Cigarette smoke is one of the most important environmental risk factors for the development of cervical intraepithelial neoplasms (CINs). Given that the prevalence of cigarette smoking in Taiwanese women is very low, compared with Caucasian women (3-4% vs. ~28%, respectively), direct smoking is likely a minimal risk factor in the Taiwanese group. However, the evaluation of the association between indirect tobacco exposure, or exposure to environmental tobacco smoke (ETS), and CIN risk in Taiwanese women may be enlightening. The authors designed a community-based nested case-control study to examine the association between ETS exposure and CIN risk among Taiwanese women who were nonsmokers. The study was conducted in Chia-Yi, a city in southwestern Taiwan. The test population comprised adult females (n = 32,466) who had undergone Pap smear screening during the 14-mo period that preceded this study. Potential cases were chosen from individuals who demonstrated positive Pap smear results, which were indicative of at least a level II cervical intraepithelial neoplasm (i.e., [greater than or equal to] CIN2), as confirmed by cervical biopsy (n = 116). The authors chose 2 matched controls for each case; criteria used were (a) the 2 controls for each case had to be approximately the same age ([+ or -] 2 yr) as the case; (b) each case and respective control lived in the same residential area; and (c) each control had a negative Pap smear, determined during the same time frame ([+ or -] 6 mo) that the matched study subject was tested. A total of 100 study subjects and 197 control females (3 cases had only 1 control) underwent interviews by public health nurses, who obtained information about the participants' active and passive smoking histories and other potential risk factors. The incidence of cigarette smoking was low (5.0% [n = 5]) in study subjects and lower in controls (3.1% [n = 6]). The authors' final analysis was restricted to 89 nonsmoking pairs (89 subjects and 175 controls). Of the affected subjects, 85.4% reported exposure to passive smoke at home during their adulthood, vs. 61.1% of controls. The authors found that, after adjusting for other covariates, there was a 2.73-fold increased risk of contracting CIN (95% confidence interval [CI] = 1.31, 5.67) among ETS-exposed individuals. The authors assessed cumulative dose in terms of pack-years (i.e., average daily exposure [number of packs smoked (multiplied by the number of years the same exposure continued). Compared with nonsmokers, women exposed to 1-20 pack-yr and more than 20 pack-yr had 1.90-fold (95% CI = 0.72, 5.03) and 2.99-fold (95% CI = 1.10, 8.09) increased risks, respectively, of developing CIN. The authors concluded that lifetime ETS exposure is a major determinant for contracting cervical neoplasms among nonsmoking women in Taiwan.
RECENTLY, CERVICAL CANCER has become the most frequently diagnosed malignancy in Taiwanese women (annual age-standardized morbidity rate: 23.8/100,000 females in the year 2000). (1) It is also the 6th leading cause of cancer mortality among women (annual age-standardized death rate: 9.0/100,000 females). Exfoliative cytology screening (i.e., Pap smears) can reveal high-grade squamous intraepithelial neoplasms, including moderate (cervical intraepithelium neoplasm II [CIN2]) and severe (CIN3) dysplasias. These neoplasms and dysplasias are precancerous cervical lesions. If left untreated, at least 25% of the dysplasias will progress to either in situ carcinoma or invasive cancer. (2)
Although cigarette smoking is 1 of the environmental risk factors most responsible for cervical neoplasms, the prevalence of cigarette smoking in Taiwanese women is low, compared with smoking in Caucasian women (34% vs. ~28%, respectively). (3-6) However, according to an annual nationwide survey of tobacco consumption in Taiwan, (7) cigarette smoking is very prevalent (55-62%) in Taiwanese men, a fact suggesting that many nonsmoking Taiwanese women may be exposed to environmental tobacco smoke (ETS). Therefore, to investigate the association between ETS exposure and CIN risk in Taiwanese women, we conducted a community-based nested case-control study.
Materials and Method
Study area. Chia-Yi, a city located on the southwestern coastal plain of Taiwan, is 15.8 km long and 10.5 km wide. The major types of industries in Chia-Yi are agriculture, business-related, and tourism and restaurant service-related; no large manufacturing or processing industries are located within the city. In June 2000, 92,019 women 19+ yr of age resided in Chia-Yi.
Participants. Between October 1999 and December 2000, 32,466 (35.3%) women 19+ yr of age underwent Pap smear screening in Chia-Yi. This cervical cancer screening program was supported by the Taiwan government in an effort to provide free testing to all women older than 19 yr who had ever had a sexual experience. It was determined that 420 of these women had newly diagnosed lesions that were category I or greater CINs, and 349 had follow-up biopsies. Of the remaining 71 women, 24 were followed-up by Pap smear and 3 by direct therapy, 5 were uncooperative, no information was available for 5, and 34 had other conditions. Of the 349 women biopsied, 116 had definite lesions that exceeded level II CINs. Women in this group served as potential subjects for this study.
We assigned 2 candidate controls for each study subject. Each control was selected randomly from women who matched each respective subject for 3 factors: (1) controls had negative Pap smears, which were determined at approximately the same time ([+ or -] 6 mo) as the matched study subject was tested, (2) controls lived in the same area of Chia-Yi (eastern vs. western administrative district) as study subjects, and (3) controls were age-matched ([+ or -] 2 yr) with the subjects. All previous Pap smear results were negative in all controls chosen; number of years tested, frequency of testing, and level of sexual activity were unknown.
Nineteen trained public health nurses interviewed our study subjects and controls, using a structured questionnaire, in their homes between October 2000 and March 2001. The duration of each interview was approximately 30 min. The 19 interviewers were not completely blind to the status of cases and controls. They knew that this activity was part of a health-promotional program; however, they did not know the main hypothesis of the present study. This study was approved by the Human Subjects Committee of the Kaohsiung Chang-Gung Memorial Hospital.
Questionnaire. Information sought through the questionnaires included demographic characteristics, education level, smoking status, history of ETS exposure (including dose), exposures to special x-ray examinations or to hair dyes, sexual and reproductive histories, date(s) on which previous cervical smears were taken, history of cooking tasks (e.g., age at which subject began cooking at home, number of years subject cooked, average hours per day spent in the kitchen, kinds of fuels used), and type of kitchen ventilation used.
A "smoker" was defined as a person who had smoked more than 1 cigarette each day for at least 1 yr (i.e., 365 consecutive days). Exposure to ETS was evaluated at home and in the workplace, as well as during 2 life periods: (1) childhood ([less than or equal to] 20 yr of age), and (2) adulthood (> 20 yr of age). We considered subjects as "passive smokers" if they had been exposed to smoke from more than 1 cigarette/day for at least 1 yr. (8,9) This same criterion was set for both the home and workplace.
If the subject had been exposed to ETS, a series of detailed follow-up questions were asked, including queries about the beginning and ending years of ETS exposure and total duration (i.e., number of years) of exposure. Lifetime dose of ETS exposure was summed for all periods and situations and was expressed in pack-years, which we calculated by multiplying the average number of packs of cigarettes smoked each day in the presence of the subject by the duration (i.e., years) of ETS exposure.
Statistical analysis. The distribution of demographic characteristics was examined by [chi square] statistics. Of the 100 study subjects and 197 controls (3 of the cases had only 1 control), 5 (5.0%) and 6 (3.1%), respectively, smoked cigarettes. Given that we focused on ETS exposure in this study, we discarded the 11 pairs for which the subject or control smoked, and we, therefore, focused on pairs (89 cases and 175 controls) who were nonsmokers. We used the generalized additive model to adjust for other confounding factors without imposing rigid parametric assumptions about dependence of confounders on CIN risk. (10)
We used multivariate conditional and unconditional logistic regressions to assess the association between subject or control status and different amounts of ETS exposures. Covariates in the final models were those that were significant or marginally significant in the univariate analysis, including education level (i.e., junior high school vs. no more than primary school; more than high school vs. no more than primary school), age (in years) at which 1st intercourse occurred ([less than or equal to] 21 vs. > 26; 22-24 vs. > 26; 25-26 vs. > 26), number of pregnancies (3 vs. [less than or equal to] 2; 4-5 vs. [less than or equal to] 2; > 5 vs. 2), and cooking in the kitchen between ages 2040 yr (yes [with ventilation] vs. no ventilation; yes [with ventilation] vs. yes [without ventilation]). The data were analyzed with the SAS and S-Plus statistical packages. (10,11) For each combination, the odds ratio (OR) and associated 95% confidence interval (CI) were calculated.
Results
A total of 116 subjects had biopsy-confirmed CINs that were at least level II. We were unable to interview 16 of these 116 subjects; 10 declined the interview, 5 individuals could not be reached on 2 separate occasions at the address given, and 1 person was deceased. The inability to interview the aforementioned 16 individuals did not bias our conclusions because the respondent (n = 100) and nonrespondent (n = 16) groups were very comparable. For example, the mean ages and education levels of the groups did not differ significantly ([p = 0.10] and [[chi square] = 1.06, df = 1, p = 0.30], respectively). The mean ages and education levels of respondents and nonrespondents were 52.5 yr (standard deviation [SD] = 13.1 yr) and 56.5 (SD = 12.1 yr), respectively. Forty-five percent of the respondents and 31.3% of the nonrespondents had education levels lower than junior high school.
Of the 100 study subjects who completed their questionnaires, 39 had CIN2, 12 had CIN3, 46 had carcinoma in situ, and 3 had invasive cancer. Distribution of demographic data and other potential confounders (Table 1) shows that members of the study group were closer in age, less educated, and had more active and passive smokers than did the control group. The number of women who had had more than 1 Pap smear during their lives was very similar statistically for both groups (82.0% of study subjects vs. 81.7% of control women). The study subjects had begun having sexual intercourse at a much younger age and became pregnant with greater frequency than women in the control group. Finally, the presence of a fume extractor in the home kitchen was a protective factor in the women aged 20-40 yr.
As stated previously, only 5 (5.0%) and 6 (3.1%) out of 100 cases and 197 controls, respectively, were cigarette smokers. Inasmuch as our study intended to examine ETS exposure, we discarded case-control pairs with cigarette smokers (n = 11) and focused on the analyses of 89 pairs (89 cases and 175 controls) of nonsmokers. The relationship between differing ETS exposure at home or in the workplace and the risk of CIN throughout life (childhood or adulthood) in the nonsmoking subjects and controls is summarized in Table 2. ETS exposure at home in adulthood was associated significantly with the development of CIN, but no such association was found between ETS exposure of adults in the workplace. Furthermore, no childhood ETS exposure, whether at home or in the workplace, was significant. Of the 89 CIN cases, 76 (85.4%), in contrast to 107 (61.1%) of the 175 controls, reported at least 1 yr of exposure to passive smoke at home during adulthood. Before and after adjusting for other covariates, the risk of contracting CIN in ETS-exposed vs. non-ETS-exposed women was 3.61-fold (95% CI: 1.84, 7.09) and 2.73-fold (95% CI: 1.31, 5.67), respectively.
Even when we dichotomized for education level, the findings remained similar. Among the 61 subjects and 83 controls with less than high school education, the risk of contracting CIN for exposed vs. nonexposed women was 3.36-fold (95% CI: 1.23, 9.17), after adjusting for covariates other than education level. Among the 28 subjects and 92 controls who had a high school education or some college education, the risk of contracting CIN for exposed vs. nonexposed women was 2.67-fold (95% CI: 0.97, 7.32) in a logistic regression model. Further adjustment for age did not alter the results appreciably (data not shown).
In addition, a dose-response trend was noted between ETS exposure at home during adulthood and risk of developing CIN (Table 3). Women who were exposed to 1-10 cigarettes/day or to more than 10 cigarettes/day at home during adulthood had a 2.13-fold (95% CI: 0.96, 4.73) and a 3.97-fold (95% CI: 1.65, 9.55), respectively, greater tendency to develop CIN (trend test, p = 0.002).
Expressing lifetime ETS exposure among nonsmoking women in terms of pack-years provided a cumulative sum of each woman's exposure to ETS. Figure 1 shows a smoothed plot of ORs vs. lifetime ETS exposure reported in this study, after adjustment for education level, number of pregnancies, age at which intercourse first occurred, and cooking activities in ventilated vs. unventilated kitchens during ages 20-40 yr. As shown, a close relationship existed between lifetime ETS exposure and CIN risk. Although some instances of high ETS exposure were associated with a decline in risk, such observations were relatively few (n = 6) and insignificant (Fig. 1). We adjusted for other potential confounders, after which we found that, compared with nonexposed women, women exposed to 1-20 pack-yr and women exposed to more than 20 pack-yr had 1.90-fold (95% CI: 0.72, 5.03) and 2.99-fold (95% CI: 1.10, 8.09) increased risk, respectively, of developing CIN.
[FIGURE 1 OMITTED]
Discussion
Although many epidemiological researchers have found that cigarette smoking, reproductive condition, and sexual behavior (likely related to infection with human papillomavirus [HPV] or herpes simplex) were the major risk factors for cervical cancer, these factors do not fully explain the entire pathogenesis of the disease. (12,13) Thus, other environmental and genetic components may play a critical role in the development of cervical carcinogenesis.
One major limitation of the present study was a lack of information about HPV infection. Sexual activity--likely accompanied with sexually transmitted infectious agents (e.g., HPV infection)--has a well-documented association with risk of cervical neoplasm. (12) In Taiwan, female virginity before marriage and fidelity subsequent to marriage are highly valued; therefore, compared with Caucasian women (who generally tend to experience intercourse several times prior to marriage), the relatively less intense sexual activity among Taiwanese women contrasts with their high incidence of cervical cancer. (14) Pao et al. (15) compared the prevalence of HPV in cervical cancer tissues of Chinese women who lived in 2 geographic locations that differed greatly with respect to cervical cancer incidence. The low-incidence group was from Taiwan (23.7/100,000 women), and the group with an extremely high incidence (i.e., 1,026/100,000 women) was from LueYang County in Shanxi Province, Peoples Republic of China; this figure was determined in an epidemiological survey of 12,980 women who resided in Shanxi Province in 1983. (15,16) Surprisingly, the positive incidence of HPV deoxyribonucleaic acid [DNA] was greater in the Taiwan group (88.5%; 54 of 61 tissue specimens) than in the Shanxi group (57.1%; 28 of 49 tissue specimens). The aforementioned findings suggest that HPV infection is an important, but not the only, risk factor for cervical neoplasms; other factors, such as environmental and host determinants, may also play important roles in the development of cervical neoplasms. (15) Alternatively, exposure to ETS over a lifetime is clearly associated with CIN risk, even after adjusting for age at 1st intercourse, which may be an alternative surrogate for sexually transmitted infectious agents, such as HPV infection.
Childhood ([less than or equal to] 20 yr of age) ETS exposure at home may be associated with risk of female lung cancer (8,9); however, we found no significance of this variable for CIN risk. This lack does not imply that no association exists between the 2 variables; memories of more remote experiences may be inaccurate and, therefore, may not be as precise as recollection of ETS exposure at home during adulthood. The resulting possible random misclassification would yield false-negative findings. We did not find significant effects of ETS exposure in the workplace--whether during childhood or adulthood--on the risk of the development of cervical neoplasms. Perhaps ETS exposure in the workplace is not as common as it is at home (Table 2), or perhaps information on ETS exposure in the workplace may be less accurate than information about home ETS exposure.
ETS contains many known human carcinogens (e.g., polycyclic aromatic hydrocarbons [PAHs], such as benzo[a]pyrene, benzo[a]anthracene, dibenz[a,h]anthracene, 5-methylchrysene, among others). (17-19) Two mechanisms of the effect of ETS exposures on cervical neoplasms seem possible: (1) direct carcinogenic action of ETS, such as through PAH-DNA adducts, in the cervix; and (2) immunosuppressant action by ETS. (20) In only a few epidemiological studies have authors examined the effect of ETS on the risk of cervical neoplasms. (20,21) Hellberg et al. (20) studied 392 women nonsmokers and found that 73 (55.3%) of 132 study subjects and 99 (38.1%) of 260 controls were exposed to ETS (p < 0.01). From 1987 to 1988, Coker et al. (21) studied 103 women with biopsy-confirmed CIN2 or CIN3, as well as 268 control women with normal cervical cytology in a case-control study. Coker et al. (21) found that 70% of the CIN2/3 patients were cigarette smokers, whereas only 30% of control subjects had ever smoked; after the authors controlled for confounders, the adjusted OR for current cigarette smoking was 3.4 (95% CI: 1.7, 7.0). However, given the small sample size in that study, the results--after controlling for confounders--showed that passive exposure to cigarette smoke was not associated significantly with CIN2/3 (adjusted OR: 1.5; 95% CI: 0.5, 4.0). Alternatively, the insignificant effect of ETS exposure on cervical neoplasms may be attributable to the greater percentage of active cigarette smokers in the Coker et al. (21) study subjects than in ours. Recently, the committee of the International Agency for Research on Cancer (IARC) of the World Health Organization in France, after reviewing all published evidence related to active and passive tobacco smoking and cancer, (17) concluded that cervical neoplasms are definitely associated with active tobacco smoking. However, the IARC's findings about the association between passive tobacco smoking and cervical neoplasms remain unresolved. (17) Our results provide additional evidence for such an association.
In Taiwan, in only a few epidemiological studies have investigators examined cigarette smoking and risk of cervical neoplasms, but no one has investigated the effect of ETS exposure on cervical neoplasm risk. (14,22-24) Given the minimal statistical power of those few studies, and in light of the small sample size of study subjects among women smokers, all of the investigators' findings were negative. From 1979 to 1984, Chou and Chen, (22) as well as Chou, (23) investigated the risk factors for CIN in 77,599 Taiwanese women via Pap smear screening. These investigators used questionnaires and found that frequency of sexual activity, onset of sexual behavior, and husbands' affairs with other women were major predictors of CIN risk. They did not find, however, a positive, significant effect of cigarette smoking on risk of CIN in Taiwanese women. Liaw et al. (24) compared 37 women who had low-grade lesions (i.e., CIN1) with 260 healthy women, and they found that, although only 6 of the women smoked cigarettes, women smokers had a 1.7fold risk (95% CI: 0.2, 20.0) of developing CIN1. Their findings, however, were not statistically significant. Similarly, Wang and Lin (14) recruited 288 women with cervical cancer and 576 controls; only 31 of the women smoked cigarettes. Although cigarette smoking was not a risk factor for cervical cancer (OR: 1.0; 95% CI: 0.7, 6.9) in the study by Wang and Lin, (14) the small number of cigarette smokers among Taiwanese women helps explain their conclusion. Therefore, our findings provide additional evidence of an association between ETS exposure and CIN risk.
Lee et al. (8,9) have reported that ETS exposure had a major significant effect on lung cancer risk among nonsmoking Taiwanese women. A competing effect of lung cancer and cervical dysplasia attributable to ETS exposure is possible; nonetheless, we still found that ETS exposure had a significant adverse effect on CIN risk. Although without treatment at least 25% of CIN2 patients will progress to carcinoma in situ or to invasive cervical cancer, (2) approximately 40% may regress to normal. Our study, which included women with CIN2 in the study group, may have, therefore, underestimated the association between ETS exposure and risk of cervical dysplasia.
In conclusion, the results of this study present evidence that lifetime ETS exposure can increase the risk of CIN among Taiwanese nonsmoking women. In addition, we noted a dose-response trend between the amount of lifetime ETS exposure and the risk of CIN.
Table 1.--Distribution of Demographic Characteristics and
Other Potential Confounders
Study
subjects Controls
(n = 100) (n = 197
Variable n % n %
Age (yr)
< 42 25 25.0 49 24.9
42-51 27 27.0 47 23.9
52-62 21 21.0 48 24.4
> 62 27 27.0 53 26.9
Cigarette smoking
Nonsmoker 12 12.0 45 22.8
Passive smoker * 83 83.0 146 74.1
Current or former smoker 5 5.0 6 3.1
Education levels
[greater than or equal to] High school 32 32.0 109 55.3
Junior high school 23 23.0 22 11.2
[less than or equal to] Primary school 45 45.0 66 33.5
No. of prior Pap smears
1 18 18.0 36 18.3
> 1 82 82.0 161 81.7
No. of lifetime sexual
partners
[less than or equal to] 1 93 93.0 189 95.9
> 1 7 7.0 8 4.1
Age at 1st intercourse (yr)
> 26 11 11.0 43 21.8
25-26 11 11.0 43 21.8
22-24 42 42.0 60 30.5
[less than or equal to] 21 36 36.0 51 25.9
No. of pregnancies
[less than or equal to] 2 14 14.0 48 24.4
3 15 15.0 47 24.9
4-5 38 38.0 70 35.5
> 5 33 33.0 32 16.2
Cooking in the kitchen
during ages 20 to 40 yr
No 10 10.0 18 9.1
Yes, with ventilation 65 72.2 151 84.4
Yes, without ventilation 25 27.8 28 15.6
Variable p
Age (yr)
< 42 0.90
42-51
52-62
> 62
Cigarette smoking
Nonsmoker 0.07
Passive smoker *
Current or former smoker
Education levels
[greater than or equal to] High school <0.01
Junior high school
[less than or equal to] Primary school
No. of prior Pap smears
1 0.95
> 1
No. of lifetime sexual
partners
[less than or equal to] 1 0.27
> 1
Age at 1st intercourse (yr)
> 26 <0.01
25-26
22-24
[less than or equal to] 21
No. of pregnancies
[less than or equal to] 2 <0.01
3
4-5
> 5
Cooking in the kitchen
during ages 20 to 40 yr
No 0.06
Yes, with ventilation
Yes, without ventilation
* Exposed to environmental tobacco smoke during childhood
or adulthood.
Table 2.--Effect of Exposure to Environmental Tobacco Smoke (ETS) and
Cervical Intraepithelial Neoplasms in Women Nonsmokers (N = 264)
Study
subjects Controls
(n = 89) (n = 175)
Variable n % n % OR
Childhood ETS exposure
At home
No 36 40.5 79 45.1 1.00
Yes 53 59.5 96 54.9 1.25
In the workplace
No 75 84.3 147 84.0 1.00
Yes 14 15.7 28 16.0 1.02
Adulthood ETS exposure
At home
No 13 14.6 68 38.9 1.00
Yes 76 85.4 107 61.1 3.61
In the workplace
No 61 68.5 130 74.3 1.00
Yes 28 31.5 45 25.7 1.34
Variable 95% CI AOR * 95% CI
Childhood ETS exposure
At home
No
Yes 0.73, 2.13 0.99 0.54, 1.83
In the workplace
No
Yes 0.50, 2.10 1.03 0.47, 2.26
Adulthood ETS exposure
At home
No
Yes 1.84, 7.09 2.73 1.31, 5.67
In the workplace
No
Yes 0.77, 2.33 1.56 0.83, 2.92
Notes: OR = odds ratio, and CI = confidence interval.
* Adjusted odds ratio, adjusted for educational levels (junior high
school vs. primary school or less; more than high school vs. primary
school or less), number of pregnancies (3 vs. [less than or equal to]
2; 4-5 vs. [less than or equal to] 2; > 5 vs. 2), age (yr) at 1st
intercourse ([less than or equal to] 21 vs. > 26; 22-24 vs. > 26; 25-26
vs. > 26), and cooking in the kitchen during ages 20-40 yr (yes with
ventilation vs. no; yes with ventilation vs. yes without ventilation).
Table 3.--Lifetime Exposure to Environmental Tobacco Smoke (ETS) and
Cervical Intraepithelial Neoplasms in Women Nonsmokers (N = 264)
Study
subjects Controls
(n = 89) (n = 175)
Variable n % n % OR
Childhood ETS exposure (cigarettes/day)
At home
No 36 40.4 80 45.7 1.00
1-10 42 47.2 70 40.0 1.35
> 10 11 12.4 25 14.3 1.02
In the workplace
No 74 83.2 146 83.4 1.00
1-10 11 12.4 19 10.9 1.25
> 10 4 4.5 10 5.7 0.79
Adulthood ETS exposure (cigarettes/day)
At home
No 13 14.6 68 38.9 1.00
1-10 37 41.6 66 37.7 2.74
> 10 39 43.8 41 23.4 5.04
In the workplace
No 61 68.5 130 74.3 1.00
1-10 12 13.5 25 14.3 1.07
> 10 16 18.0 20 11.4 1.66
Lifetime ETS exposure (pack-years)
No 7 7.9 33 18.9 1.00
1-20 40 44.9 95 54.3 1.91
> 21 42 47.2 47 26.8 4.18
Variable 95% CI AOR * 95% CI
Childhood ETS exposure (cigarettes/day)
At home
No
1-10 0.78, 2.35 1.11 0.59, 2.09
> 10 0.43, 2.40 0.82 0.31, 2.19
In the workplace
No
1-10 0.54, 2.90 1.34 0.53, 3.36
> 10 0.24, 2.55 0.79 0.21, 2.97
Adulthood ETS exposure (cigarettes/day)
At home
No
1-10 1.31, 5.72 2.13 0.96, 4.73
([dagger])
> 10 2.35, 10.81 3.97 1.65, 9.55
([dagger])
In the workplace
No
1-10 0.51, 2.24 1.47 0.64, 3.37
> 10 0.82, 3.37 1.65 0.73, 3.75
Lifetime ETS exposure (pack-years)
No
1-20 0.79, 4.64 1.90 0.72, 5.03
([double
dagger])
> 21 1.68, 10.38 2.99 1.10, 8.09
([double
dagger])
Notes: OR = odds ratio, and CI = confidence interval.
* Adjusted odds ratio, adjusted for covariate presented in Table 2.
([dagger]) Trend test, p = 0.002.
([double dagger]) Trend test, p = 0.02.
The authors gratefully acknowledge the public health nurses from Chia-Yi City Health Bureau for their assistance with field interviewing.
References
(1.) Department of Health, Executive Yuan, Republic of China. Health Statistics, 2000. Taipei: Department of Health, 2000.
(2.) Ostor AG. Studies on 200 cases of early squamous cell carcinoma of the cervix. Int J Gynecol Pathol 1993; 12: 193-207.
(3.) Clarke EA, Morgan RW, Newman AM. Smoking as a risk factor in cancer of the cervix: additional evidence from a case-control study. Am J Epidemiol 1982; 115:59-66.
(4.) Fiore MC, Novotny TE, Pierce JP, et al. Trends in cigarette smoking in the United States. The changing influence of gender and race. JAMA 1989; 261:49-55.
(5.) Gram IT, Austin H, Stalsberg H. Cigarette smoking and the incidence of cervical intraepithelial neoplasia, grade III, and cancer of the cervix uteri. Am J Epidemiol 1992; 135: 341-46.
(6.) Ko YC, Cheng LSC, Lee CH, et al. Chinese food cooking and lung cancer in women nonsmokers. Am J Epidemiol 2000; 151:140-47.
(7.) Annual Report of Tobacco, Alcohol Consumption Investigation in Taiwan Area. Taipei, Taiwan: Bureau of Tobacco and Alcohol Monopoly, Taiwan Provincial Government; 1974-1996.
(8.) Ko YC, Lee CH, Chen MJ, et al. Risk and protective factors for primary lung cancer among nonsmoking women in Taiwan. Int J Epidemiol 1997; 26:24-31.
(9.) Lee CH, Ko YC, Goggins W, et al. Lifetime environmental exposure to tobacco smoke and primary lung cancer of nonsmoking Taiwanese women. Int J Epidemiol 2000; 29: 224-31.
(10.) Hastie TJ, Tibshirani RJ. Generalized Additive Models. New York: Chapman & Hall, 1990.
(11.) SAS/STAT Guide for Personal Computers, Ver. 8. Cary, NC: SAS Institute, Inc., 2001.
(12.) Brisson J, Morin C, Fortier M, et al. Risk factors for cervical intraepithelial neoplasia: differences between low-and high-grade lesions. Am J Epidemiol 1994; 140: 700-10.
(13.) Kjellberg L, Hallmans G, Ahren AM, et al. Smoking, diet, pregnancy and oral contraceptive use as risk factors for cervical intra-epithelial neoplasia in relation to human papillomavirus infection. Br J Cancer 2000; 82:1332-38.
(14.) Wang PD, Lin RS. Risk factors for cervical intraepithelial neoplasia in Taiwan. Gynecol Oncol 1996; 62:10-18.
(15.) Pao CC, Kao SM, Tang GC, et al. Prevalence of human papillomavirus DNA sequences in an area with very high incidence of cervical carcinoma. Br J Cancer 1994; 70:694-96.
(16.) Zhang JM. The distribution and characteristics of carcinoma of the cervix uteri in LueYang County. Chin J Epidemiol 1986; 7:343-45.
(17.) IARC. Tobacco smoking and involuntary smoking. International Agency for Research on Cancer (IARC) Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 83. Lyon, France: IARC, 2002.
(18.) Kaufman DW, Palmer JR, Rosenberg L, et al. Tar content of cigarettes in relation to lung cancer. Am J Epidemiol 1989; 129:703-11.
(19.) U.S. Environmental Protection Agency (EPA). Respiratory health effects of passive smoking: lung cancer and other disorders. Smoking and Tobacco Control Monograph No. 4. Bethesda, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, 1993. NIH Publication No. 93-3605.
(20.) Hellberg D, Valentin J, Nilsson S. Smoking as risk factor in cervical neoplasia. Lancet 1983; 2:1497.
(21.) Coker AL, Rosenberg AJ, McCann MF, et al. Active and passive cigarette smoke exposure and cervical intraepithelial neoplasia. Cancer Epidemiol Biomarkers Prev 1992; 1:349-56.
(22.) Chou P, Chen V. Mass screening for cervical cancer in Taiwan from 1974 to 1984. Cancer 1998; 64:962 68.
(23.) Chou P. Review on cervical cancer screening. Zhonghua Yi Xue Za Zhi (Taipei) 1991; 48:1-12.
(24.) Liaw KL, Hsing AW, Chen CJ, et al. Human papillomavirus and cervical neoplasia: a case-control study in Taiwan. Int J Cancer 1995; 62:565-71.
This research was supported by grants from the Province of Taiwan, Republic of China, the Taiwan National Health Research Institutes (NHRI-EX92-9205PI); by the Taiwan National Science Council (NSC 91-2320-B-037-053); and by an award from the American Bureau for Medical Advancement in China, Inc. (ABMAC), for Ming-Tsang Wu.
Submitted for publication August 23, 2002; revised; accepted for publication February 24, 2003.
Requests for reprints should be sent to Trong-Neng Wu, in care of Ming-Tsang Wu, Graduate Institute of Occupational Safety and Health, Kaohsiung Medical University, 100 Shih-Chuan 1st Road, Kaohsiung, Taiwan 807, Republic of China.
E-mail: mingtsangwu@yahoo.com
MING-TSANG WU
LI-HUNG LEE
CHI-KUNG HO
CHIA-LING LIU
TRONG-NENG WU
Graduate Institute of Occupational Safety and Health
Department of Occupational Medicine
Kaohsiung Medical University
Kaohsiung, Taiwan, Republic of China
SU-CHU WU
Chia-Yi City Health Bureau
Chia-Yi, Taiwan, Republic of China
LONG-YAU LIN
Department of Gynecology and Obstetrics
Chung-Shan Medical University Hospital
Taichung, Taiwan, Republic of China
BI-HUA CHENG
Department of Gynecology and Obstetrics
Chang-Gung Memorial Hospital
Kaohsiung, Taiwan, Republic of China
CHUN-YUH YANG
Institute of Public Health
Kaohsiung Medical University
Kaohsiung, Taiwan, Republic of China