Increased deaths from fungal infections during the COVID-19 pandemic—National Vital Statistics System, United States, January 2020–December 2021

Abstract Background COVID-19–associated fungal infections cause severe illness, but comprehensive data on disease burden are lacking. We analyzed US National Vital Statistics System (NVSS) data to characterize disease burden, temporal trends, and demographic characteristics of persons dying from fungal infections during the COVID-19 pandemic. Methods Using NVSS’s January 2018–December 2021 Multiple Cause of Death Database, we examined numbers and age-adjusted rates (per 100,000 population) of fungal deaths by fungal pathogen, COVID-19 association, demographic characteristics, and year. Results Numbers and age-adjusted rates of fungal deaths increased from 2019 (n = 4,833, rate 1.2, 95% confidence interval [CI]    1.2–1.3) to 2021 (n = 7,199, rate: 1.8, 95% CI = 1.8–1.8); of 13,121 fungal deaths during 2020–2021, 2,868 (21.9%) were COVID-19–associated. Compared with non-COVID-19–associated fungal deaths (n = 10,253), COVID-19–associated fungal deaths more frequently involved Candida (n = 776 [27.1%] versus n = 2,432 [23.7%]) and Aspergillus (n = 668 [23.3%] versus n = 1,486 [14.5%]) and less frequently involved other specific fungal pathogens. Fungal death rates were generally highest in non-White and non-Asian populations. Death rates from Aspergillus infections were approximately two times higher in the Pacific US census division compared with most other divisions. Conclusions Fungal deaths increased during 2020–2021 compared with previous years, primarily driven by COVID-19–associated fungal deaths, particularly those involving Aspergillus and Candida. Our findings may inform efforts to prevent, identify, and treat severe fungal infections in COVID-19 patients, especially in certain racial/ethnic groups and geographic areas.

For fungal deaths occurring during January 2020-December 2021, we stratified data by COVID-3 19 association and examined numbers, percentages, and age-adjusted rates of death by sex, 4 race/ethnicity, US census division of residence (https://www.census.gov/programs-surveys/economic-5 census/guidance-geographies/levels.html), and type of fungal disease; we examined crude death rates 6 across 10-year age groups and urban-rural 2013 classifications of residence 7 (https://www.cdc.gov/nchs/data_access/urban_rural.htm). Crude death rates were shown for urban-rural 8 classification because age-stratified death rates by urban-rural status were not available on CDC-9 WONDER. Death rates based on counts of less than twenty were not calculated. We categorized 10 race/ethnicity as Hispanic or Latino (Hispanic), non-Hispanic White (White), non-Hispanic Black 11 (Black), non-Hispanic Asian (Asian), non-Hispanic American Indian or Alaska Native (AI/AN), non-12 Hispanic Native Hawaiian or other Pacific Islander (NHPI), non-Hispanic multiracial, and unknown. We 13 analyzed data using the six single race categories because these were the only categories for which race-14 specific population estimates for rate denominators were available on the CDC-WONDER platform.

15
Our analysis of US death certificate data found that >13,000 persons died from fungal infections 16 during 2020-2021, representing an increase in the numbers and age-adjusted rates of death from fungal 17 infections compared with previous years. This increase was primarily driven by COVID-19-associated 18 fungal deaths, particularly those involving Aspergillus and Candida, and highlights the importance of 19 considering fungal infections in patients with COVID-19. We also found striking racial/ethnic 20 disparities and geographic differences in rates of death from fungal infections.

21
In our analysis, fungal death counts rose in tandem with COVID-19 surges during January and

22
October 2021 but not during the first COVID-19 surge in April 2020. Recent analyses of testing 23 practices have documented a precipitous decrease in testing for pathogens other than SARS-CoV-2 (the virus that causes COVID-19) during April 2020, a finding that authors have attributed to strained 1 healthcare resources during the early COVID-19 pandemic [17,18]. We suspect that the absence of a 2 peak in fungal deaths during April 2020 might reflect a lack of disease detection and reporting rather 3 than a truly low number of COVID-19-associated fungal deaths. Conversely, the peaks in fungal deaths 4 that occurred during January and October 2021 might reflect increased clinician awareness and testing 5 for COVID-19-associated fungal infections, and possibly, the increased use of corticosteroids and 6 tocilizumab (both known risk factors for invasive mold infections and candidiasis) [6] to treat patients 7 with severe COVID-19.

8
Our finding that Candida and Aspergillus were the most commonly identified fungal pathogens 9 causing death is consistent with previous literature describing fungal disease epidemiology both before 10 and during the COVID-19 pandemic [6,9,12,19]. In our analysis, a higher percentage of COVID-19-11 associated fungal deaths involved Candida and Aspergillus compared with non-COVID-19-associated 12 fungal deaths, a finding that aligns with reports identifying COVID-19 as a risk factor for invasive 13 aspergillosis and candidiasis [6,12]. Although US data on the incidence of COVID-19-associated 14 fungal infections are sparse, a multicenter study from Europe found that >10% of critically ill COVID-15 19 patients might develop invasive aspergillosis, with mortality rates exceeding 40% [12]. Limited 16 reports also suggest that the incidence of invasive candidiasis might have increased during the COVID-  Although we identified an increase in the number of deaths from Mucorales spp. during 2020-22 2021 compared with previous years, yearly rates of death for this pathogen remained low throughout the 23 study period (<0.1 per 100,000 population). This finding is consistent with reports highlighting the rarity of mucormycosis in the United States [4,24,25]. Nonetheless, previous reports suggest that 1 mucormycosis can cause severe illness, disfiguration, and death in COVID-19 patients, including among 2 those who lack severe immunocompromising conditions [10,13].    was less than 20 (indicated by the symbol "-"). Rates were rounded to one decimal point; therefore, rates of "0.0" might not represent true zeros.