Baselines cannot define effects

We saw in Chapter 3 that they cannot. The effect of a treatment is the difference between what happens when treatment is given and what would have happened had treatment been denied. In controlled clinical trials we choose to measure differences between the effects of treatments the effect of giving a verum rather than a placebo. Difference from baseline cannot be used to define an effect. This does not mean that it cannot be used to measure an effect. It can be so used if we strongly believe that the baseline value is the value we would have seen had treatment been denied. However, the very fact that we feel it is necessary to carry out a clinical trial means that we lack this confidence. 

The U.S. Clean Air Amendments of 1977 define two kinds of air quality standards primary standards, levels that will protect health but not necessarily prevent the other adverse effects of air pollution, and secondary standards, levels that will prevent all the other adverse effects of air pollution (Table 22-7). The amendments also define air quality levels that cannot be exceeded in specified geographic areas for "prevention of significant deterioration" (PSD) of the air of those areas. Although they are called "increments" over "baseline air quality" in the law, they are in effect tertiary standards, which are set at lower ambient levels than either the primary or secondary standards (Table 22-8). 

---