key: cord-1044165-byelk9m4
authors: Yang, Xueying; Zhang, Jiajia; Chen, Shujie; Olatosi, Bankole; Bruner, Larisa; Diedhiou, Abdoulaye; Scott, Cheryl; Mansaray, Ali; Weissman, Sharon; Li, Xiaoming
title: Demographic disparities in clinical outcomes of COVID-19: data from a statewide cohort in South Carolina
date: 2021-08-13
journal: Open Forum Infect Dis
DOI: 10.1093/ofid/ofab428
sha: db94cfe7efe068bc38a7a05cf2f7c1fc475f50c0
doc_id: 1044165
cord_uid: byelk9m4

BACKGROUND: Current literature examining the clinical characteristics of COVID-19 patients under-represent COVID-19 cases who were either asymptomatic or had mild symtoms. METHODS: We analyzed statewide data from 280,177 COVID-19 cases from various health care facilities during March 04–December 31, 2020. Each COVID-19 case was reported using the standardized Case Report Form (CRF), which collected information on demographic characteristics, symptoms, hospitalization, and death. We used multivariable logistic regression to analyze the associations between socio-demographics and disease severity, hospitalization and mortality. RESULTS: Among a total of 280,177 COVID-19 cases, 5.2% (14,451) were hospitalized and 1.9% (5,308) died. Older adults, males, and Black individuals had higher odds for hospitalization and death from COVID-19 (Ps<0.0001). Particularly, individuals residing in rural areas experienced a high risk of death (OR: 1.16; 95%CI: 1.08, 1.25). Regarding disease severity, older adults (OR: 1.06, 95%CI: 1.03, 1.10) and Hispanic or Latino patients (OR: 2.06; 95%CI: 1.95, 2.18), had higher odds of experiencing moderate/severe symptoms, while male and Asian patients, compared to White patients, had lower odds of experiencing moderate/severe symptoms. CONCLUSIONS: As the first statewide population-based study using data from multiple healthcare systems with a long follow-up period in the US, we provide a more generalizable picture of COVID-19 symptoms and clinical outcomes. The findings from this study reinforce the fact that rural residence and racial/ethnic social determinants of health, unfortunately, remain predictors of adverse health outcomes for COVID-19 patients.

Since the first confirmed case of coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus (SARS-CoV-2) in the United States (US) on January 21, 2020, outbreaks of COVID-19 have surged quickly. The US is among the countries that were hit the hardest by the pandemic 1 . As of July 19, 2021, there were over 34 million COVID-19 cases with 612,105 deaths in the US. South Carolina, a predominately rural state with significant health care shortage regions, has reported 614,912 COVID-19 cases and 9,904 deaths as of July 29 2 , 2021.

The clinical spectrum of SARS-CoV-2 infection ranges from asymptomatic to life-threatening and death.

Studies show that the presentation of symptoms plays a fundamental role in informing the disease severity after COVID-19 diagnosis. Based on existing research, most individuals infected with SARS-CoV-2 are asymptomatic (around 40%-45% 3 ) or experience mild to moderate symptoms 1 . About 14% of all cases become severe, and 5% critical 4 . Ongoing research continues to investigate clinical outcomes of COVID-19 patients in the US. However, a large heterogeneity existed between these studies because of differences in clinical settings, sample selection methods and statistical plans, which limits the generalizability of these findings.

Characteristics and clinical outcomes of COVID-19 patients has been frequently reported in existing literature, but the data are not necessarily representative of the full spectrum of disease. Most of these studies revealed an increasing evidence that some racial and ethnic minority groups (e.g., Black, Hispanic or Latino) are overrepresented in COVID-19 cases [5] [6] [7] [8] [9] and reported a disproportionate burden of hospitalizations, but not necessarily critical illness or death 10 , in these groups 11, 12 . Despite an increasing body of US studies investigating clinical outcomes of COVID-19 patients, including hospitalization [6] [7] [8] 10, [13] [14] [15] [16] [17] , mortality 8, 10, [13] [14] [15] [16] 18 , and intensive care unit (ICU) admission 16, 18 , several knowledge gaps persist in existing research. First, the majority of patient samples were restricted to hospitalized COVID-19 cases or identified from a single health care system, which might be less representative to the majority of the outpatient COVID-19 population who are either asymptomatic or had mild illness. Second, most of these studies collected data from a city-level or hospital-level with relatively small sample sizes (305 to 78,323) and a short observational study period (1-2 months), further restricting the generalization of the findings to all COVD-19 populations. Third, presenting symptoms of COVID-19 patients, particularly nonhospitalized cases, were not previously investigated extensively. We proposed to address these gaps by using data from a population-based statewide cohort which included all adult confirmed and probable A c c e p t e d M a n u s c r i p t 5 COVID-19 cases in SC between March 04, 2020 and December 31, 2021. The present study analyzed sociodemographic characteristics, disease severity and clinical outcomes of COVID-19 patients, including hospitalization and mortality.

Data for this study were derived from the SC statewide Case Report Form (CRF) ( March 04, 2020 and December 31, 2020 were included in the current study. SC Law (44-29-10) and

Regulations (61-20) require mandatory reporting of COVID-19 to DHEC 20 . Where available, reports must include information about diagnoses, and results of specific diagnostic tests. The sources of data include clinician reporting, laboratory reporting, reporting by other entities (e.g., hospitals, veterinarians, pharmacies, poison centers), death certificates, hospital discharge, or outpatient records 19 . The criteria of case ascertainment were described in the standardized surveillance case definition of COVID-19 19 .

The study protocol received approval from the institutional review board at the University of South Carolina and relevant SC state agencies.

Information on social demographics included age (e.g., 40-49, 50-59 years old), gender (e.g., female, male, transgender), race (e.g., White, Black, Asian), ethnicity (e.g., Hispanic/Latino, non-Hispanic/Latino), and residential status. Residential status was defined according to the Rural-Urban Commuting Area (RUCA) codes, as urban areas (i.e., metropolitan) or rural areas (i.e., micropolitan, rural and small-town areas) 21 .

A c c e p t e d M a n u s c r i p t 6 

Clinical course information included symptom category during the onset of illness (i.e., symptomatic, asymptomatic, unknown), and development of pneumonia and acute respiratory distress syndrome (ARDS). For symptomatic patients, the CRF documented specific symptoms that were experienced during the illness, such as fever, chills, muscle aches, runny nose, sore throat, new olfactory and taste disorders, headache, fatigue, cough, difficulty breathing, nausea or vomiting, abdominal pain, and diarrhea ( Figure 1 ).

We analyzed three distinct outcomes: disease severity, hospitalization, and mortality of COVID-19.

Individuals were asked to indicate any of the symptoms (e.g., fever, headache, fatigue) specified on the CRF form. Each symptom had three responses, i.e., 'Yes', 'No', 'Unknown'. Based on different presenting symptoms of COVID-19 patients, disease severity was categorized into three groups. Specifically, COVID-19 patients with no symptoms were categorized as asymptomatic; individuals who have any of the various mild signs and symptoms of COVID-19 (e.g., fever, cough, sore throat, malaise, headache, muscle pain, nausea, vomiting, diarrhea, loss of taste and smell) were categorized as mild; whereas COVID-19 patients with difficulty breathing or had developed pneumonia or ARDS were categorized as moderate/severe. Because we do not have other clinical indicators to separate moderate from severe illness, we combined them together. In the CRF, hospitalization was measured with one question, i.e., "Was the patient hospitalized?", with the responses categorized as 'Yes', 'No', and 'Unknown'. Patients with no response to this question were treated as unknown. We then dichotomized the hospitalization status as 1 if the response is 'Yes', and 0 to indicate no hospital admission. Similarly, death was measured using the question, i.e., "Did the patient die as a result of this illness?", with the response categories as 'Yes", "No' and 'Unknown'. We used a similar strategy to define patient's death status where 1 indicates the death and 0 indicates no death, which includes living and unknown status.

Descriptive statistics were used to characterize the disease severity and clinical outcomes for COVID-19 cases. We used Chi-square test to compare differences between groups. We used logistic regression models to explore the association between socio-demographic characteristics and symptom severity (multinomial), hospitalization and death in COVID-19 cases. We reported A c c e p t e d M a n u s c r i p t 7 odds ratios (OR) and 95% confidence intervals (95% CI) for each model in tables and forest plots. The P-value <0.05 was considered statistically significant. All statistical analyses were performed using SAS software version 9.4 (SAS Institute, Inc., Cary, NC) and R software (version 3.6.2). Table 1 shows baseline characteristics of the population, comparing asymptomatic, mild and moderate/severe cases. Figure 1 illustrates the frequency of symptoms, with cough, headache, myalgia, taste loss, and fever as the most common ones.

Hospitalization occurred in 5.2% (n=14,451) of the COVID-19 cases and 1.9% (n=5,308) died from COVID-19 (Table 2) . Table 2 

For COVID-19 disease severity, individuals who were older (65+) Figure 1 ).

To the best of our knowledge, this is the first statewide population-based US study investigating COVID-19 disease severity and outcomes. This study provides robust evidence-based demographic disparities in COVID-19 disease severity, hospitalization, and mortality, particularly the rural-urban disparities in COVID-19 mortality, which has rarely been reported in prior studies. Previous studies focused more on the attributes of individuals who were more or less vulnerable (e.g., race, gender) and paid little attention to the distinctive prospects for urban and rural areas 22 . In our study, people living in rural areas had higher risk of death, yet they did not require more hospitalization than urban residents. It is possible that rural residents were not hospitalized due to the less health care accessibility, as is often the case. However, they risk presenting with higher disease severity later on, and are thus more likely to die due to less access to preventive therapies, such as monoclonal antibody therapy, which could slow the progression of disease. Another possible explanation is that the rural population might have more baseline comorbidities (e.g., hypertension, diabetes), which greatly increase the risk of adverse COVID-19 outcomes. These findings inform us that the uniform state policies and actions might be insensitive to vulnerable rural areas and can have unintended consequences in amplifying inequalities. National programs need concerted efforts to improve the social and economic status in lagging areas.

Studies show that asymptomatic COVID-19 infected persons play a significant role in active transmission.

According to the CDC's best estimate, the infectiousness of asymptomatic individuals relative to symptomatic is 75% 23 . Thus, asymptomatic cases "substantially contribute to community transmission, making up at least 50% of the driving force" of COVID-19 infections 3 . However, determining the actual number of asymptomatic COVID-19 cases has been a significant challenge for researchers and public health officials. The proportion of asymptomatic cases in the present study is comparable with a narrative review 4 but higher than the best estimate (30%) reported by the CDC 23 . The data sources in our study incorporate statewide data about daily testing capacity and changes in testing rates over time, so we believe the findings from this study can provide a more accurate estimation of the proportion of COVID-19 infections that are asymptomatic. Given the large proportion of asymptomatic cases, it is crucial that everyone including individuals who do not show symptoms, adhere to public health guidelines, such as mask wearing and social distancing.

A c c e p t e d M a n u s c r i p t 9 Consistent with previous studies, older age was associated with hospitalization and mortality. 6, 17, 24 Older patients were less likely to display mild symptoms but more moderate/severe symptoms of COVID-19.

Others have reported that older individuals with COVID-19 often present with non-specific and atypical symptoms such as delirium, postural instability or diarrhea [25] [26] [27] , rather than typical respiratory symptoms and fever. According to a meta-analysis, gastrointestinal symptoms including vomiting and diarrhea are strong predictors of developing severe COVID-19 illness 28 . Some of the atypical symptoms (e.g., delirium) that occurred in older adults might not be measured in our study. That may partially explain why older adults with fewer common symptoms are not necessarily less likely to develop severe symptoms. Therefor, we must prioritize the needs of older adults in the response to the COVID-19 pandemic.

While there is not a significant difference in the proportion of male and female COVID-19 cases, a disparity in hospitalization and death was observed. According to a recent meta-analysis using data from 46 countries and 44 US states, the gender differences observed in COVID-19 is a worldwide phenomenon with men being more likely to die or require intensive care unit (ICU) admission for COVID-19 29 . The driving factors behind the gender difference may include the fundamental differences in the immune response (e.g., CD4+ T cells, type 1 interferon, estrogen level), 30, 31 or cultural and behavioral differences (e.g., smoking, handwashing frequency) 32, 33 . Therefore, an appreciation of how gender influences COVID-19 outcomes will have important implications for clinical management and mitigation strategies for this disease.

The COVID-19 pandemic has highlighted persistent racial-/ethnic-health disparities in the US. Black, Asian, and Minority Ethnic groups (termed as 'BAME' in the UK 34 ) are overrepresented among cases of hospitalization and deaths from COVID-19 in the present study. This is consistent with findings from both national and international studies 7, 35 . Of note, the crude risks of COVID-19 hospitalization and death among Hispanic or Latinos were lower than non-Hispanic or Latinos. However, such differences could be explained by the differences in their demographic distributions. Hispanic or Latinos comprised of a higher proportion of younger populations than their counterpart (75.6% vs 54.0%) (Supplementary Table S1 ) and a significant interaction effect of age and ethnicity in affecting hospitalization and death were detected (Supplementary Table S2 ). This result could explain the inverse association (positive) between Hispanic or Latinos and COVID-19 hospitalization after adjusting the other variables. The likely causes of racial-/ethnic-health disparities in COVID-19 outcomes were discussed extensively in prior studies 8, 10, 13, 15, 34, [36] [37] [38] [39] [40] [41] . Some literature argued that minority communities may be more susceptible to A c c e p t e d M a n u s c r i p t 10 severe complications of COVID-19 because of existing disparities in underlying conditions known to be associated with COVID-19 mortality, including hypertension, cardiovascular disease, kidney disease, and diabetes. This might also support the assertion that existing structural determinants (e.g., housing, economic stability, and work circumstances) pervasive in Black and Hispanic communities may explain the disproportionately higher out-of-hospital deaths due to COVID-19 infections in these populations. It might indicate that Black individuals are less likely to be identified in the outpatient setting, potentially reflecting differences in health care access or utilization. Additional research is needed to fully understand the impact of race-/ethnic-disparities on COVID-19 outcomes. Given the overrepresentation of race-/ethnic-minority patients with critical outcomes within this cohort, it is important for public health officials to ensure that prevention activities prioritize communities and racial/ethnic groups most affected by COVID-19. Despite the limitations, this study is still one of the first US statewide population-based studies using the entire population to investigate the presenting symptoms and clinical outcomes of COVID-19 patients.

Such a population-based study can minimize sampling selection bias and is more representative of all COVID-19 cases. Our results revealed that severe illness was strongly associated with hospitalization and mortality. However, the differences in the symptom distribution are not reflected in disparities in hospitalization and mortality in certain gender-and racial-minority groups. The findings from this study reinforce the fact that underlying health system disparities remain a challenge. South Carolina is often reflective of the "Deep South" states. Preexisting structural disparities were exacerbated during COVID-19 and put the already vulnerable populations at higher risk. Rural residence, as well as racial and ethnic social determinants of health unfortunately remain predictors of adverse health outcomes for COVID-19 patients. The effects are ongoing, making it a priority for interventions and policies to alleviate these problems in both the short and long-term. M a n u s c r i p t 20 A c c e p t e d M a n u s c r i p t 22 

COVID-19 Dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU). Available at

COVID-19 Cases in South Carolina

Prevalence of Asymptomatic SARS-CoV-2 Infection : A Narrative Review

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72314 Cases From the Chinese Center for Disease Control and Prevention

Coronavirus Disease 2019 Case Surveillance -United States

Characteristics Associated with Hospitalization Among Patients with COVID-19

Characteristics and Clinical Outcomes of Adult Patients Hospitalized with COVID-19 -Georgia

Hospitalization and Mortality among Black Patients and White Patients with Covid-19

Assessing differential impacts of COVID-19 on black communities

Assessment of Racial/Ethnic Disparities in Hospitalization and Mortality in Patients With COVID-19 in New York City

Failing Another National Stress Test on Health Disparities

Racial Disparities in COVID-19 Hospitalization and In-hospital Mortality at the Height of the New York City Pandemic. J Racial Ethn Health Disparities

Characteristics and Risk Factors for Hospitalization and Mortality among Persons with COVID-19 in Atlanta Metropolitan Area. medRxiv

Racial/Ethnic Differences In COVID-19 Screening, Hospitalization, And Mortality In Southeast Wisconsin

Risk factors for hospitalization, intensive care, and mortality among patients with asthma and COVID-19

Hospitalization Rates and Characteristics of Patients Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019 -COVID-NET, 14 States

Risk Factors for Intensive Care Unit Admission and In-hospital Mortality among Hospitalized Adults Identified through the U.S. Coronavirus Disease 2019 (COVID-19)-Associated Hospitalization Surveillance Network (COVID-NET)

List of Rural Counties and Designated Eligible Census Tracts in Metropolitan Counties

Rural-urban inequalities amplified by COVID-19: evidence from South Africa. Area Development and Policy

COVID-19 Pandemic Planning Scenarios

Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease

Atypical presentation of COVID-19 in a frail older person

Coronavirus Disease 2019 in Geriatrics and Long-Term Care: The ABCDs of COVID-19

COVID-19 in older people: a rapid clinical review

Clinical features of severe patients infected with 2019 novel coronavirus: a systematic review and meta-analysis

Male sex identified by global COVID-19 meta-analysis as a risk factor for death and ITU admission

Sex differences in HIV RNA level and CD4 cell percentage during childhood

Human pDCs display sex-specific differences in type I interferon subtypes and interferon alpha/beta receptor expression

Experimental pretesting of handwashing interventions in a natural setting

Tobacco smoking and COVID-19 pandemic: old and new issues. A summary of the evidence from the scientific literature

The impact of the risk of COVID-19 on Black, Asian and Minority Ethnic (BAME) members of the UK dental profession

Asian and Minority Ethnic groups in England are at increased risk of death from COVID-19: indirect standardisation of NHS mortality data

This Time Must Be Different: Disparities During the COVID-19 Pandemic

Structural Racism, Social Risk Factors, and Covid-19 -A Dangerous Convergence for Black Americans

Characteristics Associated With Racial/Ethnic Disparities in COVID-19 Outcomes in an Academic Health Care System

Assessing racial and ethnic disparities using a COVID-19 outcomes continuum for New York State

Racial disparities in COVID-19 hospitalizations do not lead to disparities in outcomes

Racial and Ethnic Disparities in COVID-19 Outcomes: Social Determination of Health

03) 11,262 (19.12) 507 (65.3) 54,977 (33.83) 75,825 (46.66) 31,705 (19.51) Hispanic or Latino

Unknown 68,658 (97.83) 1,526 (2.17)

Moderate/ severe

Unknown" and missing value were all grouped into "No" category

The sample size of some variables is not equal to the total sample size due to missing data

¶ The p value is from Chi-square test results and compared the rate differences of the dichotomized hospitalization and mortality across each exposure variable

A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t