Abstract
Purpose:
Physical activity (PA) has been shown to improve QoL in predominantly White cancer survivors. Very few studies have examined the association between PA and QoL among Black breast cancer survivors (BCS). We investigated the association between PA and multiple QoL domains and the effects of race on the proposed association in a racially diverse group of BCS.
Methods:
This was an exploratory study using secondary data from a completed 12-month randomized controlled trial (RCT). Mixed effects models were tested on a subset of participants in the control and exercise groups of the RCT. The primary outcomes were changes in the QoL domains (baseline to 12 months post baseline).
Results:
There were 173 participants included in this analysis, averaging 59 years of age; about 33% of the participants were Black women. There were no significant differences in the QoL outcomes between the control and exercise groups at 12 months post baseline. Race was not a significant moderator. Exercise improved emotional/mental wellbeing and body image as it relates to social barriers at 12 months post baseline in Black and White BCS, but the changes in these outcomes were only statistically significant in White BCS (p<0.05).
Conclusions:
Results show that exercise can improve multiple QoL domains over time in Black BCS. However, the significance of the effect on QoL was isolated to White BCS. The small sample size in Black women could constrain the statistical significance of observed effects. Future studies are warranted to assess associations between exercise and QoL in larger samples of Black women.
Keywords: cancer survivor, physical activity, exercise, quality of life, breast cancer, African American
Introduction
Breast cancer is the primary type of new invasive cancer cases and the second leading cause of cancer deaths in women in the United States (US). (1, 2, 3) The number of living breast cancer survivors (BCS) will increase from 3.8 to 4.9 million by 2030. (4) Despite the increase in the number of living BCS, there remain clear racial disparities in five-year breast cancer survival and mortality rates between Black and White women. The five-year breast cancer survival rate is significantly lower in Black women (82%) compared with White women (91%). (3) Furthermore, Black women are more likely to die from breast cancer than White women (e.g., 27.8 vs 19.3 deaths per 100,000 women). (5)
In addition to racial disparities in breast cancer outcomes, Black women in comparison with White Women are more likely to experience poorer quality of life (QoL) and adverse cancer treatment effects (e.g., pain and swelling from breast-cancer related lymphedema [BCRL]). (6, 7, 8, 9, 10, 11, 12) QoL is a multidimensional and subjective concept that “encompasses a broad range of domains, including physical, functional, emotional, and social well-being,” (12, 13, 14) whereas health-related QoL is defined by the Centers for Disease Control and Prevention as “an individual’s or a group’s perceived physical and mental health over time.” (15) Black BCS are more likely to have multiple QoL issues including lower sexual function, negative feelings or perceptions about their bodies, poorer sleep quality, poorer physical/functional wellbeing, greater fatigue, and unmet needs (e.g., emotional support). (8, 9, 12, 13, 16, 17) Based on previous research, the prevalence of lymphedema (abnormal swelling in the arms due to damaged or removed lymph nodes after breast cancer therapy) in BCS has ranged from 0% to 56%. (6, 7, 18, 19) Specifically, the prevalence of lymphedema has been shown to be higher in Black than White BCS (28% vs 21%). (6, 7) BCS with arm symptoms or diagnosed with lymphedema have reported lower HRQoL (physical and mental health) than BCS without lymphedema or arm symptoms. (18)
Physical activity (PA) has been documented to improve breast cancer outcomes. (20, 21, 22, 23, 24) PA is defined as any bodily movement, while exercise is the subset of PA defined by planned activities for the purpose of improving health and fitness. (25) Exercise prescriptions or regimens consists of the following four major components: frequency (the number of days of activity per week), intensity (e.g., moderate activity), time (number of minutes or hours per session of exercise) and type (e.g., weightlifting) (FITT). (26) Research consistently reports PA as a protective factor against breast cancer incidence and mortality. (20, 21) Aerobic exercise (e.g., resistance training) has been shown to reduce the risk of breast cancer-related lymphedema (BCRL). (27)
Previous research has examined the link between PA and QoL among cancer survivors. Two studies reported positive and significant associations between PA and different aspects of QoL (e.g., social wellbeing) in predominantly Black cancer survivors. (28, 29) Another group of researchers observed direct and indirect statistically significant associations between PA and global QoL (e.g., mental wellbeing) in predominantly Black BCS. (30) Other relevant studies have included predominantly or only White cancer survivors and demonstrated similar results. (31, 32) For example, Smith and colleagues (2009) found positive and significant associations between PA and QoL (e.g., social functioning) in Black and White women. (31) Specifically, Black and White women who met the PA recommendations (at least 150 minutes of moderate to vigorous PA each week) reported better QoL. (31) Another study found similar results among predominantly White cancer survivors. (32) However, the results were not stratified by race and gender. (32) Therefore, researchers cannot determine whether there are potential underlying differences in QoL across race and gender. Very few studies have examined the longitudinal association between PA and multiple QoL domains among BCS, specifically Black women. (28)
Previous meta-analyses investigated the effects of exercise on QoL as well as the moderating effects of various factors in cancer survivors. (33, 34) The results showed that exercise significantly improved QoL (e.g., social functioning) in cancer survivors. In addition, the evidence suggested no significant moderating effects of clinical factors (e.g., body mass index), demographic factors (e.g., age), and exercise specific intervention components (e.g., exercise frequency) on the association between PA and QoL in cancer survivors. Potential moderators such as race were not examined in previous studies. Taken together, these studies demonstrate that PA alone improves multiple QoL domains in survivors of breast cancer. However, the effects of PA on multiple QoL domains in Black BCS remains unclear. These observations justified the examination of the independent effects of PA on QoL among subgroups of BCS in a completed randomized controlled trial.
This study will address some gaps in the literature. We will expand the literature on the study of PA and QoL in BCS and determine the effects of PA on QoL among an underrepresented group, specifically Black women. The aim of this paper was to investigate the association between PA and multiple QoL domains and to explore the effects of race on the proposed association among a racially diverse group of BCS in a completed 12-month randomized controlled trial. The hypotheses were as follows: (1) the exercise group will show significant improvements in the QoL outcomes compared with the control group and (2) the effect of exercise on the QoL outcomes will significantly differ between Black and White BCS. One possible reason for this hypothesis is that White BCS are more likely to meet the PA recommendations than Black BCS. (35) The findings will help researchers and healthcare professionals to better understand the proposed association and the benefits of behavioral medicine (e.g., PA) on QoL among Black BCS who disproportionately have worse breast cancer-related outcomes.
Methods
This was an exploratory study investigating the association between PA and QoL and the effects of race on the proposed association in Black and White BCS who were enrolled in a 12-month randomized controlled trial (Women in Steady Exercise Research [WISER] Survivor). This analysis focuses specifically on the participants in the exercise and control groups. Participants were recruited in Philadelphia, Pennsylvania through local hospitals and tumor registries. (36) The study was conducted between 2011–2016. The study participants provided signed informed consent and a written clearance from their doctors to participate in the WISER Survivor trial. In the WISER Survivor trial, 351 participants were recruited and randomly assigned: (1) exercise (in person and home-based exercises); (2) weight loss (using Nutrisystem); (3) combined exercise and weight loss; or (4) control group (referral to physician and health organization for health behavior resources such as diet or exercise). The study participants received 52 weeks of exercise, diet, or exercise plus diet. The exercise treatment plan consisted of nine resistance exercises (e.g., chest press) twice per week and moderate exercise (walking) consistent with the American College of Sports Medicine (ACSM) guidelines. (37, 38, 39) Participants were asked to increase their weekly minutes of walking from 90 to 180 minutes (30 minutes on most days of the week). (38) About 74% of the survivors in the exercise and combined groups in the WISER Survivor trial engaged in an average of 133 minutes of moderate exercise each week. (37) The original study was approved by the Institutional Review Board (IRB) of the University of Pennsylvania. The study is registered on ClinicalTrials.gov as NCT01515124. The study methods are described in detail elsewhere. (38) Ethical approval was required for analysis of secondary data. Informed consent was not obtained in the current study because the research involves no more than minimal risk to subjects and no procedures for which written consent is normally required outside of the research context. The current study was approved by the IRB of the Pennsylvania State University College of Medicine (STUDY00014712).
Study Population
The inclusion criteria for the WISER Survivor trial were as follows: (1) personal history of breast cancer; (2) completed treatment at least six months prior to randomization; (3) 80 years or younger; (4) body mass index of ≥ 25 kg/m2; (5) diagnosed with breast cancer-related lymphedema; (6) female; (7) sedentary (engaging in less than 3 or more times weekly aerobic activity of moderate intensity); and (8) free of cancer (no evidence of cancer or recurrence at the time or during the study). (38) Of the 351 participants, 35% identified as a Black woman. (38) The current analysis included both Black and White women. Women who did not identify as White or Black (N=4) were removed from this analysis because further evaluation of women who did not identify as Black or White was not possible due to a smaller number of participants. In addition, White BCS were the comparison group since most women identified in this group.
Measures
Primary outcomes
The primary outcomes were changes in multiple QoL domains. The major QoL domains were social functioning, social wellbeing, emotional/mental wellbeing, and body image. The body image subdomains were strength and health, social barriers, and appearance and sexuality. Very few studies have assessed these specific domains in relation to PA among Black BCS. (29–31)
QoL was assessed at baseline, 6 months post baseline, and 12 months post baseline using multiple valid and reliable questionnaires such as the Short Form Health Survey (SF-36), Body Image and Relationships Survey (BIRS), and Specific Quality of Life Scale for Upper Limb Lymphedema (ULL-27). (38, 40, 41, 44) The SF-36 survey has been used extensively in BCS. (42) The internal consistency reliability estimates for the SF-36 scales have been shown to range from 0.76 to 0.93 among diverse groups (e.g., Black patients). (43) The SF-36 survey consists of 36 questions about eight major health domains (e.g., emotional/mental wellbeing) with each domain score ranging from 0 to 100 (higher score indicates better QoL). One specific SF-36 survey question was “During the past 4 weeks, have you been a happy person”? Participants then rated their QoL on a 5 or 6 point Likert scale (e.g., 1=all of the time, 6=none of the time), depending on the QoL domain. The BIRS survey has been shown to have internal consistency and test-retest reliability ranging from 0.41 to 0.80. (40) It consists of three subscale scores and one summary score with higher scores indicating higher impairment: strength and health (ranging from 12 to 60), social barriers (ranging from 9 to 45), and appearance and sexuality (ranging from 11 to 55), and total (32–160). For the BIRS, one example item was “I felt confident I could make myself stronger.” Using a 5 point Likert scale (1=strongly disagree, 5=strongly agree), participants then rated their attitudes about health, strength, appearance, sexuality, and social barriers. The ULL-27 survey includes 27 questions and measures three major QoL domains (e.g., social wellbeing) with each domain score ranging from 0 to 6 (higher score indicates lower QoL). This scale has been shown to be precise, sensitive, and accurate. (44) One specific ULL-27 survey question was “Are you disturbed by difficulty in social life”? Participants then rated their QoL on a 6 point Likert scale (1=never, 6=constantly).
PA measure
PA levels were assessed with the Modifiable Physical Activity Questionnaire at baseline and 12 months. (38) This survey captures leisure-time, occupational, and sedentary time in the past year. The leisure activity section of this survey has been validated through comparisons with accelerometry (rho = 0.62). (45)
Other measures
A general survey with questions about demographics and a medical history survey were administered to the study participants. (38) Clinical characteristics (e.g., cancer stage) were obtained from pathology reports and other medical records. Anthropometric measurements were assessed at different time points: body weight (baseline to 12 months) and height (baseline). (38) Body weight and height were measured on a calibrated scale and stadiometer. (37, 38)
Statistical Analysis
The WISER Survivor trial was powered to detect a change in the primary outcome (breast cancer-related lymphedema interlimb volume difference). The primary outcomes in the current study consisted of the changes in the QoL outcomes (baseline to 12 months post baseline). The baseline and 12 month timepoints were used in the primary analysis. The effects of race, time, intervention condition (e.g., exercise group), and their interactions (e.g., group X time X race) were examined. Race was examined as a potential moderator. After data extraction and cleaning (including variable construction), descriptive statistics were used to summarize categorical (e.g., frequencies) and continuous variables (e.g., means) in the secondary data analysis. The primary outcomes were emotional/mental wellbeing, social wellbeing, social functioning, body image, and the BIRS subscale scores (e.g., strength and health). Demographic and clinical characteristics were compared at baseline across the intervention conditions and race. The means and SDs for each of the primary outcomes were compared across the intervention conditions and race at baseline. Comparisons across the intervention conditions and race were performed using chi-square tests or Fisher’s exact test (categorical variables) and F-tests or unpaired t-tests (continuous variables).
Multiple imputation (MI) was used for handling missing data where necessary at baseline and 12 months. (46–48) This method for analyzing missing data assumes that the data are missing at random and similar methods were used in the main results manuscript for the WISER Survivor trial. (37, 47) The total number of imputations consisted of 10 replicates. Mixed effects models with multiple imputation were used to examine associations between the predictor and outcomes. The covariates included marital status, age, mode of transportation, cancer treatments (e.g., immunotherapy), cancer stage, leisure time, retirement, and body mass index (BMI). A mixed effects model is more efficient for handling missing data and multiple imputation in SAS and appropriate for analyses of repeated measures. (49) A moderation analysis was conducted. The moderating effect of race was tested by examining interactions. The mixed effects models included the group X time and group X time X race interaction terms for the subgroup analyses. In addition, the data were analyzed with and without imputation using mixed effects models to examine the consistency of the findings for the primary outcomes. There were differences in the BIRS scores (e.g., strength and health) for imputed versus not imputed analyses. The probability (p values) did not include adjustment for multiplicity. Therefore, the findings should be interpreted carefully. All analyses and models were conducted using SAS version 9.4 software (SAS Institute Inc., Cary, NC, USA). A significance value of 0.05 for the analyses was used.
Results
Characteristics of Study Participants
The baseline characteristics of the study sample and the differences in the baseline characteristics by race can be found in Table I. This sample consisted of 173 Black women (N=57) and White women (N=116). The average age of the participants was 59 years (SD=8.27). More than half of the women were White (67%), married or living with a partner (68%), and postmenopausal (66%). About 33% of the participants were Black women. Over half of the participants (81%) had at least some college education. The average time since diagnosis was about 8 years (SD=5.23). More than half of the participants received chemotherapy (79%) and radiation (82%), while 17% of the participants received immunotherapy.
Table I.
Baseline characteristics of the study participants by race
| Variables | Total (n=173) | Blacks (n=57) | Whites (n=116) | P-value |
|---|---|---|---|---|
| Marital status | 0.004 * | |||
| Never married | 17 (10%) | 8 (14%) | 9 (8%) | |
| Currently married or living with partner | 116 (68%) | 28 (50%) | 88 (77%) | |
| Divorced or separated | 30 (17%) | 17 (31%) | 13 (11%) | |
| Widowed or widower | 8 (5%) | 3 (5%) | 5 (4%) | |
| Education | 0.771 | |||
| Less than college | 34 (19%) | 12 (21%) | 22 (19%) | |
| Some college (junior or technical college) | 55 (32%) | 20 (35%) | 35 (30%) | |
| 4 -year college grad | 48 (28%) | 13 (23%) | 35 (30%) | |
| Graduate degree | 36 (21%) | 12 (21%) | 24 (21%) | |
| Ethnicity | 0.330 | |||
| Hispanic or Latino | 2 (1%) | 0 (0%) | 2 (2%) | |
| Not Hispanic or Latino | 167 (99%) | 54 (100%) | 113 (98%) | |
| Retirement status | 0.250 | |||
| Not retired | 125 (72%) | 38 (67%) | 87 (75%) | |
| Retired | 48 (28%) | 19 (33%) | 29 (25%) | |
| Mode of Transportation | <.0001 * | |||
| Personal car | 140 (81%) | 35 (61%) | 105 (91%) | |
| Public transportation | 27 (16%) | 18 (32%) | 9 (8%) | |
| Other | 5 (3%) | 4 (7%) | 1 (1%) | |
| Leisure time (MET hr/week) | 0.282 | |||
| Mean (SD) | 7.16 (7.78) | 6.32 (6.45) | 7.57 (8.35) | |
| Body Mass Index | 0.200 | |||
| Mean (SD) | 34.25 (6.44) | 35.15 (5.80) | 33.81 (6.71) | |
| Cancer stage | 0.167 | |||
| 0-<1 | 16 (12%) | 6 (15%) | 10 (11%) | |
| 1-<2 | 37 (28%) | 9 (23%) | 28 (31%) | |
| 2-<3 | 45 (35%) | 12 (30%) | 33 (36%) | |
| 3-<4 | 26 (20%) | 8 (20%) | 18 (20%) | |
| 4 | 7 (5%) | 5 (12%) | 2 (2%) | |
| Chemotherapy | 0.732 | |||
| No chemotherapy | 36 (21%) | 11 (19%) | 25 (22%) | |
| Chemotherapy | 137 (79%) | 46 (81%) | 91 (78%) | |
| Radiation | 0.076 | |||
| No radiation | 31 (18%) | 6 (11%) | 25 (22%) | |
| Radiation | 142 (82%) | 51 (89%) | 91 (78%) | |
| Immunotherapy | 0.237 | |||
| No immunotherapy | 126 (83%) | 40 (78%) | 86 (86%) | |
| Immunotherapy | 25 (17%) | 11 (22%) | 14 (14%) | |
| Hormonal therapy | 0.080 | |||
| No hormonal therapy | 108 (65%) | 39 (75%) | 69 (61%) | |
| Hormonal therapy | 57 (35%) | 13 (25%) | 44 (39%) | |
| Social Functioning Score | 0.626 | |||
| Mean (SD) | 79.91 (22.34) | 78.73 (22.16) | 80.50 (22.50) | |
| Emotional/Mental Wellbeing Score | 0.135 | |||
| Mean (SD) | 75.72 (15.11) | 78.18 (14.36) | 74.52 (15.38) | |
| Social Wellbeing Score | 0.035 * | |||
| Mean (SD) | 1.24 (1.25) | 0.95 (1.13) | 1.38 (1.29) | |
| Body Image and Relationships-Total Score | 0.780 | |||
| Mean (SD) | 92.68 (9.75) | 92.98 (9.37) | 92.54 (9.97) | |
| Body Image and Relationships-Social Barriers Subscale Score | 0.780 | |||
| Mean (SD) | 22.30 (8.35) | 22.01 (8.44) | 22.46 (8.35) | |
| Body Image and Relationships-Strength and Health Subscale Score | 0.230 | |||
| Mean (SD) | 36.00 (4.40) | 36.57 (4.76) | 35.72 (4.21) | |
| Body Image and Relationships-Appearance and Sexuality Subscale Score | 0.069 | |||
| Mean (SD) | 33.40 (4.64) | 34.37 (4.70) | 32.90 (4.55) |
p value < 0.05 indicated statistical significance
Overall, there were significant differences in the baseline characteristics (including marital status, mode of transportation, and social wellbeing) by race (Table I). White women were more likely to be married or living with a partner compared with Black women (77% vs 50%; p=0.004). Black women were less likely to have personal transportation (e.g., private vehicle) than White women (61% vs 91%; p<.0001). At baseline, the average social wellbeing scores (p=0.035) were significantly worse in White women (mean=1.38, SD=1.29) than Black women (mean=0.95, SD=1.13).
Table II displays the participants’ characteristics by intervention condition at baseline. Overall, there was a significant difference in one of the sociodemographic variables at baseline between the intervention conditions. There was a significant difference in race between the intervention conditions (p=0.013). There were significantly more Black women in the exercise group (42%) compared with the control group (24%) due to the small sample sizes in the current study.
Table II.
Baseline characteristics of the study participants by intervention condition
| Variables | Control (N=87) |
Exercise (N=86) |
P-value |
|---|---|---|---|
| Marital status | 0.133 | ||
| Never married | 9 (11%) | 8 (9%) | |
| Currently married or living with partner | 62 (73%) | 54 (63%) | |
| Divorced or separated | 13 (15%) | 17 (20%) | |
| Widowed or widower | 1 (1%) | 7 (8%) | |
| Education | 0.421 | ||
| Less than college | 19 (22%) | 15 (17%) | |
| Some college (junior or technical college) | 27 (31%) | 28 (33%) | |
| 4 -year college grad | 20 (23%) | 28 (33%) | |
| Graduate degree | 21 (24%) | 15 (17%) | |
| Ethnicity | 0.497 | ||
| Hispanic or Latino | 2 (2%) | 0 (0%) | |
| Not Hispanic or Latino | 83 (98%) | 84 (100%) | |
| Race | 0.013 * | ||
| Black or African American | 21 (24%) | 36 (42%) | |
| White or Caucasian | 66 (76%) | 50 (58%) | |
| Retirement status | 0.468 | ||
| Not retired | 65 (75%) | 60 (70%) | |
| Retired | 22 (25%) | 26 (30%) | |
| Mode of Transportation | 0.795 | ||
| Personal car | 71 (83%) | 69 (80%) | |
| Public transportation | 12 (14%) | 15 (18%) | |
| Other | 3 (3%) | 2 (2%) | |
| Leisure time (MET hr/week) | 0.725 | ||
| Mean (SD) | 7.37 (7.77) | 6.95 (7.83) | |
| Body mass index | 0.785 | ||
| Mean (SD) | 34.38 (6.73) | 34.12 (6.16) | |
| Cancer stage | 0.286 | ||
| 0-<1 | 10 (15%) | 6 (9%) | |
| 1-<2 | 15 (23%) | 22 (33%) | |
| 2-<3 | 22 (34%) | 23 (35%) | |
| 3-<4 | 16 (25%) | 10 (15%) | |
| 4 | 2 (3%) | 5 (8%) | |
| Chemotherapy | 0.245 | ||
| No chemotherapy | 15 (17%) | 21 (24%) | |
| Chemotherapy | 72 (83%) | 65 (76%) | |
| Radiation | 0.576 | ||
| No radiation | 17 (20%) | 14 (16%) | |
| Radiation | 70 (80%) | 72 (84%) | |
| Immunotherapy | 0.912 | ||
| No immunotherapy | 62 (84%) | 64 (83%) | |
| Immunotherapy | 12 (16%) | 13 (17%) | |
| Hormonal therapy | 0.664 | ||
| No hormonal therapy | 53 (64%) | 55 (67%) | |
| Hormonal therapy | 30 (36%) | 27 (33%) | |
| Social Functioning Score | 0.360 | ||
| Mean (SD) | 81.47 (23.41) | 78.34 (21.22) | |
| Emotional/Mental Wellbeing Score | 0.872 | ||
| Mean (SD) | 75.91 (15.82) | 75.53 (14.43) | |
| Social Wellbeing Score | 0.458 | ||
| Mean (SD) | 1.17 (1.21) | 1.31 (1.29) | |
| Body Image and Relationships-Total Score | 0.148 | ||
| Mean (SD) | 91.62 (9.47) | 93.78 (9.97) | |
| Body Image and Relationships-Social Barriers Subscale Score | 0.011 | ||
| Mean (SD) | 20.42 (8.32) | 24.25 (7.99) | |
| Body Image and Relationships-Strength and Health Subscale Score | 0.780 | ||
| Mean (SD) | 35.91 (4.39) | 36.09 (4.44) | |
| Body Image and Relationships-Appearance and Sexuality Subscale Score | 0.693 | ||
| Mean (SD) | 33.55 (4.45) | 33.25 (4.86) |
p value < 0.05 indicated statistical significance
Effects of Intervention Condition on QoL
The group X time interaction effect on social functioning (p=0.164), emotional/mental wellbeing (p=0.756), social wellbeing (p=0.092), and BIRS total (p=0.138) along with the subscale scores including BIRS strength and health (p=0.634), BIRS appearance and sexuality (p=0.939), and BIRS social barriers (p=0.174) scores was not significant. After further analysis, there were no significant changes in the QoL scores between the intervention conditions at 12 months post baseline (Table III).
Table III.
Between treatment group changes in the quality of life outcomes from baseline to 12 months
| Baseline | Change at 12 months | |||
|---|---|---|---|---|
| Outcomes | Randomized group | Mean (95% CI) | Mean (95% CI) | P-value |
| Emotional/Mental Wellbeing Score (Range: 0–100, higher is better) |
||||
| Control | 72.21 (65.68, 78.73) | Reference | Reference | |
| Exercise | 70.64 (64.24, 77.04) | 1.45 (−3.81, 6.72) | 0.585 | |
| Social Wellbeing Score (Range: 0–6, higher is worse) |
||||
| Control | 1.02 (0.51, 1.52) | Reference | Reference | |
| Exercise | 1.36 (0.86, 1.85) | −0.04 (−0.36, 0.27) | 0.787 | |
| Body Image and Relationships-Total Score (Range: 32–160, higher is worse) |
||||
| Control | 93.34 (88.95, 97.73) | Reference | Reference | |
| Exercise | 95.46 (91.19, 99.72) | −1.56 (−6.54, 3.42) | 0.532 | |
| Social Functioning Score (Range: 0–100, higher is better) |
||||
| Control | 77.87 (69.11, 86.64) | Reference | Reference | |
| Exercise | 71.83 (63.30, 80.36) | −0.01 (−7.80, 7.79) | 0.999 | |
| Body Image and Relationships-Strength and Health Subscale Score (Range: 12–60, higher is worse) |
||||
| Control | 37.10 (35.15, 39.06) | Reference | Reference | |
| Exercise | 36.64 (34.73, 38.54) | −0.16 (−2.13, 1.81) | 0.875 | |
| Body Image and Relationships-Appearance and Sexuality Subscale Score (Range: 11–55, higher is worse) |
||||
| Control | 34.37 (32.32, 36.42) | Reference | Reference | |
| Exercise | 33.84 (31.86, 35.83) | 1.45 (−0.53, 3.42) | 0.151 | |
| Body Image and Relationships-Social Barriers Subscale Score (Range: 9–45, higher is worse) |
||||
| Control | 21.00 (17.61, 24.40) | Reference | Reference | |
| Exercise | 24.36 (21.06, 27.65) | −2.83 (−5.80, 0.13) | 0.061 | |
p value < 0.05 indicated statistical significance
Effects of Exercise Intervention on QoL by race
The group X time X race interaction effect on social functioning (p= 0.274), emotional/mental wellbeing (p= 0.527), social wellbeing (p= 0.057), and BIRS total (p= 0.702) along with subscale scores including BIRS strength and health (p= 0.654), BIRS appearance and sexuality (p= 0.800), and BIRS social barriers (p=0.777) was not significant. Race was not a significant moderator. Although the main effect of the three-way interaction was not significant, we explored the proposed association further and found some significant within group differences in the QoL scores by race from baseline to 12 months (Table IV).
Table IV.
Within treatment group changes in the quality of life outcomes by race from baseline to 12 months
| Baseline | Change at 12 months | ||||
|---|---|---|---|---|---|
| Outcomes | Randomized group | Race | Mean (95% CI) | Mean (95% CI) | P-value |
| Emotional/Mental Wellbeing Score (Range: 0–100, higher is better) |
|||||
| Exercise | Black | 73.13 (66.10, 80.15) | 3.32 (−1.38, 8.02) | 0.166 | |
| White | 68.15 (60.56, 75.74) | 4.69 (0.32, 9.06) | 0.036 * | ||
| Social Wellbeing Score (Range: 0–6, higher is worse) |
|||||
| Exercise | Black | 1.19 (0.65, 1.73) | 0.21 (−0.16, 0.57) | 0.260 | |
| White | 1.52 (0.93, 2.11) | −0.41 (−0.68, −0.13) | 0.004 * | ||
| Body Image and Relationships-Total Score (Range: 32–160, higher is worse) |
|||||
| Exercise | Black | 94.96 (90.25, 99.67) | −2.44 (−6.62, 1.74) | 0.252 | |
| White | 95.96 (90.96, 100.95) | −2.29 (−5.85, 1.26) | 0.205 | ||
| Social Functioning Score (Range: 0–100, higher is better) |
|||||
| Exercise | Black | 68.05 (58.76, 77.33) | 2.20 (−6.00, 10.39) | 0.597 | |
| White | 75.61 (65.39, 85.83) | 2.58 (−3.63, 8.80) | 0.415 | ||
| Body Image and Relationships-Strength and Health Subscale Score (Range: 12–60, higher is worse) |
|||||
| Exercise | Black | 36.32 (34.28, 38.37) | −0.10 (−2.20, 2.01) | 0.928 | |
| White | 36.95 (34.69, 39.22) | 0.71 (−0.83, 2.26) | 0.364 | ||
| Body Image and Relationships-Appearance and Sexuality Subscale Score (Range: 11–55, higher is worse) |
|||||
| Exercise | Black | 34.40 (32.24, 36.57) | 0.89 (−1.02, 2.79) | 0.361 | |
| White | 33.28 (30.87, 35.69) | 0.66 (−1.38, 2.71) | 0.517 | ||
| Body Image and Relationships-Social Barriers Subscale Score (Range: 9–45, higher is worse) |
|||||
| Exercise | Black | 23.92 (20.24, 27.61) | −3.06 (−6.24, 0.12) | 0.059 | |
| White | 24.79 (20.99, 28.59) | −3.97 (−6.31, −1.63) | 0.001 * | ||
p value < 0.05 indicated statistical significance
Emotional/mental wellbeing
The exercise intervention had a significant effect on emotional/mental wellbeing in White women (Table IV). There were significant improvements in the emotional/mental wellbeing scores (p=0.036) in White women at 12 months post baseline. On the other hand, there was no significant effect of the exercise intervention on emotional/mental wellbeing (p=0.166) in Black women at 12 months post baseline.
Social wellbeing
The exercise intervention had a significant effect on social wellbeing in White women (Table IV). There were significant improvements in the social wellbeing scores (p=0.004) in White women at 12 months post baseline. In contrast, the social wellbeing scores of Black women did not significantly improve from baseline to 12 months.
BIRS total
The BIRS total scores for White women (p=0.205) and Black women (p=0.252) in the exercise intervention were reduced (i.e., lower impairment) at 12 months post baseline, but the results were not statistically significant (Table IV).
BIRS strength and health
There were no significant improvements in the strength and health scores at 12 months post baseline in White women (p=0.364) and Black women (p=0.928) (Table IV). At 12 months post baseline, there was a lower impairment in Black women, whereas a higher impairment was observed in White women.
BIRS appearance and sexuality
The BIRS appearance and sexuality scores for White women (p=0.517) and Black women (p=0.361) increased (i.e., higher impairment) at 12 months post baseline (Table IV).
BIRS social barriers
The exercise intervention had a significant effect on body image as it relates to social barriers in White women (Table IV). There were significant improvements in the social barriers scores in White women (p=0.001) at 12 months post baseline. Black women in the exercise intervention demonstrated improvements in the social barriers scores (p=0.059) at 12 months post baseline, although the result was marginally significant.
Social functioning
The social functioning scores for Black women (p=0.597) and White women (p=0.415) improved at 12 months post baseline, but the results were not statistically significant (Table IV).
Discussion
In this study, we investigated the association between exercise and multiple QoL domains and the effects of race on the proposed association in a racially diverse group of BCS. We found that exercise did not significantly improve the QoL outcomes. Furthermore, race did not moderate the associations between exercise and the QoL outcomes. The current study findings do not support the study hypotheses. Although these associations were not significant, exercise improved most of the QoL outcomes (emotional/mental wellbeing, social wellbeing, body image, body image as it relates to strength and health, and social barriers). In addition, the mean differences across multiple QoL outcomes were similar for Black and White women in the exercise group. This indicates that exercise can be used to improve multiple QoL domains among Black and White women.
Previous studies have demonstrated the association between PA and QoL among cancer survivors. For instance, Diggins and colleagues (2017) found a significant and positive association between PA and social/family wellbeing at time 2 (immediately post 10-week intervention) in Black BCS (N=114) in a randomized controlled trial. (29) Survivors who engaged in any level of PA had significantly better social/family wellbeing but not emotional wellbeing at time 2. Emotional/mental wellbeing and social wellbeing were not found to be significantly associated with PA in the current sample of Black BCS, which is not consistent with the previous results. Another group of researchers showed that mental and emotional wellbeing improved significantly in White BCS after participating in a community based exercise intervention, which is consistent with our study findings. (50) A previous study consisting of a diverse group of BCS showed a significant and positive association between PA and mental health. (16) The sample included 118 Black women. Black and White BCS who met the PA recommendations had significantly better mental health than those who did not meet the PA recommendations, which is consistent with the current study findings in White BCS. (16) The inconsistent findings could be explained by the sample sizes. Compared to the previous study, the current study included a small sample of Black women (N=57). Statistically significant effects are more likely to be observed in larger sample sizes.
Beebe-Dimmer et al. conducted a prospective cohort study among Black cancer survivors (including breast) and found a positive and significant association between PA, social, mental, and emotional wellbeing at the first study follow up (2020). (28) This sample consisted of 852 Black women. Black cancer survivors who met the PA recommendations had significantly better social, mental, and emotional wellbeing. Increasing the level of moderate to vigorous PA was positively associated with improvements in social, mental, and emotional wellbeing. No significant associations between PA and QoL were observed in Black BCS in the current sample. The sample sizes could account for variations in the study results. There was a significant difference in the sample sizes of Black women in the current study compared to the previous study. Larger sample sizes are more likely to yield statistically significant effects.
Two previous meta-analyses were conducted to examine the effect of exercise on QoL among cancer survivors. Mishra and colleagues (2012) found no significant difference in body image between the exercise and control groups, (33) which is consistent with the current results. No significant effects of exercise on mental wellbeing (depression) and emotional wellbeing in BCS were observed from baseline to follow-up, which is consistent with the current study findings. Additionally, the same group of researchers found a significant effect of exercise on social functioning from baseline to 12 weeks, which is not consistent with the current study results. Differences in the exercise frequency, intensity, and type may explain the inconsistent results. For example, the level of exercise frequency, intensity, and type prescribed in the WISER Survivor trial may not have been adequate to yield significant improvements in QoL.
A meta-analysis of 34 randomized controlled trials showed that exercise, specifically supervised exercise, significantly improved the QoL of cancer survivors. (34) Potential moderator effects were identified and tested. The researchers found that the effect of exercise on QoL did not significantly differ by participant demographics (e.g., age), clinical factors (e.g., body mass index), intervention timing and duration, and exercise FITT factors (e.g., time). However, the previous meta-analysis did not examine race as a potential moderator. The current study tested the moderating effect of race on the longitudinal association between PA and QoL in BCS but did not observe a significant interaction. The WISER Survivor trial was not powered to identify and test moderator effects (e.g., race). It is important for future studies to be designed to test moderator effects and include larger samples of Black women in PA studies. Specifically, observational (prospective or retrospective) and experimental studies (e.g., randomized controlled trial) are needed to test these effects and proposed associations. This will help to determine the specific exercise prescription (including FITT) needed to achieve the optimal QoL for Black BCS.
Strengths and Limitations
This study contributes to the literature by helping researchers and public health professionals to better understand the association between exercise and QoL among a diverse group of BCS, specifically Black women. Experimental studies can be less prone to biases compared to observational studies. However, the available secondary data may embed confounding, information bias, and selection bias (loss to follow-up). All models consisted of multiple adjusters to control confounding variables. The original study design and use of randomization reduces selection bias and theoretically balances the study groups on confounding variables. However, when the study samples are small, randomization may not always work to balance the study groups on confounding variables. In the WISER Survivor trial, the control group had a moderate to high loss to follow-up rate (24%) compared with the other groups. (37) After examining the baseline characteristics (age, race, education, and marital status), the results demonstrated no significant differences between participants who completed the study and participants who were lost to follow up. Also, there was some level of missingness in the WISER Survivor trial. Multiple imputation was used to address data loss. In addition, the results from the BIRS were not robust and the findings changed after imputation. Therefore, these preliminary findings should be interpreted with caution.
Furthermore, most of the self-reported measures in this study (e.g., SF-36 survey) have been tested and validated among White cancer survivors. Therefore, the self-reported surveys might not include all the appropriate information to understand and predict behavior among Black cancer survivors. Future studies should develop and validate these study instruments among primarily Black cancer survivors. Additionally, all the participants were diagnosed with breast cancer-related lymphedema, and most were relatively well-educated. Therefore, the findings may not be generalizable to other cancer types, races, socioeconomic groups, and cancer survivors who do not have breast cancer-related lymphedema. On the other hand, targeting an underrepresented group who have worse breast cancer outcomes should be a priority of public health concern. This research may help to encourage other investigators to increase the representation of Blacks in studies focusing on health behavior change, behavioral medicine, exercise oncology, supportive care, and cancer survivorship.
Conclusions
Results show that exercise can improve multiple QoL domains over time among subgroups of BCS after their active treatment. However, the significance of the effect on QoL was isolated to White women. The small sample size in Black women could constrain the statistical significance of observed effects. Future studies are warranted to examine the associations between exercise and multiple QoL domains in larger samples of Black women. PA could play a crucial role in attaining an optimal QoL in Black women.
Acknowledgements:
Natasha Burse has moved to the University of North Carolina at Chapel Hill since completing the research. This work is part of the first author’s dissertation. The authors would like to give special thanks to the Penn State Clinical and Translational Science Institute for their assistance with this project.
Funding:
Natasha Burse was funded by the National Cancer Institute of the National Institutes of Health under Award Number F99CA253762. The WISER Survivor trial was funded by the National Cancer Institute of the National Institutes of Health under Award Number U54155850.
Footnotes
Disclosure of potential conflicts of interest: The authors declare that there is no conflict of interest.
Ethical approval: The WISER Survivor trial was approved by the IRB of the University of Pennsylvania. The current study was approved by the IRB of the Pennsylvania State University College of Medicine (STUDY00014712).
Code availability: The SAS code will not be made available.
Informed consent: The study participants provided signed informed consent and a written clearance from their doctors to participate in the WISER Survivor trial. Informed consent was not obtained in the current study because the research involves no more than minimal risk to subjects and no procedures for which written consent is normally required outside of the research context.
Disclaimer: Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number F99CA253762 and U54155850. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Availability of data and material:
The data and materials will not be made available.
References
- 1.DeSantis C, Siegel R, Bandi P, Jemal A. Breast cancer statistics, 2011. Ca-Cancer J Clin. 2011;61(6):409–18. [DOI] [PubMed] [Google Scholar]
- 2.Breast cancer facts and figures 2017–2018. Atlanta, GA: American Cancer Society Inc; 2017. [Available from: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/breast-cancer-facts-and-figures/breast-cancer-facts-and-figures-2017-2018.pdf [Google Scholar]
- 3.U.S. Cancer Statistics Working Group. U.S. Cancer Statistics Data Visualizations Tool, based on 2020 submission data (1999–2018): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; https://www.cdc.gov/cancer/dataviz. [Google Scholar]
- 4.Miller KD, Nogueira L, Mariotto AB, Rowland JH, Yabroff KR, Alfano CM, et al. Cancer treatment and survivorship statistics, 2019. Ca-Cancer J Clin. 2019;69(5):363–85. [DOI] [PubMed] [Google Scholar]
- 5.U.S. Cancer Statistics Working Group. U.S. Cancer Statistics Data Visualizations Tool, based on 2021 submission data (1999–2019): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; https://www.cdc.gov/cancer/dataviz, released in November 2022. [Google Scholar]
- 6.Eversley R, Estrin D, Dibble S, Wardlaw L, Pedrosa M, Favila-Penney W. Post-treatment symptoms among ethnic minority breast cancer survivors. Oncol Nurs Forum. 2005;32(2):250–4. [DOI] [PubMed] [Google Scholar]
- 7.Meeske KA, Sullivan-Halley J, Smith AW, McTiernan A, Baumgartner KB, Harlan LC, et al. Risk factors for arm lymphedema following breast cancer diagnosis in Black women and White women. Breast Cancer Res Tr. 2009;113(2):383–91. [DOI] [PubMed] [Google Scholar]
- 8.Samuel CA, Pinheiro LC, Reeder-Hayes KE, Walker JS, Corbie-Smith G, Fashaw SA, et al. To be young, Black, and living with breast cancer: a systematic review of health-related quality of life in young Black breast cancer survivors. Breast Cancer Research and Treatment. 2016;160(1):1–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ashing-Giwa K Quality of life and psychosocial outcomes in long-term survivors of breast cancer: A focus on African-American women. J Psychosoc Oncol. 1999;17(3–4):47–62. [Google Scholar]
- 10.Sheppard VB, Llanos AA, Hurtado-de-Mendoza A, Taylor TR, Adams-Campbell LL. Correlates of depressive symptomatology in African-American breast cancer patients. J Cancer Surviv. 2013;7(3):292–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Janz NK, Mujahid MS, Hawley ST, Griggs JJ, Alderman A, Hamilton AS, et al. Racial/ethnic differences in quality of life after diagnosis of breast cancer. Journal of Cancer Survivorship-Research and Practice. 2009;3(4):212–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Blinder VS, Griggs JJ. Health Disparities and the Cancer Survivor. Semin Oncol. 2013;40(6):796–803. [DOI] [PubMed] [Google Scholar]
- 13.Powe BD, Hamilton J, Hancock N, Johnson N, Finnie R, Ko J, et al. Quality of life of African American cancer survivors - A review of the literature. Cancer-Am Cancer Soc. 2007;109(2):435–45. [DOI] [PubMed] [Google Scholar]
- 14.Dow KH, Ferrell BR, Leigh S, Ly J, Gulasekaram P. An evaluation of the quality of life among long-term survivors of breast cancer. Breast Cancer Res Tr. 1996;39(3):261–73. [DOI] [PubMed] [Google Scholar]
- 15.Health-Related Quality of Life (HRQOL). Centers for Disease Control and Prevention. [Available from: https://www.cdc.gov/hrqol/index.htm
- 16.Paxton RJ, Phillips KL, Jones LA, Chang S, Taylor WC, Courneya KS, et al. Associations among physical activity, body mass index, and health-related quality of life by race/ethnicity in a diverse sample of breast cancer survivors. Cancer-Am Cancer Soc. 2012;118(16):4024–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Russell KM, Von Ah DM, Giesler RB, Storniolo AM, Haase JE. Quality of Life of African American Breast Cancer Survivors How Much Do We Know? Cancer Nursing. 2008;31(6):E36–E45. [DOI] [PubMed] [Google Scholar]
- 18.Ahmed RL, Prizment A, Lazovich D, Schmitz KH, Folsom AR. Lymphedema and Quality of Life in Breast Cancer Survivors: The Iowa Women’s Health Study. J Clin Oncol. 2008;26(35):5689–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bowen DJ, Alfano CM, McGregor BA, Kuniyuki A, Bernstein L, Meeske K, et al. Possible socioeconomic and ethnic disparities in quality of life in a cohort of breast cancer survivors. Breast Cancer Res Tr. 2007;106(1):85–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Abrahamson PE, Gammon MD, Lund MJ, Britton JA, Marshall SW, Flagg EW, et al. Recreational physical activity and survival among young women with breast cancer. Cancer-Am Cancer Soc. 2006;107(8):1777–85. [DOI] [PubMed] [Google Scholar]
- 21.Friedenreich CM, Cust AE. Physical activity and breast cancer risk: impact of timing, type and dose of activity and population subgroup effects. Brit J Sport Med. 2008;42(8):636–47. [DOI] [PubMed] [Google Scholar]
- 22.Holmes MD, Chen WY, Feskanich D, Kroenke CH, Colditz GA. Physical activity and survival after breast cancer diagnosis. Journal of the American Medical Association. 2005;293(20):2479–86. [DOI] [PubMed] [Google Scholar]
- 23.Sheehy S, Palmer JR, Rosenberg L. Leisure Time Physical Activity in Relation to Mortality Among African American Women. Am J Prev Med. 2020;59(5):704–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stolley MR, Sharp LK, Oh A, Schiffer L. A Weight Loss Intervention for African American Breast Cancer Survivors, 2006. Preventing Chronic Disease. 2009;6(1):11. [PMC free article] [PubMed] [Google Scholar]
- 25.Benefits and Risks Associated with Physical Activity. American College of Sports Medicine. p. 1–21. [Available from: https://www.acsm.org/docs/default-source/publications-files/acsm-guidelines-download-10th-edabf32a97415a400e9b3be594a6cd7fbf.pdf [Google Scholar]
- 26.ACSM’s exercise testing and prescription. American College of Sports Medicine. Lippincott williams & wilkins; 2018. [Google Scholar]
- 27.Schmitz KH. Balancing Lymphedema Risk: Exercise Versus Deconditioning for Breast Cancer Survivors. Exercise and Sport Sciences Reviews. 2010;38(1):17–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Beebe-Dimmer JL, Ruterbusch JJ, Harper FWK, Baird TM, Finlay DG, Rundle AG, et al. Physical activity and quality of life in African American cancer survivors: The Detroit Research on Cancer Survivors study. Cancer. 2020;126(9):1987–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Diggins AD, Hearn LE, Lechner SC, Annane D, Antoni MH, Whitehead NE. Physical activity in Black breast cancer survivors: implications for quality of life and mood at baseline and 6-month follow-up. Psycho-Oncology. 2017;26(6):822–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Meadows R, Bonner T, Dobhal M, Borra S, Killion JA, Paxton R. Pathways between physical activity and quality of life in African-American breast cancer survivors. Supportive Care in Cancer. 2017;25(2):489–95. [DOI] [PubMed] [Google Scholar]
- 31.Smith AW, Alfano CM, Reeve BB, Irwin ML, Bernstein L, Baumgartner K, et al. Race/Ethnicity, Physical Activity, and Quality of Life in Breast Cancer Survivors. Cancer Epidemiology Biomarkers & Prevention. 2009;18(2):656–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Blanchard CM, Courneya KS, Stein K. Cancer survivors’ adherence to lifestyle behavior recommendations and associations with health-related quality of life: Results from the American Cancer Society’s SCS-II. Journal of Clinical Oncology. 2008;26(13):2198–204. [DOI] [PubMed] [Google Scholar]
- 33.Mishra SI, Scherer RW, Snyder C, Geigle PM, Berlanstein DR, & Topaloglu O Exercise interventions on health-related quality of life for people with cancer during active treatment. The Cochrane database of systematic reviews, 2012(8), CD008465. 10.1002/14651858.CD008465.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Buffart LM, Kalter J, Sweegers MG, Courneya KS, Newton RU, Aaronson NK, et al. Effects and moderators of exercise on quality of life and physical function in patients with cancer: An individual patient data meta-analysis of 34 RCTs. Cancer Treat Rev. 2017;52:91–104. [DOI] [PubMed] [Google Scholar]
- 35.White A, Pollack LA, Smith JL, Thompson T, Underwood JM, Fairley T. Racial and ethnic differences in health status and health behavior among breast cancer survivors-Behavioral Risk Factor Surveillance System, 2009. Journal of Cancer Survivorship-Research and Practice. 2013;7(1):93–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Sturgeon KM, Hackley R, Fornash A, Dean LT, Laudermilk M, Brown JC, et al. Strategic recruitment of an ethnically diverse cohort of overweight survivors of breast cancer with lymphedema. Cancer-Am Cancer Soc. 2018;124(1):95–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Schmitz KH, Troxel AB, Dean LT, DeMichele A, Brown JC, Sturgeon K, et al. Effect of Home-Based Exercise and Weight Loss Programs on Breast Cancer–Related Lymphedema OutcomesAmong Overweight Breast Cancer Survivors The WISER Survivor Randomized Clinical Trial. JAMA Oncology. 2019;5(11):1605–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Winkels RM, Sturgeon KM, Kallan MJ, Dean LT, Zhang Z, Evangelisti M, et al. The women in steady exercise research (WISER) survivor trial: The innovative transdisciplinary design of a randomized controlled trial of exercise and weight-loss interventions among breast cancer survivors with lymphedema. Contemp Clin Trials. 2017;61:63–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Campbell KL, Winters-Stone KM, Wiskemann J, May AM, Schwartz AL, Courneya KS, et al. Exercise Guidelines for Cancer Survivors: Consensus Statement from International Multidisciplinary Roundtable. Med Sci Sports Exerc. 2019;51(11):2375–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Hormes JM, Lytle LA, Gross CR, Ahmed RL, Troxel AB, Schmitz KH. The body image and relationships scale: Development and validation of a measure of body image in female breast cancer survivors. J Clin Oncol. 2008;26(8):1269–74. [DOI] [PubMed] [Google Scholar]
- 41.Brazier JE, Harper R, Jones NMB, Ocathain A, Thomas KJ, Usherwood T, et al. Validating the Sf-36 Health Survey Questionnaire - New Outcome Measure for Primary Care. British Medical Journal. 1992;305(6846):160–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Connor AE, Baumgartner RN, Pinkston CM, Boone SD, Baumgartner KB. Obesity, ethnicity, and quality of life among breast cancer survivors and women without breast cancer: the long-term quality of life follow-up study. Cancer Causes Control. 2016;27(1):115–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.McHorney CA, Ware JE Jr, Lu JF, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care. 1994;32(1):40–66. [DOI] [PubMed] [Google Scholar]
- 44.Launois R, Megnigbeto AC, Pocquet K, Alliot F, editors. Progress in Lymphology XVIII International Congress of Lymphology Sept 2001. A specific quality of life scale in upper limb lymphedema: The ULL-27 questionnaire. 2001. [Google Scholar]
- 45.Kriska AM, Knowler WC, LaPorte RE, et al. Development of questionnaire to examine relationship of physical activity and diabetes in Pima Indians. Diabetes Care. 1990;13(4):401–411. [DOI] [PubMed] [Google Scholar]
- 46.Graham JW. Missing Data Analysis: Making It Work in the Real World. Annu Rev Psychol. 2009;60:549–76. [DOI] [PubMed] [Google Scholar]
- 47.Jakobsen JC, Gluud C, Wetterslev J, & Winkel P When and how should multiple imputation be used for handling missing data in randomised clinical trials - a practical guide with flowcharts. BMC medical research methodology. 2017; 17(1), 162. 10.1186/s12874-017-0442-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Little RJA, & Rubin DB. Statistical Analysis with Missing Data. John Wiley & Sons, Inc. Data. 2002. 10.1002/9781119013563 [DOI] [Google Scholar]
- 49.Walker GA. Common statistical methods for clinical research with SAS examples, Second edition. Cary, NC: SAS Institute Inc; 2002. [Google Scholar]
- 50.Knobf MT, Thompson AS, Fennie K, Erdos D. The Effect of a Community-Based Exercise Intervention on Symptoms and Quality of Life. Cancer Nurs. 2014;37(2):E43–E50. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data and materials will not be made available.