"Walnuts add a happy crunch to life, and they may even help prolong it" 
NYTimes 28 Sept 1994

"Effects of walnuts on serum lipid levels and blood pressure in normal 
men" The New England Journal of Medicine, Vol.328 No.9, pp.603-607

1a. The newspaper artucle appeared in the New York Times Spet. 25, 1994.  
However, the study was reported in the New England Journal of Medicine on 
March 4, 1993.  Why was this 1.5 year old study the topic of the 
newspaper article?

b. Where was the study conducted? Why is Dennis Balint quoted in the 
newspaper article rather than Dr. Joan Sabate?

2. Who are the subjects of the experiment?  What are the treatments?  And 
what is the outcome measurements?

3. The experiment is described as "controlled, single-blind, randomized, 
crossover design."
        Can you explain what each of these descriptive terms means in 
relation to this particular study?  For example, the experiment was 
controlled because the investigations assigned the subjects to treatment 
groups, rather than allowing the subjects to choose which treatment they 
were to receive.

4. How did the investigators ensure that the subjects adhered to the 
treatment?

5. This is a paired experiment, because the subjects received both 
treatments.  A big advantage to pairing is that it reduces the 
variability in comparing the two treatments.  Rather than looking at the 
difference between the average fatty-acid for two groups of subjects, we 
look at the average of the differences for each subject.
        Table 4 presents mean total cholesterol for each of the diets.  
The +/- figures are the standard deviations for the subjects.  Use these 
SD's to show the advantage of pairing.  That is make a 95% confidence 
interval for the difference, ignoring the fact that the data is paired.  
Compare it to the 95% confidence interval presented in Table 4.

6. A possible difficulty with the pairing is called the "carry-over 
effect."  Find where this is discussed in the report and explain what 
this effect means.

7. Explain why figure 1 shows a clear cross-over pattern.

8.  Should this study affect Public Health?  If you were a Public Health 
official, how would you react?