Abstract
BACKGROUND:
Social needs may affect cancer survivors’ health-related quality of life (HRQOL) above and beyond sociodemographic and cancer-related factors. The purpose of this study was to estimate associations between social needs and HRQOL.
METHODS:
Results included data from 1754 participants in the Detroit Research on Cancer Survivors cohort, a population-based study of African American survivors of breast, colorectal, lung, and prostate cancer. Social needs included items related to food insecurity, utility shutoffs, housing instability, not getting health care because of cost or a lack of transportation, and perceptions of neighborhood safety. HRQOL was measured with the validated Functional Assessment of Cancer Therapy–General (FACT-G). Linear regression models controlled for demographic, socioeconomic, and cancer-related factors.
RESULTS:
More than one-third of the survivors (36.3%) reported social needs including 17.1% of survivors reported 2 or more. The prevalence of social needs ranged from 14.8% for food insecurity to 8.9% for utility shutoffs. FACT-G score differences associated with social needs were −12.2 (95% confidence interval [CI] to −15.2 to −9.3) for not getting care because of a lack of transportation, −11.3 (95% CI, −14.2 to −8.4) for housing instability, −10.1 (95% CI, −12.7 to −7.4) for food insecurity, −9.8 (95% CI, −12.7 to −6.9) for feeling unsafe in the neighborhood, −8.6 (95% CI, −11.7 to −5.4) for utility shutoffs, and −6.7 (95% CI, −9.2 to −4.1) for not getting care because of cost.
CONCLUSIONS:
Social needs were common in this cohort of African American cancer survivors and were associated with clinically significant differences in HRQOL. Clinical oncology care and survivorship care planning may present opportunities to screen for and address social needs to mitigate their impact on survivors’ HRQOL.
Keywords: African American, cancer, food insecurity, quality of life, social needs, survivorship
INTRODUCTION
Although cancer mortality declined overall by 29% between 1991 and 2017, cancer mortality remains 13% higher among African Americans than Whites.1 Racial disparities in cancer mortality are related to persisting inequalities in socioeconomic factors that limit access to cancer prevention, early detection, and treatment.2,3 Cancer health disparities are increasingly understood in the context of social determinants of health, including food insecurity, housing and neighborhood conditions, access to transportation, and economic and other social factors.4–6 Within the framework for understanding and addressing social determinants for cancer control, routine screening for health-related social needs is proposed as a strategy to address cancer disparities.4
Substantial evidence demonstrates racial inequalities in unmet social needs, and these particularly affect African Americans.7 The prevalence of food insecurity is nearly 3 times as high among African American (21.2%) as White households (8.1%).7 Decades of systematic, structural discrimination in the housing market have contributed to disproportionately high levels of housing instability among African Americans, with documented negative impacts on health outcomes, including depression, anxiety, hospitalization, and barriers to care.8–10 The prevalence of homeownership is lower among African Americans (47%) than Whites (76%),11 and African Americans are nearly 7 times as likely as Whites to be evicted.12
As interventions are developed to address health-related social needs,13–15 evidence linking social needs to specific, measurable health outcomes is needed. Health-related quality of life (HRQOL) may be a major predictor of cancer-related mortality among African American cancer survivors.16 Defined as an individual’s perceived well-being regarding their mental, physical, and social health status,17 HRQOL tends to be significantly lower among African American cancer survivors in comparison with other groups.18–20 As a means of improving long-term health outcomes among African American cancer survivors, investigators should consider the underlying social needs that these survivors experience.21,22
Patient-reported outcomes, including HRQOL, are critical measures for studies of social needs14,23; however, evidence demonstrating an association between social needs and HRQOL among African American cancer survivors is currently lacking. To address this gap, we estimate associations between social needs and HRQOL in a population-based cohort of African American cancer survivors in Detroit, Michigan. Findings from this work will allow future investigators to consider the role of social needs experienced by African American cancer survivors as a means of improving HRQOL.
MATERIALS AND METHODS
Study Population
Detroit Research on Cancer Survivors (ROCS) is a population-based cohort designed to investigate associations between medical history, health behaviors, financial factors, and health-related outcomes among African American cancer survivors.24,25 Consistent with the National Cancer Institute definition of survivor, eligible participants include both survivors in active treatment and those who have completed treatment.26 Survivors were eligible to join the cohort if they were been diagnosed with primary, invasive breast, colorectal, lung, or prostate cancer since January 1, 2013, and they identified through the Metropolitan Detroit Cancer Surveillance System, a population-based cancer registry covering metropolitan Detroit and a founding participant in the National Cancer Institute’s Surveillance, Epidemiology, and End Results program. The Wayne State University institutional review board approved this research. Detroit ROCS recruitment is ongoing with the aim of recruiting more than 5500 survivors. Results presented here are based on cross-sectional data available from the first 2500 participants who completed the ROCS enrollment survey. Participants completed the survey online via Qualtrics, over the phone with an interviewer, or via a mailed paper survey and received a $25 gift card upon survey completion.
Study Measures
Social needs
Social needs information was collected as a series of yes/no questions based on the Health Leads Social Needs Screening Toolkit.27 These included measures of food insecurity (eating less than the participants felt they should because there was not enough money for food), utility shutoffs for not paying bills, and forgoing health care because of a lack of transportation or cost concerns in the past 12 months; housing instability (concern about not having housing in the next 2 months); and neighborhood safety (whether participants generally felt safe in their neighborhood). The neighborhood safety measure was reverse-coded to indicate the percentage that generally did not feel safe in their neighborhood. We also examined associations between the number of social needs (0, 1, or ≥2) and HRQOL.
Health-related quality of life
HRQOL was measured with the Functional Assessment of Cancer Therapy–General (FACT-G), including subscales for physical, functional, social, and emotional well-being.28 Each subscale includes 6 to 7 statements (eg, “I have pain”), and participants are asked to indicate how each statement applied to them over the past 7 days (not at all, a little bit, somewhat, quite a bit, or very much). Responses are coded so that higher scores indicate higher HRQOL. The FACT-G has demonstrated appropriate reliability, with α coefficients for internal consistency measured at .71 to .83,29 and consistent findings of sensitivity to changes in disease progression and performance status, such that a 5-point difference in the total FACT-G score has been associated with meaningful differences in clinical and subjective indicators.30
Participant and cancer-related characteristics
Participant characteristics included self-reported sex, educational attainment, household income, employment status, marital status, and health insurance coverage at ROCS enrollment. Self-reported cancer-related information included treatments received (any chemotherapy, surgery, and/or radiation), time since diagnosis, and treatment status at ROCS enrollment. Age at diagnosis, Surveillance, Epidemiology, and End Results summary stage, and an indicator of the percentage of residents in the survivor’s census tract with household incomes below the federal poverty level were obtained via linkage with the Metropolitan Detroit Cancer Surveillance System.
Statistical analysis
We calculated univariate descriptive statistics and marginal mean FACT-G scores and 95% confidence intervals (CIs) with robust standard errors by participant demographic and cancer-related characteristic categories. Bivariate statistics included the proportion of participants experiencing each social need by select participant characteristics and the Pearson chi-square test of differences in prevalence of social needs by participant characteristics. Differences in FACT-G scores associated with the presence (vs absence) of each social need were based on marginal means from linear regression models with each individual social need as the exposure and with the FACT-G score as the outcome of interest with robust standard errors. Adjusted models controlled for participant age, sex, education, household income, employment status, marital status, and census-tract poverty via the categories presented in Table 1. Covariates were selected with a directed acyclic graph. Additional models of the association between individual social needs and FACT-G mutually adjusted for each of the other social needs.
TABLE 1.
Participant Characteristics and Associated Mean FACT-G Scores
| Characteristic | Total |
FACT-G |
||
|---|---|---|---|---|
| No. | % | Mean | 95% CI | |
| All survivors | 1754 | 100 | 81.1 | 80.2–82.0 |
| Age | ||||
| <55 y | 468 | 26.7 | 77.4 | 75.5–79.2 |
| 55–64 y | 692 | 39.5 | 80.8 | 79.5–82.2 |
| >65 y | 594 | 33.9 | 84.3 | 83.0–85.6 |
| Sex | ||||
| Women | 863 | 49.2 | 81.1 | 79.8–82.3 |
| Men | 891 | 50.8 | 81.1 | 79.9–82.3 |
| Education | ||||
| Less than high school | 163 | 9.4 | 72.7 | 69.7–75.6 |
| High school/GED | 497 | 28.7 | 79.6 | 77.9–81.3 |
| Some college/2-y degree | 633 | 36.6 | 80.0 | 78.6–81.4 |
| College graduate | 436 | 25.2 | 87.3 | 85.8–88.8 |
| Income | ||||
| <$20,000 | 628 | 35.8 | 73.7 | 72.3–75.2 |
| $20,000-$39,999 | 394 | 22.5 | 80.4 | 78.5–82.2 |
| $40,000-$59,999 | 260 | 14.8 | 85.9 | 83.9–87.9 |
| $60,000-$79,999 | 135 | 7.7 | 88.4 | 85.9–90.9 |
| >$80,000 | 208 | 11.9 | 91.8 | 89.9–93.7 |
| Census-tract poverty | ||||
| 0% to <5% | 95 | 5.4 | 89.8 | 86.8–92.8 |
| 5% to <10% | 202 | 11.5 | 84.9 | 82.4–87.4 |
| 10% to <20% | 302 | 17.2 | 83.6 | 81.7–85.4 |
| >20% | 1153 | 65.8 | 79.1 | 78.0–80.2 |
| Employment status | ||||
| Employed | 515 | 29.4 | 89.2 | 87.9–90.5 |
| Homemaker | 34 | 1.9 | 83.5 | 78.8–88.2 |
| Unemployed | 120 | 6.8 | 67.0 | 63.5–70.4 |
| Retired | 679 | 38.7 | 83.7 | 82.4–84.9 |
| Disability | 381 | 21.7 | 70.1 | 68.2–71.9 |
| Marital status | ||||
| Married/cohabitating | 715 | 40.9 | 85.0 | 83.7–86.2 |
| Widowed | 174 | 10.0 | 80.3 | 77.8–82.9 |
| Divorced/separated | 454 | 26.0 | 79.2 | 77.5–80.9 |
| Never married | 404 | 23.1 | 76.7 | 74.8–78.7 |
| Site | ||||
| Breast | 641 | 36.6 | 81.8 | 80.4–83.2 |
| Colorectal | 231 | 13.2 | 80.1 | 77.5–82.6 |
| Lung | 177 | 10.1 | 75.1 | 72.2–78.0 |
| Prostate | 705 | 40.2 | 82.3 | 81.1–83.6 |
| Stage | ||||
| Localized | 1081 | 62.0 | 82.2 | 81.1–83.3 |
| Regional | 521 | 29.9 | 80.8 | 79.3–82.4 |
| Distant | 142 | 8.1 | 74.1 | 70.9–77.4 |
| Currently in treatment | ||||
| No | 1354 | 78.4 | 82.2 | 81.3–83.2 |
| Yes | 374 | 21.6 | 77.1 | 75.2–79.0 |
Abbreviations: CI, confidence interval; FACT-G, Functional Assessment of Cancer Therapy–General.
Column percentages may not add up to 100 because of rounding.
RESULTS
Results include responses from the first 2500 ROCS respondents. We excluded responses from 653 participants who completed an earlier version of the survey before the inclusion of the social needs questions, 6 participants missing social needs information, 85 participants missing HRQOL information, and 2 participants missing both; this left an analysis sample of 1754 participants.
Table 1 describes the sociodemographic and cancer-related characteristics of ROCS participants and associated mean FACT-G scores. Survivors were on average 62.2 years old at their cancer diagnosis and were evenly split between men and women. Approximately 40% reported no education beyond high school, and 35.8% reported household incomes of <$20,000 per year. At ROCS enrollment, 29.4% were employed, 38.7% were retired, and 21.7% were on disability; 40.9% of the participants were married or living with a partner. Prostate cancer (40.2%) and breast cancer (36.6%) were the most common cancers, and they were followed by colorectal cancer (13.2%) and lung cancer (10.1%). Most cancers (62%) were diagnosed at a localized stage. On average, 27 months elapsed between the cancer diagnosis and ROCS enrollment, and 21.6% were currently in treatment when they completed the ROCS questionnaire.
The mean FACT-G score was 81.1 (Table 1), which was comparable to the average of 80 in a normative population of US adults.31 Older age (FACT-G65–79 = 84.3), greater educational attainment (FACT-Gcollege = 87.3), and household incomes of at least $40,000 (FACT-G = 85.9–91.8) were all associated with above-average HRQOL, as were being married/cohabitating (FACT-G = 85.0), being employed (FACT-G = 89.2) or retired (FACT-G = 83.7) at ROCS enrollment, and living in the lowest poverty areas (FACT-G = 83.6–89.8). Lung cancer (FACT-G = 75.1), a distant stage at diagnosis (FACT-G = 74.1), and currently being in treatment at ROCS enrollment (FACT-G = 77.1) were associated with below-average HRQOL. CIs for each FACT-G score listed here excluded the sample mean of 81.1. Point estimates for FACT-G means that were associated with graduating from college, having a household income of at least $60,000, living in the lowest poverty census tracts (<5%), and being employed were at least 5 points higher than the population average and reflected clinically meaningful differences in HRQOL, whereas those associated with lung cancer and a distant stage at diagnosis were at least 5 points lower.
Table 2 provides the prevalence of social needs by survivor sociodemographic and cancer-related characteristics. More than 1 in 3 survivors (36.3%) experienced at least 1 social need. Food insecurity was most common (14.8%), and it was followed by not seeing a doctor because of cost (13%), housing instability (11.4%), forgoing care because of a lack of transportation (10.2%), feeling unsafe in their neighborhood (9.2%), and concern about utility shutoffs (8.9%). The prevalence of social needs was inversely associated with age, education, and income. More women (40.6%) than men (32.3%) reported any social needs; however, this difference was driven by differences in food insecurity (16.7% among women and 13% among men; P < .001) and forgoing care because of cost (15.4% among women and 10.8% among men; P = .005). The prevalence of the other social needs considered did not differ by sex. Social needs were most common among survivors who were unemployed or on disability at ROCS enrollment. Similarly, social needs were least common among those who were married or cohabitating. More survivors diagnosed with distant-stage disease reported having social needs; this was driven by differences in food insecurity (23.9% for distant-stage disease, 16.7% for regional-stage disease, and 12.7% for local-stage disease; P < .001) and utility shut-offs by stage at diagnosis (12.8% for distant-stage disease, 10.6% for regional-stage disease, and 7.5% for local-stage disease; P = .026). The prevalence of other social needs did not differ by stage.
TABLE 2.
Prevalence of Social Needs by Participant Characteristics and Cancer-Related Factors
| Characteristic | Food Insecurity (n = 258), % | Utility Shutoff (n = 155), % | Housing Instability (n = 198), % | No Doctor, Transportation (n = 178), % | No Doctor, Cost ( n = 226), % | Feel Unsafe in Neighborhood (n = 160), % | Any Social Need (n = 623), % |
|---|---|---|---|---|---|---|---|
| All | 14.8 | 8.9 | 11.4 | 10.2 | 13.0 | 9.2 | 36.3 |
| Age | |||||||
| <55 y | 23.1 | 14.8 | 17.5 | 13.6 | 15.6 | 13.4 | 46.6 |
| 55–64 y | 15.9 | 9.2 | 12.1 | 9.7 | 14.0 | 8.3 | 37.9 |
| ≥65 y | 7.1 | 3.9 | 5.8 | 8.2 | 9.9 | 7.0 | 26.3 |
| χ2P | <.001 | <.001 | <.001 | .014 | .017 | <.001 | <.001 |
| Sex | |||||||
| Women | 16.7 | 9.8 | 11.5 | 10.4 | 15.4 | 9.2 | 40.6 |
| Men | 13.0 | 8.0 | 11.3 | 10.0 | 10.8 | 9.3 | 32.3 |
| χ2P | .027 | .19 | .89 | .78 | .005 | .94 | <.001 |
| Education | |||||||
| Less than high school | 24.7 | 15.4 | 26.1 | 20.4 | 18.6 | 13.0 | 53.7 |
| High school/GED | 14.8 | 8.7 | 11.1 | 11.3 | 10.2 | 8.6 | 35.3 |
| Some college/2-y degree | 16.6 | 8.6 | 11.5 | 12.0 | 17.2 | 11.3 | 40.4 |
| College gradúate | 9.1 | 6.9 | 6.1 | 2.6 | 8.4 | 5.6 | 24.6 |
| χ2P | <.001 | .013 | <.001 | <.001 | <.001 | .005 | <.001 |
| Income | |||||||
| <$20,000 | 28.5 | 14.6 | 21.1 | 20.8 | 18.1 | 14.5 | 55.5 |
| $20,000-$39,999 | 11.7 | 8.4 | 10.0 | 8.1 | 16.2 | 8.4 | 36.8 |
| $40,000-$59,999 | 3.9 | 5.5 | 3.9 | 2.7 | 6.3 | 5.5 | 22.9 |
| ≥$60,000 | 5.4 | 6.3 | 6.4 | 0.0 | 14.5 | 11.3 | 22.2 |
| χ2P | <.001 | <.001 | <.001 | <.001 | <.001 | <.001 | <.001 |
| Census-tract poverty | |||||||
| 0% to <5% | 2.1 | 2.1 | 2.2 | 1.1 | 6.4 | 4.3 | 15.2 |
| 5% to <10% | 7.9 | 8.0 | 6.0 | 5.5 | 12.6 | 4.0 | 27.1 |
| 10% to <20% | 14.4 | 6.0 | 8.5 | 8.1 | 12.4 | 5.7 | 30.6 |
| ≥20% | 17.1 | 10.4 | 13.9 | 12.4 | 13.8 | 11.5 | 41.1 |
| χ2P | <.001 | .008 | <.001 | <.001 | .22 | <.001 | <.001 |
| Employment status | |||||||
| Employed | 8.3 | 7.1 | 6.7 | 3.1 | 11.4 | 6.5 | 27.3 |
| Homemaker | 20.6 | 17.7 | 14.7 | 11.8 | 14.7 | 17.7 | 44.1 |
| Unemployed | 39.5 | 21.0 | 37.0 | 25.2 | 18.0 | 18.8 | 65.3 |
| Retired | 8.4 | 4.3 | 5.4 | 8.8 | 11.1 | 6.4 | 27.4 |
| Disability | 26.0 | 15.7 | 20.3 | 17.5 | 16.1 | 14.1 | 53.4 |
| χ2P | <.001 | <.001 | <.001 | <.001 | .011 | <.001 | <.001 |
| Marital status | |||||||
| Married/living with partner | 7.7 | 6.3 | 6.5 | 6.0 | 9.8 | 6.9 | 24.5 |
| Widowed | 15.5 | 8.6 | 12.1 | 12.1 | 14.4 | 9.3 | 38.2 |
| Divorced/separated | 18.0 | 10.9 | 12.5 | 12.8 | 15.4 | 8.3 | 42.0 |
| Never married | 23.8 | 11.6 | 18.5 | 14.3 | 15.7 | 14.3 | 50.1 |
| χ2P | <.001 | .009 | <.001 | <.001 | .011 | <.001 | <.001 |
| Cancer site | |||||||
| Breast | 15.8 | 10.1 | 11.4 | 9.3 | 14.9 | 8.4 | 38.9 |
| Colorectal | 15.8 | 9.3 | 9.3 | 12.3 | 13.3 | 12.3 | 38.4 |
| Lung | 19.9 | 6.8 | 14.3 | 14.2 | 14.2 | 13.7 | 42.2 |
| Prostate | 12.4 | 8.3 | 11.4 | 9.4 | 10.9 | 7.9 | 31.8 |
| χ2P | .058 | .50 | .48 | .16 | .17 | .032 | .013 |
| Stage | |||||||
| Localized | 12.7 | 7.5 | 10.6 | 10.2 | 12.2 | 8.6 | 33.7 |
| Regional | 16.7 | 10.6 | 12.0 | 9.5 | 15.3 | 9.5 | 40.0 |
| Distant | 23.9 | 12.8 | 16.2 | 12.8 | 11.4 | 13.5 | 42.5 |
| χ2P | <.001 | .026 | .13 | .52 | .20 | .17 | .014 |
| In treatment | |||||||
| No | 14.2 | 8.6 | 10.5 | 9.2 | 13.4 | 8.3 | 35.0 |
| Yes | 17.2 | 9.7 | 14.5 | 13.2 | 11.6 | 12.7 | 41.0 |
| χ2P | .15 | .50 | .029 | .026 | .36 | .010 | .032 |
Abbreviation: GED, general equivalency diploma.
The presence of each social need was associated with large and clinically meaningful differences in FACT-G scores (Table 3). These differences ranged from a 10.3-point difference in FACT-G scores among survivors who did not see a doctor because of cost to differences of 18.6 points for housing instability and 18.9 points for forgoing health care because of a lack of transportation. The mean FACT-G score among survivors who reported no social needs was 86.4 (95% CI, 85.5–87.3), whereas the score was 76.3 (95% CI, 74.4–78.3) among survivors with 1 social need and 65.9 (95% CI, 63.7–68.2) among those with 2 or more. In adjusted models, the presence of each social need was associated with differences in FACT-G scores ranging from 6.7 points for not seeing a doctor because of cost to 12.2 for not seeing a doctor because of a lack of transportation. The presence of 1 or 2 or more social needs was associated with differences in FACT-G scores (in points) of 6.7 (95% CI, 4.6–8.8) and 14.0 (95% CI, 11.4–16.5), respectively. In mutually adjusted models, housing instability, forgoing care because of a lack of transportation, and not feeling safe in the neighborhood remained independently associated with clinically meaningful differences in HRQOL.
TABLE 3.
Differences in FACT-G Scores Associated With the Presence of Social Needs
| Mean FACT-G | 95% CI | Adjustecd |
Mutually Adjusted |
||||
|---|---|---|---|---|---|---|---|
| FACT-G Difference | 95% CI | FACT-G Difference | 95% CI | ||||
| Food insecurity | No | 83.7 | 82.8 to 84.5 | ||||
| Yes | 66.6 | 64.2 to 69.1 | −10.1 | −12.7 to −7.4 | −4.3 | −7.2 to −1.4 | |
| Utility shutoff | No | 82.4 | 81.6 to 83.3 | ||||
| Yes | 68.0 | 64.8 to 71.7 | −8.6 | −11.7 to −5.4 | −2.9 | −6.2 to 0.3 | |
| Housing instability | No | 83.2 | 82.4 to 84.1 | ||||
| Yes | 64.7 | 62.0 to 67.3 | −11.3 | −14.2 to −8.4 | −6.2 | −9.2 to −3.1 | |
| No doctor, transportation | No | 83.1 | 82.2 to 83.9 | ||||
| Yes | 64.2 | 61.3 to 67.1 | −12.2 | −15.2 to −9.3 | −6.8 | −10.0 to −3.6 | |
| No doctor, cost | No | 82.4 | 81.5 to 83.3 | ||||
| Yes | 72.1 | 69.5 to 74.7 | −6.7 | −9.2 to −4.1 | −2.8 | −5.4 to −0.2 | |
| Feel unsafe in neighborhood | No | 82.5 | 81.7 to 83.4 | ||||
| Yes | 67.6 | 64.5 to 70.7 | −9.8 | −12.7 to −6.9 | −7.0 | −9.8 to −4.1 | |
| No. of social needs | None | 86.4 | 85.5 to 87.3 | ||||
| 1 | 76.3 | 74.4 to 78.3 | −6.7 | −8.8 to −4.6 | |||
| ≥2 | 65.9 | 63.7 to 68.2 | −14.0 | −16.5 to −11.4 | |||
Abbreviations: CI, confidence interval; FACT-G, Functional Assessment of Cancer Therapy–General.
Adjusted models control for age, sex, education, income, employment, marital status, and census-tract poverty.
DISCUSSION
In this population-based cohort of African American cancer survivors, more than one-third experienced at least 1 unmet social need, and social needs were associated with large and clinically meaningful differences in HRQOL. These findings highlight the potential usefulness of comprehensive health-related social needs screening and intervention in oncology practice.
Research is increasingly focused on health-related social needs, their association with worse health outcomes, and the development of interventions to address them.23 In the context of cancer care, several previous studies have estimated the prevalence of food insecurity and described interventions to improve food insecurity. Our estimate that 14.8% of African American cancer survivors in Detroit experience food insecurity is similar to an estimate from Simmons et al,32 who reported that 17.4% of patients approached in waiting rooms of a university cancer center experienced food insecurity, but lower than estimates among underserved oncology patients in New York City (56% food insecure)33 or population-based cancer survivors in New Mexico (36% food insecure).34 Differences in measures of food insecurity and in populations included, particularly with respect to socioeconomic factors, food environments, and access to social programs designed to increase food security, likely contribute to these differences in estimates.
We are unaware of previous work describing multiple social needs or their association with HRQOL in the context of cancer care; however, our findings suggest that several social needs are independently associated with clinically meaningful differences in quality of life. Nearly half of the survivors who reported having social needs experienced 2 or more needs (17.1% of 36.3% reporting any social needs), and HRQOL was much worse among survivors reporting multiple social needs in comparison with those reporting 1 need.
Detroit ROCS participants represent a particularly underserved population of African American cancer survivors. Detroit consistently has much higher levels of poverty than the United States as a whole.35 Decades of race-based residential segregation have contributed to high levels of concentrated poverty that particularly affect African Americans and affect their access to economic, political, and social resources.36 In 2017, 34.5% of Detroiters lived in poverty, whereas 15.6% of Michigan residents and 15.3% of all Americans did. The majority of Detroit residents (78.1%) are Black, and 37.7% of Black Detroiters have reported household incomes below the poverty line in comparison with 25.2% of Black Americans.35 Social needs may be particularly prevalent in this population of cancer survivors; however, we expect that our findings of substantially lower HRQOL among cancer survivors experiencing social needs will apply to other survivor populations.
Screening for social needs should be performed in the context of having evidence-based interventions and referral strategies to address them and feedback systems to demonstrate that needs have been met. Identifying strategies to sustainably address health-related social needs in oncology practice is a critical area of future research. Food insecurity was the most common social need identified in this study. There is existing evidence that patients referred to food resources after a positive clinical screen for food insecurity tend to access a wider variety of foods and food resource programs, including food banks, monetary assistance, and the Supplemental Nutrition Assistance Program, which have all been shown to reduce food insecurity.37–39 When hospital-based food pantries were made available in cancer clinics in New York City, utilization was highest among the most vulnerable populations.40 A program providing a monthly $230 grocery voucher to food-insecure patients with cancer in community cancer centers found that 77% of voucher funds went to the purchase of healthy foods, including fruits and vegetables.37 These studies provide promising early results of food-insecurity interventions among patients with cancer, but longitudinal follow-up and rigorous epidemiologic evidence are needed to determine whether addressing food insecurity can improve cancer health outcomes.
Housing instability, a lack of transportation, and neighborhood safety emerged as social needs with independent associations with HRQOL in this cohort. The long history of racial residential segregation in Detroit likely contributes to these environmental social needs.36 Nearly two-thirds of ROCS participants live in census tracts with ≥20% of residents living below the federal poverty level, and these areas of concentrated poverty may expose residents to higher levels of eviction and housing instability, fewer services (including public transportation), and higher levels of crime.41 However, even these seemingly intractable problems have potential solutions. The federally funded Housing Choice Voucher Program provides low-cost housing options on the private rental market to low-income residents, and public housing options shelter approximately 1.2 million Americans.42 Despite the long wait lists for these programs, evidence suggests that once stably housed, recipients of public housing and voucher programs have more resources to spend on food, transportation, and health care.10 Programs already exist to provide transportation to patients with cancer for appointments, including the American Cancer Society’s Road to Recovery program.43 Policies such as the Justice in Policing Act of 2020 are emerging to address neighborhood safety and build trust between law enforcement and African American communities.44 Working within the cancer care delivery system, patient navigators and social workers can use comprehensive social needs screening tools to identify unmet housing and transportation needs and work to link cancer survivors to national and local resources. Although it is unlikely that all of the social needs affecting HRQOL can be addressed directly within the oncology practice setting, our results support the continued development of multi-agency partnerships between cancer centers and community organizations to link survivors to services. Ongoing work to improve access to, funding for, and usability of existing programs is needed, as are longitudinal studies linking policy and programs to changes in HRQOL and other clinical outcomes of interest and qualitative work to understand the experiences and priorities of cancer survivors facing social needs. The strengths of this work include its use of a population-based cohort of African American cancer survivors, a population that experiences a disproportionate burden of social needs and is underrepresented in many studies of cancer survivorship. The detailed survey instruments allowed for the inclusion of several social needs, a validated measure of HRQOL, and the inclusion of several potential confounding factors. The inclusion of several common cancers increases the generalizability of our findings beyond the context of a single cancer site.
Several limitations should also be taken into consideration. Although the inclusion of African American survivors provides valuable information about outcomes in a population that is at particular risk for experiencing social needs and is also underrepresented in previous work, these findings might not be generalizable to other populations. Although Detroit ROCS is population based, an average of 27 months elapsed between cancer diagnosis and ROCS enrollment, and the sickest patients with cancer and those diagnosed with rapidly fatal cancers are likely underrepresented in this work. Although this study incorporates measures of several social needs, each is measured by 1 item, and we are not able to determine the duration or severity of the social needs identified. The data included here are cross-sectional, and this limits our ability to determine the directionality of associations.
Health-related social needs were common in this population of African American cancer survivors and were associated with substantially lower HRQOL. The oncology clinical care setting and survivorship care planning present potential opportunities to identify and address social needs among cancer survivors to improve both cancer outcomes and survivors’ HRQOL beyond the context of cancer.
Acknowledgments
FUNDING SUPPORT
This study was funded by the American Cancer Society (MRSG-17-019), the Karmanos Cancer Institute, the General Motors Foundation, and the National Cancer Institute (grants/contracts U01CA199240, HHSN261261201300011I, and P30CA022453).
Footnotes
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
REFERENCES
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70:7–30. doi: 10.3322/caac.21590 [DOI] [PubMed] [Google Scholar]
- 2.Bach PB, Schrag D, Brawley OW, Galaznik A, Yakren S, Begg CB. Survival of Blacks and Whites after a cancer diagnosis. JAMA. 2002;287:2106–2113. doi: 10.1001/jama.287.16.2106 [DOI] [PubMed] [Google Scholar]
- 3.Singh GK, Williams SD, Siahpush M, Mulhollen A. Socioeconomic, rural-urban, and racial inequalities in US cancer mortality: part I—all cancers and lung cancer and part II—colorectal, prostate, breast, and cervical cancers. J Cancer Epidemiol. 2011;2011:107497. doi: 10.1155/2011/107497 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Alcaraz KI, Wiedt TL, Daniels EC, Yabroff KR, Guerra CE, Wender RC. Understanding and addressing social determinants to advance cancer health equity in the United States: a blueprint for practice, research, and policy. CA Cancer J Clin. 2020;70:31–46. doi: 10.3322/caac.21586 [DOI] [PubMed] [Google Scholar]
- 5.Alderwick H, Gottlieb LM. Meanings and misunderstandings: a social determinants of health lexicon for health care systems. Milbank Q. 2019;97:407–419. doi: 10.1111/1468-0009.12390 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hao Y, Landrine H, Smith T, Kaw C, Corral I, Stein K. Residential segregation and disparities in health-related quality of life among Black and White cancer survivors. Health Psychol. 2011;30:137–144. doi: 10.1037/a0022096 [DOI] [PubMed] [Google Scholar]
- 7.Coleman-Jensen A, Gregory C, Singh A. Household Food Security in the United States in 2013. US Department of Agriculture; 2014. doi: 10.2139/ssrn.2504067 [DOI] [Google Scholar]
- 8.Kushel MB, Gupta R, Gee L, Haas JS. Housing instability and food insecurity as barriers to health care among low-income Americans. J Gen Intern Med. 2006;21:71–77. doi: 10.1111/j.1525-1497.2005.00278.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Burgard SA, Seefeldt KS, Zelner S. Housing instability and health: findings from the Michigan recession and recovery study. Soc Sci Med. 2012;75:2215–2224. doi: 10.1016/j.socscimed.2012.08.020 [DOI] [PubMed] [Google Scholar]
- 10.Desmond M Evicted: Poverty and Profit in the American City. Crown Publishing Group; 2016. [Google Scholar]
- 11.US Census Bureau. Quarterly residential vacancies and homeownership, second quarter 2020. Accessed July 28, 2020. https://www.census.gov/housing/hvs/files/currenthvspress.pdf
- 12.Thomas TA, Toomet O, Kennedy I, Ramiller A.The State of Evictions: Results From the University of Washington Evictions Project. Accessed August 7, 2020. https://evictions.study/washington/
- 13.Alley DE, Asomugha CN, Conway PH, Sanghavi DM. Accountable health communities—addressing social needs through Medicare and Medicaid. N Engl J Med. 2016;374:8–11. doi: 10.1056/NEJMp1512532 [DOI] [PubMed] [Google Scholar]
- 14.Fleegler EW, Lieu TA, Wise PH, Muret-Wagstaff S. Families’ health-related social problems and missed referral opportunities. Pediatrics. 2007;119:e1332–e1341. doi: 10.1542/peds.2006-1505 [DOI] [PubMed] [Google Scholar]
- 15.Gany F, Ramirez J, Chen S, Leng JCF. Targeting social and economic correlates of cancer treatment appointment keeping among immigrant Chinese patients. J Urban Health. 2011;88:98–103. doi: 10.1007/s11524-010-9512-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ashing-Giwa KT. The contextual model of HRQoL: a paradigm for expanding the HRQoL framework. Qual Life Res. 2005;14:297–307. doi: 10.1007/s11136-004-0729-7 [DOI] [PubMed] [Google Scholar]
- 17.Hays RD, Reeve BB. Measurement and modeling of health-related quality of life. In: Heggenhougen HK, Quah S, eds. International Encyclopedia of Public Health. Elsevier; 2008:241–252. doi: 10.1016/B978-012373960-5.00336-1 [DOI] [Google Scholar]
- 18.Matthews AK, Tejeda S, Johnson TP, Berbaum ML, Manfredi C. Correlates of quality of life among African American and White cancer survivors. Cancer Nurs. 2012;35:355–364. doi: 10.1097/NCC.0b013e31824131d9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mollica M, Nemeth L, Newman SD, Mueller M. Quality of life in African American breast cancer survivors: an integrative literature review. Cancer Nurs. 2015;38:194–204. doi: 10.1097/NCC.0000000000000160 [DOI] [PubMed] [Google Scholar]
- 20.Hastert TA, Kyko JM, Reed AR, et al. Financial hardship and quality of life among African American and White cancer survivors: the role of limiting care due to cost. Cancer Epidemiol Biomarkers Prev. 2019;28:1202–1211. doi: 10.1158/1055-9965.EPI-18-1336 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ashing-Giwa K, Tapp C, Brown S, et al. Are survivorship care plans responsive to African-American breast cancer survivors?: voices of survivors and advocates. J Cancer Surviv. 2013;7:283–291. doi: 10.1007/s11764-013-0270-1 [DOI] [PubMed] [Google Scholar]
- 22.Sheppard VB, Llanos AA, Hurtado-de-Mendoza A, Taylor TR, AdamsCampbell LL. Correlates of depressive symptomatology in African-American breast cancer patients. J Cancer Surviv. 2013;7:292–299. doi: 10.1007/s11764-013-0273-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Berkowitz SA, Baggett TP, Edwards ST. Addressing health-related social needs: value-based care or values-based care? J Gen Intern Med. 2019;34:1916–1918. doi: 10.1007/s11606-019-05087-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Beebe-Dimmer JL, Albrecht TL, Baird TE, et al. The Detroit Research on Cancer Survivors (ROCS) pilot study: a focus on outcomes after cancer in a racially-diverse patient population. Cancer Epidemiol Biomarkers Prev. 2019;28:666–674. doi: 10.1158/1055-9965.EPI-18-0123 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Beebe-Dimmer JL, Ruterbusch JJ, Harper FWK, et al. Physical activity and quality of life in African American cancer survivors: the Detroit Research on Cancer Survivors study. Cancer. 2020;126:1987–1994. doi: 10.1002/cncr.32725 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.National Cancer Institute. Definition of survivor. Published February 2, 2011. Accessed September 17, 2020. https://www.cancer.gov/publications/dictionaries/cancer-terms/def/survivor
- 27.Health Leads. Social Needs Screening Toolkit. Published online July 2016. Accessed May 19, 2020. https://nopren.org/wp-content/uploads/2016/12/Health-Leads-Screening-Toolkit-July-2016.pdf
- 28.Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of Cancer Therapy scale: development and validation of the general measure. J Clin Oncol. 1993;11:570–579. [DOI] [PubMed] [Google Scholar]
- 29.Victorson D, Barocas J, Song J, Cella D. Reliability across studies from the Functional Assessment of Cancer Therapy–General (FACT-G) and its subscales: a reliability generalization. Qual Life Res. 2008;17:1137–1146. [DOI] [PubMed] [Google Scholar]
- 30.Brucker PS, Yost K, Cashy J, Webster K, Cella D. General population and cancer patient norms for the Functional Assessment of Cancer Therapy–General (FACT-G). Eval Health Prof. 2005;28:192–211. doi: 10.1177/0163278705275341 [DOI] [PubMed] [Google Scholar]
- 31.Webster K, Cella D, Yost K. The Functional Assessment of Chronic Illness Therapy (FACIT) measurement system: properties, applications, and interpretation. Health Qual Life Outcomes. 2003;1:79. doi: 10.1186/1477-7525-1-79 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Simmons LA, Modesitt SC, Brody AC, Leggin AB. Food insecurity among cancer patients in Kentucky: a pilot study. J Oncol Pract. 2006;2:274–279. doi: 10.1200/jop.2006.2.6.274 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gany F, Lee T, Ramirez J, et al. Do our patients have enough to eat? Food insecurity among urban low-income cancer patients. J Health Care Poor Underserved. 2014;25:1153–1168. doi: 10.1353/hpu.2014.0145 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.McDougall JA, Anderson J, Jaffe SA, et al. Food insecurity and forgone medical care among cancer survivors. JCO Oncol Pract. 2020;16:e922–e932. doi: 10.1200/JOP.19.00736 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Welfare Info. 37.9% poverty rate in Detroit, Michigan. Accessed September 18, 2020. https://www.welfareinfo.org/poverty-rate/michigan/detroit
- 36.Schulz AJ, Williams DR, Israel BA, Lempert LB. Racial and spatial relations as fundamental determinants of health in Detroit. Milbank Q. 2002;80:677–707. doi: 10.1111/1468-0009.00028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Paolantonio L, Kim SY, Ramirez J, et al. Food purchasing behavior of food insecure cancer patients receiving supplemental food vouchers. Support Care Cancer. 2020;28:3739–3746. doi: 10.1007/s00520-019-05183-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Seligman HK, Smith M, Rosenmoss S, Marshall MB, Waxman E. Comprehensive diabetes self-management support from food banks: a randomized controlled trial. Am J Public Health. 2018;108:1227–1234. doi: 10.2105/AJPH.2018.304528 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Decker D, Flynn M. Food insecurity and chronic disease: addressing food access as a healthcare issue. R I Med J. 2018;101:28–30. [PubMed] [Google Scholar]
- 40.Gany F, Lee T, Loeb R, et al. Use of hospital-based food pantries among low-income urban cancer patients. J Community Health. 2015;40:1193–1200. doi: 10.1007/s10900-015-0048-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Desmond M, Gershenson C. Who gets evicted? Assessing individual, neighborhood, and network factors. Soc Sci Res. 2017;62:362–377. doi: 10.1016/j.ssresearch.2016.08.017 [DOI] [PubMed] [Google Scholar]
- 42.Bratt RG, Stone ME, Hartman CW. A Right to Housing: Foundation for a New Social Agenda. Temple University Press; 2006. [Google Scholar]
- 43.American Cancer Society. Road to Recovery. Accessed August 7, 2020. https://www.cancer.org/treatment/support-programs-and-services/road-to-recovery.html
- 44.US House Judiciary Committee. Chair Bass, Senators Booker and Harris, and Chair Nadler introduce the Justice in Policing Act of 2020. Published June 8, 2020. Accessed August 7, 2020. https://judiciary.house.gov/news/documentsingle.aspx?DocumentID=3005