key: cord-0777267-26i5fwlu authors: Malik, A. A.; McFadden, S. M.; Elharake, J. A.; Aguolu, O. G.; Shafiq, M.; Omer, S. B. title: COVID-19 Risk Perception Among U.S. Adults: Changes from February to May 2020 date: 2020-08-23 journal: nan DOI: 10.1101/2020.08.20.20178822 sha: ec6ec7f1ea0d723576513cdf952f2db04fb04c75 doc_id: 777267 cord_uid: 26i5fwlu The COVID-19 pandemic continues to detrimentally impact the United States. Using a survey, we collected demographic and COVID-19 risk perception, behavior, knowledge, and attitude data from 672 adults across the U.S. in May 2020. These variables were compared with the results from a survey in February 2020. Participants who were older (55+ years; M = 6.3, SD = 2.0), identified as Native American/Alaska Native (M = 6.8, SD = 1.0) or Asian (M = 6.0, SD = 2.0), and those who had contracted (M = 6.8, SD = 2.0) or knew someone who had contracted COVID-19 (M = 6.2, SD = 1.7) reported higher perceived risk. Health behaviors, such as physical distancing, have shown to impact infectious disease trajectories. As the U.S. reopens its economy, public health officials and politicians must formulate culturally appropriate and evidence-based messaging and policies, based on the public's COVID-19 risk perceptions, to encourage preventive behaviors. In less than six months, the coronavirus disease 2019 (COVID- 19) pandemic has become the most challenging global health pandemic since the 1918 flu. 1 Despite local and state governments implementing measures to reduce the spread of the virus, the COVID-19 pandemic continues to threaten human life and the immediate return to normalcy. As of August 20, 2020, the U.S. leads the world in confirmed cases (5.5 million; 1.7% of U.S. population) as well as total number of confirmed deaths (175,000). 2 Although many Americans are following the Centers for Disease Control and Prevention's (CDC) COVID-19 guidelines, such as physical distancing, wearing a mask, and frequent hand washing, others have decided to forgo these safe and protective measures. 3 Americans typically make health decisions by weighing risk for consequences with benefits of action, which necessitates understanding risk perception trends across the country. 4 This need is further elevated because the number of COVID-19 cases and deaths continues to rise in the U.S. Understanding the societal and cultural factors that drive public health risk perception and associated health outcomes, therefore, becomes a high priority. Studies conducted on previous pandemics demonstrate strong evidence on how compliance with health behaviors recommended by public health officials can significantly influence the trajectory of an outbreak. [5] [6] [7] [8] [9] [10] [11] [12] [13] With COVID-19 vaccines currently under development and not likely to be available for at least several months, 14 U.S. public health officials and politicians must evaluate the public's risk perceptions of COVID-19 to formulate strategic messaging and policies in hopes of increasing safe health behaviors and effectively controlling the COVID-19 pandemic. Additionally, while all Americans are impacted by COVID-19, vulnerable populations are facing a disproportionate burden of cases and deaths. 15 Due to social and structural determinants of health and inequitable access to healthcare, 16 current data show the most alarming disparities are observed among low-socioeconomic populations and African American, Hispanic, and Native American communities. [17] [18] [19] [20] [21] [22] While considering this nuance, public health and political leaders must also develop culturally competent messaging strategies, especially given the medical community's history with conducting unethical studies within underrepresented communities, 23 that cater to vulnerable populations to mitigate these outcomes and promote health equity. Strict policies for infection prevention including quarantine, travel restrictions, closing community facilities, and cancelling major events were endorsed by most participants ( Table 2 ). Participants that agreed or strongly agreed with these policies had a higher risk score than participants who disagreed with these policies (Table 2) . Interestingly, 36% of participants agreed or strongly agreed that non-violent inmates should be released from jail or prison to decrease crowding and slow the pandemic (Table 2) . Those who agreed with releasing inmates had a higher risk perception score (M = 6.4, SD = 1.8) than those who disagreed (M = 5.6, SD = 2.1). Additionally, risk perception was higher (M = 6.2, SD = 1.9) in the 30% of participants who agreed that in the case of a pandemic it is reasonable to "temporarily discriminate" against a community based on their country of origin than in the 70% (M = 5.7, SD = 2.1) who did not agree with this statement. Most participants reported it was easy to follow the CDC public health recommendations (n = 395, 59%). For the 115 participants (17%) who reported it was hard to follow the CDC guidelines, the most common reasons were "We don't know how long this pandemic is going to last" and "I'm classified as an essential worker" (Table 3) . Excluding the use of facemasks, over 90% of our sample reported following CDC guidelines for infection prevention (S Figure 1 ). For facemasks specifically, 85% (n = 571) of our sample reported using them. Participants who practiced or planned on buying extra medicines, food and other everyday items, and disinfectants had a statistically significant higher risk perception as compared to those who did not (S Table 2 ). Regardless of employment status, most participants agreed that the COVID-19 pandemic has severely damaged the economy (n = 566, 84%) and that slowing the spread of COVID-19 is more important than the economy (n = 457, 68%). Additionally, 70% (n = 473) of participants said that public health should be prioritized over opening the economy (S Table 3) . Overall confidence in healthcare professionals (M = 4.2, SD = 1.1) and health officials (M = 4.2, SD = 1.2) was high, and confidence in social media was low (M = 2.9, SD = 1.5; S Figure 2 ). With respect to the COVID-19 pandemic specifically, participants reported high confidence in information from health professionals (M = 4.2, SD = 1.0) and their own physician (M = 4.2, SD = 1.1; S Figure 2 ). Participants reported the lowest confidence in COVID-19 pandemic information from the White House (M = 2.9, SD = 1.6) and Congress (M = 2.8, SD = 1.4; S Figure 2 ). When compared to the survey administered in February 2020, there was significant reduction in the reported confidence levels for CDC, DHHS, NIH, and other professional organizations such as AMA (S Table 4 ). We asked participants to rank who they wanted the U.S. COVID-19 response to be led by, and 35% (n = 237) of participants ranked the president as their first choice to lead the pandemic response (S Figure 3 ). This is in contrast the survey results in February 2020 where 13% (n = 91) of participants wanted the president to lead the response, (p < 0.01). Comparatively, 17% (n = 117) of participants wanted the CDC to lead the response, in contrast to the 53% (n = 382) who wanted the CDC to lead the response when surveyed in February (p < 0.01; S Figure 3 ). As sensitivity analysis, we carried out a weighted analysis for COVID-19 risk perception using U.S. demographics. Data for weighted analysis were extracted from U.S. Census data. 24 Weighting by age, gender, and race decreased the mean risk perception score from 5.9 to 5.2 and increased the percentage of population that supports "temporary discrimination" based on nationality from 30 to 38. While COVID-19 risk perceptions are high among the U.S. adult population, there is a great deal of variability among different demographic groups. Additionally, compared to our previous survey conducted in February 2020, a greater proportion of participants reported having an above average perceived knowledge about COVID-19. Although currently there are treatments, such as dexamethasone and remdesivir, that improve outcomes in patients experiencing COVID-19, at the time the survey was conducted there were no scientifically accepted treatments. However, over 15% of our participants thought there was a curative treatment or vaccine for COVID-19 and learned this information from television media. This finding is concerning and likely due to the increased level of misinformation and inconsistent messaging among various news outlets. 25 Therefore, our study demonstrates the continued need for targeted messaging about appropriate infection prevention and protective measures. Understanding COVID-19 risk perception and knowledge may help public health agencies anticipate and address potential barriers to preventive behavior to avert thousands of more infections and deaths by There is strong evidence that a perceived lack of consistent information during a crisis may lead to distrust, lack of confidence, and fear among the public. 27 While 35% of our participants ranked the president as their first choice to lead the pandemic response, participants reported the lowest confidence in COVID-19 pandemic information from the White House and Congress. Confidence in healthcare professionals and health officials was the highest; however, there was a significant reduction in the reported confidence level in the CDC compared to the February 2020 survey. These findings may be due to the contradicting COVID-19 information and lack of immediate response to the crisis from the White House, Congress, and CDC. 28 The White House, Congress, and public health organizations, especially the CDC, must agree on the information being communicated to the public and be transparent with the American people to ensure trust in public health and governmental authorities. There is great responsibility on politicians and healthcare leaders to create messaging that is clear and consistent, as well as evidenced-based and culturally competent. However, currently, there is evidence that non-Asian Americans' fear visiting local communities with higher concentration of Asian Americans, which is reflected in their hesitance in ordering from Asian-owned eateries. 29 This may be a result of fear in a subgroup of the population that there are greater chances of contracting COVID-19 from Asian-Americans than non-Asian Americans. Moreover, this notion is fueled by political leaders who repeatedly claim that China is responsible for the pandemic. 30 Our finding that 30% of our participants agreed that it is reasonable to "temporarily discriminate" against a community based on their country of origin in the case of a pandemic All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . supports this; those who agreed with "temporary discrimination" reported a higher risk perception than those who disagreed. Irrespective of employment status, 68% of our participants agreed that slowing the spread of COVID-19 was more important than the economic disruption. Furthermore, most of our sample supported the closure of community facilities and schools as well as the cancellation of large social events. This contrasts with the common narrative depicted by the media that most Americans are against the lockdown, and in doing so, the media may be spreading a narrative not supported by the majority. 31 Messaging strategies must build awareness and knowledge of risks and benefits of prevention methods, especially considering that 17% of our participants reported having difficulties following the CDC COVID-19 guidelines. These participants mostly consisted of workers classified as essential or those unable to pay their bills, which also reflects the detrimental economic impact of COVID-19. Public health leaders and politicians must identify areas of citizens' concerns and alleviate them through targeted messaging, and strongly consider providing more financial support for Americans. Our findings may be influenced by possible selection bias because participants needed a CloudResearch account and access to a smartphone or computer to participate. We had a response rate of 72% and our data are fairly representative of the U.S. adult population. Another strength of our study includes timeliness. This is also one of the first studies that looks at detailed COVID-19 risk perceptions and associated health behaviors. Moreover, due to it being a longitudinal study, we were able to identify changes in risk perception, behavior, attitude, and messaging needs among the U.S. adult population since the beginning of the COVID-19 pandemic. Consistent with previous literature, our study calls on the importance of building trust and understanding the risk perception landscape among the U.S. public. It also highlights the need for evidence-based messaging of recommended prevention measures to mitigate crises during the COVID-19 pandemic. Participants and procedure Data were collected using an electronic questionnaire via Qualtrics® (Qualtrics, Provo, UT). In early May 2020, participants completed a questionnaire on CloudResearch. 32 CloudResearch is an online survey platform that allows for representative surveying. The goal of our sampling was to be representative of the U.S. general population based on age, gender, education, race and ethnicity. Participants were eligible if they were 18 years of age or older, could read English, and had a CloudResearch account with access to the internet via computer or smartphone. Participants received compensation in the amount they agreed to with the platform through which they entered this survey. Basic demographic information was collected as well as zip code, state of residence, and employment status. Participants also completed the perceived risk scale (Cronbach's α =0.72) which had 10 survey-items (5-point Likert Scale: 0 = strongly disagree/disagree/neutral; 1 = agree/strongly agree). Variables used for risk perception score calculation are shown in S Table 5 . Participants were removed from score calculation if they selected "don't know" to one or more All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . of the items on the risk perception scale. For scoring, variables collected on Likert Scale were dichotomized as: 0 = strongly disagree/disagree/neutral; 1 = agree/strongly agree. Additionally, participants were asked about their knowledge of COVID-19, effective preventive measures, measures being practiced, confidence and the reliability of sources of information regarding the COVID-19 pandemic (5-point Likert Scale: 1 = strongly disagree to 5 strongly agree), support for restrictive infection prevention policies, and impact of COVID-19 on their employment and behavior. Finally, participants were asked to rank who they thought should lead the U.S. response to the COVID-19 pandemic. The full survey instrument is included as a supplement. This survey instrument was adapted from McFadden et al 14 and COSMO standard protocol for WHO. 33 A knowledge score consisting of 17 survey-items was constructed by adding all the points for correct items for each participant (Cronbach's α = 0.78; S Table 6 ). For scoring, variables collected on Likert Scale were dichotomized as: 0 = strongly disagree/disagree/neutral; 1 = agree/strongly agree. To compare the current risk perception and calculate change over time, we used data collected from a similar survey in early February 2020. Details of this study are reported here. 34 Descriptive statistics (frequencies, percentages) were calculated for the sample demographic characteristics, trust, attitude, and behavior of the participants. T-test and one-way ANOVA for continuous variables and chi-square test for categorical variables were used to compare the change in risk perception, knowledge, attitude, and behavior between the surveys in February and May 2020 using variables collected at both time points. To test the associations between demographic factors and exposure to COVID-19 with COVID-19 risk perception, a multi-variable linear regression analysis was used. Data were analyzed using Stata Version 16 (StataCorp, College Station, Texas). Yale University Institutional Review Board approved this study (IRB protocol number: 2000027891). Participants provided informed consent prior to data collection. The raw data will be accessible on contacting the corresponding author. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint Very concerned about this pandemic. 520 (77) Government is exaggerating the threat associated with the novel coronavirus/COVID-19. Media is exaggerating the threat associated with the novel coronavirus/COVID-19. Expect this pandemic to get larger. 407 (61) In case of a pandemic, it is reasonable to temporarily discriminate against a community based on their country of origin. Trust that health officials can effectively handle this crisis. Trust the Congress to effectively handle this crisis. 213 (32) Trust that the President can effectively handle this crisis. Trust that my state's Governor can effectively handle this crisis. United Americans who live in pandemic hotspots (such as New York City, Los Angeles, Seattle, and New Orleans) should be quarantined when traveling to other locations. Americans returning from any international location should be quarantined. Non-violent inmates should be released from jail or prison to decrease crowding and slow the spread of novel coronavirus/COVID-19. Major events should be cancelled by the organizers. 531 (79) Community facilities such as schools or kindergartens should be closed. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . (4) Categories are mutually exclusive Participants who reported a difficulty in following these recommendations (n = 115, 17.1%) are a part of a subset of the overall sample (n = 672) All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. May 2020 February 2020 Figure 1 . Comparison of knowledge score between survey administered in May to survey administered in February All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint S (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint S (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint S All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint S (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint S (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint S Figure 1 : Percent of sample who participated in CDC recommended infection prevention strategies All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20178822 doi: medRxiv preprint COVID-19: The first documented coronavirus pandemic in history COVID-19): Cases in the U Why are some people still refusing to practice social distancing? Health Risk perceptions and health behavior The effect of risk perception on the 2009 H1N1 pandemic influenza dynamics Game theory of social distancing in response to an epidemic Demographic and attitudinal determinants of protective behaviours during a pandemic: a review Public perceptions, anxiety, and behaviour change in relation to the swine flu outbreak: cross sectional telephone survey The spread of awareness and its impact on epidemic outbreaks Coupled contagion dynamics of fear and disease: mathematical and computational explorations Using social and behavioural science to support COVID-19 pandemic response How do we handle new health risks? Risk perception, optimism, and behaviors regarding the H1N1 virus Risk perceptions related to SARS and avian influenza: theoretical foundations of current empirical research COVID-19 vaccine development pipeline gears up COVID-19): COVID-19 in Racial and Ethnic Minority Groups The social determinants of health: it's time to consider the causes of the causes What do coronavirus racial disparities look like state by state? NPR Communities of color at higher risk for health and economic challenges due to COVID-19 Racial and ethinic disparities in population level COVID-19 mortality COVID-19 exacerbating inequalities in the US How COVID-19 is impacting indigenous peoples in the U.S. PBS Newshour For Native Americans, COVID-19 is 'the worst of both worlds at the same time'. The Harvard Gazette More than Tuskegee: understanding mistrust about research participation Population Estimates by Age, Sex, Race and Hispanic Origin Accessed 15 Relating Messages to Risk Perceptions: Crisis and Emergency Risk Communication Corner World Health Organization. Vaccination and trust: How concerns arise and the role of communication in mitigating crises New face mask guidance comes after battle between White House and CDC. The Washington Post Fear sent her Chinatown restaurant spiraling. The challenges to reopening feel 'just impossible Trump scapegoats China and WHO -Americans will suffer Armed protestors demonstrate against COVID-19 lockdown at Michigan capitol. The Guardian WHO Regional Office for Europe. COVID-19 Snapshot MOnitoring (COSMO Standard): Monitoring knowledge, risk perceptions, preventive behaviors, and public trust in the current coronavirus outbreak-WHO standard protocol Perceptions of the adult US population regarding the novel coronavirus outbreak