Fax: (203) 785-6279
Physical activity levels before and after a diagnosis of breast carcinoma†
The health, eating, activity, and lifestyle (HEAL) study
Article first published online: 18 MAR 2003
Copyright © 2003 American Cancer Society
Volume 97, Issue 7, pages 1746–1757, 1 April 2003
How to Cite
Irwin, M. L., Crumley, D., McTiernan, A., Bernstein, L., Baumgartner, R., Gilliland, F. D., Kriska, A. and Ballard-Barbash, R. (2003), Physical activity levels before and after a diagnosis of breast carcinoma. Cancer, 97: 1746–1757. doi: 10.1002/cncr.11227
A portion of this work was conducted through the Clinical Research Center at the University of Washington
- Issue published online: 18 MAR 2003
- Article first published online: 18 MAR 2003
- Manuscript Accepted: 18 NOV 2002
- Manuscript Revised: 8 NOV 2002
- Manuscript Received: 11 SEP 2002
- National Cancer Institute. Grant Numbers: N01-CN-75036-20, NO1-CN-05228, NO1-PC-67010, T32 CA09661
- National Institutes of Health. Grant Number: M01-RR-00037
- body weight;
Increased body weight at the time patients are diagnosed with breast carcinoma has been associated with an increased risk of recurrence and reduced survival. Weight gain also is common after diagnosis. Increasing physical activity (PA) after diagnosis may minimize these adverse outcomes. In this population-based study, the authors investigated whether PA levels after diagnosis declined from prediagnosis levels and whether any changes in PA varied by disease stage, adjuvant treatment, patient age, or body mass index (BMI) in 812 patients with incident breast carcinoma (from in situ to Stage IIIa).
Types of sports and household activities and their frequency and duration for the year prior to diagnosis and for the month prior to the interview (i.e., 4–12 months postdiagnosis) were assessed during a baseline interview.
Patients decreased their total PA by an estimated 2.0 hours per week from prediagnosis to postdiagnosis, an 11% decrease (P < 0.05). Greater decreases in sports PA were observed among women who were treated with radiation and chemotherapy (50% decrease) compared with women who underwent surgery only (24% decrease) or who were treated with radiation only (23%; (P < 0.05). Greater decreases in sports PA were observed among obese patients (41% decrease) compared with patients of normal weight (24% decrease; P < 0.05).
PA levels were reduced significantly after patients were diagnosed with breast carcinoma. Greater decreases in PA observed among heavier patients implied a potential for greater weight gain among women who already were overweight. Randomized, controlled trials are needed to evaluate how PA may improve the prognosis for patients with breast carcinoma. Cancer 2003;97:1746–57. © 2003 American Cancer Society.
Weight gain is common during the first year after patients are diagnosed with breast carcinoma, especially among women receiving systemic adjuvant treatment.1–6 In a study by Goodwin et al.2, 4 84% of patients with breast carcinoma gained weight in the first year after diagnosis, with a mean weight gain of 2.5 kg among patients receiving chemotherapy. Weight gain commonly ranges from 2.5 kg to 6.2 kg; however, greater gains are not unusual.1, 3, 5 Increased body mass and body fat at the time patients are diagnosed with breast carcinoma have been associated with an increased risk of recurrence and reduced survival.2 Less is known about the role played by weight gain in disease recurrence and survival after patients are diagnosed with breast carcinoma.
Multiple reasons for postdiagnosis weight gain in patients with breast carcinoma have been suggested, such as receiving adjuvant treatment (specifically, chemotherapy), not taking tamoxifen, decreased physical activity, and an increase in caloric intake,1–6 although these theories have not been validated scientifically. Maintaining or increasing physical activity levels after patients are diagnosed with breast carcinoma may minimize postdiagnosis weight gain, because physical activity is associated with weight maintenance in healthy women.7, 8
To our knowledge, few epidemiologic studies have been conducted that describe physical activity levels among women who have been diagnosed and treated for breast carcinoma or any malignancy.9, 10 In a study of patients with Stage I or II breast carcinoma, physical activity levels decreased significantly during chemotherapy treatment from pretreatment levels.10 Among survivors of colorectal carcinoma, treatment for their disease had a negative effect on exercise participation levels that did not recover completely posttreatment.9 We examined data collected in what we believe to be the first population-based cohort of 812 patients with breast carcinoma—the Health, Eating, Activity, and Lifestyle (HEAL) Study—to investigate changes in physical activity levels after a diagnosis of breast carcinoma relative to prediagnosis levels and to assess the influence of disease stage, adjuvant treatment, age, and body mass index (BMI) on these changes.
MATERIALS AND METHODS
Study Setting and Patients
The HEAL Study is a population-based, multicenter, multiethnic, prospective cohort study that has enrolled 1185 patients with breast carcinoma who are being followed to determine whether weight, physical activity, diet, hormone receptors, and other exposures have an effect on prognosis. Patients were recruited to the HEAL Study through the Surveillance, Epidemiology, and End Results (SEER) registries in New Mexico, Los Angeles County (CA), and Western Washington State. Comparable data on physical activity levels in patients with breast carcinoma prediagnosis and postdiagnosis are available only for New Mexico and Western Washington State. Prediagnosis data were collected for Los Angeles County participants as part of a case-control study of risk factors for breast carcinoma, and postdiagnosis data were not collected during the first year of follow-up. Therefore, these analyses were limited to patients from New Mexico and Washington State.
Recruitment and Eligibility
Different eligibility criteria were applied for New Mexico and Washington, because the New Mexico site initially was funded and started 1 year earlier than the Washington site. In both New Mexico and Washington, patients were eligible for the study if they had been diagnosed with in situ breast carcinoma to Stage IIIa breast carcinoma. However, due to other ongoing studies at the Fred Hutchinson Cancer Research Center in Seattle, Washington, a limited number of women who were treated for Stage I–IIIa breast carcinoma were eligible for the HEAL Study. Thus, a higher percent of women who were treated for in situ breast carcinoma in Washington were recruited into the study. Other eligibility criteria for recruitment into the HEAL Study required that patients were 1) diagnosed at age ≥ 18 years in New Mexico at age 40–64 years in Washington; 2) able to participate in an interview within a 12-month period after diagnosis; 3) living in defined geographic areas (King, Pierce, or Snohomish Counties in Washington and Bernalillo, Santa Fe, Sandoval, Valencia, or Taos Counties in New Mexico); and 4) diagnosed between July 1996 and March 1999 (in New Mexico) or between September 1997 and September 1998 (in Washington). Due to feasibility issues, an additional criterion for Washington participants was their willingness to come to the study site for measures. In New Mexico, if a participant was not able to come to the study site, then a staff member traveled to the participant's home to conduct the visit.
In total, 1036 age-eligible women in Washington and 1037 women in New Mexico who lived in the catchment areas were diagnosed with breast carcinoma during the specified recruitment periods. Patients were excluded from the HEAL Study for a variety of reasons: Eight hundred thirty-four women from Western Washington and 383 women from the targeted New Mexico counties were excluded due to doctor refusal (n = 59 women), patient refusal (n = 437 women), unable to locate physician or patient (n = 443 women), or interviews for other studies (n = 278 women). In total, 202 women from Washington and 654 women from New Mexico participated in the HEAL Study.
Among the 856 women enrolled, 38 women had a prior diagnosis of breast carcinoma, 3 women were diagnosed with Stage IV breast carcinoma, 1 woman had an undefined disease stage, and 2 women did not have complete physical activity data. These women were excluded, and the remaining analyses were restricted to the remaining 812 women. Written informed consent was obtained from each participant. The study was performed with the approval of the Institutional Review Boards of participating centers, in accordance with an assurance filed with and approved by the United States Department of Health and Human Services.
In Western Washington, patients completed a mailed study questionnaire, which they brought to a baseline clinic visit (scheduled within 4–12 months postdiagnosis) at the Fred Hutchinson Cancer Research Center Prevention Studies Clinic. Physical activity data were collected at an in-person interview. In-person interviews were conducted in New Mexico (scheduled within 4–12 months postdiagnosis), and physical measurements were obtained at the University of New Mexico Aging and Genetic Epidemiology Program.
Physical Activity Assessment
In both New Mexico and Washington, we collected information on physical activity using an interview-administered physical activity questionnaire at the baseline clinic visit (i.e., 4–12 months postdiagnosis).11 The questionnaire was based on the Modifiable Activity Questionnaire developed by Kriska and colleagues11 that was designed to be modified easily for use with different populations and has been demonstrated as reliable and valid. The type, duration, and frequency of activities for the year prior to breast carcinoma diagnosis and for the month prior to the interview, corresponding to the period 4–12 months postdiagnosis, were assessed. The sports/recreation and household activity section of the questionnaire addressed 29 popular activities, such as fast walking, jogging, aerobics, tennis, household cleaning, and yard work. The interviewer first read the list of activities to the participant, and the participant identified all activities that she performed on at least 10 different occasions over the past 5 years. For each activity performed, participants also were asked how often and for how long they performed the activity in the year prior to their diagnosis of breast carcinoma. We also recorded sedentary activities as the average number of hours per weekday and weekend spent watching television, sitting (not including watching television), and napping. After participation in each activity was determined for the year prior to diagnosis, the interviewer followed the exact same procedure to determine physical activity participation in the past month (i.e., postdiagnosis physical activity).
Stage of Disease and Treatment
We obtained data on stage of disease from the respective local SEER registries (the New Mexico Tumor Registry and the Cancer Surveillance System of Western Washington) and obtained information on adjuvant therapy from medical record abstraction and SEER registry records. We abstracted the radiation and/or chemotherapy sequence with surgery, type of chemotherapy agents given, and use of tamoxifen therapy. Participants also provided information on use of tamoxifen.
Trained staff measured height and weight in a standard manner at the baseline clinic visit. Weight was measured to the nearest 0.1 kg using a laboratory scale. Height was measured without shoes to the nearest 0.1 cm using a stadiometer. All measurements were performed and recorded twice in succession. The two measurements were averaged for a final value for analyses. BMI was computed as the weight in kg divided by the height in m2. Three mutually exclusive BMI groups were created: patients with normal weight (BMI < 25 kg/m2), overweight patients (25 kg/m2 ≤ BMI < 30.0 kg/m2), and obese patients (BMI ≥ 30.0 kg/m2).12
Standardized questionnaire information was collected on medical history; health habits; history of benign breast disease; age at menopause; type of menopause; hysterectomy status; family history of breast and other specific malignancies; self-reported, physician-diagnosed type 2 diabetes; smoking status; body weight 1 year prediagnosis; general health status (defined as excellent, very good, good, fair, and poor); and selected demographic data (e.g., age, education, race, ethnicity, and marital status). These forms were self-administered in Washington and were interview-administered in New Mexico.
Physical activity data
We estimated the hours per week for each activity by multiplying the frequency and duration reported. Two mutually exclusive patient groups were created based on type of activity (sports/recreation, including walking, or household/gardening). Each activity also was categorized as light intensity, moderate intensity, or vigorous intensity based on the Compendium of physical activities by Ainsworth et al.13 With an MET defined as the ratio of the associated metabolic rate for a specific activity divided by the resting metabolic rate (e.g., a 2-MET activity requires two times the resting metabolic energy expenditure of sitting quietly), light-intensity physical activity was defined as < 3 METs, moderate-intensity physical activity was defined as 3–6 METs, and vigorous-intensity physical activity was defined as > 6 METs.14 One MET also is defined as the energy expenditure for sitting quietly, which, for the average adult, is approximately 3.5 mL of oxygen per kg body weight per minute or 1 kcal per kg body weight per hour.13
The hours per week for sedentary activities also were determined by summing the duration reported for television watching, sitting, and napping by weekday (Monday through Friday) and by weekend (Saturday and Sunday). Exceptions were made for participants with schedules that differed from the norm (e.g., participants who worked Tuesday through Saturday and had Sunday and Monday off from work). Weekday and weekend sedentary hours per day were summed separately and were weighted by 5 of 7 days and 2 of 7 days, respectively, to calculate total sedentary hours per week.
Disease stage and treatment
Participants were classified with in situ breast carcinoma, Stage I breast carcinoma, or Stage II–IIIa breast carcinoma using the SEER disease classification system.15 Adjuvant treatment was categorized into four mutually exclusive groups: surgery only; surgery and radiation; surgery and chemotherapy; and surgery, radiation, and chemotherapy.
The means ± standard errors of physiologic and demographic characteristics of the study participants were calculated separately for each study site. Differences in means by study site were compared using t tests for continuous variables and chi-square analyses for categorical variable.
We used one-way analysis of covariance methods to compare prediagnosis and postdiagnosis physical activity levels by disease stage, adjuvant treatment, age, and BMI with adjustments for age; ethnicity; menopausal status; education; study site; disease stage; adjuvant treatment; tamoxifen use; season; time from diagnosis to interview; prediagnosis physical activity levels; full-time employment; marital status; smoking status; self-reported, physician-diagnosed type 2 diabetes; family history of breast carcinoma; and general health status. Least-square mean levels of physical activity for categories of disease stage, age, and BMI were fit by linear regression to determine whether the linear decrease in physical activity from prediagnosis to postdiagnosis was statistically significant. We used the Tukey honestly significant difference test to identify statistically significant differences between groups, with the overall level of statistical significance constrained to 5%.
Finally, to determine which factors predicted a decrease in physical activity levels from prediagnosis to postdiagnosis (i.e., a change from prediagnosis to postdiagnosis in the hours of physical activity per week was used as the dependent variable), we used stepwise, forward regression methods and considered the following factors as potential predictors: age; menopausal status; education; full-time employment; race; ethnicity; marital status; study site; disease stage; treatment; prediagnosis BMI; prediagnosis physical activity levels; smoking status; self-reported, physician-diagnosed type 2 diabetes; family history of breast carcinoma; number of months from diagnosis to interview; general health status; and reported weight gain during treatment. Age, prediagnosis BMI, prediagnosis physical activity levels, and number of months from diagnosis to interview were analyzed as continuous variables; the other variables were analyzed as categorical variables.
New Mexico and Washington differed on several demographic and physiologic characteristics (Table 1). Thirty-four percent, 50%, and 16% of Washington participants were treated for in situ, Stage I, or Stage II–IIIa breast carcinoma, compared with 19%, 62%, and 19% of New Mexico participants, respectively (P < 0.05). Washington participants were younger and heavier compared with New Mexico participants (P < 0.05). Only 3% of Washington participants were of Hispanic ethnicity, compared with 24% of New Mexico participants (P < 0.05).
|Characteristic||Washington (n = 197 patients)||New Mexico (n = 615) patients|
|Mean age ± SE (yrs)||52.4 ± 6.4||59.2 ± 12.6a|
|Mean weight ± SE (kg)||73.8 ± 16.2||68.7 ± 14.9a|
|Mean height ± SE (cm)||164.2 ± 6.5||161.8 ± 7.0a|
|Mean BMI ± SE (weight/height2)||27.3 ± 5.7||26.3 ± 5.7a|
|Months ± SE from diagnosis to interview||7.5 ± 2.0||5.7 ± 1.7a|
|Disease stage (%)|
|Stage II and IIIa||16||19|
|Radiation and chemotherapy||24||19a|
|Education (% high school graduates)||98||94|
|Tamoxifen users (%)||51||46a|
|Family history of breast carcinomac||21||24|
|Current smokers (%)||7||11a|
|Diagnosed with type 2 diabetes (%)||4||9a|
We found statistically significant decreases in total, moderate-intensity, vigorous-intensity, and sports/recreational physical activity from prediagnosis to postdiagnosis among patients with Stage I and Stage II–IIIa breast carcinoma adjusted for study site and other covariates (P < 0.05) (Tables 2, 3). Patients with Stage I and Stage II–IIIa breast carcinoma decreased their total physical activity from prediagnosis to postdiagnosis by 13.8% and 8.9%, respectively. Among patients who were diagnosed with in situ breast carcinoma, statistically significant decreases in vigorous-intensity and sports/recreational physical activity were observed from prediagnosis to postdiagnosis (P < 0.05). Finally, 52%, 58%, and 62% of patients with in situ, Stage I, and Stage II–IIIa breast carcinoma, respectively, decreased their total physical activity levels from prediagnosis to postdiagnosis (P < 0.05).
|Activity level||Hours per week (mean ± SE)||P value, for trend|
|In situ (n = 185 patients)||Stage I (n = 479 patients)||Stage II and IIIa (n = 148 patients)|
|Before diagnosis||18.5 ± 0.3||18.9 ± 0.1||19.2 ± 0.3||0.07|
|After diagnosis||17.8 ± 0.8||16.3 ± 0.4||17.5 ± 0.9||0.9|
|Absolute difference||−0.7 ± 0.8||−2.6 ± 0.4cd||−1.7 ± 0.9d||0.5|
|Before diagnosis||11.2 ± 0.5||11.5 ± 0.3||12.1 ± 0.5||0.2|
|After diagnosis||11.1 ± 0.6||10.1 ± 0.4||12.0 ± 0.7f||0.4|
|Absolute difference||− 0.1 ± 0.6||− 1.3 ± 0.3d||− 0.2 ± 0.7||0.9|
|% Difference||−0.9%||− 11.3%||− 1.7%||0.9|
|Before diagnosis||6.7 ± 0.4||6.8 ± 0.2||6.6 ± 0.4||0.9|
|After diagnosis||6.2 ± 0.5||5.8 ± 0.3||5.1 ± 0.6||0.2|
|Absolute difference||− 0.5 ± 0.5||− 1.0 ± 0.3d||− 1.5 ± 0.6d||0.3|
|Before diagnosis||0.6 ± 0.1||0.7 ± 0.1||0.5 ± 0.1||0.4|
|After diagnosis||0.4 ± 0.1||0.4 ± 0.1||0.2 ± 0.1cf||0.1|
|Absolute difference||− 0.2 ± 0.1d||− 0.3 ± 0.1d||− 0.3 ± 0.1d||0.5|
|Activity level||Hours per week (mean ± SE)||P value for trend|
|In situ (n = 185 patients)||Stage I (n = 479 patients)||Stage II and IIIa (n = 148 patients)|
|Before diagnosis||2.9 ± 0.3||2.8 ± 0.2||3.0 ± 0.4||0.5|
|After diagnosis||2.2 ± 0.2||2.0 ± 0.1||1.8 ± 0.2||0.3|
|Absolute difference||− 0.7 ± 0.2b||− 0.9 ± 0.1b||− 1.1 ± 0.2b||0.3|
|% Difference||− 24.1%||− 32.1%||− 36.7%||0.3|
|Before diagnosis||14.6 ± 1.1||14.4 ± 0.6||14.5 ± 1.2||0.6|
|After diagnosis||14.5 ± 0.8||13.0 ± 0.4c||13.9 ± 0.9||0.7|
|Absolute difference||− 0.1 ± 0.8||− 1.4 ± 0.4b||− 0.6 ± 0.9||0.7|
|% Difference||− 0.7%||− 9.7%||− 4.1%||0.4|
|Before diagnosis||39.7 ± 1.7||40.0 ± 0.9||40.9 ± 1.9||0.6|
|After diagnosis||43.4 ± 1.7||42.9 ± 0.9||45.7 ± 1.9||0.4|
|Absolute difference||3.7 ± 1.3bc||3.0 ± 0.7b||4.8 ± 1.4b||0.6|
Statistically significant decreases in physical activity were observed from prediagnosis to postdiagnosis when comparing patients according to adjuvant treatment (Table 4). These decreases were observed overall and among subgroups defined by intensity of activity and type of activity. For example, patients who were treated with surgery, radiation, and chemotherapy had statistically significantly greater decreases in prediagnosis to postdiagnosis sports/recreational activity (50.0% decrease) compared with patients who underwent surgery only (24.1% decrease) and patients who underwent surgery and received radiation (23.3% decrease; P < 0.05). Finally, 54%, 55%, 68%, and 65% of patients who received surgery only; surgery and radiation; surgery and chemotherapy; and surgery, radiation, and chemotherapy, respectively, decreased their total physical activity levels from prediagnosis to postdiagnosis (P < 0.05). Statistically significant decreases in total, vigorous-intensity, and sports/recreational physical activity were observed from prediagnosis to postdiagnosis by age group (P < 0.05) (Table 5). No statistically significant trends across age groups were observed in prediagnosis to postdiagnosis changes in physical activity levels (P value for trend > 0.05).
|Activity level||Hours per week (mean ± SE)|
|Surgery (n = 266 patients)||Surgery and radiation (n = 329 patients)||Surgery and chemotherapy (n = 52 patients)||Surgery, radiation, and chemotherapy (n = 165 patients)|
|Before diagnosis||18.9 ± 0.2||18.7 ± 0.2||19.7 ± 0.4||18.9 ± 0.3|
|After diagnosis||17.3 ± 0.6||17.4 ± 0.5||16.1 ± 1.4||15.3 ± 0.9|
|Absolute difference||− 1.6 ± 0.6c||− 1.3 ± 0.5||− 3.6 ± 1.4cd||− 3.5 ± 0.9cd|
|% Difference||− 8.5%||− 7.0%||− 18.3%d||− 18.5%d|
|Before diagnosis||11.7 ± 0.4||11.5 ± 0.3||12.1 ± 0.8||11.2 ± 0.5|
|After diagnosis||11.1 ± 0.5||10.9 ± 0.4||10.2 ± 1.1||9.8 ± 0.7|
|Absolute difference||− 0.7 ± 0.5||− 0.5 ± 0.4||− 1.8 ± 1.1c||− 1.4 ± 0.7|
|% Difference||− 6.0%||− 4.3%||− 14.9%||− 12.5%|
|Before diagnosis||6.6 ± 0.3||6.5 ± 0.3||7.1 ± 0.7||7.1 ± 0.4|
|After diagnosis||5.8 ± 0.4||6.0 ± 0.3||6.1 ± 0.9||4.9 ± 0.6|
|Absolute difference||− 0.8 ± 0.4c||− 0.5 ± 0.4||− 1.1 ± 0.9||− 2.2 ± 0.6cdg|
|% Difference||− 12.1%||− 7.7%||− 15.5%||− 31.0%dg|
|Before diagnosis||0.6 ± 0.1||0.7 ± 0.1||0.5 ± 0.2||0.5 ± 0.1|
|After diagnosis||0.3 ± 0.1||0.5 ± 0.1||0.1 ± 0.2||0.3 ± 0.1|
|Absolute difference||− 0.2 ± 0.1c||− 0.3 ± 0.1c||− 0.3 ± 0.2c||− 0.3 ± 0.1c|
|% Difference||− 33.3%||− 42.9%||− 60.0%||− 60.0%|
|Before diagnosis||2.9 ± 0.7||3.0 ± 0.2||2.9 ± 0.5||3.0 ± 0.4|
|After diagnosis||2.2 ± 0.1||2.2 ± 0.1||1.8 ± 0.3||1.4 ± 0.2dg|
|Absolute difference||− 0.7 ± 0.1c||− 0.7 ± 0.1c||− 1.0 ± 0.3c||− 1.5 ± 0.2cdg|
|% Difference||− 24.1%||− 23.3%||− 34.5%||− 50.0%dg|
|Before diagnosis||14.1 ± 0.8||14.5 ± 0.8||14.5 ± 1.7||14.5 ± 1.4|
|After diagnosis||13.7 ± 0.6||13.9 ± 0.5||13.1 ± 1.4||12.7 ± 0.8|
|Absolute difference||− 0.8 ± 0.6||− 0.6 ± 0.5||− 1.4 ± 1.3c||− 1.8 ± 0.8c|
|% Difference||− 5.7%||− 4.1%||− 9.7%||− 12.4%|
|Before diagnosis||41.3 ± 1.3||39.7 ± 1.1||43.4 ± 2.9||37.8 ± 1.8|
|After diagnosis||44.1 ± 1.3||41.0 ± 1.1||49.4 ± 2.9d||45.9 ± 1.8d|
|Absolute difference||2.8 ± 1.0c||1.3 ± 0.8c||6.0 ± 2.2c||8.0 ± 1.4cdg|
|Activity level||Hours per week (mean ± SE)||P value for trend|
|Age 40–49 yrs (n = 218 patients)||Age 50–59 yrs (n = 274 patients)||Age > 60 yrs (n = 320 patients)|
|Before diagnosis||18.7 ± 0.2||18.9 ± 0.2||19.0 ± 0.2||0.5|
|After diagnosis||16.8 ± 0.8||17.1 ± 0.6||16.7 ± 0.6||0.9|
|Absolute difference||− 1.9 ± 0.8c||− 1.8 ± 0.6c||− 2.2 ± 0.6c||0.8|
|% Difference||− 10.2%||− 9.5%||− 11.6%||0.8|
|Before diagnosis||11.1 ± 0.5||11.8 ± 0.3||11.6 ± 0.4||0.4|
|After diagnosis||10.1 ± 0.6||10.6 ± 0.5||11.2 ± 0.5||0.3|
|Absolute difference||− 1.0 ± 0.6||− 1.2 ± 0.5c||− 0.4 ± 0.5||0.5|
|% Difference||− 9.0%||− 10.2%||− 3.4%||0.5|
|Before diagnosis||6.7 ± 0.4||6.5 ± 0.3||6.9 ± 0.3||0.8|
|After diagnosis||6.2 ± 0.5||6.1 ± 0.4||5.1 ± 0.4||0.2|
|Absolute difference||− 0.6 ± 0.5c||− 0.4 ± 0.4||− 1.8 ± 0.4cf||0.1|
|% Difference||− 9.0%||− 6.2%||− 26.1%b||0.1|
|Before diagnosis||0.8 ± 0.1||0.7 ± 0.1||0.5 ± 0.1b||0.02|
|After diagnosis||0.4 ± 0.1||0.5 ± 0.1||0.2 ± 0.1f||0.2|
|Absolute difference||− 0.4 ± 0.1c||− 0.2 ± 0.1c||− 0.2 ± 0.1c||0.1|
|% Difference||− 50.0%||− 28.6%||− 40.0%||0.1|
|Before diagnosis||2.9 ± 0.3||2.9 ± 0.2||3.0 ± 0.4||0.4|
|After diagnosis||2.2 ± 0.2||2.1 ± 0.1||1.8 ± 0.1||0.2|
|Absolute difference||− 0.7 ± 0.2c||− 0.8 ± 0.1c||− 1.1 ± 0.1c||0.2|
|% Difference||− 24.1%||− 27.6%||− 36.7%||0.2|
|Before diagnosis||14.5 ± 1.1||14.4 ± 0.6||14.5 ± 1.2||0.3|
|After diagnosis||13.2 ± 0.7||13.8 ± 0.6||13.5 ± 0.6||0.8|
|Absolute difference||− 1.3 ± 0.7||− 0.7 ± 0.6||− 1.0 ± 0.6||0.8|
|% Difference||− 9.0%||− 4.9%||− 6.9%||0.8|
|Before diagnosis||39.4 ± 1.6||40.8 ± 1.2||39.9 ± 1.3||0.8|
|After diagnosis||40.9 ± 1.6||45.4 ± 1.2h||43.8 ± 1.3||0.2|
|Absolute difference||1.5 ± 1.2c||4.6 ± 0.9ch||3.8 ± 1.0c||0.2|
We observed decreases in total, moderate-intensity, vigorous-intensity, and sports/recreational physical activity from prediagnosis to postdiagnosis for each BMI group (P < 0.05) (Table 6). Obese patients with breast carcinoma had greater decreases in prediagnosis to postdiagnosis sports/recreational activity (41.4% decrease) compared with leanest patients (24.1% decrease) and overweight patients (35.7% decrease; P value for trend < 0.05). Increases in sedentary activity (reading, napping, watching television) from prediagnosis to postdiagnosis in the range of 1.3 hours per week to 8.0 hours per week also were observed across all stages of disease, all treatment groups, all ages, and all BMI groups (P < 0.05).
|Activity level||Hours per week (mean ± SE)||P value for trend|
|BMI < 25 (n = 401 patients)||BMI 25.0–29.9 (n = 246 patients)||BMI > 29.9 (n = 165 patients)|
|Before diagnosis||18.8 ± 0.2||19.0 ± 0.2||18.9 ± 0.2||0.7|
|After diagnosis||17.1 ± 0.5||17.0 ± 0.6||16.1 ± 0.8||0.3|
|Absolute difference||− 1.7 ± 0.5c||− 2.0 ± 0.6c||− 2.8 ± 0.8c||0.2|
|% Difference||− 9.0%||− 10.5%||− 14.8%||0.2|
|Before diagnosis||11.4 ± 0.3||11.4 ± 0.4||11.8 ± 0.4||0.5|
|After diagnosis||10.8 ± 0.4||10.7 ± 0.5||10.5 ± 0.6||0.6|
|Absolute difference||− 0.7 ± 0.4||− 0.8 ± 0.5||− 1.4 ± 0.6||0.3|
|% Difference||− 6.1%||− 7.0%||− 11.9%||0.3|
|Before diagnosis||6.6 ± 0.2||7.0 ± 0.3||6.5 ± 0.4||0.9|
|After diagnosis||5.7 ± 0.3||6.0 ± 0.4||5.4 ± 0.5||0.6|
|Absolute difference||− 0.9 ± 0.3c||− 1.0 ± 0.4c||− 1.1 ± 0.5c||0.7|
|% Difference||− 13.6%||− 14.3%||− 16.9%||0.7|
|Before diagnosis||0.7 ± 0.1||0.5 ± 0.1||0.5 ± 0.1||0.1|
|After diagnosis||0.5 ± 0.1||0.3 ± 0.1g||0.2 ± 0.1g||0.01|
|Absolute difference||− 0.2 ± 0.1c||− 0.3 ± 0.1c||− 0.3 ± 0.1c||0.5|
|% Difference||− 28.6%||− 60.0%||− 60.0%||0.5|
|Before diagnosis||2.9 ± 0.3||2.8 ± 0.2||2.9 ± 0.4||0.4|
|After diagnosis||2.2 ± 0.1||1.9 ± 0.1||1.7 ± 0.2g||0.01|
|Absolute difference||− 0.7 ± 0.1c||− 1.0 ± 0.1c||− 1.2 ± 0.2c||0.01|
|% Difference||− 24.1%||− 35.7%||− 41.4%||0.01|
|Before diagnosis||14.4 ± 1.1||14.5 ± 0.6||14.3 ± 1.2||0.5|
|After diagnosis||13.4 ± 0.5||13.8 ± 0.6||13.3 ± 0.7||0.9|
|Absolute difference||− 1.0 ± 0.5||− 0.6 ± 0.6||− 1.1 ± 0.7||0.9|
|% Difference||− 6.9%||− 4.1%||− 7.7%||0.9|
|Before diagnosis||38.9 ± 1.0||38.8 ± 1.3||45.2 ± 1.6||0.006|
|After diagnosis||41.8 ± 1.0||41.4 ± 1.3||51.1 ± 1.6||0.0001|
|Absolute difference||3.0 ± 0.8c||2.6 ± 1.0c||5.9 ± 1.2cgi||0.04|
We also examined whether any of the confounders modified the changes in physical activity from prediagnosis to postdiagnosis. Ethnicity, marital status, education, study site, tamoxifen use, and family history did not modify the changes in physical activity from prediagnosis to postdiagnosis. However, time from diagnosis to interview and prediagnosis physical activity levels were associated with different physical activity patterns. The decreases in prediagnosis to postdiagnosis sports/recreational physical activity were greater for women who were interviewed within 5 months of diagnosis (− 1.0 ± 0.1 hours per week) compared with women who were interviewed 8–12 months from diagnosis (− 0.4 ± 0.2 hours per week). Women who were the most active (i.e., the top quartile) prediagnosis had greater decreases in total physical activity postdiagnosis (− 8.2 ± 0.7 hours per week) compared with women who were the least active (i.e., the bottom quartile) prediagnosis (− 0.7 ± 0.7 hours per week).
Statistically significant predictors of a decrease in total, sports/recreational, and vigorous-intensity physical activity levels were treatment with chemotherapy and radiation and no family history of breast carcinoma (correlation coefficient [R2], 0.23; P < 0.05); treatment with chemotherapy and radiation, BMI, and age (R2, 0.39; P < 0.05); and treatment with chemotherapy and radiation, advanced disease stage, and BMI (R2, 0.44; P < 0.05), respectively (data not shown).
Patients with breast carcinoma in this population-based cohort study decreased their estimated time spent in total physical activity on average by 2.0 hours per week from the year prediagnosis to within 1 year postdiagnosis, an 11% decrease (P < 0.05). This decrease in physical activity most likely explains some of the weight gain observed postdiagnosis. Statistically significantly larger decreases in physical activity were observed among women who were treated with radiation and chemotherapy compared with women who underwent surgery only or who were treated with radiation only (P < 0.05). The decrease in physical activity levels from prediagnosis to postdiagnosis was expected given the documented negative side effects of treatment for breast carcinoma, including fatigue, nausea, and pain.10, 16–19 In smaller, nonpopulation-based studies, similar findings have been reported by Demark-Wahnefried et al.,10 Courneya and Friedenreich,9 Cooper,16 and Mock et al.17
In a study of 18 premenopausal women with breast carcinoma who received adjuvant chemotherapy, Demark-Wahnefried et al.10 reported that physical activity decreased from 2159 ± 490 kilo Joules (kJ) per day at baseline (i.e., ≈ 2 weeks after surgery and before adjuvant chemotherapy began) to 1937 ± 347 kJ per day (a 10% decrease) at the completion of chemotherapy. In a study of 130 survivors of colorectal carcinoma, Courneya and Friedenreich9 reported that treatment had a significant negative effect on physical activity participation levels that did not recover completely posttreatment. Furthermore, survivors of colorectal carcinoma exhibited four main exercise patterns across the cancer experience that were labeled maintainers (active before diagnosis, during treatment, and after treatment), temporary relapsers (decreased activity only during treatment), permanent relapsers (decreased activity during and after treatment), and nonexercisers (inactive before diagnosis, during treatment, and after treatment). Permanent relapsers reported the lowest current quality of life compared with the other three groups. These results indicate that, within 1–4 years postdiagnosis, patients who are treated for breast carcinoma and colon carcinoma still have not returned fully to their prediagnosis level of physical activity and that decreased physical activity levels are associated with a worse quality of life.
Decreases in physical activity with a diagnosis of breast carcinoma may lead to increases in body weight and body fat, which, in turn, may lead to a poorer prognosis.1, 2 Furthermore, as observed in the current study, greater decreases in physical activity from prediagnosis to postdiagnosis were observed among obese patients with breast carcinoma compared with normal weight and overweight patients (P < 0.05), implying a potential for greater weight gain among already obese women. The negative influence of increased body weight or BMI on the prognosis of patients with breast carcinoma has been reported in the vast majority of studies over the past 25 years, with many analyses controlled for recognized prognostic variables, such as comorbidities and breast cancer screening.20 Differential variations in sex hormones (e.g., estrogens) by body mass and body fat may be one explanation for the poor survival experienced by overweight patients with breast carcinoma.21–23 Specifically, it has been observed that overweight and obese postmenopausal women with breast carcinoma24 and without breast carcinoma21–23 have higher estrogen levels compared with lighter weight women. High estrogen levels have been associated with increased risk for incident breast carcinoma in several cohort studies,25 suggesting that these hormones may be breast tumor promoters.20 High estrogen levels likely represent conversion of androgens to estrogens by the enzyme aromatase, which is abundant in adipose tissue.21–23 High estrogen levels promote breast carcinoma cell growth,25 thus explaining in part the poorer prognosis of overweight or obese postmenopausal women with breast carcinoma. Other biologic markers associated with breast carcinoma are insulin-like growth factors (IGFs)26 and fasting insulin levels.27 Several prospective cohort studies have shown that levels of circulating IGF-I and fasting insulin have a positive association with the risk of breast carcinoma,26, 27 and experimental studies indicate that circulating levels of IGF-I and insulin are influenced by physical activity and obesity.28, 29 Few epidemiologic studies have addressed the role of IGFs and insulin levels in breast carcinoma and their interaction with physical activity and obesity.
The current study had several limitations that should be considered when interpreting the findings and planning future research. With regard to the physical activity assessment, prediagnosis and postdiagnosis physical activity levels were assessed in the same interview, which may result in cross-contamination of responses. Prediagnosis physical activity levels also were assessed retrospectively and, thus, are subject to recall bias. Changes in occupational physical activity from before to after breast carcinoma diagnosis was not assessed. We hypothesize that occupational physical activity, similar to sports/recreational and household activity, also would decrease from before diagnosis to after diagnosis. Information on whether the patient continued to work during treatment and, if so, whether her occupational activity level changed with diagnosis would provide a more detailed description of physical activity patterns across the cancer experience. The physical activity questionnaire, although it was based on another validated questionnaire,11 was not validated internally in the HEAL Study. Although our analyses were limited to within-cohort comparisons, we cannot be sure that these associations pertain to all patients with breast carcinoma. Finally, there is the possibility of selection bias with healthier women volunteering to participate in the HEAL Study. Healthier women may have maintained their prediagnosis physical activity levels after diagnosis; thus, the reported decreases in physical activity levels after diagnosis may be attenuated.
The study had several strengths. First, the HEAL Study is quite unique, in that it is the first cohort study that followed patients with breast carcinoma to determine the associations between diet, physical activity, body composition, and hormones on disease stage and prognosis. Second, this study is one of a handful that has examined physical activity levels across the cancer experience and the only study reporting detailed physical activity by type and intensity stratified by disease stage, adjuvant treatment, age, and BMI. Third, although the HEAL Study participants in New Mexico and Washington differed on several demographic and physiological characteristics, physical activity levels did not differ by study site. Thus, our findings of decreased physical activity levels after a diagnosis of breast carcinoma may be more generalizable to all patients with breast carcinoma because of the variability in age, disease stage, ethnicity, and body weight observed in the HEAL Study.
Overall, in the current study, physical activity levels were reduced significantly after diagnosis: This was particularly evident for activities that required greater energy expenditure. A 60% decline in vigorous-intensity physical activity was observed among women who received the most intensive treatment (chemotherapy and radiation therapy), the group most likely to gain the most weight after diagnosis, compared with a 33% decline among women who underwent surgery only. Decreases in prediagnosis to postdiagnosis physical activity levels were greater for women who were interviewed closer to diagnosis and women who were more physically active before diagnosis. When designing exercise interventions for cancer survivors, it is important to tailor the intervention to the participant's prediagnosis physical activity levels. Such intervention should focus on maintenance of physical activity levels among previously active survivors and increasing physical activity levels among inactive survivors of breast carcinoma.
The implications of decreased physical activity for the prognosis of patients with breast carcinoma have not been identified to date. However, greater body mass and body fat at the time patients are diagnosed with breast carcinoma have been associated with an increased risk of developing recurrent breast carcinoma and with poorer survival.20 Work by Courneya and Friedenreich9 and by others17–19 has demonstrated that women who undergo more intensive treatment are able to be physically active; thus, it should be feasible to avoid this decline in physical activity. These data on physical activity levels among patients with breast carcinoma provide valuable information for planning larger observational and intervention studies designed to evaluate how physical activity levels may improve prognosis and prolong survival.
- 11Modifiable activity questionnaire. Med Sci Sports Exer. 1997; 29: S73–S78..
- 12National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report. Obesity Res. 1998; 6 (Suppl 2): 51S–209S.
- 15National Cancer Institute. The SEER program code manual, June 1992. Bethesda (MD): Cancer Statistics Branch, Surveillance Program, Division of Cancer Prevention and Control, National Cancer Institute, U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health; 1992 NIH Pub. No. 92-1999.
- 16The role of physical activity in the recovery from breast cancer. Melpomene J. 1995; 14: 18–20..
- 24Adiposity and sex hormones in post menopausal breast cancer survivors. J Clin Oncol. In press., , , et al.