Vol. 5. 155-160. Marc/i /996 Cancer Epidemiology, Biomarkers & Prevention 155 Physical Activity and Breast Cancer Risk1 Barbara D’Avanzo,2 Oriana Nanni, Carlo La Vecchia, Silvia Franceschi, Eva Negri, Attilio Giacosa, Ettore Conti, Maurizio Montella, Renato Talamini, and Adriano Decarli Istituto di Ricerche Farmacologiche Mario Negri. Via Eritrea 62. 20157 Milano (B. D.. C. L. V.. E. N.(: Istituto Oncologico Romagnolo. Ospedale Pierantoni, 471(83 ForlI (0. NI: Istituto di Statistica Medica e Biometria, Universita di Milano. lstituto Nazionale dci Tumori. 20133 Milano [C. L. V.. A. D.(: Centro di Riferimento Oncologico. 33081 Aviano (PN) [S. F., R. 1.]: Istituto Nazionale per In Ricerca sul Cancro. 16132 Genova [A. G.(; Istituto Regina Elena per Ia Studio e Ia Cura dci Tumori, (30100 Roma (E. C.J; and Istituto per lo Studio e Ia Cura dci Tumor) Senatore Pascale, 80100 Napoli [M. M.(. Italy Abstract Data from a multicentric case-control study on breast cancer conducted in Italy were used to analyze the relationship of occupational and leisure-time physical activity with breast cancer risk. Cases were 2569 histologically confirmed incident breast cancer cases, and controls were 2588 patients admitted to the same network of hospitals of cases for acute, nonneoplastic, non- hormone related diseases. After allowance for major identified potential confounding factors (including an estimate of total calorie intake), the odds ratios (ORs) were 0.70, 0.71, 0.64, and 0.54 in subsequent levels of physical activity at work at ages 30-39, compared to the lowest level. The association was similar for occupational physical activity at ages 15-19 and still apparent at ages 50-59, with risk estimates of 0.86, 0.85, 0.85, and 0.62. The ORs for the highest versus the lowest category of leisure-time physical activity were also below unity (ORs for the highest level of leisure-time physical activity at ages 15-19, 0.95; at ages 30-39, 0.76; and at ages 50-59, 0.66). The protection of physical activity was apparently stronger below age 60 at diagnosis and was consistent across the strata of selected covariates, although the protection was somewhat greater for more educated women. Introduction There are various indirect indications that physical activity can reduce breast cancer risk. These include the observation that Received 6/5/95: revised I 1/3/95; accepted I 1/3/95. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. I This work was conducted within the framework of the Consiglio Nazionale delle Ricerche ( Italian National Research Council) Applied Projects “Clinical Appli- cations ofOncological Research’ (Contracts 94.01321.PF39. 94.01 1 l9.PF39. and 94.01 268.PF39. and ‘Prevention and Control of Disease Factors” (Contract 95.0()952.PF4I (. and with the contributions of the Italian Association for Cancer Research. the Italian League against Tumours. Milan, and Angela Marchegiano Borgomainerio. 2 To whom requests for reprints should be addressed. strenuous exercise in adolescence is associated with reduced breast cancer risk later in life (I ), possibly by delaying men- arche and, in general, reducing the frequency of ovulation, which is a possible correlate of a woman’s breast cancer risk (2-5). Only scanty epidemiologicab data are available on the issue. These include an investigation of athletes (1 ), indicating a lower prevalence of breast cancer among more active people; a study using occupational status as indicator of physical ac- tivity (6), which found decreased breast cancer mortality in more active women; and the First National Health and Nutrition Examination Survey (7), which reported inverse associations for either leisure time and occupational activity, in postmeno- pausal women only. The Framingham study (8), however, sug- gested a positive relationship, of borderline significance, for occupational and leisure-time physical activities combined. No adjustment for dietary factors was possible on those data. The First National Health and Nutrition Examination Survey and the Framingham study were both based on relatively few cases of breast cancer ( 122 and 1 1 7, respectively). Results from a cohort study conducted in Finland on physical education and language teachers were first reported by Vihko (9), and indicated a reduced risk of breast cancer for premenopausal physical education teachers. A subsequent re- port from the same study (10), however, showed an excess risk of breast cancer in both teacher groups as compared to the total Finnish female population and only a slight difference in breast cancer risk between the two groups. A record-linkage study conducted in Shanghai (I 1), using occupational categories as proxy indicators of physical activity, found lower standardized incidence ratios for women employed in occupations requiring short time sitting and high energy expenditure, with standard- ized incidence ratios of 87 for service workers and 91 for craftswomen, and the protection persisted in retired women. A Turkish study (12) found an OR3 of 1 . 1 for very low energy expenditure and of 1 .5 for long time spent sitting at work, but these associations were weakened after allowance for socio- economic status. Only a case-control study conducted within the Cancer Surveillance Program of the University of Southern California (4) on 545 breast cancer cases 40 years of age or under was specifically focused on physical activity. A strong protection emerged with physical activity (OR = 0.42 for women report- ing 3.8 h/week or more of physical activity compared to mac- tive ones), with a consistent trend in risk, and persisted when several confounding factors were taken into account. In that study, the protection was apparently stronger among parous women, and this was interpreted in terms of the ability of physical activity to reduce ovulatory menstrual cycles and days of exposure to estradiol and progesterone in the luteal phase (13). 3 The abbreviations used are: OR. odds ratio: BMI. body mass index: CI. confi- dence interval. on August 18, 2015. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from