Abstract
Increased physical activity has been shown to be protective for breast cancer although few studies have examined this association in black women. In addition, limited evidence to date indicates that sedentary behavior may be an independent risk factor for breast cancer. We examined sedentary behavior and physical activity in relation to subsequent incident breast cancer in a nested case-control study within 546 cases (374 among black women) and 2,184 matched controls enrolled in the Southern Community Cohort Study. Sedentary and physically active behaviors were assessed via self-report at study baseline (2002–2009) using a validated physical activity questionnaire. Conditional logistic regression was used to estimate mutually adjusted odds ratios (OR) and corresponding 95% confidence intervals (CI) for quartiles of sedentary and physical activity measures in relation to breast cancer risk. Being in the highest versus lowest quartile of total sedentary behavior (≥12 hours/day versus <5.5 hours/day) was associated with increased odds of breast cancer among white women (OR=1.94 [95% CI 1.01–3.70], p for trend=0.1) but not black women (OR=1.23 [95% CI 0.82–1.83], p for trend=0.6) after adjustment for physical activity. After adjustment for sedentary activity, greater physical activity was associated with reduced odds for breast cancer among white women (p for trend=0.03) only. In conclusion, independent of one another, sedentary behavior and physical activity are risk factors for breast cancer among white women. Differences in these associations between black and white women require further investigation. Reducing sedentary behavior and increasing physical activity are potentially independent targets for breast cancer prevention interventions.
Keywords: Breast cancer risk, African American, physical activity, sedentary behavior
INTRODUCTION
In the United States, breast cancer incidence is lower in black women than in white women, while mortality is higher in black women (1). As a full understanding of the mechanisms underlying these differences remains elusive, there is a need to examine a wide range of relevant factors for cancer initiation, progression, and survival in racially diverse populations. Physical activity is one such factor that is particularly promising because activity behaviors are potentially modifiable, unlike many other risk factors for breast cancer identified to date such as age at menarche or age at menopause (2). Moderate-to-vigorous physical activity has been consistently linked to a reduced risk of post-menopausal breast cancer (3). Recently, time spent in sedentary behaviors has also been suggested as a risk factor for breast cancer, independent of physical activity (4). Supporting this hypothesis is evidence showing that sedentary behavior is associated with several risk factors associated with breast cancer including elevated body mass index and waist circumference, circulating C-reactive protein and fasting insulin levels, and diabetes (5, 6). However, very few epidemiologic studies have examined sedentary behaviors in relation to breast cancer risk (7–9), and none have specifically examined the association in black women. Further, a limited number of studies have examined any physical activity measures in relation to breast cancer among black women (10–13). Thus, the objective of this analysis was to examine both sedentary behaviors and physical activity in relation to incident breast cancer among black and white women in a nested case-control study within the Southern Community Cohort Study.
METHODS
Study population
The Southern Community Cohort Study (SCCS) is a prospective cohort study focused on cancer disparities related to race, socioeconomic status, and other factors (14, 15). Between 2002 and 2009, nearly 86,000 residents of 12 southern states were enrolled in the cohort, most (86%) at one of 71 participating Community Health Centers (CHCs), institutions which provide basic health and preventive services mainly to low income and uninsured persons (16). An additional 14% of participants responded to mailed questionnaires sent to residents of the 12 states randomly selected from general population sources between 2004 and 2006. All SCCS participants were required to be age 40–79 years, speak English, and not be under treatment for cancer within the past 12 months. Informed consent was obtained from each participant upon enrollment into the SCCS. Institutional Review Boards at Vanderbilt University and Meharry Medical College approved the study.
Participants enrolled at CHCs were administered an in-person structured baseline interview by a trained interviewer at enrollment, while those recruited from the general population completed and returned a paper version of the same questionnaire. The study questionnaire (available online) (17) contained questions about demographic, medical, familial, lifestyle and other participant characteristics. It also contained a physical activity questionnaire (18) and an 89-item food frequency questionnaire (19, 20).
Follow-up of the cohort for ascertainment of incident cancers is carried out via linkage to state cancer registries in eleven study enrollment states (AL, AR, FL, GA, KY, LA, MS, NC, SC, TN, VA). For this analysis, a nested case-control study of incident breast cancer was conducted within the SCCS study population based on all cases of invasive breast cancer (International Classification of Disease for Oncology codes C50.0–C50.9) diagnosed after the date of SCCS enrollment. A longitudinal study design was determined to be infeasible due to uncertainties in complete follow-up time for many of the state cancer registries. Estrogen receptor (ER) status and progesterone receptor (PR) status were abstracted from the cancer registry data supplemented by available pathology reports and medical records when registry data were incomplete.
Female SCCS participants were eligible to be selected as controls for this study if they had not been diagnosed with invasive breast cancer at the time of their matched case's diagnosis. Four controls were matched to each case by age (+/− one year), race (white, black, or other), menopausal status (pre- or post-), and enrollment location (CHC for CHC-enrollees or state for general population enrollees).
Measurement of Sedentary Behavior and Physical Activity
The baseline SCCS physical activity questionnaire (PAQ) was developed specifically for the SCCS to assess both sedentary and active behaviors done at home, at work, and during leisure-time at the time of the interview. Time spent in sedentary behaviors was assessed by asking for the amount of time per day spent sitting in a car or bus, at work, watching television or seeing movies, using a computer at home, and for other reasons (examples provided were sitting at meals, talking on the phone, reading, playing games, or sewing). Times spent in light, moderate, and strenuous activity at home and work were assessed separately for weekdays and weekends and then combined using weighted averages. During the interview, handcards were given to the participants with examples of light work (examples were standing at work, light office work, shopping, cooking, child care), moderate work (examples were manufacturing work, cleaning house, gardening, mowing lawn, home repair), and strenuous work (examples were loading trucks, construction work, farming). Two questions elicited time spent in moderate sports (examples were bowling, dancing, golf, or softball) and vigorous sports (examples were jogging, aerobics, bicycling, tennis, swimming, weight lifting, or basketball). Participants were also asked to recall time spent in light, moderate, and strenuous activity at home and work as well as in sports and exercise in their 30s.
The SCCS PAQ was evaluated in 118 participants (87 black, 31 white) using test-retest reliability methodology and via comparisons with a physical activity monitor (accelerometer) and a last-month physical activity survey administered up to four times in each participant, and there was general consistency in the magnitude of correlations between blacks and whites in the validation study (18).
Sedentary behavior was analyzed in hours/day while physically active times were transformed from hours/day into summary measures of energy expended, defined as metabolic equivalent (MET)-hours per day. Two MET-hours is roughly equivalent to walking at a moderate intensity (5 METs) for 0.5 hours or jogging at a vigorous intensity (8 METs) for 0.25 hours. MET values for specific activities and intensities were based on the values suggested in Compendium of Physical Activities (21).
The two primary exposures for this analysis were summary measures calculated as 1) total sitting (sum of all individual sedentary behaviors in hours/day) and 2) total physical activity (sum of light, moderate, and strenuous household/occupational work as well as moderate and vigorous sports in MET-hours/day).
Statistical Methods
Women were categorized by quartiles of sedentary behavior (hours/day) and active behaviors (MET-hours/day) as calculated from the distribution of the control subjects. For total sports and sitting time at work, the distributions were so highly skewed that women were classified as 1) none, 2) less than the median value for non-zero values, or 3) greater than or equal to the median value for non-zero values. Cases and controls were compared using chi-square statistics for personal characteristics and using the non-parametric Wilcoxon Rank-Sum test for sedentary behavior and physical activity measures.
Conditional logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (CI) for incident breast cancer in relation to categories of physical activity or sedentary behavior. Matching factors (age, race, menopausal status, and enrollment source) were accounted for in the conditional analysis, and additional covariates included education (<high school, high school, and vocational training/some college/college/graduate school), annual household income (<$15,000, $15,000–24,999, $25,000–49,999, and $50,000+), body mass index at age 21, cigarette smoking (current, former, or never), ever use of hormone replacement therapy (yes or no), parity (nulliparous,1,2,3,4, and 5+), age at menarche (<12 or 12+), first-degree family history of breast cancer (yes or no), and having health insurance (yes or no). Alcohol consumption, BMI at study entry, total energy consumption, and age at first birth were also considered as potential covariates but their inclusion in the models did not change the ORs appreciably and thus they are not included in the final models presented here. Total sedentary behavior and total physical activity were examined in separate multivariate models and then included in a single model as well. Hormone receptor positive cancers were defined as all cases with either ER-positive or PR-positive receptors. Hormone receptor negative cancers included all cases that were both ER-negative and PR-negative. Race-stratified models examining hormone receptor positive and hormone receptor negative cancers separately in relation to total activity and total sedentary behavior were assessed. Models limited to post-menopausal women were also examined as were models including activity measures during the 30s.
All analyses were conducted using SAS/STAT software, version 9.3 of the SAS System for Windows (SAS Institute Inc., Cary, NC)
RESULTS
From July 2002 to August 2011, 546 incident breast cancer cases were ascertained among women in the cohort and were matched to 2,184 controls on race, menopausal status, age, and enrollment location. Approximately 69% of the women were black and three-quarters were post-menopausal (Table 1). Having a mother or sister with breast cancer was more common in cases (14.5%) then in controls (10.9%) (p=0.02). Among the cases, 292 (53%) were ER-positive, 150 (27%) were ER-negative, and 104 (19%) were missing ER status; 238 (44%) were PR-positive, 198 (36%) were PR-negative, and PR status was missing for 108 (20%). Jointly, 307 of the cases were hormone receptor positive (either ER-positive or PR-positive or both) while 140 were hormone receptor negative (ER-negative and PR-negative).
Table 1.
Descriptive characteristics of incident breast cancer cases and matched controls, Southern Community Cohort Study
| Cases (N=546) | Controls (N=2,184) | ||||
|---|---|---|---|---|---|
| Covariates | Mean | [std] | Mean | [std] | p-valuea |
| Age at enrollment | 55.1 | [8.9] | 55.1 | [8.8] | 0.96 |
| Body mass index at enrollment (kg/m2) | 32.4 | [8.0] | 31.7 | [8.1] | 0.06 |
| Body mass index at age 21 (kg/m2) | 22.5 | [4.7] | 22.6 | [5.3] | 0.81 |
| Age at menarche | 12.7 | [1.8] | 12.9 | [1.9] | 0.14 |
| Total energy intake (kcal/day) | 2106.4 | [1179.3] | 2146.1 | [1223.7] | 0.51 |
| N | (%) | N | (%) | ||
|---|---|---|---|---|---|
|
|
|||||
| Raceb | |||||
| White | 152 | 27.8 | 608 | 27.8 | Matched |
| Black | 374 | 68.5 | 1496 | 68.5 | |
| Other | 20 | 3.7 | 80 | 3.7 | |
| Enrollment sourceb | |||||
| Community Health Center | 464 | 85.0 | 1856 | 85.0 | Matched |
| General population | 82 | 15.0 | 328 | 15.0 | |
| Menopausal status at baselineb | |||||
| Pre | 132 | 24.2 | 528 | 24.2 | Matched |
| Post | 414 | 75.8 | 1656 | 75.8 | |
| Highest educational attainment | |||||
| Less than high school | 132 | 25.1 | 592 | 27.9 | 0.45 |
| High school | 171 | 32.5 | 670 | 31.5 | |
| Some college, college, graduate school | 223 | 42.4 | 864 | 40.6 | |
| Annual Household Income | |||||
| <$15,000 | 275 | 53.1 | 1185 | 56.4 | 0.52 |
| $15,000–24,999 | 116 | 22.4 | 422 | 20.1 | |
| $25,000–49,999 | 80 | 15.4 | 323 | 15.4 | |
| $50,000+ | 47 | 9.1 | 173 | 8.2 | |
| Health insurance coverage | |||||
| Yes | 358 | 68.2 | 1422 | 67.0 | 0.60 |
| No | 167 | 31.8 | 701 | 33.0 | |
| Cigarette smoking | |||||
| Current | 162 | 30.6 | 611 | 28.8 | 0.41 |
| Former | 115 | 21.7 | 518 | 24.4 | |
| Never | 252 | 47.6 | 995 | 46.9 | |
| Alcohol consumption (drinks/day) | |||||
| 0 | 304 | 58.7 | 1222 | 58.2 | 0.70 |
| <1 | 156 | 30.1 | 664 | 31.6 | |
| 1+ | 58 | 11.2 | 214 | 10.2 | |
| Use of Female Replacement Hormone Therapy | |||||
| Ever | 191 | 36.2 | 727 | 34.2 | 0.37 |
| Cases (N=546) | Controls (N=2,184) | ||||
|---|---|---|---|---|---|
| Never | 336 | 63.8 | 1400 | 65.8 | |
| Parity | |||||
| Nulliparous | 51 | 9.7 | 233 | 11.0 | 0.46 |
| 1 | 85 | 16.2 | 291 | 13.7 | |
| 2 | 129 | 24.5 | 500 | 23.5 | |
| 3 | 112 | 21.3 | 429 | 20.2 | |
| 4 | 64 | 12.2 | 303 | 14.3 | |
| 5+ | 85 | 16.2 | 371 | 17.4 | |
| Age at menarche (years) | |||||
| <12 | 108 | 20.7 | 417 | 19.8 | 0.67 |
| 12+ | 415 | 79.4 | 1686 | 80.2 | |
| Family history of breast cancerc | |||||
| Yes | 79 | 14.5 | 237 | 10.9 | 0.02 |
| No | 467 | 85.5 | 1947 | 89.2 | |
| Sedentary and Physical Activity measures | Mean | [std] | Mean | [std] | p-value |
|---|---|---|---|---|---|
|
|
|||||
| Sitting (hours/day) | |||||
| Car or bus | 1.3 | [1.6] | 1.2 | [1.5] | 0.25 |
| At work | 1.5 | [2.6] | 1.4 | [2.5] | 0.21 |
| TV or movies | 3.7 | [2.9] | 3.8 | [3.0] | 0.58 |
| Home computer | 0.5 | [1.1] | 0.5 | [1.2] | 0.43 |
| Otherd | 2.5 | [2.0] | 2.4 | [2.0] | 0.15 |
| TOTAL | 9.5 | [4.8] | 9.2 | [5.2] | 0.07 |
| Household/occupational activity (MET-hours/day) | |||||
| Light | 7.8 | [6.1] | 7.8 | [6.1] | 0.81 |
| Moderate | 9.2 | [8.0] | 9.1 | [7.9] | 0.82 |
| Strenuous | 2.3 | [7.0] | 2.6 | [7.8] | 0.23 |
| Sports (MET-hours/day) | |||||
| Moderate | 0.3 | [0.8] | 0.3 | [0.9] | 0.74 |
| Vigorous | 0.4 | [1.3] | 0.6 | [1.8] | 0.73 |
| Total Physical Activity (MET-hours/day)e | 20.0 | [15.5] | 20.3 | [15.9] | 0.98 |
NOTE: Among cases, missing values include N=20 for education, N=28 for income, N=21 for health insurance, N=17 for smoking, N=28 for alcohol consumption, N=19 for replacement hormone therapy, N=23 for age at menarche, and N=20 for parity. Among controls, missing values include N=58 for education, N=81 for income, N=61 for health insurance, N=60 for smoking, N=84 for alcohol consumption, N=57 for replacement hormone therapy, N=81 for age at menarche, and N=57 for parity.
p-values for covariates based on t-tests and p-values for sedentary and physical activity measures from Wilcoxon Mann-Whitney non-parametric tests.
Matching factors
Mother or full sister
Other sitting activities include sitting at meals, talking on the phone, reading, playing cards, or sewing.
Total physical activity includes light, moderate, and strenuous household/occupational activity as well as moderate and vigorous sports.
Mean time spent in sedentary behaviors was 9.5 hours/day in cases versus 9.2 hours in controls, and mean total physical activity time was 20.0 MET-hours/day in cases and controls (Table 1). Quartiles for active and sedentary measures were computed in the entire group of controls (black, white, and other race combined). Combined cut-points for the behavior measures were utilized because the race-specific cut-points were very similar to one another; for total activity (MET-hours/day), the 25th, 50th and 75th percentiles were 10.3, 17.3, and 27.4, respectively, for white controls, and 8.6, 16.5, and 27.0 for black controls, while for total sedentary behavior(hours/day) these percentile values were 5.8, 8.5, and 12.0 for white controls and 5.5, 8.0, and 12.0 for black controls.
In multivariate models, increased time in sedentary behaviors was associated with significantly increased odds of breast cancer among white women but not among black women (OR [95% CI]= 2.04 [1.07–3.86] for whites and 1.20 [0.81–1.77] for blacks) comparing the highest versus lowest quartiles of sitting) (Table 2). Further adjustment for total activity (i.e. household and occupational activity and sports/exercise activity) did not notably affect the ORs for sedentary behavior (Table 2). An inverse trend was seen for increasing total physical activity in relation to odds of breast cancer among white women (OR [95% CI]=0.54 [0.29–1.01]) but not black women (OR [95% CI]=1.14 [0.78–1.67]) (Table 2).
Table 2.
Odds Ratios (OR) and 95% confidence intervals (CI) from conditional logistic regression models examining incident breast cancer in relation to total sedentary behavior (hours/day) and total physical activity (MET-hours/day)
| All women | White Women | Black Women | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Quartilesa | Cases | OR | 95% CI | Cases | OR | 95% CI | Cases | OR | 95% CI | |||
| Total sedentary behavior (hours/day) b | ||||||||||||
| < 5.5 | 87 | 1.00 | Ref | 19 | 1.00 | Ref | 65 | 1.00 | Ref | |||
| 5.5 – 8.1 | 125 | 1.27 | 0.93 | 1.74 | 36 | 1.52 | 0.82 | 2.83 | 86 | 1.24 | 0.85 | 1.81 |
| 8.2 – 11.9 | 114 | 1.23 | 0.89 | 1.70 | 35 | 1.79 | 0.94 | 3.42 | 76 | 1.08 | 0.73 | 1.59 |
| ≥ 12 | 132 | 1.37 | 0.99 | 1.90 | 42 | 2.04 | 1.07 | 3.86 | 86 | 1.20 | 0.81 | 1.77 |
| p for trend | 0.81 | 0.06 | 0.41 | |||||||||
| Total sedentary behavior with adjustment for total activity (hours/day) | ||||||||||||
| < 5.5 | 86 | 1.00 | Ref | 19 | 1.00 | Ref | 64 | 1.00 | Ref | |||
| 5.5 – 8.1 | 124 | 1.29 | 0.94 | 1.77 | 35 | 1.50 | 0.80 | 2.81 | 86 | 1.26 | 0.86 | 1.85 |
| 8.2 – 11.9 | 113 | 1.25 | 0.90 | 1.73 | 35 | 1.71 | 0.89 | 3.30 | 75 | 1.09 | 0.74 | 1.61 |
| ≥ 12 | 132 | 1.41 | 1.01 | 1.95 | 42 | 1.94 | 1.01 | 3.70 | 86 | 1.23 | 0.82 | 1.83 |
| p for trend | 0.60 | 0.10 | 0.58 | |||||||||
| Total Activity (MET-hours/day) c | ||||||||||||
| < 9.0 | 104 | 1.00 | Ref | 35 | 1.00 | Ref | 67 | 1.00 | Ref | |||
| 9.0 – 16.5 | 129 | 1.26 | 0.94 | 1.70 | 36 | 1.04 | 0.60 | 1.83 | 89 | 1.34 | 0.93 | 1.92 |
| 16.6 – 27.0 | 120 | 1.13 | 0.83 | 1.53 | 35 | 0.75 | 0.42 | 1.32 | 81 | 1.26 | 0.86 | 1.83 |
| ≥ 27.1 | 106 | 0.97 | 0.71 | 1.33 | 26 | 0.54 | 0.29 | 1.01 | 77 | 1.14 | 0.78 | 1.67 |
| p for trend | 0.27 | 0.02 | 0.96 | |||||||||
| Total Activity with adjustment for total sedentary behavior (MET-hours/day) | ||||||||||||
| < 9.0 | 103 | 1.00 | Ref | 35 | 1.00 | Ref | 66 | 1.00 | Ref | |||
| 9.0 – 16.5 | 128 | 1.28 | 0.94 | 1.73 | 36 | 1.08 | 0.61 | 1.90 | 88 | 1.35 | 0.94 | 1.95 |
| 16.6 – 27.0 | 118 | 1.13 | 0.83 | 1.53 | 34 | 0.78 | 0.44 | 1.38 | 80 | 1.26 | 0.86 | 1.83 |
| ≥ 27.1 | 106 | 0.99 | 0.72 | 1.36 | 26 | 0.59 | 0.31 | 1.11 | 77 | 1.15 | 0.78 | 1.69 |
| p for trend | 0.27 | 0.03 | 0.93 | |||||||||
NOTE: Analysis of all women includes women of other/mixed race. Matching factors (age, race, menopausal status, and enrollment source) were accounted for in the conditional analysis. Additional covariates included in the models were education, household income, body mass index at age 21, cigarette smoking, ever use of hormone replacement therapy, parity, age at menarche, first-degree family history of breast cancer, and having health insurance (categories for all shown in Table 1).
Quartiles were determined from the distribution of the controls.
Total sedentary behavior includes sitting in a car or bus, sitting at work, watching television or movies, using a home computer, and other sitting (examples included sitting at meals, talking on the phone, reading, playing games, or sewing)
Total activity includes light, moderate, and vigorous household and occupational activity as well as moderate and vigorous sports/exercise.
Subtypes of breast cancer based on ER and PR status were examined in relation to total activity and total sedentary behavior although small numbers of cases, particularly among white women, limited precision (Table 3). Both white and black women with hormone receptor positive tumors in the highest quartile of sedentary behavior had increased odds of breast cancer (OR [95% CI]=2.86 [1.15–7.11] for white women and OR [95% CI]=1.94 [1.09–3.44] for black women). There was a significant inverse trend among white women with hormone receptor positive cancers for the association between total physical activity and odds of breast cancer (p=0.05) (Table 3).
Table 3.
Odds Ratios (OR) and 95% confidence intervals (CI) from conditional logistic regression models examining incident breast cancer in relation to total sedentary behavior (hours/day) and total physical activity (MET-hours/day) according to hormone receptor statusa
| All women | White Women | Black Women | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Quartilesb | Cases | OR | 95% CI | Cases | OR | 95% CI | Cases | OR | 95% CI | |||
| TOTAL SEDENTARY BEHAVIOR c | ||||||||||||
| Hormone Receptor Positive | ||||||||||||
| < 5.5 | 37 | 1.00 | Ref | 9 | 1.00 | Ref | 26 | 1.00 | Ref | |||
| 5.5 – 8.1 | 71 | 1.67 | 1.06 | 2.65 | 21 | 2.02 | 1.80 | 5.10 | 48 | 1.81 | 1.04 | 3.15 |
| 8.2 – 11.9 | 71 | 1.86 | 1.18 | 2.93 | 24 | 2.95 | 1.19 | 7.29 | 46 | 1.62 | 0.93 | 2.82 |
| ≥ 12 | 81 | 2.10 | 1.32 | 3.34 | 28 | 2.86 | 1.15 | 7.11 | 51 | 1.94 | 1.09 | 3.44 |
| p for trend | 0.05 | 0.04 | 0.31 | |||||||||
| Hormone Receptor Negative | ||||||||||||
| < 5.5 | 31 | 1.00 | Ref | 8 | 1.00 | Ref | 22 | 1.00 | Ref | |||
| 5.5 – 8.1 | 30 | 0.89 | 0.48 | 1.65 | 5 | 0.45 | 0.09 | 2.22 | 25 | 1.09 | 0.52 | 2.29 |
| 8.2 – 11.9 | 26 | 0.64 | 0.34 | 1.22 | 5 | 0.79 | 0.10 | 6.47 | 21 | 0.77 | 0.36 | 1.61 |
| ≥ 12 | 30 | 0.80 | 0.43 | 1.51 | 10 | 1.83 | 0.37 | 9.06 | 20 | 0.73 | 0.34 | 1.56 |
| p for trend | 0.45 | 0.34 | 0.27 | |||||||||
| TOTAL ACTIVITY d | ||||||||||||
| Hormone Receptor Positive | ||||||||||||
| < 9.0 | 52 | 1.00 | Ref | 20 | 1.00 | Ref | 31 | 1.00 | Ref | |||
| 9.0 – 16.5 | 80 | 1.63 | 1.08 | 2.45 | 28 | 1.44 | 0.68 | 3.06 | 50 | 1.84 | 1.10 | 3.09 |
| 16.6 – 27.0 | 68 | 1.22 | 0.81 | 1.85 | 19 | 0.79 | 0.37 | 1.69 | 47 | 1.51 | 0.87 | 2.57 |
| ≥ 27.1 | 60 | 1.00 | 0.65 | 1.56 | 15 | 0.61 | 0.26 | 1.47 | 43 | 1.26 | 0.72 | 2.19 |
| p for trend | 0.05 | 0.05 | 0.30 | |||||||||
| Hormone Receptor Negative | ||||||||||||
| < 9.0 | 29 | 1.00 | Ref | 6 | 1.00 | Ref | 23 | 1.00 | Ref | |||
| 9.0 – 16.5 | 31 | 1.02 | 0.55 | 1.89 | 3 | 0.40 | 0.05 | 2.87 | 28 | 1.18 | 0.59 | 2.37 |
| 16.6 – 27.0 | 25 | 0.78 | 0.41 | 1.50 | 9 | 2.21 | 0.38 | 12.90 | 15 | 0.59 | 0.27 | 1.29 |
| ≥ 27.1 | 32 | 0.96 | 0.52 | 1.76 | 10 | 2.16 | 0.37 | 12.50 | 22 | 0.84 | 0.42 | 1.70 |
| p for trend | 0.99 | 0.71 | 0.78 | |||||||||
NOTE: Analysis of all women includes women of other/mixed race. Matching factors (age, race, menopausal status, and enrollment source) were accounted for in the conditional analysis. Additional covariates included in the models were education, household income, body mass index at age 21, cigarette smoking, ever use of hormone replacement therapy, parity, age at menarche, first-degree family history of breast cancer, and having health insurance (categories for all shown in Table 1).
Hormone receptor positive cases include all cases that are either ER-positive or PR-positive. Hormone receptor negative cases include all cases that are both ER-negative and PR-negative.
Quartiles were determined from the distribution of the controls.
Total sedentary behavior includes sitting in a car or bus, sitting at work, watching television or movies, using a home computer, and other sitting (examples included sitting at meals, talking on the phone, reading, playing games, or sewing). Models adjusted for total activity.
Total activity includes light, moderate, and vigorous household and occupational activity as well as moderate and vigorous sports/exercise. Models adjusted for total sitting.
Individual components of physical activity (light household and occupational activity, moderate and vigorous household and occupational activity, and moderate and vigorous sports/exercise) and sedentary behaviors were examined in relation to breast cancer (Table 4). None of the individual physical activity behaviors was associated with significant decreases in breast cancer risk although there was some indication of reduced odds of developing breast cancer among white women with increasing moderate and vigorous activity (p for trend = 0.06). For individual sedentary behaviors, there was no increase in breast cancer risk as time spent viewing television or movies increased (Table 4).
Table 4.
Odds Ratios (OR) and 95% confidence intervals (CI) from conditional logistic regression models examining incident breast cancer in relation to individual components of sedentary behavior and physically active behaviors
| All women | White Women | Black Women | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Quartilesa | Cases | OR | 95% CI | Cases | OR | 95% CI | Cases | OR | 95% CI | |||
| SEDENTARY BEHAVIORS | ||||||||||||
| Sitting in a car or bus (hours/day) | ||||||||||||
| < 0.33 | 108 | 1.00 | Ref | 27 | 1.00 | Ref | 77 | 1.00 | Ref | |||
| 0.33 – 0.99 | 198 | 0.88 | 0.66 | 1.18 | 61 | 0.85 | 0.47 | 1.54 | 131 | 0.90 | 0.64 | 1.28 |
| 1.0 – 1.99 | 83 | 1.06 | 0.74 | 1.51 | 27 | 1.28 | 0.62 | 2.64 | 53 | 0.97 | 0.63 | 1.50 |
| ≥ 2.0 | 68 | 1.05 | 0.72 | 1.53 | 16 | 1.26 | 0.56 | 2.85 | 52 | 1.06 | 0.68 | 1.65 |
| p for trend | 0.20 | 0.20 | 0.25 | |||||||||
| Sitting at work (hours/day) | ||||||||||||
| None | 285 | 1.00 | Ref | 80 | 1.00 | Ref | 195 | 1.00 | Ref | |||
| > 0 – 2.9 | 86 | 1.15 | 0.85 | 1.56 | 18 | 1.14 | 0.58 | 2.27 | 65 | 1.13 | 0.80 | 1.61 |
| ≥ 3 | 88 | 1.13 | 0.82 | 1.56 | 34 | 1.64 | 0.90 | 2.99 | 54 | 0.96 | 0.65 | 1.44 |
| p for trend | 0.63 | 0.10 | 0.54 | |||||||||
| Watching television or movies (hours/day) | ||||||||||||
| < 2 | 182 | 1.00 | Ref | 59 | 1.00 | Ref | 119 | 1.00 | Ref | |||
| 2 – 2.9 | 88 | 1.01 | 0.75 | 1.37 | 31 | 1.41 | 0.79 | 2.52 | 55 | 0.86 | 0.59 | 1.27 |
| 3 – 4.9 | 98 | 0.91 | 0.68 | 1.23 | 19 | 0.75 | 0.39 | 1.42 | 74 | 0.90 | 0.64 | 1.28 |
| ≥ 5 | 91 | 0.97 | 0.70 | 1.35 | 23 | 1.13 | 0.59 | 2.19 | 66 | 0.95 | 0.64 | 1.39 |
| p for trend | 0.31 | 0.45 | 0.45 | |||||||||
| Other Sitting (hours/day) b | ||||||||||||
| < 1 | 112 | 1.00 | Ref | 21 | 1.00 | Ref | 90 | 1.00 | Ref | |||
| 1 – 1.9 | 111 | 1.27 | 0.93 | 1.73 | 29 | 1.42 | 0.74 | 2.72 | 79 | 1.17 | 0.82 | 1.68 |
| 2 – 3.9 | 144 | 1.30 | 0.97 | 1.74 | 54 | 1.97 | 1.08 | 3.63 | 84 | 1.03 | 0.72 | 1.46 |
| ≥ 4 | 90 | 1.28 | 0.91 | 1.80 | 28 | 1.56 | 0.77 | 3.14 | 59 | 1.13 | 0.75 | 1.71 |
| p for trend | 0.52 | 0.11 | 0.63 | |||||||||
| PHYSICALLY ACTIVE BEHAVIORS | ||||||||||||
| Light Household and Occupational Activity (MET-hours/day) c | ||||||||||||
| < 3.3 | 112 | 1.00 | Ref | 31 | 1.00 | Ref | 78 | 1.00 | Ref | |||
| 3.3 – 5.8 | 102 | 1.04 | 0.76 | 1.43 | 32 | 1.27 | 0.70 | 2.31 | 68 | 0.96 | 0.65 | 1.41 |
| 5.9 – 11.1 | 128 | 1.02 | 0.74 | 1.39 | 38 | 1.24 | 0.68 | 2.28 | 86 | 0.92 | 0.63 | 1.36 |
| ≥ 11.2 | 115 | 1.00 | 0.72 | 1.40 | 31 | 1.08 | 0.57 | 2.02 | 80 | 0.96 | 0.64 | 1.43 |
| p for trend | 0.95 | 0.82 | 0.91 | |||||||||
| Moderate and Vigorous Household and Occupational Activity (MET-hours/day) d | ||||||||||||
| < 4 | 89 | 1.00 | Ref | 33 | 1.00 | Ref | 53 | 1.00 | Ref | |||
| 4 – 7.9 | 107 | 1.15 | 0.83 | 1.61 | 33 | 1.01 | 0.54 | 1.89 | 72 | 1.24 | 0.82 | 1.88 |
| 8 – 15.6 | 158 | 1.24 | 0.91 | 1.71 | 39 | 0.73 | 0.40 | 1.33 | 114 | 1.48 | 1.00 | 2.19 |
| ≥ 15.7 | 102 | 0.97 | 0.69 | 1.38 | 27 | 0.60 | 0.31 | 1.18 | 72 | 1.14 | 0.74 | 1.76 |
| p for trend | 0.46 | 0.06 | 0.84 | |||||||||
| Moderate and Vigorous Sports/Exercise (MET-hours/day) | ||||||||||||
| None | 332 | 1.00 | Ref | 91 | 1.00 | Ref | 229 | 1.00 | Ref | |||
| > 0 – 2.0 | 63 | 1.15 | 0.83 | 1.59 | 20 | 1.97 | 1.03 | 3.74 | 42 | 0.99 | 0.67 | 1.46 |
| ≥ 2.1 | 61 | 0.93 | 0.67 | 1.29 | 20 | 0.96 | 0.52 | 1.77 | 41 | 0.98 | 0.66 | 1.47 |
| p for trend | 0.17 | 0.21 | 0.56 | |||||||||
NOTE: Each model examining an individual sedentary behavior is also adjusted for other sedentary behaviors as well as total activity. Each model examining an individual physically active behavior is also adjusted for the other physically active behaviors as well as total sedentary time. Matching factors (age, race, menopausal status, and enrollment source) were accounted for in the conditional analysis. Additional covariates included in the models were education, household income, body mass index at age 21, cigarette smoking, ever use of hormone replacement therapy, parity, age at menarche, first-degree family history of breast cancer, and having health insurance (categories for all shown in Table 1).
Quartiles were determined from the distribution of the controls. Due to highly skewed distributions, the cut-points used for total sports/exercise and sitting at work were none, <median, and median+ (median = 2.1 MET-hours/week for total sports/exercise and median=3 hours/day for sitting at work).
Other sitting includes using a computer at home, sitting at meals, talking on the phone, reading, playing games, or sewing.
Examples of light household and occupational activity were standing at work, light office work, shopping, cooking, and child care.
Examples of moderate and vigorous household and occupational activity were manufacturing work, cleaning house, gardening, mowing lawn, home repair, loading trucks, construction work, and farming.
In models limited to women who were post-menopausal at diagnosis (371 cases) (data not shown), increased time spent in sedentary behaviors was positively associated with odds of breast cancer overall (OR=1.55 [1.08–2.23]), and the difference in magnitude between race groups was amplified, with ORs [95% CI] of 2.68 (1.30–5.53) for white women and 1.24 (0.80–1.93) for black women for the highest versus lowest quartile. Total physical activity was not significantly associated with breast cancer risk in this subset of post-menopausal women in either race group although the ORs among white women with increasing activity level did show an inverse association (OR [95% CI]=0.82 [0.43–1.54], 0.75 [0.40–1.40], and 0.64 [0.32–1.27], for quartiles 2,3, and 4 compared to the lowest activity quartile; p=0.11 for trend).
Most associations between physical activity during the participants' 30s and incident breast cancer were null (Table 5). Among black women, there was some indication that women with the highest level of sports/exercise had a reduced risk of breast cancer (OR=0.73 [0.51–1.05]) compared to women reporting no sports and exercise in this decade.
Table 5.
Odds Ratios (OR) and 95% confidence intervals (CI) from conditional logistic regression models examining incident breast cancer in relation to physical activity behaviors during the 30s.
| All women | White Women | Black Women | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Quartilesa | Cases | OR | 95% CI | Cases | OR | 95% CI | Cases | OR | 95% CI | |||
| Total Activity (MET-hours/day) b | ||||||||||||
| < 30.4 | 101 | 1.00 | Ref | 22 | 1.00 | Ref | 75 | 1.00 | Ref | |||
| 30.4 – 45.6 | 122 | 1.19 | 0.87 | 1.63 | 38 | 1.69 | 0.90 | 3.19 | 83 | 1.14 | 0.77 | 1.66 |
| 45.6 – 68.0 | 117 | 1.06 | 0.77 | 1.46 | 37 | 1.27 | 0.67 | 2.42 | 76 | 1.01 | 0.69 | 1.49 |
| ≥ 68.1 | 108 | 1.02 | 0.73 | 1.45 | 33 | 1.59 | 0.78 | 3.23 | 72 | 0.92 | 0.61 | 1.39 |
| p for trend | 0.76 | 0.32 | 0.41 | |||||||||
| Light Household and Occupational Activity (MET-hours/day) c | ||||||||||||
| < 9.2 | 101 | 1.00 | Ref | 31 | 1.00 | Ref | 68 | 1.00 | Ref | |||
| 9.2 – 16.0 | 131 | 1.02 | 0.74 | 1.40 | 34 | 0.89 | 0.48 | 1.65 | 94 | 1.12 | 0.76 | 1.65 |
| 16.1 – 18.4 | 113 | 0.95 | 0.69 | 1.31 | 27 | 0.90 | 0.47 | 1.73 | 81 | 0.99 | 0.67 | 1.45 |
| ≥ 18.5 | 104 | 1.05 | 0.75 | 1.49 | 38 | 1.27 | 0.69 | 2.34 | 63 | 1.02 | 0.66 | 1.59 |
| p for trend | 0.87 | 0.96 | 0.67 | |||||||||
| Moderate and Vigorous Household and Occupational Activity (MET-hours/day) d | ||||||||||||
| < 12 | 77 | 1.00 | Ref | 21 | 1.00 | Ref | 55 | 1.00 | Ref | |||
| 12 – 27.9 | 150 | 1.43 | 1.02 | 2.00 | 39 | 1.51 | 0.78 | 2.92 | 107 | 1.47 | 0.97 | 2.21 |
| 28 – 47.9 | 109 | 1.18 | 0.83 | 1.69 | 37 | 1.83 | 0.92 | 3.64 | 67 | 0.98 | 0.64 | 1.52 |
| ≥ 48 | 113 | 1.23 | 0.85 | 1.78 | 33 | 1.74 | 0.81 | 3.74 | 77 | 1.10 | 0.71 | 1.71 |
| p for trend | 0.72 | 0.34 | 0.83 | |||||||||
| Moderate and Vigorous Sports/Exercise (MET-hours/day) | ||||||||||||
| None | 156 | 1.00 | Ref | 36 | 1.00 | Ref | 112 | 1.00 | Ref | |||
| > 0 – 2.0 | 143 | 0.90 | 0.68 | 1.19 | 42 | 1.23 | 0.70 | 2.17 | 98 | 0.85 | 0.60 | 1.19 |
| ≥ 2.1 | 149 | 0.79 | 0.59 | 1.06 | 52 | 1.09 | 0.62 | 1.90 | 96 | 0.73 | 0.51 | 1.05 |
| p for trend | 0.13 | 0.80 | 0.09 | |||||||||
NOTE: Analysis of all women includes women of other/mixed race. Matching factors (age, race, menopausal status, and enrollment source) were accounted for in the conditional analysis. Quartiles of baseline total physical activity and total sedentary behavior were included in the models. Additional covariates included in the models were education, household income, body mass index at age 21, cigarette smoking, ever use of hormone replacement therapy, parity, age at menarche, first-degree family history of breast cancer, and having health insurance (categories for all shown in Table 1).
Quartiles were determined from the distribution of the controls.
Total activity includes light, moderate, and vigorous household and occupational activity as well as moderate and vigorous sports/exercise.
Examples of light household and occupational activity were standing at work, light office work, shopping, cooking, and child care.
Examples of moderate and vigorous household and occupational activity were manufacturing work, cleaning house, gardening, mowing lawn, home repair, loading trucks, construction work, and farming.
DISCUSSION
This prospective study is to our knowledge the first to investigate associations between sedentary behaviors and breast cancer risk among black and white middle age and older women, and is one of only a few studies to date reporting on associations between physical activity and breast cancer in black women. We found that longer time spent in sedentary behaviors was associated with increased odds of developing breast cancer among white but not black women, and that increased amounts of total physical activity were associated with decreased odds of breast cancer only in white women as well.
Investigations of associations of physical activity and sedentary behaviors with breast cancer risk are of particular importance because these are potentially modifiable behaviors, in contrast to many other known risk factors for breast cancer such as family history, age at menarche, and parity (2). Although some individual studies have shown inconsistencies, the totality of the evidence indicates that physical activity reduces the risk of breast cancer (3, 22). Across 73 studies of breast cancer reviewed in 2010, an average 25% reduction in risk was observed in women engaged in the highest versus lowest level of physical activity (23). Our results for white women are consistent with the results of this review showing a modest but significant reduction in breast cancer risk in women engaged in the highest versus lowest levels of physical activity.
While there is a large body of literature regarding associations between physical activity and breast cancer in white women, few studies have been conducted in black women. We found no significant reduction in risk among black women for increased physical activity at baseline but some indication that higher levels of sports and exercise in the 30s was associated with decreased risk for developing breast cancer later in life. Two prior studies limited to black women reported reductions in breast cancer risk associated with strenuous exercise at age 21, age 30, and age 40 (post-menopausal breast cancer only) (11), and with vigorous physical activity (13). A case-control study including black and white women in the San Francisco Bay Area examined lifetime moderate and vigorous activity (including activity that was recreational, related to transportation and chores, and occupational) and found modest reductions in breast cancer risk among pre- and postmenopausal black and white women with the highest levels of activity compared to the lowest (10). In a large case-control study conducted in Atlanta, Detroit, Los Angeles, and Seattle, high lifetime exercise activity (defined as ≥ 3 hours of exercise per week versus inactive) was found to be associated with a moderate reduction in breast cancer in both black (OR=0.75 [0.61–0.93]) and white women (OR=0.83 [0.70–0.98]) (12). In contrast to these two studies, in the SCCS, we found dissimilar patterns between black and white women for total physical activity in relation to breast cancer. These differences between studies could be attributed to multiple factors such as those related to measurement including potentially differential reporting of activity between women of different races; differences in ascertainment of physical activity in terms of domain, type, duration, frequency, and timing, between studies; or study timing. Differences in underlying biologic processes whereby physical activity may influence risk of breast cancer in white versus black women (such as adipose-derived biomarkers or patterns of fat distribution) may also cause differences in observed associations, but are difficult to compare across studies.
Sedentary behavior, in contrast to physically active behaviors, is a relatively newly hypothesized risk factor for cancer. A 2010 review identified 11 studies examining some measure of sedentary behavior with a cancer outcome (not limited to breast), with 8 showing a positive association (24). To date, very few studies have examined sedentary behavior specifically in relation to breast cancer risk. In a case-control study of white women in Poland, with accelerometer-based measures of sedentary behavior, elevated odds ratios for breast cancer were found in the highest versus lowest quartile of sedentary behavior (OR=1.81), a finding very similar to our results among white women (9). Two other studies have examined self-reported television viewing time as a proxy for sedentary behavior in relation to breast cancer; in a hospital-based case-control study in India, television viewing was not associated with incident breast cancer (8), while in the US-based NIH-AARP Diet and Health Study, television viewing did show a modest, positive but not statistically significant association with incident breast cancer risk (7). Another component of sedentary behavior, occupational sitting time, has been assessed in relation to breast cancer risk. In the Shanghai Women's Health Study, women with the most sitting at work were found to have increased risk of breast cancer compared with women with more active jobs (25), while in the Netherlands Cohort Study, the association between sitting time at work and breast cancer was null (26). Few of these studies examined a broad range of sitting behaviors, and this may explain the larger magnitude of the findings in the SCCS which included multiple types of sitting activities.
Beyond the few epidemiologic studies of cancer outcomes, sedentary behavior has also been associated with physiological risk factors for breast cancer including increased adiposity, insulin resistance, and increased inflammation (5, 27). In laboratory studies, prolonged sedentary time led to fewer skeletal muscle contractions causing rapid reductions in lipoprotein lipase activity, which in turn leads to decreased HDL cholesterol production as well as reduced triglyceride uptake (28, 29). Another documented physiological consequence of sedentary behavior is increased insulin response to glucose loading (30). Sedentary behavior has also been linked to increased waist circumference, BMI, and weight gain (24, 31). These pathways, and possibly others yet to be explored may mediate the effect of increased sedentary behavior on risk of breast cancer (24).
In a recent review of studies of physical activity and breast cancer that have examined effect modification by ER/PR status, the authors determined that there was not clear evidence of effect modification by hormone receptor status (3). In the present study, there was some indication of breast cancer risk reduction related to higher physical activity levels among white women with hormone receptor positive tumors but not receptor negative tumors. This finding is consistent with some previous reports but not all (3). Our findings are among the first to examine sedentary behavior in relation to subtypes of breast cancer, and while precision was low due to the small numbers of cases, the data were suggestive that hormone receptor positive cancers may have as strong or stronger associations with sedentary behavior than hormone receptor negative cancers in white women. Additional follow-up in the SCCS as well as examination of both physical activity and sitting behaviors in other large, diverse study populations are needed to better understand the associations within specific subtypes of breast cancer.
An important strength of this study is the ascertainment of a broad range of physical activity and sedentary behaviors. In many previous reports analyzing sedentary behavior, television viewing has been used as a proxy for all sedentary behaviors. In contrast to using only this proxy measure, we collected information on sitting at work, in transportation, while viewing television or movies, using a computer, and other sitting activities. Among the controls of this study, television and movie viewing was the largest contributor to the total sedentary behavior time, but it represented only 40% of all sedentary behavior. The relatively low correlation between all sitting activities reported and television/movie viewing (0.59) indicate that television/movie time is not an ideal proxy for all sedentary behaviors. A second strength of our study is that our PAQ included questions about light, moderate, and vigorous household and occupational work as well as time spent in sports and exercise which allowed us to distinguish true sedentary behavior from light activity, a distinction that has not been possible in many previous studies.
Our study also has limitations which first include that physical activity was self-reported rather than objectively measured. Second, it is possible that the different results between races observed in this study were due to differences in reporting errors for black and white women. However, our evaluation of the PAQ used in this study as compared to an accelerometer did not reveal notable racial differences (18). Third, we did not measure lifetime physical activity level. However, we did assess physical activity during the 30s which ameliorated this deficiency at least in part. Finally, this study was limited by the lack of power to examine effect modification by potentially important factors such as body size or hormone replacement therapy.
In conclusion, this study found that increased time spent in sedentary behaviors is associated with increased breast cancer risk, and that this association may be different in black and white women of middle to older age. Unlike many known risk factors for breast cancer, physical activity and sedentary behavior represent promising targets for public health interventions to reduce the burden of breast cancer due to their modifiable nature.
ACKNOWLEDGEMENTS
Data on SCCS cancer cases used in this publication were provided by the Alabama Statewide Cancer Registry; Kentucky Cancer Registry, Lexington, KY; Tennessee Department of Health, Office of Cancer Surveillance; Florida Cancer Data System; North Carolina Central Cancer Registry, North Carolina Division of Public Health; Georgia Comprehensive Cancer Registry; Louisiana Tumor Registry; Mississippi Cancer Registry; South Carolina Central Cancer Registry; Virginia Department of Health, Virginia Cancer Registry; Arkansas Department of Health, Cancer Registry, 4815 W. Markham, Little Rock, AR 72205. The Arkansas Central Cancer Registry is fully funded by a grant from National Program of Cancer Registries, Centers for Disease Control and Prevention (CDC). Data on SCCS cancer cases from Mississippi were collected by the Mississippi Cancer Registry which participates in the National Program of Cancer Registries (NPCR) of the Centers for Disease Control and Prevention (CDC). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the CDC or the Mississippi Cancer Registry.
GRANT SUPPORT Funding for this research was provided by NIH grant R01 CA092447 and its American Reinvestment and Recovery Act Supplement 3R01 CA092447-08S1.
Footnotes
Potential conflicts of interest: None
REFERENCES
- 1.DeSantis C, Siegel R, Bandi P, Jemal A. Breast cancer statistics, 2011. CA Cancer J Clin. 2011;61(6):409–18. doi: 10.3322/caac.20134. Epub 2011/10/05. [DOI] [PubMed] [Google Scholar]
- 2.Colditz G, Baer HJ, Tammi RM. Breast Cancer. In: Schottenfeld D, Fraumeni JF, editors. Cancer Epidemiology And Prevention. Third ed. Oxford University Press; New York: 2006. [Google Scholar]
- 3.Lynch BM, Neilson HK, Friedenreich CM. Physical activity and breast cancer prevention. Recent results in cancer research Fortschritte der Krebsforschung Progres dans les recherches sur le cancer. 2011;186:13–42. doi: 10.1007/978-3-642-04231-7_2. Epub 2010/11/30. [DOI] [PubMed] [Google Scholar]
- 4.Friedenreich C, Lynch B. Can living a less sedentary life decrease breast cancer risk in women? Womens Health (Lond Engl) 2012;8(1):5–7. doi: 10.2217/whe.11.75. Epub 2011/12/17. [DOI] [PubMed] [Google Scholar]
- 5.Lynch BM, Friedenreich CM, Winkler EA, Healy GN, Vallance JK, Eakin EG, et al. Associations of objectively assessed physical activity and sedentary time with biomarkers of breast cancer risk in postmenopausal women: findings from NHANES (2003–2006) Breast Cancer Res Treat. 2011;130(1):183–94. doi: 10.1007/s10549-011-1559-2. Epub 2011/05/10. [DOI] [PubMed] [Google Scholar]
- 6.Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women. JAMA. 2003;289(14):1785–91. doi: 10.1001/jama.289.14.1785. Epub 2003/04/10. [DOI] [PubMed] [Google Scholar]
- 7.George SM, Irwin ML, Matthews CE, Mayne ST, Gail MH, Moore SC, et al. Beyond recreational physical activity: examining occupational and household activity, transportation activity, and sedentary behavior in relation to postmenopausal breast cancer risk. Am J Public Health. 2010;100(11):2288–95. doi: 10.2105/AJPH.2009.180828. Epub 2010/09/25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mathew A, Gajalakshmi V, Rajan B, Kanimozhi VC, Brennan P, Binukumar BP, et al. Physical activity levels among urban and rural women in south India and the risk of breast cancer: a case-control study. Eur J Cancer Prev. 2009;18(5):368–76. doi: 10.1097/CEJ.0b013e32832e1c46. Epub 2009/07/09. [DOI] [PubMed] [Google Scholar]
- 9.Dallal CM, Brinton LA, Matthews CE, Lissowska J, Peplonska B, Hartman TJ, et al. Accelerometer-based measures of active and sedentary behavior in relation to breast cancer risk. Breast Cancer Res Treat. 2012;134(3):1279–90. doi: 10.1007/s10549-012-2129-y. Epub 2012/07/04. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.John EM, Horn-Ross PL, Koo J. Lifetime physical activity and breast cancer risk in a multiethnic population: the San Francisco Bay area breast cancer study. Cancer Epidemiol Biomarkers Prev. 2003;12(11 Pt 1):1143–52. Epub 2003/12/04. [PubMed] [Google Scholar]
- 11.Adams-Campbell LL, Rosenberg L, Rao RS, Palmer JR. Strenuous physical activity and breast cancer risk in African-American women. J Natl Med Assoc. 2001;93(7–8):267–75. Epub 2001/08/09. [PMC free article] [PubMed] [Google Scholar]
- 12.Bernstein L, Patel AV, Ursin G, Sullivan-Halley J, Press MF, Deapen D, et al. Lifetime recreational exercise activity and breast cancer risk among black women and white women. J Natl Cancer Inst. 2005;97(22):1671–9. doi: 10.1093/jnci/dji374. Epub 2005/11/17. [DOI] [PubMed] [Google Scholar]
- 13.Sheppard VB, Makambi K, Taylor T, Wallington SF, Sween J, Adams-Campbell L. Physical activity reduces breast cancer risk in African American women. Ethn Dis. 2011;21(4):406–11. Epub 2012/03/21. [PMC free article] [PubMed] [Google Scholar]
- 14.Signorello LB, Hargreaves MK, Steinwandel MD, Zheng W, Cai Q, Schlundt DG, et al. Southern community cohort study: establishing a cohort to investigate health disparities. J Natl Med Assoc. 2005;97(7):972–9. [PMC free article] [PubMed] [Google Scholar]
- 15.Signorello LB, Hargreaves MK, Blot WJ. The Southern Community Cohort Study: investigating health disparities. J Health Care Poor Underserved. 2010;21(1 Suppl):26–37. doi: 10.1353/hpu.0.0245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hargreaves MK, Arnold C, Blot WJ. Community health centers: Their role in the treatment of minorities and in health disparities research. In: Satcher D, Pamies R, editors. Multicultural Medicine and Health Disparities. McGraw-Hill; New York: 2006. pp. 485–94. [Google Scholar]
- 17.Southern Community Cohort Study [Internet] [cited 2013 March 28] Available from: http://www.southerncommunitystudy.org.
- 18.Buchowski MS, Matthews CE, Cohen SS, Signorello LB, Fowke JH, Hargreaves MK, et al. Evaluation of a questionnaire to assess sedentary and active behaviors in the southern community cohort study. Journal of physical activity & health. 2012;9(6):765–75. doi: 10.1123/jpah.9.6.765. Epub 2012/08/18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Buchowski MS, Schlundt DG, Hargreaves MK, Hankin JH, Signorello LB, Blot WJ. Development of a culturally sensitive food frequency questionnaire for use in the Southern Community Cohort Study. Cell Mol Biol (Noisy-le-grand) 2003;49(8):1295–304. [PubMed] [Google Scholar]
- 20.Signorello LB, Munro HM, Buchowski MS, Schlundt DG, Cohen SS, Hargreaves MK, et al. Estimating nutrient intake from a food frequency questionnaire: incorporating the elements of race and geographic region. Am J Epidemiol. 2009;170(1):104–11. doi: 10.1093/aje/kwp098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc. 2000;32(9 Suppl):S498–504. doi: 10.1097/00005768-200009001-00009. [DOI] [PubMed] [Google Scholar]
- 22.World Cancer Research Fund. American Institute for Cancer Research . Food, nutrition,and physical activity, and the prevention of cancer: a global perspective. AICR; Washington, DC: 2007. [Google Scholar]
- 23.Lynch BM, Nielson HK, Friedenreich CM. Chapter 2: Physical Activity and Breast Cancer Prevention. In: Courneya KS, Friedenreich CM, editors. Physical Activity and Cancer Recent Results in Cancer Research. Springer-Verlag; Berlin Heidelberg: 2010. [Google Scholar]
- 24.Lynch BM. Sedentary behavior and cancer: a systematic review of the literature and proposed biological mechanisms. Cancer Epidemiol Biomarkers Prev. 2010;19(11):2691–709. doi: 10.1158/1055-9965.EPI-10-0815. Epub 2010/09/14. [DOI] [PubMed] [Google Scholar]
- 25.Pronk A, Ji BT, Shu XO, Chow WH, Xue S, Yang G, et al. Physical activity and breast cancer risk in Chinese women. Br J Cancer. 2011;105(9):1443–50. doi: 10.1038/bjc.2011.370. Epub 2011/09/22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Dirx MJ, Voorrips LE, Goldbohm RA, van den Brandt PA. Baseline recreational physical activity, history of sports participation, and postmenopausal breast carcinoma risk in the Netherlands Cohort Study. Cancer. 2001;92(6):1638–49. doi: 10.1002/1097-0142(20010915)92:6<1638::aid-cncr1490>3.0.co;2-q. Epub 2001/12/18. [DOI] [PubMed] [Google Scholar]
- 27.Thorp AA, Healy GN, Owen N, Salmon J, Ball K, Shaw JE, et al. Deleterious associations of sitting time and television viewing time with cardiometabolic risk biomarkers: Australian Diabetes, Obesity and Lifestyle (AusDiab) study 2004–2005. Diabetes Care. 2010;33(2):327–34. doi: 10.2337/dc09-0493. Epub 2009/11/18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bey L, Hamilton MT. Suppression of skeletal muscle lipoprotein lipase activity during physical inactivity: a molecular reason to maintain daily low-intensity activity. The Journal of physiology. 2003;551(Pt 2):673–82. doi: 10.1113/jphysiol.2003.045591. Epub 2003/06/20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hamilton MT, Hamilton DG, Zderic TW. Role of low energy expenditure and sitting in obesity, metabolic syndrome, type 2 diabetes, and cardiovascular disease. Diabetes. 2007;56(11):2655–67. doi: 10.2337/db07-0882. Epub 2007/09/11. [DOI] [PubMed] [Google Scholar]
- 30.Hamburg NM, McMackin CJ, Huang AL, Shenouda SM, Widlansky ME, Schulz E, et al. Physical inactivity rapidly induces insulin resistance and microvascular dysfunction in healthy volunteers. Arterioscler Thromb Vasc Biol. 2007;27(12):2650–6. doi: 10.1161/ATVBAHA.107.153288. Epub 2007/10/13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Blanck HM, McCullough ML, Patel AV, Gillespie C, Calle EE, Cokkinides VE, et al. Sedentary behavior, recreational physical activity, and 7-year weight gain among postmenopausal U.S. women. Obesity (Silver Spring, Md. 2007;15(6):1578–88. doi: 10.1038/oby.2007.187. Epub 2007/06/15. [DOI] [PubMed] [Google Scholar]