Breast Cancer in California Teachers- Regional Variations

Institution: Public Health Institute
Investigator(s): Peggy Reynolds, Ph.D. -
Award Cycle: 2000 (Cycle VI) Grant #: 6JB-0111 Award: $160,267
Award Type: IDEAS II
Research Priorities
Etiology and Prevention>Etiology: the role of environment and lifestyle



Initial Award Abstract (2000)
Regional variations in breast cancer incidence have traditionally been a source of public health concern. These variations have also served to generate many hypotheses about factors which might cause breast cancer. Breast cancer rates in the San Francisco Bay Area have been recently touted as "the highest in the world." This distinction is consistent with the observation of higher rates in this area over the last several decades, as well as with generally higher breast cancer morbidity and mortality rates in urban areas. Rate differences observed on the national scale are similar to rate differences seen within California.

Geographic variations in breast cancer incidence, while well-recognized, are not well understood. Most studies aimed at investigating geographic patterns of breast cancer have relied on data at the population-level with no individual-level data on personal risk factors. The study proposed here is designed to build on a rich repertoire of information available on individual women participating in the California Teachers Study (CTS). Comprised of over 133,000 active and retired female California school employees, the CTS is one of the largest studies implemented to date specifically designed to study personal risk factors for breast cancer. With study participants residing in all corners of the state, and with patterns of breast cancer incidence mirroring those seen at the state and national level, the CTS offers a unique opportunity to use individual-level data in exploring geographic patterns of breast cancer incidence. The primary goal of this study is to determine the degree to which established and suspected risk factors explain the observed geographic patterns of breast cancer incidence within the California Teachers Study (CTS) cohort.

The specific aims of this study are to: The extensive information gathered by the CTS surveys will allow us to examine the role of a myriad of established and suspected personal risk factors among the CTS members including: reproductive factors, family history, physical activity, hormonal exposures, and dietary & other ‘lifestyle’ factors. Current and historic place of residence as reported in the CTS data will be used in conjunction with statewide census and environmental monitoring data to evaluate the additional contribution of social factors and environmental exposures in explaining regional patterns of breast cancer incidence. Together these resources offer information on both traditionally-accepted and newly-postulated risk factors which when jointly evaluated may provide helpful new insights into geographic patterns in breast cancer incidence and ultimately into the causes of breast cancer.


Final Report (2004)
Breast cancer incidence rates have historically been higher in the San Francisco Bay Area and in the southern coastal area (Los Angeles, Orange and San Diego Counties) than other areas of the State. While this geographic pattern of breast cancer is well-recognized and has been observed over the past several decades, the reasons for it are not well understood. Most studies investigating geographic patterns of breast cancer incidence have relied on data at the population-level with no individual-level data on personal risk factors. The current study is designed to describe and evaluate factors related to the regional distribution of breast cancer incidence among members of the California Teachers Study (CTS), a large cohort (n = 133,479) of California women specifically designed to study breast cancer.

After excluding women residing outside of California at baseline and extensive geocoding efforts for women residing inside California, 121,742 (91%) members of the cohort were assigned to a California census block group of residence at baseline and were included in these analyses. Linkage of the cohort to 1996-1999 California Cancer Registry data identified 1,674 incident cases of invasive breast cancer. Based on the geocoded address at baseline, cohort members were assigned to one of three regions in the state: San Francisco Bay Area (San Francisco, San Mateo, Marin, Alameda, Contra Costa, and Santa Clara counties); South Coast (Los Angeles, Orange and San Diego Counties); remainder of the State (all other counties). Cox proportional hazards models were run to compute hazard ratios (HRs) for the SF Bay Area and the South Coast regions compared to the remainder of the State. After adjusting for age and race/ethnicity, the SF Bay Area had breast cancer rates approximately 20% higher (RR = 1.21, 95% CI = 1.06 – 1.38) and the South Coast rates approximately 17% higher (RR = 1.17, 95% CI = 1.05 – 1.30) than the remainder of the State. These HRs are consistent with what has been observed for the statewide population.

We considered both neighborhood (social and environmental) and individual-level (personal) risk factors as potential explanatory factors underlying the observed regional patterns in breast cancer incidence. Census data were used to create summary measures of neighborhood social characteristics (socioeconomic status and degree of urbanization). Data from the California Department of Pesticide Regulation and from the California Department of Transportation were used to explore the influence of residential proximity to agricultural pesticide use and traffic, respectively. Information on regional distributions of personal risk factors was derived from the baseline CTS survey. While notable regional differences in the distributions of these factors were found, accounting for these factors, reduced, but did not fully explain the observed regional patterns of breast cancer incidence in this cohort of women.

This study is the first to incorporate both neighborhood- and individual-level risk factors in attempting to explain regional patterns of breast cancer incidence. In so doing, we hope to provide new insights into geographic breast cancer patterns and ultimately into the causes of breast cancer.

We have completed all work towards our three specific aims. Two manuscripts have been submitted for publication: one is in press, (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14712141&dopt=Abstract) the other is provisionally accepted. We have attached the abstracts for these papers, and will forward copies of them once they are published.


Symposium Abstract (2003)
Within California, rates of breast cancer have tended to be higher in the San Francisco (SF) Bay Area and in the southern coastal area (Los Angeles, Orange and San Diego Counties) than in the remainder of the State. These variations in incidence are also evident among members of the California Teachers Study (CTS), a large cohort of 133,479 female professional school employees who have participated in a series of questionnaires and have been followed for cancer incidence since 1995. Of these, 121,742 (91%)could be assigned to a California census block group of residence at baseline and were included in these analyses. Linkage of the cohort to 1996–1999 California Cancer Registry data identified 1,674 incident cases of invasive breast cancer. Statistical models were used to compare the rates of breast cancer in the three regions of the state, after taking into account differences in age and race/ethnicity. Compared to the remainder of the state, the age and race adjusted incidence of breast cancer was 22% higher among SF Bay Area residents and 16% higher among South Coast Area residents. Women living in both the SF Bay and South Coast regions of the state were, on average, slightly older, represented a more racially/ethnically diverse population, and were more likely to live in urban and affluent neighborhoods. While these characteristics were independently associated with breast cancer risk, adjustment for these factors did not fully explain the excess risk seen in these two areas. These associations were most evident for post-menopausal members of the cohort. Associations for pre/peri-menopausal women appeared to be somewhat different, although the number of cases occurring in this group was relatively small. Further exploration of differences in regional profiles of personal and environmental risk factors thought to be associated with breast cancer is the subject of this research project.

Residential exposure to traffic in California and childhood cancer.
Periodical:Epidemiology
Index Medicus:
Authors: Reynolds P, Von Behren J, Gunier RB, Goldberg DE, Hertz A (year is 2004)
Yr: 2004 Vol: 15 Nbr: 1 Abs: Pg:6-12