Morbidity and Mortality Weekly Report

Weekly / Vol. 71 / No. 21 May 27, 2022

Barriers to and Disparities in Access to Health Care Among Adults

Aged ≥18 Years with Epilepsy — United States, 2015 and 2017
Niu Tian, MD, PhD1; Rosemarie Kobau, MPH1; Matthew M. Zack, MD1,*; Kurt J. Greenlund, PhD1

Approximately 3 million U.S. adults have active epilepsy care because of insufficient transportation. Enhancing link-
(i.e., self-reported doctor-diagnosed history of epilepsy and ages between clinical and community programs and services
currently taking epilepsy medication or have had at least one by public health practitioners and epilepsy health and social
seizure in the past year, or both) (1). One of the most com- service providers can address gaps in access to health care.
mon brain disorders, epilepsy poses a number of challenges for NHIS is an annual, nationally representative household
people living with this condition because its treatment can be survey of the U.S. civilian, noninstitutionalized population.§
complex, daily management might be inadequate to achieve Supplementary questions on epilepsy were added to the
seizure control, it limits social participation, and epilepsy is 2015 and 2017 Sample Adult Core component of NHIS,
associated with early mortality.† Previous studies indicate that which includes one randomly selected adult aged ≥18 years
persons with epilepsy are more likely to experience barriers or from each randomly selected household. Adult respondents
delays in receipt of certain types of care, including epilepsy
specialty care, and that these delays are often associated with § https://www.cdc.gov/nchs/nhis/index.htm

individual factors (e.g., seizure type) or social determinants

of health (e.g., household income or provider availability)
(2–4). To obtain updated estimates of access to health care INSIDE
among U.S. adults aged ≥18 years by epilepsy status, CDC 703 Seizure- or Epilepsy-Related Emergency
analyzed pooled data from the 2015 and 2017 National Health Department Visits Before and During the COVID-19
Interview Survey (NHIS), the most recent years with available Pandemic — United States, 2019–2021
epilepsy data. Age-adjusted analyses comparing adults with 709 Multistate Outbreak of Listeria monocytogenes
active epilepsy or inactive epilepsy (i.e., self-reported doctor- Infections Linked to Fresh, Soft Hispanic-Style
diagnosed epilepsy but not currently taking medication for Cheese — United States, 2021
epilepsy and have had no seizure in the past year) with adults 713 Post-COVID Conditions Among Adult COVID-19
without epilepsy indicated that adults with active or inactive Survivors Aged 18–64 and ≥65 Years — United
epilepsy were more likely to have Medicaid or other public States, March 2020–November 2021
insurance coverage and to report an inability to afford prescrip- 718 Notes from the Field: Self-Reported Health
tion medicine, specialty care, or vision or dental care. Adults Symptoms Following Petroleum Contamination of
with active or inactive epilepsy were more likely to take less Drinking Water System — Honolulu, Hawaii,
medication than prescribed to save money, to be in families November 2021–February 2022
having problems paying medical bills, and to report delaying 720 QuickStats

* Deceased.
† https://www.nap.edu/read/13379/chapter/3#25 Continuing Education examination available at
https://www.cdc.gov/mmwr/mmwr_continuingEducation.html

U.S. Department of Health and Human Services

Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

answered three questions about epilepsy to self-identify as a prevalences of selected access-to-care indicators§§ were com-
person with active, inactive, or no epilepsy (epilepsy status).¶ pared between adults with active epilepsy and no epilepsy and
These case-ascertainment questions have been validated for between those with inactive epilepsy and no epilepsy. Age-
use in community surveillance (5). Information about access standardized percentages of adults with active, inactive, and no
to health care and income was collected in the NHIS Sample epilepsy who were in families having problems paying medical
Adult, Person, and Imputed Income files. In 2015 and 2017, a bills in the past year were calculated by selected sociodemo-
total of 33,672 adults (final response rate = 55.2%) and 26,742 graphic characteristics. Analyses were conducted with SAS-
adults (final response rate = 53.0%), respectively, responded callable SUDAAN (version 9.4; SAS Institute) to account for
to the survey.** CDC pooled 2015 and 2017 data (combined the respondent sampling weights and NHIS complex sample
response rate = 54.1%) to increase the reliability of estimates. design. All reported differences are statistically significant
Estimates were weighted and age-standardized to the 2000 (p<0.05 by two-tailed t-tests). After excluding respondents with
U.S. Census Bureau projected adult population using three missing information on epilepsy history (i.e., respondents who
age groups: 18–44, 45–64, and ≥65 years.†† Age-standardized refused to respond or responded “don’t know” to the question
“Have you ever been told by a doctor or other health profes-
¶ 1) “Have you ever been told by a doctor or other health professional that you sional that you have a seizure disorder or epilepsy?”), the final
have a seizure disorder or epilepsy?” 2) “Are you currently taking any medicine
to control your seizure disorder or epilepsy?” 3) “Think back to last year about §§ NHIS
the same time. About how many seizures of any type have you had in the past Person file (Family questionnaire) access-to-care indicators include
year?” Active epilepsy was defined as having a diagnosis of epilepsy and either insurance type; respondent or a family member having problems paying
taking medication, having had one or more seizures in the past year, or both. medical bills; and having medical bills that cannot be paid at all. Sample
Inactive epilepsy was defined as adults who reported a history of epilepsy but Adult file access-to-care indicators include no transportation to get to a
were not taking medication for epilepsy and had not had a seizure in the past doctor’s office in the past 12 months; trouble finding a doctor/provider in
year. Adults with no epilepsy were those who answered no history of ever the past 12 months; couldn’t afford seeing a specialist in the past year; couldn’t
having received a diagnosis of epilepsy or seizure disorder by a doctor or afford mental health care or counseling in the past 12 months; had an
health professional. emergency department visit because of not having another place to go (among
** https://nhis.ipums.org/nhis/resources/srvydesc2015.pdf; https://ftp.cdc.gov/ adults who had an emergency department visit in the past year); couldn’t
pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/2017/ afford dental care or eyeglasses in the past 12 months; couldn’t afford
srvydesc.pdf prescription medicine in the past 12 months; and skipped medication doses/
†† https://www.cdc.gov/nchs/data/statnt/statnt20.pdf took less/delayed filling prescription to save money in the past 12 months.

The MMWR series of publications is published by the Center for Surveillance, Epidemiology, and Laboratory Services, Centers for Disease Control and Prevention (CDC),
U.S. Department of Health and Human Services, Atlanta, GA 30329-4027.
Suggested citation: [Author names; first three, then et al., if more than six.] [Report title]. MMWR Morb Mortal Wkly Rep 2022;71:[inclusive page numbers].
Centers for Disease Control and Prevention
Rochelle P. Walensky, MD, MPH, Director
Debra Houry, MD, MPH, Acting Principal Deputy Director
Daniel B. Jernigan, MD, MPH, Deputy Director for Public Health Science and Surveillance
Rebecca Bunnell, PhD, MEd, Director, Office of Science
Jennifer Layden, MD, PhD, Deputy Director, Office of Science
Leslie Dauphin, PhD, Director, Center for Surveillance, Epidemiology, and Laboratory Services
MMWR Editorial and Production Staff (Weekly)
Charlotte K. Kent, PhD, MPH, Editor in Chief Martha F. Boyd, Lead Visual Information Specialist Ian Branam, MA,
Brian A. King, PhD, MPH, Executive Editor Alexander J. Gottardy, Maureen A. Leahy, Acting Lead Health Communication Specialist
Jacqueline Gindler, MD, Editor Julia C. Martinroe, Stephen R. Spriggs, Tong Yang, Shelton Bartley, MPH, Leslie Hamlin,
Paul Z. Siegel, MD, MPH, Associate Editor Visual Information Specialists Lowery Johnson, Amanda Ray,
Mary Dott, MD, MPH, Online Editor Quang M. Doan, MBA, Phyllis H. King, Health Communication Specialists
Terisa F. Rutledge, Managing Editor Terraye M. Starr, Moua Yang, Will Yang, MA,
Teresa M. Hood, MS, Lead Technical Writer-Editor Information Technology Specialists Visual Information Specialist
Glenn Damon, Soumya Dunworth, PhD,
Tiana Garrett-Cherry, PhD, MPH, Srila Sen, MA,
Stacy Simon, MA, Jesse Sokolow, Morgan Thompson,
Technical Writer-Editors
MMWR Editorial Board
Timothy F. Jones, MD, Chairman
Matthew L. Boulton, MD, MPH David W. Fleming, MD Patrick L. Remington, MD, MPH
Carolyn Brooks, ScD, MA William E. Halperin, MD, DrPH, MPH Carlos Roig, MS, MA
Jay C. Butler, MD Jewel Mullen, MD, MPH, MPA William Schaffner, MD
Virginia A. Caine, MD Jeff Niederdeppe, PhD Morgan Bobb Swanson, BS
Jonathan E. Fielding, MD, MPH, MBA Celeste Philip, MD, MPH Abbigail Tumpey, MPH
Patricia Quinlisk, MD, MPH

698 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

analytical sample included 60,281 (99.0%) respondents. This care because of lack of transportation. A greater percentage of
activity was reviewed by CDC and was conducted consistent adults with active or inactive epilepsy had difficulty affording a
with applicable federal law and CDC policy.¶¶ specialist (7.5% and 7.3%, respectively) than did those without
During 2015 and 2017, adults were less likely to be unin- epilepsy (4.1%); a similar pattern was observed for affording
sured if they had active (6.5%) epilepsy compared with those mental health care.
without epilepsy (11.0%) (Table 1). Adults with active or inac- Adults with active or inactive epilepsy were more likely to
tive epilepsy were less likely to have private insurance (39.3% report an inability to afford prescription medicine (13.2% and
and 53.9%, respectively) and more likely to have Medicaid or 12.4%), skipping medication doses to save money (9.3% and
other public health insurance coverage (44.4% and 27.3%, 12.9%), delaying obtaining refills (12.2% and 14.9%), taking
respectively) than were those without epilepsy (64.9% [private] less than the prescribed dosages of medicine to save money
and 15.6% [Medicaid or other public health insurance]). More (10.8% and 11.6%), and being unable to afford dental care
adults with active epilepsy than without epilepsy had trouble (20.3% and 19.3%) compared with those without epilepsy
finding a doctor or other health care provider (5.4% versus (6.1%, 6.1%, 8.3%, 6.4%, and 10.7%, respectively). Adults with
3.1%). More adults with active epilepsy (6.9%) or inactive active epilepsy were more likely to report an inability to afford
epilepsy (4.9%) than without epilepsy (1.7%) reported delayed eyeglasses (12.5%) than were those without epilepsy (5.9%).
Adults with active or inactive epilepsy were overall signifi-
¶¶ 45C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 cantly more likely to be in families having problems paying
U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq. their medical bills (27.9% and 27.6%, respectively) than were
TABLE 1. Crude and age-standardized prevalences* of indicators of limitations in access to care among adults aged ≥18 years, by epilepsy
status — National Health Interview Survey, United States, 2015 and 2017
% (95% CI)
Active epilepsy† (n = 735) Inactive epilepsy§ (n = 456) No epilepsy¶ (n = 59,090)
Characteristic Crude Age-standardized Crude Age-standardized Crude Age-standardized
Current insurance type
Private 39.2 (34.4–44.2) 39.3 (34.4–44.4)** 55.6 (49.7–61.4) 53.9 (48.0–59.6)** 64.4 (63.7–65.1) 64.9 (64.2–65.6)
Medicaid/Other public†† 46.6 (41.7–51.5) 44.4 (39.8–49.1)** 27.9 (22.8–33.7) 27.3 (22.2–33.0)** 15.4 (14.9–15.8) 15.6 (15.1–16.1)
Medicare 8.0 (6.1–10.5) 9.8 (8.2–11.7) 5.5 (3.7–8.1) 8.0 (6.0–10.5) 10.0 (9.6–10.3) 8.6 (8.4–8.8)
Uninsured 6.3 (4.2–9.3) 6.5 (4.4–9.7)** 10.9 (10.6–15.4) 10.8 (7.6–15.3) 10.3 (9.9–10.7) 11.0 (10.5–11.4)
Reasons for not seeking care, paying medical bills, or obtaining prescriptions when needed during the last 12 months
Lack of transportation 7.11 (5.4–9.3) 6.9 (5.2–9.1)** 5.0 (3.3–7.6) 4.9 (3.2–7.4)** 1.8 (1.6–1.9) 1.7 (1.6–1.9)
Trouble finding provider who would 5.7 (3.9–8.0) 5.4 (3.7–7.8)** 4.5 (2.7–7.4) 4.4 (2.7–7.2) 3.1 (2.9–3.3) 3.1 (2.9–3.3)
see them
Could not afford to see a specialist 7.6 (5.4–10.7) 7.5 (5.3–10.6)** 7.6 (5.2–10.9) 7.3 (5.0–10.4)** 4.1 (3.9–4.4) 4.1 (3.9–4.4)
Could not afford mental health care 4.4 (2.9–6.8) 4.4 (2.8–6.8)** 5.6 (3.2–9.4) 5.4 (3.1–9.3)** 1.9 (1.8–2.1) 2.0 (1.9–2.2)
or counseling
Last ED visit because didn’t have another 36.4 (29.8–43.6) 38.4 (31.2–46.1) 44.7 (34.8–55.0) 43.7 (34.4–53.5) 39.6 (38.2–40.9) 40.3 (38.9–41.8)
place to go§§
Problems paying medical bills¶¶ 28.4 (24.0–33.2) 27.9 (23.5–32.8)** 27.8 (22.9–33.2) 27.6 (22.7–33.1)** 13.8 (13.4–14.3) 14.0 (13.6–14.5)
Could not afford prescription medicines 13.6 (10.7–17.2) 13.2 (10.3–16.7)** 12.6 (9.6–16.3) 12.4 (9.3–16.2)** 6.1 (5.8–6.3) 6.1 (5.8–6.3)
Skipped medication doses to save money 9.7 (7.1–13.2) 9.3 (6.7–12.7)** 13.1 (9.4–18.0) 12.9 (8.9–18.2)** 5.9 (5.6–6.2) 6.1 (5.8–6.5)
Took less medicine to save money 10.5 (7.8–14.0) 10.8 (8.1–14.2)** 12.2 (8.6–16.9) 11.6 (7.9–16.6)** 6.2 (5.8–6.5) 6.4 (6.1–6.8)
Delayed filling prescription to save money 12.2 (9.4–15.6) 12.2 (9.4–15.7)** 15.4 (11.5–20.5) 14.9 (10.8–20.2)** 7.7 (7.4–8.1) 8.3 (7.9–8.7)
Could not afford dental care 20.9 (17.4–25.0) 20.3 (16.8–24.3)** 19.5 (15.4–24.5) 19.3 (15.1–24.2)** 10.7 (10.3–11.0) 10.7 (10.3–11.1)
Could not afford eyeglasses 12.9 (10.1–16.3) 12.5 (9.7-–5.9)** 10.2 (7.3–14.0) 9.6 (6.0–13.4) 6.0 (5.7–6.3) 5.9 (5.6–6.3)
Abbreviation: ED = emergency department.
* The percentage estimates are weighted. Estimates are age-standardized to the 2000 U.S. Census Bureau projected population, aged ≥18 years, using three age
groups: 18–34, 35–64, and ≥65 years.
† Active epilepsy was defined as adults who answered that a doctor or health professional had ever told them they had a seizure disorder or epilepsy and also
reported taking medication, having had one or more seizures in the past year, or both.
§ Inactive epilepsy was defined as adults who reported a history of epilepsy but were not taking medication for epilepsy and had not had a seizure in the past year.
¶ No epilepsy was defined as adults who answered no history of ever having been diagnosed with epilepsy or seizure disorder by a doctor or health professional.
** A t-test was conducted to compare the prevalence estimates between adults with active epilepsy and without epilepsy and between adults with inactive epilepsy
and without epilepsy in the same category of indicator of access to care at the statistical significance level of 0.05 (p<0.05 by two-tailed t-tests).
†† Other public included state sponsored or state and federal jointly sponsored children’s health insurance program and any type of military coverage with or without
Medicare or other government programs.
§§ Among adults with at least one ED visit in the past year.
¶¶ Problems paying bills was defined as answering “yes” to any of the following questions: “Did you/anyone in the family have problems paying or were unable
to pay any medical bills in the past 12 months?” (this could include bills for doctors, dentists, hospitals, therapists, medication, equipment, nursing home, or
home care) or “Do you/does anyone in your family currently have any medical bills that you are unable to pay at all?”

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 699
 Morbidity and Mortality Weekly Report

TABLE 2. Numbers and age-standardized percentages* of living in a family having problems paying medical bills† in the past year among
adults aged ≥18 years, by epilepsy status — National Health Interview Survey, United States, 2015 and 2017
Active epilepsy§ (n = 735) Inactive epilepsy¶ (n = 456) No epilepsy** (n = 59,090)
No./Total Age-standardized No./Total Age-standardized No./Total Age-standardized
Characteristic no.†† % (95% CI) no.†† % (95% CI) no.†† % (95% CI)
Total 202/735 27.9 (23.5–32.8)§§ 112/456 27.6 (22.7–33.1)§§ 7,768/59,090 14.0 (13.6–14.5)
Age group, yrs
18–44 85/284 30.9 (24.1–38.6)§§ 47/187 28.7 (21.1–37.8)§§ 3,535/23,846 15.4 (14.7–16.0)
45–64 97/298 30.5 (24.4–37.3)§§ 50/178 27.8 (20.7–36.1)§§ 3,083/19,818 15.1 (14.4–15.9)
≥65 20/153 14.1 (7.2–26.0)¶¶ 15/91 23.7 (14.3–36.5)§§ 1,150/15,426 8.0 (7.4–8.6)
Sex
Men 81/333 24.5 (18.2–32.2)§§ 46/166 31.5 (23.0–41.3)§§ 3,098/26,606 12.61 (12.0–13.2)
Women 121/402 31.4 (25.7–37.8)§§ 66/290 25.5 (19.6–32.4)§§ 4,670/32,484 15.4 (14.8–16.0)
Race/Ethnicity
White, non-Hispanic 129/517 25.6 (20.5–31.6)§§ 78/330 26.0 (20.5–32.4)§§ 4,477/38,553 12.7 (12.2–13.2)
Black, non-Hispanic 31/99 34.3 (23.4–47.1)§§ 15/ 58 42.5 (26.9–59.8)§§ 1,391/7,131 20.3 (19.0–21.8)
Hispanic 23/71 34.6 (21.9–50.0)§§ 15/42 40.7 (26.3–56.9)§§ 1,436/8,705 16.9 (15.8–18.1)
Other*** 19/48 36.7 (20.9–55.9)§§ —††† —††† 464/4,701 10.0 (8.9–11.3)
Poverty status§§§
<100% (of FPL) 77/227 33.9 (26.6–42.0)§§ 32/107 37.7 (26.4–50.4)§§ 1,646/8,807 20.5 (19.2–21.9)
≥100% to <200% 70/196 43.4 (33.3–54.0)§§ 37/107 40.3 (29.9–51.7)§§ 2,410/11,433 23.5 (22.4–24.6)
≥200% to <300% 32/102 36.0 (23.7–50.6)§§ 21/68 32.6 (18.9–50.2) 1,571/9,697 19.0 (17.8–20.2)
≥300% to <400% 14/80 17.9 (10.3–29.3) 8/45 24.1 (11.0–45.0)¶¶ 912/7,437 14.3 (13.1–15.7)
≥400% 9/130 5.1 (2.3–10.7)¶¶ 14/129 15.2 (8.7–25.2)§§ 1,229/21,715 6.3 (5.8–6.8)
Education level
Less than HS graduate 40/152 30.7 (22.3–40.7)§§ 20/71 30.9 (19.8–44.8) 1,321/7,440 21.1 (19.6–22.6)
HS graduate or equivalent 53/217 26.2 (18.4–35.8)§§ 33/119 32.2 (22.1–44.3)§§ 2,151/14,423 16.8 (15.9–17.8)
Some college or more 106/356 26.1 (20.3–32.8)§§ 58/263 24.0 (18.0–31.1)§§ 4,270/37,015 11.7 (11.3–12.2)
Current employment
Yes 47/192 18.7 (12.9–26.2) 50/209 26.8 (19.6–35.6)§§ 4,446/34,524 12.6 (12.0–13.1)
No 155/543 31.5 (25.8–37.7)§§ 62/247 28.3 (21.4–36.4)§§ 3,320/24,543 17.8 (16.9–18.7)
Marital status
Married/Living with partner 68/268 24.9 (18.7–32.2)§§ 44/188 26.1 (19.2–34.4)§§ 3,706/29,705 13.2 (12.6–13.7)
Widowed/Divorced/Separated 76/240 33.7 (23.0–45.2)§§ 44/145 38.1 (26.0–52.0)§§ 2,340/15,911 19.4 (18.0–20.7)
Never married 57/226 26.0 (19.0–34.7)§§ 24/123 24.5 (15.3–36.7)§§ 1,709/13,357 13.5 (12.6–14.5)
Region
Northeast 20/121 12.9 (6.0–25.6)¶¶ 12/58 22.3 (11.0–40.2)¶¶ 1,092/9,727 11.4 (10.4–12.3)
Midwest 36/161 26.1 (16.2–39.2) 32/121 29.4 (20.3–40.5)§§ 1,724/13,137 15.1 (14.2–16.0)
South 97/ 287 36.3 (29.4–43.8)§§ 49/163 36.3 (27.5–46.1)§§ 3,251/21,005 16.6 (15.8–17.4)
West 49/166 24.1 (16.4–33.9)§§ 19/114 17.2 (10.0–28.1) 1,701/15,221 11.1 (10.3–12.1)
Abbreviations: FPL = federal poverty level; HS = high school.
* The percentage estimates are weighted. Age-standardized to the 2000 U.S. Census Bureau projected population, aged ≥18 years, using three age groups: 18–44,
45–64, and ≥65 years. Estimates for age groups are not age-standardized (i.e., presented as crude percentages).
† Problem paying bills was defined as answering “yes” to any of the following questions: “Did you/anyone in the family have problems paying or were unable
to pay any medical bills in the past 12 months?” (this could include bills for doctors, dentists, hospitals, therapists, medication, equipment, nursing home, or
home care); or “Do you/does anyone in your family currently have any medical bills that you are unable to pay at all?”
§ Active epilepsy was defined as having a diagnosis of epilepsy and either taking medication, having had one or more seizures in the past year, or both.
¶ Inactive epilepsy was defined as adults who reported a history of epilepsy but were not taking medication for epilepsy and had not had a seizure in the past year.
** No epilepsy was defined as adults who answered no history of ever having been diagnosed with epilepsy or seizure disorder by a doctor or health professional.
†† “Total number” represents unweighted numbers of those with active epilepsy, inactive epilepsy, or no epilepsy (denominator); “number” represents unweighted
numbers of those living in a family having problems paying bills among those with active epilepsy, inactive epilepsy, or no epilepsy (numerator). Some of the
categories do not sum to the total (e.g., education level or marital status) and categories might not sum to the sample total because of missing responses.
§§ A t-test was conducted to compare the prevalence estimates between adults with active epilepsy and without epilepsy and between adults with inactive epilepsy
and without epilepsy in the same category of characteristics at the statistical significance level of 0.05 (p<0.05 by two-tailed t-tests).
¶¶ Estimate is unreliable because the relative SE was >30% but <50%. Results should be interpreted with caution.
*** The Other race and ethnicity category includes other non-Hispanic groups (American Indian or Alaskan Native, Asian, multiple race, and race group not releasable).
††† Number and estimate were suppressed because denominator was <30 or relative SE was >50%.
§§§ Poverty status was defined as the ratio of family income to federal poverty level. Estimates were calculated from the National Health Interview Survey
income data file.

700 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

adults without epilepsy (14%) (Table 2). Selected subgroups of

Summary
adults with active or inactive epilepsy (e.g., those aged <65 years
and in a family earning <200% of the federal poverty status) What is already known about this topic?
were also more likely to be in families having problems paying Adults with epilepsy are more likely to experience barriers to
accessing health care than are adults without epilepsy.
medical bills.
What is added by this report?
Discussion In 2015 and 2017, compared with U.S. adults without epilepsy,
Healthy People 2030 objectives include reducing the propor- adults with active or inactive epilepsy were more likely to report
an inability to afford prescription medicine, specialty care, or
tion of persons who cannot obtain needed medical care and other types of care, had trouble finding a doctor, delayed care
reducing the proportion of persons who cannot obtain neces- because of transportation barriers, or were in families having
sary prescription medicines (6). Persons with epilepsy need problems paying medical bills.
access to medical care for both epilepsy (e.g., access to anti- What are the implications for public health practice?
seizure medication and neurologists) and nonepilepsy-related Public health practitioners and epilepsy health and social
medical care (e.g., access to dental and vision care) to prevent service providers can enhance linkages between clinical and
comorbidity, worsening health status, and early mortality (7). community programs and services to address gaps in access to
The findings in this study indicate a broad range of barriers for health care.
both epilepsy- and nonepilepsy-related medical care that might
complicate epilepsy management and increase comorbidities, problems paying medical bills than their counterparts without
hospitalizations, disability, and health care costs for those living epilepsy, assessing differences in problems paying medical
with the disorder as well as for those with a history of epilepsy. bills within racial and ethnic groups requires more study with
Consistent with a previous study based on 2010 and 2013 larger samples. Additional studies are warranted to examine
NHIS data, adults with active epilepsy were more likely to health inequities associated with race and ethnicity and social
be insured with Medicaid or other public insurance coverage determinants of health by epilepsy status. Finally, an epilepsy
than were those without epilepsy (3). Medicare coverage might diagnosis earlier in life has been reported to alter neurodevel-
afford some protection against problems paying medical bills opment and might limit opportunities later in life (10). More
for adults with active epilepsy aged ≥65 years compared with studies are needed to examine the challenges faced by adults
younger adults with epilepsy who are not eligible for Medicare. with inactive epilepsy.
However, a 2013 study found that fewer California adults with The findings in this report are subject to at least five limita-
active epilepsy who had Medicare or Medicaid obtained special- tions. First, because NHIS is cross-sectional, causal inferences
ized epilepsy care compared with adults with private insurance related to the association between health care access barriers and
(4). Medicaid expansion reduced cost-related barriers to care and epilepsy status cannot be made. Second, estimates are based on
was associated with improvements in selected health outcomes self-reported data and might be subject to reporting bias. Third,
among low-income adults with chronic disease (8). Medicaid because adults aged ≥65 years without private insurance can
coverage for those who qualify includes mandatory benefits (e.g., have both Medicare and Medicaid coverage, the percentages
outpatient hospital services) and optional benefits (e.g., prescrip- of adults with Medicare might include some who are eligible
tion drugs, non-emergency medical transportation, dental care, for Medicaid (i.e., “dually eligible”), potentially leading to
and optometry care), which vary by state. The extent to which an underestimate of the overall percentages with Medicaid
services are covered by Medicaid might facilitate or limit access coverage by epilepsy status. Fourth, it is not known whether
to these services for adults with active epilepsy.*** problems paying medical bills reported by a respondent with
Other individual-level factors such as sex, presence of comor- active or inactive epilepsy are related to the respondent’s own
bidities, or health literacy and contextual factors that constitute medical bills or those of other family members.††† Finally,
social determinants of health (e.g., reliable transportation, because more recent data are not available, the findings from
provider availability or cultural competency, and lower rates this analysis might not represent associations of these factors
of public insurance reimbursement) might also influence epi- beyond 2017. The ongoing efforts for data modernization
lepsy care and outcomes (4,9). Although all racial and ethnic and enhanced linkages with electronic health records might
groups with active epilepsy were more likely to report having improve availability of more data to guide public health action.

*** https://www.medicaid.gov/medicaid/benefits/index.html ††† https://www.cdc.gov/nchs/data/nhis/earlyrelease/probs_paying_medical_

bills_jan_2011_jun_2017.pdf

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 701
 Morbidity and Mortality Weekly Report

Public health practitioners and epilepsy health and social References

service providers can raise awareness of the CDC-supported 1. Zack MM, Kobau R. National and state estimates of the numbers of
Epilepsy Foundation Epilepsy and Seizures 24/7 Helpline, adults and children with active epilepsy—United States, 2015. MMWR
Morb Mortal Wkly Rep 2017;66:821–5. PMID:28796763 https://doi.
which has trained English- and Spanish-speaking informa- org/10.15585/mmwr.mm6631a1
tion specialists available 24 hours a day by phone and email 2. Kalilani L, Faught E, Kim H, et al. Assessment and effect of a gap between
to refer persons to local community-based programs such as new-onset epilepsy diagnosis and treatment in the US. Neurology
medication assistance programs, transportation services, and 2019;92:e2197–208. PMID:30971487 https://doi.org/10.1212/
WNL.0000000000007448
other resources.§§§ The Epilepsy Foundation also provides 3. Thurman DJ, Kobau R, Luo YH, Helmers SL, Zack MM. Health-care
information to assist patients in finding epilepsy centers and access among adults with epilepsy: the U.S. National Health Interview
specialists nationwide.¶¶¶ Addressing disparities in access to Survey, 2010 and 2013. Epilepsy Behav 2016;55:184–8. PMID:26627980
https://doi.org/10.1016/j.yebeh.2015.10.028
care necessitates a comprehensive approach that accounts for 4. Schiltz NK, Koroukian SM, Singer ME, Love TE, Kaiboriboon K.
social determinants of health (6,9) and intervenes to reduce Disparities in access to specialized epilepsy care. Epilepsy Res
treatment gaps. Public health practitioners and epilepsy health 2013;107:172–80. PMID:24008077 https://doi.org/10.1016/j.
eplepsyres.2013.08.003
and social service providers can enhance linkages between clini- 5. Brooks DR, Avetisyan R, Jarrett KM, et al. Validation of self-
cal and community programs and services to address gaps in reported epilepsy for purposes of community surveillance. Epilepsy
access to health care. Behav 2012;23:57–63. PMID:22189155 https://doi.org/10.1016/j.
yebeh.2011.11.002
6. Office of Disease Prevention and Health Promotion. Healthy people
§§§ https://www.epilepsy.com/connect/247-helpline 2030: access to health services. Rockville, MD: US Department of Health
¶¶¶ https://www.epilepsy.com/connect/find-epilepsy-specialist and Human Services; 2020. https://health.gov/healthypeople/objectives-
and-data/social-determinants-health/literature-summaries/
Corresponding author: Niu Tian, vii9@cdc.gov, 770-488-5679. access-health-services
7. Wilson DA, Malek AM, Wagner JL, Wannamaker BB, Selassie AW.
1Division of Population Health, National Center for Chronic Disease Prevention Mortality in people with epilepsy: a statewide retrospective cohort study.
and Health Promotion, CDC. Epilepsy Res 2016;122:7–14. PMID:26900886 https://doi.
org/10.1016/j.eplepsyres.2016.01.008
All authors have completed and submitted the International 8. Torres H, Poorman E, Tadepalli U, et al. Coverage and access for
Committee of Medical Journal Editors form for disclosure of potential Americans with chronic disease under the Affordable Care Act: a quasi-
conflicts of interest. No potential conflicts of interest were disclosed. experimental study. Ann Intern Med 2017;166:472–9. PMID:28114684
https://doi.org/10.7326/M16-1256
9. Szaflarski M. Social determinants of health in epilepsy. Epilepsy Behav
2014;41:283–9. PMID:24998313 https://doi.org/10.1016/
j.yebeh.2014.06.013
10. Hermann B, Seidenberg M, Jones J. The neurobehavioural
comorbidities of epilepsy: can a natural history be developed? Lancet
Neurol 2008;7:151–60. PMID:18207113 https://doi.org/10.1016/
S1474-4422(08)70018-8

702 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

Seizure- or Epilepsy-Related Emergency Department Visits Before and During

the COVID-19 Pandemic — United States, 2019–2021
Sanjeeb Sapkota, MBBS1; Elise Caruso, MPH2; Rosemarie Kobau, MPH3; Lakshmi Radhakrishnan, MPH4; Barbara Jobst, MD5; Jourdan DeVies, MS6;
Niu Tian, MD, PhD3; R. Edward Hogan, MD7; Matthew M. Zack, MD3,*; Daniel M. Pastula, MD3,8

Seizures, transient signs or symptoms caused by abnormal need to encourage persons to seek appropriate care for seizure-
surges of electrical activity in the brain, can result from epilepsy, related emergencies** to prevent adverse outcomes.
a neurologic disorder characterized by abnormal electrical NSSP collects deidentified electronic health record data from
brain activity causing recurrent, unprovoked seizures, or from EDs and other health care settings. ED visit data are derived
other inciting causes, such as high fever or substance abuse from a subset of approximately 71% of the nation’s nonfed-
(1). Seizures generally account for approximately 1% of all eral EDs (i.e., EDs not supported by the Veterans Health
emergency department (ED) visits (2,3). Persons of any age Administration or U.S. Department of Defense). Diagnosis
can experience seizures, and outcomes might range from no codes from the International Classification of Diseases, Ninth
complications for those with a single seizure to increased risk Revision, Clinical Modification (ICD-9-CM) and International
for injury, comorbidity, impaired quality of life, and early mor- Classification of Diseases, Tenth Revision, Clinical Modification
tality for those with epilepsy (4). To examine trends in weekly (ICD-10-CM), Systematized Nomenclature of Medicine, and
seizure- or epilepsy-related (seizure-related) ED visits† in the relevant free-text reason for visit (chief complaint) terms were
United States before and during the COVID-19 pandemic, used to identify seizure-related ED visits (Supplementary Table,
CDC analyzed data from the National Syndromic Surveillance https://stacks.cdc.gov/view/cdc/117412) (Supplementary
Program (NSSP).§ Seizure-related ED visits decreased abruptly Box, https://stacks.cdc.gov/view/cdc/117573). All analyses
during the early pandemic period. By the end of 2020, seizure- were restricted to EDs that reported consistently more com-
related ED visits returned almost to prepandemic levels for per- plete data throughout the study period (January 1, 2019–
sons of all ages, except children aged 0–9 years. By mid-2021, December 31, 2021); 56% of EDs sharing data with NSSP met
however, this age group gradually returned to baseline as well. these criteria.†† CDC assessed trends by six age groups (0–9,
Reasons for the decrease in seizure-related ED visits in 2020 10–19, 20–39, 40–59, 60–69, and ≥70 years) and visualized
among all age groups and the slow return to baseline among age-specific trends of weekly seizure-related ED visits during
children aged 0–9 years compared with other age groups are 2019–2021. Using R (version 4.1.2; The R Foundation), CDC
unclear. The decrease might have been associated with fear of quantified change in mean weekly seizure-related ED visits
exposure to COVID-19 infection in EDs deterring parents or during April 1–December 29 across 3 years: 2019, 2020, and
guardians of children from seeking care, adherence to mitiga- 2021; results were stratified by age group and sex. Percentage
tion measures including avoiding public settings such as EDs, change in mean weekly seizure-related ED visits was assessed by
or increased access to telehealth services decreasing the need comparing 2020 data with corresponding data from 2019 and
for ED visits (5). These findings reinforce the importance of 2021. This activity was reviewed by CDC and was conducted
understanding factors associated with ED avoidance among consistent with applicable federal law and CDC policy.§§
persons with epilepsy or seizure, the importance that all eligible All ED visits, including seizure-related ED visits, decreased
persons be up to date¶ with COVID-19 vaccination, and the among all age groups and among both males and females
during the pandemic period April 1–December 29, 2020,
* Deceased. compared with the corresponding period in 2019 (Table). The
† Analysis was limited to ED encounters. As of December 31, 2021, the median
number of facilities included in the analysis was 2,031 (range = 1,986–2,038),
including data from 56% of all nonfederal EDs sharing data with NSSP. ** Includes a first-time seizure and status epilepticus, which is defined as a
§ NSSP is a collaboration among CDC, federal partners, local and state health
continuous seizure lasting >5 minutes or recurrent seizures without regaining
departments, and academic and private sector partners. NSSP receives consciousness between seizures.
deidentified electronic health data from 50 states representing approximately †† To limit the impact of data quality on trends, all analyses were restricted to
71% of nonfederal EDs nationwide, although <50% of ED facilities from facilities with a coefficient of variation ≤40 and percentage of weekly average
California, Hawaii, Iowa, Minnesota, Ohio, and Oklahoma currently informative discharge diagnosis ≥75 throughout the analysis period
participate in NSSP at the time of this analysis. (January 2019–December 2021) so that only consistently reporting facilities
¶ https://www.cdc.gov/coronavirus/2019-ncov/vaccines/stay-up-to-date.html
with more complete data were included. EDs that met these data quality
control criteria were included in the analysis.
§§ 45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5
U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 703
 Morbidity and Mortality Weekly Report

TABLE. Mean weekly seizure- or epilepsy-related emergency department visits and overall emergency department visits, by age and sex, and
percentage change* — National Syndromic Surveillance Program,† United States, April 1–December 29, 2019–2021
Mean weekly visits, no. (95% CI)§ % Change
Characteristic 2019 2020 2021 2019–2020 2020–2021
Seizure or epilepsy ED visits
Age group, yrs
0–9 2,759 (2,660–2,864) 1,553 (1,504–1,593) 2,528 (2,462–2,593) −44 63
10–19 1,893 (1,846–1,940) 1,413 (1,356–1,469) 1,749 (1,710–1,786) −25 24
20–39 7,102 (7,037–7,165) 6,143 (5,957–6,316) 6,579 (6,478–6,680) −13 7
40–59 6,476 (6,412–6,539) 5,701 (5,548–5,838) 5,769 (5,678–5,860) −12 1
60–69 2,588 (2,561–2,617) 2,423 (2,373–2,467) 2,495 (2,468–2,524) −6 3
≥70 2,641 (2,604–2,679) 2,504(2,441–2,561) 2,583 (2,557–2,613) −5 3
Sex
Female 11,422 (11,344–11,501) 9,327 (9,044–9,579) 10,373 (10,280–10,470) −18 11
Male 12,128 (12,039–12,236) 10,462 (10,214–10,694) 11,387 (11,296–11,470) −14 9
Total 23,588 (23,429–23,739) 19,824 (19,295–20,311) 21,800 (21,614–21,969) −16 10
All-cause ED visits
Age group, yrs
0–9 162,711 (154,767–171,195) 71,131 (67,015–74,824) 142,868 (137,805–147,822) −56 101
10–19 127,264 (123,781–130,677) 79,594 (74,870–84,171) 114,353 (111,036–117,884) −37 44
20–39 416,652 (413,210–420,159) 336,598 (322,674–348,693) 401,671 (394,081–409,796) −19 19
40–59 347,606 (344,299–350,816) 288,453 (278,532–297,426) 337,317 (331,750–342,781) −17 17
60–69 157,694 (156,596–158,946) 135,574 (130,804–139,547) 161,899 (160,116–163,865) −14 19
≥70 231,619 (230,000–233,699) 193,202 (185,523–199,808) 231,799 (229,713–233,852) −17 20
Sex
Female 797,473 (791,101–804,433) 593,418 (568,244–615,384) 755,769 (745,392–766,769) −26 27
Male 651,555 (646,948–656,594) 513,365 (494,989–530,303) 636,576 (627,504–646,651) −21 24
Total 1,451,717 (1,441,285–1,463,581) 1,109,069 (1,067,564–1,148,844) 1,395,349 (1,374,389–1,415,093) −24 26
Abbreviation: ED = emergency department.
* The percentage change in visits between the surveillance and reference periods (2019 [reference] versus 2020 [surveillance] and 2020 [reference] versus 2021
[surveillance]) was calculated as (ED visits during surveillance period – ED visits during reference period)/ED visits during reference period x 100%.
† The National Syndromic Surveillance Program receives anonymized medical record information from approximately 71% of nonfederal EDs nationwide. To reduce
artifactual impact from changes in reporting patterns, analyses were restricted to facilities with more consistent reporting of more complete data (coefficient of
variation ≤40 and average weekly informative discharge diagnosis ≥75% complete during 2019–2021).
§ CIs were constructed using the percentile bootstrap method using 1,000 replicate samples of the weekly counts. CIs were formed using the 2.5th and 97.5th
percentiles of the bootstrap distributions.

largest decline in seizure-related ED visits, noted as early as among whom the rebound to prepandemic levels was delayed
February 2020, was observed among children aged 0–9 years until approximately week 25 of 2021 (Figure 1). To examine
(Figure 1) (Figure 2). During April 1–December 29, 2020, the whether the decrease among children aged 0–9 years was asso-
number of weekly seizure-related ED visits declined by 16% ciated with pediatric febrile seizure burden, a posthoc analysis
overall to 19,824, from 23,588 during the same period¶¶ in was conducted. In children aged 0–9 years, febrile seizures
2019 (Table). Among children aged 0–9 years, the number of accounted for approximately one third of all seizure-related
seizure-related weekly ED visits declined by 44% to 1,553, ED visits in all 3 years (approximately 35%, 31%, and 33%
compared with 2,759 visits during the same period in 2019; in 2019, 2020, and 2021, respectively).
overall ED visits among children aged 0–9 years declined by
Discussion
56%, from 162,711 visits in 2019 to 71,131 in 2020. By the
first week of 2021, the number of seizure-related ED visits In this study of trends in seizure-related ED visits during the
among all age groups was close to respective prepandemic COVID-19 pandemic, seizure-related ED visits during the ini-
levels in 2019, with the exception of children aged 0–9 years, tial COVID-19 waves declined among all age groups, especially
among children aged 0–9 years. These findings are consistent
¶¶ Percentage change in visits during surveillance periods compared with reference with several other studies (6–8). In one analysis of U.S. ED vis-
periods (surveillance period April 1–December 29, 2020, compared with its during January 2019–May 2020, the number of weekly all-
reference period April 1–December 29, 2019, and surveillance period April 1– cause ED visits declined abruptly during March 29–April 25,
December 29, 2021, compared with reference period April 1–December 29,
2020) was calculated as (ED visits for seizures or epilepsy during surveillance 2020, along with a decline in ED visits among children aged
period – ED visits for seizures or epilepsy during reference period)/ED visits 0–9 years attributable to common conditions, including influ-
for seizures or epilepsy during reference period x 100%. enza, otitis media, upper respiratory conditions, asthma, viral

704 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

FIGURE 1. Weekly seizure- or epilepsy-related emergency department visits among persons aged <40 years, by age group* — National
Syndromic Surveillance Program,† United States, 2019–2021

4,000
0–9 yrs
3,500

3,000
No. of ED visits

2,500

2,000

1,500

1,000

500
2019 2020 2021
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Weeks
3,000
10–19 yrs

2,500

2,000
No. of ED visits

1,500

1,000

500
2019 2020 2021

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Weeks
9,000
20–39 yrs
8,000

7,000

6,000
No. of ED visits

5,000

4,000

3,000

2,000

1,000 2019 2020 2021

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Weeks

Abbreviation: ED = emergency department.

* The y-axis range differs for different age groups to account for different numbers of ED visits by these groups and to facilitate visualization of changes over time.
† The National Syndromic Surveillance Program receives deidentified medical record information from approximately 71% of nonfederal EDs nationwide. To reduce
artifactual impact from changes in reporting patterns, analyses were restricted to facilities with more consistent reporting of more complete data (coefficient of
variation ≤40 and average weekly informative discharge diagnosis ≥75% complete during 2019–2021).

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 705
 Morbidity and Mortality Weekly Report

FIGURE 2. Weekly seizure- or epilepsy-related emergency department visits among persons aged ≥40 years, by age group* — National
Syndromic Surveillance Program,† United States, 2019–2021

8,000
40–59 yrs
7,000

6,000
No. of ED visits

5,000

4,000

3,000

2,000

1,000 2019 2020 2021

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Weeks
4,000
60–69 yrs
3,500

3,000
No. of ED visits

2,500

2,000

1,500

1,000

500
2019 2020 2021
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Weeks
4,000
≥70 yrs
3,500

3,000

2,500
No. of ED visits

2,000

1,500

1,000

500 2019 2020 2021

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53
Weeks

Abbreviation: ED = emergency department.

* The y-axis range differs for different age groups to account for different numbers of ED visits by these groups and to facilitate visualization of changes over time.
† The National Syndromic Surveillance Program receives deidentified medical record information from approximately 71% of nonfederal EDs nationwide. To reduce
artifactual impact from changes in reporting patterns, analyses were restricted to facilities with more consistent reporting of more complete data (coefficient of
variation ≤40 and average weekly informative discharge diagnosis ≥75% complete during 2019–2021).

706 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

infection, respiratory symptoms, and fever (6). International

Summary
studies have described a reduction in seizure-related ED visits
What is already known about this topic?
among children during the COVID-19 pandemic, with one
study reporting a notable decline in febrile seizure–related ED Seizures or epilepsy account for 1% of annual emergency
department (ED) visits. Data on seizure- or epilepsy-related
visits among children aged 0–6 years (7,8). ED visits during the COVID-19 pandemic are limited.
The percentages of ED visits attributable to febrile seizures
What is added by this report?
among children aged 0–9 years in this study were relatively
Weekly seizure- or epilepsy-related ED visits decreased sharply
stable, therefore any changes in ED visits for febrile seizures
during the early pandemic period among all age groups,
during the study period were unlikely to explain the overall especially children aged 0–9 years. The return to prepandemic
change of trend in seizure-related ED visits in this age group. baseline in this group was delayed until mid-2021, longer than
Researchers in Italy examined selected causes for seizure-related other age groups.
ED visits during February 23–April 21, 2020 (e.g., first episode What are the implications for public health practice?
or breakthrough seizure), but could not attribute the observed These findings reinforce the importance of understanding factors
decrease in seizure-related ED visits to seizure type (e.g., febrile associated with ED avoidance among persons with epilepsy or
versus first episode seizures) (7). However, a limitation of the seizure, the importance that all eligible persons be up to date
Italian study was small sample size; thus, the findings warrant with COVID-19 vaccination, and the need to encourage persons
to seek appropriate care for seizure-related emergencies.
additional study. The findings related to febrile seizure–attrib-
utable ED use in the current report differ from, but supplement
growing research in this area (8). and do not capture treatment sought for seizures in other set-
In the present study, school closures and the need to shelter at tings. Finally, distinguishing initial seizure-related visits from
home could have facilitated heightened supervision of children subsequent visits was not possible, therefore the numbers of ED
while at home, including increased monitoring and promotion visits reported might represent multiple visits by one person.
of healthful behaviors reducing seizure risk (e.g., medication These findings reinforce the importance of understanding
adherence and regular sleep) or seizure sequelae (e.g., injury), factors associated with ED avoidance among persons with
thereby reducing the need for ED care (7,9). The decrease in epilepsy or seizures, and any alternative care approaches among
weekly seizure-related ED visits among children aged 0–9 years persons with epilepsy or seizures and the need to encourage
might also have been associated with concern about risk for persons to seek appropriate care for seizure-related emergen-
COVID-19 in EDs, deterring parents or guardians from seek- cies. Vaccination against SARS-CoV-2, the virus that causes
ing care for their children. It is also possible that expanded COVID-19, of all age-eligible persons, including those with
access and increased use of telehealth facilitated triaged tele- epilepsy, is recommended to protect against the adverse effects
phone support or virtual health care encounters, especially for of COVID-19 (9).
children with epilepsy and high-risk comorbidities, otherwise
Acknowledgments
obtained in EDs (5,10). Additional studies are warranted to
determine whether decreased in-person ED care for children Kathleen Hartnett, Michael Sheppard, Jonathan Wortham.
with seizures or epilepsy during the initial COVID-19 pan- Corresponding author: Sanjeeb Sapkota, auu6@cdc.gov.
demic was associated with any differences in risk for infection, 1Office of the Director, Center for Global Health, CDC; 2Division of
injury, or delayed care, seizure type, or other factors and any Tuberculosis Elimination, National Center for HIV/AIDS, Viral Hepatitis,
associations between these factors and adverse outcomes. STD, and TB Prevention, CDC; 3Division of Population Health, National
Center for Chronic Disease Prevention and Health Promotion, CDC; 4Division
The findings in this report are subject to at least four limi- of Health Informatics and Surveillance, Center for Surveillance, Epidemiology,
tations. First, because NSSP coverage varies both within and and Laboratory Services, CDC; 5Geisel School of Medicine, University of
Dartmouth, Hanover, New Hampshire; 6ICF International, Fairfax, Virginia;
across states, NSSP data are not nationally representative. In 7Washington University, St. Louis, Missouri; 8University of Colorado School
some states nearly all hospitals report, while in others only those of Medicine and Colorado School of Public Health, Aurora, Colorado.
in certain counties or health care systems report. Thus, these All authors have completed and submitted the International
findings might not be generalizable. Second, differences in Committee of Medical Journal Editors form for disclosure of
availability, coding practices, and reporting of chief complaints potential conflicts of interest. R. Edward Hogan reports institutional
and discharge diagnoses from facilities might influence trends. support for clinical trials from Otsuka Pharmaceutical Development,
To limit the impact of changing data volume and underlying Cerevel Therapeutics, and Biogen, Inc. Barbara Jobst reports grants
data quality on results, only data from hospitals with consis- from National Institutes of Health, U.S. Department of Defense,
tent reporting and more complete data were included in this Harvard Pilgrim, Inc, and Neuropace, Inc. No other potential
analysis. Third, trends displayed are restricted to ED visits only, conflicts of interest were disclosed.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 707
 Morbidity and Mortality Weekly Report

References 6. Hartnett KP, Kite-Powell A, DeVies J, et al.; National Syndromic

Surveillance Program Community of Practice. Impact of the COVID-19
1. Fisher RS, Cross JH, French JA, et al. Operational classification of seizure
pandemic on emergency department visits—United States, January 1,
types by the International League Against Epilepsy: position paper of
2019–May 30, 2020. MMWR Morb Mortal Wkly Rep 2020;69:699–704.
the ILAE Commission for Classification and Terminology. Epilepsia
PMID:32525856 https://doi.org/10.15585/mmwr.mm6923e1
2017;58:522–30. PMID:28276060 https://doi.org/10.1111/epi.13670
7. Davico C, Marcotulli D, Lux C, et al. Where have the children with
2. Bank AM, Bazil CW. Emergency management of epilepsy and seizures.
epilepsy gone? An observational study of seizure-related accesses to
Semin Neurol 2019;39:73–81. PMID:30743294 https://doi.
emergency department at the time of COVID-19. Seizure 2020;83:38–40.
org/10.1055/s-0038-1677008
PMID:33080483 https://doi.org/10.1016/j.seizure.2020.09.025
3. Pallin DJ, Goldstein JN, Moussally JS, Pelletier AJ, Green AR,
8. Chiu TGA, Leung WCY, Zhang Q, et al. Changes in pediatric seizure-
Camargo CA Jr. Seizure visits in US emergency departments:
related emergency department attendances during COVID-19 - a territory-
epidemiology and potential disparities in care. Int J Emerg Med
wide observational study. J Formos Med Assoc 2021;120:1647–51.
2008;1:97–105. PMID:19384659 https://doi.org/10.1007/
PMID:33248859 https://doi.org/10.1016/j.jfma.2020.11.006
s12245-008-0024-4
9. CDC. Guidance for COVID-19 prevention in K–12 schools. Atlanta,
4. Koh HK, Kobau R, Whittemore VH, et al. Toward an integrated public
GA: US Department of Health and Human Services, CDC; 2021.
health approach for epilepsy in the 21st century. Prev Chronic Dis
Accessed February 25, 2022. https://www.cdc.gov/coronavirus/2019-
2014;11:E146. PMID:25167091 https://doi.org/10.5888/pcd11.140270
ncov/community/schools-childcare/k-12-guidance.html
5. Cross JH, Kwon CS, Asadi-Pooya AA, et al.; ILAE Task Forces on
10. Kikuchi K, Hamano SI, Horiguchi A, et al. Telemedicine in epilepsy
COVID-19, Telemedicine. Epilepsy care during the COVID-19
management during the coronavirus disease 2019 pandemic. Pediatr Int
pandemic. Epilepsia 2021;62:2322–32. PMID:34428314 https://doi.
(Roma) 2022;64:e14972. PMID:34460985 https://doi.org/10.1111/
org/10.1111/epi.17045
ped.14972

708 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

Multistate Outbreak of Listeria monocytogenes Infections Linked to Fresh, Soft

Hispanic-Style Cheese — United States, 2021
Alexandra Palacios, MPH1,2; Mark Otto, MSPH3; Eileen Flaherty4; Michelle M. Boyle, MPH5; Lenka Malec, MPH6; Kelsey Holloman, MPH7;
Mabel Low, MPH1,2; Allison Wellman, MPH3; Corinne Newhart, MPH3; Lauren Gollarza, MHS1; Tracey Weeks, MS4; Anthony Muyombwe, PhD4;
Kristen Lozinak, MS5; Erin Kafka, MA5; Daniel O’Halloran, MPH6; Teresa Rozza6; David Nicholas, MPH9,10; Stranjae Ivory, MPH3;
Katherine Kreil, MPH3; Jasmine Huffman1,8; Laura Gieraltowski, PhD1; Amanda Conrad, MPH1

Listeriosis is a serious infection usually caused by eating food All three patients were hospitalized; no deaths were reported.
contaminated with the bacterium Listeria monocytogenes. An Patients were aged 45–69 years, and one patient was female. All
estimated 1,600 persons become ill with listeriosis each year, three patients reported Hispanic ethnicity. State partners inter-
among whom approximately 260 die. Persons at higher risk viewed patients or their surrogates using the Listeria Initiative
for listeriosis include pregnant persons and their newborns, questionnaire for hypothesis generation (4). All three patients
adults aged ≥65 years, and persons with weakened immune reported consuming fresh, soft Hispanic-style cheeses before
systems. Persons with invasive listeriosis usually report symp- becoming ill; two reported consuming queso fresco, a type of
toms starting 1–4 weeks after eating food contaminated fresh, soft Hispanic-style cheese. In this outbreak, a case was
with L. monocytogenes; however, some persons who become defined as an infection in a person with a clinical isolate related
infected have reported symptoms starting as late as 70 days within five allele differences by WGS and a specimen collection
after exposure or as early as the same day of exposure (1). On date from October 20, 2020, to March 17, 2021 (Figure).
January 29, 2021, PulseNet, the national molecular subtyp- Based on food histories from the three index patients,
ing surveillance network coordinated by CDC, identified a their reported Hispanic ethnicity, and the known associa-
multistate cluster of three L. monocytogenes infections: two tion between L. monocytogenes and fresh, soft Hispanic-style
from Maryland and one from Connecticut (2). CDC, the cheeses, CDC asked the Connecticut Department of Public
Food and Drug Administration (FDA), and state and local Health (CDPH) to contact the Connecticut patient for brand
partners began an investigation on February 1, 2021. A total information. During re-interview, the patient reported con-
of 13 outbreak-related cases were eventually identified from suming brand A queso fresco. CDC conducted a case-case
four states. All patients reported Hispanic ethnicity; 12 patients analysis comparing food exposures for four listeriosis patients
were hospitalized, and one died. Rapid food testing and record included in the outbreak (outbreak cases) with completed
collection by regulatory agencies enabled investigators to Listeria Initiative questionnaires to exposures for listeriosis
identify a brand of queso fresco made with pasteurized milk as patients not associated with an outbreak by WGS from the
the likely source of the outbreak, leading to an initial product same states as outbreak cases (control cases). An exact odds
recall on February 19, 2021. Fresh, soft Hispanic-style cheeses ratio analysis was conducted using SAS software (version 9.4;
made with pasteurized milk are a well-documented source of SAS Institute). Consumption of fresh, soft Hispanic-style
listeriosis outbreaks. These cheeses can be contaminated with cheeses (odds ratio [OR] = 30.4; p<0.001) and queso fresco
L. monocytogenes unless stringent hygienic controls are imple- (OR = 51.2; p = 0.002) were both statistically significant. This
mented, and the processing environment is monitored for activity was reviewed by CDC and was conducted consistent
contamination (3). U.S. public health agencies should establish with applicable federal law and CDC policy.*
or improve communications, including new methods of dis- A total of 13 L. monocytogenes infections that met the case
seminating information that also effectively reach Hispanic definition were reported from four states (Connecticut [one],
populations, to emphasize the risk from eating fresh, soft Maryland [five], New York [four], and Virginia [three]). Patients
Hispanic-style cheeses, even those made with pasteurized milk. ranged in age from <1 year to 75 years (median = 51 years). All
patients reported Hispanic ethnicity; seven were female. Twelve
Investigation and Results patients were hospitalized; one died. Four patients became ill
On February 1, 2021, CDC notified state and federal part- during pregnancy, resulting in two pregnancy losses and one
ners of three listeriosis illnesses from Maryland (two cases) premature birth; one patient remained pregnant after becom-
and Connecticut (one case) uploaded to PulseNet within the ing ill. Among the eleven patients who completed the Listeria
previous 120 days that were highly related (i.e., within four
alleles by whole genome sequencing [WGS]). Specimen collec- * 45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C.
Sect. 552a; 44 U.S.C. Sect. 3501 et seq.
tion dates ranged from October 20, 2020, to January 6, 2021.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 709
 Morbidity and Mortality Weekly Report

Initiative questionnaire, eight reported consuming fresh, soft

Summary
Hispanic-style cheeses; seven reported consuming queso fresco.
What is already known about this topic?
Four patients reported consuming brands of cheeses manufac-
tured by firm A, the firm that produces brand A queso fresco. Listeriosis outbreaks are frequently associated with consump-
tion of fresh, soft Hispanic-style cheeses.
The Connecticut Food Protection Program and Maryland
Rapid Response Team collected samples of fresh, soft Hispanic- What is added by this report?
style cheeses at stores reported by patients, including brand A In early 2021, a multistate outbreak of listeriosis involving
13 cases in four states occurred, resulting in 12 hospitalizations
queso fresco. Connecticut and Maryland collected and tested
and one death. The outbreak was linked to Hispanic-style
61 fresh, soft Hispanic-style cheese samples; two yielded cheese within 19 days of cluster detection. Rapid food testing
L. monocytogenes. CDPH identified the outbreak strain of by regulatory agencies in response to the investigation
L. monocytogenes in two samples of brand A queso fresco. WGS identified the implicated cheese.
analysis of isolates from the CDPH samples showed they were What are the implications for public health practice?
closely related to the clinical isolates (0–4 allele differences), To prevent severe health outcomes among persons at increased
suggesting that patients became ill from brand A queso fresco. risk for listeriosis, public health agencies should improve
All 13 clinical isolates and two cheese isolates were related communications, including implementing new methods of
within five allele differences by WGS. dissemination to emphasize the risk from eating fresh, soft
Hispanic-style cheeses, even those made with pasteurized milk.
Public Health Response
FDA determined that brand A queso fresco was produced been identified as the outbreak source as quickly. The public
by firm A, located in New Jersey, and initiated an inspec- health actions taken within 19 days of cluster identification,
tion. Firm A produced or handled various types of fresh, soft firm A’s voluntary recalls, and outbreak notices likely prevented
Hispanic-style cheeses under its own brand name and for additional illnesses or deaths.
private label brands. Firm A agreed to recall brand A queso In early 2020, during an unrelated outbreak of listeriosis,
fresco products with expiration dates from February 26 to Listeria grayi and Listeria innocua, typically nonpathogenic to
March 13, 2021. The initial recall and outbreak investigation humans, were found in firm A’s processing areas. The presence
were announced on February 19, 2021. Because of cross- of Listeria species in a processing environment indicates that
contamination concerns, firm A agreed on February 26 to L. monocytogenes could survive in that same environment. FDA
expand the recall to all types of Hispanic-style cheeses produced issued a warning letter to firm A in 2020 because of violations
or handled in the facility: queso fresco, requesón, and quesillo. of Current Good Manufacturing Practice regulations and a
As a result of this investigation, firm A ceased production, lack of hazard analysis and preventive control programs (6).
repackaging, and distribution of all products manufactured Fresh, soft Hispanic-style cheeses made with pasteurized milk
at the facility. continue to constitute a serious risk for listeriosis because cheeses
CDC published seven outbreak notices; FDA posted nine can become contaminated during the production process (after
outbreak advisories, two recall notices, and two lists of retail milk pasteurization). High moisture, low salt content, and low
establishments that received recalled product. CDC and FDA acidity support growth of L. monocytogenes in these cheeses dur-
communications were available in both English and Spanish. In ing refrigerated storage, thereby increasing the risk for illness
addition, Connecticut published four public communications. (7). A study of U.S. listeriosis outbreaks associated with soft
Two patients who became ill after the expanded recall, both with cheeses during 1998–2014 found that soft cheeses made with
specimen collection dates of March 17, 2021, likely purchased pasteurized milk are implicated in more outbreaks than soft
and consumed the queso fresco before the recall given their cheeses made with unpasteurized milk, which might be related
illness dates and the long incubation period for listeriosis (5). to higher consumption of cheese made with pasteurized milk or
to public health messages advising persons at higher risk for liste-
Discussion riosis not to eat cheeses made with unpasteurized milk. Among
Patients in this outbreak were more likely to consume fresh, 17 outbreaks linked to soft cheeses during 1998–2014, eleven
soft Hispanic-style cheeses, including queso fresco, compared were linked to Hispanic-style cheeses, three of which included
with patients with sporadic Listeria infections reported from cheeses made with unpasteurized milk. The six outbreaks not
the same states. In listeriosis outbreaks, prompt, epidemio- linked to Hispanic-style cheeses included sheep’s milk, Middle
logically directed food sampling plays a key role in identify- Eastern-style, Eastern European-style, Italian-style, blue-veined,
ing the source of illness. Without the rapid identification of and soft-ripened cheeses (8).
L. monocytogenes in firm A’s queso fresco, firm A would not have

710 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

FIGURE. Number of persons infected with the outbreak strain of Listeria monocytogenes, by date of specimen collection (n = 13) — United States,
October 20, 2020–March 17, 2021
3
Feb 11
Hispanic–style cheese Feb 16 Feb 17
and queso fresco CT cheese Firm A
statistically significant; samples initial recall
MD collects samples yield Lm (queso fresco)

Feb 1 Feb 10 Feb 12

2 Multistate CT collects CDC and FDA
investigation samples post first
No. of persons

begins outbreak Feb 19

notice Cheese
Jan 29 isolates
PulseNet match
detects outbreak
cluster strain
1
Feb 26
Firm A
expands
recall

0
20 20 20 10 12 14 15 16 22 26 1 6 9 14 23 9 17
Oct Nov Dec Jan Feb Mar
2020 2021
Specimen collection date
Abbreviations: CT = Connecticut; FDA = Food and Drug Administration; Lm = Listeria monocytogenes; MD = Maryland.

Fresh, soft Hispanic-style cheeses, especially those produced Corresponding author: Alexandra Palacios, omq5@cdc.gov, 470-426-4170.
in facilities with unhygienic processing conditions, have fre- 1Division of Foodborne, Waterborne, and Environmental Diseases, National
quently led to listeriosis outbreaks during the last two decades Center for Emerging and Zoonotic Infectious Diseases, CDC; 2Oak Ridge
(8). Rapid food testing by regulatory agencies in response to Institute for Science and Education, Oak Ridge, Tennessee; 3Food and Drug
Administration, Silver Spring, Maryland; 4Connecticut Department of Public
this outbreak investigation identified the implicated cheese. Health; 5Maryland Department of Health; 6New York City Department of
Public health agencies should establish or improve communica- Health and Mental Hygiene, New York; 7Virginia Department of Health;
8Weems Design Studio, Suwanee, Georgia; 9New York State Department of
tions, including new methods for disseminating information Health; 10Department of Epidemiology & Biostatistics, School of Public Health,
to emphasize the risk from eating fresh, soft Hispanic-style University at Albany, Rensselaer, New York.
cheeses, even those made with pasteurized milk, to persons at All authors have completed and submitted the International
higher risk for listeriosis, including pregnant persons and their Committee of Medical Journal Editors form for disclosure of potential
newborns, adults aged ≥65 years, and persons with weakened conflicts of interest. No potential conflicts of interest were disclosed.
immune systems.
References
Acknowledgments 1. CDC. Listeria (listeriosis) questions and answers. Atlanta, GA: US
Christine Applewhite, Kimberly Holmes-Talbot, Lauren Mank, Department of Health and Human Services, CDC; 2016. https://www.
Quyen Phan, Christina Turner, Connecticut Department of Public cdc.gov/listeria/faq.html
2. CDC. PulseNet: whole genome sequencing (WGS). Atlanta, GA: US
Health; Laura Kasper, Tamara Rissman, Connecticut Emerging Department of Health and Human Services, CDC; 2016. https://www.
Infections Program; Kyle Shannon, Maryland Department of cdc.gov/pulsenet/pathogens/wgs.html
Health; Lauren Turner, Virginia Division of Consolidated Laboratory 3. Food and Drug Administration. Draft guidance for industry: control of Listeria
Services; Christy Brennan, Virginia Department of Agriculture and monocytogenes in ready-to-eat foods. Silver Spring, MD: US Department of
Consumer Services; Jaime Ahn, Jennifer Corrieri, Helena Kim,  Loel Health and Human Services, Food and Drug Administration; 2017. https://
www.fda.gov/regulatory-information/search-fda-guidance-documents/
Muetter, Alan Talarsky, New Jersey Department of Health; Karen draft-guidance-industry-control-listeria-monocytogenes-ready-eat-foods
Anthony, Cornelius Brown, III, Andrew Ciaccia, Joanne DeClement, 4. CDC. Listeria (listeriosis) prevention. Atlanta, GA: US Department
Cary Green, Martin Guardia, Kyle Hobson, Laishan Lam, Ruark of Health and Human Services, CDC; 2019. https://www.cdc.gov/
Lanham, Melissa Libby, Jessica Monteiro, Valeria Moore, William listeria/prevention.html
Muszynski, Dhaval Patel, Robin Rivers, Brian Young, Food and Drug 5. Angelo KM, Jackson KA, Wong KK, Hoekstra RM, Jackson BR.
Assessment of the incubation period for invasive listeriosis. Clin Infect
Administration Office of Regulatory Affairs. Dis 2016;63:1487–9. PMID:27535950 https://doi.org/10.1093/cid/
ciw569

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 711
 Morbidity and Mortality Weekly Report

6. Food and Drug Administration. FSMA final rule for preventive 8. Jackson KA, Gould LH, Hunter JC, Kucerova Z, Jackson B. Listeriosis
controls for human food. Silver Spring, MD: US Department of outbreaks associated with soft cheeses, United States, 1998–2014. Emerg
Health and Human Services, Food and Drug Administration; 2020. Infect Dis 2018;24:1116–8. PMID:29774843 https://doi.org/10.3201/
https://www.fda.gov/food/food-safety-modernization-act-fsma/ eid2406.171051
fsma-final-rule-preventive-controls-human-food
7. Ibarra-Sánchez LA, Van Tassell ML, Miller MJ. Invited review: Hispanic-
style cheeses and their association with Listeria monocytogenes. J Dairy Sci
2017;100:2421–32. PMID:28189316 https://doi.org/10.3168/
jds.2016-12116

712 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

Post-COVID Conditions Among Adult COVID-19 Survivors Aged 18–64 and

≥65 Years — United States, March 2020–November 2021
Lara Bull-Otterson, PhD1; Sarah Baca1,2; Sharon Saydah, PhD1; Tegan K. Boehmer, PhD1; Stacey Adjei, MPH1; Simone Gray, PhD1; Aaron M. Harris, MD1

On May 24, 2022, this report was posted as an MMWR Early Cerner Real-World Data,* a national, deidentified data set
Release on the MMWR website (https://www.cdc.gov/mmwr). of approximately 63.4 million unique adult records from
A growing number of persons previously infected with 110 data contributors in the 50 states. Case-patients (353,164)
SARS-CoV-2, the virus that causes COVID-19, have reported were adults aged ≥18 years who received either a diagnosis of
persistent symptoms, or the onset of long-term symptoms, COVID-19 or a positive SARS-CoV-2 test result† (case-patient
≥4 weeks after acute COVID-19; these symptoms are index encounter) in an inpatient, emergency department, or
commonly referred to as post-COVID conditions, or long outpatient settings within a subset of health care facilities
COVID (1). Electronic health record (EHR) data during that use Cerner EHRs. Control patients (1,640,776) had a
March 2020–November 2021, for persons in the United States visit in the same month as the matched case-patient (control
aged ≥18 years were used to assess the incidence of 26 condi- index encounter) and did not receive a COVID-19 diagnosis
tions often attributable to post-COVID (hereafter also referred or a positive SARS-CoV-2 test result during the observation
to as incident conditions) among patients who had received period. Controls were matched 5:1 with case-patients. All
a previous COVID-19 diagnosis (case-patients) compared patients included in the analysis were required to have at least
with the incidence among matched patients without evidence one encounter in their EHR during the year preceding and
of COVID-19 in the EHR (control patients). The analysis the year after the index encounter.
was stratified by two age groups (persons aged 18–64 and The occurrence of 26 clinical conditions previously
≥65 years). Patients were followed for 30–365 days after the attributed to post-COVID illness was assessed by review
index encounter until one or more incident conditions were of the scientific literature§ (3–5) (Supplementary Table 1,
observed or through October 31, 2021 (whichever occurred https://stacks.cdc.gov/view/cdc/117411). Patients were followed
first). Among all patients aged ≥18 years, 38% of case-patients for 30–365 days after the index encounter until the first occur-
experienced an incident condition compared with 16% of rence of an incident condition or until October 31, 2021,
controls; conditions affected multiple systems, and included whichever occurred first. Case-patients or control patients
cardiovascular, pulmonary, hematologic, renal, endocrine, with a previous history of one of the included condi-
gastrointestinal, musculoskeletal, neurologic, and psychiatric tions in the year before the index encounter were excluded
signs and symptoms. By age group, the highest risk ratios (478,072 patients). The analysis was stratified by age into two
(RRs) were for acute pulmonary embolism (RR = 2.1 and 2.2 groups: adults aged 18–64 and adults aged ≥65 years. Incidence
among persons aged 18–64 and ≥65 years, respectively) and rates per 100 person-months, and RRs with 95% CIs, were
respiratory signs and symptoms (RR = 2.1 in both age groups). calculated. The number of COVID-19 case-patients having
Among those aged 18–64 years, 35.4% of case-patients
* h t t p s : / / w w w. c e r n e r . c o m / s o l u t i o n s / r e a l - w o r l d - d a t a ? _ g a =​
experienced an incident condition compared with 14.6% of 2.134259058.2081252678.1649198012-1806687702.1649105445
controls. Among those aged ≥65 years, 45.4% of case-patients † COVID-19 cases with associated positive test result were identified by the

experienced an incident condition compared with 18.5% of following: Systematized Nomenclature of Medicine (SNOMED) codes
840533007, 840535000, 840539006, and 840546002; International
controls. These findings translate to one in five COVID-19 Classification of Diseases, Tenth Edition, Clinical Modification (ICD-10-CM)
survivors aged 18–64 years, and one in four survivors aged codes B97.29 (March, 2020) and U07.1 (April–May 2020); and Logical
≥65 years experiencing an incident condition that might Observation Identifiers Names and Codes (LOINC) codes 68993–5, 92142–9,
92141–1, 94309–2, 94307–6, 94308–4, 94500–6, 94502–2, 94533–7,
be attributable to previous COVID-19. Implementation of 94534–5, 94559–2, 94756–4, 94757–2, 94758–0, 94845–5, 95406–5,
COVID-19 prevention strategies, as well as routine assess- 95409–9, 96091–4, 95425–5, 95423–0, and 96448–6.
§ Acute myocardial infarction, cardiac dysrhythmias, cardiovascular disease, heart
ment for post-COVID conditions among persons who survive failure, myocarditis and cardiomyopathy, acute pulmonary embolism,
COVID-19, is critical to reducing the incidence and impact respiratory symptoms, asthma, renal failure, chronic kidney disease,
of post-COVID, particularly among adults aged ≥65 years (2). thromboembolic event, cerebrovascular disease, coagulation and hemorrhagic
conditions, gastrointestinal and esophageal conditions, neurologic conditions,
A retrospective matched cohort design was used to ana- smell and taste disturbances, mood disorders, other mental conditions, anxiety
lyze EHRs during March 2020–November 2021, from and fear-related conditions, sleeping disorders, substance-related disorders,
malaise and fatigue, muscle disorders, musculoskeletal pain, diabetes type 2,
and diabetes type 1.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 713
 Morbidity and Mortality Weekly Report

experienced an incident condition was also estimated by TABLE. Percentage of adult COVID-19 case-patients and control
age group.¶ Nonoverlapping CIs between age groups were patients with ≥1 post-COVID–attributable incident conditions and
estimated number of COVID-19 survivors who will experience a
considered statistically significant. Analyses were performed post-COVID condition — United States, March 2020–November 2021
using RStudio Workbench (version 3.0; RStudio). This activ- No. of patients with
No. of
ity was reviewed by CDC and was conducted consistent with ≥1 incident
COVID-19
No. of patients condition
applicable federal law and CDC policy.** (column %) (column %*) survivors
Among all patients aged ≥18 years, 38.2% of case-patients Age Absolute with a
group, Case- Control Case- Control risk post-COVID
and 16.0% of controls experienced at least one incident con- yrs patients patients patients patients difference† condition§
dition (Table). Among persons aged 18–64 years, 35.4% of 18–64 254,345 1,051,588 90,111 154,011 20.8 1/5
case-patients and 14.6% of controls experienced at least one (72.0) (64.1) (35.4) (14.6)
incident condition. Among persons aged ≥65 years, 45.4% ≥65 98,819 589,188 44,840 108,850 26.9 1/4
(28.0) (35.9) (45.4) (18.5)
of case-patients and 18.5% of controls experienced at least Total 353,164 1,640,776 134,951 262,861 22.2 1/4–5
one incident condition. The absolute risk difference between (100) (100) (38.2) (16.0)
the percentage of case-patients and controls who developed * Percentage of COVID-19 case-patients or control patients with ≥1 incident
condition divided by the total study COVID-19 cohort or control cohort row’s
an incident condition was 20.8 percentage points for those age group total.
aged 18–64 years and 26.9 percentage points for those aged † Percentage point difference between COVID-19 case-patients and control

≥65 years. This finding translates to one in five COVID-19 patients (e.g., the value 20.8 is calculated as 35.4 minus 14.6).
§ Number of COVID-19 survivors who experienced a post-COVID condition
survivors aged 18–64 years and one in four survivors aged estimated as the inverse of the absolute risk difference.
≥65 years experiencing an incident condition that might be
attributable to previous COVID-19. renal failure, thromboembolic events, cerebrovascular disease,
The most common incident conditions in both age type 2 diabetes, muscle disorders, neurologic conditions, and
groups were respiratory symptoms and musculoskeletal mental health conditions (including mood disorders, anxiety,
pain (Supplementary Table 2, https://stacks.cdc.gov/view/ other mental conditions, and substance-related disorders).
cdc/117411). Among both age groups, the highest RRs were
for incident conditions involving the pulmonary system, Discussion
including acute pulmonary embolism (RR = 2.2 [patients The findings from this analysis of a large EHR-based database
aged ≥65 years] and 2.1 [patients aged 18–64 years]) and of U.S. adults indicated that COVID-19 survivors were signifi-
respiratory symptoms (RR = 2.1, both age groups) (Figure). cantly more likely than were control patients to have incident
Among patients aged ≥65 years, the risks were higher among conditions that might be attributable to previous COVID-19.
case-patients than among controls for all 26 incident condi- One in five COVID-19 survivors aged 18–64 years and one in
tions, with RRs ranging from 1.2 (substance-related disorder) four survivors aged ≥65 years experienced at least one incident
to 2.2 (acute pulmonary embolism). Among patients aged condition that might be attributable to previous COVID-19.
18–64 years, the risks were higher among case-patients than Independent of age group, the highest RRs were for acute
among controls for 22 incident conditions, with RRs ranging pulmonary embolism and respiratory symptoms.
from 1.1 (anxiety) to 2.1 (acute pulmonary embolism); no These findings are consistent with those from several large
significant difference was observed for cerebrovascular disease studies that indicated that post-COVID incident conditions
or mental health conditions, such as mood disorders, other occur in 20%–30% of patients (6,7), and that a proportion
mental conditions, and substance-related disorders. of patients require expanded follow-up care after the initial
Differences by age group were noted. The RR for cardiac infection. COVID-19 severity and illness duration can affect
dysrhythmia was significantly higher among patients aged patients’ health care needs and economic well-being (8). The
18–64 years (RR = 1.7) compared with those aged ≥65 years occurrence of incident conditions following infection might
(1.5). Similarly, the RR for musculoskeletal pain was higher also affect a patient’s ability to contribute to the workforce and
among patients aged 18–64 years (1.6) than among those aged might have economic consequences for survivors and their
≥65 years (1.4). Among case-patients, the RRs for 10 incident dependents, particularly among adults aged 18–64 years (5).
conditions was significantly higher among those aged ≥65 years In addition, care requirements might place a strain on health
than among those aged 18–64 years; these conditions were services after acute illness in communities that experience heavy
COVID-19 case surges.
¶ Calculated as the reciprocal of the absolute risk difference of COVID-19 case-patients
and non–COVID-19 controls that experience at least one incident condition.
COVID-19 survivors aged ≥65 years in this study were at
** 45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); increased risk for neurologic conditions, as well as for four
5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq. of five mental health conditions (mood disorders, other

714 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

FIGURE. Risk ratios* for developing post-COVID conditions among adults aged 18–64 years and ≥65 years — United States, March 2020–
November 2021

Acute myocardial infarction ≥65 yrs

18–64 yrs
Cardiac dysrhythmias
Cardiovascular

Cardiovascular disease

Heart failure

Myocarditis and cardiomyopathy

Acute pulmonary embolism

Pulmonary

Respiratory symptoms

Asthma

Renal failure
Renal

Chronic kidney disease

Thromboembolic event
and vascular
Hemolytic

Cerebrovascular disease
Condition

Coagulation and hemorrhagic

Gastrointestinal and esophageal

GI
Neurologic

Neurologic conditions

Smell and taste disturbances

Mood disorders

Other mental conditions

Mental health

Anxiety and fear-related

Sleeping disorders

Substance-related disorders

Malaise and fatigue

Musculoskeletal

Muscle disorders

Musculoskeletal pain
Endocrine

Diabetes type 2

Diabetes type 1

0 0.5 1.0 1.5 2.0 2.5 3.0

Risk ratio
Abbreviation: GI = gastrointestinal.
* With CIs indicated by error bars; some error bars are not visible because of small CIs.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 715
 Morbidity and Mortality Weekly Report

Clinical Modification (ICD-10-CM) codes were used to

Summary
identify COVID-19 case-patients, and misclassification of
What is already known about this topic?
controls is possible. However, the inclusion of laboratory data
As more persons are exposed to and infected by SARS-CoV-2, to identify case-patients and exclude controls helped to limit
reports of patients who experience persistent symptoms or
organ dysfunction after acute COVID-19 and develop
the potential for such misclassification. Finally, the study only
post-COVID conditions have increased. assessed conditions thought to be attributable to COVID-19
What is added by this report?
or post-COVID illness, which might have biased RRs away
from the null. For example, clinicians might have been more
COVID-19 survivors have twice the risk for developing
pulmonary embolism or respiratory conditions; one in five likely to document possible post-COVID conditions among
COVID-19 survivors aged 18–64 years and one in four survivors case-patients. In addition, because several conditions exam-
aged ≥65 years experienced at least one incident condition ined are also risk factors for moderate to severe COVID-19,
that might be attributable to previous COVID-19. it is possible that case-patients were more likely to have had
What are the implications for public health practice? an existing condition that was not documented in their EHR
Implementation of COVID-19 prevention strategies, as well as during the year preceding their COVID-19 diagnosis, resulting
routine assessment for post-COVID conditions among persons in overestimated risk for this group.
who survive COVID-19, is critical to reducing the incidence and As the cumulative number of persons ever having been
impact of post-COVID conditions, particularly among adults
infected with SARS-CoV-2 increases, the number of survivors
aged ≥65 years.
suffering post-COVID conditions is also likely to increase.
Therefore, implementation of COVID-19 prevention strate-
mental conditions, anxiety, and substance-related disorders). gies, as well as routine assessment for post-COVID conditions
Neurocognitive symptoms have been reported to persist for among persons who survive COVID-19, is critical to reduc-
up to 1 year after acute infection and might persist longer (9). ing the incidence and impact of post-COVID conditions,
Overall, 45.4% of survivors aged ≥65 years in this study had particularly among adults aged ≥65 years (2). These findings
incident conditions. Among adults aged ≥65 years, who are can increase awareness for post-COVID conditions and improve
already at higher risk for stroke and neurocognitive impair- post-acute care and management of patients after illness. Further
ment, post-COVID conditions affecting the nervous system investigation is warranted to understand the pathophysiologic
are of particular concern because these conditions can lead to mechanisms associated with increased risk for post-COVID
early entry into supportive services or investment of additional conditions, including by age and type of condition.
resources into care (10). Corresponding author: Lara Bull-Otterson, lbull@cdc.gov.
The findings in this study are subject to at least five limitations.
1 CDC COVID-19 Emergency Response Team; 2GAP Solutions, Inc.,
First, patient data were limited to those seen at facilities serviced
Herndon, Virginia.
by Cerner EHR network during January 2020–November 2021;
therefore, the findings might not be representative of the entire All authors have completed and submitted the International
U.S. adult population or of COVID-19 case patients infected with Committee of Medical Journal Editors form for disclosure of potential
recent variants. Second, the incidence of new conditions after an conflicts of interest. No potential conflicts of interest were disclosed.
acute COVID-19 infection might be biased toward a population References
that is seeking care, either as a follow-up to a previous complaint 1. CDC. Long COVID or post-COVID conditions. Atlanta, GA: US
(including COVID-19) or for another medical complaint, which Department of Health and Human Services, CDC; 2022. Accessed
might result in a “sicker” control group leading to underestima- April 22, 2022. https://www.cdc.gov/coronavirus/2019-ncov/long-term-
tion of observed risk. Third, COVID-19 vaccination status was effects/index.html
2. Antonelli M, Penfold RS, Merino J, et al. Risk factors and disease profile
not considered in this analysis, nor were potentially confounding of post-vaccination SARS-CoV-2 infection in UK users of the COVID
factors (e.g., SARS-CoV-2 variant, sex, race, ethnicity, health care Symptom Study app: a prospective, community-based, nested, case-
entity, or geographic region), because data were not available, control study. Lancet Infect Dis 2022;22:43-55. PMID:34480857
https://doi.org/10.1016/S1473-3099(21)00460-6
were inconsistent, or included a high proportion of missing or 3. Al-Aly Z, Xie Y, Bowe B. High-dimensional characterization of post-
unknown values; for example, data were not matched by data acute sequelae of COVID-19. Nature 2021;594:259–64.
contributors, so controls were not necessarily from the same PMID:33887749 https://doi.org/10.1038/s41586-021-03553-9
4. Cohen K, Ren S, Heath K, et al. Risk of persistent and new clinical
health care entity or region of the country. Differences between sequelae among adults aged 65 years and older during the post-acute
the groups might influence the risks associated with survival from phase of SARS-CoV-2 infection: retrospective cohort study. BMJ
COVID-19 and incident conditions, which require further study. 2022;376:e068414. PMID:35140117 https://doi.org/10.1136/
Fourth, International Classification of Disease, Tenth Revision, bmj-2021-068414

716 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

5. Rajan S, Khunti K, Alwan N, et al. In the wake of the pandemic: 8. CDC. Science brief: indicators for monitoring COVID-19 community
preparing for long COVID. Copenhagen, Denmark: European levels and making public health recommendations. Atlanta, GA: US
Observatory on Health Systems and Policies; 2021. PMID:33877759 Department of Health and Human Services; 2022. https://www.cdc.
6. Ayoubkhani D, Khunti K, Nafilyan V, et al. Post-COVID syndrome in gov/coronavirus/2019-ncov/science/science-briefs/indicators-
individuals admitted to hospital with COVID-19: retrospective cohort study. monitoring-community-levels.html
BMJ 2021;372:n693. PMID:33789877 https://doi.org/10.1136/bmj.n693 9. Mueller AL, McNamara MS, Sinclair DA. Why does COVID-19
7. Donnelly JP, Wang XQ, Iwashyna TJ, Prescott HC. Readmission and disproportionately affect older people? Aging (Albany NY) 2020;12:9959–81.
death after initial hospital discharge among patients with COVID-19 PMID:32470948 https://doi.org/10.18632/aging.103344
in a large multihospital system. JAMA 2021;325:304–6. PMID:33315057 10. Mohamed MS, Johansson A, Jonsson J, Schiöth HB. Dissecting the
https://doi.org/10.1001/jama.2020.21465 molecular mechanisms surrounding post-COVID-19 syndrome and
neurological features. Int J Mol Sci 2022;23:4275. PMID:35457093
https://doi.org/10.3390/ijms23084275

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 717
 Morbidity and Mortality Weekly Report

Notes from the Field

Self-Reported Health Symptoms Following CDC and was conducted consistent with applicable federal
Petroleum Contamination of a Drinking Water law and CDC policy.§
System — Oahu, Hawaii, November 2021– A total of 2,289 eligible participants submitted surveys,
February 2022 with at least one household member participating from 1,389
Alyssa N. Troeschel, PhD1; Ben Gerhardstein, MPH 2;
(14%) of 9,694 estimated affected households. The median
Alex Poniatowski, MS3; Diana Felton, MD4; Amanda Smith, participant age was 33 years (range = 1–84 years). Participants
PhD1; Krishna Surasi, MD1; Alyson M. Cavanaugh, DPT, PhD1; were predominantly female (59%), non-Hispanic (81%),
Shanna Miko, DNP1; Michele Bolduc, PhD1; Vidisha Parasram, DrPH1; military-affiliated (88%), and identified their race as White
Charles Edge, MSN, MS5; Renée Funk, DVM3; Maureen Orr, MS5
(74%). Among all participants who were residents in the
In late November 2021, the Hawaii Department of Health affected area, 1,115 (52%) reported at least one indication that
(HDOH) received reports from Oahu residents of a fuel-like their water was contaminated (i.e., petroleum smell or taste, or
odor coming from their drinking water (1), which was later visible oily sheen). Participants indicated that they ingested the
determined to be related to a November 20, 2021, petroleum potentially contaminated water through oral hygiene (1,821;
(jet fuel) leak at the Red Hill Bulk Fuel Storage Facility. The 80%), drinking (1,650; 72%), and cooking (1,629; 71%).
petroleum leak contaminated the Joint Base Pearl Harbor- Most participants (2,123; 93%) switched to an alternative
Hickam water system,* which supplies an estimated 9,694 water source after learning of the incident.
civilian and military households (2), in addition to schools Most participants reported experiencing one or more new or
and workplaces. HDOH issued a drinking water advisory worsened symptoms after the incident (1,980; 87%), many of
on November 30, 2021 (1), which was not lifted for all whom reported symptoms lasting ≥30 days (1,493; 75%). The
affected zones until March 18, 2022.† Persons in thousands largest percentages of reported symptoms were those related
of households were offered temporary housing, and alterna- to the nervous system (62%), followed by the gastrointestinal
tive drinking water was provided to users of affected water. system (58%), skin (58%), ear, nose, and throat (47%), men-
HDOH requested epidemiologic assistance (Epi-Aid) from tal health (46%), eyes (42%), and respiratory system (31%)
CDC/Agency for Toxic Substances and Disease Registry (Table). Medical care was sought by 853 (37%) of partici-
(ATSDR) to assess the incident’s impact on civilian health in pants after the incident, including 17 who were hospitalized
the affected area; this was later expanded to include military- overnight. Among symptomatic participants, 1,591 of 1,980
affiliated persons. symptomatic participants (80%) reported improvement in
The team adapted an interviewer-administered survey from symptoms after switching to an alternative water source. In
the ATSDR Assessment of Chemical Exposures (ACE) toolkit an open-text field, 53 (2%) participants expressed concerns
to collect information about potential exposure to contami- about possible long-term health effects.
nated water, health symptoms experienced, and medical care This novel incident of jet fuel–contaminated drinking
sought. The survey was modified to be self-administered water disrupted the lives of thousands of persons. An online
online, similar to a previous ACE investigation (3). Persons survey paired with robust in-person and electronic promotion
present in the affected area after the incident were eligible to facilitated rapid information collection from many affected
complete the survey during January 7–February 10, 2022. persons across a wide geographic area, including many who
Parents and guardians completed the survey on behalf of were displaced from their homes. This survey method did not
persons aged <18 years. The survey was promoted through allow for prevalence estimates, nor did it capture the full scope
electronic and in-person outreach. Household-level response of health impacts. Reported symptoms, such as those related to
rates were calculated using ArcGIS Pro and U.S. Navy data the respiratory system, gastrointestinal tract, nervous system,
(3). Descriptive statistics were calculated using R software and mental health, were consistent with previous studies of
(version 4.1.1; R Foundation). This activity was reviewed by exposure to petroleum hydrocarbons¶ (4,5), and accounts of
some relief from symptoms after switching to an alternative
* https://www.cpf.navy.mil/News/Article/2870459/opening-statements-at-
hawaii-state-legislature-briefing/msclkid/opening-statements-at-hawaii- § 45 C.F.R. part 46, 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect.
state-legislature-briefing/ 552a; 44 U.S.C. Sect. 3501 et seq.
† https://health.hawaii.gov/news/newsroom/doh-declares-four-navy-drinking- ¶ https://wwwn.cdc.gov/TSP/ToxFAQs/ToxFAQsDetails.aspx?faqid=772&toxid=150
water-distribution-system-zones-safe/

718 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

TABLE. Occurrence of new or worsened symptoms, and symptoms Acknowledgments

persisting for ≥30 days after the contamination of a water system
by a petroleum leak on November 20, 2021, self-reported by Kaitlin Arita-Chang, Jennifer Espiritu, Maddy Jarvis, Sarah
participants of the Hawaii Assessment of Chemical Exposures survey Kemble, Nicole Mintz, Jared Wagner, Hawaii Department of
(N = 2,289) — Oahu, Hawaii, November 2021–February 2022 Health; Jose Aponte, Bryn Davis, Nina D. Dutton, Joanna Gaines,
No. (%) of survey participants William Jones, Susan McBreairty, Jamie Rayman, Libby Vianu, CDC
Experiencing new Experiencing
Epi-Aid Response Team, CDC/ATSDR; Bryan Schumacher, U.S.
or worsened symptoms for Department of Defense.
Self-reported symptom symptoms ≥30 days* Corresponding author: Alyssa N. Troeschel, rjz5@cdc.gov.
Eyes 967 (42) 514/967 (53)
1Epidemic Intelligence Service, CDC; 2Office of Community Health and
Increased tearing 498 (22) 303/498 (61)
Irritation/Pain/Burning of eyes 879 (38) 453/879 (52) Hazard Assessment, Agency for Toxic Substances and Disease Registry, San
Ear, nose, and throat 1,078 (47) 553/1,078 (51) Francisco, California; 3Office of Emergency Management, CDC/ATSDR;
4Hawaii Department of Health; 5Office of Innovation and Analytics, Agency
Runny nose 715 (31) 388/715 (54)
Nose bleeds 191 (8) 86/191 (45) for Toxic Substances and Disease Registry, Atlanta, Georgia.
Burning nose or throat 739 (32) 87/739 (12)
Ringing in ears 405 (18) 263/405 (65) All authors have completed and submitted the International
Nervous system 1,428 (62) 959/1,428 (67) Committee of Medical Journal Editors form for disclosure of
Headache 1,318 (58) 726/1,318 (55) potential conflicts of interest. Diana Felton reports grants from
Dizziness/Lightheadedness 875 (38) 463/875 (53) the Environmental Protection Agency for the Water Infrastructure
Seizures/Convulsions 23 (1) 18/23 (78) Improvements for the Nation Act, multipurpose grants, grants from
Feeling fatigued 1,016 (44) 696/1,016 (69)
Loss of consciousness/Fainting 52 (2) 29/52 (56) the State Response Program, and travel support from the University of
Confusion 424 (19) 271/424 (64) California at San Francisco (UCSF) Department of Occupational and
Difficulty concentrating 738 (32) 530/738 (72) Environmental Medicine (OEM) for travel to the 2022 UCSF OEM
Difficulty remembering things 644 (28) 483/644 (75)
annual meeting and from the Association of State and Territorial
Respiratory/Cardiovascular 719 (31) 463/719 (64)
Chest tightness or pain/Angina 362 (16) 206/362 (57)
Health Officials for attendance at the State Environmental Health
Wheezing in chest 189 (8) 126/189 (67) Directors Annual Meeting (June 2022). No other potential conflicts
Difficulty breathing/Feeling 416 (18) 271/416 (65) of interest were disclosed.
out-of-breath
Coughing 522 (23) 303/522 (58) References
Burning lungs 185 (8) 107/185 (58)
Gastrointestinal 1,332 (58) 566/1,332 (43) 1. Hawaii Department of Health. Hawai’i Department of Health advises Navy
Nausea 929 (41) 391/929 (42) water system consumers not to drink, consume tap water (Press release).
Vomiting 370 (16) 100/370 (27) Honolulu, HI: Hawaii Department of Health; November 30, 2021. https://
Diarrhea 1,121 (49) 397/1,121 (35) health.hawaii.gov/news/newsroom/hawaii-department-of-health-advises-
Dermatologic 1,322 (58) 880/1,322 (67) navy-water-system-consumers-not-to-drink-consume-tap-water/
Irritation/Pain/Burning of skin 859 (38) 476/859 (55) 2. Joint Base Pearl Harbor-Hickam Interagency Drinking Water
Skin rash 925 (40) 506/925 (55) System Team. Water system flushing zone map. Oahu, HI. Accessed
Skin blisters 169 (7) 101/169 (60) March 14, 2022. https://www.cpf.navy.mil/Portals/52/Downloads/
Dry or itchy skin 1,144 (50) 771/1,144 (67) J B PH H -Wa t e r - Up d a t e s / 2 0 2 2 0 1 3 0 _ Fl u s h i n g _ Ma p _ St a t u s .
Mental health 1,049 (46) 865/1,049 (83)
pdf?ver=z8TDJP-ippC8g5Qg9j3RKg%3d%3d
Anxiety 839 (37) 667/839 (80) 3. Surasi K, Nakayama JY, Johnson M, et al. Notes from the field: deployment
Agitation/Irritability 696 (30) 549/696 (79) of an electronic self-administered survey to assess human health effects
Difficulty sleeping 744 (33) 590/744 (79) of an industrial chemical facility fire—Winnebago County, Illinois,
Feeling depressed 463 (20) 364/463 (79) June–July 2021. MMWR Morb Mortal Wkly Rep 2021;70:1715–6.
Paranoia 226 (10) 179/226 (79) PMID:34882658 https://doi.org/10.15585/mmwr.mm7049a4
Tension/Nervousness 656 (29) 524/656 (80) 4. Bendtsen KM, Bengtsen E, Saber AT, Vogel U. A review of health effects
Other† 360 (16) 236/360 (66) associated with exposure to jet engine emissions in and around airports.
Environ Health 2021;20:10–20. PMID:33549096 https://doi.
* Among those who reported experiencing symptom. org/10.1186/s12940-020-00690-y
† Participants could report up to four additional symptoms not listed in the
5. D’Andrea MA, Reddy GK. The development of long-term adverse health
symptoms section of the survey.
effects in oil spill cleanup workers of the Deepwater Horizon offshore
drilling rig disaster. Front Public Health 2018;6:117. PMID:29755965
water source support exposure-related health effects. These https://doi.org/10.3389/fpubh.2018.00117
results highlight the need for preventing exposure to petro-
leum products and might aid public health professionals and
clinicians in detecting and responding to future similar inci-
dents. Exposure levels, duration, and long-term health effects,
however, are uncertain. Additional follow-up of the affected
population might improve understanding of the overall health
impact of this and other petroleum exposure incidents.

US Department of Health and Human Services/Centers for Disease Control and Prevention MMWR / May 27, 2022 / Vol. 71 / No. 21 719
 Morbidity and Mortality Weekly Report

QuickStats

FROM THE NATIONAL CENTER FOR HEALTH STATISTICS

Percentage Distribution of Deaths Involving Injuries from Recreational and

Nonrecreational Use of Watercraft,* by Month — United States, 2018–2020
100

20
Percentage

15

10

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month

* Deaths were identified using International Classification of Diseases, Tenth Revision underlying cause of death
codes V90–V94 (water transport) for a total of 1,508 deaths during 2018–2020. Water transport includes
recreational and nonrecreational use of motorized (e.g., merchant ship, ferry, passenger ship, fishing boat,
and jet ski) and nonmotorized (e.g., canoe, kayak, inflatable craft, surfboard, and windsurfer) watercraft. Deaths
resulted from drowning, submersion, and other types of injuries. All water transport deaths were unintentional.

During 2018–2020, 1,508 deaths occurred involving injuries from recreational and nonrecreational use of watercraft. The
percentage of deaths each month ranged from 3.0% in December to 16.6% in July. Most deaths (68.6%) occurred during
May–September.
Source: National Vital Statistics System, Mortality Data. https://www.cdc.gov/nchs/nvss/deaths.htm
Reported by: Matthew F. Garnett, MPH, Mgarnett@cdc.gov, 301-458-4383; Merianne R. Spencer, MPH.

720 MMWR / May 27, 2022 / Vol. 71 / No. 21 US Department of Health and Human Services/Centers for Disease Control and Prevention
 Morbidity and Mortality Weekly Report

The Morbidity and Mortality Weekly Report (MMWR) Series is prepared by the Centers for Disease Control and Prevention (CDC) and is available free
of charge in electronic format. To receive an electronic copy each week, visit MMWR at https://www.cdc.gov/mmwr/index.html.
Readers who have difficulty accessing this PDF file may access the HTML file at https://www.cdc.gov/mmwr/index2022.html. Address all inquiries about
the MMWR Series to Editor-in-Chief, MMWR Series, Mailstop V25-5, CDC, 1600 Clifton Rd., N.E., Atlanta, GA 30329-4027 or to mmwrq@cdc.gov.
All material in the MMWR Series is in the public domain and may be used and reprinted without permission; citation as to source, however, is appreciated.
MMWR and Morbidity and Mortality Weekly Report are service marks of the U.S. Department of Health and Human Services.
Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services.
References to non-CDC sites on the Internet are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations
or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of these sites. URL addresses
listed in MMWR were current as of the date of publication.

ISSN: 0149-2195 (Print)