As of March 7, 2023, a total of 30,235 confirmed and probable monkeypox (mpox) cases
were reported in the United States,
†
predominantly among cisgender men
§
who reported recent sexual contact with another man (
1
). Although most mpox cases during the current outbreak have been self-limited, cases
of severe illness and death have been reported (
2
–
4
). During May 10, 2022–March 7, 2023, 38 deaths among persons with probable or confirmed
mpox
¶
(1.3 per 1,000 mpox cases) were reported to CDC and classified as mpox-associated
(i.e., mpox was listed as a contributing or causal factor). Among the 38 mpox-associated
deaths, 94.7% occurred in cisgender men (median age = 34 years); 86.8% occurred in
non-Hispanic Black or African American (Black) persons. The median interval from symptom
onset to death was 68 days (IQR = 50–86 days). Among 33 decedents with available information,
93.9% were immunocompromised because of HIV. Public health actions to prevent mpox
deaths include integrated testing, diagnosis, and early treatment for mpox and HIV,
and ensuring equitable access to both mpox and HIV prevention and treatment, such
as antiretroviral therapy (ART) (
5
).
Data included in this report were collected during May 10, 2022–March 7, 2023. Jurisdictional
health departments electronically reported confirmed and probable mpox cases and associated
deaths as part of national case surveillance. Case data were shared with CDC through
a standardized case report form or through the National Notifiable Diseases Surveillance
System.** Additional data (e.g., clinical course, co-occurring health conditions,
and treatments received) about some decedents were collected during consultations
between treating clinicians and CDC clinical officers.
††
Cause of death was most commonly determined by the treating health care provider and
reported on the death certificate. Jurisdictional health departments shared cause-of-death
data as reported on the death certificate to support classification of deaths. Deaths
were classified as mpox-associated if mpox was listed on Part I or Part II of the
death certificate (chain of events that directly caused the death or significant conditions
contributing to death, respectively). Deaths were classified as non–mpox-associated
if mpox was not listed on the death certificate, or if mpox appeared to be incidental
to death. Deaths were classified as being under investigation
§§
if jurisdictions were reviewing the cause-of-death with health care providers at the
time of this analysis.
Descriptive statistics on demographics (e.g., age, gender identity, race and ethnicity,
and state of residence) and clinical characteristics (i.e., treatments received and
timing of treatments) were calculated for all patients with an mpox-associated death
and were compared between decedents and patients who are presumed to have survived
after confirmed or probable mpox (survivors). SAS statistical software (version 9.4;
SAS Institute) was used for all analyses. This activity was reviewed by CDC and was
conducted consistent with applicable federal law and CDC policy.
¶¶
CDC received reports of 52 deaths among persons with confirmed or probable mpox. Thirty-eight
deaths (73.1%) were classified as mpox-associated, three (5.8%) were non–mpox-associated,***
and 11 (21.2%) deaths remain under investigation. Among the 38 mpox-associated deaths,
25 (65.8%) occurred during October–November 2022 (Figure).
FIGURE
Weekly confirmed and probable* mpox cases (A)
†
and mpox-associated deaths (B),
§
by week — United States, May 10, 2022—March 7, 2023
* https://www.cdc.gov/poxvirus/mpox/clinicians/case-definition.html
† The earliest date available regarding the case, beginning with date of illness or
rash onset, diagnosis date, positive laboratory test report date, CDC call center
reporting date, or case data entry date into the National Notifiable Diseases Surveillance
System.
§ Date of death as reported on the death certificate.
This figure is a set of two histograms showing the weekly confirmed and probable mpox
cases and mpox-associated deaths, by week, in the United States, during May 10, 2022—March
7, 2023.
The median age of decedents was 34 years (range = 22–58 years), and 36 (94.7%) were
cisgender men (Table 1). A higher proportion of decedents than survivors were Black
(86.8% versus 32.9%). Nearly one half of decedents (47.4%) resided in the U.S. Census
Bureau South Region
†††
compared with 39.4% of survivors who lived in this region. Nine of the 10 decedents
with recent sexual or intimate contact information had sexual or intimate contact
with cisgender men in the 3 weeks preceding symptom onset. Two decedents reported
nonsexual close contact exposures (e.g., co-sleeping and caring for a household member)
to persons with mpox. Five of 11 decedents with available data were experiencing homelessness.
Among 87% of decedents and 45% of survivors with available information, HIV infection
was more prevalent among decedents than survivors (93.9% versus 38.3%).
TABLE 1
Demographic and epidemiologic characteristics of persons who survived or died* from
mpox-related illness — United States, May 10, 2022–March 7, 2023
Characteristic
Mpox cases, no. (%)†
Survivors
(n = 30,183)
Decedents
(n = 38)
Demographic
Median age, yrs (range)
34 (0–89)
34 (22–58)
Sex or gender,§ total
26,082 (86.4)
38 (100.0)
Cisgender man
24,759 (94.9)
36 (94.7)
Cisgender woman
806 (3.1)
1 (2.6)
Transgender man
55 (0.2)
0 (—)
Transgender woman
227 (0.9)
1 (2.6)
Another gender identity
235 (0.9)
0 (—)
Unknown
4,101
0
Race and ethnicity, total
28,233 (93.5)
38 (100.0)
American Indian or Alaska Native, non-Hispanic
115 (0.4)
0 (—)
Asian, non-Hispanic
786 (2.8)
0 (—)
Black or African American, non-Hispanic
9,295 (32.9)
33 (86.8)
Native Hawaiian or other Pacific Islander, non-Hispanic
68 (0.2)
0 (—)
White, non-Hispanic
8,277 (29.3)
3 (7.9)
Hispanic or Latino
8,849 (31.3)
2 (5.3)
Other race, non-Hispanic
668 (2.4)
0 (—)
Multiple races, non-Hispanic
175 (0.6)
0 (—)
Unknown
1,950
0
U.S. Census Bureau region,¶
total
30,183 (100.0)
38 (100.0)
Northeast
6,600 (21.9)
6 (15.8)
Midwest
3,164 (10.5)
9 (23.7)
South
11,882 (39.4)
18 (47.4)
West
8,330 (27.6)
5 (13.2)
Puerto Rico
207 (0.7)
0 (—)
Experiencing homelessness
NA
11 (40.7)
Yes
NA
5 (45.5)
No
NA
6 (54.5)
Unknown
30,183
16
Clinical
Sexual or intimate contact in the 3 wks before symptom onset
18,385 (60.9)
16 (42.1)
Sexual contact in the past 3 wks
15,061 (81.9)
10 (62.5)
Recent partners exclusively men
12,759 (69.4)
9 (56.2)
Recent partners women or other genders (no men)
898 (4.9)
0 (—)
Recent partners include men and other genders
503 (2.7)
0 (—)
Gender of all partners unknown/Not specified
901 (4.9)
1 (6.3)
No recent sexual contact
3,324 (18.1)
6 (37.5)
Unknown
11,798
22
Interval from illness onset to testing, days, median (IQR)**
7 (4–10)
7 (3–10)
HIV-positive or immunocompromised††
13,549 (44.9)
33 (86.8)
Yes, HIV-positive
5,186 (38.3)
31 (93.9)
Yes, other immunocompromising conditions
654 (4.8)
2 (9.1)
No
7,709 (56.9)
0 (—)
Unknown
16,634
5
Received JYNNEOS vaccine
§§
11,316 (37.5)
13 (34.2)
Yes
8,238 (72.8)
1 (7.7)
No
3,078 (27.2)
12 (92.3)
Unknown
18,867
25
Abbreviation: NA = not available.
* Mpox was listed as directly causing or significantly contributing to death on the
death certificate.
†
Percentages were calculated with nonmissing data.
§ Among decedents whose death was mpox-associated (38), self-reported data on gender
identity was available for 28 (73.7%) decedents. Sex assigned at birth was substituted
for 10 (26.3%) decedents for whom gender identity was missing.
¶
https://www2.census.gov/geo/pdfs/maps-data/maps/reference/us_regdiv.pdf (Accessed
March 24, 2023). Puerto Rico is not included in any U.S. Census Bureau region.
** Illness onset is the date any reported symptoms first started.
††
Non-HIV immunocompromising conditions included renal transplant (one) and uncontrolled
diabetes (one).
§§ At least 1 dose.
Among 27 (71.0%) decedents with available clinical data, the median interval from
symptom onset to death was 68 days (range = 1–146 days; IQR = 50–86 days), and 87.0%
(20 of 23) were admitted to an intensive care unit during hospitalization (Table 2).
Overall, 25 (92.6%) of 27 decedents with available treatment information received
mpox-directed medical therapeutics
§§§
at some point during their clinical care, including 25 who received tecovirimat, 18
of 24 who received vaccinia immunoglobulin intravenous (VIGIV), nine of 22 who received
cidofovir, and six of 15 who received brincidofovir. Among the 25 decedents who received
tecovirimat, 15 (60.0%) received it within 3 days of mpox diagnosis; however, for
six (24.0%) decedents, tecovirimat was not administered until ≥3 weeks after diagnosis
(median overall interval = 2 days; IQR = 0–20 days). Among 24 decedents with available
data, all had lesions described as necrotic, diffuse, or worsened after a 14-day course
of tecovirimat.
TABLE 2
Selected clinical characteristics* of mpox-associated deaths with available clinical
data (N = 27) — United States, May 10, 2022–March 7, 2023
Characteristic (no. with information)
Mpox-associated deaths, no. (%)†
Clinical
Days from symptom onset to death (26), median (IQR)
68 (50–86)
Admitted to ICU
23 (85.2)
Yes
20 (87.0)
No
3 (13.0)
Unknown
4
Medical treatment§
Received any mpox-directed treatment
27 (100.0)
Yes
25 (92.6)
No
2 (7.4)
Unknown
0
Days from first mpox treatment date to death (21), median (IQR)
58 (32–66)
Specific mpox-directed treatment received
Tecovirimat
27 (100.0)
Yes
25 (92.6)
No
2 (7.4)
Unknown
0
Interval from mpox diagnosis to initiation of tecovirimat treatment (20), days, median
(IQR)
2 (0–20)
VIGIV
24 (88.9)
Yes
18 (75.0)
No
6 (25.0)
Unknown
3
Cidofovir
22 (81.5)
Yes
9 (40.9)
No
13 (59.1)
Unknown
5
Brincidofovir
15 (55.6)
Yes
6 (40.0)
No
9 (60.0)
Unknown
12
Necrotic, diffuse, or worsened lesions¶
24 (88.9)
Yes
24 (100.0)
No
0 (—)
Unknown
3
Received steroids for mpox complications or IRIS concerns
24 (88.9)
Yes
13 (54.2)
No
11 (45.8)
Unknown
3
HIV-positive or immunocompromised**
27 (100.0)
HIV-positive
25 (92.6)
CD4 ≥500
0 (—)
CD4 ≥200 to <500
0 (—)
CD4 ≥50 to <200
1 (4.2)
CD4 <50
23 (95.8)
CD4 Unknown
1
Immunocompromised (HIV-negative)
2 (7.4)
Unknown
0
Receiving ART (HIV-positive persons)
22 (88.0)
Yes, before mpox diagnosis
2 (9.1)
Yes, after mpox diagnosis
19 (86.4)
No, refused
1 (4.5)
Unknown
3
Interval from mpox diagnosis to initiation of ART (9), days, median (IQR)
15 (5–26)
Abbreviations: ART = antiretroviral treatment; ICU = intensive care unit; IRIS = immune
reconstitution inflammatory syndrome; VIGIV = vaccinia immune globulin intravenous.
* Reported or collected during clinical consult with treating physician.
†
Percentages calculated with nonmissing data. Percentages calculated based on total
number in each subgroup.
§ Treatment classification includes report of tecovirimat (oral or intravenous), VIGIV,
cidofovir, or brincidofovir in the clinical consultation data.
¶ Reported during clinical consultation with treating physician or through expert
assessment of photographs.
** Non-HIV immunocompromising conditions included renal transplant (one) and uncontrolled
diabetes (one).
Two decedents did not receive any mpox therapeutics based on clinician discretion;
in one case because of concerns regarding contraindication associated with other comorbidities.
The other decedent was on ART for HIV with an undetectable viral load at the time
of initial assessment for mpox care. Approximately 1 month later, at a wellness check,
the patient was found deceased with diffuse lesions characteristic of mpox. Seven
of 27 decedents for whom information was available refused therapeutic or intravenous
medications or left the hospital against medical advice at some point during their
clinical course of treatment. Among 24 decedents with information on receipt of steroids,
more than one half (54.2%) received steroids for mpox complications or concerns about
immune reconstitution inflammatory syndrome (IRIS), a hyperinflammatory response that
can occur in patients with HIV during the first 6 months of ART (
6
).
Nearly all decedents with complete data on HIV infection were HIV-positive (93.9%;
31 of 33). Among 24 decedents with HIV and available data, all (100%) had CD4 counts
<200 cells/mm3; 23 (95.8%) had CD4 counts <50 cells/mm3. Among the two immunocompromised
decedents who did not have HIV, one was presumed to have been immunocompromised as
a consequence of undiagnosed diabetes; the patient experienced diabetic ketoacidosis,
a severe life-threatening complication of diabetes, at the time of mpox diagnosis.
The second decedent was severely immunocompromised because of a recent renal transplant
complicated by acute rejection.
Only two of the 25 persons with HIV (8.0%) reported taking ART before mpox diagnosis;
in one of these decedents, HIV was poorly controlled. ART was initiated for 19 of
the 20 decedents who were not receiving ART, including one decedent who received a
diagnosis of HIV infection 5 days after the mpox diagnosis. One decedent refused treatment
for advanced HIV. ART treatment status was not known for three of the 25 decedents
with HIV. ART was either delayed or interrupted for seven decedents because of clinician
concerns about IRIS.
Discussion
During the 2022 mpox outbreak, 1.3 mpox-associated deaths per 1,000 cases occurred
in the United States and approximately 1.2 deaths per 1,000 mpox cases occurred worldwide.
¶¶¶
Nearly all U.S. mpox decedents were immunocompromised at the time of diagnosis. Almost
90% of U.S. mpox-associated deaths occurred in Black men, whereas fewer than one in
three mpox survivors were Black men. Although mpox exposure data were not available
for all decedents, nearly all cisgender male decedents who reported recent sexual
contact reported cisgender male partners.
Most decedents received one or more prompt mpox-directed treatments and intensive
care. However, nearly one quarter of decedents experienced delays of 3–7 weeks between
diagnosis and treatment, and two patients did not receive any mpox-directed treatment.
Before the 2022 mpox epidemic, severe cases mainly occurred in resource-constrained
settings where transmission is endemic and where treatments, such as tecovirimat,
cidofovir, and VIGIV, are not routinely available (
6
).
The lengthy course of illness experienced by most decedents is likely related to a
reduced capacity to respond to infection because of co-occurring immunocompromise.
When mpox is suspected, providers should consider early treatment with mpox-directed
therapy, especially for patients with immunocompromising conditions who are at highest
risk for severe outcomes.
The gender and racial disparities in mpox-associated deaths align with previous reports,
in which most patients hospitalized for severe manifestations of mpox were Black men
with uncontrolled HIV (
4
) and parallel racial and ethnic disparities in HIV infection and mortality. In 2020,
75% of all-cause deaths among adults with HIV occurred in males, 39% of whom were
Black males.**** Disparities and barriers are apparent at all levels of HIV care including
recognition of HIV risk, access to testing, and access to and receipt of preexposure
prophylaxis and ART (
7
).
Most decedents with advanced HIV were not on ART at the time mpox was diagnosed. Boosting
immune function via ART and avoidance of immunosuppressives, such as steroids, are
critical to mpox recovery (
6
). ART was delayed or discontinued for several decedents because of concerns of IRIS,
wherein a sudden improvement of immune function paradoxically worsens an infection.
To date, no evidence exists indicating that IRIS contributes to poor mpox outcomes.
The Ending the HIV Epidemic initiative
††††
has highlighted the need to address HIV transmission in the United States through
response, prevention, diagnosis, and treatment efforts. All patients with suspected
mpox should be evaluated for preexisting immunocompromising conditions such as HIV
(7). Persons with diagnosed mpox who have HIV (even if newly diagnosed) who are not
on ART should be started on ART as soon as possible. For persons who are HIV-negative,
indications for HIV pre-exposure prophylaxis should be assessed.
Previous studies of mpox patients, in Nigeria, have documented anxiety, depression
(
8
), and suicide (
9
). It has been suggested (
9
) that these outcomes resulted from stigma and the implications of acquiring a sexually
associated infection, such as mpox, as well as the isolation experienced during prolonged
treatment. Black and Hispanic or Latino males in the United States might face additional
psychosocial pressures related to language barriers, homophobia, and discrimination
(
10
). One suicide was noted among the decedents of non–mpox-associated deaths, and one
decedent persistently declined treatment for HIV. These data highlight the need to
strengthen psychosocial support services as part of the mpox epidemic response.
The findings in this report are subject to at least two limitations. First, deaths
might be undercounted. Delays in reporting laboratory results, cases, and deaths are
expected because reporting guidance was developed concurrently with the outbreak.
Because reports from clinical consults were passively obtained, some deaths related
to patients with mpox might not have been recorded. Second, some data (e.g., housing
status, treatment, HIV status, and CD4 count) were more frequently available for decedents
than for survivors because of additional reporting requirements for those who received
intensive mpox treatments, thus limiting the comparisons between these two groups.
These findings highlight the importance of integrating prevention, testing, and treatment
for multiple sexually associated infections (e.g., mpox and HIV, among others). Equitable
access to prevention, treatment, and engagement and retention in care for both mpox
and HIV should be prioritized, particularly among Black men and other persons at risk
for sexually associated infections. These results underscore previous recommendations
that providers offer HIV testing to all patients with probable or confirmed mpox and
consider early mpox-directed treatment in highly immunocompromised patients (
4
,
5
,
7
). Further, combining therapies for mpox and boosting immune function might also reduce
mortality from severe mpox (
5
).
Summary
What is already known about this topic?
Severe manifestations of mpox have occurred in the United States, particularly among
persons with uncontrolled viral spread resulting from moderately to severely immunocompromising
conditions.
What is added by this report?
Thirty-eight mpox-associated deaths occurred in the United States during May 10, 2022–March
7, 2023 (1.3 mpox-associated deaths per 1,000 cases). Most decedents were non-Hispanic
Black or African American (87%) persons and cisgender men (95%). Among 24 decedents
with HIV for whom data were available, all had advanced HIV, typically with a CD4
count <50.
What are the implications for public health practice?
Equitable and early access to prevention and treatment for both mpox and HIV is critical
to reducing mpox-related mortality.