The study was carried out from January 2017 to December 2018, and was performed in 36 dairy farms located in the province of Ragusa, Italy (36°55’48” N, 14°44’24” E and 515 m above sea level), where the climate is warm and temperate. A total of 4081 cows were enrolled in this study. The examined farms had a variable consistency of lactating cows belonging to the following breeds: Italian Friesian, Italian Brown, Red Pied Fleckvieh, Jersey and crossbreeds. The age ranged from 6 months to 12 years.
The farms were divided into two groups on the basis of different managements: Group 1, represented by 17 farms, had an extensive housing system; Group 2, represented by 19 farms, had an intensive housing system.
Group 1 cows were kept in a grazing area of about 5–7 ha, at least 10 hours a day. In these grazing areas, there are herbs typical of the Ragusa plateau, Carob trees, which act also as shelter, and large pools of water. These areas are bordered by stone walls about one and a half meters high, built with an ancient technique. The cows spent the rest of the day in an area with a barn that serves as a refuge (from heat in summer and cold in winter). These barns are generally close to the milking parlour. Artificial insemination was practiced in some farms while in others natural fertilization with bull was performed. They were fed from May to September with 10 kg of fodder, 15 kg of hay (vetch, oats and barley) and 15 kg of silage (corn or silo grass) on average; while, from October to April, they were fed with the same diet, except silage as the season allows a lusher pasture. Group 2 cows were kept in a stable with a surface area providing between 6 and 7 m2/head or as many usable cubicles according to the number of animals. They were fed with 8 kg of fodder, 14 kg of hay (vetch, oats and barley) and 23 kg of silage (corn or silo grass) on average. Each farm then had different supplements of 1 to 2 kg (soy, beet, sunflower, cotton, alfalfa). In both groups, water was available ad libitum and the milking routine included pre and post dipping.
The average daily milk production was 27 litres in Group 1 and 32 litres in Group 2. In bulk tank milk the average milk fat composition was 3.68% in Group 1 and 3.60% in Group 2, and the value of milk proteins was 3.50% in Group 1 and 3.40% in Group 2.
The protocol of this study was reviewed and approved in accordance with the standards recommended by the Guide for the Care and Use of Laboratory Animals and the Directive 2010/63/EU. This study did not involve experimental animals. The animals did not suffer as it was performed a single blood sample in the caudal vein and the compilation of a check list was based on the visual observation of the animals. The research was performed as part of the research project with grant number IZSSI RC 12/15 approved by the Ministry of Health. Informed consent was obtained from all the farmers involved, who joined the project voluntarily.
Animal Welfare Assessment
The method we used in this study is based on the analysis of two data groups: the first group refers to the assessment of the hazards occurring as a result of environmental conditions (facilities, equipment, management and microclimatic conditions); the second group refers to the assessment of the risks, with the concerned adverse effects (health consequences). Hazard assessment is performed using parameters divided into five areas, respectively: Area A - “Farm management and personnel”; Area B - “Facilities and equipment”; Area C - “Animal-based measures”, for carrying out the assessment of the risk and of the consequent negative effects on cattle; Area D - “Inspection of microclimatic environmental conditions and alarm systems”, in the event of serious negative events (e.g. fire); Area E - “Biosecurity”, to assess the level of prevention against the introduction and/or spread of infectious diseases in the cattle-shed.
The check-list used consists of 90 items both for extensive and intensive housing systems. Each item has three (negative, acceptable, positive) or two (negative and positive) choice options. Not all the inspection items have the same weight in determining the final welfare score: some assessments have a multiplication or division algorithm that increases or reduces the importance when determining the final welfare score. This protocol was applied by a trained veterinarian on each farm and each check-list, filled in all its parts, was placed on-line on the appropriate site created at the portal of the CReNBA, which issued a certificate of “Animal Welfare and Biosecurity Assessment” by assigning a score for each of the parameters and an overall score to each farm. The set of these evaluations was subsequently entered into an algorithm that returns a value expressed as a percentage (on a scale from 0 to 100), able to identify the general welfare conditions of the herd. The final calculation of the scores in the various areas and those of the overall welfare and biosecurity was not carried out by the assessor, but by a specific CReNBA software, available through the website http://benessereruminanti.izsler.it. After the final aggregation of measures an overall score was reached to assign the assessed farm to one of the four general welfare categories (“not classified”, “acceptable”, “enhanced” and “excellent”). Each assessment took 2–3 hours and was carried out around two hours after milking, between 10 and 11 am. Milking took approximately two hours in each farm. All farms used milking machines.
Blood Sample Collection
Blood samples from all animals present in the examined farms were collected after animal welfare assessments by coccygeal venipuncture. They were put into vacuum tubes (Vacuette™, Greiner Bio-One) with no anticoagulant additive and centrifuged at 3500 rpm for 10 minutes. The obtained sera were transferred into plastic tubes. These were analysed for the detection of Mycobacterium avium subsp. paratuberculosis, Chlamydiophila abortus, Neospora caninum, Bovine Herpesvirus specific antibodies IgB and IgE, Bovine Viral Diarrhea virus antibodies, using an indirect Enzyme Linked Immunosorbent Assay (ELISA), as per the manufacturer’s instructions (ID.Vet, Grabels, France). Each serum sample was tested in duplicate and the final results were read by a spectrophotometer, measuring the optical density (OD) at 450 nm.
The data collected from the check-list drawn up by the CReNBA and the laboratory assays were entered and stored in a Microsoft Excel spreadsheet, screened for proper coding and errors, and analysis was done.
The obtained data were expressed as Mean ± Standard Deviation (SD). They were analysed for normality by Shapiro-Wilk test and for homoscedasticity by Bartlett test. Unpaired t-Test was applied to assess differences in the studied parameters between the two experimental groups. P values < 0.05 were considered statistically significant. Moreover, Kendall’s Tau coefficient (T) was determined to investigate the relationship between the areas and the studied infectious diseases. Statistical analysis was performed using the STATISTICA software package (STATISTICA 7 Stat Software Inc., Tulsa, Oklahoma).