3109 counties in the USA to analyze environmental andsocio-economic correlates with rates of infection and to produce a map of locations of increased risk forcanine leptospirosis. Boosted regression trees were used to identify the probability of a dog testing pos-itive for leptospirosis based on microscopic agglutination test (MAT) results, and environmental and socio-economic data. The Midwest, East and Southwest were more likely to yield positive tests for leptospirosis,although specific counties in Appalachia had some of the highest predicted probabilities. Location (sub-urban areas or areas with deciduous forest) and climate (precipitation and temperature) were predictorsfor positive MAT results for leptospirosis, although the precise direction and strength of the effects wasdifficult to interpret. Wide geographic variation in predicted risk was identified. This risk mapping ap-proach may provide opportunities for improved diagnosis, control and prevention of leptospirosis in dogs.

© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Leptospirosis is a common and widespread zoonotic disease, withreservoirs in domestic and wild animals (Waitkins, 1985; Bharti et al.,2003; Nelson and Couto, 2003; Heymann, 2008; Costa et al., 2015).The disease is caused by spirochaetal bacteria belonging to the genusLeptospira, which infect a range of mammals, including humans, live-stock (e.g. cattle pigs and goats) and companion animals (e.g. dogsand horses) (Nelson and Couto, 2003; Heymann, 2008)1. Infectionis typically transmitted through direct contact of oral or nasalmucosa, or broken skin, with contaminated urine or water, and dogsare at risk of infection from drinking contaminated water (Nelsonand Couto, 2003; Heymann, 2008). Leptospirosis can cause severeclinical disease in dogs, including acute hepatic and/or renal failure.It can also produce a chronic carrier status, presenting as idio-pathic polyuria/polydipsia, which may not be preceded by severehepatic or renal disease (Ward, 2002b; Nelson and Couto, 2003).

Canine leptospirosis has been reported in the USA for more than100 years (Bolin, 1996) and the prevalence of leptospirosis is reportedto be increasing; the rate increased by 1.2 cases/100,000 dogs/yearfrom 1983 to 1998 (Ward et al., 2002). The national proportion of

positive microscopic agglutination test (MAT) results increasedfrom 8.7% to 12% from 2002 to 2004 (Glickman et al., 2006; Mooreet al., 2006). Clusters of cases of canine leptospirosis have beendetected in Texas, California and the upper Midwest, suggesting that,whilst leptospirosis is ubiquitous across the USA, some areas aredisproportionately affected (Ward, 2002a; Gautam et al., 2010;Hennebelle et al., 2013).

Environmental and socio-economic factors are thought to explainthe distribution and transmission of leptospirosis (Gubler et al., 2001;Sehgal, 2006; Reis et al., 2008). Flooding has been linked to out-breaks of leptospirosis (Reis et al., 2008). The risk of leptospirosisis also related to land cover (e.g. evergreen forests, percentage ofwetlands and public open spaces) and proximity to forests(Tangkanakul et al., 2000; Ward et al., 2004; Ahern et al., 2005;Ghneim et al., 2007; Alton et al., 2009; Raghavan et al., 2011, 2012a,2012b, 2013). To date, there has been no systematic analysis of thedistribution of leptospirosis in the USA and risk factors associatedwith the occurrence of the disease.

In the current study, we used novel statistical approaches toanalyze large sets of leptospirosis test data. We tested for thecorrelation of positive results with a wide range of hypothesizeddrivers and used the results to identify ‘hotspots’ of leptospirosis(areas of relative higher probability for leptospirosis cases) andthe factors that are likely associated with them. Based on theecology and epidemiology of leptospirosis, we hypothesized thatabiotic factors, such as precipitation and temperature, as well asanthropic activities, measured as change in land use cover, can be

* Corresponding author.E-mail address: zambrana@ecohealthalliance.org (C. Zambrana-Torrelio).

1 See: CDC, 2014. Leptospirosis. https://www.cdc.gov/leptospirosis/ (accessed 10May 2016).

http://dx.doi.org/10.1016/j.tvjl.2017.02.0091090-0233/© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

The Veterinary Journal 222 (2017) 29–35

Contents lists available at ScienceDirect

The Veterinary Journal

journal homepage: www.elsevier.com/ locate / tv j l

used to detect areas of higher prevalence (i.e. hotspots) of canineleptospirosis in the USA.

Materials and methods

Literature review

Literature related to the prevalence of canine leptospirosis and associated factorswas reviewed to identify variables for selection and to assist model building. PubMedand ISI Web of Knowledge data bases were searched for the terms ‘(leptospirosisOR Leptospira) AND (dog OR dogs OR canine OR Canis familiaris)’. Articles used werelimited to those published from 2000 onwards that were published in English, Frenchor Spanish that focused on North America (Canada, USA and Mexico); most data werefrom post-2000, although some publications included earlier data, extending as farback as 1970.

Data sources and explanatory variables

We obtained serological canine leptospirosis MAT results in the USA throughan agreement with IDEXX Laboratories. A total of 87,355 tests were available in theproprietary database; however, vaccine history and whether samples were submit-ted for paired MATs were unknown. All MAT results for tests conducted from 2000to 2014 at IDEXX Laboratories were included. The sensitivity of MAT has been es-timated to be 50–67% at 1:400 dilution and 22–67% at 1:800 dilution, while thespecificity ranges from 69–93% at 1:400 dilution and 69–100% for 1:800 dilutionin dogs with confirmed leptospirosis, and dogs with clinical and laboratory indica-tors for leptospirosis, but where ultimately leptospirosis was ruled out (Miller et al.,2008). MATs performed at the same laboratories on samples from specific patho-gen free dogs were 100% specific at both 1:400 and 1:800 dilutions (Miller et al.,2008). Although the sensitivity and specificity of both titers were similar, a cut-offof ≥1:800 dilution was selected for this study to limit the effect of unknown vacci-nation status; this was also the cut-off value used by IDEXX for diagnostic purposes.All serovars were included in our analysis without distinction. All data was anonymizedand geo-referenced at USA zip code level.

To control the effect of dog population size on the number of samples submit-ted for testing, we estimated the dog population at county level. We weighted totalstate level dog population data (American Veterinary Medical Association, 2012) bycounty level human population data2, assuming the proportion of dog owners isuniform within a state’s counties.

Variables for the spatial analyses were compiled in four main categories:(1) climate; (2) land cover type; (3) ecology; and (4) socio-economic status. Resultsfrom the literature review were used to provide information for selection of vari-ables. Climatic variables included precipitation and temperature across the USA overa 30 year period (1981–2010) (Hijmans et al., 2005)3. Land cover type and percent-age cover were collected from the 2011 National Land Cover Database (Jin et al., 2013).

The mean number of rodent species per county distributed across the USA wasused as an ecological indicator of host diversity under the assumption that higherdiversity increases the likelihood of transmission of leptospirosis, in the absence ofviable data on the number of rodents in a region (which is also likely to fluctuatewidely over time).

Median household income and percentage of residents with a Bachelor’s degreeor higher were used as socio-economic variables4. All variables were aggregatedat county level using US Census 2010 county boundaries5. Since the number of re-quested tests varied highly among counties, we determined the percentage of positivetests by county and used this proportion for all our analyses.

Modeling

Predictive models were constructed using boosted regression trees (BRTs)(Leathwick et al., 2006) with internal cross-validation, previously used to model speciesdistributions (Leathwick et al., 2006; De’ath, 2007; Pittman et al., 2009) and diseaseranges (Hay et al., 2013). Boosted regression trees fit an ensemble of regression treesto data in a stepwise fashion, up-weighting poorly predicted data points at each stepand monitoring predictive accuracy to prevent over-fitting. Compared to tradition-al statistical techniques, they predict patterns in complex data sets accurately andare robust for use on data sets with many interacting variables or non-linear rela-tionships; no P values, regression coefficients or confidence intervals are reported(Cutler et al., 2007; Elith et al., 2008).

The BRT used 31 county level variables for climate and precipitation, dog own-ership, landscape composition and rodent diversity. Data analysis was performed

using R (Jorge et al., 2015) and the R program dismo (R package, version 1.0–12).The predicted probability per county was mapped using ArcGIS version 10.2(Environmental Systems Research Institute).

Results

Literature review

Five hundred articles were identified in the initial search, 474of which were excluded on the basis of geography and relevance(e.g. randomized control trials, vaccine studies, laboratory diag-nostic development, case studies and studies examining acute renalfailure). We identified 26 peer reviewed publications with de-tailed testing data (see Appendix: Supplementary material). Thesample size in these articles ranged from 32 (Martin et al., 2014)to over 1 million dogs (Ward et al., 2002). We identified a range ofclimatic, land-cover and socio-economic factors associated withcanine leptospirosis (Tangkanakul et al., 2000; Ward et al., 2004;Ahern et al., 2005; Ghneim et al., 2007; Alton et al., 2009; Raghavanet al., 2011, 2012a, 2012b, 2013) and included these variables in themodel when data was available (Table S1; see Appendix: Supple-mentary material).

Descriptive statistics

A total of 12,317/87,355 (14.1%) MAT tests performed by IDEXXin the USA from 2000 to 2014 were positive for leptospirosis. Atleast one MAT result was available for 1260/3109 (40.5%) countiesin the contiguous USA (Fig. 1) from which 746/1260 (59.2%) coun-ties had one or more positive tests. The number of samples submittedfor MAT from a county ranged from 1 to 3761 (mean 62) and themaximum number of positive tests in one county was 670 (CookCounty, Illinois).

Modeling

Values from predictive modeling were interpreted as the prob-ability that a diagnostic test in a given county would be positive;for example, a probability of 0.14 indicates that there is a 14%chance that any given MAT will be positive. The Midwest, Eastand Southwest were more likely to yield positive tests for lepto-spirosis, although specific counties in Appalachia had some of thehighest predicted probabilities (Table 1). The highest predictedprobability observed was 0.37 in Webster County, West Virginia,whilst the lowest was 0.02 in Harney County, Oregon. The overallmedian predicted probability was 0.12. Eighty-four (2.7%) coun-ties had predicted probabilities >0.2. The final model fitted 1200trees; the cross-validation accuracy was 0.58, which is better thanrandom chance, but closer to random chance than perfect accura-cy. Hotspot maps of the relative risk of leptospirosis were producedfrom the BRT model for MAT results (Figs. 2 and 3). These show

2 See: US Census Bureau, 2010. 2010 American Community Survey 1-Year Estimates,Washington, DC. https://factfinder.census.gov (accessed 10 May 2016).

3 See: PRISM Climate Group. http://prism.oregonstate.edu (accessed 10 May 2016).4 See: US Census Bureau, 2010. 2010 American Community Survey 1-Year Estimates,

Washington, DC. https://factfinder.census.gov (accessed 10 May 2016).5 See: US Census Bureau, 2014. 2014 TIGER/Line Shapefiles. https://www

.census.gov/geo/maps-data/data/tiger-line.html (accessed 10 May 2016).

Table 1Counties with the highest predicted probabilities for canine leptospirosis in the USA.

Rank County State Predictedprobability

Dogs expected totest positive

1 Webster County West Virginia 0.370 1/2.72 Marion County Indiana 0.34 1/2.93 Harrisonburg City Virginia 0.33 1/3.04 Staunton City Virginia 0.33 1/3.05 Waynesboro City Virginia 0.31 1/3.26 Adair County Kentucky 0.31 1/3.27 Covington City Virginia 0.30 1/3.38 Curry County Oregon 0.30 1/3.39 Nicholas County West Virginia 0.30 1/3.3

10 Bedford City Virginia 0.29 1/3.4

30 A.M. White et al. / The Veterinary Journal 222 (2017) 29–35

Fig. 1. Percent of microscopic agglutination tests (MAT) results positive by county in the USA in 2000–2014. Darker colors indicate a greater proportion of positive tests.Counties with no data indicate that no MATs were submitted to the reference laboratory during the study period.

Fig. 2. Predicted probability of a positive microscopic agglutination test (MAT) result for canine leptospirosis in the continental USA. Predicted probabilities range from0.023 to 0.371, indicating that approximately 1/3 dogs tested is expected to be positive for leptospirosis. Scale is green to red where green indicates lower probability andred indicates higher probability.

31A.M. White et al. / The Veterinary Journal 222 (2017) 29–35

the probability by county that a given test will be positive forleptospirosis. Since the BRT model analyses complex interactionsand non-linear relationships, the directionality and precise rela-tionship between any individual variable and the outcome is tooidiosyncratic and context dependent to be meaningful on its own;for example, there appears to be a slightly decreased probabilityof a positive test at higher ranges of precipitation in the coldestquarter of the year, conditional on other variables, and there appearsto be a modest increase as the proportion of a county covered bylow intensity developed land approaches 0.3. Instead, the overallpredictions of a model for a county should be considered. Vari-ables important to the model’s overall output included precipitation,temperature, deciduous forested land, and low density developedland (e.g. residential areas with houses built on large lots or areaswith 20–49% of surface areas covered with impervious material)(Table 2).

Discussion

This analysis has produced the first comprehensive predictiverisk map for canine leptospirosis in the contiguous USA and statistical

Fig. 3. Insets from Figs. 1 and 2, showing the counties for Maine (above) and Idaho (below). The left side shows the percent of positive microscopic agglutination test (MAT)results and the right side shows the predicted probability of a positive MAT result.

Table 2Relative influence of the five most important variables on the boosted regressiontree model.

Variable Group Percentage relativeinfluence a

Deciduous forest Land cover 10.9Precipitation (coldest quarter mean) Bioclimate 9.0Scrubland and shrub land Land cover 6.30Developed (low intensity) Land cover 5.7Temperature (mean) Bioclimate 5.0

a Greater relative influence indicates greater contribution to model results.

32 A.M. White et al. / The Veterinary Journal 222 (2017) 29–35

support for some previously hypothesized risk factors. The varia-tion in canine leptospirosis risk in specific counties and regions ofthe USA appears to be mainly influenced by environmental and landuse factors. Our model suggests that landscape and environmen-tal factors, specifically low density developed land (e.g. residentialareas with houses built on large lots or areas with 20–49% of surfaceareas covered with impervious material), deciduous forested land,precipitation and temperature, are useful in predicting MAT testresults. However, as stated above, the model’s predictions for in-dividual variables are too idiosyncratic and context dependent tobe broadly applicable or simply summarized.

Whilst boosted regression trees allow for the building of complexmodels with multiple variables, one of the major drawbacks of thisapproach is that the individual effects of any one specific variableare not easily interpretable. Given the complexity of our model andthe number of variables included (e.g. multiple land cover vari-ables), we could identify important variables, but could not describethe precise relationship between one variable and the risk of lep-tospirosis. Despite the complexity, the analysis allowed us to indicatehow much influence a particular variable might have on the modelprediction (Table 2).

The mechanistic relationship between different vegetation typesand the risk of leptospirosis is unclear and may be a reflection thatrural regions provide more opportunity for rodent reservoirs to trans-mit leptospirosis to dogs. Deciduous forest cover, which contributedmost to the predictive power of the model, has not been proposedpreviously as a risk factor for canine leptospirosis. It is likely thatdeciduous forest is a proxy for other conditions favorable for lep-tospirosis transmission, such as specific precipitation regimes, higherrodent density and specific dog behavior.

One drawback of our model is that it does not account for flood-ing; therefore, it is possible that variables representing precipitationcould explain some of the leptospirosis risk attributed to floodingin the literature (Ahern et al., 2005; Raghavan et al., 2012a). Ourfinding of the risk for leptospirosis in spacious residential areas(lower density developed land) aligns with the findings of previ-ous studies (Ward et al., 2004; Ghneim et al., 2007; Raghavan et al.,2011), whilst only one study has cited rural areas as a risk for lep-tospirosis (Alton et al., 2009). Lower density developed land providesa combination of impervious surfaces that can concentrate rain waterrun-off and acts as a good habitat for reservoir animals, such asrodents and raccoons, hence simulating the effects of ‘flooding’without requiring a specific level of annual rainfall.

Many of the counties with the highest overall predicted riskare in Appalachia (i.e. West Virginia, Eastern Kentucky and WesternVirginia). The Appalachian Mountains receive high annualprecipitation6 and contain predominantly deciduous forests. Clus-ters of leptospirosis cases have been identified in Illinois andMichigan in other studies (Ward, 2002a; Gautam et al., 2010) andcoincide with the results of our modeling. Significant spatial clus-ters were observed throughout the USA, including in Texas, California,the greater Chicago area and some regions of the upper Midwest(Ward, 2002a; Gautam et al., 2010; Hennebelle et al., 2013). Thisincreased risk may be explained by the proximity to the Great Lakesand the moderately high annual precipitation in these areas.

In addition, we identified several counties in Kansas andNebraska at higher risk for leptospirosis, despite the absence ofavailable testing data for many counties in those states. The pres-ence of leptospirosis in these states is supported by the literature.Raghavan et al. (2011) Raghavan et al., (2012a), Raghavan et al.,(2012b) examined specific risk factors (hydrologic, environmental

and socio-economic factors) for canine leptospirosis in Kansas andNebraska. In the study of Harkin et al. (2003), 41/500 (8.2%) dogswere shedding leptospires.

Our analytical approach has some important limitations. Firstly,due to a lack of available testing data, some areas of the contiguousUSA (i.e. the upper Midwest and parts of the Southeast) were poorlyrepresented in our testing data. Whilst the use of predictive methodsallows us to make predictions even when data are lacking, the in-clusion of data from these areas would increase predictive accuracy.Additional analyses using testing data from these areas would providea means to refine and externally validate our results.

Secondly, we used a titer threshold ≥1:800 to select positive testsfor our model. This cut-off threshold is expected to identify moreexposures, whilst minimizing the risk of including positive titersdue to vaccination. When vaccinated dogs were monitored for 1 yearpost-vaccination, some dogs developed titers ≥1:800 by weeks 7–15,whereas by weeks 29–52 none of the dogs had titers ≥1:400 andonly a small percentage had titers ≥1:100 (Martin et al., 2014). Ourliterature review indicated that MAT titer thresholds for consider-ing a test to be positive ranged from ≥1:100 to ≥1:3,200. The 2010American College of Veterinary Internal Medicine (ACVIM) SmallAnimal Consensus Statement on Leptospirosis (Sykes et al., 2011)states that there is a lack of consensus as to which titer level shouldbe used to determine that a test is negative for leptospirosis andalso that single positive titers must be considered with clinical signsand a paired titer, since even a titer ≥1:800 does not confirm a di-agnosis of leptospirosis. We followed the recommendation of thediagnostic laboratory (positive titer ≥1:800), which fits with ourdesire to exclude animals vaccinated within the past year and tohave enough confidence in the likelihood of exposure to proceedwith the epidemiological analysis.

Given the relatively low vaccination rates for canine leptospi-rosis in the USA (<4% of dogs in 2011) (American Veterinary MedicalAssociation, 2012), we do not expect this to have a major impacton the results (Klaasen et al., 2003). Although vaccine history andwhether samples were submitted for paired MATs were unknown,given the overall size of the data set, this was considered unlikelyto have an impact on results.

Data were aggregated at county level due to data availability andto estimate leptospirosis risk in an easily interpretable manner toaid veterinarians in describing risks to owners. County lines are oftenarbitrary and may have changed during the course of our study. Usingan arbitrary unit of area (county) may have augmented some of theobserved effects. This would be likely to have the greatest impactin the Western USA, where the counties are large.

Our analysis provided some counterintuitive results. Several coun-ties with high-predicted values were found in areas where clustersof leptospirosis had not previously been identified. In some of thesecounties, the available data did not have a high proportion of pos-itive tests, indicating differences between the predicted values andexisting leptospirosis clusters (Figs. 2 and 3) (Gautam et al., 2010;Hennebelle et al., 2013). Counties with low risk sometimes adjoincounties at high risk; for example, in Virginia, small counties withhigh risk are sometimes entirely surrounded by larger counties withlower risk. This may be due to smaller counties being populationcenters or due to artifacts in the data. These specific counties shouldbe examined further to determine whether the low level of re-ported leptospirosis is due to lack of data, lack of testing or bettervaccination coverage.

Areas identified by our model as higher risk areas differed fromthe overall distribution of proportion of available positive teststhroughout the USA (Figs. 1 and 2), indicating that this methodprovides utility compared to analyzing historical testing data alone.Identifying the county level risk for leptospirosis can contribute todetermining where to implement prevention and control mea-sures, testing and vaccination.6 See: PRISM Climate Group. http://prism.oregonstate.edu (accessed 10 May 2016).

33A.M. White et al. / The Veterinary Journal 222 (2017) 29–35

Leptospirosis is a disease of increasing concern for both peopleand dogs. Identified canine leptospirosis incidences in the USA haveranged from 0.04% in a study of hospital prevalence from 1970–1998 across the USA, to as high as 29% in a study examining testssubmitted to the veterinary diagnostic lab in Illinois from 1996 to2001. This variation makes it important to identify risk areas forthis disease (Ward et al., 2002; Boutilier et al., 2003). The US Centersfor Disease Control and Prevention recently reinstated leptospiro-sis in human beings as a nationally notifiable disease; in 2014, therewere 21 human cases of leptospirosis (Centers for Disease Control,2015). Dogs play an important role as potential indicators of areaswith high endemicity for leptospirosis and, although infrequent, zoo-notic transmission of leptospirosis from dogs to human beings canoccur. Thus, recognizing and preventing canine leptospirosis has im-plications for human health as well as dogs.

Conclusions

Our model can be used to characterize the risk of canine lepto-spirosis across the USA and can be applied to improve delivery ofveterinary services, including diagnostics and vaccination, and toidentify areas for increased research and surveillance efforts. Canineleptospirosis remains an important disease, widely distributed inthe USA, with varying levels of disease risk. Recognizing and iden-tifying the areas most at risk will help to provide improved veterinarycare and control of this disease.

Conflict of interest statement

Funds for this project were provided by Zoetis, which currentlymarkets a vaccine against canine leptospirosis. Andrea Wright andEileen Ball are employees of Zoetis. None of the other authors ofthis paper have a financial or personal relationship with other peopleor organizations that could inappropriately influence or bias thecontent of the paper.

Acknowledgements

Funding for this project was provided by Zoetis. The authorswould like to thank IDEXX Laboratories for providing the data forthis project.

Appendix: Supplementary material

Supplementary data to this article can be found online atdoi:10.1016/j.tvjl.2017.02.009.

References

Ahern, M., Kovats, R.S., Wilkinson, P., Few, R., Matthies, F., 2005. Global health impactsof floods: Epidemiologic evidence. Epidemiologic Reviews 27, 36–46.

Alton, G.D., Berke, O., Reid-Smith, R., Ojkic, D., Prescott, J.F., 2009. Increase inseroprevalence of canine leptospirosis and its risk factors, Ontario 1998–2006.Canadian Journal of Veterinary Research 73, 167–175.

American Veterinary Medical Association (AVMA), 2012. US Pet Ownership andDemographics Sourcebook 2012. AVMA, Schaumberg, IL, pp. 21–23.

Bharti, A.R., Nally, J.E., Ricaldi, J.N., Matthias, M.A., Diaz, M.M., Lovett, M.A., Levett,P.N., Gilman, R.H., Willig, M.R., Gotuzzo, E., et al., 2003. Leptospirosis: A zoonoticdisease of global importance. The Lancet Infectious Diseases 3, 757–771.

Bolin, C.A., 1996. Diagnosis of leptospirosis: A reemerging disease of companionanimals. Seminars in Veterinary Medicine and Surgery (Small Animal) 11,166–171.

Boutilier, P., Carr, A., Schulman, R.L., 2003. Leptospirosis in dogs: A serologic surveyand case series 1996 to 2001. Veterinary Therapeutics: Research in AppliedVeterinary Medicine 4, 387–396.

Centers for Disease Control, 2015. Notifiable diseases and mortality tables. Morbidityand Mortality Weekly Report (MMWR) 63, ND-719–ND-732.

Costa, F., Hagan, J.E., Calcagno, J., Kane, M., Torgerson, P., Martinez-Silveira, M.S., Stein,C., Abela-Ridder, B., Ko, A.I., 2015. Global morbidity and mortality of leptospirosis:A systematic review. PLoS Neglected Tropical Diseases 9, e0003898.

Cutler, D.R., Edwards, T.C., Beard, K.H., Cutler, A., Hess, K.T., Gibson, J., Lawler, J.J., 2007.Random forests for classification in ecology. Ecology 88, 2783–2792.

De’ath, G., 2007. Boosted trees for ecological modeling and prediction. Ecology 88,243–251.

Elith, J., Leathwick, J.R., Hastie, T., 2008. A working guide to boosted regression trees.Journal of Animal Ecology 77, 802–813.

Gautam, R., Guptill, L.F., Wu, C.C., Potter, A., Moore, G.E., 2010. Spatial and spatio-temporal clustering of overall and serovar-specific Leptospira microscopicagglutination test (MAT) seropositivity among dogs in the United States from2000 through 2007. Preventive Veterinary Medicine 96, 122–131.

Ghneim, G.S., Viers, J.H., Chomel, B.B., Kass, P.H., Descollonges, D.A., Johnson, M.L.,2007. Use of a case-control study and geographic information systems todetermine environmental and demographic risk factors for canine leptospirosis.Veterinary Research 38, 37–50.

Glickman, L.T., Moore, G.E., Glickman, N.W., Caldanaro, R.J., Aucoin, D., Lewis, H.B.,2006. Purdue University-Banfield National Companion Animal SurveillanceProgram for emerging and zoonotic diseases. Vector Borne and Zoonotic Diseases6, 14–23.

Gubler, D.J., Reiter, P., Ebi, K.L., Yap, W., Nasci, R., Patz, J.A., 2001. Climatevariability and change in the United States: Potential impacts on vector- androdent-borne diseases. Environmental Health Perspectives 109 (Suppl. 2),223–233.

Harkin, K.R., Roshto, Y.M., Sullivan, J.T., Purvis, T.J., Chengappa, M.M., 2003. Comparisonof polymerase chain reaction assay, bacteriologic culture, and serologic testingin assessment of prevalence of urinary shedding of leptospires in dogs. Journalof the American Veterinary Medical Association 222, 1230–1233.

Hay, S.I., George, D.B., Moyes, C.L., Brownstein, J.S., 2013. Big data opportunities forglobal infectious disease surveillance. PLoS Medicine 10, e1001413.

Hennebelle, J.H., Sykes, J.E., Carpenter, T.E., Foley, J., 2013. Spatial and temporalpatterns of Leptospira infection in dogs from northern California: 67 cases(2001–2010). Journal of the American Veterinary Medical Association 242,941–947.

Heymann, D., 2008. Control of Communicable Diseases Manual, 19th Ed. AmericanPublic Health Association, Washington, DC.

Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolutioninterpolated climate surfaces for global land areas. International Journal ofClimatology 25, 1965–1978.

Jin, S., Yang, L., Danielson, P., Homer, C., Fry, J., Xian, G., 2013. A comprehensive changedetection method for updating the National Land Cover Database to circa 2011.Remote Sensing of Environment 132, 159–175.

Jorge, S., Monte, L.G., De Oliveira, N.R., Collares, T.F., Roloff, B.C., Gomes, C.K., Hartwig,D.D., Dellagostin, O.A., Hartleben, C.P., 2015. Phenotypic and molecularcharacterization of Leptospira interrogans isolated from Canis familiaris in SouthernBrazil. Current Microbiology 71, 496–500.

Klaasen, H.L., Molkenboer, M.J., Vrijenhoek, M.P., Kaashoek, M.J., 2003. Duration ofimmunity in dogs vaccinated against leptospirosis with a bivalent inactivatedvaccine. Veterinary Microbiology 95, 121–132.

Leathwick, J.R., Elith, J., Francis, M.P., Hastie, T., Taylor, P., 2006. Variation in demersalfish species richness in the oceans surrounding New Zealand: An analysis usingboosted regression trees. Marine Ecology Progress Series 321, 267–281.

Martin, L.E., Wiggans, K.T., Wennogle, S.A., Curtis, K., Chandrashekar, R., Lappin, M.R.,2014. Vaccine-associated Leptospira antibodies in client-owned dogs. Journal ofVeterinary Internal Medicine 28, 789–792.

Miller, M., Annis, K., Lappin, M., Gill, M., Lunn, K., 2008. Sensitivity and specificityof the microscopic agglutination test for the diagnosis of leptospirosis in dogs.Journal of Veterinary Internal Medicine 22, 787–788, (Abstract 287).

Moore, G.E., Guptill, L.F., Glickman, N.W., Caldanaro, R.J., Aucoin, D., Glickman, L.T.,2006. Canine leptospirosis, United States, 2002–2004. Emerging InfectiousDiseases 12, 501–503.

Nelson, R.W., Couto, C.G., 2003. Small Animal Medicine, 3rd Ed. Mosby, St Louis, MO.Pittman, S.J., Costa, B.M., Battista, T.A., 2009. Using LiDAR bathymetry and boosted

regression trees to predict the diversity and abundance of fish and corals. Journalof Coastal Research 53, 27–38.

Raghavan, R., Brenner, K., Higgins, J., Van der Merwe, D., Harkin, K.R., 2011. Evaluationsof land cover risk factors for canine leptospirosis: 94 cases (2002–2009).Preventive Veterinary Medicine 101, 241–249.

Raghavan, R.K., Brenner, K.M., Higgins, J.J., Hutchinson, J.M., Harkin, K.R., 2012a.Evaluations of hydrologic risk factors for canine leptospirosis: 94 cases (2002–2009). Preventive Veterinary Medicine 107, 105–109.

Raghavan, R.K., Brenner, K.M., Higgins, J.J., Shawn Hutchinson, J.M., Harkin, K.R., 2012b.Neighborhood-level socioeconomic and urban land use risk factors of canineleptospirosis: 94 cases (2002–2009). Preventive Veterinary Medicine 106,324–331.

Raghavan, R.K., Brenner, K.M., Harrington, J.A., Jr., Higgins, J.J., Harkin, K.R., 2013.Spatial scale effects in environmental risk-factor modelling for diseases. GeospatialHealth 7, 169–182.

Reis, R.B., Ribeiro, G.S., Felzemburgh, R.D.M., Santana, F.S., Mohr, S., Melendez, A.X.T.O.,Queiroz, A., Santos, A.C., Ravines, R.R., Tassinari, W.S., et al., 2008. Impact ofenvironment and social gradient on Leptospira infection in urban slums. PLoSNeglected Tropical Diseases 2, e228.

Sehgal, S.C., 2006. Epidemiological patterns of leptospirosis. Indian Journal of MedicalMicrobiology 24, 310–311.

Sykes, J.E., Hartmann, K., Lunn, K.F., Moore, G.E., Stoddard, R.A., Goldstein, R.E., 2011.2010 ACVIM Small Animal Consensus Statement on Leptospirosis: Diagnosis,epidemiology, treatment, and prevention. Journal of Veterinary Internal Medicine25, 1–13.

34 A.M. White et al. / The Veterinary Journal 222 (2017) 29–35

Tangkanakul, W., Tharmaphornpil, P., Plikaytis, B.D., Bragg, S., Poonsuksombat, D.,Choomkasien, P., Kingnate, D., Ashford, D.A., 2000. Risk factors associated withleptospirosis in northeastern Thailand, 1998. American Journal of TropicalMedicine and Hygiene 63, 204–208.

Waitkins, S.A., 1985. From the PHLS. Update on leptospirosis. British Medical Journal(Clinical Research Edition) 290, 1502–1503.

Ward, M.P., 2002a. Clustering of reported cases of leptospirosis among dogs in theUnited States and Canada. Preventive Veterinary Medicine 56, 215–226.

Ward, M.P., 2002b. Seasonality of canine leptospirosis in the United States and Canadaand its association with rainfall. Preventive Veterinary Medicine 56, 203–213.

Ward, M.P., Glickman, L.T., Guptill, L.E., 2002. Prevalence of and risk factors forleptospirosis among dogs in the United States and Canada: 677 cases (1970–1998). Journal of the American Veterinary Medical Association 220, 53–58.

Ward, M.P., Guptill, L.F., Wu, C.C., 2004. Evaluation of environmental risk factors forleptospirosis in dogs: 36 cases (1997–2002). Journal of the American VeterinaryMedical Association 225, 72–77.

35A.M. White et al. / The Veterinary Journal 222 (2017) 29–35