Source: UNIVERSITY OF FLORIDA submitted to
SPATIAL EPIDEMIOLOGY OF BOVINE ANAPLAMOSIS AMONG DAIRY CATTLE IN PUERTO RICO
 
PROJECT DIRECTOR: Melendez, P. Urdaz, J. H.
 
PERFORMING ORGANIZATION
COLLEGE OF VETERINARY MEDICINE
UNIVERSITY OF FLORIDA
GAINESVILLE,FL 32610
 
NON TECHNICAL SUMMARY: Today, uncontrolled tick populations and bovine anaplasmosis (BA) in cattle are major concerns among livestock producers in Puerto Rico (PR). The livestock industry, particularly the dairy cattle industry, is continuously facing major economic constraints, which are threatening the stability and the economy of the agricultural sector in the island. Major economic losses include high mortality in adult cattle, abortion, poor growth performance, reduction in milk production, and poor fertility rates. An estimated economic loss of US $20 million was reported in 1989 in PR due to the presence of anaplasmosis, babesiosis, and Boophilus microplus (Canestrini). A.This project will determine the seroprevalence of bovine anaplasmosis among dairy cattle at the herd and individual animal level in Puerto Rico. B. This project will determine spatial patterns and risk factors for bovine anaplasmosis in Puerto Rico. C. This project will determine the distribution of Boophilus microplus, the tick vector for BA, in Puerto Rico. D. The purpose of this study is to learn about the current epidemiology of bovine anaplasmosis (BA) among dairy cattle in Puerto Rico (PR) and to establish national spatio-epidemiological data that may help to formulate and implement efficient and strategic control plans for BA and B. microplus in PR.
 
OBJECTIVES: Objectives: To investigate the current epidemiology of bovine anaplasmosis (BA) among dairy cattle in Puerto Rico (PR) and to establish national spatio-epidemiological data that may help to formulate and implement efficient and strategic control plans for BA and B. microplus in PR. The specific objectives of the project are: OBJECTIVE 1- To determine the serological prevalence of antibodies against Anaplasma marginale among dairy cattle at herd and individual animal levels in Puerto Rico using MSP-5 competitive ELISA. OBJECTIVE 2- To determine spatial patterns for antibody levels against Anaplasma marginale and identify risk factors associated with cattle premises, location, cattle demographics, and management factors, on cluster of bovine anaplasmosis in Puerto Rico by means of geographic information systems and spatial data analyses. OBJECTIVE 3- To determine the influence of ecological factors, temperature, humidity, and type of vegetation, within geographic and agricultural areas of Puerto Rico on the distribution of Boophilus microplus by means of remote satellite imagery.
 
APPROACH: Description of epidemiological dynamics of bovine anaplasmosis (BA) in dairy cattle in Puerto Rico (PR) will allow the quantification of economic losses and the establishment of risk factors to be used in future preventive and control programs. To date the current epidemiological situation of BA in PR is poorly reported and no large scale studies on this disease have been undertaken in the island. Several studies will be conducted to determine the epidemiological patterns of BA in dairy cattle in PR to update information about this disease. Procedures: I. Objective 1 - We will determine the seroprevalence of antibodies against Anaplasma marginale (causative agent of BA) among dairy cattle at the herd and individual animal level. 1. Measurements a. We will use a commercially available competitive MSP-5 ELISA serological test kit. 2. Procedure and data a. One blood sample will be collected for each selected animal in the herd. 3. Design of statistical analyses a. Cattle and farms will be randomly selected. b. Descriptive statistics will be used to analyze serological prevalence at individual and herd level according to breed, class of animal, and region. II. Objective 2 A. We will determine spatial patterns for antibody levels against BA at the herd level 1. Measurements a. We will use global positioning system (GPS) units to position selected farms and the serological information obtained from Objective 1 to determine the BA prevalence levels and spatial patterns within these farms and identify spatial clusters based on different prevalence levels in the farms. c. Provide a questionnaire (personal interview) for each farm which will supply information to identify risk factors. 2. Data a. Farms coordinates and survey data will be recorded for each farm and will be transferred into ArcGIS 9.0 TM software for analyses. 3. Design of statistical analyses a. Clusters in GPS point data will be analyzed by spatial point patterns statistical analyses, Cusick and Edward's test for inhomogeneous population and spatial scan statistic. b. Each variable collected from the survey will be tested by logistic regression correcting from potential confounders. III. Objective 3 A. We will determine the influence of ecological factors, temperature, humidity, and type of vegetation, within geographic and agricultural areas of Puerto Rico on the distribution of Boophilus microplus by means of remote satellite imagery 1. Measurements a. Collection sites will be randomly selected based on basic geographical and ecological measurements; coordinates (longitude, latitude, and elevation), climatic variables (wind, cloud cover, moon phase, temperature, and relative humidity), and land cover. 2. Data a. Ticks will be collected monthly by dragging (flagging) technique from designated collection sites. 3. Design of statistical analyses a. spatial prediction of the distribution of B. microplus will be analyzed by semivariogram-model fitting, co-kriging, and multiple regressions for predicting mapping. These techniques take into consideration temperature, elevation, precipitation, etc. and the number of ticks found in a specific site to perform the extrapolation procedures.
 
CRIS NUMBER: 0204349 SUBFILE: CRIS
PROJECT NUMBER: FLA-VME-04300 SPONSOR AGENCY: CSREES
PROJECT TYPE: SPECIAL GRANT PROJECT STATUS: TERMINATED MULTI-STATE PROJECT NUMBER: (N/A)
START DATE: Jul 15, 2005 TERMINATION DATE: Jul 14, 2008

GRANT PROGRAM: TROPICAL AND SUBTROPICAL RESEARCH
GRANT PROGRAM AREA: Special Research Grant

CLASSIFICATION
Knowledge Area (KA)Subject (S)Science (F)Objective (G)Percent
311341011704.250%
312341011704.250%

CLASSIFICATION HEADINGS
KA311 - Animal Diseases
KA312 - External Parasites and Pests of Animals
S3410 - Dairy cattle, live animal
F1170 - Epidemiology
G4.2 - Reduce Number and Severity of Pest and Disease Outbreaks


RESEARCH EFFORT CATEGORIES
BASIC 50%
APPLIED 50%
DEVELOPMENTAL (N/A)%

KEYWORDS: puerto rico; dairy cattle; boophilus microplus; epidemiology; anaplasmosis; disease vectors; spatial analysis; geographic distribution; disease control; pest control; antibodies; elisa (assay); disease detection; risk; location; demography; geographic information systems; satellites; remote sensing; temperature; humidity; vegetation; bacterial diseases (animals)

PROGRESS: Jul 15, 2006 TO Jul 14, 2007
Seventy-eight randomly selected dairy farms were visited from August 23, 2005 to December 4, 2006. Serum samples were obtained from 2,475 adult lactating cows and analyzed by a commercial MSP-5 cELISA for Anaplasma marginale. Preliminary statistical analyses have been performed and the overall prevalence for Anaplasma marginale in Puerto Rico was 30.6%, SE 0.023 and 95% CI of 26.1 to 35.1 in 2005-06. In addition, prevalence calculations have been performed for all four major ecological zones in Puerto Rico. The North Coastal Region had a prevalence of 28.9%, SE 0.024 and 95% CI of 24.2 to 33.6. The Western Interior had a prevalence of 38.7%, SE 0.064 and 95% CI of 16.3 to 41.2. The South Coastal Region had a prevalence of 41.2%, SE 0.072 and 95% CI of 27.1 to 55.3. The Eastern Interior had a prevalence of 36.7% SE 0.106 and 95% CI of 15.9 to 57.6. GIS data were obtained from two different sources in PR (the Department of Urban Planning and the International Institute of Tropical Forestry). Shapefiles include but are not limited to major roads, rivers, lakes, barrios, and PR Holdridge life zones, etc. Cluster identification and major geographical risk factors will be obtained from these files and field data. Data from 78 in-person questionnaires (including GPS positions) were obtained from each farm visited during the last 18 months. Preliminary logistic regression methods were used to analyze these data. The outcome variable was dichotomized using the median seroprevalence for A. marginale from the sampled farms. Those herds with a median seroprevalence more than 25.4 were considered high prevalence and those farms with a median seroprevalence less than 25.4 were considered low prevalence. Crude odds ratios were calculated and those variables with a P value less than 0.20 were selected. Significant variables included type of housing, use of fly control methods, source of replacement heifers, use of on-farm irrigation systems, and type of grass. Forty-five percent of the farms had open pasture grazing, 38% had a loose housing shed with partial access to pastures, 12% had dry-lots, and 5% had free-stalls. Seventy-one percent of the farmers did not utilize any methods for fly control in the farms, while 29 % used products like permethrin. Fifty percent farms raised their own replacements while other farms either buy from a different farm, a local market, or import cattle from United States. Ticks collections were performed during two different seasons to determine the population density fluctuation (March and August). Based on these facts, 97 sites were visited in PR from March 4 to April 1, 2007 for tick larvae collection. A single disposable 1-m2 white flannel cloth was dragged through vegetation over 50-m by 1-m straight paths at each site. Preliminary analyses were performed on this collection but completion of the analyses will be done after the second collection takes place. Crude odds ratios were calculated and those variables with a P≤ 0.20 were selected.Significant variables included the presence of cattle or horse manure, the percent of shrubs, and the percent of grass.

IMPACT: 2006-07-15 TO 2007-07-14 The seroprevalence of A. marginale in Puerto Rico (31%) was lower than expected when compared with similar studies in the Caribbean and Central America (61-72%). Preliminary serological results indicated that A. marginale is widespread in Puerto Rico. All participating farms were positive for antibodies against the organism. Preliminary spatial analysis for clustering of A. marginale antibodies (I = -0.022, z value = -0.103) suggest that the pattern is neither clustered nor dispersed. However, local clusters may exist. Further analyses with all the farms included in the study are still required to determine if clusters of high seroprevalence for A. marginale exist in Puerto Rico. Many of the risk factors variables found to be significant in preliminary, unadjusted analyses are similar to those found by a study performed in lactating dairy cattle in Costa Rica. Multivariate logistic models that are capable of estimating adjusted odds ratios will be constructed upon completion of data collection. Results obtained in this study will change and improve current methods used to control bovine anaplasmosis at the farm, region, and state levels.

PUBLICATION INFORMATION: 2006-07-15 TO 2007-07-14
Urdaz-Rodriguez JH. Seroprevalence of Bovine Anaplasmosis and Babesiosis among Lactating Dairy Cows in Puerto Rico. 11th International Symposium for Veterinary Epidemiology and Economics, Cairns, Australia, August 6-11, 2006, page 228.

PROJECT CONTACT INFORMATION
NAME: Urdaz, J. H.
PHONE: 787-242-1399
FAX: 787-833-7765