RESPONSE: Re: Active Smoking, Household Passive Smoking, and Breast Cancer: Evidence From the California Teachers Study

We thank Dr. Johnson for his thoughtful comments regarding our analysis of active smoking and passive household smoking and breast cancer in the California Teachers Study (CTS). To further evaluate the influence of active smoking with respect to a first full-term pregnancy, Dr. Johnson suggested presenting our risk estimates among nulliparous women only and also stratified by the number of years of pre-partum smoking. In response, we have conducted these analyses and present the results below. As can be seen in Table 1 , the patterns of risk estimates for active smoking among nulliparous women are similar to those reported for the full sample of women in our original analysis ( 1 ). Table 1 also shows our risk estimates for parous women stratified by how long a woman smoked before her first full-term pregnancy. These analyses are suggestive of a stronger smoking effect among women with longer smoking histories before their first pregnancy. However, the number of cases among women who smoked for less than 5 years pre-partum was small. When we restricted the analyses to postmenopausal breast cancers, the results were essentially the same (data not shown). The difference in risk estimates between those who smoked pre-partum for shorter compared to longer periods of time are not as dramatic as those reported by Johnson et al. in their recent research ( 2 ). With respect to our passive smoking analysis, which was limited to summary indicators of household sources, we agree with Dr. Johnson that it would be valuable to take a comprehensive look at passive smoking that includes quantitative measures from all sources. Our rationale for using household passive smoking measures only was twofold. First, household passive smoking is the metric that appears most often in the literature on passive smoking exposure. Thus, our results can be compared with those from a large body of literature. Second, the detailed passive smoking data collected subsequent to the baseline questionnaire are available for only a subset of the cohort, whereas the summary indicators of household passive smoking are available for the entire cohort. A recent descriptive analysis of the detailed passive smoking exposures in this cohort ( 3 ) suggested that, until the 1980s, the household was the major source of passive smoking exposures. Furthermore, the household was by far the most common setting for exposures experienced during childhood. If early-life exposures are most important in determining breast cancer risk, as the results of our active smoking analysis suggest, then it is likely that the summary measures of household passive smoking exposures that we used capture the majority of the passive smoking exposures of importance to breast cancer risk. Nevertheless, Dr. Johnson is correct in surmising that this cohort may have experienced significant occupational exposure to passive smoking. In fact, during the 1980s, the workplace replaced the home as the primary source of passive smoking exposures in this cohort ( 3 ). Extremely detailed information on lifetime passive smoking exposures in the CTS is now available, and we hope to analyze these data in the near future. These data will allow us to consider quantitative measures of passive smoke exposures from household, occupational, and other settings and for various time frames during a woman’s life, which may be especially important. We also look forward to comparing these results with those from the Canadian study ( 2 ). Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for invasive breast cancer associated with active smoking history among members of the California Teachers Study (CTS). * Limited to CTS cohort members living in California at baseline without a previous history of breast cancer who provided valid smoking data. Full study sample includes women with unknown parity. Includes 55109 never smokers (referent group). Postpartum-only smokers (n = 2261) were not included in this analysis. Number of cases do not always add up to total because of missing values for some smoking variables. For the full study sample, hazard ratios were adjusted for age, race, family history of breast cancer, age at menarche, physical activity, alcohol consumption, body mass index (BMI), menopausal status, BMI and menopausal status interaction, hormone therapy, parity, and age at first full-term pregnancy. For nulliparous women, there was no adjustment for parity or age at first full-term pregnancy. Pack-years is defined as packs per day smoked times number of years smoked. Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for invasive breast cancer associated with active smoking history among members of the California Teachers Study (CTS). * Limited to CTS cohort members living in California at baseline without a previous history of breast cancer who provided valid smoking data. Full study sample includes women with unknown parity. Includes 55109 never smokers (referent group). Postpartum-only smokers (n = 2261) were not included in this analysis. Number of cases do not always add up to total because of missing values for some smoking variables. For the full study sample, hazard ratios were adjusted for age, race, family history of breast cancer, age at menarche, physical activity, alcohol consumption, body mass index (BMI), menopausal status, BMI and menopausal status interaction, hormone therapy, parity, and age at first full-term pregnancy. For nulliparous women, there was no adjustment for parity or age at first full-term pregnancy. Pack-years is defined as packs per day smoked times number of years smoked.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.