Smoking and Cancer of the Lung. By Joseph Berkson. Proc Mayo Clin 35:367 (June 22) 1960.
The apparent unity of Australian medical scientists about the evils of smoking was shattered in November, 1962, when Dr. Charles Mayo of Rochester, Minn, said at a farewell press conference in Brisbane that he did not believe smoking played any part in causing lung cancer. A few days later in an excellent letter to the Melbourne Age, Mr. P. D. Finch, a University statistician, showed the extreme vulnerability of a local medical proclamation in relying on statistical tables to prove that smoking causes cancer.
Whereas the collection of data such as statistics, says Finch, is the province of medicine, the prime responsibility in the analysis of the statistics lies with the statistician. Further, a distinction must be made between the nature of the evidence required to convince a scientist that there is an association