Skip to main content
NCBI home page
Neuro-Oncology Practice logoLink to Neuro-Oncology Practice
. 2025 Jul 18;13(1):189–205. doi: 10.1093/nop/npaf069

The impact of neighborhood-level disadvantage on symptoms, mood, and health-related quality of life among the primary brain tumor population

Zuena Karim 1, Kimberly Robins 2, Orieta Celiku 3, Yeonju Kim 4, Hope Miller 5, Elizabeth Vera 6, Jacqueline B Vo 7, Mark R Gilbert 8, Terri S Armstrong 9, Macy L Stockdill 10,✉
PMCID: PMC12965653  PMID: 41798120

Abstract

Background

Evidence assessing the impact of neighborhood-level disadvantage and population density on outcomes of individuals with cancer is growing but has not been evaluated among the primary brain tumor (PBT) population. We evaluated associations of neighborhood-level disadvantage and population density with symptoms and health-related quality of life (HRQOL) among adults with a PBT.

Methods

Neighborhood-level disadvantage, measured using the Area Deprivation Index (ADI), and population density were evaluated against symptoms (MDASI-BT and PROMIS Anxiety/Depression Short Forms v1.0 8a) using linear regression models and against HRQOL (EQ-5D-3L) using logistic regression models. Models were adjusted for age, sex, race/ethnicity, tumor grade, functional status, and tumor recurrence. Results were further stratified by tumor grade.

Results

Of 643 participants, 24% lived in more disadvantaged neighborhoods, while 39% resided in non-urbanized areas. Patients in more disadvantaged neighborhoods reported greater symptom severity (β = 1.10, 95%CI [1.00, 1.20], P = .041) and activity-related interference (β = 1.11, 95%CI [1.01, 1.22], P = .03) than those in less disadvantaged neighborhoods. Among patients with low-grade tumors, living in more disadvantaged neighborhoods was associated with worse symptom interference (β=1.24, 95%CI [1.04, 1.50], P = .020), anxiety (β=1.03, 95%CI [1.01, 1.06], P = .023), and difficulties with mobility (OR = 3.30, 95%CI [1.09, 10.01], P = .035) and self-care (OR = 4.68, 95%CI [1.46, 14.96], P = .009). Patients in non-urbanized areas were more likely to experience difficulties with self-care (OR = 2.07, 95%CI [1.23, 3.48], P = .006) than those in urbanized areas.

Conclusion

Future studies should consider evaluating PBT symptoms with data on regional resources other than individual income as the next step to inform interventional work with under-represented and under-resourced populations.

Trial Registration Number NCT#: NCT02851706.

Keywords: primary brain tumors, signs and symptoms, health disparities, neighborhood characteristics

Key Points.

  • Based on the Area Deprivation Index, patients enrolled in the Neuro-Oncology Branch-Natural History Study living in more disadvantaged neighborhoods reported greater symptom severity and activity-related interference.

  • Patients diagnosed with low-grade PBTs living in more disadvantaged neighborhoods reported greater symptom severity, anxiety, and worse HRQOL.

Importance of the Study.

Neighborhood disadvantage has been linked to worse pain and anxiety among cancer survivors; however, little is known about the effects of neighborhood disadvantage on primary brain tumor (PBT) symptoms. Therefore, we evaluated associations of neighborhood-level disadvantage and population density with symptoms and health-related quality of life (HRQOL) among a cohort of adult participants with a PBT in a large observational study at the National Institutes of Health.

Among 643 participants diagnosed with a PBT, 24% lived in more disadvantaged neighborhoods. Those living in more disadvantaged neighborhoods reported greater symptom severity and activity-related interference. Among patients with low-grade tumors, living in more disadvantaged neighborhoods was associated with worse symptom interference, anxiety, and difficulties with mobility and self-care. Residing in less disadvantaged neighborhoods may lead to better healthcare access and better-managed patient symptoms. Understanding the impact of neighborhood deprivation on symptoms could reveal interventional targets to mitigate patient outcome disparities.

Despite advancements in diagnosis and treatment modalities, primary brain tumors (PBTs) remain associated with high mortality rates.1,2 Primary brain tumors represent a diverse group of central nervous system (CNS) tumors, each with its own unique histological features and clinical presentation.3 Patients with PBTs present with various symptoms, including seizures, headaches, affective symptoms, and neurologic and cognitive impairments, which can continue over the course of the disease trajectory and impact daily functioning and health-related quality of life (HRQOL).4,5 Other than tumor type and location, symptom presentation may be affected by patient access to healthcare and other available resources,3,6,7 highlighting the need for continued research to enhance patient care and improve patient outcomes.

Patient outcomes (eg, symptom burden, anxiety, depression, HRQOL) serve as valuable indicators of the effect of PBTs on an individual’s health,8–11 social functioning,12 and even prognosis.4 Socioeconomic factors have been implicated in shaping health-related outcomes, i.e., symptom burden and psychological distress, in cancer populations.13 Over time, studies have shifted from focusing on single-metric socioeconomic variables to broader evaluations of regional characteristics to better understand the multifactorial contributions of socioeconomic disadvantage to health inequities.13–15 Neighborhood-level disadvantage, one type of composite measure of area-level deprivation, provides a comprehensive assessment of socioeconomic factors that influence health-related outcomes.14,16 Prior studies have demonstrated that living in more socioeconomically disadvantaged neighborhoods is related to worse quality of life and patient outcomes among cancer survivors,17 and those with breast18 or prostate cancer.19 Furthermore, higher levels of neighborhood-level disadvantage have been associated with worse anxiety among patients with advanced cancer.13

Individuals living in more disadvantaged neighborhoods are less likely to access healthcare services, treatment, and clinical trials.20,21 Similarly, individuals living in rural areas or far from specialized cancer centers face similar geographic- and socioeconomic-related barriers and are less likely to have timely access to healthcare resources.22,23 Research indicates that patients with cancer living in rural or non-urbanized areas are less likely to receive treatment or participate in clinical trials, with distance and associated costs being potential barriers.18,20,21,24 Similarly, for patients with PBTs, clinical trials are usually located in metropolitan areas and often require longer and more costly travel for non-urban residents which can consequently contribute to lower enrollment rates in clinical trials and worse health-related outcomes for patients.23–25

While previous studies have explored the impact of geography-based characteristics on survival among patients with PBTs,26–28 there is a paucity of literature regarding their influence on patient symptoms, which have been linked to prognosis and HRQOL in the PBT population.4 Further characterizing these relationships may shed light on PBT health-related outcomes disparities associated with area of residence and may inform targeted interventions aimed at improving symptoms and HRQOL. Therefore, we explored the associations between neighborhood-level disadvantage and population density with patient-reported outcomes (PROs) (i e, symptom burden, anxiety, depression, and HRQOL) among a cohort of 643 adults with a PBT enrolled in a large observational research study.

Methods

Study Population

This analysis utilized study entry data of a primary CNS tumor patient cohort (N = 933) from the National Cancer Institute (NCI) Neuro-Oncology Branch’s Natural History Study (NOB-NHS; NCT#: NCT02851706; PI: Terri S. Armstrong)-an observational, longitudinal study including clinical data and PRO measures. Upon referral, participants could enroll in the NOB-NHS at any point along their disease trajectory (at the time of diagnosis, active treatment, surveillance, etc.). This study was approved by the Institutional Review Board, and individuals provided written informed consent. Cohort individuals who were ≥18 years old and had available zip code and PRO data were included in the analysis (N = 643) (Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Flow diagram outlining the Neuro-Oncology Branch Natural History Study participants with study entry data included in the analysis.

*“Other tumor types” indicates participants with a spine or brain and spine tumor.

Exposures

Area Deprivation Index

Neighborhood-level disadvantage was measured by the Area Deprivation Index (ADI). The ADI is a composite measure of neighborhood-level socioeconomic disadvantage. The ADI is comprised of 17 socioeconomic indicators abstracted from the U.S. census.16,29 Scores range from 0 to 100, with higher scores indicating more disadvantage. Consistent with previous literature,20,30 we dichotomized ADI according to the first national-level ADI quartile (40) into those who lived in more disadvantaged neighborhoods (ADI≥40) and in less disadvantaged neighborhoods (ADI < 40) using participant zip codes. Zip codes are a sequence of numbers assigned to geographical areas in the United States (U.S.) that the U.S. Postal Service uses to sort and organize the delivery of mail. Zip codes were available for participants in the NOB-NHS cohort.

Population Density

Classifications of zip codes as urbanized or non-urbanized, representing both rural and suburban populations, were based on the U.S. Department of Agriculture criteria, which defines urban as meeting a population density threshold of 1,000 people per square mile.31

Outcome Variables

Symptom burden and interference were measured by the MD Anderson Symptom Inventory-Brain Tumor (MDASI-BT).32 The MDASI-BT is a self-reported measure that provides a symptom severity score comprised of affective, cognitive, neurologic, treatment-related, general disease, and gastrointestinal domains, as well as an interference score composed of activity- and mood-related interference. Symptom severity and interference items are measured on a scale ranging from 0 to 10, with higher scores indicating worse symptoms and interference.

HRQOL was measured by the EQ-5D-3L,33 a self-reported measure of overall health based on the 5 dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension is rated a score of 1 (no problems), 2 (some problems), or 3 (extreme problems).

Anxiety and depressive symptoms were measured by the 14-item Patient-Reported Outcomes Measurement Information System (PROMIS) Short Form v1.0-Anxiety 8a and Depression 8a.34 For both measurements, the rating of each item is self-selected from 1 (never) to 5 (always). Higher scores indicate greater anxiety or depressive symptoms.

Statistical Analyses

Descriptive statistics summarized the patient sample and provided PRO summary scores. Demographic, clinical characteristics, and PROs were compared by ADI and population density levels using Students’ t-tests, chi-square tests, or Fisher exact tests. Effect size was reported using Hedges’ g, and either Phi or Cramer’s V. ADI and population density were used as explanatory variables in separate models to assess the impact of neighborhood-level disadvantage and population density on patient symptoms, anxiety, depression, and HRQOL. MDASI-BT symptom factors and interference subscales and PROMIS-Anxiety and -Depression t-scores were modeled as continuous outcome variables, while the EQ-5D-3L domains were modeled as binary outcomes (none/mild vs. moderate/severe). For all outcome variables, 3 different models were generated—(1) an unadjusted model, (2) a model adjusted only for age and sex (male, female), and (3) a model adjusted for age, sex, race/ethnicity (non-Hispanic White, Hispanic [all races], non-Hispanic Asian, non-Hispanic Black, Other), tumor grade (low, high), tumor recurrence (yes, no), and functional status (poor, good) as measured by Karnofsky Performance Scores (KPS). The KPS is a widely used tool to assess the functional status of patients and is scored from 0 (dead) to 100 (normal functioning).35 We defined patients with a KPS of 90–100 as having good functional status and those with a KPS of ≤80 as having poor functional status.36 A sub-analysis was completed to compare patients with low- vs. high-grade tumors. Data transformations were completed to meet model assumptions, as appropriate. Linear regression generated beta and standard error coefficients for the models with continuous outcomes. For the EQ-5D-3L models, logistic regression generated odds ratios (OR) and 95% confidence intervals (CI). All statistical analyses were performed utilizing IBM SPSS Version 29.0.1.1, and statistical tests were two-sided with a significance level of 0.05.37

Results

Sample Characteristics and Presenting Symptoms

Among the 643 individuals with a PBT, 24% (n = 153) lived in more disadvantaged neighborhoods, and 39% (n = 253) resided in non-urbanized areas. Overall, the mean age of the cohort was 44.2±15.29, 57.08% were male, 75.89% were Non-Hispanic White, 69.67% had a high-grade PBT, and 71.22% had a bachelor’s or advanced degree (Table 1). Supplementary Figure 1 graphically presents the geographic location of included study participants. Patients who were from more disadvantaged neighborhoods tended to be younger (Hedges’ g = 0.39, P < .001), have low-grade (I/II) tumors (Phi = 0.10, P = .017), underwent more surgeries (Cramer’s V = 0.10, P = .044), had lower incomes (Cramer’s V = 0.18, P < .001), and lower levels of educational attainment (Cramer’s V = 0.19, P < .001) than patients from less disadvantaged neighborhoods (Table 1).

Table 1.

Descriptive Characteristics of Primary Brain Tumor Patients (N = 643).

Overall Cohort (N = 643) Less disadvantaged ADI < 40 (n = 490) More disadvantaged ADI ≥ 40 (n = 153) P-value
N % N % N %
Age at Cancer Diagnosis <.001***
   Mean (SD) 44.20 (15.29) 45.61 (15.45) 39.70 (13.89)
   Range 5.05-79.09 5.05-79.09 13.18-77.93
   Median 43.09 45.56 37.46
Sex .663
   Female 276 42.92 208 42.45 68 44.44
   Male 367 57.08 282 57.55 85 55.56
Race/Ethnicity .080
 Hispanic (all races) 51 7.93 42 8.57 9 5.88
 NH-Asian 36 5.60 32 6.53 4 2.61
 NH-Black 45 7.00 38 7.76 7 4.58
 NH-Pacific Islander, AI/AN, Other, Missing 23 3.58 19 3.88 4 2.61
 NH-White 488 75.89 359 73.27 129 84.31
Income <.001***
   <$50,000 87 13.53 57 11.63 30 19.61
   $50,000-$149,000 147 22.86 106 21.63 41 26.80
   ≥$150,000 112 17.42 102 20.82 10 6.54
   Prefer not to answer 48 7.47 38 7.76 10 6.54
   Missing 249 38.72 187 38.16 62 40.52
Employment Status .183
   Employed 316 49.14 245 50.00 71 46.41
   Retired 112 17.42 92 18.78 20 13.07
   Unemployed 53 8.24 40 8.16 13 8.50
   Unemployed due to tumor diagnosis 155 24.11 108 22.04 47 30.72
   Missing 7 1.09 5 1.02 2 1.31
Education <.001***
   High school or less 81 12.60 55 11.22 26 16.99
   Associate or any college 98 15.24 63 12.86 35 22.88
   Bachelor’s degree 220 34.21 164 33.47 56 36.60
   Advanced degree 238 37.01 204 41.63 34 22.22
   Missing 6 0.93 4 0.82 2 1.31
Marital Status .687
   Unpartnereda 200 31.10 150 30.61 50 32.68
   Partneredb 435 67.65 333 67.96 102 66.67
   Missing 8 1.24 7 1.43 1 0.65
Karnofsky Performance Scale .936
   90-100 430 66.87 330 67.35 100 65.36
   50-80 174 27.06 133 27.14 41 26.80
   Missing 39 6.07 27 5.51 12 7.84
CCI Age Adjusted Score .195
 0 221 34.37 159 32.45 62 40.52
 1 97 15.09 76 15.51 21 13.73
 2+ 299 46.50 237 48.37 62 40.52
 Missing 26 4.04 18 3.67 8 5.23
Tumor Grade .017*
  Low grade (I/II) 154 23.95 106 21.63 48 31.37
  High grade (III/IV) 448 69.67 351 71.63 97 63.40
  No tissue diagnosis 24 3.73 19 3.88 5 3.27
  Not assigned 17 2.64 14 2.86 3 1.96
Received Treatment .140
 No 177 27.53 142 28.98 35 22.88
 Yes 466 72.47 348 71.02 118 77.12
Number of Surgeries .044*
 0 18 2.80 13 2.65 5 3.27
 1 313 48.68 252 51.43 61 39.87
 2+ 312 48.52 225 45.92 87 56.86
Number of Radiation Treatments .418
 0 106 16.49 76 15.51 30 19.61
 1 412 64.07 320 65.31 92 60.13
 2+ 125 19.44 94 19.18 31 20.26
Had Recurrence .063
 No 260 40.44 208 42.45 52 33.99
 Yes 383 59.56 282 57.55 101 66.01
Time with Symptoms Prior to Diagnosis .177
 Less than 6 months 386 60.03 292 59.59 94 61.44
 6 months to 1 year 84 13.06 58 11.84 26 16.99
 More than 1 year 70 10.89 55 11.22 15 9.80
 No symptoms/Don’t know/Missing 103 16.02 85 17.35 18 11.76
Open in a new tab

Notes. Abbreviations: ADI = Area Deprivation Index; SD = Standard Deviation; NH = Non-Hispanic; AI/AN = American Indian/Alaskan Native; CCI = Charlson Comorbidity Index.

*P < .05; **P < .01; ***P < .001.

a“Partnered” participants include those who reported themselves as single, divorced, separated, or widowed.

b“Unpartnered” participants include those who reported themselves as either being married or partnered.

The four most common tumor types among the sample were glioblastoma (n = 234), astrocytoma (n = 134), oligodendroglioma (n = 77), and ependymoma (n = 44), together comprising 76% of the sample. Patients with low- and high-grade tumors predominantly presented at diagnosis with neurologic dysfunction (n = 293), headaches (n = 212), or seizures (n = 210). Individuals living in more disadvantaged neighborhoods presented with more seizures (35% vs. 32%) and less cognitive dysfunction (22% vs. 24%). Patients with low-grade PBTs reported more seizures compared to patients with high-grade PBTs (38% vs. 34%), while those with high-grade PBTs reported greater levels of neurologic dysfunction (51% vs. 43%), headaches (38% vs. 29%), and cognitive dysfunction (27% vs. 16%).

Symptom Burden

Patients living in more disadvantaged neighborhoods reported higher symptom severity (β = 1.10, 95%CI [1.00, 1.20], P = .041) on the MDASI-BT than those living in less disadvantaged neighborhoods, with younger age at diagnosis (P < .004) and worse KPS scores (P < .001) being associated with higher symptom severity. Patients from more disadvantaged neighborhoods were more likely to report higher fatigue (meanmore disadvantaged: 4.06 ± 2.79, meanless disadvantaged: 3.49 ± 3.00, t(df) = −2.10(641), P = .036), pain (meanmore disadvantaged: 2.01 ± 2.80, meanless disadvantaged: 1.44 ± 2.41, t(df) = −2.43(641), P = .015), and drowsiness (meanmore disadvantaged: 3.16 ± 2.81, meanless disadvantaged: 2.50 ± 2.80, t(df) = −2.54(641), P = .011) compared to those from less disadvantaged neighborhoods. Although individuals living in more disadvantaged neighborhoods did not report more symptom interference, they did report more activity-related interference than those living in less disadvantaged neighborhoods (β = 1.11, 95%CI [1.01, 1.22], P = .032) (Table 2). Among patients with low-grade tumors, those who resided in more disadvantaged neighborhoods had more severe affective symptoms (β = 1.24, 95%CI [1.06, 1.45], P = .007), treatment-related symptoms (β = 1.21, 95%CI [1.03, 1.42], P = .021), general disease symptoms (β = 1.26, 95%CI [1.04, 1.51], P = .016), and symptom interference (β = 1.24, 95%CI [1.04, 1.50], P = .020), including both activity-related (β = 1.26, 95%CI [1.07, 1.49], P = .006) and mood-related interference (β = 1.20, 95%CI [1.01, 1.42], P = .039) than patients residing in less disadvantaged neighborhoods (Table 2). Poor functional status was significantly associated with symptom interference for patients with low-grade tumors living in more disadvantaged neighborhoods (P < .001). Among those with high-grade tumors, there were no differences in the MDASI-BT based on neighborhood-level disadvantage (Table 2).

Table 2.

Multivariable Models Assessing the Association Between MDASI-BT and Neighborhood-level Disadvantage, with Sub-analysis for Low- and High-grade Tumor.

Model 1: Unadjusted Model 3: Adjusted for age, sex, race/ethnicity, tumor grade (low vs high), KPS, and recurrence a
Patient characteristics β 95% CI P-value β 95% CI P-value
MDASI-BT Symptom Severity Overall Sample (n = 595) (n = 524)
Constant 1.04 (−1.00, 1.09) .059 1.49 (1.26, 1.77) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.12 (1.03, 1.22) .009** 1.10 (1.00, 1.20) .041*
Adj R2 = .010, F(1,593) = 6.96; P = .009 Adj R2 = .125, F(10,513) = 8.48; P < .001
Low-grade tumor b (n = 141) (n = 133)
Constant -.99 (-1.11, 1.08) .761 1.35 (-1.04, 1.90) .086
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.21 (1.03, 1.42) .019* 1.18 (-1.00, 1.39) .053
Adj R2 = .032, F(1,139) = 5.60; P = .019 Adj R2 = .115, F(9,123) = 2.90; P = .004
High-grade tumor b (n = 417) (n = 391)
Constant 1.06 (1.01, 1.12) .015 1.59 (1.31, 1.92) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.08 (-1.02, 1.21) .130 1.05 (-1.06, 1.17) .346
Adj R2 = .003, F(1,415) = 2.30; P = .130 Adj R2 = .122, F(9,381) = 7.02; P < .001
MDASI-BT Symptom Interference Overall Sample (n = 510) (n = 451)
Constant 1.30 (1.24, 1.37) <.001 1.52 (1.24, 1.86) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.05 (-1.05, 1.16) .307 1.07 (-1.04, 1.19) .213
Adj R2 = .000, F(1,508) = 1.04; P = .307 Adj R2 = .069, F(10,440) = 4.35; P < .001
Low-grade tumor b (n = 119) (n = 112)
Constant 1.24 (1.12, 1.37) <.001 1.29 (-1.14, 1.89) .191
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.23 (1.03, 1.47) .025 1.24 (1.04, 1.50) .020*
Adj R2 = .034, F(1,117) = 0.042; P = .025 Adj R2 = .143, F(9,102) = 3.06; P = .003
High-grade tumor b (n = 362) (n = 339)
Constant 1.32 (1.24, 1.40) <.001 1.59 (1.27, 2.00) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 -1.00 (-1.14, 1.13) .945 -1.00 (-1.14, 1.13) .970
Adj R2 = -.003, F(1,360) = 0.005; P = .945 Adj R2 = .054, F(9,329) = 3.16; P = .001
MDASI-BT Affective Overall Sample (n = 569) (n = 500)
Constant 1.33 (1.28, 1.38) <.001 1.72 (1.46, 2.02) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.09 (1.00, 1.17) .043* 1.07 (-1.01, 1.17) .096
Adj R2 = .005, F(1,567) = 4.12; P = .043 Adj R2 = .057, F(10,489) = 4.03; P < .001
Low-grade tumor b (n = 136) (n = 128)
Constant 1.24 (1.14, 1.35) <.001 1.56 (1.12, 2.17) .009
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.26 (1.08, 1.47) .003** 1.24 (1.06, 1.45) .007**
Adj R2 = .057, F(1,134) = 9.21; P = .003 Adj R2 = .076, F(9,118) = 2.16; P = .030
High-grade tumor b (n = 398) (n = 372)
Constant 1.34 (1.28, 1.41) <.001 1.85 (1.54, 2.21) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.02 (-1.08, 1.13) .625 1.00 (-1.10, 1.11) .952
Adj R2 = -.002, F(1,396) = 0.24; P = .625 Adj R2 = .063, F(9,362) = 3.77; P < .001
MDASI-BT Cognitive Overall Sample (n = 485) (n = 426)
Constant 1.21 (1.16, 1.27) <.001 1.62 (1.34, 1.95) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.05 (-1.04, 1.15) .300 1.03 (-1.06, 1.14) .489
Adj R2 = .000, F(1,483) = 1.070; P = .300 Adj R2 = .080, F(10,415) = 4.70; P < .001
Low-grade tumor b (n = 106) (n = 99)
Constant 1.18 (1.06, 1.31) .003 1.30 (-1.18, 1.99) .229
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.11 (-1.09, 1.34) .272 1.16 (-1.05, 1.42) .135
Adj R2 = .002, F(1,104) = 1.22; P = .272 Adj R2 = .074, F(9,89) = 1.87; P = .067
High-grade tumor b (n = 350) (n = 327)
Constant 1.24 (1.17, 1.30) <.001 1.74 (1.42, 2.13) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.03 (-1.08, 1.15) .573 -.99 (-1.13, 1.11) .907
Adj R2 = -.002, F(1,348) = .32; P = .573 Adj R2 = .086, F(9,317) = 4.41; P < .001
MDASI-BT Neurologic Overall Sample (n = 417) (n = 371)
Constant 1.08 (1.03, 1.13) .003 1.46 (1.20, 1.76) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.06 (-1.04, 1.17) .221 1.04 (-1.06, 1.15) .421
Adj R2 = .001, F(1,415) = 1.50; P = .221 Adj R2 = .104, F(10,360) = 5.32; P < .001
Low-grade tumor b (n = 101) (n = 95)
Constant 1.06 (-1.05, 1.17) .279 1.18 (-1.22, 1.69) .376
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.06 (-1.12, 1.26) .483 1.04 (-1.14, 1.23) .661
Adj R2 = -.005, F(1,99) = 0.50; P = .483 Adj R2 = .107, F(9,85) = 2.26; P = .026
High-grade tumor b (n = 296) (n = 276)
Constant 1.08 (1.02, 1.14) .009 1.56 (1.26, 1.94) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.06 (-1.06, 1.19) .331 1.04 (-1.09, 1.17) .534
Adj R2 = .000, F(1,294) = .95; P = .331 Adj R2 = .119, F(9,266) = 5.14; P < .001
MDASI-BT Treatment-related Overall Sample (n = 495) (n = 436)
Constant 1.17 (1.13, 1.23) <.001 1.46 (1.22, 1.74) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.06 (-1.02, 1.15) .152 1.07 (-1.02, 1.16) .149
Adj R2 = .002, F(1,493) = 2.06; P = .152 Adj R2 = .068, F(10,425) = 4.16; P < .001
Low-grade tumor b (n = 116) (n = 109)
Constant 1.12 (1.02, 1.23) .014 1.33 (-1.07, 1.88) .108
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.19 (1.02, 1.40) .028* 1.21 (1.03, 1.42) .021*
Adj R2 = .033, F(1,114) = 4.97; P = .028 Adj R2 = .143, F(9,99) = 3.01; P = .003
High-grade tumor b (n = 351) (n = 327)
Constant 1.20 (1.14, 1.26) <.001 1.55 (1.28, 1.88) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.02 (-1.09, 1.12) .760 1.01 (-1.10, 1.12) .862
Adj R2 = -.003, F(1,349) = .09; P = .760 Adj R2 = .065, F(9,317) = 3.52; P < .001
MDASI-BT General Disease Overall Sample (n = 435) (n = 382)
Constant 1.06 (1.01, 1.11) .021 1.26 (1.04, 1.53) .018
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.06 (-1.04, 1.16) .228 1.04 (-1.05, 1.15) .377
Adj R2 = .001, F(1,433) = 1.46; P = .228 Adj R2 = .072, F(10,371) = 3.96; P < .001
Low-grade tumor b (n = 96) (n = 90)
Constant 1.01 (-1.10, 1.12) .905 1.19 (-1.22, 1.74) .350
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.27 (1.07, 1.53) .009** 1.26 (1.04, 1.51) .016*
Adj R2 = .061, F(1,94) = 7.20; P = .009 Adj R2 = .118, F(9,80) = 2.32; P = .022
High-grade tumor b (n = 314) (n = 292)
Constant 1.07 (1.02, 1.13) .727 1.30 (1.05, 1.60) .014
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 -.98 (-1.14, 1.09) .727 -0.96 (-1.16, 1.08) .533
Adj R2 = -.003, F(1,312) = .12; P = .727 Adj R2 = .084, F(9,282) = 3.95; P < .001
MDASI-BT Gastrointestinal Overall Sample (n = 187) (n = 164)
Constant 1.24 (1.16, 1.33) <.001 1.56 (1.17, 2.07) .003
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.00 (-1.14, 1.14) .974 -0.98 (-1.17, 1.13) .809
Adj R2 = -.005, F(1,185) = .00; P = .974 Adj R2 = .089, F(10,153) = 2.60; P = .006
Low-grade tumor b (n = 46) (n = 44)
Constant 1.20 (1.02, 1.42) .030 1.85 (-1.19, 4.08) .123
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.02 (-1.25, 1.32) .848 -.98 (-1.37, 1.31) .888
Adj R2 = -.022, F(1,44) = .04; P = .848 Adj R2 = -.122, F(9,34) = .48; P = .877
High-grade tumor b (n = 129) (n = 120)
Constant 1.27 (1.17, 1.26) <.001 1.60 (1.20, 2.15) .002
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 -0.97 (-1.22, 1.15) .737 -0.94 (-1.25, 1.11) .470
Adj R2 = -.007, F(1,127) = 0\.11; P = .737 Adj R2 = .187, F(9,110) = 4.04; P < .001
MDASI-BT Activity-related Interference Overall Sample (n = 450) (n = 399)
Constant 1.47 (1.41, 1.54) <.001 1.53 (1.28, 1.85) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.08 (-1.02, 1.18) .117 1.11 (1.01, 1.22) .032*
Adj R2 = .003, F(1,448) = 2.46; P = .117 Adj R2 = .089, F(10,388) = 4.87; P < .001
Low-grade tumor b (n = 106) (n = 99)
Constant 1.40 (1.27, 1.53) <.001 1.42 (1.01, 2.00) .046
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.22 (1.03, 1.44) .020* 1.26 (1.07, 1.49) .006**
Adj R2 = 0.042, F(1,104) = 5.56; P = .020 Adj R2 = .199, F(9,89) = 3.71; P < .001
High-grade tumor b (n = 320) (n = 300)
Constant 1.49 (1.42, 1.57) <.001 1.59 (1.30, 1.96) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.04 (-1.08, 1.16) .510 1.05 (-1.07, 1.18) .394
Adj R2 = -.002, F(1,318) = .43; P = .510 Adj R2 = .065, F(9,290) = 3.32; P < .001
MDASI-BT Mood-related Interference Overall Sample (n = 484) (n = 427)
Constant 1.33 (1.27, 1.39) <.001 1.47 (1.21, 1.78) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.05 (-1.04, 1.15) .276 1.06 (-1.04, 1.17) .259
Adj R2 = .000, F(1,482) = 1.19; P = .276 Adj R2 = .021, F(10,416) = 1.90; P = .044
Low-grade tumor b (n = 115) (n = 108)
Constant 1.25 (1.14, 1.38) <.001 1.43 (-1.03, 2.11) .069
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.20 (1.02, 1.41) .032* 1.20 (1.01, 1.42) .039*
Adj R2 = .031, F(1,113) = 4.70; P = .032 Adj R2 = .074, F(9,98) = 1.95; P = .053
High-grade tumor b (n = 340) (n = 319)
Constant 1.34 (1.28, 1.42) <.001 1.51 (1.23, 1.87) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 -1.00 (-1.13, 1.12) .955 -1.00 (-1.13, 1.13) .999
Adj R2 = -.003, F(1,338) = .00; P = .955 Adj R2 = .007, F(9,309) = 1.25; P = .262
Open in a new tab

Notes. Abbreviations: CI = Confidence interval; REF = References; SE = Standard Error; Adj = Adjusted; KPS = Karnofsky Performance Score; ADI = Area Deprivation Index.

*P < .05; **P < .01; ***P < .001. Bolded values indicate significant differences between the more disadvantaged and less disadvantaged groups in unadjusted and adjusted models.

aModels were adjusted for age at diagnosis, sex, race/ethnicity, tumor grade, KPS, and tumor recurrence.

bSub-analysis comparing low- and high-grade.

Patients with high-grade PBTs living in non-urbanized areas experienced greater neurologic symptoms than those in urbanized areas (β = 0.81, 95%CI [0.57, 1.11], P = .034) (Supplementary Table 1). Younger age at diagnosis, poor functional status, and being non-Hispanic Asian compared with non-Hispanic White were significantly associated with more neurologic symptoms among those with high-grade tumors in non-urbanized areas (P = .003, P < .001, and P = .026, respectively). No other differences in the MDASI-BT were found based on population density among the overall cohort or by tumor grade.

Anxiety and Depression

There were no differences in anxiety or depression based on neighborhood-level disadvantage or population density among the overall sample or those with high-grade tumors (Table 3 and  Supplementary Table 2). Patients with low-grade tumors from more disadvantaged neighborhoods reported higher levels of anxiety (β = 1.03, 95%CI [1.01, 1.06], P = .023) than those from less disadvantaged neighborhoods, but not depression (Table 3). Younger age at diagnosis (P = .035) and Hispanic ethnicity compared to non-Hispanic White (P = .044) were significantly associated with anxiety.

Table 3.

Multivariable Models Assessing the Association Between PROMIS-Anxiety and -Depression and Neighborhood-level Disadvantage, with Sub-analysis for Low- and High-grade Tumor.

Model 1: Unadjusted Model 3: Adjusted for age, sex, race/ethnicity, tumor grade (low vs high), KPS, and recurrence a
Patient Characteristics β 95% CI P-value β 95% CI P-value
PROMIS-Anxiety Overall Sample (N = 643) (n = 567)
Constant 5.47 (5.42, 5.51) <.001 5.75 (5.58, 5.94) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.01 (-1.00, 1.03) .163 1.01 (-1.01, 1.02) .359
Adj R2 = .001, F(1,641) = 1.95; P = .163 Adj R2 = .036, F(10,556) = 3.12; P < .001
Low-grade tumor b (n = 154) (n = 146)
Constant 5.47 (5.39, 1.19) <.001 5.85 (5.50, 6.22) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 1.03 (1.00, 1.06) .036* 1.03 (1.01, 1.06) .023*
Adj R2 = .022, F(1,152) = 4.50; P = .036 Adj R2 = .072, F(9,136) = 2.25; P = .023
High-grade tumor b (n = 448) (n = 421)
Constant 5.46 (5.41, 5.51) <.001 5.67 (5.47, 5.87) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 -1.00 (-1.02, 1.02) .954 -1.00 (-1.02, 1.02) .950
Adj R2 = -.002, F(1,446) = .00; P = .954 Adj R2 = .027, F(9,411) = 2.28; P = .017
PROMIS-Depression Overall Sample (N = 643) (n = 567)
Constant 49.78 (48.98, 50.58) <.001 56.56 (53.16, 59.97) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 .93 (-.71, 2.57) .267 .48 (-1.25, 2.21) .586
Adj R2 = .000, F(1,641) = 1.23; P = .267 Adj R2 = .059, F(10,556) = 4.55; P < .001
Low-grade tumor b (n = 154) (n = 146)
Constant 50.09 (48.28, 51.91) <.001 55.47 (48.35, 62.58) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 2.03 (-1.22, 5.28) .220 1.67 (-1.65, 4.99) .322
Adj R2 = .003, F(1,152) = 1.52; P = .220 Adj R2 = .064, F(9,136) = 2.10; P = .034
High-grade tumor b (n = 448) (n = 421)
Constant 49.65 (48.73, 50.58) <.001 55.51 (51.78, 59.25) <.001
ADI
Less disadvantaged, ADI < 40 1.00 (REF) -- 1.00 (REF) --
More disadvantaged, ADI ≥ 40 .36 (-1.63, 2.35) .720 .03 (-2.04, 2.09) .981
Adj R2 = -.002, F(1,446) = 0.13; P = .720 Adj R2 = .046, F(9,411) = 3.25; P < .001
Open in a new tab

Notes. Abbreviations: CI = Confidence interval; REF = References; SE = Standard Error; Adj = Adjusted; KPS = Karnofsky Performance Score; ADI = Area Deprivation Index.

*P < 0.05; **P < 0.01; ***P < 0.001. Bolded values indicate significant differences between the more disadvantaged and less disadvantaged groups in unadjusted and adjusted models.

aModels were adjusted for age at diagnosis, sex, race/ethnicity, tumor grade, KPS, and tumor recurrence.

bSub-analysis comparing low- and high-grade.

Health-Related Quality of Life

No significant associations were observed between the 5 dimensions of the EQ-5D-3L and neighborhood-level disadvantage, overall or among those with high-grade tumors (Figure 2a–e). However, those with low-grade tumors living in more disadvantaged neighborhoods had 3.30 times the odds of experiencing moderate-severe difficulty with mobility (95%CI [1.09, 10.01]) and 4.68 times the odds of experiencing moderate-severe difficulty completing self-care tasks (95%CI [1.46, 14.96]) than those in less disadvantaged neighborhoods. Poor functional status was significantly associated with difficulties with mobility (P < .001) and self-care (P < .005).

Figure 2.

Figure 2.

Open in a new tab

Forest plots illustrating the relationship between neighborhood-level disadvantage and HRQOL dimensions measured by the EQ-5D-3L: (A) Mobility, (B) Self-care, (C) Usual activities, (D) Pain/Discomfort, and (E) Anxiety/Depression. Each plot depicts odds ratios among patients with primary brain tumors, overall and by low- and high-grade tumor subgroups. Patients with low-grade tumors living in more disadvantaged neighborhoods were more likely to experience difficulties with mobility and self-care than those in less disadvantaged neighborhoods. Abbreviations: OR = Odds Ratio, CI = Confidence Interval. Models were adjusted for age, sex, race/ethnicity, tumor grade, tumor recurrence, and functional status. *Indicates significant values.

Patients residing in non-urbanized areas were found to have 2.07 times the odds of having moderate-severe difficulty completing self-care tasks (95%CI [1.23,3.48]) compared to those in more urbanized areas (Supplementary Figure 2a–e). Similarly, patients with high-grade tumors living in non-urbanized areas reported 2.08 times the odds of having moderate-severe difficulty completing self-care tasks (95%CI [1.14,3.78]) than those in urbanized areas (Supplementary Figure 2a–e). Older age at diagnosis and poor functional status were significantly associated with difficulties with self-care among patients living in non-urbanized areas, both overall (P = .007 and P < .001, respectively) and for those with high-grade PBTs (P = .018 and P < .001, respectively).

Discussion

We found that among a cohort of 643 individuals with a PBT, those living in more disadvantaged neighborhoods reported more severe symptoms than those living in less disadvantaged neighborhoods (P = .041). While they did not report higher levels of overall symptom interference, those from more disadvantaged neighborhoods did report more activity-related interference (P = .032) than those in less disadvantaged neighborhoods. To further explore which symptoms were worse among those living in more disadvantaged neighborhoods and to limit the risk of overinterpretation, we identified pain, fatigue, and drowsiness in bivariate analyses as the more severe symptoms reported among those living in more disadvantaged neighborhoods. These symptoms have been previously documented as among the most prevalent and severe symptoms experienced by patients with PBTs across tumor grades.4 Pain, in particular, is an important concern for patients’ well-being and often presents as headaches in patients with PBTs.5,38 Prior research has shown that socioeconomic disadvantage is associated with more severe pain and inadequate pain management in individuals with chronic pain.39 Cancer-related fatigue is associated with worse HRQOL among patients with PBTs, and poor functional status is an important risk factor for cancer-related fatigue.40 The severity of these key symptoms may shape the functional abilities of patients with PBTs, and those living in more socioeconomically disadvantaged neighborhoods may experience these symptoms as heightened, potentially due to a lack of access to specialized care and an inability to take time off from work for treatment.41

Special attention and additional analyses to validate our findings in relation to symptoms—especially for pain, fatigue, and drowsiness—is warranted as they may serve as targets for interventions aiming to mitigate health disparities. Specifically, more longitudinal and interventional research is needed to describe the effect of socioeconomic-related health disparities on PBT-related symptoms. Neuro-oncology researchers and clinicians can use successful recruitment strategies from other cancer populations to ensure not only the recruitment of historically marginalized groups to research studies but also their inclusion in the development and implementation of supportive resources in more disadvantaged communities. Furthermore, including relevant socioeconomic disparity measures like ADI, income, or financial toxicity when measuring symptoms will be necessary to understand the impact of interventions among populations from socioeconomically disadvantaged neighborhoods.

Most differences by neighborhood disadvantage were observed in stratified analyses by tumor grade. Previous work led by our group found that the time to treatment was longer for patients with low-grade tumors living in more disadvantaged areas.42 In our current analysis, we found that patients with low-grade PBTs living in more disadvantaged neighborhoods faced issues with more symptom burden, anxiety, and symptom interference. Among our sample, patients with low-grade PBTs were younger at diagnosis than those with high-grade PBTs (mean±SD: 40.91±14.01 vs. 45.52±15.87), and patients diagnosed with low-grade PBTs living in more disadvantaged areas had the youngest age at diagnosis (Mean±SD: 38.82±12.38). Greater symptom severity and mood-related interference may have been found among those with low-grade tumors living in more disadvantaged neighborhoods because those with low-grade PBTs experienced longer time to treatment, and those with low-grade PBTs are more likely to be diagnosed in younger individuals who face a longer disease duration compared to those with high-grade PBTs.2,43 Additionally, younger age and Hispanic compared to non-Hispanic White ethnicity were associated with anxiety. This observation builds upon previous literature documenting the high prevalence of anxiety among Hispanic patients with cancer,44 pointing towards the need for further investigation into the differential impact of anxiety among this population.

Patients with low-grade PBTs from more disadvantaged neighborhoods experienced more difficulties with mobility and self-care than those from less disadvantaged neighborhoods. As has been previously reported, patients with low-grade PBTs reported more seizures.42,43,45 Frequent epileptic seizures are commonly associated with impairments in memory and mobility, which may have life-altering implications, especially for younger, employed adults with a PBT, as these symptoms impact their careers, daily activities, and ability to support a family.46,47 Furthermore, patients with low-grade PBTs may experience a decline in HRQOL due to their prolonged survival time and need to navigate the ongoing burden of various treatments and potential treatment late effects.43 Individuals living in more disadvantaged neighborhoods and diagnosed with a low-grade PBT may have less supportive resources and more time for their symptoms, both physical and affective, to cause additional interference in their day-to-day lives. We did not observe similar results among patients with high-grade PBTs based on ADI, potentially due to late disease presentation upon their enrollment in the NOB-NHS and limited treatment efficacy, which may be similar regardless of ADI scores.

Residence in non-urbanized areas, compared to urbanized areas, was associated with worse HRQOL (i e, self-care) among the overall sample and those with high-grade tumors, and with more neurologic symptoms among those with high-grade tumors. Poor functional status and older age at diagnosis were associated with difficulties in self-care for patients living in non-urbanized areas, possibly because these patients experience greater physical limitations,48 particularly among older individuals with cancer who may report worse HRQOL. Difficulties in caring for oneself may be exacerbated by the fact that residents of rural areas are more likely to be older, work physically demanding jobs, engage in health-risk lifestyle behaviors, and have a higher incidence of non-cancer comorbidities.39,49 Furthermore, greater neurologic symptoms among those with high-grade PBTs were associated with poor functional status, younger age at diagnosis, and identification as non-Hispanic Asian. Previous reports indicate that patients with glioblastoma living in urbanized areas are more likely to undergo surgical treatment than those in nonmetropolitan regions, with younger patients receiving more complete tumor removals due to better functional status at presentation.50 Worse neurologic symptoms among patients with high-grade PBTs in non-urbanized areas may be related to delayed detection, diagnosis, and intervention or treatment limitations caused by poor functional status. However, the relationship of population density with PROs is complex and multidimensional; therefore, further consideration is needed for factors beyond our analysis, including social support and access to symptom management medication.

Limitations

Limitations of this study include the cross-sectional, observational nature of our data, which precludes causal inferences. The generalizability of our findings may be limited as our sample was predominantly non-Hispanic White, and 71.22% had either a bachelor’s or advanced degree, which is higher than the general U.S. population. Therefore, the effects in our sample may have been underestimated compared to the general population. In the future, it will be essential to pursue strategies to address the underrepresentation of racial and ethnic groups in research studies, including Hispanics and NH-Asians with PBTs. Additionally, our sample consisted of patients enrolled across the disease trajectory (ie, either at diagnosis, during treatment, or during surveillance) with heterogeneous malignant and non-malignant PBT types. To address this limitation, we conducted stratified analyses of all outcome variables by tumor grade and adjusted our models for KPS and recurrence status to further account for disease stage. Our sample may not be representative of the general PBT population, as our participants consisted of individuals who were enrolled in an observational research study, while previous studies indicate less than 30% of patients with PBTs are referred to participate in therapeutic clinical trials.51 Although travel and lodging support was provided, participants in our sample may have had more resources to allow their access to care at specialized Neuro-Oncology centers and the NCI, while also having been stable enough to travel. Future supportive interventions should focus on increasing clinical trial referrals and developing survivorship programs for patients with low-grade tumors and those residing in more disadvantaged neighborhoods and non-urbanized areas through strategies such as the implementation of neighborhood-level cancer care resources.

Finally, while ADI offers a composite measure of neighborhood-level socioeconomic status, it may not fully capture individual-level factors, such as stress which can affect outcomes, or personal income-which we were not able to include in our analysis due to a high percentage of missing data related to personal income. Inclusion of other measures of socioeconomic status may be considered, as previous studies have found that the ADI as a measure of neighborhood-level disadvantage status may underestimate deprivation levels due to an overestimation of the influence of markers of income and housing.52 This is particularly true in communities experiencing gentrification in areas like New York City or the District of Columbia.52 Further analysis is needed to confirm our findings and explore the effects of individual and neighborhood-level disadvantage throughout different community types.

Conclusion

For individuals diagnosed with a PBT, living in more disadvantaged neighborhoods and non-urbanized areas was found to negatively impact symptoms and HRQOL, especially among those with low-grade PBTs. Those with PBTs residing in less disadvantaged neighborhoods may experience better access to treatment and supportive services, contributing to fewer or better-managed patient symptoms. Understanding the impact of socioeconomic and geographic characteristics on not only survival, but also symptoms and HRQOL can help reveal interventional targets to mitigate disparities in patient outcomes. Future research should further track symptoms, including symptom severity, mood-related interference, as well as mobility and self-care domains of HRQOL among those with low-grade PBTs, to confirm our findings. Working towards earlier intervention and follow-up for those with low-grade PBTs living in more disadvantaged neighborhoods may help to mitigate some of the long-term effects of a PBT diagnosis.

Supplementary material

Supplementary material is available online at Neuro-Oncology Practice (https://academic.oup.com/nop/).

npaf069_Supplementary_Figures_1-2
npaf069_supplementary_figures_1-2.docx (284.3KB, docx)
npaf069_Supplementary_Tables_1-2
npaf069_supplementary_tables_1-2.docx (82.9KB, docx)

Acknowledgments

We would like to thank the patients who have participated in the Natural History Study and the National Cancer Institute Neuro-Oncology Branch clinicians for making this work possible. The funder did not play a role in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; and the decision to submit the manuscript for publication.

Contributor Information

Zuena Karim, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Kimberly Robins, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Orieta Celiku, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Yeonju Kim, School of Medicine, Wayne State University, Detroit, MI, USA.

Hope Miller, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Elizabeth Vera, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Jacqueline B Vo, Radiation Epidemiology Branch, Division of Cancer Epidemiology & Genetics, National Cancer Institute, National Institutes of Health, MD, USA.

Mark R Gilbert, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Terri S Armstrong, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Macy L Stockdill, Neuro-Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.

Funding

This research was supported in part by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, and by a National Institutes of Health Center for Cancer Research Health Disparities Award. Macy L. Stockdill, PhD, was a post-doctoral fellow supported by the National Cancer Institute’s Intramural Continuing Umbrella for Research Experiences (iCURE) program. Macy Stockdill is supported in part by the National Cancer Institute of the National Institutes of Health under Award Number U54CA280770 and U54CA118948. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The Natural History Study project is supported by Intramural Project [1ZIABC011768-03 to T.S.A.]

Ethics Approval

This study was approved by the Institutional Review Board, and informed consent was obtained from all individual participants included in the study.

Conflict of interest statement. The authors have no relevant financial or non-financial interests to disclose.

Author Contributions

Study Contribution and design: MLS, JV, OC, YK, ZK, EV, KR, HM, MG, TA; Data Collection: MLS, JV, OC, YK, ZK, EV, MG, TA; Analysis and Interpretation of Results: MLS, JV, OC, YK, ZK, EV, KR, HM, MG, TA; Draft manuscript preparation and editing: MLS, JV, OC, YK, ZK, EV, KR, HM, MG, TA.

Data Availability

The data underlying this article is available upon request due to privacy and ethical restrictions.

References

  • 1. Schaff  LR, Mellinghoff  IK.  Glioblastoma and other primary brain malignancies in adults: a review. JAMA.  2023;329(7):574–587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Ostrom  QT, Price  M, Neff  C, et al.  CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2016-2020. Neuro Oncol. 2023;25(12 Suppl 2):iv1–iv99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Louis  DN, Perry  A, Wesseling  P, et al.  The 2021 WHO classification of tumors of the central nervous system: a summary. Neuro Oncol. 2021;23(8):1231–1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Armstrong  TS, Vera-Bolanos  E, Acquaye  AA, et al.  The symptom burden of primary brain tumors: evidence for a core set of tumor- and treatment-related symptoms. Neuro Oncol. 2016;18(2):252–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Liu  R, Page  M, Solheim  K, Fox  S, Chang  SM.  Quality of life in adults with brain tumors: current knowledge and future directions. Neuro Oncol. 2009;11(3):330–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Mukherjee  D, Zaidi  HA, Kosztowski  T, et al.  Disparities in access to neuro-oncologic care in the United States. Arch Surg.  2010;145(3):247–253. [DOI] [PubMed] [Google Scholar]
  • 7. Perkins  A, Liu  G.  Primary brain tumors in adults: diagnosis and treatment. Am Fam Physician.  2016;93(3):211–217. [PubMed] [Google Scholar]
  • 8. Osoba  D, Brada  M, Prados  MD, Yung  WK.  Effect of disease burden on health-related quality of life in patients with malignant gliomas. Neuro Oncol. 2000;2(4):221–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Chaitoff  A, Sarmey  N, Thompson  NR, et al.  Quality of life outcomes in patients presenting for evaluation of CNS tumors. Neurol Clin Pract. 2019;9(1):32–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Gustafsson  M, Edvardsson  T, Ahlström  G.  The relationship between function, quality of life and coping in patients with low-grade gliomas. Support Care Cancer.  2006;14(12):1205–1212. [DOI] [PubMed] [Google Scholar]
  • 11. Pelletier  G, Verhoef  MJ, Khatri  N, Hagen  N.  Quality of life in brain tumor patients: the relative contributions of depression, fatigue, emotional distress, and existential issues. J Neurooncol.  2002;57(1):41–49. [DOI] [PubMed] [Google Scholar]
  • 12. Pertz  M, Schlegel  U, Thoma  P.  Sociocognitive functioning and psychosocial burden in patients with brain tumors. Cancers (Basel). 2022;14(3):767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Rosenzweig  MQ, Althouse  AD, Sabik  L, et al.  The association between area deprivation index and patient-reported outcomes in patients with advanced cancer. Health Equity. 2021;5(1):8–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Singh  GK.  Area deprivation and widening inequalities in US mortality, 1969-1998. Am J Public Health.  2003;93(7):1137–1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Unger  JM, Moseley  AB, Cheung  CK, et al.  Persistent disparity: socioeconomic deprivation and cancer outcomes in patients treated in clinical trials. J Clin Oncol.  2021;39(12):1339–1348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. University of Wisconsin School of Medicine and Public Health (2018) Area Deprivation Index. https://www.neighborhoodatlas.medicine.wisc.edu/. Accessed 2024. [Google Scholar]
  • 17. Li  CC, Matthews  AK, Asthana  A, Shah  RC.  The impact of neighborhood disadvantage on health-related quality of life among African American and White cancer survivors. Transl Cancer Res. 2019;8(Suppl 4):S313–S322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Hassan  AM, Nguyen  HT, Corkum  JP, et al.  Area deprivation index is associated with variation in quality of life and psychosocial well-being following breast cancer surgery. Ann Surg Oncol.  2023;30(1):80–87. [DOI] [PubMed] [Google Scholar]
  • 19. Bai  J, Pugh  SL, Eldridge  R, et al.  Neighborhood deprivation and rurality associated with patient-reported outcomes and survival in men with prostate cancer in NRG oncology RTOG 0415. Int J Radiat Oncol Biol Phys.  2023;116(1):39–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Rivera Perla  KM, Tang  OY, Durfey  SNM, et al.  Predicting access to postoperative treatment after glioblastoma resection: an analysis of neighborhood-level disadvantage using the Area Deprivation Index (ADI). J Neurooncol.  2022;158(3):349–357. [DOI] [PubMed] [Google Scholar]
  • 21.CancerDisparitiesProgressReport.org [Internet]. Philadelphia: American Association for Cancer Research; ©2024. http://www.CancerDisparitiesProgressReport.org/. Accessed 2024. [Google Scholar]
  • 22. Cyr  ME, Etchin  AG, Guthrie  BJ, Benneyan  JC.  Access to specialty healthcare in urban versus rural US populations: a systematic literature review. BMC Health Serv Res.  2019;19(1):974. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Levit  LA, Byatt  L, Lyss  AP, et al.  Closing the rural cancer care gap: three institutional approaches. JCO Oncol Pract. 2020;16(7):422–430. [DOI] [PubMed] [Google Scholar]
  • 24. Caston  NE, Williams  CP, Wan  C, et al.  Associations between geography, decision-making style, and interest in cancer clinical trial participation. Cancer.  2022;128(22):3977–3984. [DOI] [PubMed] [Google Scholar]
  • 25. Meilleur  A, Subramanian  SV, Plascak  JJ, et al.  Rural residence and cancer outcomes in the United States: issues and challenges. Cancer Epidemiol Biomarkers Prev.  2013;22(10):1657–1667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Cote  DJ, Ostrom  QT, Gittleman  H, et al.  Glioma incidence and survival variations by county-level socioeconomic measures. Cancer.  2019;125(19):3390–3400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Curry  WT, Jr, Barker  FG.  ) Racial, ethnic and socioeconomic disparities in the treatment of brain tumors. J Neurooncol.  2009;93(1):25–39. [DOI] [PubMed] [Google Scholar]
  • 28. Xu  H, Chen  J, Xu  H, Qin  Z.  Geographic variations in the incidence of glioblastoma and prognostic factors predictive of overall survival in US adults from 2004-2013. Front Aging Neurosci.  2017;9:352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Kind  AJH, Buckingham  WR.  Making neighborhood-disadvantage metrics accessible - the neighborhood atlas. N Engl J Med.  2018;378(26):2456–2458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Hu  J, Kind  AJH, Nerenz  D.  Area deprivation index predicts readmission risk at an urban teaching hospital. Am J Med Qual.  2018;33(5):493–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. U.S. Census Bureau. Redefining Urban Areas follow the 2020 Census. https://www.census.gov/newsroom/blogs/random-samplings/2022/12/redefining-urban-areas-following-2020-census.html. Accessed 2024. [Google Scholar]
  • 32. Cleeland  CS. The M. D. Anderson Symptom Inventory User Guide Version 1. 2009; https://www.mdanderson.org/content/dam/mdanderson/documents/Departments-and-Divisions/Symptom-Research/MDASI_userguide.pdf [Google Scholar]
  • 33. EuroQol Research Foundation. EQ-5D-3L User Guide. 2018. https://euroqol.org/publications/user-guides [Google Scholar]
  • 34. U. S. Department of Health and Human Services. Patient-Reported Outcomes Measurement Information System. 2023. https://www.healthmeasures.net/score-and-interpret/interpret-scores/promis. Accessed 2024. [Google Scholar]
  • 35. Mor  V, Laliberte  L, Morris  JN, Wiemann  M.  The Karnofsky Performance Status Scale. An examination of its reliability and validity in a research setting. Cancer.  1984;53(9):2002–2007. [DOI] [PubMed] [Google Scholar]
  • 36. Armstrong  TS, Mendoza  T, Gning  I, et al.  Validation of the M.D. anderson symptom inventory brain tumor module (MDASI-BT). J Neurooncol.  2006;80(1):27–35. [DOI] [PubMed] [Google Scholar]
  • 37. IBM Corp. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.; Released 2012. 2012.
  • 38. Goffaux  P, Fortin  D.  Brain tumor headaches: from bedside to bench. Neurosurgery.  2010;67(2):459–466. [DOI] [PubMed] [Google Scholar]
  • 39. Prego-Domínguez  J, Khazaeipour  Z, Mallah  N, Takkouche  B.  Socioeconomic status and occurrence of chronic pain: a meta-analysis. Rheumatology (Oxford). 2021;60(3):1091–1105. [DOI] [PubMed] [Google Scholar]
  • 40. Asher  A, Fu  JB, Bailey  C, Hughes  JK.  Fatigue among patients with brain tumors. CNS Oncol. 2016;5(2):91–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Blinder  VS, Gany  FM.  Impact of cancer on employment. J Clin Oncol.  2020;38(4):302–309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Stockdill  ML, Vo  JB, Celiku  O, et al.  Neighborhood disadvantage is associated with treatment access outcomes and survival among individuals with a primary brain tumor. Neurooncol. Pract..  2024;12(2):313–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Forst  DA, Nahed  BV, Loeffler  JS, Batchelor  TT.  Low-grade gliomas. Oncologist. 2014;19(4):403–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Lee  MS, Tyson  DM, Gonzalez  BD, et al.  Anxiety and depression in Spanish-speaking Latina cancer patients prior to starting chemotherapy. Psychooncology.  2018;27(1):333–338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Vecht  CJ, Kerkhof  M, Duran-Pena  A.  Seizure prognosis in brain tumors: new insights and evidence-based management. Oncologist.  2014;19(7):751–759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Krajewski  S, Wójcik  M, Harat  M, Furtak  J.  Influence of epilepsy on the quality of life of patients with brain tumors. Int J Environ Res Public Health.  2021;18(12):6390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Rimmer  B, Bolnykh  I, Dutton  L, et al.  Health-related quality of life in adults with low-grade gliomas: a systematic review. Qual Life Res.  2023;32(3):625–651. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Mackworth  N, Fobair  P, Prados  MD.  Quality of life self-reports from 200 brain tumor patients: comparisons with Karnofsky performance scores. J Neurooncol.  1992;14(3):243–253. [DOI] [PubMed] [Google Scholar]
  • 49. Weaver  KE, Geiger  AM, Lu  L, Case  LD.  Rural-urban disparities in health status among US cancer survivors. Cancer.  2013;119(5):1050–1057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Cioffi  G, Cote  DJ, Ostrom  QT, Kruchko  C, Barnholtz-Sloan  JS.  Association between urbanicity and surgical treatment among patients with primary glioblastoma in the United States. Neurooncol. Pract..  2020;7(3):299–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Rogers  JL, Acquaye  A, Vera  E, et al.  Provider-reported challenges and barriers to referring patients to neuro-oncology clinical trials: a report from the Society for Neuro-Oncology member survey. Neurooncol. Pract..  2020;7(1):38–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Petterson  S.  Deciphering the neighborhood atlas area deprivation index: the consequences of not standardizing. Health affairs scholar. 2023;1(5):qxad063. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

npaf069_Supplementary_Figures_1-2
npaf069_supplementary_figures_1-2.docx (284.3KB, docx)
npaf069_Supplementary_Tables_1-2
npaf069_supplementary_tables_1-2.docx (82.9KB, docx)

Data Availability Statement

The data underlying this article is available upon request due to privacy and ethical restrictions.

Articles from Neuro-Oncology Practice are provided here courtesy of Oxford University Press

RESOURCES