Differences in lung cancer risk from smoking among Japanese, Chinese and Hawaiian women in Hawaii

Our plain-English summary of Differences in lung cancer risk from smoking among Japanese, Chinese and Hawaiian women in Hawaii, by M.W. Hinds, G.N. Stemmermann, H.-Y. Yang, et al (1981).

The study

• Authors: M.W. Hinds, G.N. Stemmermann, H.-Y. Yang, L.N. Kolonel, J. Lee
• Published: Int. J. Cancer, 27, 297-302 (1981)
• Type of article: Journal article
• Our copy: from Johns-Manville's own medical library. The article is still under copyright, so we describe it here rather than post it.

In plain English

This case-control study examined the link between cigarette smoking and lung cancer in Japanese, Chinese and Hawaiian (including part-Hawaiian) women in Hawaii. It does not discuss asbestos. The authors reviewed hospital records of 375 women with lung cancer diagnosed from 1968 to 1978 (176 Japanese, 132 Hawaiian, 67 Chinese). They compared their smoking histories with those of 2,404 women from the general population, adjusting for age and a socioeconomic index. Ever-smokers had odds of lung cancer 10.5 times those of never-smokers among Hawaiian women, 4.9 times among Japanese women and 1.8 times among Chinese women. For Chinese women the 95% confidence interval included 1.0, so that result was not statistically significant. The link was stronger for epidermoid and small-cell cancers than for adenocarcinoma. The authors estimated, as approximations, that smoking accounted for 79% of lung cancer in Hawaiian women, 44% in Japanese women and 11% in Chinese women. They noted that smoking was recorded differently for cases and controls and was missing for 22.1% of cases. The authors concluded that only a minority of lung cancers in Chinese and Japanese women could be attributed to smoking.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Retrospective case-control study using hospital records for cases and a population health survey for controls
• Who was studied: 375 women with primary lung cancer from the Hawaii Tumor Registry, diagnosed 1968-1978: 176 Japanese, 132 Hawaiian/part-Hawaiian and 67 Chinese. Hospital records were found for these 375 of 397 eligible cases (94.5%). Controls were 2,404 sex- and race-matched women from the general population, interviewed in 1975-1977 in the Hawaii State Department of Health's Health Surveillance Program.
• Exposure: Cigarette smoking (ever-smokers versus never-smokers), taken from hospital records for cases. A history counted as negative only if the record said the person had never smoked. A socioeconomic index based on census tract was also analyzed. No asbestos or other occupational exposure was studied.
• Main results:
  • All lung cancer types combined, ever-smokers versus never-smokers, adjusted for age and socioeconomic index: odds ratio 10.5 for Hawaiian women.
  • All lung cancer types combined, ever-smokers versus never-smokers: odds ratio 4.9 for Japanese women and 1.8 for Chinese women.
  • For Chinese women, the 95% confidence interval for all lung cancer included 1.0. The authors said smoking could not be clearly incriminated as a significant risk factor in this group.
  • For epidermoid plus small-cell lung cancer, Japanese and Chinese women had essentially the same odds ratio, while Hawaiian women had an odds ratio of 92.4.
  • Of 48 Hawaiian women with epidermoid or small-cell lung cancer, only one had never smoked, compared with 61% of Hawaiian control women.
  • In all three ethnic groups, adenocarcinoma was consistently associated with a lower odds ratio than the other histologic types.
  • Among Chinese women, the only odds ratio significantly greater than 1.0 was for epidermoid plus small-cell lung cancer. For Japanese and Hawaiian women, the odds ratios for all histologic types were significantly greater than 1.0.
  • Hawaiian cases were more often epidermoid or small-cell type. Almost two-thirds of Japanese and Chinese cases were adenocarcinoma.
  • Among cases with quantitative smoking histories (44 Japanese, 8 Chinese, 78 Hawaiian), cigarettes smoked per day did not differ significantly between the ethnic groups (p=0.7). The authors said heavier smoking did not appear to explain the higher Hawaiian odds ratios.
  • For all lung cancer, the smoker versus never-smoker odds ratio did not differ significantly between Japanese women born in Hawaii (odds ratio 2.9) and Japanese women born in Japan (odds ratio 4.1).
  • After adjusting for smoking and age, Japanese women born in Japan with a high socioeconomic index were at significantly greater risk of lung cancer than those with a low index. This was not the case for Japanese women born in Hawaii (odds ratio 1.2; 95% CI 0.77-1.9).
  • The socioeconomic index was not significantly associated with all lung cancer in any ethnic group. Hawaiian women with a high index were at significantly greater risk of adenocarcinoma than those with a low or medium index.
  • Estimated share of lung cancer attributable to smoking: 79% for Hawaiian women, 44% for Japanese women and 11% for Chinese women.
• The authors' conclusion: The authors concluded that only a minority of lung cancers among Chinese and Japanese women in Hawaii could be attributed to smoking. They also concluded that Hawaiian women who smoke were at greater risk, particularly of epidermoid or small-cell lung cancer, than Japanese or Chinese women who smoke. They said other factors should be sought and could only speculate about genetic or environmental causes.
• Limitations the authors state:
  • Smoking information was gathered differently for cases and controls. Cases were mostly asked by their physicians, while controls were interviewed by lay persons in a general health survey.
  • No smoking history was recorded in 22.1% of the cases' hospital records.
  • The authors judged that cases with unknown smoking history were probably more often never-smokers. If so, their odds ratios may be somewhat too high, particularly for Japanese and Chinese women.
  • The odds ratios had wide 95% confidence intervals, so the attributable-risk percentages should be viewed only as approximations.

Topics

Medical terms in this study, explained in our Lexicon: Lung Cancer, Case-Control Study, confidence interval, Adenocarcinoma.

Browse return to medical research.