Racial, Ethnic, and Socioeconomic Disparities in Meeting Physical Activity Guidelines among Female Breast Cancer Survivors in the United States

Abstract

Background:

Cancer survivors show low physical activity participation rates in the United States. However, there are limited national-level data on disparities in the prevalence of meeting physical activity guidelines among women with and without breast cancer. We aimed to evaluate national-level trends in meeting physical activity guidelines across demographic and socioeconomic characteristics of breast cancer survivors and women without cancer.

Methods:

Data for women ≥35 years of age with and without breast cancer were obtained from the 2004 to 2018 National Health Interview Survey. We used National Health Interview Survey sample weights to generate national-level prevalence estimates and calculate absolute and relative indices of disparity for breast cancer survivors and women without cancer meeting aerobic (150 minutes/week) and muscle-strengthening guidelines (2 sessions/week) stratified by demographic (e.g., race/ethnicity) and socioeconomic (e.g., homeownership) characteristics.

Results:

We included 5,845 breast cancer survivors and 160,162 women without cancer. The weighted percentage of breast cancer survivors meeting aerobic guidelines was 37.7% compared with 40.9% of women without cancer. Fewer women met muscle-strengthening guidelines. There were lower proportions of women who were younger (<50-years), were non-Hispanic Black, were Hispanic, worked 35+ hours/week, or rented their home among breast cancer survivors meeting aerobic guidelines compared with women without cancer meeting aerobic guidelines.

Conclusions:

Breast cancer survivors were less likely to meet physical activity guidelines compared with women without cancer. Demographic and socioeconomic disparities may exist among breast cancer survivors and women without cancer meeting physical activity guidelines.

Impact:

Targeted interventions may be necessary to address low physical activity participation among breast cancer survivors.

Introduction

President Biden’s Cancer Moonshot initiative aims to reduce age-standardized cancer mortality rates in the United States by 50% in the next 25 years (1). The initiative consists of several mechanisms to achieve this goal, and one potential intervention outlined by researchers includes an evaluation of current strategies to increase physical activity and decrease obesity among breast cancer survivors (2). Pre- and post-diagnosis physical activity may offer various clinical benefits to breast cancer survivors, including an increase in cardiovascular health and strength (3) and reductions in treatment-related side effects (e.g., fatigue; ref. 4), comorbidities (5), recurrence (6–9), and mortality (6, 8–10). These favorable outcomes associated with physical activity may be explained by several biological mechanisms associated with changes in adipose tissue, circulating insulin, and natural killer (NK) cell cytotoxic activity (11–15). Therefore, setting fitness goals to promote physical activity is an important aspect of breast cancer survivorship care (16).

The American Cancer Society, the American College of Sports Medicine, and the American Institute for Cancer Research recommend that cancer survivors engage in the equivalent of 150 to 300 minutes of moderate-intensity physical activity combined with at least two days of muscle-strengthening physical activity each week (17–22). Additionally, the American Society of Clinical Oncology released guidelines in 2022 that recommend that all patients with cancer undergoing treatment should receive physical activity counseling (3). Studies suggest that the prevalence of meeting physical activity guidelines is lower in cancer survivors compared with adult women without cancer (23, 24), and individual factors such as body mass index, treatment, and comorbidities may be associated with physical activity participation of cancer survivors (25–27). Currently, only 20% to 54% of breast cancer survivors meet aerobic guidelines (25, 26, 28, 29), and even fewer, only 12%, meet muscle-strengthening guidelines (25). Low physical activity participation among breast cancer survivors may be related to various clinical, demographic, and socioeconomic characteristics. For instance, pain, discomfort, and lower physical functioning following breast cancer treatment may result in low physical activity participation among breast cancer survivors (30). Similar to the general population, socioeconomic barriers related to education, income, employment, access to fitness facilities, and walkable/safe neighborhoods may also influence physical activity participation among breast cancer survivors (27, 31, 32).

However, to our knowledge, there are limited data evaluating demographic and socioeconomic disparities in meeting physical activity guidelines among breast cancer survivors and adult women without cancer at the national level. Therefore, in our study, we aimed to address this knowledge gap by comparing national-level estimates for meeting physical activity guidelines among breast cancer survivors and adult women without cancer across individual demographic (age, race, ethnicity, and marital status) and socioeconomic (education, employment, health insurance, income, poverty level, homeownership, and transportation) characteristics. In a second aim, we also assessed the severity of disparities in meeting physical activity guidelines across age, race and ethnicity, poverty, education, employment, and homeownership in breast cancer survivors using the NCI’s Health Disparities Calculator (https://seer.cancer.gov/hdcalc/). These data may provide critical information to develop targeted interventions to reduce disparities and improve physical activity participation among breast cancer survivors.

Materials and Methods

Data source

This ecological study used data from the Centers for Disease Control and Prevention’s National Health Interview Survey (NHIS; https://www.cdc.gov/nchs/nhis/index.htm). The NHIS is a cross-sectional survey of the US civilian, noninstitutionalized population that collects information on a broad range of health outcomes via household interviews. Informed consent was obtained from all NHIS participants, and the survey has been approved by the National Center for Health Statistics Ethics Review Board. This study was approved by the NIH Institutional Review Board and was considered as exempt research based on the use of de-identified pre-existing data. This study was conducted in accordance with the US Common Rule ethical guidelines.

Study population

The study population included adult female NHIS participants who completed in-person interviews in 2004 to 2018 and had self-reported physical activity as well as demographic and socioeconomic characteristics of interest available. Breast cancer survivors included female participants who had a current or past breast cancer diagnosis at the time of the survey, were ≥35 years of age at diagnosis, and had not been diagnosed with any other cancer besides skin cancer or melanoma (Fig. 1). The rate of breast cancer in women <35 years of age is low (<4%; ref. 33); however, breast cancer in younger women is more likely to require more aggressive treatment due to the disease being of a high grade and an advanced stage at diagnosis and the patient’s low hormone sensitivity (34). Therefore, we excluded women who were <35 years because younger breast cancer survivors may face different circumstances that could affect their ability to participate in physical activity. We excluded 65,494 women <35 years of age, 788 women who were diagnosed with another cancer besides breast cancer, skin cancer, or melanoma, and 149 women with missing physical activity variables. Adult women without cancer included women who had no cancer history besides skin cancer or melanoma and were ≥35 years of age (Supplementary Fig. S1).

Figure 1.

Inclusion/exclusion criteria for breast cancer survivors. This figure depicts the inclusion and exclusion criteria used for breast cancer survivors. Our final analytic sample consisted of 5,845 female adults ages 35+, were diagnosed with breast cancer as the primary cancer, had no other cancer besides skin cancer, and had no missing physical activity variables.

Primary exposure of interest

The primary exposure of interest was breast cancer diagnosis. Self-reported cancer diagnosis was collected by asking participants if they had ever had cancer of any kind. Participants who answered “no” were included in the “adult women without cancer” sample. Participants who answered “yes” were then asked what kind of cancer it was. Participants who answered only “skin (nonmelanoma),” “skin (do not know what kind),” or “melanoma” were included in the “adult women without cancer” sample. Participants were included in the “breast cancer survivors” sample if they answered only “breast” or answered “breast” and “skin (nonmelanoma)”/“skin (do not know what kind)”/“melanoma.”

Primary outcome of interest

The primary outcome of interest was the prevalence of meeting physical activity guidelines, which was created using self-reported minutes/week of moderate and vigorous leisure-time physical activities, as well as the frequency/week of leisure-time strength training activity. An aerobic physical activity variable was created by converting minutes/week for both moderate and vigorous physical activities to weekly metabolic equivalent of task (MET) minutes for each activity type reported and then adding these MET minutes together to get the total weekly MET minutes (35). Individuals met aerobic guidelines if they participated in ≥150 minutes of moderate-intensity physical activity per week, ≥75 minutes of vigorous intensity physical activity per week, or an equivalent combination of the two (i.e., ≥600 weekly MET minutes; refs. 18, 36). Moderate and vigorous physical activities were converted to weekly MET minutes based on the analysis guidelines (35, 37) of the International Physical Activity Questionnaire (IPAQ) and the Global Physical Activity Questionnaire (GPAQ; refs. 38, 39). Individuals were considered to meet muscle-strengthening guidelines if they reported participating in two or more sessions per week. Participants could either meet both guidelines, meet aerobic guidelines (with or without muscle-strengthening guidelines), meet muscle-strengthening guidelines (with or without aerobic guidelines), or meet neither.

Demographic and socioeconomic characteristics for subgroup analyses

We included demographic and socioeconomic variables as covariates. Variables of interest included self-reported age, race/ethnicity, marital status, education level, employment status, insurance, family income, ratio of family income to poverty threshold, homeownership, and transportation. The NHIS survey questions used to create the variables are described in Supplementary Materials S1.

Studies show that physical activity levels tend to decrease with increasing age (30). Therefore, we included age in our analysis and used self-reported age to create a categorical variable for ages 35 to 49, 50 to 64, and ≥65. Physical activity participation may vary by race and ethnicity (40). For instance, Asian American women have reported cultural reasons, lack of time due to keeping their traditions, and feeling like physical activity was not appropriate for girls as reasons for lower physical activity participation rates (41). We used two questions related to race and ethnicity to create a combined race/ethnicity variable (Hispanic, non-Hispanic Asian, non-Hispanic American Indian/Alaska Native, non-Hispanic Black, non-Hispanic White, and non-Hispanic other/multiple races).

Social support and time available to engage in physical activity could also be associated with marital status, education, employment, and family income (32, 42, 43). Therefore, these variables were included in our analysis. For marital status, we considered married and unmarried categories. We considered three groups for education level—less than high school, high school or general education diploma, and greater than high school. Employment status included the number of hours a participant had worked the previous week. This variable was split into categories (0, 1–34, and ≥35) according to those who did not work, those who worked part-time, and those who worked full-time. The NHIS gathers self-reported best estimates of participants’ total family income from all sources, before taxes, in the past year. We created income categories for $0 to $34,999, $35,000 to $74,999, and $75,000 or more. The ratio of imputed family income to the poverty threshold was calculated to determine poverty status, as poverty may be associated with barriers to physical activity, such as lack of time and access to fitness facilities (31). Income used for this ratio was imputed due to missing data (44). A ratio of less than 1.50 indicates that the individual has an income less than 150% of the federal poverty level, and a ratio greater than 1.50 indicates that family income is greater than 150% of the federal poverty level. Government agencies typically use 150% of the poverty level as an indicator of whether or not a family is eligible for assistance (23, 45).

Homeownership was defined by whether homes were owned or rented or whether individuals had other living arrangements. Homeownership may indicate a positive neighborhood environment associated with increased rates of physical activity (46). Transportation barriers may decrease levels of physical activity if individuals cannot travel to access green spaces, parks, or fitness facilities (47). Some health insurance plans include a fitness benefit, allowing members to access gyms or participate in fitness classes for free or a reduced cost, increasing their likelihood of participating in physical activity (48). Therefore, self-reported health insurance was also included in the analysis.

Statistical analysis

National-level estimates for meeting physical activity guidelines

The NHIS uses a complex, multistage probability sampling design that incorporates stratification, clustering, and oversampling of some subpopulations (e.g., Black, Hispanic, and Asian) in certain years (49). We used NHIS sample weights to generate national-level estimates for the percentage of women meeting aerobic, muscle-strengthening, or both guidelines for breast cancer survivors and adult women without cancer stratified by the aforementioned demographic and socioeconomic characteristics. The sample weights are the inverse of the probability of participant selection into the sample adjusted for nonresponse, potential bias due to sample undercoverage, and poststratification by age, race/ethnicity, and sex using US Census data. The weights are constructed to ensure that each survey response could be expanded to represent other individuals, families, or households in the United States (50). Absolute differences were calculated by comparing the weighted proportion in a particular sociodemographic group among breast cancer survivors who met aerobic or muscle-strengthening physical activity guidelines to the weighted proportion in the same sociodemographic group among adult women without cancer who met aerobic or muscle-strengthening physical activity guidelines. The absolute differences were considered statistically significant if the 95% confidence intervals (CI) of each group did not overlap or if the 95% CIs of an absolute difference crossed zero. All analyses were conducted using SAS, version 9.4, SUDAAN, version 11.0.4, and RStudio, version 2023.06.1 (RRID: SCR_000432).

NCI Health Disparity Calculator

We used the NCI’s Health Disparity Calculator (HD^∗Calc) to assess the severity of demographic and socioeconomic disparities in meeting physical activity guidelines between the groups. The HD^∗Calc is an extension of the Surveillance Epidemiology and End Results^∗Stat statistical software which can be used to estimate absolute and relative indices of disparities. These indices could help researchers monitor health disparities between groups, and the calculator can be used with population-based datasets to calculate absolute and relative measures of disparity. Accordingly, we used the calculator to estimate a “between-group variance (BGV),” which is an absolute measure of disparity, and “mean log deviation (MLD),” which is a relative measure of disparity. These estimates were calculated for groups stratified by age, race and ethnicity, ratio of imputed family income to poverty level, employment, education level, and homeownership for breast cancer survivors who met aerobic or muscle-strengthening guidelines. Larger BGV and MLD values indicate larger disparities within a given group (51).

BGV measures the variance in a population if each individual had the same mean health status as their social group (51). It can be calculated by adding the squared differences in health status, in which µ_j is the average health status of group j, p_j is the population size of group j, and µ is the average health status in a population (51, 52).

$$\text{BGV}\mathit{=}\mathit{\hspace{0.17em}}{\displaystyle \mathrm{\sum }}_{\mathit{j}\mathit{=}\mathit{1}}^{\mathit{J}}{{\mathit{p}}_{\mathit{j}}\left({\mathit{\mu }}_{\mathit{j}}\mathit{-}\mathit{\mu }\right)}^{\mathit{2}}\mathit{.}$$

MLD measures the disproportionality of a group (51). It can be calculated by taking the natural logarithm of the shares of health and the shares of population. Here, p_j is the population share of the group j, µ_j is the health status of the group j, and µ is the average health status in the population (51, 52).

$$\text{MLD}\mathit{=}\mathit{\hspace{0.17em}}\mathit{-}{\displaystyle \mathrm{\sum }}_{\mathit{j}\mathit{=}\mathit{1}}^{\mathit{J}}{\mathit{p}}_{\mathit{j}}\ln \mathrm{}\left(\frac{{\mathit{\mu }}_{\mathit{j}}}{\mathit{\mu }}\right)\mathit{.}$$

Sensitivity analysis

In our primary analyses, we used 600 MET minutes/week as the level for “meeting physical activity guidelines” according to the established GPAQ (37) and IPAQ (35) questionnaires. However, other guidelines, such as the Physical Activity Guidelines for Americans (22), report 450 MET minutes/week as the minimum level of physical activity for meeting guidelines. Therefore, in a sensitivity analysis, we generated national-level estimates for the percentage of women meeting aerobic guidelines for breast cancer survivors and adult women without cancer stratified by the aforementioned demographic and socioeconomic characteristics using 450 MET minutes/week as the minimum number needed to meet guidelines.

Data availability

All data used for analysis are publicly available on the NHIS website (https://www.cdc.gov/nchs/nhis/index.htm). We used Integrated Public Use Microdata Series NHIS to assist in harmonizing and pooling data across survey years (53).

Results

There were a total of 5,845 breast cancer survivors and 160,162 adult women without cancer in the sample (Table 1). The sample consisted of mostly non-Hispanic White women (adult women without cancer, 69.7%; breast cancer survivors, 81.5%), women with high school or higher education (adult women without cancer, 85.1%; breast cancer survivors, 86.8%), women with a family income to poverty ratio of 1.50+ (adult women without cancer, 79.2%; breast cancer survivors, 82.5%), and women who owned a home (adult women without cancer, 75.1%; breast cancer survivors, 81.8%). There was a higher percentage of women 65+ years of age among breast cancer survivors (56.9%) compared with adult women without cancer (25.2%).

Table 1.

Sample characteristics of breast cancer survivors and adult women without cancer^a in the United States.

Variable	Adult women without cancer (n = 160,162)		Breast cancer survivors (n = 5,845)
	n	Weighted col %b	n	Weighted col %b
Race/ethnicity
Hispanic	23,447	12.2	407	5.8
Non-Hispanic Asian	7,832	5.2	221	3.6
Non-Hispanic AIAN	1,183	0.7	43	0.6
Non-Hispanic Black	24,598	12.1	650	8.2
Non-Hispanic White	102,564	69.7	4,505	81.5
Non-Hispanic other/multiple races	538	0.3	19	0.2
Age
35–49	58,099	39.5	384	7.7
50–64	53,485	35.3	1,851	35.4
≥65	48,578	25.2	3,610	56.9
Marital status
Married	72,833	59.2	2,268	53.7
Unmarried	86,674	40.5	3,565	46.1
Unknown	655	0.3	12	0.2
Education level
< High school/GED	26,438	14.1	855	12.7
High school/GED	42,986	27.0	1,695	29.4
> High school/GED	89,707	58.1	3,267	57.4
Unknown	1,031	0.7	28	0.5
Hours worked the previous week
0	77,582	45.4	4,052	66.6
1–34	21,376	14.3	597	10.8
≥35	60,432	39.8	1,181	22.3
Unknown	772	0.5	15	0.2
Health insurance coverage
No	17,659	10.5	135	2.2
Yes	142,071	89.2	-d	-
Unknown	432	0.3	-d	-
Family income
$0–$34,999	62,496	29.6	2,415	32.1
$35,000–$74,999	41,944	27.1	1,547	27.7
≥$75,000	38,907	31.9	1,217	27.6
Unknown	16,815	11.3	666	12.6
Ratio of imputed family income to poverty threshold
<1.50	42,641	20.8	1,303	17.6
1.50–3.49	53,310	32.4	2,131	34.7
≥3.50	64,211	46.8	2,411	47.8
Homeownership
Owned or being bought	110,443	75.1	4,474	81.8
Rented	45,495	22.8	1,203	16.0
Other arrangement/unknown	4,224	2.2	168	2.2
Delayed medical care in the past 12 months due to lack of transportation
No	155,290	97.5	5,661	97.3
Yes	4,263	2.1	157	2.2
Unknown	609	0.4	27	0.5

	n	Weighted % met guidelines	n	Weighted % met guidelines
Meeting exercise guidelinesc
Met aerobic guidelines (with or without muscle-strengthening guidelines)	62,840	40.9	2,083	37.7
Met muscle-strengthening guidelines (with or without aerobic guidelines)	28,430	18.6	987	17.6
Met both guidelines	22,429	14.9	713	13.3

Abbreviations: AIAN, American Indian/Alaska Native; GED: general education diploma.

^aNo cancer diagnosis at the time of the survey.

^bPercentages represent weighted column percentages, which is calculated as the ratio of the count of a single cell to the total count for the column that contains that cell, weighted by sample weight. Ratios are represented as percentages.

^cIndividuals met aerobic guidelines if they participated in ≥150 minutes of moderate-intensity physical activity per week, ≥75 minutes of vigorous-intensity physical activity per week, or an equivalent combination of the two (i.e., ≥600 weekly MET minutes). Individuals were considered to meet muscle-strengthening guidelines if they reported participating in two or more sessions per week.

^dCategories were suppressed when the number of cases or events in any cell was less than 10 to reduce the likelihood of a breach of confidentiality.

National-level estimates for meeting physical activity guidelines

A total of 62,840 adult women without cancer and 2,083 breast cancer survivors met aerobic physical activity guidelines (with or without meeting muscle-strengthening guidelines; Table 2). Breast cancer survivors (37.7%; 95% CI, 36.1%–39.4%) had a lower prevalence of meeting aerobic physical activity guidelines compared with adult women without breast cancer (40.9%; 95% CI, 40.5%–41.4%). The patterns varied by race and ethnicity. The proportion of women who were Hispanic among breast cancer survivors meeting aerobic guidelines (4.2%; 95% CI, 3.4%–5.2%) was significantly lower compared with the proportion of women who were Hispanic among adult women without cancer meeting aerobic guidelines (10.1%; 95% CI, 9.7%–10.5%), for an absolute difference of 5.9 percentage points (% pts; 95% CI, 4.9%–6.8% pts). Similarly, the proportion of women who were non-Hispanic Black among breast cancer survivors meeting aerobic guidelines was significantly lower compared with adult women without cancer meeting aerobic guidelines, for an absolute difference of 3.2% pts (95% CI, 2.1%–4.4% pts). This pattern was reversed for non-Hispanic White women, in which a higher proportion of breast cancer survivors meeting aerobic guidelines were non-Hispanic White compared with the proportion of women without cancer meeting aerobic guidelines who were non-Hispanic White.

Table 2.

National-level (weighted) estimates and absolute differences among female breast cancer survivors and adult women without cancer^a who met aerobic physical activity guidelines^b.

Variable	Adult women without cancera who met aerobic recommendationsb (n = 62,840)				Breast cancer survivors who met aerobic recommendationsb (n = 2,083)				Absolute differencesc
	N	Weighted Col %d	95% CIs		N	Weighted Col %d	95% CIs
			Lower	Upper			Lower	Upper	%pts	Lower 95% CI	Upper 95% CI
Race/ethnicity
Hispanic	7,570	10.1	9.7	10.5	114	4.2	3.4	5.2	5.9	4.9	6.8
Non-Hispanic Asian	3,146	5.1	4.8	5.3	96	4.4	3.3	5.8	0.7	−0.6	1.9
Non-Hispanic AIAN	460	0.6	0.5	0.8	—e	—	—	—	—	—	—
Non-Hispanic Black	7,349	9.2	8.9	9.6	168	6.0	5.0	7.3	3.2	2.1	4.4
Non-Hispanic White	44,094	74.7	74.1	75.3	1,688	84.7	82.7	86.5	−10.0	−11.9	−8.1
Non-Hispanic other/multiple races	221	0.3	0.2	0.3	—e	—	—	—	—	—	—
Age
35–49	26,531	45.1	44.6	45.7	174	9.6	8.0	11.4	35.6	33.8	37.4
50–64	21,930	36.3	35.8	36.8	795	42.5	39.8	45.4	−6.2	−9.0	−3.4
≥65	14,379	18.5	18.1	18.9	1,114	47.9	45.3	50.5	−29.4	−32.0	−26.7
Marital status
Married	32,400	65.1	64.5	65.7	960	61.8	59.2	64.4	3.2	0.6	5.8
Unmarried	30,214	34.7	34.1	35.3	—e	—	—	—	—	—	—
Unknown	226	0.3	0.2	0.3	—e	—	—	—	—	—	—
Education level
< High school/GED	5,606	7.6	7.3	7.9	153	6.3	5.2	7.7	1.3	0.0	2.6
High school/GED	13,197	21.0	20.5	21.5	—e	—	—	—	—	—	—
> High school/GED	43,819	71.0	70.5	71.6	1,459	71.6	69.2	73.9	−0.5	−2.9	1.8
Unknown	218	0.4	0.3	0.4	—e	—	—	—	—	—	—
Hours worked the previous week
0	24,913	37.9	37.4	38.4	1,275	57.9	55.3	60.6	−20.0	−22.7	−17.4
1–34	9,814	16.5	16.1	16.9	—e	—	—	—	—	—	—
≥35	27,819	45.1	44.5	45.6	543	28.3	25.9	30.8	16.8	14.3	19.2
Unknown	294	0.5	0.4	0.6	—e	—	—	—	—	—	—
Health insurance coverage
No	5,933	8.5	8.2	8.8	41	2.0	1.4	2.8	6.5	5.8	7.2
Yes	56,754	91.2	90.9	91.5	—e	—	—	—	—	—	—
Unknown	153	0.3	0.2	0.4	—e	—	—	—	—	—	—
Ratio of imputed family income to poverty threshold
<1.50	10,660	12.9	12.5	13.3	273	9.7	8.2	11.3	3.2	1.7	4.8
1.50–3.49	18,684	27.5	27.0	28.1	637	26.8	24.6	29.2	0.7	−1.6	3.0
≥3.50	33,496	59.6	58.9	60.2	1,173	63.5	60.9	66.0	−3.9	−6.5	−1.3
Homeownership
Owned or being bought	47,248	80.0	79.6	80.5	1,725	86.9	84.9	88.6	−6.8	−8.7	−5.0
Rented	14,312	18.3	17.8	18.8	311	11.4	9.7	13.3	6.9	5.1	8.7
Other arrangement/unknown	1,280	1.6	1.5	1.8	47	1.7	1.2	2.6	−0.1	−0.8	0.6
Delayed medical care in the past 12 months due to lack of transportation
No/unknown	61,920	98.9	98.8	99.0	2,054	98.7	97.9	99.2	0.2	−0.5	0.8
Yes	920	1.1	1.0	1.2	29	1.3	0.8	2.1	−0.2	−0.8	0.5

Bold text indicates that the absolute difference between the two groups is significant.

Abbreviations: AIAN, American Indian/Alaska Native; GED, general education diploma.

^aNo cancer diagnosis at the time of the survey.

^bParticipants met aerobic guidelines if they participated in ≥150 minutes/week of moderate-intensity aerobic activity (with or without meeting muscle-strengthening recommendations).

^cAbsolute differences = weighted column percentages for adult women without cancer − weighted column percentages for breast cancer survivors.

^dPercentages represent weighted column percentages, which is calculated as the ratio of the count of a single cell to the total count for the column that contains that cell, weighted by sample weight. Ratios are represented as percentages.

^eCategories were suppressed when the number of individuals in any cell was less than 10 to reduce the likelihood of a breach of confidentiality.

Comparing breast cancer survivors meeting aerobic physical activity guidelines to adult women without cancer meeting aerobic physical activity guidelines, the breast cancer survivors meeting aerobic guidelines had a significantly lower percentage of individuals who were married (absolute difference: 3.2% pts; 95% CI, 0.6%–5.8% pts), worked 35+ hours/week (absolute difference: 16.8% pts; 95% CI, 14.3%–19.2% pts), had <1.50 family income to poverty ratio (absolute difference: 3.2% pts; 95% CI, 1.7%–4.8% pts), resided in a rented home (absolute difference: 6.9% pts; 95% CI, 5.1%–8.7% pts), and had no health insurance coverage (absolute difference: 6.5% pts; 95% CI, 5.8%–7.2% pts; Table 2). Conversely, a significantly higher proportion of breast cancer survivors meeting aerobic physical activity guidelines compared with adult women without cancer meeting aerobic physical activity guidelines were non-Hispanic White, ≥50 years of age, reported zero hours of work/week, or owned a home.

A total of 987 breast cancer survivors and 28,430 adult women without cancer met muscle-strengthening guidelines (with or without meeting aerobic guidelines; Table 3). Overall, fewer women met muscle-strengthening recommendations compared with aerobic recommendations. The prevalence of meeting muscle-strengthening guidelines was 17.6% (95% CI, 16.3%–19.0%) in breast cancer survivors and 18.6% (95% CI, 18.3%–18.9%) in adult women without cancer. The proportion of Hispanic women among breast cancer survivors meeting muscle-strengthening guidelines (4.2%; 95% CI, 3.1%–5.6%) was significantly lower compared with the proportion of Hispanic women among adult women without cancer meeting muscle-strengthening guidelines (8.4%; 95% CI, 7.9%–8.9%), with an absolute difference of 4.2% pts (95% CI, 2.9%–5.5%). Among breast cancer survivors who met muscle-strengthening guidelines, 60.8% (95% CI, 57.1%–64.5%) were married, whereas 64.9% (95% CI, 64.2%–65.6%) were married among adult women without cancer meeting muscle-strengthening guidelines. Finally, a significantly higher proportion of women without cancer who met muscle-strengthening guidelines worked ≥35 hours per week or lived in a rented household compared with breast cancer survivors who met muscle-strengthening guidelines, with an absolute difference of 16.1% pts (95% CI, 12.7%–19.6% pts) and 6.0% pts (95% CI, 3.7%–8.3% pts), respectively. Conversely, a significantly higher proportion of breast cancer survivors meeting muscle-strengthening guidelines were of ages ≥65, reported zero hours of work in a given week, or owned a home. The results for meeting both guidelines are described in Supplementary Table S1. Similar patterns were observed for meeting both muscle-strengthening and aerobic guidelines.

Table 3.

National-level (weighted) estimates and absolute differences among female breast cancer survivors and adult women without cancer^a who met muscle-strengthening physical activity guidelines^b.

Variable	Adult women without cancera who met muscle-strengthening recommendationsb (n = 28,430)				Breast cancer survivors who met muscle-strengthening recommendationsb (n = 987)				Absolute differencesc
	N	Weighted Col %d	95% CIs		N	Weighted Col %d	95% CIs
			Lower	Upper			Lower	Upper	%pts	Lower 95% CI	Upper 95% CI
Race/ethnicity
Hispanic	2,852	8.4	7.9	8.9	53	4.2	3.1	5.6	4.2	2.9	5.5
Non-Hispanic Asian	1,103	3.9	3.6	4.2	35	3.1	2.0	4.7	0.8	−0.6	2.1
Non-Hispanic AIAN	203	0.6	0.5	0.8	—e	—	—	—	—	—	—
Non-Hispanic Black	3,205	9.0	8.5	9.5	72	4.8	3.6	6.3	4.2	2.8	5.6
Non-Hispanic White	20,960	77.9	77.1	78.6	818	87.4	85.1	89.4	−9.5	−11.8	−7.3
Non-Hispanic Other/multiple races	107	0.3	0.2	0.4	—e	—	—	—	—	—	—
Age
35–49	12,027	45.5	44.7	46.3	78	10.4	8.0	13.4	35.1	32.3	37.9
50–64	9,794	35.8	35.0	36.5	358	40.4	36.4	44.5	−4.6	−8.7	−0.5
≥65	6,609	18.7	18.2	19.4	551	49.3	45.1	53.4	−30.5	−34.7	−26.4
Marital status
Married	14,445	64.9	64.2	65.6	446	60.8	57.1	64.5	4.0	0.3	7.8
Unmarried	13,883	34.8	34.1	35.6	541	39.2	35.5	42.9	−4.3	−8.1	−0.6
Unknown	102	0.3	0.2	0.3	0	0	—	—	0.3	—	—
Education level
< High School/GED	1,809	5.3	5.0	5.7	49	3.7	2.6	5.1	1.6	0.3	2.9
High School/GED	5,036	17.6	17.0	18.2	—e	—	—	—	—	—	—
> High School/GED	21,505	76.8	76.0	77.5	760	77.3	73.7	80.4	−0.5	−3.9	2.9
Unknown	80	0.3	0.2	0.5	—e	—	—	—	—	—	—
Hours worked the previous week
0	10,921	36.4	35.7	37.1	—e	—	—	—	—	—	—
1–34	4,583	17.5	16.9	18.1	104	11.7	9.4	14.5	5.8	3.2	8.4
≥35	12,814	45.6	44.9	46.4	258	29.5	26.3	33.0	16.1	12.7	19.6
Unknown	112	0.4	0.3	0.6	—e	—	—	—	—	—	—
Health insurance coverage
No	2,066	6.6	6.2	7.0	15	1.4	0.8	2.4	5.2	4.3	6.1
Yes	26,304	93.1	92.8	93.5	—e	—	—	—	—	—	—
Unknown	60	0.3	0.2	0.4	—e	—	—	—	—	—	—
Ratio of imputed family income to poverty threshold
<1.50	3,916	10.3	9.9	10.8	89	6.6	4.9	8.6	3.8	1.9	5.6
1.50–3.49	7,861	24.8	24.1	25.6	284	24.7	21.6	28.1	0.1	−3.2	3.5
≥3.50	16,653	64.8	64.0	65.7	614	68.7	65.2	72.1	−3.9	−7.4	−0.4
Homeownership
Owned or being bought	21,789	81.5	80.9	82.2	830	87.2	84.6	89.5	−5.7	−8.1	−3.3
Rented	6,132	17.0	16.4	17.7	139	11.0	9.0	13.5	6.0	3.7	8.3
Other arrangement/unknown	509	1.4	1.3	1.6	18	1.7	0.9	3.2	−0.3	−1.3	0.8
Delayed medical care in the past 12 months due to lack of transportation
No/unknown	27,987	98.7	98.6	98.9	976	99.1	98.1	99.6	−0.4	−1.1	0.3
Yes	443	1.3	1.1	1.4	11	0.9	0.4	1.9	0.4	−0.3	1.1

Bold text indicates that the absolute difference between the two groups is significant.

Abbreviations: AIAN, American Indian/Alaska Native; GED: general education diploma.

^aNo cancer diagnosis at the time of the survey.

^bParticipants met muscle-strengthening guidelines if they participated in ≥2 sessions/week of muscle-strengthening activity (with or without meeting aerobic recommendations).

^cAbsolute differences = weighted column percentages for adult women without cancer − weighted column percentages for breast cancer survivors.

^dPercentages represent weighted column percentages, which is calculated as the ratio of the count of a single cell to the total count for the column that contains that cell, weighted by sample weight. Ratios are represented as percentages.

^eCategories were suppressed when the number of individuals in any cell was less than 10 to reduce the likelihood of a breach of confidentiality.

Figure 2 shows the distribution of meeting aerobic and/or muscle-strengthening physical activity guidelines stratified by age, race/ethnicity, education, and ratio of income to poverty threshold for breast cancer survivors. Overall, women of ages ≥65 showed lower rates for meeting aerobic or muscle-strengthening physical activity guidelines compared with younger women. The proportion of breast cancer survivors who met aerobic guidelines was lower among non-Hispanic Black (27.8%; 95% CI, 23.6%–32.5%) and Hispanic (27.2%; 95% CI, 22.1%–32.9%) women compared with the overall population of breast cancer survivors (37.7%; 95% CI, 36.1%–39.4%). A smaller proportion of breast cancer survivors who had less than a high school education met aerobic guidelines (18.7%; 95% CI, 15.5%–22.4%) or muscle-strengthening guidelines (5.1%; 95% CI, 3.6%–7.1%) compared with the overall population of breast cancer survivors (aerobic: 37.7%; 95% CI, 36.1%–39.4%; muscle strengthening: 17.6%; 95% CI, 16.3%–19.0%). A lower percentage of survivors with a ratio of imputed family income to poverty threshold of <1.50 met aerobic guidelines (20.8%; 95% CI, 18.1%–23.7%) or muscle-strengthening guidelines (6.6%; 95% CI, 5.0%–8.6%) compared with survivors with a ratio of imputed family income to poverty threshold of 1.50 to 3.49 or ≥3.50. The distributions stratified by insurance status, homeownership, and marital status are provided in Supplementary Fig. S2.

Figure 2.

Percentage of breast cancer survivors who met aerobic guidelines and/or muscle-strengthening guidelines. A, The percentage of breast cancer survivors who met aerobic guidelines, stratified by age. B, The percentage of breast cancer survivors who met muscle-strengthening guidelines, stratified by age. C, The percentage of breast cancer survivors who met aerobic guidelines, stratified by race/ethnicity. D, The percentage of breast cancer survivors who met muscle-strengthening guidelines, stratified by race/ethnicity. E, The percentage of breast cancer survivors who met aerobic guidelines, stratified by education level. F, The percentage of breast cancer survivors who met muscle-strengthening guidelines, stratified by education level. G, The percentage of breast cancer survivors who met aerobic guidelines, stratified by the ratio of income to poverty threshold. H, The percentage of breast cancer survivors who met muscle-strengthening guidelines, stratified by the ratio of income to poverty threshold. Breast cancer survivors met aerobic guidelines if they participated in the equivalent of ≥150 minutes per week of activity and met muscle-strengthening guidelines if they participated in two or more strength training sessions per week. Percentages represent weighted percentages among breast cancer survivors, calculated as the ratio of the count of a single cell to the total count for the row that contains that cell, weighted by survey year. Ratios are represented as percentages. *, categories were suppressed when the number of individuals in any cell was less than 10 to reduce the likelihood of a breach of confidentiality. AI/AN, American Indian/Alaska Native; GED, general educational development; HS, high school.

NCI HD*Calc

The HD^∗Calc generated both absolute (BGV) and relative (MLD) disparity indices, in which higher indices represented more severe disparities between the groups. For breast cancer survivors who met aerobic or muscle-strengthening recommendations, the most severe disparities were observed across groups stratified by race and ethnicity (BGV: 1006.3 and 1148.9; MLD: 1.2 and 1.5) and homeownership (BGV: 1413.4 and 1305.2; MLD: 1.0 and 1.1), followed by employment (BGV: 775.6 and 384.6; MLD: 0.5 and 0.2), education level (BGV: 604.3 and 619.0; MLD: 0.3 and 0.4), ratio of income level to poverty threshold (BGV: 342.4 and 237.6; MLD: 0.2 and 0.1), and age (BGV: 320.0 and 541.7; MLD: 0.3 and 0.4; Table 4).

Table 4.

HD*Calc results for absolute and relative disparity indices for breast cancer survivors.

Group	Met aerobic recommendations, with or without muscle strengthening				Met muscle-strengthening recommendations, with or without aerobic
	BGV	95% CI	MLD	95% CI	BGV	95% CI	MLD	95% CI
Race/ethnicity
Ref: non-Hispanic White	1,006.3	(−466.5 to 2,479.2)	1.2	(0.5 to 1.9)	1,148.9	(−533.8 to 2,831.6)	1.5	(0.7 to 2.3)
Ratio of imputed family income to poverty threshold
Ref: ≥3.50	342.4	(6.9 to 677.9)	0.2	(−0.01 to 0.4)	237.6	(4.8 to 470.5)	0.1	(−0.03 to 0.03)
Employment
Ref: ≥35 hours	775.6	(15.5 to 1,535.6)	0.5	(−0.01 to 0.9)	384.6	(7.7 to 761.6)	0.2	(0.02 to 0.3)
Education level
Ref: > high school	604.3	(12.1 to 1,196.6)	0.3	(−0.005 to 0.7)	619.0	(12.4 to 1,225.6)	0.4	(−0.03 to 0.7)
Age
Ref: 35–49	320.0	(6.4 to 633.5)	0.3	(−0.1 to 0.6)	541.7	(10.8 to 1,072.5)	0.4	(−0.1 to 0.9)
Homeownership
Ref: own	1,413.4	(28.3 to 2,798.4)	1.0	(−0.03 to 2.1)	1,305.2	(26.1 to 2,584.2)	1.1	(−0.2, 2.3)

Sensitivity analysis

The national-level estimates using 450 MET minutes/week were similar to the primary analyses using 600 MET minutes/week (Supplementary Table S2).

Discussion

Our study showed that the national-level estimates for the prevalence of breast cancer survivors meeting aerobic and/or muscle-strengthening physical activity guidelines were lower compared with adult women without cancer. Of note, we found that breast cancer survivors meeting aerobic guidelines had a lower percentage of non-Hispanic Black and Hispanic women compared with adult women without cancer meeting aerobic guidelines. Conversely, breast cancer survivors meeting aerobic guidelines had a higher percentage of non-Hispanic White women compared with adult women without cancer. These findings may highlight the need for targeted physical activity interventions for non-Hispanic Black and Hispanic breast cancer survivors. Moreover, the analyses among breast cancer survivors also showed that a lower percentage of those who were non-Hispanic Black, were Hispanic, had less than a high school education, had a <1.50 ratio of imputed family income to poverty threshold, had no health insurance, or lived in a rented home met aerobic guidelines or muscle-strengthening guidelines compared with the overall population of breast cancer survivors, further highlighting the potential need for targeted physical activity interventions for these women.

The exercise participation rates for breast cancer survivors reported in our study are consistent with previous studies of cancer survivors (28, 54). However, it is important to note that previous studies have provided limited data comparing exercise participation rates among breast cancer survivors and adult women without cancer. To our knowledge, this is the first study to quantify and compare the differences in exercise participation rates among breast cancer survivors and adult women without cancer across individual demographic and socioeconomic characteristics. Accordingly, our findings may indicate that the barriers to physical activity prevalent in the general population may have a larger impact on breast cancer survivors in terms of further reducing their ability to engage in physical activity.

Breast cancer survivors may encounter unique challenges and barriers to participating in physical activity. For instance, studies have shown that about 75% of women who have undergone mastectomy report negative body image, dissatisfaction with their physical appearance, and feelings of loss of femininity (55–57). These body image issues may lead to low self-esteem and confidence in these women’s ability to participate in physical activity (58). Breast cancer survivors have also reported low awareness on the benefits of physical activity and high concern about additional risks associated with treatment, such as weight loss, anemia, and immunocompromisation (59, 60). Furthermore, breast cancer treatment such as targeted therapy, hormonal therapy, and chemotherapy are associated with cardiotoxicity (61, 62). Although physical activity may reduce the risk of cardiovascular events (63), when recommending exercise, clinicians may consider survivors’ cardiovascular risk profile following treatment. Studies also show that physical activity may reduce the risk of recurrence and mortality and also improve the overall physical and immune functions in breast cancer survivors (4, 5, 64). Therefore, targeted interventions may be necessary to improve physical activity participation and help women diagnosed with breast cancer achieve the benefits of physical activity during survivorship care.

In both breast cancer survivors and adult women without cancer, fewer women met muscle-strengthening guidelines than aerobic guidelines. As shown in previous studies, women typically show lower adherence rates than men to muscle-strengthening physical activity guidelines (65). These differences could be related to negative stigma associated with women lifting weights, boredom, and lack of strength training knowledge (66, 67). However, participation in muscle-strengthening physical activity may have beneficial effects on bone health among women (68, 69). Muscle strengthening is even more important for patients with cancer, as many cancer treatments are associated with poor bone health, muscle atrophy, and loss of strength (70, 71). Muscle-strengthening physical activity can reduce or delay muscle wasting in patients with chronic diseases (72, 73). Incorporating muscle-strengthening recommendations into cancer survivorship care could help increase participation in these activities among breast cancer survivors.

Our results showed that, among breast cancer survivors, a lower proportion of older women (ages ≥65) met aerobic or muscle-strengthening guidelines compared with younger women (ages 35–49 and 50–64; Fig. 2). Older breast cancer survivors may report low physical activity rates due to increased comorbidities, intimidation of physical activity, or lack of guidance and support (74). Similar trends have also been reported for all cancer survivors using the NHIS and other data sources (23). For example, studies using NHIS and community health data found that physical activity rates among cancer survivors may decline with age due to increased comorbidities, loss of strength, and poorer quality of life (75, 76). Older breast cancer survivors may need extra guidance and support to engage in physical activity.

The racial and ethnic differences in meeting physical activity guidelines shown in this study are consistent with the current literature showing the complex and multidimensional nature of health disparities related to physical activity (77–79). Structural racism may play an important role in adherence to physical activity (80), in which racial and ethnic differences in the number of work hours, income, homeownership, living in areas with limited access to green spaces or recreational areas, access to safe infrastructure, walkability, and neighborhood crime rates may also contribute to disparities in physical activity participation (81–84). Previous studies have found that non-Hispanic Black, Hispanic, and Asian individuals are less likely to engage in physical activity compared with non-Hispanic White individuals, which could be due to racial and ethnic differences in work hours, income, education, insurance, access to safe neighborhoods, and fitness facilities (23, 76, 84).

Our results showed that breast cancer survivors who had lower than a high school education, lower family income, no health insurance, and lived in rented homes had lower levels of adherence to aerobic or muscle-strengthening guidelines. Individuals with lower socioeconomic characteristics may encounter barriers to physical activity, such as lack of access to fitness facilities, recreation spaces, walkable and safe neighborhoods, adequate childcare, and fewer flexible work arrangements (85–87). Therefore, physical activity interventions that incorporate broader contextual factors associated with where women live and work may help address disparities in physical activity participation.

Our study presents several limitations. First, NHIS data are self-reported; therefore, there may be measurement errors, social desirability bias, and/or recall bias, which may have affected the validity of our results. There are several national surveys, including the NHIS, the Behavioral Risk Factor Surveillance System, and the National Health and Nutrition Examination Survey, that provide information on physical activity participation in the United States. We used NHIS data for this study as they provided detailed information on the frequency, intensity, time, and the type of physical activity to evaluate the prevalence of meeting aerobic and muscle-strengthening physical activity guidelines for women with breast cancer compared with women without breast cancer (88). In contrast, the Behavioral Risk Factor Surveillance System data provided a binary variable, in which participants were asked every other year if they did physical activity or not (89). The National Health and Nutrition Examination Survey dataset did not include information on the type of physical activity (i.e., muscle-strengthening or aerobic activity; ref. 90). Second, NHIS data do not provide information on the type of melanoma; therefore, we were unable to exclude women with a history of melanoma that may have required treatment beyond surgery. However, this may have had a limited impact on our overall findings because melanoma accounts only for about 1% of skin cancers (91). Third, the sample sizes for several categories in our analysis, especially those from underrepresented and minority groups, were small (<10 individuals). We did not show categories in our results that had less than 10 individuals, and our findings highlight the need for greater representation (e.g., American Indian/Alaska Native) in data collection. Fourth, we considered weekly MET minutes for physical activity guidelines that were outlined by the GPAQ and IPAQ (35, 37). However, it is important to acknowledge that there are other methods to determine what physical activity guidelines are and how to calculate weekly MET minutes, and some consider the minimum weekly MET minutes to meet guidelines to be as low as 450 (22). Our results from the sensitivity analyses show similar patterns for meeting aerobic physical activity guidelines for 450 MET minutes/week and 600 MET minutes/week. Finally, because the NHIS does not collect data on clinical characteristics (e.g., stage at diagnosis, tumor grade, treatment, etc.), we were unable to investigate how clinical factors may influence meeting physical activity guidelines and disparities among breast cancer survivors.

Overall, the percentage of women meeting current physical activity guidelines was lower among breast cancer survivors compared with adult women without cancer. Lower adherence to physical activity guidelines may be driven by various demographic, clinical, and socioeconomic characteristics. Future public health investments and clinical decision tools are needed to individualize physical activity prescriptions that incorporate both individual and contextual factors that may impact physical activity participation.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S1: Inclusion/Exclusion Criteria for Adult Women without Cancer

Supplementary Figure S2

Supplementary Figure S2: Percentage of Breast Cancer Survivors Who Met Aerobic Guidelines and/or Muscle Strengthening Guidelines.

Supplementary Methods S1

Supplementary Methods S1: NHIS Survey Questions

Supplementary Table S1

Supplementary Table S1: National-level (weighted) Estimates and Absolute Differences among Female Breast Cancer Survivors and Adult Women without Cancer Who Met Both Aerobic and Muscle Strengthening Physical Activity Guidelines

Supplementary Table S2

Supplementary Table S2: Sensitivity Analysis: National-level (weighted) Estimates and Absolute Differences among Female Breast Cancer Survivors and Adult Women without Cancer Who Met Aerobic Physical Activity Guidelines

Authors’ Disclosures

J. Jayasekera reports funding from the Division of Intramural Research, NIH, and the NIH Distinguished Scholars Program during the conduct of the study. No disclosures were reported by the other authors.

Disclaimer

Opinions and comments expressed in this article belong to the authors and do not necessarily reflect those of the US Government, the Department of Health and Human Services, the NIH, the NCI, or the National Institute on Minority Health and Health Disparities. The study funders had no role in the design of the study, the collection, analysis, or interpretation of the data, the writing of the manuscript, or the decision to submit the manuscript for publication.

Authors’ Contributions

K.M. Wojcik: Formal analysis, investigation, methodology, writing–original draft, writing–review and editing. O.W.A. Wilson: Investigation, writing–review and editing. M.S. Shiels: Methodology, writing–original draft, writing–review and editing. V.B. Sheppard: Investigation, writing–review and editing. J. Jayasekera: Conceptualization, supervision, funding acquisition, investigation, methodology, writing–original draft, project administration, writing–review and editing.