Question  Did racial and ethnic minority adults with cancer in the United States experience more cancer care delays and adverse social and economic effects than White adults during the COVID-19 pandemic?

Findings  In this survey study of 1240 US adults with cancer, Black and Latinx adults reported experiencing higher rates of delayed cancer care and more adverse social and economic effects than White adults.

Meaning  This study suggests that the COVID-19 pandemic is associated with disparities in the receipt of timely cancer care among Black and Latinx adults.

Importance  The full effect of the COVID-19 pandemic on cancer care disparities, particularly by race and ethnicity, remains unknown.

Objectives  To assess whether the race and ethnicity of patients with cancer was associated with disparities in cancer treatment delays, adverse social and economic effects, and concerns during the COVID-19 pandemic and to evaluate trusted sources of COVID-19 information by race and ethnicity.

Design, Setting, and Participants  This national survey study of US adults with cancer compared treatment delays, adverse social and economic effects, concerns, and trusted sources of COVID-19 information by race and ethnicity from September 1, 2020, to January 12, 2021.

Exposures  The COVID-19 pandemic.

Main Outcomes and Measures  The primary outcome was delay in cancer treatment by race and ethnicity. Secondary outcomes were duration of delay, adverse social and economic effects, concerns, and trusted sources of COVID-19 information.

Results  Of 1639 invited respondents, 1240 participated (75.7% response rate) from 50 US states, the District of Columbia, and 5 US territories (744 female respondents [60.0%]; median age, 60 years [range, 24-92 years]; 266 African American or Black [hereafter referred to as Black] respondents [21.5%]; 186 Asian respondents [15.0%]; 232 Hispanic or Latinx [hereafter referred to as Latinx] respondents [18.7%]; 29 American Indian or Alaska Native, Native Hawaiian, or multiple races [hereafter referred to as other] respondents [2.3%]; and 527 White respondents [42.5%]). Compared with White respondents, Black respondents (odds ratio [OR], 6.13 [95% CI, 3.50-10.74]) and Latinx respondents (OR, 2.77 [95% CI, 1.49-5.14]) had greater odds of involuntary treatment delays, and Black respondents had greater odds of treatment delays greater than 4 weeks (OR, 3.13 [95% CI, 1.11-8.81]). Compared with White respondents, Black respondents (OR, 4.32 [95% CI, 2.65-7.04]) and Latinx respondents (OR, 6.13 [95% CI, 3.57-10.53]) had greater odds of food insecurity and concerns regarding food security (Black respondents: OR, 2.02 [95% CI, 1.34-3.04]; Latinx respondents: OR, 2.94 [95% CI, [1.86-4.66]), financial stability (Black respondents: OR, 3.56 [95% CI, 1.79-7.08]; Latinx respondents: OR, 4.29 [95% CI, 1.98-9.29]), and affordability of cancer treatment (Black respondents: OR, 4.27 [95% CI, 2.20-8.28]; Latinx respondents: OR, 2.81 [95% CI, 1.48-5.36]). Trusted sources of COVID-19 information varied significantly by race and ethnicity.

Conclusions and Relevance  In this survey of US adults with cancer, the COVID-19 pandemic was associated with treatment delay disparities and adverse social and economic effects among Black and Latinx adults. Partnering with trusted sources may be an opportunity to overcome such disparities.

The COVID-19 pandemic delayed cancer screening1,2 and surgical procedures3 worldwide, with unknown implications for cancer mortality rates. Before the COVID-19 pandemic, race and ethnicity–based disparities were prevalent in the United States, with lower rates of timely, evidence-based cancer care4-6 and higher premature death rates among Black and Latinx adults than other racial and ethnic groups.7,8 The COVID-19 pandemic highlighted the association of systemic racism and associated inequities in the structural, economic, and socioenvironmental system with these long-standing health disparities.9,10 However, the full effect of the COVID-19 pandemic on disparities in cancer care and cancer deaths remains unknown.11,12

Since March 2020, Black and Latinx adults have experienced the highest rates of excess morbidity and mortality from COVID-1913 and other causes14 as well as unemployment.15 Medical care disruptions, including cancer screening and treatment delays among Black and Latinx adults with cancer,16,17 and disparate adverse social and economic effects fuel concerns regarding the presence of more advanced cancer stages at diagnosis, avoidable cancer deaths,18-20 and widening of existing race and ethnicity–based disparities in cancer.

To our knowledge, this is the first national study evaluating the association of the COVID-19 pandemic with patient-reported cancer care delays, concerns, and adverse social economic effects by race and ethnicity. Specifically, we sought to evaluate whether racial and ethnic minority adults with cancer, compared with White adults with cancer, experienced more cancer care disruptions, concerns regarding health outcomes, adverse social and economic effects, and concerns regarding the association of COVID-19 with adverse social and economic factors. We also sought to evaluate trusted sources of COVID-19 information by race and ethnicity.

We conducted a cross-sectional survey study using a 74-question online survey in English, Spanish, Vietnamese, Chinese, and Hindi to assess patient-reported experiences from September 1, 2020, to January 12, 2021. The survey collected the following: (1) demographic characteristics (eg, gender identity, race and ethnicity, age, income, educational level, and insurance status), (2) clinical characteristics (eg, cancer diagnosis, stage, and treatment), (3) modifications in care, (4) adverse social and economic effects, (5) concerns regarding cancer and other health outcomes, (6) concerns regarding future adverse social and economic effects, and (7) trusted sources of COVID-19 information (eAppendix in the Supplement). Individuals 18 years of age or older with cancer who consented to the study procedures were invited to participate and received no compensation for their participation. Participants were provided with a consent form online and if they participated in the survey, they had consented to participation. Date of diagnosis was not a consideration for inclusion. All survey responses were anonymous unless participants voluntarily provided their name and contact information. The Stanford University School of Medicine institutional review board reviewed and approved the study. This study followed the American Association for Public Opinion Research (AAPOR) reporting guideline for survey studies.21

We used a variety of techniques to distribute the survey, including a virtual snowball sampling technique. We distributed the survey link directly to participants in collaboration with clinicians and through online listservs in collaboration with patient advocacy groups and organizations, including the American Society of Clinical Oncology, the Latino Cancer Institute, the Susan G. Komen Foundation, and the International Association for the Study of Lung Cancer, with an emphasis on recruitment of underrepresented racial and ethnic groups. We also promoted the survey on social media. The response rate was calculated among those who directly received the survey link.

We used multivariable logistic regression to estimate odds ratios (ORs) and 95% CIs for 21 factors associated with delays in cancer care, concerns regarding health outcomes, experiences and concerns regarding adverse social and economic effects, and trusted sources of COVID-19 information. We combined self-reported race and ethnicity to create categories of African American or Black (hereafter referred to as Black), Asian, Hispanic or Latinx (hereafter referred to as Latinx), other (comprised of American Indian or Alaska Native, Native Hawaiian, or multiple races), and White non-Latinx (hereafter referred to as White). We calculated all ORs relative to White participants. We adjusted models for a priori selected sociodemographic variables (gender identity, age, educational level, income, insurance status, and place of residence) and clinical variables (cancer diagnosis and stage) associated with delays in care.22,23

To account for respondents with unknown cancer stage, we imputed unknown values for cancer stage using multiple imputation24 over 100 simulated iterations based on the age, gender identity, place of residence, income, educational level, insurance, and cancer diagnosis of each respondent. In 2 sensitivity analyses, we coded unknown cancer stage as a separate category within the cancer stage variable and conducted a complete-case analysis among participants with a reported known cancer stage. All analyses were 2-sided and conducted using Stata, version 16 (StataCorp LLC), with P 10.1002/(SICI)1097-0142(19961001)78:73.0.CO;2-K PubMedGoogle Scholar32.Carey  P, Gjerdingen  DK.  Follow-up of abnormal Papanicolaou smears among women of different races.   J Fam Pract. 1993;37(6):583-587.PubMedGoogle Scholar33.Bleicher  RJ, Ruth  K, Sigurdson  ER,  et al.  Time to surgery and breast cancer survival in the United States.   JAMA Oncol. 2016;2(3):330-339. doi:10.1001/jamaoncol.2015.4508 PubMedGoogle Scholar34.Fedewa  SA, Edge  SB, Stewart  AK, Halpern  MT, Marlow  NM, Ward  EM.  Race and ethnicity are associated with delays in breast cancer treatment (2003-2006).   J Health Care Poor Underserved. 2011;22(1):128-141.PubMedGoogle Scholar35.Borrayo  EA, Scott  KL, Drennen  AR, MacDonald  T, Nguyen  J.  Determinants of treatment delays among underserved Hispanics with lung and head and neck cancers.   Cancer Control. 2016;23(4):390-400. doi:10.1177/107327481602300410 PubMedGoogle Scholar36.Bandinelli  L, Ornell  F, von Diemen  L, Kessler  FHP.  The sum of fears in cancer patients inside the context of the COVID-19.   Front Psychiatry. 2021;12:557834. doi:10.3389/fpsyt.2021.557834 PubMedGoogle Scholar37.Islam  JY, Vidot  DC, Camacho-Rivera  M.  Evaluating mental health–related symptoms among cancer survivors during the COVID-19 pandemic: an analysis of the COVID impact survey.   JCO Oncol Pract. 2021;17(9):e1258-e1269. doi:10.1200/OP.20.00752 PubMedGoogle Scholar38.Hamlish  T, Papautsky  EL.  Differences in emotional distress among Black and White breast cancer survivors during the COVID-19 pandemic: a national survey.   J Racial Ethn Health Disparities. 2021;9(2):576-580. doi:10.1007/s40615-021-00990-0PubMedGoogle Scholar39.Goldmann  E, Hagen  D, Khoury  EE, Owens  M, Misra  S, Thrul  J.  An examination of racial and ethnic disparities in mental health during the COVID-19 pandemic in the U.S. South.   J Affect Disord. 2021;295:471-478. doi:10.1016/j.jad.2021.08.047 PubMedGoogle Scholar40.Flowers  P, Riddell  J, Boydell  N, Teal  G, Coia  N, McDaid  L.  What are mass media interventions made of? exploring the active content of interventions designed to increase HIV testing in gay men within a systematic review.   Br J Health Psychol. 2019;24(3):704-737. doi:10.1111/bjhp.12377 PubMedGoogle Scholar41.Rubin  GJ, Amlôt  R, Page  L, Wessely  S.  Public perceptions, anxiety, and behaviour change in relation to the swine flu outbreak: cross sectional telephone survey.   BMJ. 2009;339:b2651-b2651. doi:10.1136/bmj.b2651 PubMedGoogle Scholar42.Armstrong  K, Ravenell  KL, McMurphy  S, Putt  M.  Racial/ethnic differences in physician distrust in the United States.   Am J Public Health. 2007;97(7):1283-1289. doi:10.2105/AJPH.2005.080762 PubMedGoogle Scholar43.Robinson  RV, Jackson  EF.  Is trust in others declining in America? an age–period–cohort analysis.   Soc Sci Res. 2001;30(1):117-145. doi:10.1006/ssre.2000.0692 Google Scholar44.Abdul-Mutakabbir  JC, Casey  S, Jews  V,  et al.  A three-tiered approach to address barriers to COVID-19 vaccine delivery in the Black community.   Lancet Glob Health. 2021;9(6):e749-e750. doi:10.1016/S2214-109X(21)00099-1 PubMedGoogle Scholar45.Hung  MC, Ekwueme  DU, Rim  SH, White  A.  Racial/ethnicity disparities in invasive breast cancer among younger and older women: an analysis using multiple measures of population health.   Cancer Epidemiol. 2016;45:112-118. doi:10.1016/j.canep.2016.10.013 PubMedGoogle Scholar46.Pinheiro  PS, Callahan  KE, Siegel  RL,  et al.  Cancer mortality in Hispanic ethnic groups.   Cancer Epidemiol Biomarkers Prev. 2017;26(3):376-382. doi:10.1158/1055-9965.EPI-16-0684 PubMedGoogle Scholar47.Merrill  RM, Sloan  A, Anderson  AE, Ryker  K.  Unstaged cancer in the United States: a population-based study.   BMC Cancer. 2011;11:402-402. doi:10.1186/1471-2407-11-402 PubMedGoogle Scholar48.Adams  J, Audisio  RA, White  M, Forman  D.  Age-related variations in progression of cancer at diagnosis and completeness of cancer registry data.   Surg Oncol. 2004;13(4):175-179. doi:10.1016/j.suronc.2004.08.007 PubMedGoogle Scholar49.Worthington  JL, Koroukian  SM, Cooper  GS.  Examining the characteristics of unstaged colon and rectal cancer cases.   Cancer Detect Prev. 2008;32(3):251-258. doi:10.1016/j.cdp.2008.08.006 PubMedGoogle Scholar50.Bhaskaran  K, Smeeth  L.  What is the difference between missing completely at random and missing at random?   Int J Epidemiol. 2014;43(4):1336-1339. doi:10.1093/ije/dyu080 PubMedGoogle Scholar51.Sterne  JA, White  IR, Carlin  JB,  et al.  Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls.   BMJ. 2009;338:b2393. doi:10.1136/bmj.b2393 PubMedGoogle Scholar52.O’Kelly  M.  Multiple Imputation and Its Application. Carpenter, James and Kenward, Michael (2013). Chichester: John Wiley & Sons. 345 pages, ISBN: 9780470740521.   Biom J. 2014;56(2):352-353. doi:10.1002/bimj.201300188 Google Scholar53.Toutenburg H, Rubin DB.  Multiple imputation for nonresponse in surveys.   Stat Hefte. 1990;31(1):180. Google Scholar54.Greenland  S, Finkle  WD.  A critical look at methods for handling missing covariates in epidemiologic regression analyses.   Am J Epidemiol. 1995;142(12):1255-1264. doi:10.1093/oxfordjournals.aje.a117592 PubMedGoogle Scholar