Segura-Correa JC, Segura-Correa VM, Solís-Calderón JJ

Language: Spanish
References: 8
Page: 33-36
PDF size: 25.24 Kb.

ABSTRACT

Introduction. The sample size to estimate the prevalence of a disease is generally greater for a cluster sampling design than for a simple sampling design, due to the intracluster correlation. The number of times the sample size increases depends on the design effect (D). The objective of this study was to estimate the intracluster correlation and D for 14 serovars of Leptospira in state of Yucatan, Mexico. Material and Methods. Two cross-sectional studies with two sampling stages and a constant number of animals (7 cows) per cluster were carried out. The first study was done in the livestock region of Yucatan from April to June of 1997 (53 ranchs) and the second one in the agricultural and central regions from September to December of 1998 (52 ranchs). Blood samples were collected from the coccigeal vein of each animal and sera were diluted 1:100 using an microscopic aglutination test against 14 antigens of L interrogans serogroups. Positive sera were those with a 50% or more aglutination. The values for r_ho were calculated from variance components using restricted maximum likelihood procedures. The design effect values were calculated with the formula D= 1 + (b-1)r_ho, where b refers to the average number of cows sampled per ranch. Results and Discussion. The r_ho values ranged from 0.10 a 0.52, which indicates that clusters of 7 cows required sample sizes 1.7 to 4.6 times greater than for the simple sampling design. The greater design effects corresponded to L. javanica, L. wolffi, L. hardjo and L. tarassovi. It is recommended that estimates for r_ho and D for different diseases in the tropics and in different production systems be obtained to facilitate the planning of future surveys.

REFERENCES

1. 1.- Segura JC, Honhold N. Métodos de muestreo para producción y salud animal. Mérida: Universidad Autónoma de Yucatán; 2000. p. 139.

2. 2.- Bennett S, Woods T, Liyanage WM, Smith DL. A simplified general method for cluster surveys of health in developing countries. World Health Statistics Quarterly 1991; 44: 98-106.

3. 3.- Snedecor GW, Cochran WG. Métodos Estadísticos.México: Compañía Editorial Continental; 1978. p. xxx

4. 4.- Instituto Nacional de Estadísitica, Geografía e Informática (INEGI). Yucatán. Consulta Electrónica de Tabulados. Resultados definitivos del VII Censo Agrícola- Ganadero. Aguascalientes, México: INEGI; 1994.

5. 5.- Cole JR, Ellinghausen HC, Rubin HL. Laboratory diagnosis of leptospirosis of domestic animals. Proc US Animal Health Assoc. 1979. 83: 189-199.

6. 6.- SAS Institute Inc. SAS user’s guide, Statistics. Cary:SAS Institute Inc., 1989. p. 846.

7. 7.- Orjuela J, Navarrete M, Betancourt A, Roqueme L,Cortez E, Morrison RB. Salud y productividad en bovinos de la costa norte de Colombia. 2. Hallazgos serológicos, bacteriológicos y parasitológicos. World Animal Review 1991; 69: 7-14.

8. 8.- Otte MJ, Gumm ID. Intra-cluster correlation coefficient of 20 infections calculated from results of cluster-sample surveys. Prevent Vet Med 1997; 31: 147-50.