Background: Physical activity has been found to be associated with decreased risk of breast cancer in postmenopausal women in the majority of epidemiologic studies, but the association is inconsistent in premenopausal women.
Methods: We studied the effect of physical activity at various ages on the incidence of breast cancer in 99,504 women from 30 to 49 years of age at enrollment in the Women's Lifestyle and Health Study, a prospective population-based cohort study in Norway and Sweden. Physical activity level on an ordinal scale at age 14, age 30, and age at enrollment, participation in competitive sports, as well as information on other covariates was obtained using a self-administered questionnaire. Complete follow-up with data on incident invasive breast cancer and mortality was collected by linkage to national registries. The relation between physical activity and time to breast cancer development was assessed using Cox proportional hazard models, controlling for potential confounders.
Results: During an average 9.1 years of follow-up, there were 1,166 incident breast cancer cases. The mean age of the women was 41 years at enrollment, and the mean age at breast cancer diagnosis was 48 years. Compared to inactive women, women with higher levels of physical activity at enrollment had a similar risk of incident breast cancer (adjusted relative risk, 1.24 for vigorous activity compared with no activity; 95% CI, 0.85-1.82). Physical activity at age 30 or at age 14 also did not afford any significant protection from breast cancer, nor did a consistently high level of activity from younger ages to enrollment.
Conclusions: We did not find evidence of a protective effect of physical activity on breast cancer risk in this group of primarily premenopausal women.
- breast neoplasms
- physical activity
- cohort study
- Breast cancer
- Diet, alcohol, smoking, and other lifestyle risk factors
Of the known and suspected risk factors for breast cancer, physical activity is one of the few that can be modified (1). Physical activity has been found in the majority of published studies to be associated with decreased risk of breast cancer (2, 3). Several biologically plausible mechanisms have been proposed, including decreased estrogen and progesterone levels, enhanced natural immunity, and reduction in levels of insulin and other growth factors (4). Physical activity might also prevent women from weight gain and development of obesity, an established breast cancer risk factor after menopause (2, 5).
There is no consensus on the critical exposure period, intensity, frequency, or consistency of physical activity that might be required to protect against breast cancer. Although studies have examined different populations at various ages and used widely divergent methods to measure physical activity, there is disagreement even among well-designed studies using similar methods (2). However, in contrast to the findings in postmenopausal women, the association of physical activity and breast cancer incidence in premenopausal women is particularly inconsistent between studies (6-11).
We therefore studied the association between physical activity and incident invasive breast cancer over an average of 9.1 years of follow-up in a cohort of women enrolled in the large, population-based prospective Norwegian-Swedish Women's Lifestyle and Health Cohort Study. At enrollment, 99,504 participants, ages 30 to 49, answered questions about their physical activity levels at age 14, age 30, and age at enrollment, allowing an assessment of the effect of age at exposure and consistency of physical activity.
Materials and Methods
The present investigation is based on data from the Women's Lifestyle and Health Study, a prospective cohort study in Norway and Sweden, described in detail previously (12, 13). The study population includes 57,582 Norwegian women from 34 to 49 years of age at recruitment, and 49,259 Swedish women from 30 to 49 years of age, who were monitored with regard to vital status, incident cancer, and emigration from 1991 to 1992. In Norway, the source population was the entire country during that time. The source population in Sweden consisted of all women who were 30 to 49 years old and residents in the Uppsala Health Care Region from 1990 to 1991. Women were randomly selected from the populations within four 5-year birth cohorts (ages 30-34, 35-39, 40-44, and 45-49 years). A total of 196,000 women were invited to participate, and 106,841 (54.5%), returned the questionnaire.
From the initial cohort, we excluded 1,645 women with a diagnosis of malignant cancer other than nonmelanoma skin cancer at enrollment. We further excuded 5 women due to lack of information on vital status, 13 women who had emigrated before the start of follow-up, and 5,674 women who were missing information on all physical activity variables. Thus, the final analysis was based on data from 99,504 (51,429 Norwegian and 48,075 Swedish) women. The responsible data monitoring boards and ethical committees in both countries approved the study design, and all women gave informed consent prior to participating in the study.
Exposure and covariate data in the cohort are based on self-reported information collected in a questionnaire given at cohort enrollment. The questionnaires sent to Norwegian and Swedish women included a core set of similar questions, including information on physical activity, demographic characteristics, active and passive smoking exposure, alcohol intake, medical history, height, and weight.
The women rated their level of physical activity at three time points: age 14, age 30, and at enrollment. Norwegian women ranked activity on a scale of 1 (very low) to 10 (very high). On the Swedish questionnaire, women ranked their level of physical activity on a 5-point scale: 1 (very low) was described as mainly sitting, 3 (normal) as several long walks per week, and 5 (very high) as sports or jogging several times per week. We collapsed the 10-level Norwegian scale into 5 levels to be comparable to the Swedish scale. Women were also asked whether they had participated in competitive sports, and if so, the number of years they participated.
Using the self-reported data at age 14, age 30, and at enrollment, we further categorized individuals on changes in physical activity over time. First, we dichotomized the levels of physical activity into women who participated in no or low physical activity (inactive), and those who participated in moderate, high, or vigorous activity (active) for each time point. We then compared physical activity levels between ages 14 and 30, age 14 and enrollment, and between age 30 and enrollment. From this comparison, women could be categorized as those who remained inactive, women who were active and became inactive, women who were inactive and became active, and those who remained active over time. Women who reported a natural menopause or a bilateral oophorectomy at enrollment were considered postmenopausal during the follow-up. All other women were considered premenopausal, regardless of age, hysterectomy, or use of hormone therapy.
Follow-up of the cohort was achieved through linkages with existing nationwide health registers. Because each resident in Norway and Sweden is assigned a unique national registration number, one can link the data from the cohort with these registers for virtually complete follow-up with respect to death and emigration. From the total population registers, we received information on the dates of death for women who died during the follow-up period, and dates of emigration for women who moved out of their respective countries. The population registers were updated through January 2003 for Norway and through June 2003 for Sweden. The national cancer registries, established in the 1950s in both countries, provided data on prevalent cancer cases at cohort enrollment and on incident invasive breast cancers as well as other cancers diagnosed in the cohort during follow-up. During the period studied, these registries are also estimated to be close to 100% complete (14-16). The start of follow-up was defined as the date of return of the questionnaire from 1991 to 1992. Observation time was calculated from date of entry into the cohort until the occurrence of incident breast cancer, or censoring on account of other cancer, emigration, death, or end of the observation period (December 31, 2000).
The relation between physical activity and invasive breast cancer was assessed using time-to-event analyses. The cumulative risk of breast cancer during follow-up was estimated using Kaplan-Meier methods.
Proportional hazard models (Proc PHREG, SAS Version 8.2) were used to estimate the hazard ratio as a measure of relative risk (RR) and with 95% CI, controlling for potential confounders. In each comparison, groups with low levels of physical activity were used as the reference. The following covariates assessed at enrollment were considered as potential confounders of the association between physical activity and breast cancer: age at enrollment, number of years of education, body mass index (BMI, categorized as < 25.0, 25.0-29.9, 30.0+), height, smoking status (current, past, or never), alcohol intake (0 drinks per day, < 1, 1 or more), age at menarche, parity, age at first birth, months of breast-feeding, oral contraceptive use (current, past, or never), sister or mother with breast cancer, menopausal status at enrollment, and country of origin (Norway versus Sweden).
To assess possible effect modification, we first estimated hazard ratios comparing active versus inactive women, and stratified models by age at enrollment, menopausal status, age at breast cancer diagnosis, smoking status, BMI, education, and country of origin. We then assessed effect modification on a multiplicative scale by modeling interaction terms and calculating log-likelihood ratio tests to assess whether interaction terms were significantly different than 0. Because we do not have information about menopausal status after the start of follow-up, the age of 50 was chosen as a proxy to distinguish premenopausal from postmenopausal breast cancer cases, based on the average age at menopause (approximately at age 50 in Sweden and at age 49 in Norway; refs. 17, 18).
In our cohort of 99,504 women during an average of 9.1 years of follow-up, altogether there were 1,166 incident invasive breast cancer cases. The annual incidence of breast cancer in the cohort per 1,000 cases was 1.38 in Norway, 1.18 in Sweden, and 1.29 overall. The mean age at breast cancer diagnosis was 48.1 years, and 42% of the cancers occurred in women 50 and older.
Table 1 shows demographic characteristics and breast cancer risk factors by physical activity level at enrollment and for the entire cohort. The mean age of the study participants at enrollment was 41 years, and only 6.9% of the cohort was postmenopausal at enrollment. More active women were slightly younger, had lower BMI (P < 0.0001), were less likely to smoke (P < 0.0001), consumed less alcohol, and were more likely to use oral contraceptives. There were no clear trends across physical activity levels for age at menarche, parity, age at first birth, months of breast-feeding, or family history of breast cancer.
About 80% of the women reported at least moderate levels of activity at enrollment, with even higher proportions at younger ages (Table 2). About 16% of the women reported participating in competitive sports, and nearly 10% had competed for 5 or more years. More than 70% of the women reported sustained moderate to vigorous activity from age 14 to later time points. A change from inactive (no or low physical activity) to active (moderate, high or vigorous physical activity) was reported by 8.2% from age 14 to age 30, and by 4.9% from age 30 to enrollment. Conversely, 9.9% who were active at age 14 became inactive by age 30, and 11.4% became inactive from age 30 to enrollment.
Compared to inactive women, women with higher levels of physical activity at enrollment did not have a lower risk of incident breast cancer (Table 3). Extensive adjustment for potential confounders changed these risks minimally. There was no evidence of a trend over categories of increasing levels of physical activity (P for trend, 0.85), and the RR for the highest compared with the lowest category of physical activity was 1.24 (95% CI, 0.85-1.82). When all breast cancer cases occurring in the first 2 years were deleted, the results were unchanged: the RR for the highest compared with the lowest category of physical activity was 1.12 (95% CI, 0.70-1.68). Likewise, physical activity at age 30, at age 14, or participation in competitive sports did not afford protection from breast cancer (Table 3). Although we did not find that a consistently high level of physical activity was beneficial (Table 4), a change from being physically inactive to active from age 30 to enrollment was associated with significantly lower breast cancer risk (RR, 0.66; 95% CI, 0.44-0.96; P = 0.03).
There was no evidence that physical activity at enrollment provided protection from incident breast cancer in subgroups defined by age at enrollment, age at follow-up, smoking, BMI, education, and country of origin. The RR estimates were all very close to unity, and all tests for interaction were nonsignificant. However, there was a trend toward lower breast cancer risk among physically active women who were postmenopausal at enrollment (RR, 0.66; 95% CI, 0.42-1.04), although no such trend was evident for premenopausal women (RR, 1.02; 95% CI, 0.87-1.19; P for interaction 0.07).
Our data from a large prospective cohort study of primarily premenopausal women do not show any evidence that physical activity in adult life reduces the risk of breast cancer occurring in the first decade of follow-up. There was also no evidence of benefit from physical activity done at younger ages, or from a consistently high level of activity. More specifically, our results pertain to physical activity assessed before age 50 and to breast cancer diagnosed before age 60, but chiefly in premenopausal women. Risk estimates seemed largely unconfounded by other established breast cancer risk factors. There was some evidence of effect modification by menopausal status at enrollment, but this was of borderline statistical significance.
A large number of epidemiologic studies have now examined the association between physical activity and the risk of breast cancer. When these were recently reviewed, 8 of 14 cohort studies, and 14 of 19 case control studies observed an inverse association between physical activity and breast cancer incidence (2). Nine of the cohort studies measured leisure time physical activity, either alone or in addition to occupational physical activity (6-8, 19-24). In this review, no clear pattern of risk by menopausal status at the time of measurement of physical activity or at cancer incidence was evident. The Iowa Women's Health Study enrolled exclusively postmenopausal women, and found no association between physical activity and breast cancer (21), whereas several other smaller studies with primarily postmenopausal cancer cases found a protective effect of physical activity (20, 22). In a population-based Norwegian study, risk of breast cancer was reduced at higher levels of physical activity and the effect was at least as large in premenopausal women and women under age 45 (6). The Nurses' Health Study, with the largest number of cases, found an approximately 20% reduction in breast cancer risk in the highest category of physical activity, and also reported similar findings in both premenopausal and postmenopausal women(8). However, the Nurses' Health Study II, which enrolled much younger women with primarily premenopausal cancer found no protective effect of physical activity during either 6 or 10 years of follow-up (7, 25).
Subsequently, seven large cohort studies have been newly published or updated, contributing information on thousands of additional cases of breast cancer. Of these recent cohort studies, three studied exclusively postmenopausal women and found about a 25% to 30% reduction in the risk of breast cancer in the most compared with the least physically active groups (26-28). The remaining studies enrolled both premenopausal and postmenopausal women, and three found that a protective effect of physical activity on breast cancer was restricted to women who were either postmenopausal or at least 50 years old at enrollment (9-11). The only null study was a population-based study in Finland, which found no association of physical activity with breast cancer in any age group (29). Thus, taking into account recent reports, a reduction of postmenopausal breast cancer risk by physical activity seems quite well-established and consistent with our finding of possible effect modification by menopausal status at enrollment. Our null study contributes data on a large cohort of primarily premenopausal women to the cohort study literature.
Disparate results between the effect of physical activity on premenopausal and postmenopausal breast cancer have been seen in several population-based case-control studies using comparable methods (30-32). Similarly, Friedenreich et al. (33, 34) found a 30% decreased risk of breast cancer among postmenopausal women who had engaged in high levels of physical activity throughout their lifetime, but no risk reduction was observed in premenopausal women. In contrast, several population-based case-control studies have found strong protective effects of physical activity among younger women (35-37).
Several lines of reasoning might explain a difference in the strength of an association between physical activity and risk of breast cancer depending on menopausal status. Rockhill et al. have hypothesized that the dissimilar results from the Nurses' Health Study I (8) and II (7) might have the result of a higher proportion of cancer due to highly penetrant genetic mutations in younger, premenopausal women (38). These types of tumors may be less amenable to prevention. Alternatively, adiposity might be in the causal pathway between physical activity and breast cancer. Adiposity is a risk factor for postmenopausal breast cancer and a protective factor in premenopausal breast cancer (2, 5). In postmenopausal women, adiposity increases the conversion of androgen to estrogen, in contrast to premenopausal women. In postmenopausal women who do not use exogenous estrogens, this conversion in adipose tissue is the main source of circulating estrogens (39).
We acknowledge that our study has several limitations. Although sedentary behavior and physical activity levels reported by the women in our study are similar to those reported in other Scandinavian studies (2, 40), we did not directly validate our measure of self-reported physical activity. It is possible that different results might have been obtained with a more sophisticated measurement. However, several studies that showed a strong inverse relationship between physical activity and breast cancer risk used a similar instrument to measure physical activity (6, 11). Also, we found linear inverse trends with BMI and the proportion of smokers with increasing activity levels, which lends validity to the measure. Furthermore, our study found a strong negative association between physical activity and total mortality7, suggesting that the measurement was sufficient to detect important variation. Although breast cancer ascertainment is likely to be nearly complete and unbiased (14, 15), we do not have data on in situ breast cancers or other important characteristics of the cancers, including size, stage or estrogen/progestin receptor status, as these are not included in the cancer registries in Norway or Sweden. Finally, menopausal status at diagnosis was estimated by using the average age of menopause (approximately at age 50 in Sweden and at age 49 in Norway; refs. 17, 18)
In summary, we did not find evidence of a protective effect of physical activity on incident breast cancer in this cohort of relatively young, moderately active, and primarily premenopausal women. Updating of exposure and risk factor data is now under way for the entire cohort. This study, as well as further follow-up of other cohorts may resolve the present uncertainties about the effect of physical activity on breast cancer in younger women.
We thank all of the women who contributed to the study.
↵7 Unpublished observation.
Grant support: In Norway, the survey was supported by grants from the National Cancer Institute of the U.S. (grant CA 52449), the Norwegian Cancer Society (grant DNK 90050), and the Aakre Foundation. In Sweden, the survey was supported by the Swedish Council for Planning and Coordination of Research, the Swedish Cancer Society, STINT (The Swedish Foundation for International Cooperation in Research and Higher Education), Organon, Pharmacia, Medical Products Agency, and Schering-Plough. K.L. Margolis received support from an award from the National Heart, Lung, and Blood Institute (K23 HL03996) and E. Weiderpass is supported by the Swedish Cancer Society.
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.
- Received March 25, 2004.
- Revision received June 1, 2004.
- Accepted August 16, 2004.
- American Association for Cancer Research