Pasteurization and Public Health: A study of city milk supply and infant mortality in the US 1920 to 1930

Infant mortality rate is an indicator of public health and social welfare (Preston 1975; Haines 1985, 2011; Lee 1991). In the US and some European countries, urban infant mortality rates have been gradually declining since the middle 1800s. Several factors accounted for this trend, e.g. the economic growth, improved public sanitation and medical provisions, spread of child care knowledge, and better food and nutrition conditions (McKeown 1976; Preston and Haines 1991; Fogel 1994; Millward and Bell 1998; Cutler and Miller 2005; Lee 2007). Specifically, the improvement of city milk supply was emphasized a lot by some researchers, as adulteration and contamination were two fatal milk problem to infants in those years (Beaver 1973; Selitzer 1976; Meckel 1990; Block 1999; Atkins 2003; Lee 2007; Tunick 2009). Fortunately, milk adulteration was largely controlled after many cities introduced municipal pure milk regulations around 1890s (Law 2003). Another threat, the bacterial contamination, was also noticeably reduced with the spread of pasteurization in early 1900s. Thus, some researchers like Beaver (1973), Block (1999), Atkins (2003) and Lee (2007) argued the improvement in market milk quality was the critical factor led to decline in urban infant mortality rate. As an remarkable technical breakthrough in the dairy industry, the public health impacts of pasteurization have been an important topic in epidemiology, food science and nutrition, and public policy since early 1900s. However, there are two major limitations in prior studies. First, most of them were in a narrative nature. There were no persuasive quantitative methods to measure the number of infants were saved by the technology from serious milk diseases, such as diarrhea, typhoid fever and tuberculosis. Second, this question became more complicated because of the endogeneity problem. In history, income or economic condition was an important sources of endogeneity, since it decided both sides of the extent of pasteurized milk in market and the status of public health in cities. For example, Lee (2007) referred income as “a mediating factor in diarrheal infant mortality”. Even some literature noticed the problem, no effort has been made to separate the impacts of income in previous discussions. The share of pasteurized milk was related with income because certain costs arose in process of pasteurization and bottling. Moreover, pasteurized milk production also was related with the manufacture in cities. Even some of above literature mentioned the problem, no effort has been made to separate the impacts of income in previous discussions. This paper contributes to existing knowledge in three aspects. First, it focuses on the measurement of the public health impact of pasteurization instead of just narrating the historical facts. Consequently, some important data were first explored. For example, the US Public Health Service (US