Epidemiological Investigations Utilizing Industry Abattoir Data - a Study in Finishing Pigs

Routine post-mortem inspections are enforced by governments worldwide to maintain meat hygiene standards. Through these systems, several pathologies and abnormalities are detected. In parallel, the industries of a number of countries have developed their own abattoir-based health schemes. These programs provide an integrated system to optimise the post-mortem pathological detection and the reporting of abattoir health information which is widely disseminated to both participating producers and veterinarians. In Great Britain, two initiatives have been implemented by the pig industry – Wholesome Pigs Scotland (WPS) and BPEX Pig Health Scheme (BPHS). This thesis brought together existing information from the health schemes to identify disease determinants for major pig abattoir conditions. Enzootic pneumonia-like (EP-like) lesions, pleurisy and ascariosis (identified through the presence of milk spot livers) are the three most prevalent conditions reported by the British pig health schemes, and their presence is associated with reduction in performance traits. This thesis combined records (from 2005 to 2007) on EP-like lesion, pleurisy and milk spots from 129,819 slaughtered pigs with information from the (505) farms of origin in order to identify those production characteristics that may influence the prevalence of these lesions. Geographical location of the finishing unit appeared to be a statistically significant determinant for the presence of EP-like lesions, pleurisy and milk spots. Part-slatted floors also appeared as a potential risk factor for the presence of EP-like lesions and pleurisy, versus the use of solid floor with bedding which appeared protective. For milk spots, the use of bedding and/or outdoor production appeared as a risk factor. Time-series analyses were conducted to identify trends and the presence of seasonal variations for EP-like lesions and ascariosis. For visual analytics, the monthly prevalence for EP-like lesions (from July 2005 to June 2011) and for milk spots (from July 2005 to December 2010) were modelled using STL, a seasonal-trend decomposition method based on locally-weighted regression. Over 12,000 pigs in average were assessed per month across 12 pig abattoirs over the studied periods. A trend toward reduction in the prevalence of EP-like lesions during the first three years of BPHS, followed by an increasing trend, was identified with November and December appearing as higher seasonal peaks of the occurrence of EP-like lesions. A mild overall decrease in the prevalence of milk spots over time was identified as well as a seasonal variation which showed peaks in September and October. This thesis also determined how different abattoir conditions are interrelated between them by the application of multi-dimensional machine learning methodology to the records from 6,485 batches of slaughtered finishing pigs. The resulting network indicates that pathologies potentially associated with septicaemia (e.g. pericarditis, peritonitis) appear interrelated, suggesting on-going bacterial challenges by pathogens such as Haemophilus parasuis and Streptococcus suis. This thesis contributes to demonstrate the potential of combining animal health abattoir data to perform epidemiological analyses which may lead the industry to a better understanding of how the farm characteristics, geographic location and temporal distribution could influence the prevalence of EP-like lesions, pleurisy and ascariosis.