Challenges Experienced by Black Women with Breast Cancer During Active Treatment: Relationship to Treatment Adherence

Abstract

Background

Although rates of death from breast cancer have declined in the USA for both Black and White women since 1990, mortality rates for Black women remain strikingly higher — 40% higher compared to White women (American Cancer Society 1). The barriers and challenges that may be triggering unfavorable treatment-related outcomes and diminished treatment adherence among Black women are not well understood.

Methods

We recruited 25 Black women with breast cancer who were to receive surgery and chemotherapy and/or radiation therapy. Through weekly electronic surveys, we assessed types and severity of challenges across various life domains. Because the participants rarely missed treatments or appointments, we examined the impact of severity of weekly challenges on thoughts of skipping treatment or appointment with their cancer care team using a mixed-effects location scale model.

Results

Both a higher average severity of challenges and a higher deviation of severity reported across weeks were associated with increased thoughts on skipping treatment or appointment. The correlation between the random location and scale effects was positive; thus, those women that reported more thoughts on skipping a dose of medicine or appointment were also more unpredictable with respect to the severity of challenges reported.

Conclusions

Black women with breast cancer are impacted by familial, social, work-related, and medical care factors, and these may in turn affect adherence to treatment. Providers are encouraged to actively screen and communicate with patients regarding life challenges and to build networks of support within the medical care team and social community that can help patients successfully complete treatment as planned.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40615-023-01537-1.

Disparities in breast cancer incidence and mortality between White and Black women highlight the need for further research in this area. In the years from 2001 to 2015, breast cancer incidence declined for White women and increased for Black women; since 2012, the rates have converged and are similar for both groups (126.7/100,100 vs. 130.8/100,000) [1]. If this trend continues, there will be higher incidence rates among Black women than White women for the first time since rates have been recorded. Rates of death from breast cancer have declined in the USA for both groups since 1990; however, mortality rates for Black women remain strikingly higher — 40% higher compared to White women [1]. Although the changes in incidence have yet to be explored, the disparities in mortality have been studied by researchers from varied fields.

Some of these disparities in mortality rates may be explained by differences in tumor characteristics. Among White women, 76% are diagnosed with HR + /HER2- cancers known as Liminal A, associated with a 92% 5-year survival. Seventy-six percent of White women have this type of tumor, versus 61% of Black women [2]. A more deadly type of cancer, HR-/HER2-, also known as Basal-like, has a 77% 5-year survival; this type is diagnosed among 10% of White women and 21% of Black women [2]. Triple negative tumors, which are the most aggressive subtype of HR-HER2- cancers, are more common in Black women, and among young Black women, it is at twice the incidence compared to White women [3].

Stage at diagnosis may also explain some of the differences in mortality. The extent of disease at diagnosis is localized in 66% of White women and 56% of Black women, regional in 26% of White women and 33% of Black women, and distant in 5% of White women and 8% of Black women. Even when diagnosed at the same stage, 5-year survival for Black women is lower than for White women [4].

The social position and context of women’s lives also play a role. Having lower educational attainment and living in poverty are associated with higher mortality rates irrespective of race. However, when examining socioeconomic status (SES) by race, at each quintile of SES, cancer survival for all sites and both sexes is lower for Blacks than non-Hispanic whites [5]. These differences by race and SES remain even after controlling for stage of cancer at diagnosis [5].

Systemic and structural racism play a role in producing these differences in mortality rates, through their association with socioeconomic factors as well as quality and satisfaction with care [6]. Structural racism “involves the complex interplay of macro-level systems, policies as well as institutional practices and processes which accumulate over time and result in unfair advantage of one racial group over another” (p. 2) [7], which may result in disparities in treatment approaches.

Once breast cancer treatment is initiated, guidelines and algorithms that determine the type of chemotherapy that patients receive should eliminate disparities in treatment approach [8], although research has found that Black women are more likely to receive non-guideline concordant treatment [9]. Other aspects of medical care may be experienced differently by Black women compared to White women. Black women experience greater delay in follow up after abnormal mammograms, delay in initiating chemotherapy after surgery, and are more likely to discontinue chemotherapy [8], often because of more serious reactions to treatment [6]. More recent research has focused on differential outcomes for Black women compared to White women for certain standard chemotherapies. For example, chemotherapy-induced neuropathy, a very painful side effect, is higher among Black than White women when using certain agents [10].

Despite all this evidence, limited research examines the treatment trajectory of Black women and specifically focuses on the multiple settings in which barriers and challenges may arise. Therefore, the barriers and challenges that may be triggering unfavorable treatment-related outcomes and diminished treatment adherence are not well understood. This hinders our understanding of the mechanisms contributing to treatment and survival disparities and action to address them. Furthermore, there is recognition of the need to look not only at factors within the medical care system as is typically done but also well beyond it into the everyday lives of these women and the resources available to them through social networks and other means to overcome challenges.

Through weekly patient prompts using electronic surveys, we investigated the barriers and challenges faced by Black women across various realms of their life while undergoing treatment for breast cancer. We assessed types and severity of challenges across the life domains of home/family, social support, work, unmet needs, mental health, physical health, and medical care. We also determined which of these challenges were more prevalent and whether there were any patterns regarding the experiences of challenges as treatment progressed. Our aim was to better understand the challenges experienced by Black women as they progressed through breast cancer treatment, and whether these were related to thoughts about interruption of prescribed treatment.

Methods

Recruitment and Consent

Following IRB review and approval, Black women newly diagnosed with breast cancer were recruited through the assistance of community partners and organizations serving cancer patients such as Little Red Door. Pink-4-Ever Ending Disparities, a non-profit whose mission is to eliminate breast cancer disparities for Black women, partnered with us in this research project. Women were invited to consider participation in the study by a staff member of these organizations, who provided and reviewed a study flyer and a study information sheet with each potential participant. If the potential participant expressed interest, the staff member connected them with an online link to the eligibility survey within REDCap, a secure, HIPAA-compliant research data capture tool. The potential participant was given the option to complete the survey at the community site or from their home.

Eligible potential participants were (1) female, (2) age ≥ 21 years, (3) African American or Black, (4) diagnosed with breast cancer for the first time within the past 3 months, (5) receiving or planning to receive primary treatment that included surgery as well as chemotherapy (neoadjuvant or adjuvant) and/or radiation therapy, and (6) adequate English fluency for completion of surveys. Potential participants were excluded if they had a diagnosis of other active malignancy, though prior treated malignancies other than breast cancer were acceptable.

A member of the study team reviewed the submitted REDCap surveys and then called each potential participant to discuss their eligibility status. If eligible, the study team member explained the study in full and any questions the potential participant asked. Then, the study team member obtained the preliminary verbal consent of the participant and scheduled a time for an initial in-person meeting at their home. At the beginning of the subsequent in-person meeting, each potential participant was given another opportunity to ask any questions about the study, and the informed consent was signed and dated prior to proceeding with any study procedures.

Procedures

Participants recruited to this study were anticipated to undergo standard of care primary treatment for breast cancer with no limitations by stage but which would include surgery as well as chemotherapy (neoadjuvant or adjuvant) and/or radiation therapy typically lasting around 6–9 months. The study procedures were designed to capture their experiences during this period of initial treatment for breast cancer. Participants were not followed during post-treatment adjuvant systemic therapy such as endocrine or immunotherapy.

Initial Survey

Participants completed an initial survey at the time of enrollment collecting (1) demographics, (2) details regarding their breast cancer diagnosis and treatment plan, and (3) screening for depression and anxiety using the Patient Health Questionnaire-4 (PHQ-4) [11], so that the participant would be connected with low-cost mental health resources if the screening tool suggested the need. The baseline survey was completed online with REDCap during the initial in-person meeting. The participants were given an internet-enabled tablet to make it easy for them to complete study surveys and which they were allowed to keep at the end of their study participation. Additionally, participants were given a US $30 gift card as incentive for completing the initial survey. The initial in-person meeting was usually at the participant’s home or location of her preference.

Weekly Experiences Survey

Participants were sent an email with a personal survey link each week throughout their primary course of treatment. This survey was accessible on their mobile devices (phone, tablet) and took approximately 5 minutes to complete via REDCap. This weekly survey was designed for participants to capture their experiences as they occurred and to investigate influential characteristics or conditions that impacted their treatment.

We used the National Comprehensive Cancer Network (NCCN) Distress Thermometer [12] as a guide to develop our Weekly Experiences Survey. The NCCN Distress Thermometer measures, on a scale from 0 to 10, physical, emotional, social, practical, and spiritual concerns as well as problems with communication with care team. For each of these categories, the thermometer lists a number of potential issues. We left our categories open so as not to limit the potential issues that patients may consider when responding. We adapted it based on the extant literature that shows that Black women, in general, place caring for their families above caring for themselves [13], and that social support is as important for Black women with breast cancer [14] as it is for the general population. Rosenzweig et al. [15] found that Black women reported damage in relationships because of the cancer and financial stressors that are a consequence of a cancer diagnosis and its social consequences. They also found that Black patients worried about their community’s reactions to their disease. In studies of communication between Black patients and medical providers, patients report having problems with communication and with receiving proper care in general [16, 17]. Rozenweig et al. found that Black patients of low income expressed frustration with provider’s communication of treatment goals and medical care. All these studies suggested the need for more recent exploration of specific breast cancer challenges experienced by Black women with breast cancer. The weekly survey began with one question asking about life overall, followed by 6 questions asking the participant to rate their experiences in various dimensions of life over the past week (7-day period), using a Visual Analog Scale (VAS) anchored at 0 (best response) and 100 (worst response). The VAS option within the REDCap data capture system automatically scores the point along the horizontal line between anchors that is selected by the participant from 0 to 100, a common range for these scales [18, 19]. In this section of 6 questions assessing weekly challenges, participants were asked about life overall, family life, work life, general needs, emotional support, and physical and emotional state.

We created the total severity index by averaging the weekly challenges minus the question about work life, since several participants did not work. The 5-item weekly challenges section of the survey scale (total severity index) demonstrated good internal consistency with a Cronbach alpha of 0.81, 95% CI = (0.66, 0.91) at baseline, and a Cronbach alpha ranging from 0.63 to 0.94 across all weeks in this study. Test–retest for total severity (mean of the 5 questions) as estimated with a Pearson correlation between total severity measured at baseline and week 1 is r = 0.44, baseline and week 2 is r = 0.55, and week 1 and week 2 is r = 0.66. Low test–retest is to be expected among participants navigating this phase of life/treatment. Treatments often occur in cycles; effects intensify at times, and then return more to normal.

Experiences regarding care were also assessed weekly with a VAS. Specifically, in the weekly survey, participants were asked three questions: “Did your cancer care team take good care of you (doctors, nurses, or others)?” “How well did communication between you and your cancer care team go this week (such as to discuss problems or plans)?” and “Did you think about skipping a dose of medicine or an appointment this week? (see Table 1).

Table 1.

Weekly survey

Question	Visual analog scale anchors [0, 100]
1. Life overall — Overall, how was your life this past week?	0 Fine, no issues $\to$ 100 One of the hardest weeks of my life
2. Family life — How was family life this week?	0 Fine, no issues $\to$ 100 Terrible week, various family issues
3. Work life — (If you do not work, skip) How was work this week?	0 Fine, no issues $\to$ 100 Terrible work week, lots of concerns
4. Needs met — How well were your general needs met this week?	0 I had everything I needed $\to$ 100 Not at all, most needs not met
5. Feel emotionally — Overall, how did you feel emotionally this week?	0 I felt calm, going about my life as usual $\to$ 100 Very depressed, sad, fearful, or worried-No interest in doing my usual activities
6. Social support — Did you get the social and emotional support you needed this week?	0 Yes, I had good support from others $\to$ 100 No, I felt like I am facing this alone
7. Physical — Overall, how did you feel physically this week?	0 Fine, no issues $\to$ 100 One of the hardest weeks of my life
8. Cancer care — Did your cancer care team take good care of you (doctors, nurses, or others)?	0 Yes, the best care →100 Not at all
9. Cancer communication — How well did communication between you and your cancer care team go this week (such as to discuss problems or plans)?	0 Very well, they listened and explained things clearly →100 Not well; they didn’t listen or explain things clearly
10. Thoughts on adherence — Did you think about skipping a dose of cancer medicine or an appointment with your cancer care team this week?	0 No, not at all $\to$ Yes, I thought about it a lot

Finally, participants were asked a few questions about the cancer treatment they had during the past week (last 7 days). Each participant was asked two questions weekly to determine whether or not they missed therapeutic care (in contrast to only thoughts about skipping care): (1) Did you miss or skip a dose of medicine prescribed by your cancer care team this week? and (2) Did you miss or delay an appointment with your cancer care team this week? In both cases, the response options allowed the participant to distinguish missed medicine or appointments that were “because my healthcare team told me I should” versus other reasons (e.g., “my own choice due to side effects or other life problems”). Treatment could be broader than medication, to include surgery or radiation therapy. We were intending to identify any missed therapeutic treatment for the cancer, and we asked the two questions above to get at this. Among medications specifically, we were aiming to capture medications prescribed to treat the cancer, not those given for co-morbidities.

We offered weekly phone assessment as an alternative to the electronic survey. Participants received a US $10 gift card for each weekly survey they completed. When necessary, particularly at treatment inflection points (e.g., beginning or ending of chemotherapy), a research team member contacted participants by phone to clarify their weekly survey responses.

Analysis

This study collected ecological momentary assessment (EMA) data from a weekly survey specifically asking each individual multiple questions which were collected using a VAS ranging from 0 to 100. The primary outcome was how the participant answered the question assessing whether they considered skipping a dose of cancer medicine or an appointment with their cancer care team in the last week (thoughts on adherence). The primary question is whether the severity of challenges reported (obtained from questions described above and listed in Table 1) was associated with their thoughts on adherence.

The original study had planned to enroll 38 participants to have n = 30 participants complete the study accounting for 20% attrition. However, after recruiting 25 participants, recruitment was halted because of COVID-19. Hedeker noted that simulations using the mixed location scale models with small sample sizes (e.g., 20 subjects with 5 data points) often resulted in non-convergence but the non-convergence issue quickly dissipates as numbers increase (e.g., 100 subjects with 10 observations) [20, 21]. As we obtained 25 subjects with approximately 25 data points, basic mixed location scale models reached convergence.

Based on the EMA data, spaghetti plots of the number of challenges across the 24-weeks were created. Additionally, the mean level of severity and standard error for each type of challenge reported were plotted across all weeks. The mean VAS rating assessing patient care (“Did your cancer care team take good care of you (doctors, nurses, or others)?” and “How well did communication between you and your cancer care team go this week (such as to discuss problems or plans)”) were also plotted over time. Baseline descriptive statistics for demographic and breast cancer characteristics were summarized for the cohort of Black women. Additionally, depression (PHQ4) scores were estimated at baseline.

Outcomes of interest in the original proposal were thoughts on skipping an appointment or medication, overall satisfaction with care, missing or skipping a dose of cancer medicine (1/0), and missing or skipping an appointment with the cancer care team (1/0). As this patient population rarely missed a dose of medicine or appointment, the analysis was focused on thoughts on adherence and whether weekly severity of challenges reported were associated with their thoughts on adherence. To investigate the relationship between challenges experienced by Black women during breast cancer treatment and their thoughts on adherence, a mixed-effects location scale model was fit to the continuous outcome of thoughts on adherence, which is an extension of the conventional linear mixed model [20]. As there are several observations nested within each person, a random subject effect is included in the regression model. Both the between-subject and within-subject variances are also modeled as a function of covariates [21]. In this way, how subject-level covariates affect the between-subjects variance in thoughts on adherence can be determined. Also, we can determine how subject-varying and time-varying covariates such as severity of challenges influence the within-subject variance.

The primary outcome for the model was thoughts on skipping a dose of cancer medication or appointment (0–100). For analysis, severity was broken into two components, the component that varies between subjects only (the mean severity for each subject across all time points) and the component that varies within-subjects (how much on a given occasion a subjects’ severity is higher or lower than their mean level). With only 25 patients in the cohort, this model is not adjusting for patient covariates other than week, the mean severity and deviation of severity. Pearson correlations were reported between the proportion of weeks that a dose of cancer medication or medical appointment was missed and the average total severity of challenges and the average individual challenges reported for the entire study period.

Results

Participant Demographic and Disease Characteristics

This cohort of 25 Black women with breast cancer was on average 55 years old (SD = 9.7) with most having at least some college (72%), and more than half supporting only themself financially (60%) (Table 1). Of note, the cohort was mostly low income (78% below US $40,000/year income). Average depression (PHQ4) was 3.9 (SD = 3.3) indicating none to mild depression [11]. The majority of women (68%) had stage 1 or 2 breast cancer. Table 2 includes detailed demographic and disease characteristics.

Table 2.

Baseline patient characteristics

	Total (n = 25)
What is your current age (in years)?
Mean (SD)	55.1 (9.68)
Median (range)	58.0 (32.0, 72.0)
Patient race, n (%)
Black or African American	25 (100.0%)
Are you Hispanic or Latino? n (%)
No	24 (96.0%)
Skip question	1 (4.0%)
What is the highest grade of school you completed or the highest degree you received?, n (%)
High school diploma or GED	7 (28.0%)
Some college but no degree	9 (36.0%)
Associate’s degree	3 (12.0%)
Bachelor’s degree	5 (20.0%)
Advanced degree (Master’s, Doctorate, Professional)	1 (4.0%)
What is your approximate annual household income from all sources (before taxes)? n (%)
Less than US $20,000	8 (32.0%)
US $20,000–$39,999	11 (44.0%)
US $40,000–$49,000	0 (0%)
US $50,000–$69,999	4 (16.0%)
US $70,000–$89,999	3 (12.0%)
On average, how many people, including yourself, does your total family income support? n (%)
1	15 (60.0%)
2	5 (20.0%)
3	4 (16.0%)
6	1 (4.0%)
What is your current work status? n (%)
Employed for wages, full time	6 (24.0%)
Employed for wages, part time	5 (20.0%)
Out of work for < 6 months	1 (4.0%)
Out of work for > 6 months	1 (4.0%)
A homemaker or full-time caregiver	1 (4.0%)
Retired	3 (12.0%)
Unable to work, disabled	8 (32.0%)
Have any of these BLOOD family relatives been diagnosed with breast cancer — mother, grandmother, sister, aunt, daughter, first cousin? n (%)
Yes	14 (58.3%)
No	9 (37.5%)
I don’t know	1 (4.2%)
Depression subscale (PHQ4) range [0–12]
Mean (SD)	3.9 (3.26)
Median (range)	4.0 (0.0, 11.0)
What stage of breast cancer do you have? n (%)
Stage 1	8 (32.0%)
Stage 2	9 (36.0%)
Stage 3	5 (20.0%)
Stage 4	3 (12.0%)

Types and Severity of Challenges Faced by Participants

The five different challenges reported are described in Table 2 (family life, needs met, feel emotionally, social support, and feel physically), with total severity comprised of all these challenges. Descriptive statistics are provided for all types of challenges across study weeks in Supplementary Table 1. Mean total severity at baseline (within the first 3 months after diagnosis) was 26.0 (SD = 19.9) with a range of 2 to 70.2 across the 25 study participants. The mean total severity was the highest for week 3 with a mean of 35.3 (n = 20, SD = 25.9).

Figure 1 provides the weekly mean level of reported challenges by type (family life, work life, needs met, feel emotionally, social support, physical) and life overall for all women obtained from the 0 to 100 VAS. The mean challenge reported for life overall was 31.8 (SD = 20.1) across all weeks. Based on the challenges reported weekly, the lowest average challenges were reported for whether social and emotional support needs were met (n = 25, mean = 19.9, SD = 14.7), general needs were met (n = 25, mean = 21.1, SD = 18.5), and family life (n = 25, mean = 21.0, SD = 15.8) which can also be seen in Fig. 1. The highest mean reported challenges were for how women felt emotionally (mean = 35.3, SD = 20.6) and physically (mean = 34.8, SD = 18.0). Across all weeks (and particularly higher in the first 12 weeks as seen in Fig. 1), challenges related to work-life were high among those employed (n = 14, mean = 37.6, SD = 21.8). Across the five types of challenges, 15 women (60%) had the highest mean across weeks for physical challenges, 5 (20%) for how they felt emotionally, 4 (16%) for whether their general needs were met, and 1 (4%) reported the highest challenge for whether their social and emotional needs were met.

Fig. 1.

Average of challenges reported each week by type

Patterns of Challenges Throughout Treatment

Individual spaghetti plots of total severity by week were created for each participant with indicators if participant had surgery (red), chemotherapy (blue), or radiation (green) during the week. Provided in Supplemental Figs. 1, 2, and 3. Overall, severity fluctuated quite a bit by participant without any visual pattern observed during weeks of treatment.

Relationship Between Challenges and Treatment Adherence

The mixed-effects location scale model (Supplementary material) fit to the outcome of weekly thoughts on adherence (i.e., thoughts on skipping a dose of cancer medication or appointment with cancer care team) included up to 25 weeks on 25 participants; thus, the model contained 426 observations out of 525 (25 × 25) possible. Initial models which were fit did not have covariates in the within-subject or between-subject variance terms and demonstrated a significantly worse fit based on likelihood ratio tests (Supplemental material).

From the final model results (Table 3), after controlling for week (as continuous covariate), both a higher average severity of challenges (p = 0.001) and a higher deviation of severity (p = 0.002) reported across weeks is associated with increased thoughts on skipping a dose or visit. The within-subject (WS) variance decreased over the weeks (p < 0.0001) and significantly increased with deviance of severity (p < 0.0001) but was not significantly affected by the mean level of severity (p = 0.079), although it almost met significance. Exponentiating the estimate for deviance of severity (e^.0443 = 1.045) provides the variance ratio estimate. Thus, with every 1 unit increase in the deviance of severity, the WS variance increases by 4.5% after adjusting for covariates. The between-subject (BS) variance also significantly increased with deviance of severity (p-value = 0.001) but was not associated with week or mean level of severity. The correlation between the random location and scale effects (r = 0.93) is positive; thus, those women that report more thoughts on skipping a dose or visit are also more unpredictable with respect to the severity of challenges reported. These results demonstrate that a woman’s thoughts on skipping a dose of cancer medicine or an appointment with their cancer care team are influenced by both the mean level of severity of challenges reported and deviance of severity.

Table 3.

Mixed-effects location scale model for analysis of ecological momentary assessment (EMA) data. Modeling the effects of average weekly severity reported and the deviation of weekly severity reported on thoughts on skipping a dose or visit. Severity is 0–100 scale of 5 items and outcome is thoughts on skipping a dose or visit (0–100)

Parameter	Estimate	se	t Value	p-value
Mean
Intercept β0	1.4623	0.6126	2.39	0.026
Week β1	0.0159	0.0130	1.22	0.233
Avg severity β3	0.0847	0.0227	3.73	0.001
Deviance severity β4	0.0659	0.0192	3.43	0.002
Within-subject variance
Intercept τ0	1.2936	1.2717	1.02	0.320
Week τ1	− 0.0741	0.0143	− 5.17	< 0.0001
Avg severity τ3	0.0856	0.0466	1.84	0.079
Deviance severity τ4	0.0443	0.0067	6.67	< 0.0001
Between-subject variance
Intercept α0	0.4399	0.9234	0.48	0.638
Week α1	0.0235	0.0123	1.90	0.069
Avg severity α3	0.0439	0.0316	1.39	0.178
Deviance severity α4	0.0385	0.0099	3.89	0.001
Scale
Ln BS variance of scale σw2	2.0572	0.3107	6.62	< 0.0001
Covariance σvw (correlation)	0.9310	0.0496	18.76	< 0.0001

AIC = 2950.5; -2LL = 2562.5; parameters = 14; number of observations = 426; number of participants = 25

Overall, the prevalence of missed/skipped doses of cancer medicine each week was low, from 0% to at most 22.2% (4/18) across the 25 weeks, although most weeks (20/25) at least 1 participant missed a dose. Similarly, the prevalence of missed appointments remained low, from 0% to 14.3% (3/21). Thoughts of skipping, measured on a visual analog scale from not at all (0) to a lot (100), were assessed weekly. Weekly means for all participants ranged from a low of 1.41 (SD = 2.35) to a high of 17.2 (SD = 33.4). Throughout the study, all-participant responses ranged from 0 to 99. Thoughts on skipping were not statistically correlated to actual missed or skipped doses of medication (r = 0.205, p = 0.326) or missed or skipped medical appointments (r = 0.194, p = 0.352), possibly because of the low prevalence of missed or skipped doses or appointments in this sample. The proportion of weeks a medical appointment was reported missed or skipped was significantly correlated with challenges related to how a woman felt emotionally (r = 0.425, p = 0.034) and physically (r = 0.453, p = 0.023), although the correlation of missed medical appointments with mean total severity did not quite meet statistical significance (r = 0.337, p = 0.100). Missing or skipping a dose of cancer medicine was not found to be significantly correlated to total severity of challenges reported (r = 0.049, p = 0.817).

Mean rating for thoughts on adherence was significantly correlated with the mean level of total severity (r = 0.413, p = 0.04) which agrees with our primary model results. Additionally, the specific challenges found to be significantly correlated with thoughts on adherence were how well general needs were met (r = 0.474, p = 0.017) and whether they received the social and emotional support they needed (r = 0.453, p = 0.023). These results are exploratory and were not adjusted for multiple comparisons; thus, results should be interpreted with caution.

Relationship Between Experiences with Care Team and Treatment Adherence

Participants’ experiences regarding care  [“Did your cancer care team take good care of you (doctors, nurses, or others)?” and “How well did communication between you and your cancer care team go this week (such as to discuss problems or plans)”] obtained in the weekly survey are displayed in supplementary information (SI Figs. 4 and 5). As with the earlier challenges, lower scores suggested better care and better communication. The mean of these items across weeks was not correlated with actually skipping a dose or an appointment but both were correlated with thoughts on skipping a dose or appointment, with r = 0.397 (p = 0.049) and r = 0.417 (p = 0.038), respectively for satisfaction with care team and communication with care team (Table 4).

Table 4.

Pearson correlation between average severity and average challenges reported with the proportion of weeks a medical appointment was reported missed, the proportion of weeks a dose was reported missed, and mean thoughts on skipping a dose or visit

Mean challenge reported over study period	Proportion of weeks a medical appt was reported missed (n = 25)	Proportion of weeks a dose was reported missed (n = 25)	Mean thoughts on skipping a dose or visit† (n = 25)
Thoughts on skipping a dose or visita	0.194	0.205	–
Total severity	0.337	0.049	0.413*
Perceived life overall	0.169	0.071	0.257
How was family life this week?	0.304	− 0.077	0.372
How well were your general needs met this week? (e.g., food, transportation, money, other types of help)	− 0.011	− 0.224	0.474*
Overall, how did you feel emotionally this week?	0.425*	0.222	0.316
Did you get the social and emotional support you needed this week?	0.254	− 0.081	0.453*
Overall, how did you feel physically this week?	0.453*	0.324	0.160
Satisfaction with cancer care teamb	0.073	− 0.050	0.397*
Communication with cancer care teamc	0.136	− 0.045	0.417*

^a”Did you consider skipping a dose of cancer medicine or an appointment with your cancer care team this week?” ^b”Did your cancer care team take good care of you (doctors, nurses, or others)?” ^c”How well did communication between you and your cancer care team go this week (such as to discuss problems or plans)?” *p-value < 0.05

Discussion

The study reported here focused on psychosocial and medical care factors associated with treatment behavior and outcomes among Black women with breast cancer. We collected weekly information from Black women with breast cancer from the time of or close to diagnosis to completion of active treatment. We assessed personal challenges such as life overall and family life, including social and emotional support, work-related challenges, and physical and emotional challenges. We also assessed experiences with the health care team, both in feeling taken care of and in effective communication.

Our main aim was understanding skipping treatments or appointments, as extant research suggests this is more common among Black women than White women and could potentially be associated with disparities in mortality. Few of the 25 participants who were followed throughout treatment missed cancer care appointments or medications. Those who did miss planned care also reported feeling poorly both emotionally and physically during the week where the appointment or treatment was missed.

We also assessed thoughts on skipping an appointment or medication, which was much more common among participants. Such thoughts are concerning as they often lead to actual behavior. In an extensive review of the literature, the evidence for thought leading to behavior was described as “profound, extensive, adaptive, multifaceted and empirically strong” (p. 351)[22]. Given the importance of adherence to cancer treatment, such thoughts signify consideration of high-risk behavior and therefore are important to understand as they may give insight into the challenges patients may be facing. Among our participants, those who had such thoughts were having challenges meeting general needs or social and emotional support. Interestingly, those most likely to report thoughts about skipping appointments or medication had the most variability in their weekly reported challenges over the study period, suggesting that their personal life situation might be more unstable than that of others, and increasing the risk for such thoughts to result in actual skipping.

Personal and interpersonal challenges have been frequently reported by Black women with cancer. Challenges such as physical distress, social distress, and fear of gossip/stigmatization have been reported among low-income Black women [15]. Extant research participants have described difficulties dealing with the physical symptoms associated with breast cancer and treatment and being overwhelmed to the point of no action. Their reports of social distress were associated with fear of gossip and stigmatization [15] and are common among Black women with breast cancer. Their fear of being stigmatized because of the diagnosis contributed to their social isolation at a time when social support is especially necessary [15]. These unique challenges are not surprising if one considers that Black women, especially if they have low income like the majority of our sample [23], experience higher psychosocial stressors than the general population. These psychosocial stressors may lead to chronic activation of the stress response, known as allostatic load [24]. Obeng-Gyasi et al. (2021) describe the impact of allostatic load in tumor initiation and progression, and the effects of stress hormones on the body system [24]. Chronic stress results in lower ability to fight challenges to the body, both psychologically and physically, and results in more symptomatology among these women.

Concerns about medical care among patients are important to assess. Relationships between clinicians and other care team members and patients determine care outcome and disparities [25, 26]. The questions regarding experiences with the medical team correlated concurrently with thoughts on skipping appointments or medication. Specifically, thoughts on skipping were associated with not feeling taken care of and ineffective communication with the medical team. Importantly, both questions about medical care were associated with severity of challenges. These findings show that medical providers cannot expect their care to be independent of the patients’ life experiences. Patients who were having especially challenging weeks also reported challenges with medical care and possibly skipping appointments or medication.

Underlying these participants’ concerns about medical care may be a well-understood problem of mistrust in the quality of care they receive. Extant literature has identified mistrust as a factor among Black patients with breast cancer [27, 28]. Black women mistrust providers in many areas including those of technical expertise, interpersonal relationships, and access to quality care [29, 30]. This mistrust is not misplaced. Black women are more likely to receive non-guideline concordant treatment [9]. McCall et al. developed a model of disparity of breast cancer treatment in Black women, the SEMOARS + GEM, which considers symptom experience according to race and social determinants plus genomic, epigenomics, and metabolomics [26]. Their model recognizes the unique experiences that Black women bring to the cancer journey, including that of living within the constraints of structural racism, which may result in more problems with certain treatments, more physical distress, and more problematic experiences with the care team, which in turn may contribute to poorer quality care [26].

It is important to note that in assessing challenges, we were also assessing strengths. Lower scores suggested fewer challenges. Our VAS for weekly challenges ranged from a zero indicating there were no problems to 100 indicating significant challenges. It is reasonable to assume that scoring at or close to zero on any of the challenges identified would suggest that participants were doing well in spite of the cancer and its treatment. In our sample, over the weeks assessed, our participants often reported low to no challenges in various areas, suggesting that our participants were able to find support, strength, and resilience throughout their breast cancer treatment. Future research exploring resilience in more depth may suggest further recommendations for providers, families, and employers of Black women with breast cancer.

Future research might also explore how the experiences of these Black women are similar to or differ from other groups. There is a paucity of recent research focused on experiences during treatment for racial and ethnic groups. Earlier studies have identified some similarities to our findings. Latina women of low income, especially those with limited English proficiency, reported problems with communication with providers and associated to that, low satisfaction with care, along with issues with work, family, and emotional wellbeing [31]. Asian American women report unmet physical and emotional needs and challenges [32]. A consistent finding across all groups is the participants’ concerns about family and care of children [31, 33]. Most studies report qualitative findings. A study such as the one reported here that includes a diversity of racial and ethnic groups would aid in better understanding similarities and differences in their breast cancer experiences.

Limitations

The study included a small sample of Black women with breast cancer. Larger studies may yield additional findings not identified here because of statistical power, including a larger percentage of participants who do not adhere to treatment recommendations. The sample was diverse in terms of demographics; however, all participants were living in a Midwestern city and these findings may not be generalizable to Black women with breast cancer located in other communities with more or less support for breast care. Our sample was mostly low income; Black women with breast cancer at higher income levels may respond differently to these types of assessment. However, although our findings may not be generalizable to higher income women, they do capture a hard-to-reach patient population and one that may be especially in need of study to understand better their experiences. A majority of participants had a family history of breast cancer; thus, these women may have been more motivated to adhere to treatment than the general population. Although we sought to be comprehensive in identifying potential challenges, it is possible that other challenges not identified here may be impacting Black women.

Our measures were self-report, including our outcome measure of missing or skipping or thoughts of skipping therapeutic care (medication or appointment). Even though our question specifically asked about skipping cancer medicine or appointment, it is possible that some participants did not correctly understand whether a particular medication was for treating the cancer or for other purposes. Our weekly assessment provided very useful and time sensitive information on the challenges and thoughts of participants. However, such weekly assessments, and the expectation of having to complete one on any given week, may have influenced the decision to miss or skip an appointment among participants. We do not know whether this is the case, but it is a possibility and a limitation of the otherwise strong design.

Conclusions and Recommendations

The findings reported here improve understanding of the context in which Black women receive treatment, both within the medical setting and in their daily lives. It is clear from the study that Black women with breast cancer are impacted by familial, social, work-related, and medical care factors, and that these may in turn affect adherence to treatment. Providers are encouraged to actively screen and communicate with patients regarding life challenges, potentially using weekly electronic tools such as the one used for this study, and to build networks of support within the medical care team that can help patients navigate these challenges successfully so that they complete treatment as planned.

Furthermore, stakeholders throughout patients’ network of support, including broader health care and social services environment (social service providers, advocates, insurers, faith-based organizations and culturally sensitive support groups) should collectively create interventions and systems to help women better manage other life challenges confronting them during treatment — such as on-the-job issues, financial problems, or strained relationships — so that they can better navigate the unique challenge of breast cancer treatment. Those stakeholders should also advocate for strict enforcement of state and federal laws that provide protections against workplace discrimination, such as the Americans With Disabilities Act and the Federal Family and Medical Leave Act and actively support patients needing work accommodations during treatment, such as periodic breaks or provision of a private area to rest or take medication, adjustments to work schedule, permission to use work telephone to call doctors, or permission to work from home. Better supporting Black women throughout treatment for breast cancer may not only improve their ability to complete treatment but also mitigate known collateral harms, such as unemployment [34], to survivors’ quality of life.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contribution

All authors contributed equally to this work. Silvia Bigatti and Tess Weathers conceptualized the study and planned the methodology. Tess Weathers and Lisa Hayes recruited participants. Silvia Bigatti and Lisa Hayes collected baseline and final data. Tess Weathers collected weekly data and managed data and attention to and retention of participants. Joanne Daggy completed all statistical analyses. All authors contributed to data interpretation, writing, and reviewing the manuscript, and have read and approved the final version.

Funding

This study was funded by a grant from the Walther Cancer Institute.

Data availability

The datasets used in the analyses are available from the corresponding author upon reasonable request.

Declarations

Ethics Approval

This study was performed in line with the principles of the Declaration of Helsinki. The study was approved by the IUPUI Institutional Review Board protocol number 1603281630.

Consent to Participate

All participants provided signed informed consent before completing the study.

Consent for Publication

Not applicable.

Competing Interests

The authors declare no competing interests.