To the Editor.— The readers of the recent editorial "Cholesterol, Coronary Disease, and Cancer" by Richard J. Jones, MD (1981;245:2060), deserve to know more details about some studies that have led him to a warning with which he "reinforces doubts that we yet know enough about the consequences of lowering the serum cholesterol level to consider its reduction in the general population by either drug or dietary means." Dr Jones is citing the Los Angeles Veterans Administration study1 as evidence of an increased risk "of both respiratory and gastrointestinal malignant neoplasms in the treatment group" without mentioning the obvious explanation of the difference in cancer mortality between the diet and the control group. Twelve cancer cases were recorded in men who adhered less than 20% of the time to the diet (adherence, calculated from attendance records, was expressed as a percentage of the maximum number of meals that could