Case Study 3 – Chapter 8

Abstract

We examined associations of health insurance status with self-perceived poor/fair health and frequent mental distress (FMD) among working-aged US adults from 42 states and the District of Columbia using data from the 2014 Behavioral Risk Factor Surveillance System. After multiple-variable adjustment, compared with adequately insured adults, underinsured and never insured adults were 39% and 59% more likely to report poor/fair health, respectively, and 38% more likely to report FMD. Compared with working-aged adults with employer-based insurance, adults with Medicaid/Medicare or other public insurance coverage were 28% and 13% more likely to report poor/fair health, respectively, and 15% more likely to report FMD. Increasing insurance coverage and reducing cost barriers to care may improve general and mental health.

Questions:

After reading the full paper, answer the following questions. Provide detailed information. The full paper is in the module section.

1. What is the role of epidemiology in finance?

2. Based on table 1 in the full paper, of the 201,423 participants who reported poor/fair health status, what percentages of participants are unemployed, and leave at less than 100% poverty level?

3. By comparing smokers or obese participants (Mass Index (BMI) more or equal to 30kg/m2), how many times more likely is poor/fair health among those who either smoke or obese compared to those who do not smoke or are not obese? (compare smokers to non-smokers and obese to non-obese)

4. Assuming that the average cost of being hospitalized for mental distress is $3,985, and that people with chronic conditions more or equal to 3 have 50 percent annual risk of being hospitalized, what is the total cost of hospitalization paid by people with chronic conditions more or equal to 3 in 2014?

5. Assuming that each person with mental health distress visits a physician once a year with an average cost of $120 per visit. Assuming that the risk of hospitalization drops to “0” if each person visits a physician once a year. What would be the cost savings among people with chronic conditions more or equal to 3?

6. Do you think capitation system based on the prevalence of risk factors is actuarially fair? (provide detailed explanation)

Reference

Zhao, G., Okoro, C. A., Hsia, J., & Town, M. (2018). Self-Perceived Poor/Fair Health, Frequent Mental

Distress, and Health Insurance Status Among Working-Aged US Adults. Preventing chronic

disease, 15, E95. https://doi.org/10.5888/pcd15.170523

PREVENTING CHRONIC DISEASE
P U B L I C H E A L T H R E S E A R C H , P R A C T I C E , A N D P O L I C Y

Volume 15, E95 JULY 2018

RESEARCH BRIEF

Self-Perceived Poor/Fair Health, Frequent
Mental Distress, and Health Insurance

Status Among Working-Aged US Adults

Guixiang Zhao, MD, PhD1; Catherine A. Okoro1; Jason Hsia1; Machell Town1

Accessible Version: www.cdc.gov/pcd/issues/2018/17_0523.htm

Suggested citation for this article: Zhao G, Okoro CA, Hsia J,
Town M. Self-Perceived Poor/Fair Health, Frequent Mental
Distress, and Health Insurance Status Among Working-Aged US
Adults. Prev Chronic Dis 2018;15:170523. DOI: https://doi.org/
10.5888/pcd15.170523.

PEER REVIEWED

Abstract
We examined associations of health insurance status with self-per-
ceived poor/fair health and frequent mental distress (FMD) among
working-aged US adults from 42 states and the District of
Columbia using data from the 2014 Behavioral Risk Factor Sur-
veillance System. After multiple-variable adjustment, compared
with adequately insured adults, underinsured and never insured
adults were 39% and 59% more likely to report poor/fair health,
respectively, and 38% more likely to report FMD. Compared with
working-aged adults with employer-based insurance, adults with
Medicaid/Medicare or other public insurance coverage were 28%
and 13% more likely to report poor/fair health, respectively, and
15% more likely to report FMD. Increasing insurance coverage
and reducing cost barriers to care may improve general and men-
tal health.

Objective
Self-rated health and health-related quality of life (HRQOL) are
commonly used measures of overall well-being, physical health
conditions, and functioning (1,2) and are key indicators for assess-
ing national health in Healthy People objectives. Socioeconomic
status, lifestyle factors, and chronic conditions affect self-rated
health and HRQOL (3,4).

Health insurance improves access to and affordability of care,
which can be critical to managing chronic conditions; lacking ad-

equate coverage and not being able to afford care can cause men-
tal distress (5,6). We examined associations of health insurance
coverage and type of coverage with self-perceived poor/fair health
and frequent mental distress (FMD) among working-aged US
adults.

Methods
The Behavioral Risk Factor Surveillance System (BRFSS) is a
state-based, landline-telephone and cellular-telephone survey of
noninstitutionalized civilian US adults (7). In 2014, 42 states and
the District of Columbia collected health care access data through
both core and module questions, which were used for this analysis.
The BRFSS protocol was approved by the Centers for Disease
Control and Prevention institutional review board. The median re-
sponse rate was 47.0% in 2014.

Survey participants’ self-perceived health was dichotomized into
poor/fair and good/very good/excellent. FMD was defined as hav-
ing 14 or more days of poor mental health (including stress, de-
pression, and problems with emotions) in the past 30 days.

The survey questions and categorization of participants’ insurance
status and type of insurance are described elsewhere (8). Health
insurance coverage over the past 12 months was categorized as ad-
equately insured, underinsured, and never insured. Type of cover-
age was categorized as employer-based, self-purchased, Medicaid
or Medicare; other public or some other source, and not currently
insured.

Study covariates were age, sex, race/ethnicity, education level,
marital status, employment status, federal poverty level (FPL),
current smoking, leisure-time physical activity, body mass index,
and the number of chronic conditions/diseases including diabetes,
coronary heart disease, stroke, current asthma, arthritis, cancer,
chronic obstructive pulmonary disease, history of depression, kid-
ney disease, and disability.

The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health

and Human Services,

the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.

www.cdc.gov/pcd/issues/2018/17_0523.htm • Centers for Disease Control and Prevention 1

Participants who responded “don’t know/not sure,” refused to an-
swer, or had missing responses to any study covariates mentioned
above were excluded, leaving 201,781 working-aged adults (aged
18–64 years) in the analytic sample. After further excluding parti-
cipants with missing data on the 2 outcome variables, 201,423 par-
ticipants remained for the analysis for self-rated poor/fair health
and 199,709 participants remained for the analysis for FMD.

We estimated the weighted prevalence for self-perceived poor/fair
health and FMD by health insurance coverage and type of cover-
age. Log-linear regression analyses were conducted to estimate ad-
justed prevalence ratios with 95% confidence intervals (CIs) while
controlling for study covariates. SAS-callable SUDAAN (Re-
search Triangle Institute) was used to account for the complex sur-
vey design. Significance was set at P < .05.

Results
Of 201,781 working-aged adults, the mean age was 41 years;
49.0% were women; 69.9% were non-Hispanic white; 13.5% were
non-Hispanic black; and 9.8% were Hispanic. Approximately
59.9% had a degree higher than high school; 14.8% lived below
the poverty level (ie, household income <100% of FPL); 20.6% were current smokers; 20.6% were physically inactive; 29.8% were obese; and 48.6% had at least 1 chronic condition or disabil- ity.

For insurance status, 57.3% were adequately insured, 32.3% un-
derinsured, and 10.4% never insured. Approximately 56.2% had
employer-based coverage; 9.5% had self-purchased coverage;
13.8% had Medicaid or Medicare; 5.3% had other public insur-
ance; and 15.2% were not currently insured.

Overall, 14.8% (95% CI, 14.6%–15.1%) of adults reported poor/
fair health, and 12.5% (95% CI, 12.2%–12.7%) reported

FMD

(Table 1). Prevalence differed by sociodemographic characterist-
ics, lifestyle risk factors, and number of chronic conditions and
disability.

The age-adjusted prevalence of poor/fair health and FMD was sig-
nificantly higher among underinsured adults (21.8% and 18.8%,
respectively) and never insured adults (23.4% and 16.8%, respect-
ively) compared with adequately insured adults (8.9% and 8.5%,
respectively) (Table 2). The age-adjusted prevalence of poor/fair
health and FMD were lowest among adults with employer-based
insurance (7.8% and 8.5%, respectively) or self-purchased cover-
age (9.3% and 8.8%, respectively) and highest among adults with
Medicaid/Medicare (36.5% and 27.2%, respectively).

After multiple-variable adjustment for study covariates, compared
with adequately insured adults, underinsured and never insured

adults were 39% (P < .001) and 59% (P < .001), respectively, more likely to report poor/fair health, and 38% (P < .001 for both) more likely to report FMD (Table 2). Compared with adults with employer-based insurance, those with Medicaid/Medicare or other public coverage were more likely to report poor/fair health (28% and 13%, respectively, P < .001) and FMD (15% for both, P < .001).

Discussion
Self-assessed health status reflects concurrent decrements in health
associated with physical functional status and certain chronic ill-
nesses (1), and lower self-rated health predicts increased mortality
(2,9). Poor mental health is associated with risky health behaviors
and social burden (6,10). Our results from a large population sur-
vey demonstrated that self-perceived health and FMD were signi-
ficantly associated with health insurance status, independent of so-
cioeconomic status; behavioral risk factors; and multiple chronic
conditions and disability. Working-aged adults who were under-
insured and never insured or who had Medicaid/Medicare or other
public insurance were more likely to rate their general health as
poor/fair and to report FMD than their counterparts who were ad-
equately insured or had private insurance.

Research indicates that having insurance coverage predicts better
self-rated health (3,11,12) and that prevalence of FMD is signific-
antly higher among adults with no health insurance coverage (5,6)
or with financial barriers to needed medical care (5). These results
are consistent with our findings that the prevalence of poor/fair
health and FMD were significantly higher among underinsured
and uninsured adults. Moreover, our results further demonstrated
that poor/fair health and FMD were significantly higher among
adults with Medicaid/Medicare or other public insurance than
among those with private coverage, either employer-based or self-
purchased. Adults younger than 65 years who have Medicare or
Medicaid are likely to have permanent disabilities or certain ter-
minal diseases, or live in poverty; all of these may contribute to a
higher prevalence of poor/fair health or FMD.

The strength of our study is that results are based on a large popu-
lation surveillance system. BRFSS data, however, are self-repor-
ted and subject to recall bias. Also, because BRFSS excluded in-
stitutionalized adults, prevalence of poor/fair health and FMD may
be underestimated. In addition, data were from 42 states and the
District of Columbia, which limits generalizability of the study
results to the entire US working-aged population.

Although uninsurance rates are declining in the United States (13),
the trend in the rates of being underinsured or having private cov-

PREVENTING CHRONIC DISEASE VOLUME 15, E95
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JULY 2018

The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,

the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.

2 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2018/17_0523.htm

erage are largely unknown. Continuing efforts to increase health
insurance coverage and reduce cost barriers to needed medical
care may help the US population achieve optimal overall health
and reduce mental distress.

Acknowledgments
The authors thank the BRFSS coordinators from state health de-
partments for their participation in data collection. There was no
financial support associated with this study. There were no copy-
righted materials or copyrighted surveys/instruments/tools used in
this study. The findings and conclusions in this report are those of
the authors and do not necessarily represent the official position of
the Centers for Disease Control and Prevention.

Author Information
Corresponding Author: Guixiang Zhao, MD, PhD, Division of
Population Health, Centers for Disease Control and Prevention,
4770 Buford Hwy, MS F–78, Atlanta, GA 30341. Telephone:
770–488–4450. Email: fwj4@cdc.gov.

Author Affiliations: 1Division of Population Health, National
Center for Chronic Disease Prevention and Health Promotion,
Centers for Disease Control and Prevention, Atlanta, Georgia.

References
Gold M, Franks P, Erickson P. Assessing the health of the
nation. The predictive validity of a preference-based measure
and self-rated health. Med Care 1996;34(2):163–77.

1.

Franks P, Gold MR, Fiscella K. Sociodemographics, self-rated
health, and mortality in the US. Soc Sci Med 2003;
56(12):2505–14.

2.

Kim J. Socioeconomic inequalities in self-rated health among
middle-aged and older adults. Soc Work Health Care 2011;
50(2):124–42.

3.

Tsai J, Ford ES, Li C, Zhao G, Pearson WS, Balluz LS.
Multiple healthy behaviors and optimal self-rated health:
findings from the 2007 Behavioral Risk Factor Surveillance
System Survey. Prev Med 2010;51(3-4):268–74.

4.

Bruning J, Arif AA, Rohrer JE. Medical cost and frequent
mental distress among the non-elderly US adult population. J
Public Health (Oxf) 2014;36(1):134–9.

5.

Pearson WS, Dhingra SS, Strine TW, Liang YW, Berry JT,
Mokdad AH. Relationships between serious psychological
distress and the use of health services in the United States:
findings from the Behavioral Risk Factor Surveillance System.
Int J Public Health 2009;54(S1,Suppl 1):23–9.

6.

Centers for Disease Control and Prevention. Behavioral Risk
Factor Surveillance System. https://www.cdc.gov/brfss/2017.

7.

Zhao G, Okoro CA, Li J, Town M. Health insurance status and
clinical cancer screenings among US Adults. Am J Prev Med
2017;54(1):11–19.

8.

Regidor E, Guallar-Castillón P, Gutiérrez-Fisac JL, Banegas
JR, Rodríguez-Artalejo F. Socioeconomic variation in the
magnitude of the association between self-rated health and
mortality. Ann Epidemiol 2010;20(5):395–400.

9.

Strine TW, Balluz L, Chapman DP, Moriarty DG, Owens M,
Mokdad AH. Risk behaviors and healthcare coverage among
adults by frequent mental distress status, 2001. Am J Prev Med
2004;26(3):213–6.

10.

Alexopoulos EC, Geitona M. Self-rated health: inequalities and
potential determinants. Int J Environ Res Public Health 2009;
6(9):2456–69.

11.

Franks P, Clancy CM, Gold MR, Nutting PA. Health insurance
and subjective health status: data from the 1987 National
Medical Expenditure survey. Am J Public Health 1993;
83(9):1295–9.

12.

Zhao G, Okoro CA, Dhingra SS, Xu F, Zack M. Trends of lack
of health insurance among US adults aged 18-64 years:
findings from the Behavioral Risk Factor Surveillance System,
1993-2014. Public Health 2017;146:108–17.

13.

PREVENTING CHRONIC DISEASE VOLUME 15, E95
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JULY 2018
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.

www.cdc.gov/pcd/issues/2018/17_0523.htm • Centers for Disease Control and Prevention 3

Tables

Table 1. Crude Prevalence of Self-Perceived Poor/Fair Health and FMD Among Working-Aged Adults in 42 States and the District of Columbia, by Sociodemograph-
ic Characteristics, Health-Related Behaviors, and Chronic Conditions and Disability, Behavioral Risk Factor Surveillance System, 2014

Characteristic

Poor/Fair Health FMD

No. % (95% CI) No. % (95% CI)

Overall 201,423 14.8 (14.6–15.1) 199,709 12.5 (12.2–12.7)

Age, y

18–25 17,710 8.1 (7.5–8.8) 17,580 13.1 (12.3–13.9)

26–44 61,137 12.3 (11.9–12.8) 60,708 12.6 (12.1–13.0)

45–64 122,576 19.9 (19.5–20.3) 121,421 12.1 (11.8–12.4)

Sex

Male 89,898 14.2 (13.8–14.6) 89,154 10.2 (9.8–10.6)

Female 111,525 15.6 (15.2–15.9) 110,555 14.8 (14.4–15.2)

Race/ethnicity

Non-Hispanic white 158,790 13.0 (12.8–13.3) 157,563 12.3 (12.0–12.6)

Non-Hispanic black 18,678 19.3 (18.4–20.2) 18,494 13.8 (13.0–14.7)

Hispanic 12,595 22.6 (21.4–23.9) 12,450 12.4 (11.5–13.4)

Other 11,360 13.4 (12.3–14.5) 11,202 11.2 (10.3–12.2)

Education level

High school graduate/GED 53,224 18.1 (17.6–18.6) 52,602 13.9 (13.4–14.4)

>High school graduate 135,989 9.6 (9.3–9.8) 135,164 10.1 (9.8–10.4)

Marital status

Married 115,424 11.9 (11.6–12.2) 114,687 8.9 (8.6–9.2)

Previously married 39,528 26.3 (25.6–27.1) 39,010 20.1 (19.4–20.8)

Never married/live with a partner 46,471 14.1 (13.5–14.6) 46,012 14.4 (13.9–15.0)

Employment status

Employed 138,721 8.9 (8.6–9.2) 137,869 8.6 (8.4–8.9)

Unemployed 11,858 22.7 (21.5–24.0) 11,713 21.9 (20.6–23.2)

Not in labor force 50,844 28.7 (28.0–29.4) 50,127 20.2 (19.6–20.8)

Federal poverty level, %

<100 23,075 31.0 (30.0–32.0) 22,759 23.6 (22.7–24.6)

100–400 73,013 16.6 (16.1–17.1) 72,367 13.6 (13.1–14.0)

>400 82,601 6.0 (5.7–6.3) 82,259 6.5 (6.2–6.9)

Unknown 22,734 15.8 (15.1–16.6) 22,324 13.1 (12.3–13.8)

Abbreviations: CI, confidence interval; FMD, frequent mental distress; GED, general education diploma.
a Including diabetes, coronary heart disease, stroke, current asthma, arthritis, cancer, chronic obstructive pulmonary disease, history of depression, kidney disease,
and disability. Disability was defined as respondents who were limited in any way in any activities because of physical, mental, or emotional problems, or who had
any health problem that required them to use special equipment (eg, cane, wheelchair, special bed, special telephone).

(continued on next page)

PREVENTING CHRONIC DISEASE VOLUME 15, E95
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JULY 2018
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.

4 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2018/17_0523.htm

(continued)

Table 1. Crude Prevalence of Self-Perceived Poor/Fair Health and FMD Among Working-Aged Adults in 42 States and the District of Columbia, by Sociodemograph-
ic Characteristics, Health-Related Behaviors, and Chronic Conditions and Disability, Behavioral Risk Factor Surveillance System, 2014
Characteristic
Poor/Fair Health FMD
No. % (95% CI) No. % (95% CI)

Current smoking

Yes 37,329 24.9 (24.1–25.6) 36,834 23.3 (22.6–24.1)

No 164,094 12.2 (12.0–12.5) 162,875 9.6 (9.4–9.9)

Leisure-time exercise

Yes 161,356 10.8 (10.6–11.1) 160,202 10.4 (10.1–10.6)

No 40,067 30.4 (29.6–31.2) 39,507 20.6 (19.9–21.3)

Body mass index, kg/m2

<25.0 68,283 10.1 (9.7–10.5) 67,700 11.3 (10.9–11.8)

25.0–29.9 70,543 12.2 (11.8–12.6) 69,973 10.8 (10.3–11.2)

≥30.0 62,597 23.7 (23.1–24.3) 62,036 15.8 (15.3–16.3)

Number of chronic conditions and disabilitya

0 92,513 4.4 (4.2–4.7) 92,021 4.8 (4.6–5.1)

1 49,088 10.6 (10.1–11.2) 48,688 11.4 (10.9–12.0)

2 27,245 24.5 (23.6–25.5) 26,951 20.4 (19.5–21.3)

≥3 32,577 54.5 (53.6–55.4) 32,049 37.2 (36.3–38.2)

Abbreviations: CI, confidence interval; FMD, frequent mental distress; GED, general education diploma.
a Including diabetes, coronary heart disease, stroke, current asthma, arthritis, cancer, chronic obstructive pulmonary disease, history of depression, kidney disease,
and disability. Disability was defined as respondents who were limited in any way in any activities because of physical, mental, or emotional problems, or who had
any health problem that required them to use special equipment (eg, cane, wheelchair, special bed, special telephone).
PREVENTING CHRONIC DISEASE VOLUME 15, E95
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JULY 2018
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.

www.cdc.gov/pcd/issues/2018/17_0523.htm • Centers for Disease Control and Prevention 5

Table 2. Crude and Age-Adjusted Prevalence and APRs for Self-Perceived Poor/Fair Health and FMD Among Adults Aged 18 to 64 Years in 42 States and the Dis-
trict of Columbia, by Insurance Status and Type of Insurance, Behavioral Risk Factor Surveillance System, 2014

Insurance Status No.

% (95% Confidence Interval

APRb (%95 CI)Crude Age Adjusteda

Poor/Fair Health

Insurance coverage during the past 12 months

Adequately insured 118,446 9.7 (9.4–10.0) 8.9 (8.6–9.2) 1.00

Underinsured 59,921 22.7 (22.2–23.3) 21.8 (21.3–22.4) 1.39 (1.33–1.44)

Never insured 14,794 22.9 (21.8–24.1) 23.4 (22.3–24.6) 1.59 (1.50–1.68)

Insurance type

Employer-based insurance 116,080 8.3 (8.0–8.6) 7.8 (7.5–8.1) 1.00

Self-purchased plan 19,732 9.8 (9.1–10.5) 9.3 (8.6–10.1) 1.00 (0.93–1.08)

Medicaid or Medicare 26,045 39.2 (38.1–40.2) 36.5 (35.5–37.6) 1.28 (1.22–1.36)

Other public insurance 11,107 18.8 (17.6–20.1) 17.3 (16.1–18.7) 1.13 (1.06–1.21)

Not currently insured 21,368 21.4 (20.5–22.4) 22.5 (21.6–23.4) 1.48 (1.39–1.57)

FMD
Insurance coverage during the past 12 months

Adequately insured 117,630 8.3 (8.1–8.7) 8.5 (8.1–8.8) 1.00

Underinsured 59,296 18.8 (18.3–19.4) 18.8 (18.2–19.3) 1.38 (1.31–1.44)

Never insured 14,625 16.7 (15.7–17.8) 16.8 (15.8–17.8) 1.38 (1.29–1.48)

Insurance type

Employer-based insurance 115,424 8.2 (8.0–8.5) 8.5 (8.2–8.9) 1.00

Self-purchased plan 19,606 9.3 (8.6–10.1) 8.8 (8.1–9.6) 0.95 (0.87–1.04)

Medicaid or Medicare 25,568 27.4 (26.4–28.4) 27.2 (26.1–28.2) 1.15 (1.08–1.23)

Other public insurancec 10,976 15.6 (14.5–16.8) 15.8 (14.6–17.2) 1.15 (1.05–1.25)

Not currently insured 21,123 16.7 (15.9–17.6) 16.8 (16.0–17.7) 1.24 (1.16–1.33)

Abbreviation: APR, adjusted prevalence ratio; CI, confidence interval; FMD, frequent mental distress.
a Age adjusted (age groups of 18–25 y, 26–44 y, and 45–64 y were used) to the 2000 projected US population.
b Adjusted for age, sex, race/ethnicity, education level, marital status, employment status, federal poverty level, current smoking, leisure-time physical activity, body
mass index, and the number of self-reported, physician-diagnosed chronic conditions (including diabetes, coronary heart disease, stroke, current asthma, arthritis,
cancer, chronic obstructive pulmonary disease, history of depression, kidney disease, and disability). Disability was defined as respondents who were limited in any
way in any activities because of physical, mental, or emotional problems, or who had any health problem that required them to use special equipment, such as a
cane, wheelchair, special bed, or special telephone.
c Includes TRICARE (formerly CHAMPUS), Veterans Affairs, or military plan; Alaska Native, Indian Health Service, Tribal Health Services, or some other source.

PREVENTING CHRONIC DISEASE VOLUME 15, E95
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JULY 2018
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.

6 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2018/17_0523.htm