2008, Number 2
<< Back Next >>
Vet Mex 2008; 39 (2)
Comparison of four regression models for the study of herd risk factors for binomial correlated data
Segura CJC, Salinas-Meléndez A, Solís CJJ
Language: English/Spanish
References: 20
Page: 129-137
PDF size: 376.84 Kb.
ABSTRACT
The objectives of this study were to compare four linear or logistic regression models and to determine their effects on the level
of signifi cance and parameter estimates, using the data from a study on seroprevalence of brucellosis in goats. Information on
5 114 does tested during 2002-2003 from 79 herds in the Bajio region in Michoacan, Mexico was used. The models were: the
prevalence of seropositive animals per herd (V1), analyzed by a general linear model (GLM), herds with at least one seropositive
animal, analyzed by standard logistic regression (SLRH); V1 analyzed by standard logistic regression (SLR), assuming independence
among results within a same herd (SLRA); and V1 analyzed by mixed LR, considering the herd as random effect (MLR). The
risk factors included in the four models were: the presence of abortions the year previous to the study, cleanness of the corral
(hygiene) and length of lactation. The V1 variable transformed to arcsine-square root did not show a normal distribution. SLRH
model (SLR assuming the herd as the unit of interest) and MLR were not compared because they were not nested models. MLR
model adjusted the data better than the SLRA model. The deviance (−2LL) from model SLRH (70.6) was similar to their degrees
of freedom (75), suggesting that the model adjusted the data very well. Levels of signifi cance for the risk factors were different,
depending of the model used. GLM and SLRH models showed signifi cant effects (P ‹ 0.02) only for the presence of abortions;
SLRA model showed signifi cant effect (P ‹ 0.05) for the three risk factors, and MLR, effects of the presence of abortions and lactation
length, but not for hygiene. The values for the odd ratios (OR) for the SLRA and MLR models were different; the narrowest
confi dence intervals corresponded to the SLRA model, and the widest to the SLRH model.
REFERENCES
Schukken YH, Grohn YT, McDermott B, McDermott JJ. Analysis of correlated discrete observations: background, examples and solutions. Prev Vet Med 2003;59:223-240.
2.McDermott JJ, Schukken YH. A review of methods used to adjust for cluster effects in explanatory epidemiological studies of animal populations. Prev Vet Med 1994;18: 155-173.
3.McDermott JJ, Schukken YH, Shoukri MM. Study design and analytic methods for data collected from clusters of animals. Prev Vet Med 1994;18: 175-191.
4.Curtis CR, Mauritsen RH, Kass PH, Salman MD, Erb HN. Ordinary versus random-effects logistic regression for analyzing herd-level calf morbidity and mortality data. Prev Vet Med 1993;16:207-22.
5.Solorio-Rivera JL, Rodriguez-Vivas RI, Perez-Gutierrez E, Wagner G. Management factors associated with Babesia bovis seroprevalence in cattle from eastern Yucatan, Mexico. Prev Vet Med 1999;40: 261-269.
6.Riveriego FJ, Moreno MA, Dominguez L. Risk factors for brucellosis seroprevalence of sheep and goat fl ocks in Spain. Prev Vet Med 2000;44:167-173.
7.Al-Talafhah AH, Lafi SQ, Al-Tarazi Y. Epidemiology of ovine brucellosis in Awassi sheep in Northern Jordan. Prev Vet Med 2003;60:297-306.
8.Steel RGD, Torrie JH. Principles and Procedures of Statistics. A Biometrical Approach. 2nd ed. New York: McGraw-Hill Book Company. 1980.
9.Solorio-Rivera JL, Segura Correa JC, Sanchez-Gil LG. Seroprevalence of antibodies and risk factors for brucellosis of goats in the Bajio region of Michoacan, Mexico. Prev Vet Med 2007, 82 (In press).
10.Cochran Ch, Coull B, Patel A. EGRET Users Manual for Windows (Version 2.0.3) Seattle WA: Cytel Software Corporation. 1999.
11.SAS. SAS/STAT User’s Guide (Version 8.1) Cary NC,USA: SAS Inst. Inc. 2000.
12.EGRET for Windows (Version 2.0.3) Seattle WA: Cytel Software Corporation. 1999.
13.Breslow NE, Day NE. Statistical Methods in Cancer Research. Vol. I. The analysis of Case-Control Studies.International Agency for Research on Cancer. Lyon,France: Scientifi c Publications No. 32, 1980.
14.Zhao L, Chen Y, Schaffner DW. Comparison of logistic regression and lineal regression in modeling percentage data. Appl Environ Microbiol 2001; 67: 2129-2135.
15.Hosmer DW, Lemeshow S. Applied Logistic Regression.New York: Wiley, 1989.
16.Segura-Correa JC, Alzina-López A, Solorio-Rivera JL.Evaluación de tres modelos y factores de riesgo asociados a la mortalidad de lechones al nacimiento en el trópico de México. Téc Pecu Méx 2007; 45:227-236.
17.Noordhulzen JPTM, Frankena K, Van der Hoofd CM,Graat EAM. Application of Quantitative Methods in Veterinary Epidemiology. Wageningen: Wageningen Pers, 1997.
18.McDermott JJ, Kadohira M, O´Callaghan CJ, Shoukri MM. A comparison of different models for assessing variation in the sero-prevalence of infectious bovine rhinotracheitis by farm, area and district in Kenya. Prev Vet Med 1997;32: 219-234.
19.Bendixen PH, Vilson B, Ekesbo I, Astrand DB. Disease frequencies of tied zero-grazed dairy cows and of dairy cows on pasture during summer and tied during winter. Prev Vet Med 1986; 4: 291-306.
20.Condon J, Kelly G, Bradshow B, Leonard N. Estimation of infection prevalence from correlated binomial samples. Prev Vet Med 2004;64: 1-14.
Full text