Mortality in relation to smoking: 50 years' observations on male British doctors
BMJ 2004; 328 doi: https://doi.org/10.1136/bmj.38142.554479.AE (Published 24 June 2004) Cite this as: BMJ 2004;328:1519
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Could someone tell me if my analysis of the BDS lung cancer
statistics is correct or incorrect? My e-mail is jjosephson@shaw.ca.
Table 1 reports that 2.49 current male smokers out of "1000 men/year"
die of lung cancer. My interpretation, therefore, is that 997.51 smokers
per 1000 men/year do not contract lung cancer.
In terms of percentages, this suggests to me that 99.751% of current
male smokers per year do not get lung cancer.
Am I correct in concluding that 99.751% of smokers never contract
lung cancer? Or, does the "per year" of the "1000 men/year" make a
difference?
If so, can someone tell me precisely what percentage of smokers
actually develop lung cancer?
Thanks and sincerely, M.I. (Joe) Josephson, Ph.D.
Competing interests:
None declared
Competing interests: No competing interests
I find this 10 year average difference difficult to believe,
especially since Cuyler Hammond's earlier study--and no-control-of-
variables study--purported a much smaller approximate 5 year average
difference.
I'm also concerned about the use of the word "about" in "about 10
years." Does this mean 9 years or 8 years or 11 years or 12 years? This
does not sound very scientific.
Therefore, if 25,346 doctors died during the 50 years of this study,
why didn't Doll et al report the actual average ages at death of the non-
smokers vs. the smokers?
For example, if the non-smokers lived to be on average 81.5 years and
the smokers lived to be on average 71.4 years, then why not publish these
figures in the study and say, precisely, that "cigarette smokers die 10.1
years younger than non-smokers"?
Moreover, even though I've read the article several times, I can't
find any table to substantiate the sweeping statement that there is an
"average" loss of life of 10 years, a loss that applies to all 25,346 now-
deceased doctors. Have I missed something?
Competing interests:
None declared
Competing interests: No competing interests
Good grief!, Dr Glantz advocates elimination of public smoking to
reduce teen's exposure to Big Tobacco's efforts to retain them as
customers. Right in his own backyard, www.lao.ca.gov/cup0396.html,
Legislative Analyst's office, Mar 1996 the State of California found
despite tax hikes, teen smoking increased. Of course, many public smoking
bans had also already been passed by various legislative authorities
within the state which, according to Dr Glantz should also have cut teen
smoking.
Same thing in the USA overall, despite the price hike due to the
Tobacco Settlement, and various excise tax increases, and many public
smoking bans, it is a well known fact teen smoking increased up to
approximately 1999.
Education, and respect for authority ( laws that don't allow teens to
smoke) are the only ways to cut teen smoking. Try to use the force of law
to cut teen smoking results in rebellion, and actually hikes teen smoking.
Lets not use this paper as a platform to promote more persecution of
adults in the name of cutting teen smoking.
Competing interests:
None declared
Competing interests: No competing interests
In my previous response I suggested that with the large odds ratio(RR=24) that smoking has on Lung Cancer (LC), that it would be as difficult to disprove smoking as a causative factor. However, Silva’s(1) recently published analysis of asthma and Chronic Obstructive Pulmonary Disease (COPD) found that when asthma was included in the model, the relative risk for smokings' influence on COPD dropped from the usual large ratio of 6-12 reported in previous studies to a still significant but meager 3.
From Santillan’s(2) meta-analysis, asthma is also an independent risk factor(RR=1.8) for LC. And while I still suspect that smoking is still a considerable causative risk factor, we would indeed need to include asthma (as well as the stress of ACE’s) in the model in order to accurately assess the true effect, and of course, the population attributable risk (PAR).
Jay R. Schrand
References:
Silva GE, Sherrill DL, Guerra S, Barbee RA.
Asthma as a Risk Factor for COPD in a Longitudinal Study.
Chest. 2004 Jul;126(1):59-65.
Santillan AA, Camargo CA Jr, Colditz GA.
A meta-analysis of asthma and risk of lung cancer (United States).
Cancer Causes Control. 2003 May;14(4):327-34.
Competing interests:
: The author is a Systems Engineer, Independent Researcher, and Veteran, and has no financial interest in the tobacco, food or pharmaceutical industries other than as a consumer of their products.
Competing interests: No competing interests
Sir John Crofton's picture on the cover of 26 June issue reminds me
of his lecture in Japan. He attended the 6th World Conference on Smoking
and Health held in Tokyo in 1987. I was deeply impressed with the anti-
smoking activities in the United Kingdom. British doctors got to know the
hazards of cigarette smoking by their own findings. Sixteen years later in
2003, as a member of the committee on smoking problems in the Japanese
Respiratory Society, I could take part in making a declaration of
opposition to smoking[1]. The Japanese Respiratory Society has drawn up a
set of principles and guidelines on the elimination of smoking. We aim to
reduce the number of not only smoking-related disease cases but also
premature deaths in Japan.
Hiroshi Kawane
The Japanese Red Cross Hiroshima College of Nursing
Reference
[1]The Japanese Respiratory Society. Declaration of Opposition to Smoking.
http://www.jrs.or.jp/citizen/topics/sengen_e.html(accessed 14 July)
Competing interests:
None declared
Competing interests: No competing interests
It would be a mistake to believe that stopping smoking at age 30
prevents most excess deaths from smoking. The Doll study did not include
the effects of parental smoking on the doctors' children. It is well
established that tobacco smoking increases the risk of unexplained
infertility, recurrent miscarriages, premature births, pre-eclampsia and
stillbirths. Damage to the health of the future generations may be the
most harmful and important effect of smoking.
Competing interests:
None declared
Competing interests: No competing interests
The new findings from the UK doctors study are of great importance.
There is one aspect of the findings where there are grounds for caution in
generalising to the current population of smokers. That is the finding
that there was no net loss of life in those who quit smoking before 30 (25
-34). The average age of initiation in the doctors study is reported as
18. Depending on how age of initiation is defined, this is probably about
3 years older than is the case today for many countries. 1,2 If it is
years smoked that is critical to when years of life begin to be lost, then
the critical age will be lower. Further, if there is potential for
smoking to do greater damage when it occurs during the growth phase of
adolescence, then there might be an even greater reduction in the age to
no damage. Research is needed on health outcomes as a result of age of
uptake and of years smoked before firm conclusions can be reached about
the size of any relative safe period for smoking. Safe in this context,
is if the person quits, something the addictive nature of nicotine makes
problematic even for those who may start out with such intentions.
References:
1 Hill D, Borland R. Adults' accounts of Onset of Regular Smoking:
Influences of School, Work, and Other Settings. Public Health Reports
1991; 106(2): 181-186
2 Giovino GA Epidemiology of tobacco use among US adolescents. Nicotine
and Tobacco Research 1999, 1,S31-S40
Competing interests:
None declared
Competing interests: No competing interests
I recognize the courageous efforts of the distinguished Sir Richard Doll, in persisting against such insurmountable odds. After 50 years, smoking should be considered a major causative factor in Lung Cancer(LC)(1). Theoretically there is always a possibility for an unknown confounder. However, to dispute the large odds ratio(RR=24) would be as difficult as trying to prove causation for ETS with its’ low odds ratio(RR=1.2) and numerous potential confounders.
I would question why the mortality risk increased and the differential in survival rates for smokers VS never smokers increased from 7 to 10? Traumatic and chronic stress have a serious effect on the human system. Anda et al report that tobacco initiation and use has a strong and graded relationship with the number of Adverse Childhood Experiences (ACE)(2); being exposed to the stress of child abuse or coming from a dysfunctional family. Nicotine reduces the effects of stress. From the same 1996 survey (Dube et al table 4)(3), it appears that the risk between smoking and ACE's $ 3 increased in the latest 1962-1978 relative to previous birth cohorts. While not yet significant, this suspicious trend may indicate that those exposed to ACE's are recently under even more stress, which encourages tobacco use in the first place.
Stress does not appear to influence the risk for LC incidence. However, the stigma that patients report(4) certainly does effect treatment, survival and eventually mortality. The helping professions(5) and the military/veteran populations have an attraction for those with ACE's, and are now exposed to occupational sleep disruption and stress. In the past few decades, it is likely that Physicians who smoke, encountered enormous stress from their peers, especially if still in practice. How much professional courtesy, much less empathy, is a physician with LC going to receive from his peers? This recent stigma may explain the large increase in the percentage of surveyed physicians in this study who requested no further questionnaires after 1991(1). Yes, tobacco is a causative factor in LC. But, how much of the mortality, especially in the healing and peacekeeping occupations is due to childhood, adult or tobacco control program related stress? Neither the increased mortality risk nor differential in survival rates in Physicians may generalize to the general population.
Who is responsible and to blame? In 1998, the average age of death from LC was 71 compared to 78(6) for all causes. In 1900 life expectancy was 47 years(7). It takes around 70 pack/years to develop LC. It was not apparent until the 1930's when advances in medicine and health care increased life-span. It would be tempting to blame health care, as patients tend to do. However, Physicians healing talents in particular, improve life. Yes, the tobacco companies' aggressive marketing has increased consumption of this legal product. However, there is no indication that they introduced a specific cancerous agent or process that increased this particular risk. It's likely been there all along. Those who smoke are easy targets. However, they are simply using a 2000 year old product that, as compulsive as it is, has utility.
But, just who owns the problem? Prevention only deters the easy ones, and is a dismal failure. It is health care’s advancement in treatment that has made the greatest improvements in other areas. However, the amount of money spent on research for treatment of LC has been a paltry $1,740 compared to the $4-13,000 per death for designer cancers. If the money wasted on prevention was spent on honest treatment research, LC would not be a problem. It would also help the 13% of LC patients who do not smoke. We all own the problem and need to work together.
Pudgy former smokers produced by the tobacco control efforts, and are now targets of the anti-obesity campaign. Can the program to prevent the scourge of poor penmanship be far behind? One would think that Physicians who enjoyed smoking for 50 years would be especially interested in more research for better LC treatment. Medice cura te ipsum.
Jay R. Schrand
Port Heuneme, CA
schrand@fcc.net
References:
1. Doll R, Peto R, Boreham J, Sutherland I.
Mortality in relation to smoking: 50 years' observations on male British doctors.
BMJ 2004;328: 1519-33.
2. Anda RF, Croft JB, Felitti VJ, Nordenberg D, Giles WH, et al.
Adverse Childhood Experiences and Smoking During Adolescence and Adulthood
JAMA 1999;282:1652-1658
ACE Study: http://acestudy.org/
3. Dube SR, Felitti VJ, Dong M, Giles WH, Anda RF.
The impact of adverse childhood experiences on health problems: evidence from four birth cohorts dating back to 1900.
Prev Med. 2003 Sep;37(3):268-77.
4. Nuttall R, Jackson H
Personal history of childhood abuse among clinicians.
Child Abuse Negl 1994 May;18(5):455-72
5. Chapple A, Ziebland S, McPherson A.
Stigma, shame, and blame experienced by patients with lung cancer: qualitative study.
BMJ 2004; 328:1470-4
6. Computed from:
National Center for Health Statistics - DataWharehouse
GMWK I Total Deaths For Each Cause By 5-Year Age Groups, United States, 1998
http://www.cdc.gov/nchs/datawh/statab/unpubd/mortabs/gmwki.htm
7. U.S. Bureau of the Census.
Historical Statistics of the United States, Colonial Times to 1970
Competing interests:
The author is a Systems Engineer, Independent Researcher, and Veteran, and has no financial interest in the tobacco, food or pharmaceutical industries other than as a consumer of their products.
Competing interests: No competing interests
The study by Doll, et al (1) reinforces the fact that physicians and
public health officials should invest more energy in smoking cessation and
prevention for young adults because helping someone stop smoking before
age 30 avoids most of the long term damage. While most cessation efforts
are directed at people in middle age, the highest spontaneous quit rates
are actually among young adults (2). While people generally begin
experimenting with cigarettes as teenagers, it is during the period of
young adulthood that most people make the transition from experimentation
to consistent addicted cigarette consumption. This transition is
characterized by stopping and starting smoking, and the tobacco industry
invests considerable resources in creating a social environment to
recapture these transient quitters (2).
Cessation efforts directed at this group should capitalize on the
fact that many young adults transiently quit and seek to maintain their
smoke-free status. In addition to focusing clinical resources on this
subpopulation, public health interventions such as smoke-free pubs would
make it more difficult for the tobacco industry to recapture these victims
(3).
References
1. Doll R, Peto R, Boreham J, Sutherland I. Mortality in Relation to
Smoking: 50 Years' Observations on Male British Doctors. Bmj
2004;328(7455):1519.
2. Ling PM, Glantz SA. Tobacco Industry Research on Smoking
Cessation. Recapturing Young Adults and Other Recent Quitters. J Gen
Intern Med 2004;19(5 Pt 1):419-26.
3. Sepe E, Ling PM, Glantz SA. Smooth Moves: Bar and Nightclub
Tobacco Promotions That Target Young Adults. Am J Public Health
2002;92(3):414-9.
Competing interests:
None declared
Competing interests: No competing interests
This Study's Death Rate Stastistics
I absolutely do not believe this study's overall conclusion that the
average smoker will die "about 10 years" sooner than the average non-
smoker. I have corresponded with Doll/Peto asking them to provide me with
the actual average age of death of the non-smoking doctors in their study
vs. the smoking doctors and they have refused to provide me with that
information.
Why? I suspect that the average loss of life is about one year and
certainly not 10 years and I have to say also that the BDS is the most
biased and unscientific study I've ever read.
So whoever reads this, and I would urge the British Medical Journal
to do so, could you please urge the authors of the BDS to release the
information re actual death rates?
And I would in particular like to receive this information and my e-
mail address is jjosephson@shaw.ca.
Sincerely, Mundi Irving (Joe) Josephson, Ph.D.
Competing interests:
None declared
Competing interests: No competing interests