Raw milk quality assessment: the risk of multiple exceeding of the somatic cell count limits is greater than pure chance

A raw milk sample with more than 350,000 somatic cells per millilitre (SCC/ml) is considered to be substandard in Switzerland. In every case of non-conformity an investigation is carried out into how often the milk producer in question has exceeded the tolerance limits in the pre-vious five months. This acts as the criterion under private law for any price deductions and, in case of five incidences of non-conformity in a row, for imposition of the milk delivery ban. The expected values for single and multiple non-conformities, calculated according to the „same risk for all“ principle, differ markedly from the actual counts recorded. Non-conformities are therefore not chance events. The relatively small number of single non-conformities and the high number of multiple ones are an indication of the causal link with supplier-dependent factors. In 2001 in Switzerland, 13,015 instead of the expected 16,134 suppliers were counted with a single non-conformity after a previous period free of non-conformity of at least five inspections, that is only 81 % of the expected value. In contrast there were 65 % more suppliers than expected with two or more non-conformities, specifically 7,873 instead of 4,753. The quality statistics for the years 1997 to 2004 show similar conditions. The rates of non-conformity also depend on season of the year and geographical region. Increased somatic cell counts are registered especially in late summer. Non-conformities are generally less common in the Central Plateau and more frequent in certain rather more remote regions. If these variable factors are included in calculation of the expected frequencies, then several modifications arise. The risk of one non-conformity is dynamic and contains at least three significant components: a) the season, b) the region and above all c) the medium-term history of sample quality of the holding in question. Whilst with a) and b) improvements of a general nature are possible (general hygiene, cooling chain, logistics), as far as c) is concerned, specific weak points must be looked for in the affected holdings. The quality control of commercial milk, which takes account of previous history, is a successful tool for objective assessment and for identifying unfavourable situationsthe graduated weighting is selective, informative and effective.