Main Menu            Health Library

SMOKING, CANCER and CARDIOVASCULAR DISEASE

Abstract:  Cigarette smoking increases the risk of heart and other cardiovascular disease by 2 to 3 times, can increase the risk of lung cancer by up to 50 times, and increases risk of nearly every type of cancer and other cause of premature death.  Risks from smoking develop slowly and usually become substantial only after about 30 pack years (as smoking 20 cigarettes per day for 30 years).  Health risk is related directly to years of smoking and to the the square root of number of cigarettes smoked, and thus the number smoked must be reduced from 40 to 10 per day or from 20 to 5 per day to halve smoking risk. Lower tar and filtering has had little if any benefit, but the amount of smoke inhaled can have a quite substantial effect on risk.  Risk for most diseases drops about 16% each year after smoking is stopped, but some risk can persist for up to 20 years for former heavy smokers. Cigar and pipe smokers have a much smaller increase in risk than do cigarette smokers.   The vast research available on smoking provides basic information on the mechanisms by which carcinogens produce cancer that has been widely overlooked by researchers.

The #1 Research Study of Smoking:  The premier and now classic research data on smoking were obtained by Hammond on more than a million men and women in the 1960’s. This study was published in various papers, and specifically in a National Cancer Institute Monograph #19.  This data set appears unique in its scope.  It identified all-cause deaths, deaths from lung cancer and most other causes of cancer, and deaths from cardiovascular and other diseases.  Further, it related these deaths to amounts of cigarettes and other types of smoking, to length of time cigarettes were smoked, to amount of smoke inhaled, to age of individuals and age smoking started, to time after smoking was stopped, and all this for both men and women.  No other data obtained since challenges the scope of or conclusions obtainable from this massive study.  Rather most other studies of smoking confirm or extend its conclusions somewhat. During the 1980's the NCI produced  another study of a million men and women. The results not only affirmed those of the first study but showed an even larger risk for smoking than that measured earlier.

More than 50 thousand research studies and hundreds of population studies have been published on cigarette smoking.  This research has  invariably shown that risks of smoking on the key cardiovascular diseases and cancer are so high and so consistently confirmed that people who have any serious interest in health and life simply should not smoke. Perhaps because this has  been so evident, most researchers have been content to publish only the simple risk ratios found confirming that it is 'bad'.  Little study appears to have been done about how smoking develops cancer and other diseases at differing rates from causative mechanisms during life. More detailed and quantitative studies of smoking risks were developed in the Life Ahead project for two reasons. 

First, US and other populations today include varying proportions of smokers that smoke differing numbers of cigarettes at each different age. The vast available data bases of disease and death from diseases suffered at age, and from which popular results such of life expectancy are derived are really not valid for any individual.  This is because these data bases really result from a confounded combination of the risks of smokers and non-smokers in the populations studied.  A set of these key disease and life expectancy rates for a population free of smoking is needed to assess the risk or outlook of a non-smoking person - and today this includes most of those in the US.  And a data base for individuals that smoke some specific amount is needed to assess the real long range risk for a smoker. These much needed results have not been available.  And any serious model of disease and health progression needs these more useful results for a any valid projection of future health and likely length of life.  To obtain an assessment of the true disease risks of a smoke-free population we must recognize the varying proportions of smokers in the population at each age, and have accurate risk ratios for the effect of smoking on every disease involved at every age.  The true smoke-free disease and death risks of some diseases at age can be a third or less of those now reported by the NIH for today's actual total US populations

Second, a Life Ahead program objective is to produce state-of-art projections of an individual's mean actual risks from all combinations of known health factors. These more incisive analyses can suggest that the true risks of smokers that can develop when in combination with other risk factors can be vastly higher than would be expected from conventional health risk factor advice. Thus we need to know actual risks of smoking for every involved disease for each age and time and amount of smoking, and how smoking risks combine with those from other factors.

The Life Ahead Analysis:  The Hammond data were a key basis for the Life Ahead model of cigarette and other smoking extent and duration.  A first analysis was made of the results for all causes of death because the data on this were most extensive.  Hammond provided a total of 110 different risk ratios on cigarettes vs all causes of death for men and women for different amounts smoked, for various smoking ages of starting, stopping, and durations, for amounts inhaled and at differing ages.  Many more risk ratios were included for deaths from Lung Cancer, Coronary Disease, and other causes. These data was first studied using dozens of stepwise multiple regressions to identify the probable causal relationships involved and how these causes operate to cause premature death.

A primary cause of death from cigarettes was the product of number of cigarettes smoking and the time they were smoked.  This has widely been confirmed by researchers as the “Pack-Years” effect of smoking. (A pack is 20 cigarettes). The problem in using a crude ratio such such a "Pack-Years" is that the ratio assumes that the effect of each percentage increase in amount smoked is duplicated exactly by a similar change in time smoked. This is not true.  Rather, risk related with best significance to the 0.5 power or square root of number of cigarettes smoked times the length of time smoked.   It takes a reduction from 2 packs to one-half pack per day of smoking to reduce risk in half.  Or to halve risk, smoking must be reduced from 20 cigarettes per day to 5 per day. But smoking risk is quite directly related to time or years of smoking.  In fact there may even be an induction period of some years before risk initiates. For example, in the large NCI study of the 1980's, smokers experienced 1781 lung cancer deaths. Only one of these deaths occurred for men during a first 23 years of smoking. 260 deaths occurred in the 5 year period after 40 years of smoking. Measured risk ratios increased from near unity in earliest years to 38 during the age 50's.

The Formula for Cigarettes and Death from All Causes:   A more accurate function of "Pack-Years" thus becomes No of Cigarettes or packs smoked to the 0.5 power times years of smoking.  All other smoking factors must in return related to this quite basic function because at a zero level of either numbers of cigarettes smoked or number of years smoked the risk ratio of smoking for any disease or other combination of factors must become unity, and risk from smoking per se becomes zero.  Conventional textbook regression or statistical methods cannot easily value this type of a relationship together with other factors, and thus stepwise  methods of analysis were required. 

Because of this complexity, most of the risk factors for smoking identified in various population research studies have only very narrow validity.  True risks of smoking for every disease are changing continually with time of and amount of smoking and user age.  The formula type finally developed for relating risk of death to cigarettes is appended. The 'factor' in this formula will depend on the disease involved, the amount inhaled, age, gender, and any other risk determining factor. Actually measured average smoking risks vary from near zero to more than fifty  times for differing diseases, times of smoking and age during life.  Determining a most probable and accurate future risk of smoking requires use of a life-cycle model such as Life Ahead that can produce a sum of the differing risks that occur at each age of life for each, and that can sum these risks for  individual diseases. 

       Another confounding factor in determining a true risk of smoking is that its basic risk increases with time or years of smoking, and decreases with an individual's actual age.  Although actual overall risk of cancer increases steadily with age, the actual rate of progress of most cancer declines modestly with increase in age.  For cancer and most other diseases this decline in rate is about 1.5% per year of age. Thus there is a serious confounding of time of smoking with user age because although users can start smoking at different ages, time of regular smoking usually increases with age during life. Although the effect of years of smoking is usually much larger than the offsetting effect of age,  the true effect of years of smoking  must be determined from individuals of a given age that have smoked differing numbers of years. Fortunately there is sufficient research data reported on this in the above first major study of cigarettes and cancer to accomplish this, and verify the above relationship of modified pack-years. Most other studies did not include this depth of needed data.

       A surprisingly large effect on the risk from cigarettes was found for answers to simple questions about how much smoke is inhaled.  Although answers to such a non-quantified subjective measure would be expected to be diffuse, those citing lowest amount of inhaling obtained risks to 50% lower than those citing substantial inhaling. This means that the inhale habits of a smoker may be far more important than the amount of tar or use of filters, or within limits, the exact number of cigarettes smoked.  Women typically report lower amounts of inhaling than do men.

The Risk of Lung Cancer from Cigarettes:   By far the highest risk for smoking is for Lung Cancer.  The 1960's study revealed risks for average male smokers of up to 15 times higher at age 65.  The 1980's study showed substantially higher risks, with risks for average smokers up to an incredible 38 times at  lower age 55.  The added risk for lung cancer for average middle age women smokers had moved up to 17 times from less than 4 times in 1960.  Smoking risks were even higher for men smoking more than their average of about 25 cigarettes per day and for women smoking more than their average of about 20 per day.

The average amount of tar in cigarettes had declined from 25mg in 1960 to about 14 mg in 1985. This plus extensive use of filters had led to widespread assumptions that this should have reduced smoking risk.  How - in spite of this - could this much higher actual risk for smokers have occurred?  Many explanations have been advanced. People tend to smoke more and inhale more when tar and nicotine is reduced.  Women also did smoke several more cigarettes on average per day in and had smoked longer in the 1980's than in the 1960's.  And men did smoke a bit more and a had smoked somewhat longer in the 1980's.  Yet adjustment for these known differences still fails to explain the higher smoking risk of the 1980's and that probably remains true today. One explanation might be that smokers inhale the equivalent of a cup of tar each year even from from the reduced tar, and even the lowered amounts were more than sufficient to produce the potential increase in risk. Another reason probably is that despite study estimates, most smokers in 1960 and especially women had not then smoked long enough to obtain the full impact of long term cigarette risk. Another factor: In the later years nearly everyone was exposed to the cigarette smoking of others far more than they were in earlier years.  And they lived in this smoke-heavy environment for many more years than did those in earlier years. The passive smoking of others not only produces disease in non-smokers but increases the risk of everyone that smokes.

Risk of Cardiovascular Disease and Death from Cigarettes:  Coronary Heart disease produces far more premature deaths than does lung cancer.  Smoking men experienced about twice the number of Coronary deaths as did non-smoking men in the 1960's, and two and a half times non-smoker deaths in the 1980's.  Women's coronary death rate was about two times in the 1960's and up to about 3 times in the 1980's.  Again, risks for heart disease and stroke were higher in the 80's despite the lowered tar and use of filters. 

The Risk for Breast Cancer from Cigarettes:  There long has been controversy about the effect of cigarettes on breast cancer.  Most studies published prior to about 1990 showed little or no effect of smoking on the risk of breast cancer.  Other more recent studies (Palmer, Am J Epidemiol 134:1-13 and Calle, Am J Epidemiology 139:1001) found significant risks for smoking in the range of 1.7-1.8 times.  A Swiss study (Morabia, J Epidemiology 143:918) found even larger risks. A brief review of 15 studies of smoking and breast cancer identified a probable reason for this discrepancy.

Cigarette smoking requires a long duration of time before cancer develops.  And the risk of smoking develops as a function of pack years of smoking. For example, only a limited fraction of smokers develop lung cancer much before 40 pack years of smoking.  This is particularly true for the development of breast cancer that develops much more more slowly than does lung cancer.  Most breast cancer patients studied in research on smoking were in the age range of 30-50 years of age, and smoked about 15 cigarettes per day.  Assuming an average starting age of 17-20, this will give only about 15 pack years of exposure, a far from adequate exposure for a development of breast cancer.  It hardly is surprising that such studies would find little or no effect of smoking within their usual margin of error.  Further, some researchers failed even to recognize the key importance of pack years of smoking.  Attempting to relate risk separately to years smoked and to numbers of cigarettes smoked lowers even further the statistical power of the relationships.

Calle found a breast cancer risk ratio of 1.74 for the smoking of 40+ cigarettes per day for a probable 40 pack years of exposure.  Palmer found from each of two studies, one in the US and another in Canada, that smokers of more than 25 cigarettes per day that started before age 16 developed breast cancer risk ratios of 1.7(1.0-2.9) and 1.8(1.0-3.4) respectively.  The large Nurse’s study (London J Natl Cancer Inst 18:1625)  found no effect on breast cancer for an exposure of perhaps 30 pack years, but found a development at this time of tumors with a risk ratio of 1.38(1.04-1.84).   A formula for breast cancer used in Life Ahead explains the risk ratios of the 15 studies reviewed on breast cancer reasonably well within their margins of error.  Times and amounts of smoking are only roughly estimated by some of the research used in this analysis, and a more detailed and complete global analysis of smoking and breast cancer is needed.  This formula produces a conservative risk ratio of just 1.5 for 40 pack years of smoking, and is used only to a maximum risk ratio of 2.0.  The majority of women that experience breast cancer during their years before age 50 would not acquire a sufficient pack years of smoking for it to be a substantial contributor to their cancer.  But as they smoke for much longer times this can become a significant additional contributor to breast cancer risk in their later years and significantly reduce their potential Well Days of life

Risks for Other Causes of Disease from Smoking:  Most formulas for Life Ahead were derived initially from Hammond as modifications of formula (1) above, and updated to the 1980's study results.  Medline indexed research was searched and used for each other Life Ahead monitored disease as Prostrate Cancer, Female Genital Cancer, Colon Cancer, All Other Cancer, All Other Disease Death, Motor Vehicle Deaths, and Non-Disease Deaths.  Effects of smoking on genital, prostate, and colorectal cancer from Hammond and other research appear to be from nil to small.

A summary of effect of smoking on risk ratios of various disease causes of death for a man and woman of age 60 that has smoked 30 cigarettes per day from age 17 follows.  These ratios assume similar exposure to cigarettes for each gender.

Years of Smoking             Men         Women

Disease:

Death from All Causes        3.5          3.5                    

Coronary Heart Disease       3.0          5.2

Stroke                       3.8          7.1

Lung Cancer                  49           49              

Breast Cancer                 -           1.5

Prostate Cancer              1.4           -              

Genital Cancer (Women)        -           1.0

Colorectal Cancer            1.1          1.1

Other Cancer                 3.4          3.4

Motor Vehicle                1.2          1.2

The risks for both men women in this table are higher than those reported in research because the assumed 30 per day represents higher amounts of smoking than those on average researched.   The risks for women are computed for the same number of cigarettes, smoking years, and inhaling as for men. Actual measured risks for women that were lower were obtained for fewer years of smoking, fewer cigarettes per day, and lesser amounts of inhaling than those measured for men.  The higher risk ratios of women for cardiovascular disease were due to the much lower basic risk of non-smoking women than that for non-smoking men.  Actual overall cardiovascular risks for women smokers were lower than those for men smokers. Generally, lung and other cancer risks for women were similar to those for men when considered for same amounts of smoking. A curious anomaly was the confirmed research results showing that women's risk of endometrial cancer was reduced by smoking.  The average risk factor of 1.0 now used for women's genital cancer recognizes that smoking risk for other female cancer genital sites may be substantially positive.

Risks for Pipes or Cigar Smoking:   Smoking of pipes and cigars does create risks of both cancer  and heart disease, but at a  lesser level than that from cigarettes.  This probably is due to the much lower amounts of such smoke that is inhaled.  Life Ahead uses a simple factor of general pipe and/or cigar smoking as an added equivalent of 2.5 cigarettes per day.  Although this is a rough approximation only, the risk level here does not deserve much refinement.  The other factors as age and time of smoking apply similarly as for cigarettes. Keep in mind that computed risk  for 2.5 cigarettes per day is about half the risk of 10 cigarettes per day.

The Risk after Smoking is Stopped:  The risk of most diseases from smoking declines at about 16% per year after smoking is stopped.  After 1 year, the risk is 84% of that previous.  After 5 years the risk drops below half to 42% of that when smoking.  After 10 years only 18% of prior risk remains.  This risk from lung cancer drops slightly faster at 18% per year.  But because risk from this cause can become so high, some appreciable excess risk of lung cancer can remain 10 years after smoking is stopped.

Smoking has Declined Remarkably, but a Substantial Health Impact Continues: During the 1960's nearly 60% of all men smoked cigarettes. Every business meeting and restaurant reeked of smoke.  Nearly 80% of all men born in the 1920's smoked at one time. The now classic surgeon general's report in 1964 finally convinced many that smoking was harmful, and smoking started to decrease. As of 2002 about a quarter of all men and about 20% of women continued to smoke. Some most recent published results were:

      Percent  of Population and Amounts of Cigarette Smoking 

Even with only 20-30% of individuals still smoking, their increased risk of up 2.5 to 3 times in of heart disease produces a substantial contribution to the total heart disease deaths at most ages of our population. And the large majority of deaths from lung cancer and lung diseases still are obtained from those that still smoke.  Thus conclusions from our national vital statistics are distorted by this, and popular values of the life expectancy of individuals are seriously confused by this problem. The more useful and valid Life Ahead data bases of disease for non-smoking individuals required computer derived adjustments for the proportion, amount, extent, and duration of population smoking at every age for every disease.  This proved to be a time consuming but very worthwhile project. 

The Research on Cigarettes provides a Major Source of Information:  The available research on how cigarettes develop cancer provides an important information about cancer development. Surprisingly, there appears to have been little global study of this important data beyond its use in roughly verifying the health debits of smoking.  The research on cigarettes can tell us much more about cancer than the just the statistical risks of smoking.  This research provides quantitative insights on how cancer develops, how cancer develops over time from exposure to carcinogenic agents, and how its risk declines after a major carcinogen is eliminated.  These insights from cigarette research probably apply to the development of and regression of cancer from diet and other causative agents.  

Particularly important is the long induction period usually needed to produce human cancer.  This means that the 3 to 5 year clinical studies revered for obtaining health information on medicines and other health factors can be useless for measuring the effect of a factor on cancer. Many researchers today appear unaware of this problem.  Such clinical studies probably would fail to find any effect of cigarettes on cancer during this time period.  Life Ahead shows that duration of exposure can be a factor a key importance to risk level of most health factors that can affect healthy human life, and especially those that involve vitamins and antioxidants.  

Smoking produces tars that include carcinogens that contact directly the mouth, throat, and lungs, and digestive system.  The section on cancer notes that the effect of carcinogens is reduced by frequent washing that does occur in the digestive tract.  But this washing is largely absent in the lungs, and carcinogens that gradually coat lung tissue create the greatest increase in risk at this site.  The breast and genital organs are not so directly contacted by tar, and smoking produces a much lower risk at these sites.

Smoking increases risk of Cardiovascular Diseases (CVD)  by increasing the rate at which atherosclerosis forms. (See the section on Atherosclerosis).  This clogs arteries faster and appears due to the fact that smoking creates a pro-oxidant environment that increases the deleterious effect of LDL cholesterol..  Thus smoking risk may be related to the level of LDL.  This interaction with LDL has been confirmed for the effect of C-reactive protein (CRP), and has yet to be studied for smoking. Also, smoking can increase the clotting tendency of blood that can cause heart attacks.

The slow decline of risk after smoking is stopped is important from the standpoint of basic science.  This demonstrates that even after a complete removal of a specific carcinogen  - or after removal of an agent that increases atherosclerosis – the risk to health of this agent does not immediately disappear. Why does this happen?   Conventional risk factors and most health risk models assume in serious error that risk is immediately associated with a risk factor.  Rather, risks of both cancer and cardiovascular diseases gradually build up over long periods of time and then gradually move down during the decade or two following removal of this factor.  This behavior is specifically and perhaps now uniquely quantified in Life Ahead via its life cycle computations for both cardiovascular diseases and cancer.

Estimating Risks of Cancer using Life Ahead:  The present Life Ahead program produces the risk of each of 19 causes of major disease and death, and death from all causes from any amount, duration of smoking, and years stopped for either men or women having all types of other risks.  A striking new finding from the program is how cigarette smoking can combine with other risks, as lack of exercise, poor diet, high cholesterol, or family risks. Combinations of any one of these other risks with smoking multiplies its risk, and a combination of smoking with two or more other risks often can predictably produce those suddenly fatal heart attack that occur now so surprisingly during the low 40's or earlier.  Some can survive smoking if the other risks are low, but few will survive to older age if smoking combines with other major risks.

To view smoking risks, the benefit of stopping it, and how risks can be reduced via improving other health factors download the free Life Ahead computer program at Download Life Ahead.  The Demo Program will not value lifetime smoking because it is set up to identify risks for a non-smoking person at present age and that change habits at this age.  Thus the Demo will not value the effect of lifetime smoking from early age. To value the effect of lifetime smoking, enter a name, sex, and complete all possible entries in the top or first program option to enter Wellness factors. You do not need a diet entry and do not have to complete all entries shown because the program will assume average population values for factors not entered. (You do have to complete the exercise entries, however).  Once complete, you can explore the improvements in risk of diseases and Well-Days of future life for starting smoking at any age, for stopping smoking, for reducing its amount, or for changing any other contributing health habit. You can change risks of stopping smoking starting at present age by accessing the 'Change Weight or Smoking' option from the result screen.  But to explore the effect of changing smoking before present age you must revise the smoking entry at the third wellness factor entry screen. The specific state-of-art risks of smoking estimated from Life Ahead may be different and far more serious than those estimated just from general advice on smoking risks.

 _____________________________________________________________________________________

Formula Method for Valuing Smoking Risk::

      (1)  factor = exp(c1  + c2 *age + c5*gender + c3 * (inhale factor - c4 )      

            Risk of death from disease =  Exp (factor * (cigyrs ^ 0.5 *years smoked)

         where c1 and c2 depend on disease, c3 is a constant usually 0.20, c4 is average inhale of 2.74 inhale index, c5 is factor for gender,, male=0; female=1.  c1 and  

            c2 are normally negative coefficients with lower risk for higher negative values. C5 is positive only for CVD diseases. Otherwise it becomes zero.