S James Kilpatrick

From SourceWatch
(Redirected from S. James Kilpatrick)
Jump to: navigation, search

Also known as Jim Kilpatrick or James Kilpatrick Jr. He was a Professor of Biostatistics at Virginia Commonwealth University Medical School who worked for the Tobacco Institute, and wrote a series of reports for the tobacco law firm Covington & Burling. From 1965 to 1983 he was the Chairman of the Department of Biostatistics at the Medical College of Virginia. [1]

He is also listed as a member of the Indoor Air Pollution Advisory Group (IAPAG) [2]. IAPAG was the American version of the European Whitecoats consulting groups. In the USA, it was run by Sorrell Schwartz and Nancy Balter out of Georgetown University (within a division known as CEHHT), and controlled by John Rupp at Covington & Burling.

Hirayama report

In 1980, Takeshi Hirayama, chief of epidemiology at Tokyo's National Cancer Centre Research Institute in Japan produced a devastating report on the incidence of lung-cancer among the non-smoking wives of Japanese male smokers. This was a very large, very robust study (he tracked almost 100,000 non-smoking women for 14 years), which settled any lingering doubts that Environmental Tobacco Smoke (ETS) could have a long-term impact on the health of those who were forced into passive smoking by their physical environment. [3] It was effectively the 'smoking gun' in the politics of second-hand smoke.

Hirayama followed his initial study up with further reports in 1982, 1984 and 1987 (active smoking only), which confirmed his original findings, and added to the robustness of the claims. This series of reports created consternation in the tobacco industry and triggered a massive effort by cigarette companies and their trade associations around the world to attack both the man and his findings. This led, in turn, to the expansion of the Whitecoats program around the world and a major recruitment of consultants by the companies.

However it was only after his 1984 report and the report of the US Surgeon General at that time implicating passive smoking, that the media and politicians began to take the question seriously. In this third report Hirayama reinforced his original finding and demonstrated "a significantly increased risk of lung cancer mortality in relation to the extent of the husband's smoking". The association was significant when observed by age of husbands [] and also by age of wives."' [4]

John Viren, a statistician employed by RJ Reynolds Tobacco at this time, thought he had detected a fatal flaw (a "bombshell", so one executive claimed) in Hirayama's statistical methods: In HIrayama's primary data he had sometimes relied on an estimate of the age of the woman. Older Japanese women are often reluctant to give their age to a stranger, and so Hirayama had sometimes resorted to using the husbands age (this was usually given accurately) as a way to estimate the woman's age. Viren collaberated with Jim Kilpatrick on a publishable research paper that attempted to refute Hirayama's findings

They concluded that:

Our analysis shows that husband's age is not a surrogate for wife's age, in spite of its common use as such by Dr Hirayama. Since age at entry is confounded with cohort effects, it is also likely further that interactions occur between age and calendar period and between age and other covariates not represented in the model.

As a consequence of this finding, our conclusion is that the results of Hirayama's large study on ETS and lung cancer should not be pooled with other ETS/lung cancer studies to form a global estimate of relative risk. The relevance of this Japanese study's findings to public policy is questioned in

the absence of more detailed information on exposure and follow-up, and more extensive analysis of the data at the level of the individual subject.


Kilpatrick also billed Shook Hardy & Bacon for his work on preparing and giving paper at the London Imperial College Indoor Air Conference run in 1987 by Professor Roger Perry for the tobacco industry. [1] World Health Organization scientists pulled out at the last moment, when they realized the conference had been stacked in favor of the tobacco lobby.

In 1988 Mary Ward of RJR wrote to Marvin Kastenbaum the staff statistician at the Tobacco Institute asking him to check Verin re-analysis of Hirayama's work. The Board of the new Center for Indoor Air Research was considering funding a statistical expert, Dr. Michael E. Ginevan of Environ Corporation to confirm and to refine John Viren's analysis, ""with a view of co-authoring a publication with John. Ginevan's work would include sensitivity analyses to determine whether there are any logical reasons for the severe mortality deficits John has uncovered. before it was used." [2]

Not long after, Kirkpatrick published another attempt at refuting Hirayama under the title: '"Model Specification Effects in ETS/Nutrition Research". The draft copy of the report in the files of RJ Reynolds lists the author's address at the Medical College of Virginia and has a marginal hand note "Send proofs to this address." [3]

This paper also credits John Viren (who was still working for RJ Reynolds) and Dr Michael E Ginevan of Environ Corporation who was contracted ($60,000) to the CIAR [4]

EPA risk assessment

In 1990 the Environmental Protection Agency announced that it was conducting a risk-assessment of passive smoking, and asked for scientific input. James Kilpatrick was one of the scientific witnesses paid by the Tobacco Institute to submit. His efforts [[5]] were totally directed at attacking the statitical basis of Hirayama's work, even though statisticians employed by the tobacco industry had long agreed that his methodology was unassailable.


  1. Curriculum Vitae for S. James Kilpatrick: From The Legacy Tobacco Documents Library, Accessed April 6th, 2012
  2. IAPGA's Allocation of Responsibilities for TI Scientific Witness Project: From the Legacy Tobacco Documents Library, Accessed April 6th, 2012
  3. Hirayama T: "Non-smoking wives of heavy smokers have a higher risk of lung cancer" a study from Japan. British Medical Journal, 1981, 282: 183–185.
  4. Hirayama T: "Lung Cancer in Japan: Effects of Nutrition and Smoking" Originally published in: "Lung Cancer, Causes and Prevention"
  5. S.James Kilpatrick & John Viren: "Age As a Modifying Factor in the Association Between Lung Cancer in Nonsmoking Women and their Husbands' Smoking Status" Medical College of Virginia, Richmond, Virginia 23298-00032, and R.J.Reynolds Tobacco Company, Winston-Salem, N .C. 27102. From The Legacy Tobacco Documents Library.