Knowledge, Perceptions, and Preferred Information Sources Related to COVID-19 Among Healthcare Workers: Results of a Cross Sectional Survey

Abstract

Purpose
To compare COVID-19 related knowledge, perceptions, and preferred information sources between healthcare workers and non-healthcare workers.

Design
Cross-sectional survey.

Setting
Web-based.

Subjects
Convenience sample of Pennsylvanian adults.

Measures
Primary outcomes were binary responses to 15 COVID-19 knowledge questions weighted by a Likert scale assessing response confidence.

Analysis
Generalized linear mixed-effects models to assess comparisons between clinical decision makers (CDM), non-clinical decision makers working in healthcare (non-CDM) and non-healthcare workers (non-HCW).

Results
CDMs (n = 91) had higher overall knowledge than non-CDMs (n = 854; OR 1.81 [1.51, 2.17], p < .05). Overall knowledge scores were not significantly different between non-CDMs (n = 854) and non-HCW (n = 4,966; OR 1.03 [0.97, 1.09], p > .05).

Conclusion
The findings suggest a need for improved education about COVID-19 for healthcare workers who are not clinical decision makers, as they play key roles in patient perceptions and compliance with preventive medicine during primary care visits.

Full text

Quantitative Research
Knowledge, Perceptions, and Preferred
Information Sources Related to COVID-19
Among Healthcare Workers: Results
of a Cross Sectional Survey
Shyama Sathianathan, BS
1
, Lauren Jodi Van Scoy, MD
2,3,4,5
, Surav Man Sakya, BS
1
,
Erin Miller, BS
5
, Bethany Snyder, MPH
2,5
, Emily Wasserman, MAS
4,5
,
Vernon M. Chinchilli, PhD
4
, John Garman, BS
1
, and Robert P. Lennon, MD, JD
6,7
Abstract
Purpose: To compare COVID-19 related knowledge, perceptions, and preferred information sources between healthcare
workers and non-healthcare workers.
Design: Cross-sectional survey.
Setting: Web-based.
Subjects: Convenience sample of Pennsylvanian adults.
Measures: Primary outcomes were binary responses to 15 COVID-19 knowledge questions weighted by a Likert scale assessing
response confidence.
Analysis: Generalized linear mixed-effects models to assess comparisons between clinical decision makers (CDM), non-clinical
decision makers working in healthcare (non-CDM) and non-healthcare workers (non-HCW).
Results: CDMs (n ¼91) had higher overall knowledge than non-CDMs (n ¼854; OR 1.81 [1.51, 2.17], p< .05). Overall
knowledge scores were not significantly different between non-CDMs (n ¼854) and non-HCW (n ¼4,966; OR 1.03 [0.97, 1.09],
p> .05).
Conclusion: The findings suggest a need for improved education about COVID-19 for healthcare workers who are not clinical
decision makers, as they play key roles in patient perceptions and compliance with preventive medicine during primary care visits.
Keywords
coronavirus, COVID19, health care workers, health personnel, surveys and questionnaires, knowledge, comprehension,
perception, news sources
Purpose
Healthcare worker (HCW) knowledge gaps in pandemic man-
agement may lead to delayed diagnosis, disease spread, and
poor infection control practices.
1
Previous studies have evalu-
ated HCW knowledge about COVID-19,
2-6
but little is known
about the scope of HCW knowledge about COVID-19 com-
pared with non-HCWs. This study compares HCW understand-
ing of COVID-19 topics pertaining to treatment, risk, severity,
prevention, and testing, to that of non-HCWs.
Methods
Design
This cross-sectional online survey was administered March
25-31, 2020; details of the survey design and validation are
1
School of Medicine, Penn State College of Medicine, Hershey, PA, USA
2
Department of Medicine, Penn State College of Medicine, Hershey, PA, USA
3
Department of Humanities, Penn State College of Medicine, Hershey, PA,
USA
4
Department of Public Health Sciences, Penn State College of Medicine,
Hershey, PA, USA
5
Qualitative and Mixed Methods Core, Penn State College of Medicine,
Hershey, PA, USA
6
Department of Family and Community Medicine, Penn State College of
Medicine, Hershey, PA, USA
7
Penn State Law, The Pennsylvania State University, University Park, PA, USA
Corresponding Author:
Shyama Sathianathan, School of Medicine, Penn State College of Medicine,
500 University Dr, Hershey, PA 17033, USA.
Email: ssathianathan@pennstatehealth.psu.edu
American Journal of Health Promotion
1-4
ªThe Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0890117120982416
journals.sagepub.com/home/ahp

described elsewhere.
7
Briefly, in the absence of any validated
COVID-19 survey, we modified the European “Standard
Questionnaire on Risk Perception of an Infectious Disease Out-
break.”
8
Thirteen cognitive interviews were conducted to
assess sensibility of questions; then, 1,000 responses from the
current survey pool (not included in our results) were used to
confirm knowledge differentiation was met.
Measures
Correct responses to 15 true-false questions were weighted by
confidence in the correct response assessed on a 5-point Likert
scale. In order to discriminate knowledge levels between
groups, a range of “difficulties” were included in questions,
making some questions less relevant to certain groups but
allowing for clearer discrimination of knowledge patterns.
Analysis
HCWs self-identified, and were analyzed as clinical decision
makers (CDMs; physicians, nurse practitioners, or physician
assistants), or non-clinical decision makers (non-CDM; all
other HCW). All others were classified as non-HCW.
Binary knowledge answers were analyzed via a generalized
linear mixed-effects model (GLMM) with a logistic link func-
tion and a random effect for the participant, modeling the prob-
ability of a correct response using SAS Version 9.4. The
inverse of the confidence score was used as the weighting
variable to allow responses answered in the greatest confidence
to carry more weight. The false-discovery rate was applied to
the evaluation of each of the individual 15 items, resulting in
adjusted p-values. The Penn State College of Medicine Insti-
tutional Review Board approved this study.
Table 1. COVID-19 Knowledge Compared Between Groups.
# Questions and Correct Responses
OR (95% CL)
CDM vs. Non-HCW Non-CDM vs. Non-HCW CDM vs. Non-CDM
1 Treatments for the symptoms of COVID-19 are available
without a prescription ¼TRUE
3.33 (2.01, 5.53) 1.24 (1.03, 1.49) 2.69 (1.59, 4.57)
2 Most hospitalized patients with COVID-19 should be treated in
an ICU ¼FALSE
3.39 (1.63, 7.05) 1.00 (0.81, 1.23) 3.39 (1.60, 7.20)
3 The CDC recommends using corticosteroids for COVID-19
patients with acute respiratory distress syndrome
(ARDS) ¼FALSE
2.60 (1.32, 5.12) 0.80 (0.63, 1.02) 3.25 (1.61, 6.57)
4 COVID-19 is the first coronavirus to cause disease in
humans ¼FALSE
2.88 (0.62, 13.43) 0.90 (0.62, 1.29) 3.21 (0.67, 15.44)
5 Patients with shortness of breath, fever, and cough should call
the emergency room prior to arrival ¼TRUE
1.17 (0.54, 2.57) 1.21 (0.90, 1.61) 0.97 (0.43, 2.22)
6 Patients whose first (early) symptoms are severe are more
likely to die from COVID-19 than those whose first (early)
symptoms are less severe ¼FALSE
0.56 (0.29, 1.05) 0.70 (0.55, 0.89) 0.80 (0.41, 1.55)
7 Children ages 5 and under are at higher risk of death from
COVID-19 ¼FALSE
2.14 (0.80, 5.69) 0.93 (0.70, 1.24) 2.29 (0.84, 6.28)
8 In someone who has not received the measles vaccine, measles
is more contagious than COVID-19 ¼TRUE
2.64 (1.49, 4.66) 1.05 (0.85, 1.30) 2.51 (1.38, 4.56)
9 The incubation period for the coronavirus that causes
COVID-19 is up to 21 days ¼FALSE
0.94 (0.55, 1.62) 1.02 (0.83, 1.24) 0.93 (0.52, 1.64)
10 Healthy people should wear facemasks to help prevent the
spread of COVID-19 ¼FALSE
2.01 (0.85, 4.72) 1.22 (0.94, 1.58) 1.64 (0.68, 3.97)
11 A vaccine for COVID-19 should be available within
approximately 3 months ¼FALSE
3.27 (0.80, 13.42) 0.82 (0.60, 1.13) 3.97 (0.94, 16.68)
12 CDC recommends the use of alcohol-based hand sanitizers
with greater than 60% ethanol or 70% isopropanol ¼TRUE
1.86 (0.53, 6.60) 0.90 (0.63, 1.28) 2.07 (0.57, 7.58)
13 Currently, the CDC recommends that everyone with
COVID-19 symptoms should get tested ¼FALSE
3.76 (2.01, 7.01) 1.47 (1.21, 1.79) 2.56 (1.34, 4.87)
14 Everyone who tests positive for COVID-19 should be treated
with hydroxychloroquine (Plaquenil
®
)or
chloroquine ¼FALSE
2.36 (0.57, 9.70) 1.06 (0.72, 1.57) 2.23 (0.52, 9.53)
15 COVID-19 testing is not recommended for individuals with
no symptoms, even if they were exposed to someone with
confirmed COVID-19 within the past 2 weeks ¼TRUE
2.84 (1.29, 6.25) 1.37 (1.09, 1.72) 2.07 (0.92, 4.66)
16 Total Score (15-Item) 1.86 (1.56, 2.22) 1.03 (0.97, 1.09) 1.81 (1.51, 2.17)
Non-HCW ¼a person not working in healthcare. Non-CDM ¼a person working in healthcare who is not a clinical decision maker. CDM ¼a person working in
healthcare who is a clinical decision maker (MD, DO, CRNP, PA). OR ¼Odds Ratio. CL ¼Confidence Limits. Statistically significant comparisons are bolded
(p < .05).
2American Journal of Health Promotion XX(X)

Results
Of 5,948 respondents,
7
37 were excluded for missing medical
profession status, for a net sample size of 5,911 (73%of those
who opened the survey, 4.9%of the entire list emailed).
Respondents were primarily white women; non-HCWs were
generally older and less educated than HCWs. Further demo-
graphics are reported elsewhere.
7
CDMs had the highest
knowledge (85%correct) compared to non-CDMs and
non-HCWs (75%correct). Questions and odds ratios (OR) of
probable correct responses between groups are reported in
Table 1. Table 2 reports the single most preferred information
sources as government websites (e.g. CDC) for all groups.
Television news channels were trusted among more
non-HCWs and non-CDMs than CDMs.
Discussion
To our knowledge, this is the first study of U.S. HCW’s knowl-
edge of COVID-19 topics related to treatment, risk, severity,
prevention, and testing in comparison with the general popula-
tion. CDMs demonstrated higher knowledge than non-HCWs
(OR 1.86 [1.56, 2.22], p< .05) (Table 1). Non-CDMs, however,
did not demonstrate significantly higher knowledge than
non-HCWs (OR 1.03 [0.97, 1.09], p > .05) (Table 1). This is
notable because non-CDMs are front-line staff (nurses and med-
ical assistants) who play an important role in public health
through extensive patient interactions—including influencing
vaccine uptake.
9
Improving knowledge in these key, influential
groups may be helpful for improving compliance with public
health recommendations.
CDMs were more knowledgeable about prognosis and treat-
ment of COVID-19, topics that were among the most debated
topics early in the pandemic, with constantly evolving infor-
mation released by government agencies, media outlets, and
scientific journals.
10
CDMs used government websites more
and television news channels less than non-CDMs and
non-HCWs.
Use of hydroxychloroquine became a focus of debate not
only in the scientific literature, but also in television news. This
may explain why non-HCW’s knowledge on the hydroxychlor-
oquine treatment question was not different than HCW’s
(Question 14, Table 1).
Probability of correct responses to questions about epide-
miology and transmission of coronavirus were similar between
groups, again reflecting their prominence in mass media early
in the pandemic. In situations where media and government
websites contradicted, for example, when news articles por-
trayed COVID-19 as more infectious than measles
11,12
,
non-CDMs and non-HCWs were less likely to give a correct
answer than CDMs (Question 8, Table 1).
Finally, CDMs were also significantly more knowledgeable
about COVID-19 testing than non-CDMs and non-HCWs. In
addition to the types of information sources used, contradicting
messages from state and national health institutions,
13
and con-
fusion from a lack of universal testing guidelines may also play
a role in different levels of knowledge among groups.
Limitations
Our study population includes a majority white population liv-
ing in a rural/suburban setting in a single state, which may limit
generalizability to racial minorities and urban residents. Our
study is also susceptible to selection bias as individuals who are
more interested in COVID-19 issues may be more likely to
Table 2. Current, Single Most Trusted Source for Information About
COVID-19 Among Groups.
Information source
Non-HCW
(N ¼4951)
Non-CDM
(N ¼850)
CDM
(N ¼91)
Social Media (Facebook,
Instagram, Twitter, etc.)
2% 2% 0%
Government Websites
(CDC, NIH, WHO)
41% 54% 66%
Television News channels 29% 17% 7%
Other (Family, Friends, Internet
News Websites, Radio, etc.)
28% 27% 27%
Non-HCW ¼a person not working in healthcare. Non-CDM ¼aperson
working in healthcare who is not a clinical decision maker. CDM ¼a person
working in healthcare who is a clinical decision maker (MD, DO, CRNP, PA).
So What?
What is already known on this topic?
To our knowledge there is no reported data on U.S.
healthcare worker knowledge of COVID-19. Studies in
other countries show a range of knowledge across dis-
ciplines, level of training, and world region.
2,3,5,6
What does this article add?
Our study is among the largest pandemic information
studies reported to date. This article identifies health-
care worker COVID-19 knowledge in a U.S. sample from
central Pennsylvania. It indicates that while clinical deci-
sion makers in healthcarehavegreaterCOVID-19
knowledge than the public, non-clinical decision makers
in healthcare do not. Further, it identifies the preferred
information source of healthcare workers, which
informs the ideal dissemination venue to provide them
education.
What are the implications for health promotion
practice or research?
Targeted education for allied health professionals is
needed to ensure that they can fulfill their vital role in
influencing public attitudes toward COVID-19 preven-
tion strategies and, ultimately, COVID-19 vaccination
uptake.
Sathianathan et al. 3

voluntarily respond to COVID-19 surveys. Our results are from
one point in time, and may not be generalizable as the pan-
demic evolves. Also, some correct answers to the knowledge
test are now outdated. However, while the recommendations
may change, recognizing differences in how well these groups
understood CDC recommendations remains meaningful.
Finally, while pilot-testing that included cognitive interviewing
was used to help ensure that knowledge questions were clear
and discriminated across various levels of knowledge, different
groups might have interpreted questions in different ways, con-
founding results.
Significance
Our data suggests a need to improve the COVID-19 knowledge
for HCW non-CDMs, as low health literacy is a major barrier to
containment of disease and public health,
14
and HCW
non-CDMs play a critical and influential role in overcoming
this barrier. Those who rely on news outlets versus government
websites have lower COVID-19 knowledge, highlighting the
need for public health education efforts to overcome the con-
flicting information and messages presented by news outlets.
Authorship Statement
Experimental design and data acquisition: L.J.V.S., S.M.S., E.M.,
B.S., E.W., V.M.C., J.G., and R.P.L. Analysis and interpretation of
data: S.S., L.J.V.S., E.M., B.S., E.W., V.M.C., and R.P.L. Manuscript
writing: S.S., L.J.V.S., E.M., B.S., E.W., and R.P.L. Manuscript revi-
sion and approval: All authors contributed to manuscript revision and
approval.
Consent to Participate
Participants provided written informed consent to participate in this
voluntary study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for
the research, authorship, and/or publication of this article: This study
was funded by the Huck Institutes of Life Sciences and the Social
Science Research Institute of Pennsylvania State University, and the
Department of Family and Community Medicine, Pennsylvania State
College of Medicine.
ORCID iD
Shyama Sathianathan, BS https://orcid.org/0000-0003-3988-5910
References
1. Omrani AS, Shalhoub S. Middle East respiratory syndrome
coronavirus (MERS-CoV): what lessons can we learn? J Hosp
Infect. 2015;91(3):188-196. doi:10.1016/j.jhin.2015.08.002
2. Ayinde OO, Usman AB, Posi A, Gbolahan A. A cross-sectional
study on Oyo state health care worker’s knowledge, attitude and
practice regarding corona virus disease 2019 (COVID-19). Adv
Infec Dis. 2020;10(3):6-15. doi:10.4236/aid.2020.103002
3. Olum R, Chekwech G, Wekha G, Nassozi DR, Bongomin F.Cor-
onavirus disease-2019: knowledge, attitude, and practices of
health care workers at Makerere University Teaching Hospitals,
Uganda. Front Public Health. 2020;8(April):181. doi:10.3389/
fpubh.2020.00181
4. Prescott K, Baxter E, Lynch C, Jassal S, Bashir A, Gray J.
COVID-19: how prepared are front-line healthcare workers in
England? J Hosp Infect. 2020;105(2):142-145. doi:10.1016/j.
jhin.2020.04.031
5. Saqlain M, Munir MM, Rehman SU, et al. Knowledge, attitude,
practice and perceived barriers among healthcare workers regard-
ing COVID-19: a cross-sectional survey from Pakistan. J Hosp
Infect. 2020;105(3):419-423. doi:10.1016/j.jhin.2020.05.007
6. Zhang M, Zhou M, Tang F, et al. Knowledge, attitude, and practice
regarding COVID-19 among healthcare workers in Henan, China.
JHospInfect. 2020;105(2):183-187. doi:10.1016/j.jhin.2020.04.012
7. Van Scoy LJ, Miller EL, Snyder B, et al. Knowledge, perceptions,
and preferred information sources related to COVID-19 among cen-
tral Pennsylvania adults early in the pandemic: results of a mixed
methods cross sectional survey. Ann Fam Med. 2020. In Press.
8. Europe ECiOMf. 2015. Standard questionnaire on risk perception
of an infectious disease outbreak. [online]. Accessed August 23,
2020. http://ecomeu.info/wp-content/uploads/2015/11/Standard-
questionnaire-risk-perception-ECOM-november-2015.pdf
9. Centers for Disease Control and Prevention. How nurses and
medical assistants can foster a culture of immunization in the
practice. 2019. [online]. Accessed August 23, 2020. https://
www.cdc.gov/vaccines/ed/vaccine-communication/foster-cul
ture-of-immunization.html
10. Zarocostas J. How to fight an infodemic. Lancet. 2020;
395(10225):676.
11. Flanagan R. How contagious is COVID-19 compared to other
viral diseases? 2020. [online] Coronavirus. Accessed August 23,
2020. https://www.ctvnews.ca/health/coronavirus/how-conta
gious-is-covid-19-compared-to-other-viral-diseases-1.4836734
12. Centers for Disease Control and Prevention. Measles is easily
transmitted. 2020. [online]. Accessed August 23, 2020. https://
www.cdc.gov/measles/transmission.html
13. Ian Freckelton QC. COVID-19: fear, quackery, false representa-
tions and the law. Int J Law Psychiatry. 2020;72:101611. [Online
ahead of print.].
14. Paakkari L, Okan O. COVID-19: health literacy is an underesti-
mated problem. Lancet Public Health. 2020;5(5):e249-e250.
4American Journal of Health Promotion XX(X)