SUMMARY
Objective:
To examine the associations between demographic/medical and geographic factors with follow-up medical care and health-related quality of life (HRQoL) among cancer survivors during the SARS-CoV-2 pandemic.
Study design:
Cross-sectional survey.
Methods:
An online survey was sent to cancer survivors between May 2020 and January 2021, exploring their experience with SARS-CoV-2, follow-up care, and HRQoL. PolicyMap was used to geocode home addresses. Both geographic and demographic/medical factors were examined for their associations with SARS-CoV-2 experience, follow-up care, and HRQoL (FACT-G7).
Results:
Geographic data was available for 9,651 participants. Patients living in the highest area deprivation index (ADI) neighborhoods (most deprived) had higher odds of avoiding in-person general (odds ratio [OR] = 7.20; 95% confidence interval [CI] = 2.79–18.60), cancer (OR = 8.47; 95% CI = 3.73–19.30), and emergency (OR = 14.2; 95% CI = 5.57–36.30) medical care, as well as lower odds of using telemedicine (OR = 0.61; 95% CI = 0.52–0.73) compared to the lowest ADI group. Race/ethnicity was not associated with follow-up care after controlling for ADI. The effect of ADI on HRQoL was generally in the expected direction, with higher ADI being associated with worse HRQoL.
Conclusions:
ADI influenced follow-up medical care more than age, race/ethnicity, or health insurance type. Healthcare providers and institutions should focus on decreasing barriers to in-person and telemedicine health care that disproportionally impact those living in more deprived communities, which are exacerbated by healthcare disruptions like those caused by the SARS-CoV-2 pandemic.
Keywords: Area Deprivation Index, cancer, follow-up medical care, SARS-CoV-2, healthcare access
Introduction
The SARS-CoV-2 pandemic caused significant interruptions in all facets of life. Time-sensitive cancer care was one of the most disrupted sectors, as was found in a study conducted in India.1 Non-critical appointments were canceled by the medical system or patients, missing routine follow-up visits, post-treatment screening, emergency room visits, or other medical visits in the US,2 Europe,3 and Australia.4 For cancer survivors already vulnerable to severe illness, concern about contracting SARS-CoV-2 was wise because a concurrent SARS-CoV-2 diagnosis was associated with significantly more hospitalizations, ICU admissions, and mortality than for those without cancer.5 A meta-analysis of European and North American WHO regions found having active cancer was significantly associated with a greater risk of death from SARS-CoV-2.6 Although racial disparities in health in the US were made clear during the SARS-CoV-2 pandemic, less research has specifically examined how racial and socioeconomic factors influenced cancer survivors during the SARS-CoV-2 pandemic.
SARS-CoV-2 CDC data reported that non-Hispanic Black or African Americans experienced a two times greater risk of hospitalization and a 60% greater risk of death. In comparison, Hispanics were at 1.8 times greater risk of hospitalization and 70% greater risk of death compared to non-Hispanic Whites.7 Similarly, there are significant oncology disparities, with Non-Hispanic Black8 and Hispanic9 cancer death rates higher than non-Hispanic Whites. SARS-CoV-2 outcomes among Non-Hispanic Black patients with cancer were similarly worse than their non-Hispanic White cancer patient counterparts.10 Qualitative UK data describes the exacerbating effect of SARS-CoV-2 on health outcomes,11 with significantly greater deaths occurring in more socioeconomically deprived communities than less socioeconomically deprived communities in England and Wales,12 Ireland,13 as well as in Korea.14 One factor linking race with health outcomes is the association between race and economic deprivation. Economic deprivation can be measured using the area deprivation index (ADI). The 17-item geographic metric includes education, housing quality, employment, and poverty data. ADI is a percentile score in relation to the nation, ranging from 0 (least disadvantaged) to 100 (most disadvantaged).15
Some research has examined the association between ADI and race/ethnicity during the SARS-CoV-2 pandemic. When controlling for ADI, researchers found the odds were higher for both contracting SARS-CoV-2 and SARS-CoV-2-related ICU admissions among those living in “Non-Hispanic Black majority neighborhoods” compared to non-Hispanic White majority neighborhoods.16 For SARS-CoV-2 death, a significant ADI-race interaction effect was found in “Non-Hispanic Black majority neighborhoods,” where a more substantial increase in death was found per decile increase in ADI level.16 Compared to the lowest ADI quintile, the odds of dying from SARS-CoV-2 were 74% higher for those living in the most deprived quintile.17 The role of race/ethnicity and ADI in cancer survivors as it relates to their follow-up cancer care during SARS-CoV-2 has not been explored.
The current study examined the association between race/ethnicity and ADI and how they related to follow-up care and health-related quality of life (HRQoL) in a cohort of cancer survivors. We hypothesized that after adjusting for relevant clinical and demographic factors (e.g., cancer type, stage, treatment, age, and sex), Non-Hispanic Black race, Hispanic ethnicity, and high ADI would each be associated with greater medical care disruptions due to SARS-CoV-2 and worse HRQoL. In addition, we hypothesized that ADI would be the strongest predictor of medical care disruptions and HRQoL in multivariate analyses that included clinical/demographic factors, race, ethnicity, and ADI.
Method
This is a secondary analysis of an online survey conducted between May 2020 and January 2021 to investigate the impact of the SARS-CoV-2 pandemic on cancer patients. Participants were recruited from two National Cancer Institute-designated comprehensive cancer centers in the Southern US. Inclusion criteria were: ≥18 years old; English- or Spanish-speaking; cancer center patients; attended a clinical visit within the past five years; ICD-10 confirmed cancer diagnosis; and contactable via email or MyChart. Patients received an email or MyChart message with the link to the online REDCap survey. REDCap eConsent Framework provided an electronic informed consent document to sign, followed by access to the brief electronic questionnaire comprised of three sections: (1) exposures and practical experiences related to the SARS-CoV-2 pandemic (“COVID-related exposures”); (2) SARS-CoV-2-related Psychosocial and Practical Experiences questionnaire (COVID-PPE); and (3) HRQoL.18 Patients whose mailing addresses were available to obtain ADI were included in this secondary analysis.
Participants reported on four healthcare disruption items: 1) decided not to attend an in-person general medical appointment; 2) decided not to attend an in-person cancer care appointment; 3) decided not to seek emergency/urgent care; and 3) participated in a telehealth medical visit. These four items were among 19 SARS-CoV-2-related exposure items, including testing and results in self/family/friends, hospitalization, symptoms, death of family or friends, days of social isolation, risk factors, and SARS-CoV-2 effects on finances and daily life. Outcomes and psychometrics for these other items and the COVID-PPE are reported elsewhere.18, 19 HRQoL was measured using the Functional Assessment of Cancer Therapy General-7 Item Version (FACT-G7). The total score (α=.80) provides an index of overall HRQoL, and the cutoff score of 13 and under was used to indicate low HRQoL.20
The primary exposure was ADI. Demographics and clinical characteristics, including age, sex, race/ethnicity, insurance status, primary cancer site, and cancer stage, were obtained from an electronic data warehouse (EDW). Geographic information systems (GIS) are computer tools increasingly used in social science and healthcare research that uses information about people based on physical spaces or environmental attributes to aid the visualization and analysis of environmental factors and health.21 A Census Block Group (CBG) is one geographic unit containing 600–3,000 people and comprises several Census Blocks.22 The CBG is considered the best approximation of a neighborhood,23 and is the area unit used in this study. Geocoding was done using PolicyMap,24 a GIS, to match current home addresses to the US CBG. Participants were excluded from geocoding if the home address was from another country, Puerto Rico or other U.S. territory without CBG geographic units, P.O. Box,25, 26 or missing addresses that prevented geocoding. Each participant was assigned the estimated median household income and ADI score of the CBG (their neighborhood).27 In addition, the residential area was categorized into three groups: 1) metropolitan (The 12 counties of the Greater Houston Region;28 Key West-Key Largo, FL Micropolitan Statistical Area; and Miami-Fort Lauderdale-West Palm Beach, FL Metropolitan Statistical Area29); 2) non-metropolitan but within the state; and 3) outside of the state as proximity of distance to the respective cancer centers.
ADI ranks CBGs by their socioeconomic advantage from 1–100.23 The rankings are scored using 17 factors (see Table 1).15, 30 The ADI score of each CBG was calculated by the Center for Health Disparities Research, University of Wisconsin School of Medicine and Public Health.23 All CBG neighborhood data came from the US Census American Communities Survey (ACS) from the five years between 2017 and 2021.23, 31 ADI are categorized into five established pre-defined groups32, 33 from the least to the most socioeconomically deprived areas (ADI 1–20, 21–40, 41–60, 61–80, 81–100).
Table 1.
Area Deprivation Index variable from the American Communities Survey
| Financial | Median family income Income disparity Percent population below 150% poverty threshold Percent families below poverty |
| Housing Cost | Median gross rent Homeownership rate Median home value Median monthly mortgage |
| Housing Quality | Household crowding Single-parent household rate Percent occupied housing units without complete plumbing Percent household without a telephone |
| Employment, Commuting, Education | Percent household without a motor vehicle Percent of people employed in white-collar occupations Unemployment rate Educational distribution (percentage of the population with less than 9 years and with 12 or more years of education) |
Chi-square tests compared patient demographics, other patient characteristics, healthcare disruptions, and low HRQoL by ADI groups. Logistic regression models examined ADI and other patient factors impacting the outcomes. Factors investigated in multivariate logistic regression were chosen based on the descriptive analysis, and the interaction between race/ethnicity and ADI was analyzed using multiplicative interaction terms. To correct the family-wise error rate due to multiple testing, Bonferroni correction adjusted p values were used. After Bonferroni correction, a significance level <0.006 was used for the Chi-squared tests comparing the healthcare disruption outcomes among demographics, and a significance level of <0.01 was applied to the logistic regression to evaluate the association between healthcare disruption outcomes and covariates. Data management and statistical analysis were performed with SAS-v9.4 (SAS Institute Inc, Cary, NC).
Results
Among 11,325 patients who responded to the questionnaire, 9651 patients with valid mailing addresses were included in this secondary analysis. Patients were primarily female (n=5534/57.3%), non-Hispanic White (n=7942/82.3%), insured (n=9502/98.6%), and had non-metastatic cancers (n=7462/77.3%). Breast cancer was the most common malignancy (22.7%), followed by hematological cancers (14.5%) and male genitourinary cancers (11.4%), as well as another 12 cancer categories each under 9%, with 22.7% reporting having stage IV disease. Patient demographics and cancer type/stage in each ADI group are shown in Table 2. Twenty-six percent (n=2496) of participants lived in low ADI areas (ADI 1–20); 30.8% (n=2974) in ADI 21–40, 22.2% (n=2140) in ADI 41–60, 13.4% (n=1298) in ADI 61–80, and 7.7% (n=743) lived in the highest ADI areas (ADI 81–100). It was noted that more non-Hispanic Black (22.1%) and Hispanics (16.4%) lived in high ADI areas than other race/ethnic groups (6.2% for non-Hispanic White and 4.4% for non-Hispanic, other). Also, more people with Medicaid (20.8%) or no insurance (33.3%) were observed in high ADI areas than people with managed care (7.3%) and Medicare (7.9%).
Table 2.
Demographics by ADI groups (N=9651)
| ADI | |||||||
|---|---|---|---|---|---|---|---|
| 1–20 | 21–40 | 41–60 | 61–80 | 81–100 | |||
| Total | N (%) | N (%) | N (%) | N (%) | N (%) | P value | |
| 9651 (100) | 2496 (25.9) | 2974 (30.8) | 2140 (22.2) | 1298 (13.4) | 743 (7.7) | ||
| Sex | <0.0001 | ||||||
| Male | 4117 (42.7) | 1092 (26.5) | 1339 (32.5) | 857 (20.8) | 559 (13.6) | 270 (6.6) | |
| Female | 5534 (57.3) | 1404 (25.4) | 1635 (29.5) | 1283 (23.2) | 739 (13.4) | 473 (8.5) | |
| Race/Ethnicity | <0.0001 | ||||||
| Non-Hispanic White | 7942 (82.3) | 2100 (26.4) | 2518 (31.7) | 1795 (22.6) | 1036 (13.0) | 493 (6.2) | |
| Non-Hispanic Black | 371 (3.8) | 48 (12.9) | 86 (23.2) | 84 (22.6) | 71 (19.1) | 82 (22.1) | |
| Non-Hispanic Other | 274 (2.8) | 109 (39.8) | 83 (30.3) | 48 (17.5) | 22 (8.3) | 12 (4.4) | |
| Hispanic Any Race | 897 (9.3) | 189 (21.1) | 241 (26.7) | 172 (19.2) | 148 (16.5) | 147 (16.4) | |
| Unknown/refused/missing | 167 (1.7) | 50 (29.9) | 46 (27.5) | 41 (24.6) | 21 (12.6) | 9 (5.3) | |
| Health Insurance | <0.0001 | ||||||
| Managed care | 4855 (50.3) | 1342 (27.6) | 1492 (30.7) | 1039 (21.4) | 630 (13.0) | 352 (7.3) | |
| Medicare | 4420 (45.8) | 1070 (24.2) | 1369 (31.0) | 1026 (23.2) | 608 (13.6) | 347 (7.9) | |
| Medicaid/Other governmental | 210 (2.2) | 29 (13.8) | 63 (30.0) | 52 (24.8) | 39 (18.6) | 27 (12.9) | |
| Other | 166 (1.7) | 55 (30.1) | 50 (30.1) | 23 (13.9) | 21 (12.7) | 17 (10.2) | |
| Location | <0.001 | ||||||
| Miami/Houston metro area | 3400 (35.2) | 1125 (33.1) | 1237 (36.4) | 627 (18.4) | 296 (8.7) | 112 (3.4) | |
| Non-metro area in TX/FL | 2625 (27.2) | 479 (18.3) | 695 (26.5) | 653 (24.9) | 465 (17.7) | 333 (12.7) | |
| Outside of TX/FL | 3469 (35.9) | 843 (24.3) | 1009 (29.1) | 824 (23.8) | 516 (14.9) | 277 (8.0) | |
| Unknown | 157 (1.6) | 49 (31.2) | 33 (21.0) | 36 (22.9) | 21 (13.4) | 18 (11.5) | |
| Age M (SD) | 63.8 (12.3) | 63.7 (12.6) | 64.2 (11.9) | 63.7 (12.4) | 63.8 (11.8) | 63.2 (13.0) | 0.31 |
| Estimated Median Income (dollars) M (SD) | 92437.1 (45632.3) | 144784.3 (46659.9) | 94968.6 (23845.3) | 69502.4 (15221.3) | 52939.0 (11980.2) | 37722.9 (9988.0) | <0.0001 |
| N (%) | N (%) | N (%) | N (%) | N (%) | |||
| Cancer Type | 0.03 | ||||||
| Hematologic Malignancy | 1400 (14.5) | 385 (27.5) | 434 (31.0) | 300 (21.4) | 184 (13.1) | 97 (6.9) | |
| Breast | 2186 (22.7) | 591 (27.0) | 643 (29.4) | 495 (22.6) | 274 (12.5) | 183 (8.4) | |
| Genital System, Male | 1092 (11.3) | 293 (26.8) | 357 (32.7) | 228 (20.9) | 147 (13.5) | 67 (6.1) | |
| Abdominal Cavity Viscera | 814 (8.4) | 183 (22.5) | 248 (30.5) | 192 (23.6) | 107 (13.1) | 84 (10.3) | |
| Genital System, Female | 730 (7.6) | 172 (23.6) | 221 (30.3) | 175 (24.0) | 90 (12.3) | 72 (9.9) | |
| Chest Cavity Viscera | 659 (6.8) | 154 (23.4) | 190 (28.8) | 158 (24.0) | 101 (15.3) | 56 (8.5) | |
| Urinary Tract | 512 (5.3) | 110 (21.5) | 160 (31.2) | 127 (54.8) | 80 (15.6) | 35 (6.8) | |
| Head and Neck | 501 (5.2) | 140 (27.9) | 158 (31.5) | 109 (21.6) | 69 (13.8) | 25 (5.0) | |
| Skin | 427 (4.4) | 105 (24.6) | 148 (34.7) | 86 (20.1) | 65 (15.2) | 23 (5.4) | |
| Endocrine | 364 (3.8) | 93 (25.4) | 110 (30.2) | 66 (18.1) | 61 (9.3) | 34 (9.3) | |
| Soft Tissue | 298 (3.1) | 85 (28.5) | 90 (30.2) | 62 (20.8) | 41 (10.4) | 20 (6.7) | |
| Nervous System | 140 (1.5) | 30 (21.4) | 50 (35.7) | 35 (25.0) | 13 (9.3) | 12 (8.6) | |
| Neuroendocrine | 115 (1.2) | 29 (25.2) | 38 (33.0) | 28 (24.3) | 12 (10.4) | 8 (7.0) | |
| Bone/Skeletal System | 86 (0.9) | 20 (23.3) | 28 (32.6) | 16 (18.6) | 15 (17.4) | 7 (8.1) | |
| Eye | 31 (0.3) | 7 (22.6) | 11 (35.5) | 8 (25.8) | 3 (9.7) | 2 (6.5) | |
| Other &Ill-Defined | 194 (2.0) | 65 (33.5) | 56 (28.9) | 38 (19.6) | 24 (12.4) | 11 (5.7) | |
| Metastatic cancer, unknown primary | 102 (1.1) | 34 (33.3) | 32 (31.4) | 17 (16.7) | 12 (11.8) | 7 (6.9) | |
| Cancer Stage | 0.08 | ||||||
| Stage 0-III | 7462 (77.3) | 1962 (26.3) | 2324 (31.1) | 1633 (21.9) | 979 (13.1) | 564 (7.6) | |
| Stage IV | 2189 (22.7) | 534 (24.4) | 650 (29.7) | 507 (23.2) | 319 (14.6) | 179 (8.2) | |
FL: Florida, TX: Texas
Among patients who answered the questions about healthcare disruption, 95.5% (4187/4384) decided not to attend in-person general medical appointments, 91.1% (2908/3192) chose not to participate in in-person cancer care appointments, 68.5% (747/1090) decided to avoid seeking emergency room visits, and 57.4% (5525/9609) participated in telemedicine. Overall, 15.0% (1437/9571) had a low FACT-G7 score. Differences in healthcare disruption and HRQoL by ADI and other patient characteristics are described in Table 3. There were significant differences in healthcare disruptions among ADI groups. The frequency of cancer patients who avoided in-person medical care increased, while the frequency of cancer patients who participated in telehealth was lower in the more deprived neighborhoods. The frequency of cancer patients with low HRQoL scores was higher in more disadvantaged communities. There were significant differences in healthcare disruptions by race/ethnicity as well. The frequency of patients who avoided in-person medical care was higher in non-Hispanic patients than in Hispanic patients. However, there was no difference in HRQoL among race/ethnicity.
Table 3.
Differences in healthcare disruption and health-related quality of life by patient characteristics
| Decided not to attend an in-person general medical appointment | Decided not to attend an in-person cancer care appointment | Decided not to seek emergency/urgent care | Participated telehealth | Low HRQOL (FACT G7≤13) |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| n/N (%) | P-value | n/N (%) | P-value | n/N (%) | P-value | n/N (%) | P-value | n/N (%) | P-value | |
| Sex | NS | NS | NS | <0.0001* | <0.0001* | |||||
| Males | 1687/1772 (95.2) | 1227/1346 (91.2) | 270/414 (65.2) | 2248/4096 (54.9) | 522/4093 (12.8) | |||||
| Females | 2500/2612 (95.7) | 1681/1846 (91.1) | 477/676 (70.6) | 3277/5513 (59.4) | 915/5478 (16.7) | |||||
| Race/ethnicity | <0.0001* | <0.0001* | <0.0001* | NS | NS | |||||
| Non-Hispanic White | 3443/3554 (96.9) | 2397/2574 (93.1) | 570/782 (73.0) | 4542/7923 (57.3) | 1164/7882 (14.8) | |||||
| Non-Hispanic Non-Hispanic Black | 173/177 (97.7) | 126/132 (95.5) | 33/43 (76.7) | 208/369 (56.4) | 53/366 (14.5) | |||||
| Non-Hispanic, all other races | 123/127 (96.9) | 91/98 (92.9) | 36/43 (83.7) | 163/272 (59.9) | 35/7882 (12.9) | |||||
| Hispanic, any race | 369/447 (82.6) | 229/322 (71.1) | 92/205 (44.9) | 523/878 (59.6) | 157/887 (17.7) | |||||
| Age | NS | NS | NS | NS | <0.0001* | |||||
| <65 | 1772/1869 (94.8) | 1266/1402 (60.3) | 372/535 (69.5) | 58.7 (2592/4413) | 776/4397 (17.7) | |||||
| >=65 | 2438/2538 (96.1) | 1633/1811 (91.8) | 379/559 (67.8) | 56.4 (2954/5242) | 675/5221 (12.9) | |||||
| Cancer Types | 0.19 | 0.76 | 0.08 | 0.94 | 0.25 | |||||
| Hematological malignancies | 661/699 (94.6) | 490/540 (90.7) | 106/169 (62.7) | 804/1396 (57.6) | 223/1390 (16.0) | |||||
| Solid tumor/other | 3526/3685 (95.7) | 2418/2652 (91.2) | 641/921 (69.6) | 4721/8213 (57.5) | 1214/8181(14.8) | |||||
| Cancer stage | <0.0001* | <0.0001* | <0.0001* | <0.0001* | <0.0001* | |||||
| Non-metastatic | 3319/3534 (94.4) | 2186/2470 (88.5) | 574/917 (62.6) | 4115/7433 (55.4) | 1030/7393 (14.0) | |||||
| Metastatic | 868/868 (100) | 722/722 (100) | 173/173 (100) | 1410/2176 (64.8) | 407/2178 (18.7) | |||||
| Health insurance type | <0.0001* | <0.0001* | 0.0001* | NS | <0.0001* | |||||
| Managed care | 1960/2057 (95.3) | 1358/1493 (91.0) | 369/532 (69.4) | 2769/4833 (57.3) | 715/4816 (14.9) | |||||
| Medicare | 2073/2138 (97.0) | 1438/1539 (93.4) | 328/456 (71.9) | 2516/4401 (57.2) | 625/4384 (14.3) | |||||
| Medicaid/Other governmental | 8995 (93.7) | 6271 (87.3) | 23/31 (74.2) | 132/209 (63.2) | 20/47 (42.6) | |||||
| Other | 65/94 (69.2) | 50/89 (56.2) | 27/71 (38.0) | 108/166 (65.1) | 77/324 (23.8) | |||||
| Location | <0.0001* | <0.0001* | <0.0001* | 0.003* | 0.003* | |||||
| Miami/Houston metro area | 1368/1545 (88.5) | 847/1101 (76.9) | 262/560 (46.8) | 2027/3383 (59.9) | 473/3368 (14.0) | |||||
| Non-metro area in TX/FL | 1107/1123 (98.6) | 742/768 (96.6) | 180/219 (82.2) | 1486/2611 (56.9) | 374/2303 (14.4) | |||||
| Outside of TX/FL | 1648/1652 (99.8) | 1276/1280 (99.7) | 297/303 (98.0) | 1936/3458 (56.0) | 580/3447 (16.8) | |||||
| ADI | <0.0001* | <0.0001* | <0.0001* | <0.0001* | 0.003* | |||||
| 1–20 | 1105/1189 (92.9) | 730/864 (84.5) | 163/322 (50.6) | 1524/2482 (61.4) | 319/2467 (12.9) | |||||
| 21–40 | 1276/1351 (94.5) | 897/988 (90.8) | 241/356 (67.7) | 1760/2963 (59.4) | 434/2945 (14.7) | |||||
| 41–60 | 917/935 (98.1) | 652/682 (95.6) | 194/231 (84.0) | 1161/2133 (54.4) | 340/2131 (16.0) | |||||
| 61–80 | 566/581 (97.4) | 405/427 (94.9) | 96/122 (79.7) | 706/1292 (54.6) | 221/1291 (17.1) | |||||
| 81–100 | 323/328 (98.5) | 224/231 (97.0) | 53/59 (89.8) | 374/739 (50.6) | 123/737 (16.7) | |||||
p<0.006 (0.05/8) is accepted as statistically significant after Bonferroni correction.
ADI: area deprivation index, FACT G7: Functional Assessment of Cancer Therapy- General- 7 Item Version, HRQOL: health-related quality of life, NS: not significant
Table 4 summarizes the univariate and multivariate logistic regression results examining the association of ADI and other patient demographics with healthcare disruptions and low HRQoL. The median income was also removed because it correlated with ADI, given that both variables were obtained using census block groups (Pearson Correlation −0.79, p<0.001). All metastatic cancer patients reported avoiding attending an in-person general medical appointment and cancer care, as well as avoiding seeking emergency care. Therefore, the odds ratios were not calculated, and the cancer stage was removed from the multivariate logistic regression models for these three outcomes. The interaction between race/ethnicity and AID was not statistically significant.
Table 4.
Association between healthcare utilization/health-related quality of life and patient demographics
| Health care-related outcomes | Health-related Quality of Life | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Predictor╲Outcomes | Decided not to attend an in-person general medical appointment OR (95%CI) |
Decided not to attend an in-person cancer care appointment OR (95%CI) |
Decided not to seek emergency/urgent care OR (95%CI) |
Participated in Telemedicine OR (95%CI) |
FACT G7 ≤ 13 OR (95%CI) |
|||||
| Univariate | Multivariate | Univariate | Multivariate | Univariate | Multivariate | Univariate | Multivariate | Univariate | Multivariate | |
| ADI | ||||||||||
| 1–20 (ref) | - | - | - | - | - | - | - | - | - | - |
| 21–40 | 1.29 (0.94–1.78) | 1.14 (0.80–1.62) | 1.81 (1.36–2.40) * | 1.76 (1.27–2.44) * | 2.04 (1.50–2.79) * | 2.19 (1.51–3.17) * | 0.92 (0.83–1.03) | 0.91 (0.81–1.01) | 1.16 (0.99–1.36) | 1.17 (0.99–1.37) |
| 41–60 | 3.87 (2.31–6.49) * | 2.27 (1.31–3.93) * | 3.99 (2.65–6.01) * | 2.37 (1.51–3.73) * | 5.12 (3.38–7.74) * | 3.19 (1.97–5.17) * | 0.75 (0.67–0.84) * | 0.72 (0.64–0.81) * | 1.28 (1.08–1.51) * | 1.23 (1.03–1.46) |
| 61–80 | 2.87 (1.64–5.02) * | 1.82 (0.98–3.37) | 3.38 (2.12–5.39) * | 2.18 (1.27–3.73) * | 3.60 (2.22–5.85) * | 2.20 (1.21–4.02) | 0.76 (0.66–0.87) * | 0.73 (0.63–0.84) * | 1.39 (1.15–1.68) * | 1.31 (1.07–1.59) * |
| 81–100 | 4.91 (1.98–12.20) * | 2.85 (1.06–7.65) | 5.87 (2.71–12.70) * | 3.22 (1.33–7.82) * | 8.62 (3.60–20.60) * | 7.49 (2.77–20.3) * | 0.64 (0.55–0.76) * | 0.62 (0.52–0.74) * | 1.35 (1.08–1.69) * | 1.26 (0.99–1.60) |
| Age | - | - | - | - | - | - | - | - | - | - |
| ≥65 | 1.33 (1.00–1.77) | 0.62 (0.40–0.96) | 1.20 (0.94–1.54) | 0.56 (0.38–0.84) * | 0.92 (0.71–1.19) | 0.50 (0.32–0.78) * | 0.91 (0.93–0.98) | 0.86 (0.76–0.98) | 0.69 (0.62–0.77) * | 0.48 (0.39–0.58) * |
| Sex | ||||||||||
| Male (ref.) | - | - | - | - | - | - | - | - | - | - |
| Female | 1.13 (0.84–1.50) | 1.22 (0.88–1.69) | 0.99 (0.77–1.27) | 1.07 (0.80–1.43) | 1.28 (0.98–1.67) | 1.29 (0.93–1.80) | 1.24 (1.11–1.31) * | 1.23 (1.12–1.34) * | 1.37 (0.22–1.54) * | 1.31 (1.16–1.48) * |
| Race/Ethnicity | ||||||||||
| Non-Hispanic, White (ref) | - | - | - | - | - | - | - | - | - | - |
| Non-Hispanic Black | 1.39 (0.51–3.83) | 1.83 (0.65–5.17) | 1.55 (0.67–3.57) | 2.29 (0.95–5.52) | 1.23 (0.60–2.53) | 1.40 (0.61–3.23) | 0.96 (0.78–1.19) | 0.96 (0.77–1.20) | 0.98 (0.73–1.32) | 0.86 (0.63–1.17) |
| Non-Hispanic, Other | 0.99 (0.36–2.73) | 1.04 (0.37–2.95) | 0.96 (0.44–2.10) | 1.31 (0.57–3.02) | 1.91 (0.84–4.36) | 2.74 (1.12–6.73) | 1.11 (0.87–1.42) | 1.02 (0.79–1.31) | 0.86 (0.60–1.23) | 0.81 (0.56–1.16) |
| Hispanic | 0.15 (0.11–0.21) * | 0.21 (0.15–0.30) * | 0.18 (0.14–0.24) * | 0.30 (0.21–0.42) * | 0.30 (0.22–0.42) * | 0.41 (0.27–0.60) * | 1.10 (0.95–1.26) | 1.08 (0.93–1.26) | 1.24 (1.03–1.49) | 1.16 (0.95–1.41) |
| Median Income | ||||||||||
| (unit $10,000) | 1.01 (0.97–1.04) | - | 0.99 (0.97–1.02) | - | 0.99 (0.96–1.02) | - | 1.02 (1.01–1.03) * | - | 0.97 (0.95–0.98) * | - |
| Health Insurance Type | ||||||||||
| Managed (ref) | - | - | - | - | - | - | - | - | - | - |
| Medicare | 1.58 (1.15–2.17) * | 1.72 (1.08–2.73) | 1.42 (1.08–1.85) | 1.76 (1.17–2.66) * | 1.13 (0.86–1.49) | 1.47 (0.92–2.36) | 0.99 (0.92–1.08) | 1.17 (1.03–1.33) | 0.95 (0.85–1.07) | 1.77 (1.46–2.16) * |
| Medicaid/Other governmental | 0.73 (0.31–1.72) | 0.71 (0.27–1.83) | 0.69 (0.33–1.41) * | 0.67 (0.28–1.59) | 1.27 (0.42–2.90) | 1.69 (0.62–4.66) | 1.28 (0.96–1.70) | 1.39 (1.04–1.87) | 2.20 (1.61–3.01) * | 2.32 (1.68–3.20) * |
| *Other | 0.11 (0.07–0.18) * | 0.18 (0.10–0.31) * | 0.13 (0.08–0.20) * | 0.18 (0.10–0.31) * | 0.27 (0.16–0.45) * | 0.41 (0.22–0.75) * | 1.39 (1.00–1.92) | 1.35 (0.97–1.88) | 1.82 (1.26–2.63) * | 2.05 (1.40–3.00) * |
| Cancer Stage | ||||||||||
| Non- metastatic (ref) | - | - | - | - | - | - | - | - | - | - |
| Metastatic | NA | - | NA | - | NA | - | 1.45 (1.34–1.64) * | 1.59 (1.44–1.77) * | 1.42 (1.25–1.61) * | 1.42 (1.24–1.62) * |
| Cancer Type | ||||||||||
| Solid tumor (ref) | - | - | - | - | - | - | - | - | - | - |
| Hematologic malignancy | 0.78 (0.55–1.13) | 0.58 (0.39–0.88) * | 0.95 (0.69–1.31) | 0.71 (0.49–1.04) | 0.74 (0.52–1.04) | 0.62 (0.40–0.97) | 1.10 (0.96–1.26) | 1.14 (1.01–1.29) | 1.01 (0.90–1.13) | 1.28 (1.08–1.51) * |
| Location | ||||||||||
| Miami/Houston metro area (ref) | - | - | - | - | - | - | - | - | - | - |
| Non-metro area in FL/TX | 8.95 (5.33–15.0) * | 8.00 (4.67–13.7) * | 8.56 (5.65–13.0) * | 7.9 (5.03–12.1) * | 5.25 (3.58–7.71) * | 4.50 (2.97–6.80) * | 0.88 (0.80–0.98) | 0.94 (0.84–1.05) | 1.03 (0.89–1.19) | 0.94 (0.80–1.09) |
| Outside of FL/TX | 53.30 (19.7–143) * | 36.9 (13.6–100) * | 95.7 (35.5–257) * | 72.4 (26.7–196) * | 56.3 (24.7–128) * | 42.6 (18.4–98.4) | 0.85 (0.77–0.94) * | 0.87 (0.78–0.96) * | 1.24 (1.09–1.41) * | 1.18 (1.03–1.36) |
P<0.01 (0.05/5) is accepted as statistically significant after Bonferroni correction.
ADI: area deprivation index, CI: confidence interval, FACT G7: Functional Assessment of Cancer Therapy- General- 7 Item Version, NA: not applicable, OR: odds ratio, ref: reference.
ADI was the strongest predictor of study outcomes. Patients with higher ADI were more likely to avoid in-person cancer care compared to the lowest ADI group (0–20, least deprived neighborhoods) (Adjusted odds ratio [aOR]: 1.76, 2.37, 2.18, and 3.22, 95% confidence interval [CI];.1.27–2.40, 1.51–3.73, 1.27–3.73, 1.33–7.82, for ADI 21–40, 41–60, 61–80, and 81–100, respectively). Similarly, the patients in the middle to high ADI groups (41–60, 61–80, 81–100) were less likely to participate in telemedicine compared to the lowest ADI group (aOR; 0.72, 0.73, and 0.62, 95%CI; 0.64–0.81, 0.63–0.84, 0.52–0.74, respectively). These results suggest that high ADI was associated with more medical care disruptions. Unexpectedly, Hispanic patients were less likely to experience medical care disruption compared to non-Hispanic White patients (aOR for decision not to attend an in-person general medical appointment; 0.15, 95%CI; 0.11 – 0.21, aOR for decision not to attend an in-person cancer care appointment; 0.30, 95%CI; 0.21–0.42, aOR for decision not to seek emergency/urgent care; 0.41, 95%CI; 0.27–0.60). Patients with hematologic malignancy were less likely to decide not to attend in-person medical appointments and not to seek emergency care compared to patients with solid tumors. The location of residence was significantly associated with the decisions not to attend in-person medical care. The high OR is because almost all patients living out-of-state responded “Yes” to missing in-person health care visits.
The only significant association between low HRQoL and ADI groups showed that the patients in the second highest ADI (61–80) group were more likely to have low HRQoL compared to the lowest ADI (1–20) group (aOR=1.31, 95%CI=1.07–1.59). Other outcomes were in the expected direction, with higher ADI associated with worse HRQoL. Race/ethnicity was not significantly associated with low HRQoL. Patients classified as having metastatic disease versus non-metastatic or being on Medicare, Medicaid, or other insurance relative to managed care reported worse HRQoL; female patients had worse HRQoL than male patients, and those 65 years old or older reported better HRQoL. Patients with hematological malignancies reported worse HRQoL relative to patients with solid tumors.
Discussion
SARS-CoV-2 disrupted health care worldwide, especially for cancer patients. Here, nearly all cancer survivors avoided some form of follow-up health care. Supporting our hypotheses, ADI was significantly associated with healthcare disruption, even after controlling for demographic (including race/ethnicity) and medical characteristics. The most deprived, relative to the least deprived neighborhoods, had greater odds of missing appointments: 3.2 times cancer-specific medical appointments, 7.5 times emergency/urgent care, and 0.6 times fewer telemedicine appointments. Neighborhoods in the second to fourth ADI quintiles were at significantly greater odds of choosing to miss in-person general medical care, cancer care, and emergency/urgent care than the lowest ADI quintiles. Compared to non-Hispanic Whites, Hispanics were less likely to avoid in-person general medical appointments (aOR=0.21, 95%CI=0.15–0.30), less likely to avoid cancer-specific medical appointments (aOR=0.30, 95%CI=0.21–0.42), and less likely to avoid emergency/urgent care (aOR=0.41, 95%CI=0.27–0.06) in multivariate analyses. There were no significant differences between non-Hispanic Whites and non-Hispanic Blacks or non-Hispanic Whites and the Other racial group in healthcare utilization in multivariate analyses. Breast cancer patients missed more appointments relative to the other cancer groups, even after controlling for race/ethnicity, ADI, and other characteristics. In multivariate analyses, the main factors associated with worse HRQoL were being in the second highest ADI category (although the other ADI groups followed a similar pattern), having metastatic disease, being under 65 years of age, having breast cancer, and being on Medicare/Medicaid.
The current results mirror other research finding substantial interruptions of crucial medical care during the pandemic in Australia4 Belgium,3 India,1 and the US.2 While prior research found 64% reported that missing oncology appointments/treatments threatened their health more than contracting SARS-CoV-2,34 this study’s results found much greater willingness to disregard oncologic and emergency department care, with 100% of people with metastatic disease in our study avoiding in-person general medical care, cancer care appointments, and emergency/urgent care. We found up to 57% of respondents utilized telehealth visits, more than the 25% reported in other studies.35, 36 This is supported by international findings where most study participants valued telephone consults that replace some in-person consults including general health care patients in Toronto,37 and hematologic patients in Denmark.38 However, less deprived neighborhoods used telemedicine more than the most disadvantaged neighborhoods. This matches Hassan et al., who found a 3% decrease in telemedicine use for every 10-unit increase in deprivation score.39 Qualitative data from Murphy et al. observed cancer care stakeholders found lower socioeconomic status groups experienced worse cancer care disruptions than their more advantaged peers.13 Mora et al. found less chemotherapy use in higher ADI neighborhoods in a pre-pandemic sample, and they similarly found greater odds of avoiding appointments for those living in the most deprived neighborhoods than those living in the least disadvantaged neighborhoods.40 We found that Hispanic Americans had lower odds of experiencing healthcare disruptions than non-Hispanic Whites, and there were no healthcare disruption differences between non-Hispanic Whites and non-Hispanic Blacks in multivariate analyses. Similarly, Wenger et al. found weak associations between postponed/canceled appointments and race/ethnicity.41 Conversely, Conley et al. found that non-Hispanic White participants experienced fewer healthcare disruptions than patients from other racial or ethnic backgrounds.42 Similar to our own small but statistically significant findings, the fourth ADI quintile had greater odds of a worse HRQoL; prior research found living in more deprived neighborhoods (higher ADI) was associated with greater odds of a worse HRQoL and greater anxiety.34 The location of residence was significantly associated with the outcomes. However, careful interpretation is needed given the lack of response variability (99.8% of patients living out-of-state reported that they decided not to attend an in-person general medical appointment).
Social gradients of health entered the healthcare lexicon decades ago, and our findings mirror the extensive research by Sir Michael Marmot and colleagues showing an association between fewer community resources and greater healthcare disruption.43 Ecologically,44 individual forgetfulness, family duties like childcare, and access to functional personal/public transportation in the community affect getting to the doctor,45 with reduced resources to overcome these obstacles in high ADI communities. This multifactorial burden on cancer patients, weighing heavier in more deprived communities, may explain the differences in appointment attendance in our study. Similar patterns were noted for the association of increasing ADI and worse HRQoL, but those associations diminished when controlling for race/ethnicity and other medical and demographic factors. However, the magnitude of the effect was substantial. The strongest association with worse HRQoL was being on Medicaid, however, only 48 patients were on Medicaid.
Unlike what was expected, we did not find significant negative associations between non-Hispanic Black and Hispanic racial/ethnic identities and missed appointments of all types in the multivariable analyses after adjusting for ADI and other covariates, even though this was the case with other SARS-CoV-2-related research.41 However, we found significant associations between the non-Hispanic Black race and residing in higher deprivation index communities in univariate analyses, which may be reflected in worse mortality outcomes for non-Hispanic Black people than for other races.8, 46 Our findings of significantly lower odds of missed appointments for non-White Hispanics contrast epidemiological findings of lower rates of localized breast cancer diagnoses, a proxy for timely cancer diagnosis screenings, compared to non-Hispanic Whites.9, 47 There is also conflicting evidence to suggest that Hispanics underutilize health care, even with health insurance,47 and therefore there are fewer healthcare disruptions due to less overall utilization. One possible reason Hispanic cancer survivors reported less healthcare disruption in the current study is that culturally, interactions with medical providers are valued and prioritized because the quality of healthcare delivered is seen as broadly positive.48 However, the current study did not assess healthcare utilization per se, and further research is needed in this area. We found no associations between race/ethnicity and HRQoL, perhaps because a small portion of the sample reported poor HRQoL; 15% reported a FACT-G7 score of 13 or less.
This study had several limitations. The correlational, cross-sectional nature of the data makes inferences on causality impossible. The sample was overrepresented by non-Hispanic Whites. Non-Hispanic Black participants and higher ADI communities were underrepresented. However, 21% of the sample did fall within the top two quintiles. Ninety-four percent of the sample comprised managed care and Medicare health insurance users. Few participants used Medicaid. The weak association between ADI, race/ethnicity, and HRQoL is likely due to a ceiling effect. Most participants reported high HRQoL. Finally, study participants were treated in two Southern cancer hospitals, which may not reflect the medical care behaviors of cancer treatment usage in other parts of the US.
Nevertheless, this study represents one of the most extensive assessments of cancer survivors’ medical behaviors during the SARS-CoV-2 pandemic and the influence of neighborhood deprivation. Notwithstanding the seriousness of the medical condition and the need for continuity of care, most cancer survivors reported missing all types of critical medical appointments. Even though there was an association between ADI and race/ethnicity, with more Non-Hispanic blacks and Hispanics living in high ADI areas than other race/ethnic groups, ADI was the most influential factor associated with follow-up medical care in multivariable models, more so than age, race/ethnicity, or health insurance type. Practitioners should consider novel ways to increase appointment attendance that center on overcoming community deprivation.
Acknowledgments
The authors would like to thank the study participants for contributing to the research. The authors wish to thank Rosalinda Engle and Telma I. Gomez for their data management assistance.
Funding
This work was supported in part by the National Institutes of Health through the University of Miami Sylvester Comprehensive Cancer Center’s Cancer Center Support Grant CA240139 and MD Anderson’s Cancer Center Support Grant CA016672, the Duncan Family Institute for Cancer Prevention and Risk Assessment, and the Richard E. Haynes Distinguished Professorship for Clinical Cancer Prevention at The University of Texas MD Anderson Cancer Center (L. Cohen).
Footnotes
Ethical approval
This study received ethics approval from the institutional review boards at the University of Miami Sylvester Comprehensive Cancer Center: #202000450, and MD Anderson Cancer Center: #2020–0508.
Competing interests
None declared.
REFERENCES
- 1.Mitra S, Simson DK, Khurana H, Tandon S, Ahlawat P, Bansal N, et al. Treatment Delay during Radiotherapy of Cancer Patients due to COVID-19 Pandemic. Asian Pac J Cancer Prev. 2022; 23:2415–20. 10.31557/apjcp.2022.23.7.2415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Czeisler MÉ, Marynak K, Clarke KEN, Salah Z, Shakya I, Thierry JM, et al. Delay or Avoidance of Medical Care Because of COVID-19–Related Concerns — United States. MMWR Morb Mortal Wkly Rep, CDC;2020. Contract No.: 36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kenis I, Theys S, Hermie E, Foulon V, Van Hecke A. Impact of COVID-19 on the Organization of Cancer Care in Belgium: Lessons Learned for the (Post-)Pandemic Future. Int J Environ Res Public Health. 2022; 19. 10.3390/ijerph191912456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Davis C, Cadet T, Rune K, Wilson P, Banks J. Exploring the challenges in accessing care and support for cancer survivors in Australia during COVID-19. J Psychosoc Oncol. 2021; 39:479–84. 10.1080/07347332.2021.1904086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chavez-MacGregor M, Lei X, Zhao H, Scheet P, Giordano SH. Evaluation of COVID-19 Mortality and Adverse Outcomes in US Patients With or Without Cancer. JAMA Oncology. 2022; 8:69–78. 10.1001/jamaoncol.2021.5148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Treskova-Schwarzbach M, Haas L, Reda S, Pilic A, Borodova A, Karimi K, et al. Pre-existing health conditions and severe COVID-19 outcomes: an umbrella review approach and meta-analysis of global evidence. BMC Med. 2021; 19:212. 10.1186/s12916-021-02058-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.CDC. Risk for COVID-19 Infection, Hospitalization, and Death By Race/Ethnicity. 2023. [updated May 25, 2023Aug 14, 2023]; Available from: https://archive.cdc.gov/#/details?url=https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigations-discovery/hospitalization-death-by-race-ethnicity.html.
- 8.Giaquinto AN, Miller KD, Tossas KY, Winn RA, Jemal A, Siegel RL. Cancer statistics for African American/Black People 2022. CA: A Cancer Journal for Clinicians. 2022; 72:202–29. 10.3322/caac.21718. [DOI] [PubMed] [Google Scholar]
- 9.Miller KD, Ortiz AP, Pinheiro PS, Bandi P, Minihan A, Fuchs HE, et al. Cancer statistics for the US Hispanic/Latino population, 2021. CA: A Cancer Journal for Clinicians. 2021; 71:466–87. 10.3322/caac.21695. [DOI] [PubMed] [Google Scholar]
- 10.Fu J, Reid SA, French B, Hennessy C, Hwang C, Gatson NT, et al. Racial Disparities in COVID-19 Outcomes Among Black and White Patients With Cancer. JAMA Network Open. 2022; 5:e224304–e. 10.1001/jamanetworkopen.2022.4304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gogoi M, Chaloner J, Qureshi I, Wobi F, Al-Oraibi A, Wilson H, et al. One virus, many lives: a qualitative study of lived experiences and quality of life of adults from diverse backgrounds living in the UK during the COVID-19 pandemic. BMJ Open. 2023; 13:e067569. 10.1136/bmjopen-2022-067569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kontopantelis E, Mamas MA, Webb RT, Castro A, Rutter MK, Gale CP, et al. Excess years of life lost to COVID-19 and other causes of death by sex, neighbourhood deprivation, and region in England and Wales during 2020: A registry-based study. PLoS Med. 2022; 19:e1003904. 10.1371/journal.pmed.1003904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Murphy A, Lawlor A, Kirby A, Drummond FJ. A pragmatic dialogue amongst stakeholders on the impact of COVID-19 on Irish cancer patients and healthcare services and lessons learned. Support Care Cancer. 2023; 32:19. 10.1007/s00520-023-08227-y. [DOI] [PubMed] [Google Scholar]
- 14.Jeon J, Park J, Choi MH, Choi H, Kim MH. Unequal burdens of COVID-19 infection: a nationwide cohort study of COVID-19-related health inequalities in Korea. Epidemiol Health. 2023; 45:e2023068. 10.4178/epih.e2023068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kind AJH, Buckingham WR. Making Neighborhood-Disadvantage Metrics Accessible - The Neighborhood Atlas. N Engl J Med. 2018; 378:2456–8. 10.1056/NEJMp1802313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tung EL, Peek ME, Rivas MA, Yang JP, Volerman A. Association Of Neighborhood Disadvantage With Racial Disparities In COVID-19 Positivity In Chicago. Health Aff (Millwood). 2021; 40:1784–91. 10.1377/hlthaff.2021.00695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hu J, Bartels CM, Rovin RA, Lamb LE, Kind AJH, Nerenz DR. Race, Ethnicity, Neighborhood Characteristics, and In-Hospital Coronavirus Disease-2019 Mortality. Med Care. 2021; 59:888–92. 10.1097/mlr.0000000000001624. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sáez-Clarke E, Otto AK, Prinsloo S, Natori A, Wagner RW, Gomez TI, et al. Development and initial psychometric evaluation of a COVID-related psychosocial experiences questionnaire for cancer survivors. Quality of Life Research (under review). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Otto AK, Prinsloo S, Natori A, Wagner RW, Gomez TI, Ochoa JM, et al. Impact of COVID-19-related experiences on health-related quality of life in cancer survivors. PLOS ONE (under review). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pearman T, Yanez B, Peipert J, Wortman K, Beaumont J, Cella D. Ambulatory cancer and US general population reference values and cutoff scores for the functional assessment of cancer therapy. Cancer. 2014; 120:2902–9. 10.1002/cncr.28758. [DOI] [PubMed] [Google Scholar]
- 21.Center for Disease Control and Prevention. What is GIS. 2019; Available from: https://www.cdc.gov/gis/what-is-gis.htm.
- 22.United States Census Bureau. Glossary. 2022. [January 24, 2023]; Available from: https://www.census.gov/programs-surveys/geography/about/glossary.html.
- 23.Center for Health Disparities Research University of Wisconsin School of Medicine and Public Health. Neighborhood Atlas. N/D [February 22, 2023]; Available from: https://www.neighborhoodatlas.medicine.wisc.edu/.
- 24.PolicyMap. PolicyMap Dig Deeper. 2022. [April 12, 2022]; Available from: https://www.policymap.com/.
- 25.Hurley SE, Saunders TM, Nivas R, Hertz A, Reynolds P. Post Office Box Addresses: A Challenge for Geographic Information System-Based Studies. Epidemiology. 2003; 14:386–91. [DOI] [PubMed] [Google Scholar]
- 26.Robinson JC, Wyatt SB, Hickson D, Gwinn D, Faruque F, Sims M, et al. Methods for retrospective geocoding in population studies: the Jackson Heart Study. J Urban Health. 2010; 87:136–50. 10.1007/s11524-009-9403-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.PolicyMap. Estimated median income of a household, between 2016–2020. 2022. [4/12/2022]; Available from: www.policymap.com.
- 28.Greater Houston Partnership. The 12 Counties of the Greater Houston Region. 2024. [January 18, 2024]; Available from: https://www.houston.org/why-houston/explore-regions.
- 29.Office of Management and Budget (OMB) Bulletins. July 2023. (OMB Bulletin No. 23 01). 2023 [February 2, 2024]; Available from: https://www.census.gov/programs-surveys/metro-micro/about/omb-bulletins.html.
- 30.Singh GK. Area deprivation and widening inequalities in US mortality, 1969–1998. Am J Public Health. 2003; 93:1137–43. 10.2105/ajph.93.7.1137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.United States Census Bureau. American Community Survey (ACS). N/D; Available from: https://www.census.gov/programs-surveys/acs.
- 32.Holman D, Bell A, Green M, Salway S. Neighbourhood deprivation and intersectional inequalities in biomarkers of healthy ageing in England. Health & Place. 2022; 77:102871. 10.1016/j.healthplace.2022.102871. [DOI] [PubMed] [Google Scholar]
- 33.Oluyomi A, Aldrich KD, Foster KL, Badr H, Kamdar KY, Scheurer ME, et al. Neighborhood deprivation index is associated with weight status among long-term survivors of childhood acute lymphoblastic leukemia. J Cancer Surviv. 2021; 15:767–75. 10.1007/s11764-020-00968-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Rosenzweig MQ, Althouse AD, Sabik L, Arnold R, Chu E, Smith TJ, et al. The Association Between Area Deprivation Index and Patient-Reported Outcomes in Patients with Advanced Cancer. Health Equity. 2021; 5:8–16. 10.1089/heq.2020.0037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Safavi AH, Lovas M, Liu ZA, Melwani S, Truong T, Devonish S, et al. Virtual Care and Electronic Patient Communication During COVID-19: Cross-sectional Study of Inequities Across a Canadian Tertiary Cancer Center. J Med Internet Res. 2022; 24:e39728. 10.2196/39728. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Worster B, Waldman L, Garber G, Zhan T, Lopez A, Trachtenberg O, et al. Increasing equitable access to telehealth oncology care in the COVID-19 National Emergency: Creation of a telehealth task force. Cancer Med. 2023; 12:2842–9. 10.1002/cam4.5176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ashcroft R, Saluja K, Ryan B, Mehta K, Brown J, Lam S, et al. Recommendations for virtual care in primary care practices: a survey of patients across Ontario, Canada. Ann Fam Med. 2022; 20. 10.1370/afm.20.s1.2947. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ellehuus C, Jeppesen SS, Eckhoff L, Vestergaard H, Therkildsen DN, Frederiksen H, et al. Health-related quality of life of patients with haematologic cancer during COVID-19 and their opinions on telehealth consultations - a Danish single site cross-sectional survey. Acta Oncol. 2021; 60:872–80. 10.1080/0284186x.2021.1902565. [DOI] [PubMed] [Google Scholar]
- 39.Hassan AM, Chu CK, Liu J, Angove R, Rocque G, Gallagher KD, et al. Determinants of telemedicine adoption among financially distressed patients with cancer during the COVID-19 pandemic: insights from a nationwide study. Support Care Cancer. 2022; 30:7665–78. 10.1007/s00520-022-07204-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Mora J, Krepline AN, Aldakkak M, Christians KK, George B, Hall WA, et al. Adjuvant therapy rates and overall survival in patients with localized pancreatic cancer from high Area Deprivation Index neighborhoods. The American Journal of Surgery. 2021; 222:10–7. 10.1016/j.amjsurg.2020.12.001. [DOI] [PubMed] [Google Scholar]
- 41.Wenger NS, Stanton AL, Baxter-King R, Sepucha K, Vavreck L, Naeim A. The Impact of COVID-19 on Routine Medical Care and Cancer Screening. J Gen Intern Med. 2022; 37:1450–6. 10.1007/s11606-021-07254-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Conley CC, Rodriguez JD, Brownstein NC, O’Neill SC, Vadaparampil ST. Characteristics associated with healthcare disruptions during the COVID-19 pandemic for women in the United States at high risk for breast cancer. Prev Med Rep. 2022; 30:101975. 10.1016/j.pmedr.2022.101975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Marmot MG, Smith GD, Stansfeld S, Patel C, North F, Head J, et al. Health inequalities among British civil servants: the Whitehall II study. Lancet. 1991; 337:1387–93. 10.1016/0140-6736(91)93068-k. [DOI] [PubMed] [Google Scholar]
- 44.Bronfenbrenner U, Ceci SJ. Heredity, environment, and the question “How?”: A first approximation. Nature, nurture & psychology. Washington, DC, US: American Psychological Association; 1993. p. 313–24. [Google Scholar]
- 45.Joanne P, Carol B, Helen A. Which patients miss appointments with general practice and the reasons why: a systematic review. British Journal of General Practice. 2021; 71:e406. 10.3399/BJGP.2020.1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.O’Keefe EB, Meltzer JP, Bethea TN. Health disparities and cancer: racial disparities in cancer mortality in the United States, 2000–2010. Front Public Health. 2015; 3:51. 10.3389/fpubh.2015.00051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Velasco-Mondragon E, Jimenez A, Palladino-Davis AG, Davis D, Escamilla-Cejudo JA. Hispanic health in the USA: a scoping review of the literature. Public Health Reviews. 2016; 37:31. 10.1186/s40985-016-0043-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Funk C, Lopez MH. Hispanic Americans’ trust in and engagement with science. Pew Research Center; 2022. [May 19, 2023]; Available from: https://www.pewresearch.org/science/2022/06/14/hispanic-americans-experiences-with-health-care. [Google Scholar]