Article Tools
ORIGINAL REPORTS
Special Series: COVID-19 and Cancer Care
Article Tools
SARS-CoV-2 Testing, Positivity Rates, and Healthcare Outcomes in a Cohort of 22,481 Breast Cancer Survivors
Abstract
As health inequities during the pandemic have been magnified, we evaluated how use of SARS-CoV-2 testing differed by race or ethnicity in a large cohort of breast cancer survivors and examined the correlates of testing positive.
We conducted a retrospective cohort study of 22,481 adult breast cancer survivors who were active members of a large California integrated healthcare plan in 2020. We collected data on their breast cancer diagnosis, comorbidity, and demographic characteristics. We examined SARS-CoV-2 testing utilization between March 2020 and September 2020 by race or ethnicity, comorbidity, and other patient characteristics. We also examined the correlates of a having a positive SARS-CoV-2 test result. We conducted bivariable and multivariable logistic regression to identify correlates of testing utilization and test positivity.
Of these 22,481 women, 3,288 (14.6%) underwent SARS-CoV-2 testing. The cohort included 51.8% women of color. Of the 3,288 tested, 264 (8.0%) women had a positive test result. In multivariable analyses, Latinx survivors were more likely (adjusted odds ratio [OR], 1.23; 95% CI, 1.12 to 1.34) to undergo testing than White survivors; however, Asian or Pacific Islander survivors were 16% less likely to get tested (adjusted OR, 0.84; 95% CI, 0.75 to 0.94). Compared to White survivors, Latinx survivors were 3.5 times (adjusted OR, 3.47; 95% CI, 2.52 to 4.77) and Asian or Pacific Islander or Other survivors were 2.2-fold (adjusted OR, 2.23; 95% CI, 1.49 to 3.34) more likely to test positive. Being overweight (adjusted OR, 1.83; 95% CI, 1.24 to 2.72) or obese (adjusted OR, 2.04; 95% CI, 1.39 to 2.98) were also strongly associated with SARS-CoV-2 positivity.
Even in an integrated healthcare system, Asian or Pacific Islander patients were less likely to undergo SARS-CoV-2 testing than White survivors, but more likely to test positive. Additionally, Latinx ethnicity and high body mass index were strongly correlated with a greater odds of SARS-CoV-2 test positivity.
Cancer survivors represent a susceptible group for COVID-19 caused by the SARS-CoV-2 virus, given their potential immunocompromised status1-3; however, little is known about SARS-CoV-2 testing utilization, infection rates, and healthcare outcomes, for example, hospitalization, admission into intensive care units (ICU), and emergency department visits in such survivors. In parallel, health inequities in the United States during the pandemic have been magnified. For example, the Centers for Disease Control4 have identified that factors that enhance risk of SARS-CoV-2 infection and death in racial or ethnic minority groups include reduced healthcare access,5,6 providers' implicit bias,7-11 and socioeconomic factors such as lower income and education.12 Because many cancer outcome studies included patients with and without health insurance, it is difficult to disentangle the effects of insurance coverage from other social determinants of health.13-15 Thus, our goal was to examine whether disparities exist in SARS-CoV-2 testing as well as in SARS-CoV-2 test positivity in an insured group of breast cancer survivors.
Key Objective
To determine disparities in SARS-CoV-2 testing utilization in 22,481 diverse breast cancer survivors in a large integrated healthcare system and to identify correlates of test positivity.
Knowledge Generated
Of these 22,481 survivors, 3,288 patients (14.6%) underwent SARS-CoV-2 testing, and 264 (8.0%) tested positive. African American and Latinx survivors were more likely to undergo testing than White survivors. Although Asian or Pacific Islander survivors were 16% less likely to get tested, they were 2.2-fold more likely to test positive than White survivors. Being overweight or obese was also strongly associated with SARS-CoV-2 positivity.
Relevance
Programs to increase SARS-CoV-2 testing in Asian or Pacific Islander breast cancer survivors should be implemented as they had a markedly increased occurrence of positive tests than White survivors, but they were less likely to undergo testing. Furthermore, Latinx survivors and those with higher body mass index were dramatically more likely to have a positive test result, suggesting that these survivors might be more susceptible SARS-CoV-2 infection.
We assembled a cohort of 22,481 female breast cancer survivors (18+ years, diagnosed since 2009 with stage 0-IV disease) who were active members of a not-for-profit integrated healthcare system, Kaiser Permanente Southern California (KPSC), in 2020. KPSC serves 4.7 million patients and contributes data to the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program. We examined the SARS-CoV-2 testing use between March 15, 2020, and September 30, 2020. All data elements were extracted from electronic health records, including polymerase chain reaction-based SARS-CoV-2 tests (nasal swab collection, 97% of the tests), comorbidities, healthcare utilization, and deaths. Data on cancer stage, primary cancer treatments, and race or ethnicity were captured from the health plan’s SEER-affiliated cancer registry. The primary study outcomes were (1) receipt of SARS-CoV-2 test and (2) positive SARS-CoV-2 tests as identified from laboratory results and confirmed with International Classification of Diseases (ICD)-10-CM diagnosis codes of SARS-CoV-2 (ICD-10-CM: B34.2, B97.29, J12.89, J20.8, J22, J80, and U07.1). For secondary study outcomes, we captured hospitalization, ICU admissions, mechanical ventilation during hospitalizations, and all-cause deaths in patients who underwent SARS-CoV-2 testing. Correlates associated with SARS-CoV-2 testing we extracted included race or ethnicity; age at test date; tumor factors (cancer stage, survivorship length, and adjuvant therapies); geocoded education and median household income based on the census block level; body mass index (BMI); smoking history; and comorbidities in the past year using the Charlson Comorbidity Index (CCI).16 For those who did not undergo SARS-CoV-2 testing, we used January 1, 2020, as the index date for capturing these variables. We calculated overall and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) to identify the correlates associated with undergoing SARS-CoV-2 testing using logistic regression. Multivariable logistic regression models were partially adjusted for age in 2020; stage of initial cancer diagnosis; and race or ethnicity, except for the model in which we examined race or ethnicity as the main independent variable in which we adjusted for age at testing and stage. Similarly, in the model in which we examined the correlation between age and receipt of testing, we adjusted for race or ethnicity and stage. In the model that examined the association between cancer stage and receipt of testing, we adjusted for race or ethnicity and age at testing. In the 3,288 patients who underwent SARS-CoV-2 testing, we determined the correlates of a positive result using the same analytic methods. In the analyses for secondary outcomes of healthcare outcomes (hospitalizations, ICU admissions, mechanical ventilation, and all-cause deaths), we examined frequencies and percent distribution of these events.
Table 1 displays the patient and clinical characteristics and their association with receipt of SARS-CoV-2 testing. The cohort of 22,481 breast cancer survivors included 11,641 (51.8%) women of color. Of these 22,481 survivors, 3,288 women (14.6%) underwent SARS-CoV-2 testing. Odds of undergoing SARS-CoV-2 testing was 17%-30% lower in older survivors ≥ 50 years of age compared with younger survivors < 50 years of age. Latinx survivors were 23% more likely to undergo testing (adjusted odds ratio [OR], 1.23; 95% CI, 1.12 to 1.34) compared with White survivors survivors (Table 1). Odds of undergoing testing was similar in African American or Black survivors (adjusted OR, 1.07; 95% CI, 0.95 to 1.21) as for White survivors; however, breast cancer survivors in the Asian or Pacific Islander or Other category were 16% less likely to undergo SARS-CoV-2 testing (adjusted OR, 0.84; 95% CI, 0.75 to 0.94). Of note, the most common ethnicity in the Asian or Pacific Islander or Other group was Filipino and Chinese, and the percentage of women in the Other category was small (0.92%) in the full cohort (data not shown). Survivors who were former smokers were 30% more likely to get tested (adjusted OR, 1.29; 95% CI, 1.18 to 1.41) versus never smokers. Survivors with more comorbidities were two-fold more likely to undergo testing (adjusted OR, 1.98; 95% CI, 1.80 to 2.18 for CCI > 3). Specifically, survivors with hypertension (adjusted OR, 1.12; 95% CI, 1.03 to 1.21) and diabetes (adjusted OR, 1.33; 95% CI, 1.22 to 1.45) were more likely to undergo SARS-CoV-2 testing than those without such conditions. Women who were initially diagnosed with stage III-IV breast cancer were 16% more likely (adjusted OR, 1.16; 95% CI, 1.01 to 1.32) to undergo testing than those diagnosed with stage 0-II disease. Compared to recent survivors (< 3 years), longer-term survivors (> 3-5, > 5 years) were two-fold more likely to get tested (adjusted OR, 1.92; 95% CI, 1.67 to 2.21 and adjusted OR, 1.92; 95% CI, 1.68 to 2.19, respectively).
|
Table 2 shows the variables correlated with a positive SARS-CoV-2 test result. Of the 3,288 breast cancer survivors tested, 264 (8.0%) had a positive test result. Compared to White survivors, Latinx survivors were nearly 3.5-times (adjusted OR, 3.47; 95% CI, 2.52 to 4.77) and Asian or Pacific Islander or Other survivors were 2.2-times more likely (adjusted OR, 2.23; 95% CI, 1.49 to 3.34) to test positive (Table 2, Fig 1). Higher BMIs (overweight: adjusted OR, 1.83; 95% CI, 1.24 to 2.72 and overweight: adjusted OR, 2.04; 95% CI, 1.39 to 2.98) were dramatically associated with SARS-CoV-2 positivity as compared to having healthy BMI. Geocoded education and income were not associated with SARS-CoV-2 positivity. The odds of SARS-CoV-2 positivity were also lower in breast cancer survivors who were former smokers; however, their smoking habits reflected tobacco use many years prior to breast cancer diagnosis.
|
FIG 1. Forest plots of adjusted odds ratios for SARS-CoV-2 test utilization and SARS-CoV-2 positivity.
Table 3 shows the secondary healthcare outcomes we examined between March 15, 2020, and September 30, 2020, in the 3,288 patients who underwent SARS-CoV-2 testing: hospitalizations, ICU admissions, mechanical ventilation during hospitalization, and all-cause deaths. Of these 3,288 breast cancer survivors, 564 (17.2%) women were hospitalized; some were hospitalized multiple times for a total count of 799 hospitalizations among these 564 women. During hospitalization, 72 survivors (2.2%) underwent mechanical ventilation. Additionally, 80 survivors (2.4%) were admitted into the ICU; the total number of ICU admissions during the study period was 167. A total of 43 survivors (1.3%) died in hospital through September 30, 2020, whereas an additional 80 women (2.4%) died 30-days post-discharge, reflecting an overall 3.7% risk of death in the 3,288 patients who underwent SARS-CoV-2 testing. We did not find any differences in hospitalizations or deaths by race or ethnicity.
|
In this large diverse cohort of 22,481 breast cancer survivors, we observed that Latinx women were more likely to get SARS-CoV-2 testing, whereas Asian or Pacific Islander women were less likely to undergo testing compared with their White survivors counterparts. Other factors associated with SARS-CoV-2 testing use were younger age (possibly because older women were encouraged to use telemedicine); being a former smoker; having hypertension or diabetes; longer cancer survivorship length; and having had been initially diagnosed stage III-IV breast cancer. Among the 3,288 women who underwent testing, the percent positivity was 8.0%. Although Asian or Pacific Islanders survivors were 2.2-fold more likely to have a positive result (adjusted OR, 2.23; 95% CI, 1.49 to 3.34), their likelihood of getting tested was 16% lower than for White survivors (adjusted OR, 0.84; 95% CI, 0.75 to 94, Fig 1) even in this integrated healthcare system in which financial barriers are mitigated. Reasons for Asian or Pacific Islander patients undergoing less testing in this equal-access integrated healthcare system is not clear, but may be partly attributed to the of fear of discrimination during the pandemic as reported in a recent study of Chinese Americans.17 Our data suggest that programs to increase SARS-CoV-2 testing in Asian or Pacific Islander women should be implemented as they had a markedly increased occurrence of positive tests than White survivors women; this may help mitigate the health inequities seen in this group. Furthermore, we found that breast cancer survivors with higher BMI were markedly more likely to have a positive test result, suggesting that these survivors might be more susceptible to SARS-CoV-2 infection compared to those with healthy BMI. Additionally, the overall prevalence of testing utilization (14.6%) in the full cohort was low possibly because of the health plan’s transitioning follow-up cancer survivorship care to telehealth visits at the beginning of the pandemic.
Strengths of this study include a large cohort of breast cancer survivors with data on SARS-CoV-2 testing use. Data were extracted from one of largest integrated healthcare systems in California where patients get virtually all their care from within this system, and this enabled us to comprehensively identify correlates of SARS-CoV-2 positivity and healthcare outcomes using a combination of procedure data and diagnostic codes. Certain limitations must also be considered. Although we found that Latinx and Asian or Pacific Islanders breast cancer survivors were nearly 2.2- to 3.5-fold were more likely, respectively, to have a positive SARS-CoV-2 result than White survivors, we could not determine the reasons for this. For example, they might have been more recent immigrants who live in multifamily dwellings or could not work from home during the pandemic. To better understand this, patient surveys need to be conducted to better understand the role of other social determinants of health. In addition, although breast cancer survivors with hypertension and diabetes were somewhat more likely to have a positive test result, these results were not statistically significant possibly because of the low prevalence of these conditions in the cohort. Furthermore, we did not find disparities the secondary study healthcare outcomes (hospitalizations, ICU admission, use of mechanical ventilation, and deaths) by race or ethnicity, possibly because of small numbers of these events that occurred during the study period. Even larger cohorts with longer follow-up will be needed to robustly assess this. Also, regarding identification of deaths as secondary outcomes, some might have occurred outside the hospital; however, we also reviewed the membership files to capture these events. Nonetheless, deaths might have been underestimated.
In summary, this retrospective cohort study of 22,481 breast cancer survivors determined that certain racial or ethnic groups, such as Asian or Pacific Islander patients, were less likely to undergo SARS-CoV-2 testing than White survivors, but more likely to test positive. Additionally, Latinx women were more likely to test positive for SARS-CoV-2. Overall, among the 3,288 breast cancer survivors who underwent SARS-CoV-2 testing, 17.2% of these women were hospitalized, but we did not find differences in the healthcare outcomes (hospitalizations, ICU admissions, mechanical ventilation, and deaths) by race or ethnicity, possibly because of the small numbers of such events. Furthermore, breast cancer survivors with higher BMIs may be more susceptible to SARS-CoV-2, given their greater odds of positive results.
Supported by Kaiser Permanente Southern California, a not-for-profit integrated healthcare system. The funder had no role in the study design, analysis, data interpretation, manuscript writing, or decision to submit for publication.
The deidentified data underlying this article will be shared on reasonable request to the corresponding author pending data use agreements and approval by the KPSC IRB.
Conception and design: Reina Haque
Financial support: Reina Haque
Administrative support: Reina Haque
Provision of study materials or patients: Reina Haque
Collection and assembly of data: All authors
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/cci/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
Research Funding: Puma Biotechnology
No other potential conflicts of interest were reported.
| 1. | Zhang H, Han H, He T, et al: Clinical characteristics and outcomes of COVID-19 infected cancer patients: A systematic review and meta-analysis. J Natl Cancer Inst 10.1093/jnci/djaa168 [epub ahead of print on November 2, 2020] Google Scholar |
| 2. | Liang W, Guan W, Chen R, et al: Cancer patients in SARS-CoV-2 infection: A nationwide analysis in China. Lancet Oncol 21:335-337, 2020 Crossref, Medline, Google Scholar |
| 3. | Yu J, Ouyang W, Chua MLK, et al: SARS-CoV-2 Transmission in patients with cancer at a tertiary care hospital in Wuhan, China. JAMA Oncol 6:1108-1110, 2020 Crossref, Medline, Google Scholar |
| 4. | Centers for Disease Control: Health Equity Considerations and Racial and Ethnic Minority Groups. https://www.cdc.gov/coronavirus/2019-ncov/community/health-equity/race-ethnicity.html Google Scholar |
| 5. | Berchick ER, Barnett JC, Upton RD: Current Population Reports, P60-267(RV), Health Insurance Coverage in the United States: 2018. Washington, DC, US Government Printing Office, 2019 Google Scholar |
| 6. | Institute of Medicine (US): Committee on the Consequences of Uninsurance: Care Without Coverage: Too Little, Too Late. Washington, DC, National Academies Press (US), 2002. https://doi.org/10.17226/10367externalicon Google Scholar |
| 7. | Price-Haygood EG, Burton J, Fort D, et al. Hospitalization and mortality among black patients and white patients with SARS-CoV-2. N Engl J Med 382:2534-2543, 2020 Crossref, Medline, Google Scholar |
| 8. | Millet GA, Jones AT, Benkeser D, et al: Assessing differential impacts of SARS-CoV-2 on black communities. Ann Epidemiol 47:37-44, 2020 Crossref, Medline, Google Scholar |
| 9. | Paradies Y: A systematic review of empirical research on self-reported racism and health. Int J Epidemiol 35:888-901, 2006 Crossref, Medline, Google Scholar |
| 10. | Simons RL, Lei MK, Beach SRH, et al: Discrimination, segregation, and chronic inflammation: Testing the weathering explanation for the poor health of Black Americans. Dev Psychol 54:1993-2006, 2018 Crossref, Medline, Google Scholar |
| 11. | Cordes J, Castro MC: Spatial analysis of SARS-CV-2 clusters and contextual factors in New York City. Spat Spatiotemporal Epidemiol 34:100355, 2020 Crossref, Medline, Google Scholar |
| 12. | The Annie E. Casey Foundation: Unequal Opportunities in Education [online]. 2006. https://www.aecf.org/m/resourcedoc/aecf-racemattersEDUCATION-2006.pdf Google Scholar |
| 13. | Institute of Medicine, Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care, Smedley BD, Stith AY, Nelson AR (eds): Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC, National Academies Press (US), 2003 Google Scholar |
| 14. | Ward E, Jemal A, Cokkinides V, et al: Cancer disparities by race/ethnicity and socioeconomic status. CA Cancer J Clin 54:78-93, 2004 Crossref, Medline, Google Scholar |
| 15. | Ward E, Halpern M, Schrag N, et al: Association of insurance with cancer care utilization and outcomes. CA Cancer J Clin 58:9-31, 2008 Crossref, Medline, Google Scholar |
| 16. | Quan H, Sundararajan V, Halfon P, et al: Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10-CM administrative data. Med Care 43:1130-1139, 2005 Crossref, Medline, Google Scholar |
| 17. | Cheah CSL, Wang C, Ren H, et al: COVID-19 racism and mental health in Chinese American families. Pediatrics 146:e2020021816, 2020 Crossref, Medline, Google Scholar |
