Skip to main content

Advertisement

Table 2 Summary of evidence extracted from two Surgeon General reports

From: Two Surgeon General's reports on smoking and cancer: a historical investigation of the practice of causal inference

1964 1982
  Lung Laryngeal Esophageal Bladder Esophageal Bladder
Consistency: CC 24 studies/24 positive 10 studies/10 positive 6 studies/6 positive 4 studies/4 positive 9 studies/All positive 10 studies/8 positive
Consistency: Cohort 7 Studies/7 positive 7 studies/4 positive, 3 neutral 7 studies/6 positive, 1 negative 7 studies/6 positive, 1 negative 7 studies/All positive 8 studies/All positive
Strength: CC 5.43 (95% CI: 4.82 – 6.11) 4.52 (95% CI: 3.69 – 5.53) 2.38 (95% CI: 1.62 – 3.49) 2.28 (95% CI: 1.79 – 2.89) 3.11 (95% CI: 2.54 – 3.81) 2.02 (95% CI: 1.74 – 2.34)
Strength: Cohort 10.8 (range: 4.9 – 20.2) 5.4 (range: 1.5 – 13.1) 3.4 (range: .7 – 6.6) 1.9 (range: 0.9 – 6.0) Range: 1.8 – 6.4 Prospective Studies: 1.4 – 2.9
Specificity "90% [of lung cancer] is associated with cigarette smoking."(184) "the number of disease in which the ratios remain significantly high ... is not so great as to cast serious doubt on the causal hypothesis."(185) Not stated Not stated Not stated "Specificity ... is evidenced by substantial differences in the mortality ratios ... for esophageal cancer compared to other smoking related cancers."(96) "lower order of ... specificity for bladder cancer than for cancers of the lung, larynx, oral cavity or esophagus suggests that factors other than smoking may be associated etiologically with bladder cancer."(108)
Temporality "The early exposure to tobacco smoke and late manifestation of lung cancer among smokers, seem, at least superficially, to fulfill this condition." (185) Not stated Not stated Not stated "The temporal relationship of the association is supported by the prospective studies" "In addition, there are histological data suggesting that smoking predates premalignant and malignant transformation." (97) "Evidence for the temporal relationship of the association is provided by the prospective studies" "Reliable histological studies of bladder epithelium in smokers compared with nonsmokers have not been reported."(110)
Dose Response "In almost every study for which data were adequate" a gradient was observed with amount of smoking, duration, age when started, ex smokers, and inhalation Gradient also observed with increased amounts of smoking and inhalation. "The parallelism with lung cancer, though not as complete because of a smaller amount of material, is remarkable." Pooled cohort data revealed gradient for heavy versus moderate smokers. Gradient with amount smoked observed in only two of seven case control studies. Pooled cohort data revealed gradient for heavy versus moderate smokers. Gradient with amount smoked observed in only two of four case control studies. Dose response observed in retrospective and prospective studies Modest dose response "however this relationship is not as strong as that noted between smoking and lung, laryngeal, oral, and esophageal cancers."(108)
Biologic Plausibility: Animal/Lab Studies Application of tobacco smoke or condensates to lungs or tracheobronchial tree in animals failed to induce lung tumors (except possibly in dogs) [106, 248, 206, 224, 205a, 273, 274, 275, 29, 289] "No attempts to induce carcinoma of the larynx by tobacco smoker or smoke condensates have been reported."(210) "No attempts to induce carcinoma of the esophagus by tobacco smoker or smoke condensates have been reported."(217) Three teams of investigators have studied bladder cancer in mice treated with tobacco tars, but findings were inconsistent and their significance unclear [177, 75, 295] "There is experimental evidence that benzo[a]pyrene is able to penetrate the cell membranes of the esophageal epithelium," producing papillomas and carcinoma.(101) Not stated
Biologic Plausibility: Human Pathology Histopathologic changes in lungs of smokers in a controlled blind study of 402 male patients Histopathologic changes in larynx of smokers(271) Not Stated Not Stated Examination of autopsy tissue from 1,268 men showed atypical cells much more common in smokers [see temporality above]
Biologic Plausibility: Localization of cancer "Localization of cancer in relation to type of smoking"(188) "Localization of lesions"(210) Not Stated Not States Not Stated Not Stated
Coherence Time Trends Sex Differential Urban-rural differences Socio-economic differential Time Trends Sex Differential "Mortality from esophageal cancer in the United States had shown a tendency to rise slightly among whites in the last 30 years."(218) "information is lacking for an intelligent discussion of the sex differential" "urban rural differential is virtually non-existent"(225) Sex differential Lower mortality rates in low smoking populations (Mormons and 7th Day Adventists) Lower risk in ex-smokers Smoking acts synergistically in combination with alcohol. Sex differential Lower mortality rates in low smoking populations (Mormons and 7th Day Adventists) Lower risk in ex-smokers Occupational exposures associated with bladder cancer
Causal? Yes Yes No No Yes No
  1. Numbers in parentheses are references to page numbers in the Surgeon General reports