COMMENT ON "HEALTH EFFECTS OF EXPOSURE TO ENVIRONMENTAL TOBACCO SMOKE"

CHAPTER 7
Carcinogenic Effects

7.2 ETS and Lung Cancer

The case is absolutely not yet ready to be closed on the subject of whether ETS causes lung cancer in non-smokers. A review of the studies shows that the odds ratios for "previous lung disease" (by various definitions) generally exceed those for secondhand smoke, although results have been inconsistent. A recently-discovered pathogen, Chlamydia pneumoniae, has been linked not only to acute disease, pneumonia, but also to chronic illness, i.e., asthma and heart disease. It has a number of suspicion-provoking attributes besides the capacity to produce chronic infection. Unlike most bacteria, C pneumoniae multiplies intracellularly, like a virus. There is the possibility that this could cause genetic damage. Most CP infection is mild and may hardly be noticed by the patient. This would lead to underreporting, which could conceal a high odds ratio. Socioeconomic similarities between smokers and passive smokers, and their difference from non-smokers, in rates of CP infection would lead to spurious associations between passive smoking and lung cancer.

From my crude notes on NSLC associations with previous lung disease:

AKIBA- No data. BROWNSON (1987 Denver)- No data. BROWNSON -> ALAVANJA (Missouri 1992). (Layard said it was adjusted for previous lung disease.) "There were 12 cases who had a diagnosis of nonmalignant lung disease within 3 years preceding the diagnosis of lung cancer; these were not counted as cases with previous lung disease, because the nonmalignant disease might be related to early symptoms of lung cancer." Also 18 controls. p 627 lifetime nonsmokers n=432 no prev lung dis n=88 ASTHMA 2.7 (1.4-5.4) chronic bronchitis 1.4 (0.7-2.5) emphysema 0.7 (0.1-5.5) pleurisy 1.3 (0.8-2.3) PNEUMONIA 1.5 (1.0-2.1) tuberculosis 1.0 (0.2-4.5) ANY PREVIOUS L DIS 1.4 (1.0-2.0) p 628 Table 5 percentage of attributable risk, lifetime nonsmokers, previous lung disease 13% (1-26%).

Review of the literature - "inconclusive" ASTHMA Robinette & Fraumeni, J Chron Dis 1977: sig assoc asthma, ? smoking bias Reynolds & Kaplan; Vena, Am J Epidemiol 1987: asthma +lc in s & ns males but not females. Osann, Cancer Res 1991: LC neg assoc with asthma, but they included hay fever Alderson, Lancet 1974: neg assoc, did not adj for smoking Ford, Ann Allergy 1978: not adj for smoking PNEUMONIA Gao YT, Int J Cancer 1987: + assoc Wu AH, JNCI 1985: + assoc

"Since pneumonia is as prevalent as all other nonmalignant respiratory diseases combined in this study, it is estimated to account for half the attributable risk of lung cancer associated with previous lung disease." EMPHYSEMA & BRONCHITIS Gao YT, Int J Cancer 1987; Wu, Cancer Res 1988; Wu-Wil, Br J Cancer 1990: pos assoc.

BUFFLER (Texas)- no data on lung disease. CHAN (Hong Kong, Br J Cancer 1979)- no data on lung disease CAPEWELL (Scotland, Thorax 1991)- no data on lung disease CORREA (Louisiana, Lancet 1983)- no data on lung disease DALAGER (pooled from Buffler TX, Correa LA, Ziegler NJ)- no data on lung disease. FONTHAM (5 study centers, JAMA 1994)- "Previous lung disease and [dietary variables] were also evaluated, but were not included in the final models because they did not contribute further to the association between ETS and lung cancer." ???? weaken the association? GAO (Shanghai, Int J Cancer 1987)- Previous lung diseases none cases 418 1.0 ref TUBERCULOSIS 80 1.7 (1.1-2.4) PNEUMONIA 65 1.9 (1.2-3.0) EMPHYSEMA 37 2.0 (1.0-3.7) ch bron no emp 112 1.2 (0.8-1.7) others 30 1.3 (0.7-2.2) (adjusted for smoking according to text) "emphysema and chronic bronchitis were associated only with squamous and oat cell types" > Non-malignant lung diseases GARFINKEL (JNCI 1981)- no data on lung disease. Layard said ps was adjusted for GARFINKEL (Prev Med 1984)- no data on lung disease. GARFINKEL (JNCI 1985)- no data on lung disease. GENG (Tianjin, Smoking & Health 1987)- no data on lung disease. HE (Xuan Wei County, Yunnan Province, Environ Health Perspect 1991)- Family and personal medical history was taken. Table 6 males- history of chronic bronchitis 9.66 p>0.01 56 cases Table 7 females- history of chronic bronchitis 7.78 p>0.01 54 cases adjusted for smoking HIRAYAMA (BMJ Jan 17 1981)- passive smoking RRs HIRAYAMA (BMJ Oct 3 1981 correspondence)- no data on lung disease HIRAYAMA (Lancet Dec 17 1983)- no data on lung disease HIRAYAMA (Prev Med 1984) cohort- no data on lung disease HIRAYAMA (Lung Cancer: Causes and Prevention) again claim of increased death rates from emphysema & chronic bronchitis; no data on previous lung disease & lc.