Homogeneous subject populations

Homogeneity of the subject population is an important factor in controlling variability. The more homogeneous the subject population generating the data, the more informative it is. Thus, fewer subjects are required to achieve the desired control... 

In addition to the greater risk in patients, results in patients are frequently confounded by the effects of disease, concomitant medication, age and other variables. By contrast, healthy subjects are much more homogeneous and subjects are studied under standardised conditions. It is sometimes argued that healthy volunteers are not representative of the patient population and therefore that the studies are of less... 

Figure 6.6 shows a generalized dose-response curve. Such a plot may be obtained, for example, by administering different doses of a poison in a uniform manner to a homogeneous population of test animals and plotting the cumulative percentage of deaths as a function of the log of the dose. The result is normally an S-shaped curve, as shown in Figure 6.6. The dose corresponding to the midpoint (inflection point) of such a curve is the statistical estimate of the dose that would cause death in 50% of the subjects and is designated as LD50. The estimated doses at which 5% (LD5)...

Population characteristics -h Measured in humans -h Large population size possible -h Full range of sensitive subpopulations possible + Measured in humans - Limited number of subjects + Possible to study sensitive subpopulations - Extrapolation to humans + Large group size possible - Homogeneity of animal model population and environmental factors-relevance to human ... 

Another example of distinct residual error models is when data from two different populations are studied. For example, data from healthy volunteers and subjects may be analyzed together. For whatever reason, possibly due to better control over dosing and sampling in healthy volunteers or because of inherent homogeneity in healthy volunteers, data from healthy volunteers have smaller residual variability than subjects. Hence, a residual error of the type... 

---