Varela Study

From SourceWatch
Jump to navigation Jump to search

This stub is a work-in-progress by the journalists's group. We are indexing the millions of documents stored at the San Francisco Uni's Legacy Tobacco Archive [1] With some entries you'll need to go to this site and type into the Search panel a (multi-digit) Bates number. You can search on names for other documents also.     Send any corrections or additions to


This article is part of the Tobacco portal on Sourcewatch funded from 2006 - 2009 by the American Legacy Foundation.

Luis R Varela's 1987 doctoral thesis 'Assessment of the association between passive smoking and lung cancer' was supposedly discovered in the files of this dead Yale student (he died in February 1990). It was a long-term study that had not been finalised or peer-reviewed.

However, the collection of data for this incomplete study had begun in 1980 (by Dwight Janerich) using National Cancer Institute contacts, and the preliminary material used in the dissertation by Varela had proved to be inconclusive. However, from the tobacco industry viewpoint, it had not demonstrated any strong linkage between passive smoking and lung cancer among either the wives of smokers or their co-workers. The numbers investigated were statistically relatively low (439 case-controlled families only -- 191 cases of primary lung-cancer) -- but they could still claim it to be the largest study of the kind every done, and they could exploit the NCI connection.

This was, of course, completely unjustified. The part of the study done by Varela was relatively low in numbers, and the NCI had had no part in the study itself. THIS IS HOWEVER A CASE STUDY IN HOW CORPORATIONS MANIPULATE INCONSEQUENTIAL FINDINGS, AND HOW THE PUBLIC RELATIONS INDUSTRY DISTORTS AND EXPLOITS THEM.

The Hirayama Study of non-smoking Japanese wives

The tobacco industry came to hear about this study at a time they were under considerable pressure -- following a couple or studies (mainly Hirayama) which had clearly demonstrated a link between passive smoking and health consequences. The Japanese scientist Hirayama had calculated that non-smoking wives of smokers had much higher rates of lung-cancer than normal. This was particularly relevant since the older Japanese wives, in the decades before this time, had cooked meals over burning braziers inside their houses and apartments (very little air-conditioning) and so it could be expected that cigarette smoke was only one of the household's normal load of similar air pollutants. As a consequence, therefore, cigarettes could only elevate the lung-cancer rates to the levels Hirayama detected if their second-hand smoke was particularly toxic.

Since Varela was not around to object, the tobacco industry (mainly Philip Morris's own issues scientist Thomas Borelli) tried to promote his study as proof that ETS had no adverse health effects on the spouses of smokers, as a counter to the Hirayama study. Philip Morris passed the idea over to the Tobacco institute, who did nothing with it -- then they decided to use it themselves; maintaining the line that the Varela study had been deliberately suppressed by health activists and the medical establishment. (It had taken a year to pass through the publication process, mainly because it required revisions)

Janerich's refutation

Varela's Yale academic supervisor Dr Dwight T Janerich had begun the study before Varela came to Yale. He had passed his data to Varela. Now he re-examined Varela's thesis and published a rebuttal of tobacco industry claims. He released an extension of study with another 191 patients who had been diagnosed with lung cancer from 1982 through 1984. Thes patients had either never smoked, or had smoked at one time but not more than 100 cigarettes in the decade prior to their diagnosis. They were compared to an equal number of people without lung cancer who had never smoked: with the aim of getting an estimate of the relative effect of passive smoking in the aetiology of lung cancer.

The researchers calculated the smoke load on individuals by the number of years lived in a house, multiplied by the number of smokers (calculated smoker/years of exposure). The researchers found that household exposure of 25 or more smoker/years during childhood and adolescence doubled the risk of lung cancer. This confirmed the EPA figures for passive-smoking illnesses among non-smokers-- and particularly, the problems with children.

The estimate from the study suggests that 1,700 cases of lung cancer each year result from childhood exposure to secondhand smoke, according to figures from the EPA. The agency said that of the 150,000 new cases of lung cancer diagnosed each year, about 25,000 involve people who are not currently smokers and about 10,000 who never smoked. (New York Times)

[Note that some of the documents refer to the full study as the Janerich Study, and others also use this term for the much smaller Varela Study.

Janerich had also started a study of 400 never-smokers in New York State when he worked for the New York State Department of Health, and it was continued with eight colleagues from the National Cancer Institute, Memorial Sloan-Kettering Cancer Center, Albany Medical College and the University of Southern Maine. [2] See published final study [3]

Documents & Timeline

1990 Mar 23 The NPL has now joined forces with PR firm Edelman to devise a plan to communicate the Varela Study to the public, media and politicians in order to counter the EPA's ETS Risk Assessment. This spells out an elaborate plan in 11 pages. [4]

1990 Apr 4 Note from Newman Partners Ltd to Tom Borreli on how to used the Varela Study.

RE: Variations on Varela
SITUATION: It is still desireable to bring the Varela Study to the attention of the scientific world and the general public because * Varela shows that the ETS-Lung Cancer issue is not resolved.

  • Varela offers comfort to smokers and PM employees who have been the target of the anti-ETS campaign.
  • Varela's advisor (Janerich) is about to publish "reworked data" alleging that ETS is linked to lung cancer. (They had plans to limit his criticism, since Yale had given Varela his PhD) Janerich had actually collected the basic data himself before handing it to Varela.
  • The EPA Risk Assessment is about to say that the ETS-Lung Cancer issue is resolved and ETS is killing 3000+ people a year.

They want to bypass the normal media processes and give their version of the story to only Lawrence Altman of the New York Times. If they get a negative reaction they will dump the Varela Study story; it i is positive, they will run the Times story as an advertisement claiming that the story was being suppressed by the health zealots and the media. There is already a plan for the Tobacco Institute to utilize the Varela story. It should be linked to George L Carlo's 'bias study' which will 'prove' that the media is biased against tobacco good news (although it wasn't then finished)


1990 Mar 29 Comments on the Varela dissertation; The paper is being circulated through the Scientific Issues division of Philip Morris, and checked by its lawyers Shook Hardy & Bacon. They list a couple of ways the study can be challenged. [6] The communications strategy is already being organised: [7]

1990 March 19 Tom Borelli gets a memo from the Newman Partners Ltd. public relations company

NPL RE: Varela news conference

  1. Announce existence and results of largest study ever done on potential health risks of environmental tobacco smoke (ETS).
  2. Pre-empt or counteract EPA Risk Assessment ...etc


1990 Mar /E This announcement was followed by a Public Education Program on Varela claiming that the medical establishment was deliberately withholding evidence that ETS (Environmental Tobacco Smoke) was not a health risk.

  • Varela is a major, significant ETS study which has been buried in the file drawer since it was completed in 1987. It runs counter to the general trend. The file drawer problem and publication bias are issues which affect all areas of scientific research ... [9]
  • STRATEGY OUTLINE: The thrust of our efforts following announcement of the Varela study will be to demonstrate the limitations of using epidemiology to develop a "fixed-number" death toll for exposure to Substance X. It is not possible to specifically ... [10]