Validation of parameters in HACCP veri®cation using univariate and multivariate statistics. Application to the ®nal phases of poultry meat production M.L. Gonz alez-Miret a , M.T. Coello b , S. Alonso a , F.J. Heredia a, * a  Area de Nutrici on y Bromatolog  õa, Facultad de Farmacia, Universidad de Sevilla, Sevilla, Spain b Dpto. Metodolog  õa de las Ciencias del Comportamiento, Facultad de Psicolog  õa, Universidad Complutense, Madrid, Spain Received 24 November 1999; received in revised form 1 December 2000; accepted 14 December 2000 Abstract For Hazard Analysis and Critical Control Points HACCP) systems veri®cation in food industries, the selection of parameters that oer signi®cant information about control system at each production stages is necessary. Total Count TC), Pseudomonas PS), Enterobacteriaceae EB) and Staphylococcus aureus ST) are organisms frequently measured on carcasses at slaughters. Their in- clusion in HACCP system can be assessed by means of statistical techniques. Univariate and multivariate statistical methods are of great importance in order to identify most useful variables. In this study the use of these microbiological parameters in veri®cation of refrigeration, and cutup and packaging operations has been evaluated. It has established the usefulness of these variables using statistical techniques. It makes clear the most discriminant eect of EB and TC to establish dierences between the stages studied in this process. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: HACCP veri®cation; Multivariate statistics; Poultry 1. Introduction In designing the Hazard Analysis and Critical Con- trol Points HACCP) system in food industry, one of the most important facts to be concerned about is the proper selection of veri®cation parameters which must warrant the maximum information about control pro- cedures at each and every one of the phases of the production line ICMSF, 1991; Mortimore & Wallace, 1996). Frequent estimates of the number of several micro- organisms can be regarded as counts of indicator or- ganisms. In the production of poultry carcasses, the skin is not removed so that the number of organisms on the skin is the re¯ection of the destruction of the organisms by scalding and of recontamination in the abattoir Grau, 1986). The number of aerobic mesophiles Total Count, TC) on carcasses is usually taken in sanitation during slaughter and dressing. In the stages after evisceration considerable increases in psychrotrophs occur. Spoilage of chilled meats is caused mainly by the growth and metabolic activities of psychrotrophic aerobic organisms such as Pseudomonas PS), the principal ¯ora that de- velops during chilled storage. There are two main groups of microorganisms that, if present in a high degree in food, do reveal a somewhat faulty handling of the process: Enterobacteriaceae EB) and Staphylococcus aureus ST). EB is a well-known index of faecal contamination in food that provides a good target to assure processing eciency at critical control points and conformity with the veri®cation process of HACCP Robach, 1996). Because of this one uses this parameter as index for veri®cation of food process. On the other hand, ST is a very heat sensitive bacteria and disinfectant-resistant. Some of their strains may produce heat-resistant enterotoxins which if present in food are very likely to produce intoxications upon ingestion of such foods. Enterotoxigenic Staphylococcus may be found in food, especially in the one of animal origin, either before reception in factory or being con- taminated during dierent processes. It is possible to set a growth control of staphylococcus on foods by an Food Control 12 2001) 261±268 www.elsevier.com/locate/foodcont * Corresponding author. Tel.: +34-95-4556495; fax: +34-95- 4557017. E-mail address: heredia@fafar.us.es F.J. Heredia). 0956-7135/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 9 5 6 - 7 1 3 5  0 1 ) 0 0 0 0 7 - X