Book Description
This book analyses patterns in rural China in the late 1980s: patterns of causes of death, of what people ate, what they smoked and drank, what kinds of houses they lived in, what they worked at, their education, and many measurements of their blood (for cholesterol, vitamins, evidence of infectious disease) and urine (for food metabolites and other factors). The variation is examined at the level of counties scattered all over mainland China and Taiwan, representing the extremes
of values for deaths from specific cancers; ie the counties with the highest and the lowest rates of lung cancer, or the highest and lowest for liver cancer. Coincidentally, this covers the extremes of many of the other variables, such as the intake of fresh fruits and vegetables, and smoking rates. The analysis that fills the pages is the correlation of all of these patterns, one variable at a time, with all the others. The question it answers is, "How well does the variation among the
counties for one variable (eg cholesterol in the blood) correlate with the variation across China in deaths from different diseases (eg heart disease)?". If the correlation is strong, it may mean that the variables are related in some causal sense, although this cannot be assumed. If the correlation is weak, it means that the variation must be caused mainly by other factors. Importantly, if the correlation is weak, it does not necessarily mean that the two variables are not related; for
example, a weak correlation between blood cholesterol and deaths from heart disease does not mean that cholesterol is not implicated in heart disease, but that in China other factors are more important. Each variable page is similarly arranged, and there are keys to interpreting each element at the beginning of major sections.
of values for deaths from specific cancers; ie the counties with the highest and the lowest rates of lung cancer, or the highest and lowest for liver cancer. Coincidentally, this covers the extremes of many of the other variables, such as the intake of fresh fruits and vegetables, and smoking rates. The analysis that fills the pages is the correlation of all of these patterns, one variable at a time, with all the others. The question it answers is, "How well does the variation among the
counties for one variable (eg cholesterol in the blood) correlate with the variation across China in deaths from different diseases (eg heart disease)?". If the correlation is strong, it may mean that the variables are related in some causal sense, although this cannot be assumed. If the correlation is weak, it means that the variation must be caused mainly by other factors. Importantly, if the correlation is weak, it does not necessarily mean that the two variables are not related; for
example, a weak correlation between blood cholesterol and deaths from heart disease does not mean that cholesterol is not implicated in heart disease, but that in China other factors are more important. Each variable page is similarly arranged, and there are keys to interpreting each element at the beginning of major sections.
The book also includes numerous extra tables in the back that give mean values for many variables. These can be useful as many of these values in China are so different from the much more available and common Western values. We tend to think of the range of Western variables as somehow 'normal', without realizing that in China the mean value may not even be within the generally accepted normal range that we are used to.
