American Cancer Society & CDC Withheld Evidence on Safer Smokeless Products

The Centers for Disease Control (CDC) and the American Cancer Society (ACS) have a long history of suppressing information about the relative risks of smokeless tobacco (ST) products. (CDC example here)

CDC and ACS researchers in 2007 jointly published a mortality study of smokers who had switched to ST (here).  Using data from the ACS 2^nd Cancer Prevention Survey (CPS-II), they reported that smokers who switched to ST had slightly higher risks than smokers who became abstinent.  However, the authors omitted a critically important comparison group: smokers who continued smoking. 

The ACS and the CDC had always avoided direct comparison of smokers with ST users because they knew that ST users incurred vastly lower risks.  This time, their omission was glaring.   

In February 2007, I filed a federal Freedom of Information Act request for access to the ACS dataset.  The CDC denied my request seven months later, claiming that the CDC authors had not had access to the data.

A month later, in October 2007, I filed an appeal with the HHS Deputy Assistant Secretary for Public Affairs, raising two issues:

• CDC co-authors would not willingly violate the Uniform Requirements for Manuscripts Submitted to Biomedical Journals, a mandate that authorship be based on substantial contribution to the conception and design, acquisition of data, or analysis and interpretation of data.  My position was that CDC authors were effectively co-owners of the data.

• There was incontrovertible evidence that the CDC possessed or had access to the CPS-II Dataset.  I noted, “The CDC maintains the SAMMEC website, which ‘calculates annual smoking-attributable deaths, years of potential life lost, smoking-attributable expenditures, and productivity losses for adults in the United States, individual states, and user-defined populations.’  The methodology section of SAMMEC states that it ‘uses unpublished age-adjusted RR estimates for persons aged 35 years and older from the second wave of the American Cancer Society's Cancer Prevention Study (CPS-II), 6-year follow-up.’”

More than two years later, in June 2010, HHS denied my appeal, asserting, “These primary data are in the possession of the ACS. CDC did not have any role in collecting or defining these data and never had possession of it. ACS never sent CDC its CPS-II primary data and it never allowed CDC to access the data or to submit computer programs to access the data. The ACS never showed any intent to relinquish control over the data.”

For 30 years, the CDC has been the source for smoking-related U.S. and global mortality estimates (discussed here).  Critical measures used to generate these numbers are the relative risk estimates from CPS-II, which are apparently the private property of the ACS and which have never been validated by external experts, even those at the CDC. 

I wrote ACS CEO John Seffrin in 2010: “I urge the ACS to release comparable risk estimates for male smokers, smokeless tobacco users and switchers in CPS-II…In addition, the ACS should release the CPS-II data, so that the risk estimates for all tobacco users can be validated by external scientists.”  I pointed out, “Earlier this year, Lawrence Deyton, director of the FDA Center for Tobacco Products, Principal Deputy Commissioner Joshua Sharfstein and Commissioner Margaret Hamburg authored a commentary about tobacco regulation for the New England Journal of Medicine, in which smoking and tobacco use were used synonymously (here). These physicians wrote that ‘tobacco use causes more than 400,000 deaths in the United States annually…,’ and they described ‘tobacco products’ as ‘the leading cause of preventable death in the United States.’”

I explained, “When FDA officials characterize all tobacco products as equally lethal, they are effectively denying smokers lifesaving information about safer cigarette substitutes…The public health is not well served when information that can save lives is buried and/or ignored. The selective and incomplete provision of risk information by the ACS is contrary to its public health mission and will only serve to hamper the FDA from undertaking an independent assessment of this issue in pursuit of its overarching mission: reducing the harm associated with tobacco use.”

ACS Chief Medical Officer Otis Brawley denied my information request, writing, “We have not found it productive to share the CPS-II data with researchers supported by the tobacco industry, because of their repeated misuse and misrepresentation of the data.”  Both letters were copied to Deyton, Sharfstein and Hamburg.

I described the ACS correspondence in my blog (here), where I noted, “I have conducted research on tobacco harm reduction for almost 20 years; published numerous studies in professional journals; written dozens of articles in the general media; and given hundreds of lectures. No one has ever accused me of misusing or misrepresenting any data.”

This is how the CDC and ACS have for years ignored and buried evidence documenting that smokeless tobacco use is vastly safer than smoking.

FDA Cigar Study Part II: Heart & Lung Disease Nearly Zero for 1-2 Cigars Per Day

As noted in my last blog post (here), data presented in a recent FDA review of cigar health risks (here) showed that smoking up to two cigars per day, while not completely safe, is neither associated with significantly increased risks for death from all causes, nor smoking-related cancers.       

Now we examine the FDA data regarding cigar-related circulatory and lung diseases. Here are the overall results:

Table 1. Relative Risks for Mortality From Circulatory Diseases and Emphysema Among Men Who Smoke Cigars
Disease	Study, Year	Relative Risk (95% CI)
Coronary heart disease	Kahn, 1966	1.04 (0.96 – 1.13)
	Carstensen, 1987	1.16 (0.84 – 1.57)
	Ben-Schlomo, 1994	0.45 (0.17 – 1.22)
	Shanks, 1998	1.05 (1.00 – 1.11)
	Jacobs, 1999	1.30 (1.05 – 1.62) 1
	Jacobs, 1999	0.93 (0.72 – 1.21) 2
Stroke	Kahn, 1966	1.08 (0.91 – 1.28)
	Shanks, 1998	0.96 (0.87 – 1.06)
Aortic Aneurysm	Kahn, 1966	2.06 (1.32 – 3.07)
	Carstensen, 1987	5.10 (1.33 – 13.19)
	Shanks, 1998	1.76 (1.29 – 2.35)
Atherosclerosis	Kahn, 1966	0.97 (0.69 – 1.33)
Emphysema	Kahn, 1966	0.79 (0.25 – 1.86)
	Carstensen, 1987	1.30 (0.00 – 7.45)
	Lange, 1992	3.70 (1.10 – 12.00)
	Shanks, 1998	1.42 (0.96 – 2.03)

¹Age 30-74 years

²Age 75+ years

    

Compared with never smokers, cigar smokers in most studies had no elevated risks for coronary heart disease, which is one of the most common smoking-related causes of death in the U.S.; the single exception was a subgroup of men, age 30-74 years, in the Jacobs study.  Cigar smokers also did not have increased risks for two other frequent causes of death, stroke and atherosclerosis (hardening of the arteries).

Aortic aneurysm – a bulge in the heart’s main artery – is the only disease risk that is consistently elevated in cigar smokers.  It is a serious disorder but a distinctly uncommon cause of death; the mortality rate due to aortic aneurysm among those 45 and older dropped precipitously from 16 deaths per 100,000 in 2000 to 7.4 in 2014.

For men who smoke only one or two cigars a day, the health risks are even lower.

Table 2. Relative Risks for Mortality From Circulatory Diseases and Emphysema Among Men Who Smoke 1 or 2 Cigars Per Day
Disease	Shanks, 1998	Other Studies
Coronary heart disease	0.98 (0.91 – 1.07)	1.00 (0.90 – 1.10)1
		1.18 (0.76 – 1.82)2
Stroke	1.01 (0.88 – 1.17)
Aortic Aneurysm	1.82 (1.11 – 2.81)
Emphysema	1.39 (0.74 – 2.38)

¹Kahn 1966, fewer than 5 cigars per day.

²Jacobs 1999, 1 cigar per day.

No elevated risks for coronary heart disease, stroke or emphysema among men smoking 1-2 cigars per day.  The only disease that was significantly elevated was aortic aneurysm.

The Take-Home Message for Cigar Smokers

Puffing and/or inhaling the smoke of burning tobacco is not without risks.

The FDA, which now regulates tobacco products, seems inclined to treat cigars the same as cigarettes.  FDA staff wrote in their cigar study that “…cigar smoking carries many of the same health risks as cigarette smoking…We have observed that some risks associated with cigar smoking can be as high or higher than those associated with cigarette smoking, especially at the highest doses and levels of inhalation for cigar smoking.”

The problem with such a sweeping indictment is that it ignores scientific evidence and misleads cigar smokers who could substantially benefit from truthful harm reduction guidance.

Like any risky behavior, the degree of risk is proportional to dose and duration of exposure.  In other words, risk is based on frequency of cigar smoking and the degree to which smoke is puffed and/or inhaled.

FDA Study: Cancer Risks Nearly Nil for 1-2 Cigars Per Day

The FDA, which now regulates cigars, has taken the position that “cigar smoking carries many of the same health risks as cigarette smoking.”  The agency is requiring cigar packages and ads to display six new warnings, including:

• Cigar smoking can cause cancers of the mouth and throat, even if you do not inhale.

• Cigar smoking can cause lung cancer and heart disease.

• Cigars are not a safe alternative to cigarettes.

However, an FDA staff report shows that smoking up to two cigars a day is associated with minimal significant health risks.

Last year, FDA staff, led by Cindy Chang, published in BMC Public Health (available here) a systematic review of the risks of cigar smoking, declaring that their action was “not a formal dissemination of information by the FDA and does not represent agency position or policy.” 

The authors reviewed 22 prospective epidemiologic studies on cigars and health outcomes; they produced tables of results for many smoking-related diseases. 

This entry will explore results for deaths from all causes and from cancers among men who are primary cigar smokers (no history of cigarette or pipe smoking).  Because the 22 studies employed different methods, and because they controlled or adjusted for various factors, the FDA authors did not provide overall summary risk estimates.

First, let’s look at mortality for all causes of death among cigar smokers, compared with never smokers.  (A relative risk, or RR, of 1 is no risk, and an RR confidence interval that includes 1 is not statistically significant.)

Table 1. Relative Risk Estimates For All-Cause Mortality Among Men Who Smoke Cigars
Publication, Year	Relative Risk (95% Confidence Interval, CI)
Best, 1966	1.06 (0.92 – 1.22)
Kahn, 1966	1.10 (1.05 – 1.16)
Cole, 1974	1.15 (0.70 – 1.90)
Carstensen, 1987	1.39 (1.16 – 1.65)
Lange, 1992	1.60 (1.30 – 2.00)
Ben-Schlomo, 1994	0.48 (0.25 – 0.93)
Shanks, 1998	1.08 (1.05 – 1.12)

Table 1 shows that cigar smokers have an elevated risk for death from all causes.  With the exception of the Ben-Schlomo study, all of the estimates are elevated, and four are statistically significant.  So it appears that there is a modest increase in risk among cigar smokers for all causes of death.     

Next, let’s look at the risks for individual cancers. 

Table 2. Relative Risks For Cancer Mortality Among Men Who Smoke Cigars
Cancer Site	Relative Risk Range	No. of Studies / No. of Significant Elevated Estimates
Mouth/throat	4.0 – 7.9	3 / 3
Esophagus	1.8 – 6.5	4 / 2
Stomach	1.2 – 2.3	2 / 1
Liver	3.1 – 7.2	2 / 2
Pancreas	1.0 – 1.6	4 / 1
Larynx	10.0 – 10.3	3 / 3
Lung	1.6 – 7.6	6 / 5
Bladder	0.9 – 1.9	4 / 0

It appears that, like cigarette smokers, cigar smokers have elevated cancer risks especially at places in contact with smoke, such as the mouth/throat, esophagus, larynx and lung.  For other sites, like the stomach, pancreas and bladder, elevated risks are minimal and/or based on very limited data.

In other research published by the FDA (here), primary cigar smokers consumed on average about 1 ½ cigars per day when they smoked.  This is relevant to a key epidemiology principle: the level of risk is related to the level of exposure.  Several studies in the Chang review reported results for smoking one or two cigars per day.  Here are those results: 

Table 3. Relative Risks (95% CI) for Mortality Among Men Smoking 1 or 2 Cigars Per Day
	Shanks, 1998	Shapiro, 2000	Other Studies
All Causes	1.02 (0.97 – 1.07)		1.04 (0.98 – 1.11)1
Cancer
Mouth/throat	2.12 (0.43 – 6.18)	0 (0.00 – 0.00)
Esophagus	2.28 (0.74 – 5.33)	1.80 (0.60 – 5.00)
Stomach			1.68 (0.95 – 2.97)2
Pancreas	1.18 (0.69 – 1.89)	0.60 (0.30 – 1.40)
Larynx	6.45 (0.72 – 23.3)	6.00 (0.70 – 53.5)
Lung	0.90 (0.54 – 1.66)	1.30 (0.70 – 2.40)	1.14 (0.59 – 2.00)1
Bladder	0.78 (0.29 – 1.71)	0 (0.00 – 0.00)

¹Kahn 1966, 5 cigars per day or fewer.

²Chao 2002, 5 cigars per day or fewer.

While some risk estimates are elevated, none are statistically significant.  The highest risks are for larynx cancer, but they are based on only two deaths in Shanks and one death in Shapiro.

What are the important take-home messages for cigar smokers with respect to risks for all causes of death and for smoking-related cancers?  FDA staff concluded that “cigar smoking carries many of the same health risks as cigarette smoking. Mortality risks from cigar smoking vary by level of exposure as measured by cigars per day and inhalation level [not discussed here] and can be as high as or exceed those of cigarette smoking.”

However, the data indicates that consumption of up to two cigars per day, while not completely safe, is neither associated with significantly increased risks for death from all causes, nor smoking-related cancers.        

In a subsequent post, I will present results for other diseases, including lung and cardiovascular disorders.