Characteristics of Reported Deaths Among Fully Vaccinated Persons With Coronavirus Disease 2019—United States, January–April 2021

Abstract

Background

Vaccines against coronavirus disease 2019 (COVID-19) are highly efficacious, but severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections do occur after vaccination. We characterized COVID-19 cases among fully vaccinated persons with an outcome of death.

Methods

We analyzed COVID-19 cases voluntarily reported to the Centers for Disease Control and Prevention by US health departments from 1 January to 30 April 2021. We included cases among US residents with a positive SARS-CoV-2 test result ≥14 days after completion of an authorized primary vaccine series and who had a known outcome (alive or dead) as of 31 May 2021. When available, specimens were sequenced for viral lineage and death certificates were reviewed for cause(s) of death.

Results

Of 8084 fully vaccinated persons with reported COVID-19 during the surveillance period, 245 (3.0%) died. Compared with patients who remained alive, those who died were older (median age, 82 vs 57 years;), more likely to reside in a long-term care facility (51% vs 18%), and more likely to have ≥1 underlying health condition associated with risk for severe disease (64% vs 24%) (all P < .01). Among 245 deaths, 191 (78%) were classified as COVID-19 related. Of 106 deaths with available death certificates, COVID-19 was listed for 81 deaths (77%). There were no differences in the type of vaccine administered or the most common viral lineage (B.1.1.7).

Conclusions

COVID-19 deaths are rare in fully vaccinated persons, occurring most commonly in those with risk factors for severe disease, including older age and underlying health conditions. All eligible persons should be fully vaccinated against COVID-19 and follow other prevention measures to mitigate exposure risk.

Breakthrough, COVID-19, death, postvaccination, SARS-CoV-2, Vaccine

As of 27 December 2021, the United States reported approximately 51 million coronavirus disease 2019 (COVID-19) cases and 809 300 deaths [1]. Approximately 62% of the US population (71% of those ≥12 years of age) have received the recommended primary series with 1 of 3 COVID-19 vaccines (Pfizer-BioNTech [BNT162b2], Moderna [mRNA-1273], and Johnson & Johnson’s Janssen [JNJ-78436735]) that are authorized or approved by the US Food and Drug Administration. Clinical trials involving ≥76 000 participants demonstrated these vaccines to be safe and highly efficacious in preventing symptomatic COVID-19, and subsequent observational studies confirmed these findings [2–9]. Real-world data provided further evidence that these vaccines are highly effective in preventing hospitalizations and deaths from severe COVID-19 [10–15]. High vaccine effectiveness against severe COVID-19 is maintained even after infection with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Delta variant, demonstrating the continued critical role of vaccines in controlling the pandemic [16–22].

Because no vaccine is 100% effective, SARS-CoV-2 infections among fully vaccinated persons are expected and have been documented [23–29]. However, fully vaccinated people remain at much lower risk for hospitalization or death than unvaccinated or partially vaccinated persons with similar risk factors [1, 17–20, 30, 31]. Nevertheless, understanding the characteristics of the small proportion of COVID-19 cases among fully vaccinated persons that result in severe disease is of clinical and public health importance.

Beginning in January 2021, the Centers for Disease Control and Prevention (CDC) implemented a passive surveillance system to collect reports of COVID-19 infection among fully vaccinated persons (also called vaccine breakthrough cases). State and territorial health departments voluntarily reported COVID-19 cases among fully vaccinated persons to the CDC until 30 April 2021 [32]. After that time, surveillance of postvaccination SARS-CoV-2 infections focused on cases resulting in hospitalization or death, allowing for more complete information on cases of highest clinical and public health significance. The characteristics of COVID-19 cases among fully vaccinated persons reported to the CDC through 30 April 2021 have been reported elsewhere [23]. We describe in more detail the clinical and epidemiologic characteristics of cases with an outcome of death reported during the same time period.

METHODS

Case Definition

For this analysis, a COVID-19 case in a fully vaccinated person was defined as the detection of SARS-CoV-2 RNA or antigen in a respiratory specimen collected from the person ≥14 days after receipt of recommended primary series of a COVID-19 vaccine authorized or approved by the Food and Drug Administration. Cases meeting this definition that were reported to the CDC by 30 April 2021 were included in this analysis if outcomes (ie, alive or dead from any cause) were known as of 31 May 2021. Cases were excluded if a previous positive SARS-CoV-2 result was documented within 45 days of the first postvaccination positive test, if information on the vaccine manufacturer or administration dates was missing, or if the schedule was inconsistent with CDC recommendation (ie, receipt of 2 doses of vaccines from different manufacturers or administration of a second dose more than 4 days before the manufacturer’s recommended spacing between doses: <24 days for Moderna or <17 days for Pfizer-BioNTech)) [33]. This activity was conducted consistent with applicable federal law and CDC policy and was determined to be nonresearch and exempt from further institutional review board review at the CDC.

Data Collection and Analysis

Data were collected from state health departments and managed through a Research Electronic Data Capture (REDCap) secure platform and data capture tools hosted at the CDC [34]. Reported variables included demographics (age, sex, and race/ethnicity), vaccination information (dates and manufacturer), SARS-CoV-2 testing results (including cycle threshold [Ct] values from the first positive test and whole-genome sequencing results when available), underlying medical conditions associated with risk for severe disease [35], clinical course (including symptom status, hospitalization status, intensive care unit admission, and requirement for mechanical ventilation), and outcome (ie, alive or dead). A subset of available residual respiratory specimens, not sequenced within the state, was sent to the CDC for whole-genome sequencing. We defined strains based on CDC classification (as of November 2021) using the PANGO nomenclature as variants of concern if the lineage was Delta (B.1.617.2 and AY lineages) and as a variant being monitored if the lineage was Alpha (B.1.1.7 and Q lineages), Beta (B.1.351 and descendent lineages), Gamma (P.1 and descendent lineages), Epsilon (B.1.427, B.1.429), Eta (B.1.525), Iota (B.1.526), Kappa (B.1.617.1, B.1.617.3), Mu (B.1.621, B.1.621.1), or Zeta (P.2) [36].

For cases with an outcome of death, health departments reported whether the cause was COVID-19 related and how this determination was made (eg, death certificate, medical records, medical examiner, or provider report). We reviewed death certificates (ie, US Standard Certificate of Death Rev 11/2003) and medical records when available to verify the cause(s) of death and completed missing or unreported information. When the death certificate was available for review, the death was classified as COVID-19 related if COVID-19 was listed as an immediate or underlying cause of death (part I) or as a significant condition contributing to death (part II). If COVID-19 was not listed on the death certificate or the death certificate was not available for review, the death was still classified as COVID-19 related if the health department reported it to be COVID-19 related based on their review of the death certificate or medical examiner or provider report. We documented all conditions listed on the death certificate (parts I and II) for both COVID-19–related and non–COVID-19–related deaths.

We used descriptive statistics to assess the demographic characteristics, clinical features and course, and SARS-CoV-2 lineage of fully vaccinated people with COVID-19. We compared cases with an outcome of death versus alive and cases where cause of death was deemed COVID-19 related versus unrelated. Distributions of categorical variables were compared using χ² or Fisher exact tests. Distributions of continuous variables were compared using Wilcoxon rank sum tests. Differences were considered significant at P < .05. All analyses were performed using SAS software, version 9.4 (SAS Institute).

RESULTS

Among 12 417 cases reported to CDC as of 30 April 2021, 8084 (65%) were included in our analysis. The remaining 4333 reported cases were excluded because of an unknown outcome as of 31 May 2021 (n = 3649), no positive SARS-CoV-2 test (no test, no result, or negative result; n = 258), insufficient information to determine that an authorized vaccine dosing schedule was followed (n = 211), <14 days or an unknown number of days between completion of primary vaccine series and positive SARS-CoV-2 test (n = 146), case record not finalized by the state health department (n = 62), or a positive SARS-CoV-2 test within 45 days of the first postvaccination positive test (n = 7). The 8084 cases included in our analysis were reported to the CDC from 44 states, Washington, DC, and the US Virgin Islands; however, 10 states, Washington, DC, and the US Virgin Islands contributed <20 cases each. As of 31 May 2021, 245 (3%) of the 8084 reported COVID-19 cases in fully vaccinated persons had an outcome of death.

Among cases ending in death, dates of positive SARS-CoV-2 test ranged from 15 January to 26 April, a median of 33 days (interquartile range [IQR], 22–47 days) after vaccine series completion. Dates of death ranged from 3 February to 19 May, and patients died a median of 10 days after the positive SARS-CoV-2 test (IQR, 4–17 days). For 22 patients (9%), the positive COVID-19 test was obtained on the day of death or up to 5 days later, suggesting possible postmortem diagnosis. Compared with patients who lived, those who died were older (median age, 82 vs 57 years, respectively) and were more likely to be male (51% vs 34%), to reside in a nursing home or long-term care facility (51% vs 18%), and to have ≥1 underlying medical condition associated with risk for severe disease (64% vs 24%) (all P < .01) (Table 1).

Table 1.

Characteristics of Coronavirus Disease 2019 Cases Among Fully Vaccinated Persons by Outcome—United States, Reported January–April 2021 (n = 8084)

Characteristic	Deada (n = 245)		Alive (n = 7839)		P Value
	No./Total No.b	%	No./Total No.b	%
Age group, y
16–49	4/244	2	3046/7793	39	<.01
50–64	23/244	9	1740/7793	22
65–74	48/244	20	1203/7793	15
75–84	70/244	29	1040/7793	13
≥85	99/244	41	764/7793	10
Male sex	125/245	51	2672/7805	34	<.01
Race/ethnicity
White, non-Hispanic	189/233	81	4780/6351	75	.04
Black, non-Hispanic	25/233	11	526/6351	8
Asian, non-Hispanic	2/233	1	172/6351	3
AIAN, non-Hispanic	0/233	0	12/6351	0
Pacific Islander, non-Hispanic	3/233	1	170/6351	3
Multirace/other, non-Hispanic	6/233	3	193/6351	3
Hispanic or Latino (any race)	8/233	3	498/6351	8
LTCF or nursing home resident	112/220	51	1010/5657	18	<.01
Symptomatic SARS-CoV-2 infection	185/221	84	4211/6662	63	<.01
Hospitalization	186/235	79	870/7031	12	<.01
ICU	84/155	54	109/620	18	<.01
Mechanical ventilation	34/141	24	30/567	5	<.01
No. of underlying medical conditionsc
≥1	146/227	64	1488/6099	24	<.01
≥2	82/227	36	411/6099	7	<.01
Vaccine type
Pfizer-BioNTech	169/245	69	5017/7839	64	.30
Moderna	60/245	24	2175/7839	28
Janssen	16/245	7	647/7839	8
Test type
Nucleic acid amplification	226/245	92	6681/7774	86	<.01
Antigen	19/245	8	1093/7774	14

Abbreviations: AIAN, American Indian or Alaskan native; ICU, intensive care unit; long-term care facility; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Death from any cause (regardless of relation to coronavirus disease 2019).

Denominators represent total no. with data for each characteristic.

Including pregnancy, diabetes, chronic renal disease, chronic liver disease, autoimmune disease, immunosuppressive condition, or use of immunosuppressive medications.

The most common underlying conditions were the same in both groups but were more common among those who died: diabetes (44% vs 18% in those who survived), chronic renal disease (37% vs 7%), and immunosuppressive condition (33% vs 7%) (all P < .01) (data not shown). Patients who died compared with those who lived were more likely to have COVID-19–like symptoms (84% vs 63%), to require hospitalization (79% vs 12%), and intensive care (54% vs 18% among those hospitalized), and to experience symptoms associated with severe disease such as fever (40% vs 22%), shortness of breath (67% vs 24%), and difficulty breathing (51% vs 15%) (all P < .01). Milder symptoms, including headache (19% vs 24%), sore throat (6% vs 32%), runny nose (23% vs 62%), and loss of smell or taste (9% vs 29%) (all P < .01) were more common among patients who lived (data not shown). There were no significant differences between the 2 groups in the type of vaccine administered.

SARS-CoV-2 sequence data were available for 54 (22%) of 245 patients who died and 945 (12%) of 7839 patients who lived (Supplementary Table). The Alpha (B.1.1.7) lineage was the most common among both patients who died and those who lived (43% vs 41%, respectively), and the breakdown of variants being monitored did not differ significantly between groups (P = .34). The median Ct value among those specimens with sequence data was lower (associated with higher levels of viral genetic material) for those who died than for those who lived (median [IQR], 18.6 [14.7–24.6] vs 28.0 [20.6–33.2], respectively; P < .01). Only 4 infections caused by the Delta variant were observed during this time, all among those who remained alive. The proportion of infections caused by other variants being monitored was the same among patients who died and those who lived (72% for both; P = .98).

Of the 245 total deaths, 191 (78%) were classified as COVID-19 related, 41 (17%) were classified as not COVID-19 related, and 13 (5%) could not be classified. COVID-19 relatedness was determined by review of death certificates for 106 (46%) patients and reported by state health public health officials for the remainder; the most common sources of information used by health officials were death certificates, medical records, and reports from medical examiners or other healthcare professionals. Patients whose deaths were deemed COVID-19 related had similar distributions of age, sex, and race/ethnicity to those whose deaths were not considered COVID-19 related. However, compared with patients with non–COVID-19–related deaths, patients with COVID-19–related deaths were less likely to be residents of a long-term care facility (46% vs 76%; P < .01) and more likely to have ≥1 underlying medical condition (69% vs 47%; P = .01) or symptoms of COVID-19 (90% vs 57%; P < .01), to be hospitalized (83% vs 68%; P < .01), or to have SARS-CoV-2 sequence data available (26% vs 7%; P < .01), although not more likely to be infected with a variant being monitored (Table 2).

Table 2.

Characteristics of Coronavirus Disease 2019 (COVID-19) Deaths Among Fully Vaccinated Persons by COVID-19 Relatedness—United States, Reported January–April 2021 (n = 232)

Characteristic	COVID-19–Related Deaths (n = 191)a		Non–COVID-19–Related Deaths (n = 41)		P Value
	No./Total No.b	%	No./Total No.b	%
Age group, y
16–49	1/191	<1	2/40	5	.14
50–64	18/191	9	4/40	10
65–74	40/191	21	7/40	17
75–84	57/191	30	8/40	20
≥85	75/191	39	19/40	48
Male sex	101/191	53	25/41	61	.11
Race/ethnicity
White, non-Hispanic	147/183	80	32/38	84	.12
Black, non-Hispanic	22/183	12	1/38	3
Asian, non-Hispanic	2/183	1	0/38	0
AIAN, non-Hispanic	3/183	2	0/38	0
Pacific Islander, non-Hispanic	0/183	0	0/38	0
Multirace/other, non-Hispanic	3/183	2	3/38	8
Hispanic or Latino any race	6/183	3	2/38	5
LTCF or nursing home resident	79/171	46	29/38	76	<.01
Hospitalization	152/184	83	27/40	68	.03
ICU	72/126	57	9/23	39	…
Mechanical ventilation	29/113	26	4/22	18	…
COVID-19 symptoms	156/174	90	21/37	57
Fever	54/130	42	7/20	35	<.01
Chills	26/113	23	3/19	16
Rigors	4/91	4	1/19	5
Myalgias	33/115	29	2/19	11
Headache	24/117	21	2/18	11
Sore throat	6/114	5	1/19	5
Nausea/vomiting	20/121	17	3/20	15
Diarrhea	18/117	15	6/20	30
Fatigue	70/120	58	9/19	47
Runny nose	28/115	24	3/20	15
Cough	77/130	59	7/20	35
Shortness of breath	98/138	71	10/20	50
Difficulty breathing	65/121	54	6/18	33
Loss of taste	10/109	9	1/17	6
≥1 underlying medical conditionc	124/181	69	17/36	47	.01
Pregnancy	0/169	0	0/35	0	…
Diabetes mellitus	73/162	45	10/33	30	…
Chronic kidney disease	61/156	39	7/33	21	…
Chronic liver disease	6/139	4	3/32	9	…
Autoimmune disease	6/130	5	1/30	3	…
Immunosuppressive condition	49/140	35	6/30	20	…
Immunosuppressive medications	27/131	21	3/28	11	…
SARS-CoV-2 lineage identified	50/191	26	3/41	7	<.01
B.1.617.2 (Delta)	0/50	0	0/3	0	…
Variant monitoredd	37/50	74	2/3	66	…
Other lineage	13/50	26	1/3	33	…

Abbreviations: AIAN, American Indian or Alaskan native; COVID-19, coronavirus disease 2019; ICU, intensive care unit; LTCF, long-term care facility; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

COVID-19 relatedness was determined by review of death certificates for 106 patients and reports from state health public health officials for the remainder (based on death certificates, medical records, and healthcare professional reports).

Denominators represent total no. with data for each characteristic.

Including pregnancy, diabetes, chronic renal disease, chronic liver disease, autoimmune disease, immunosuppressive condition, or use of immunosuppressive medications.

Variants monitored included Alpha (B.1.1.7), Beta (B.1.351 and descendent lineages), Gamma (P.1), and Epsilon (B.1.427, B.1.429).

Of the 106 death certificates available for review (43%), 72 (68%) listed COVID-19 in the chain of events leading to the immediate cause of death, 9 (8%) listed COVID-19 as a contributing condition, and 25 (24%) did not have COVID-19 listed on the death certificate. Other common causes of death listed on the available death certificates included other respiratory conditions, cardiovascular conditions, diabetes, and sepsis (Table 3). Of note, among 25 patients for whom COVID-19 was not listed on the death certificate, 8 (32%) had documentation of either pneumonia or respiratory failure.

Table 3.

Causes of Death Among Fully Vaccinated Persons With Coronavirus Disease 2019 and Available Death Certificates—United States January–April 2021

Causes of Death	Total Deaths (n = 106)		COVID-19 Listed on Death Certificate (n = 81)		COVID-19 Not Listed on Death Certificate (n = 25)
	No.	%	No.	%	No.	%
COVID-19
In chain of eventsa	72	68	72	89	…	…
As a contributing conditionb	9	8	9	11	…	…
Other conditions on death certificate
Respiratory conditions
ARDS or respiratory failure	36	34	31	38	5	20
COPD	10	9	9	11	1	4
Pneumonia	56	53	51	63	5	20
Pulmonary embolism	2	2	1	1	1	4
Cardiovascular conditions
Arteriosclerosisc	15	14	13	16	2	8
Atrial fibrillation	4	4	4	5	0	0
Cardiac arrest	8	8	6	7	2	8
Endocarditis	2	2	1	1	1	4
Heart failure or cardiomyopathy	18	17	14	17	4	16
Hypertension	20	19	18	22	2	8
Stroke	9	8	5	6	4	16
Other cardiovascular conditionsd	3	3	3	4	0	0
Sepsis	15	14	11	14	4	16
Other chronic conditionse	49	46	42	52	7	28
Other acute conditionsf	23	22	16	20	7	28

Abbreviations: ARDS, acute respiratory distress syndrome; COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 2019.

Refers to part I of the death certificate: “Enter the chain of events-—diseases, injuries, or complications—that directly caused the death.”

Refers to part II of the death certificate: “Enter other significant conditions contributing to death but not resulting in the underlying cause given in Part I.”

Arteriosclerosis includes myocardial infarction, coronary artery disease, and peripheral vascular disease.

Other cardiovascular conditions included aortic valve replacement, arteriosclerosis, sick sinus syndrome, and supraventricular tachycardia (each n = 1).

Other chronic conditions included diabetes (n = 18); chronic renal disease (n = 13); dementia (n = 12); leukemia or lymphoma (n = 12); liver disease (n = 7); solid tumor cancer (n = 3); alcohol abuse , obesity, and rheumatoid arthritis (each n = 2); and anemia, asthma, kidney and liver transplant, myotonic muscular dystrophy, polymyalgia rheumatica, protein calorie malnutrition, seizure disorder, and tobacco abuse (n = 1)

Other acute conditions included acute renal failure (7); encephalitis (4); gastrointestinal bleed (3); bowel perforation or peritonitis, fall, and urinary tract infection (each n = 2); and acute blood loss, ankle fracture, cervical injuries, cholecystitis, disseminated intravascular disease, gangrene, influenza B virus, methamphetamine positive, sacral decubitus ulcer, and thermal burns from stove (each n = 1).

Of the 245 patients who died, 231 (94%) had ≥1 underlying condition, resided in a long-term care facility or nursing home, or had age ≥75 years. Only 2 deaths occurred in patients who were <50 years old and had no reported underlying conditions, and only 1 of these was classified as COVID-19 related, with septic shock reported as the immediate cause of death on the death certificate.

DISCUSSION

In this analysis of COVID-19 cases among fully vaccinated persons voluntarily reported during January–April 2021, we found that COVID-19 deaths occurred more commonly among persons with risk factors for severe disease, including older age (especially among those aged >75 years) and underlying health conditions such as diabetes, chronic renal disease, and immunosuppressive conditions. These deaths represented a very small proportion of the reported COVID-19 cases among fully vaccinated persons and occurred among the >100 million Americans fully vaccinated during the same period [1]. The characteristics and causes of death in fully vaccinated patients resembled those who died of COVID-19 before vaccination [37, 38], highlighting the importance of multiple exposure mitigation strategies and full vaccination in populations at high risk for severe outcomes of COVID-19.

Previous reports have documented that SARS-CoV-2 infections among fully vaccinated individuals are largely not severe, and infrequent cases of severe COVID-19, including from Delta variant, occurred in medically vulnerable individuals with underlying conditions that predisposed them to severe COVID-19—findings similar to that of our analysis [25, 27, 28, 30, 39–41]. The overall characteristics of cases included in our analysis were similar to those in the population prioritized for vaccination in the first 4 months of 2021, with a predominance of older adults, persons with underlying conditions, and residents of long-term care facilities. Likewise, the distribution of vaccine types represented what was administered nationwide [1], and we did not find any significant differences in vaccine type between fully vaccinated persons who died and those who lived. The viral lineage of specimens collected during the analytic period also reflected those that were circulating in the United States at the time [1], with no evidence that specific lineages were associated with death, though the number of records with sequence information was limited.

Roughly 4 of 5 deaths were reported as COVID-19 related by state health departments; a review of a subset of cases with available death certificates found that COVID-19 was listed as a cause or contributor of death in a similar proportion of cases. Among death certificates in which COVID-19 was not listed, one-third of the cases documented pneumonia or respiratory failure as the cause of death, which is consistent with COVID-19. The remaining two-thirds of non–COVID-19–related deaths appeared to be among persons who were medically fragile with conditions that placed them at high risk of death, and the positive SARS-CoV-2 test result may have been incidental.

Approximately one-quarter of patients tested positive within the third week (day 15–21) after vaccine series completion, suggesting possible acquisition of infection before full immunity was reached. Prior studies have shown an inverse relationship between time since vaccination and the likelihood of a positive SARS-CoV-2 test [24, 27, 42]. However, it is possible that we observed fewer postvaccination infections further out from the time of full vaccination because the analytic period was too early in the national vaccination campaign to capture waning immunity and the follow-up time (ie, known outcome as of 31 May 2021) was <4 months.

This analysis is subject to several limitations. First, our findings may not be generalizable because the passive surveillance system for COVID-19 cases among fully vaccinated persons did not receive case reports from all states, represented a time before the widespread circulation of the Delta or Omicron variants, and focused on a period before vaccines were approved for use in children <16 years old or were widely available to adults at lower risk of occupational exposure or severe COVID. Second, fully vaccinated individuals with COVID-19 who died after 31 May 2021 could have been misclassified as alive in this analysis; however, given the additional month of follow-up time allowed, we expect these misclassifications to be rare. Third, we did not obtain symptom onset date, which could have led to the inappropriate inclusion of some patients who contracted COVID-19 before becoming fully vaccinated. We were unable to assess Ct values in relation to illness onset, although viral loads are known to vary according to the timing of testing. Fourth, the methods and timing of testing were not standardized across laboratories. Ct values may differ with the type of test, limiting their usefulness as a marker for viral load, and more severe cases might have been diagnosed earlier, resulting in lower Ct values. Fifth, missing data were more common among patients who lived, which could have affected comparisons with patients who died. Finally, SARS-CoV-2 lineage (all cases) and death certificates (among patients who died) were available for only a small number of cases. Nevertheless, this analysis adds to our understanding of the most severe cases of COVID-19 in fully vaccinated individuals.

COVID-19 vaccines are safe and effective, but COVID-19 cases among fully vaccinated individuals will continue to occur, particularly while community transmission remains high. Deaths among fully vaccinated persons were most common among older adults (especially those >75 years old) and those with ≥1 underlying health condition during the early postvaccination period before widespread transmission of the Delta variant B.1.617.2 (January–April 2021). Our findings highlight the importance of complete vaccination of all persons eligible for COVID-19 vaccination (including a third or booster dose), especially for those at increased risk for severe disease, and the continued use of recommended prevention measures (eg, masking and social distancing). High population-level vaccination coverage will help reduce community-level transmission and reduce the risk of COVID-19 among fully vaccinated persons.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. The authors acknowledge the support and contributions of other members of the US Centers for Disease Control and Prevention (CDC) COVID-19 Vaccine Breakthrough Case Investigations Team, including Gregory C. Chang, Thomas A. Clark, Taylor Eisenstein, Jason M. Goldstein, Gayle E. Langley, K. Danielle Lecy, Stacey W. Martin, Felicita Medalla, Robin Spratling, Gail C. Thompson, and A. Angelica Trujillo; the CDC COVID-19 Strain Surveillance and Emerging Variant Team, Laboratory and Testing Task Force; and epidemiologists and laboratory scientists at US state and local public health departments.

Disclaimer. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC. This work was performed at the CDC.

Potential conflicts of interest. The authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

Author notes

L. K. F. W. and K. M. contributed equally to this work.