This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.
Although timely and accurate information during the COVID-19 pandemic is essential for containing the disease and reducing mental distress, an infodemic, which refers to an overabundance of information, may trigger unpleasant emotions and reduce compliance. Prior research has shown the negative consequences of an infodemic during the pandemic; however, we know less about which subpopulations are more exposed to the infodemic and are more vulnerable to the adverse psychological and behavioral effects.
This study aimed to examine how sociodemographic factors and information-seeking behaviors affect the perceived information overload during the COVID-19 pandemic. We also investigated the effect of perceived information overload on psychological distress and protective behavior and analyzed the socioeconomic differences in the effects.
The data for this study were obtained from a cross-national survey of residents in 6 jurisdictions in Asia in May 2020. The survey targeted residents aged 18 years or older. A probability-based quota sampling strategy was adopted to ensure that the selected samples matched the population’s geographical and demographic characteristics released by the latest available census in each jurisdiction. The final sample included 10,063 respondents. Information overload about COVID-19 was measured by asking the respondents to what extent they feel overwhelmed by news related to COVID-19. The measure of psychological distress was adapted from the Posttraumatic Stress Disorder Checklist for Diagnostic and Statistical Manual of Mental Disorders 5 (DSM-5). Protective behaviors included personal hygienic behavior and compliance with social distancing measures.
Younger respondents and women (b=0.20, 95% CI 0.14 to 0.26) were more likely to perceive information overload. Participants self-perceived as upper or upper-middle class (b=0.19, 95% CI 0.09 to 0.30) and those with full-time jobs (b=0.11, 95% CI 0.04 to 0.17) tended to perceive higher information overload. Respondents who more frequently sought COVID-19 information from newspapers (b=0.12, 95% CI 0.11 to 0.14), television (b=0.07, 95% CI 0.05 to 0.09), and family and friends (b=0.11, 95% CI 0.09 to 0.14) were more likely to feel overwhelmed. In contrast, obtaining COVID-19 information from online news outlets and social media was not associated with perceived information overload. There was a positive relationship between perceived information overload and psychological distress (b=2.18, 95% CI 2.09 to 2.26). Such an association was stronger among urban residents, full-time employees, and those living in privately owned housing. The effect of perceived information overload on protective behavior was not significant.
Our findings revealed that respondents who were younger, were female, had a higher socioeconomic status (SES), and had vulnerable populations in the household were more likely to feel overwhelmed by COVID-19 information. Perceived information overload tended to increase psychological distress, and people with higher SES were more vulnerable to this adverse psychological consequence. Effective policies and interventions should be promoted to target vulnerable populations who are more susceptible to the occurrence and negative psychological influence of perceived information overload.
The COVID-19 pandemic has posed unprecedented challenges to public health and daily life worldwide. A cluster of COVID-19 cases was first reported in December 2019 in Wuhan of Hubei Province in China. Due to the proximity and various links to China, COVID-19 badly hit Asia early on. By imposing strict public health measures, some countries in Asia and the Pacific had better performance in containing the spread of COVID-19 compared with the rest of the world. However, the outbreaks of the Delta variant in several Asian countries and regions, including India, Singapore, Taiwan, and Thailand [
The unpredictable course of the pandemic along with prolonged social distancing can negatively affect people’s mental health, regardless of exposure to the disease itself [
In addition, the psychological impact of COVID-19 has been fueled by an “infodemic,” which refers to “an overabundance of information—some accurate and some not—that makes it hard for people to find trustworthy sources and reliable guidance when they need it” [
Although timely and accurate information during the pandemic helps individuals develop adequate risk perceptions, take preventive measures, and reduce mental distress, an infodemic and information overload can trigger unpleasant emotions, cause confusion and distrust among people, and impede effective public health responses [
To date, only a few studies have investigated the prevalence and consequences of perceived information overload during the COVID-19 pandemic, most of which discussed the effect of information overload on compliance with protective behaviors [
Besides a lack of research on the psychological consequences of information overload about COVID-19, most existing studies only examined the impact of information overload in the general population while ignoring the potentially stratified impact of information overload in different subpopulations. Moreover, the scope of previous studies on information overload is limited. A recent systematic review of health information overload pointed out that most research has been conducted in the United States, focusing on cancer information overload perceived by cancer patients and thus called for extending the scope to other health issues in different contexts [
To fill the gaps, this paper aimed to systematically examine the level, associated factors, and psychological and behavioral consequences of perceived information overload about COVID-19 and explore the sociodemographic variances in the susceptibility to and impact of information overload. The data were collected from a large-scale, cross-sectional survey of 10,063 residents in 6 jurisdictions in Asia, including Hong Kong, Taiwan, Japan, South Korea, Singapore, and Thailand, in May 2020. We focused on 4 research questions: (1) Which segments of the population perceived higher levels of information overload during the COVID-19 pandemic? (2) How has information-seeking behavior (ie, the frequency of accessing and perceived trustworthiness of COVID-19 information from different sources) affected the perception of information overload? (3) Would perceived information overload negatively affect psychological well-being and preventative behavior during the pandemic? (4) If so, which subpopulations are more vulnerable to the psychological and behavioral consequences of information overload? The findings would help identify vulnerable groups who are more susceptible to information overload about COVID-19 and its psychological and behavioral consequences and thus contribute to a nuanced understanding of the correlates and consequences of an infodemic and information overload during the COVID-19 pandemic.
As information overload arises when there is much more information available than an individual’s information processing capacity [
The empirical results of the relationship between media exposure and information overload were mixed [
Although access to timely and quality health information during outbreaks of infectious diseases can effectively contain the spread of diseases and reduce depressive and anxious feelings [
Researchers have also investigated the role of various information sources in psychological well-being and coping behavior during the COVID-19 pandemic. Previous studies in mainland China [
Despite the well-documented relationship between perceived information overload and psychological distress, limited studies have investigated the socioeconomic differences in the relationship between perceived information overload and psychological well-being. Prior research suggested that confusing and ambiguous information is especially problematic for those experiencing communication inequality, such as the lack of access to relevant health information or the ability to make sense of information [
In addition to psychological responses to information overload, people’s health behavior is also likely to be influenced by information overload. Previous studies on health information overload have consistently shown that those who perceive higher information overload are less likely to perform health behaviors [
The data for this study were obtained from a cross-national survey of public attitudes and responses toward COVID-19 in 6 jurisdictions in East and Southeast Asia, including Hong Kong, Taiwan, Japan, South Korea, Singapore, and Thailand. The survey was conducted by a group of scholars at the City University of Hong Kong between May 11, 2020 and May 26, 2020. The 6 regions were selected due to their geographical proximity to mainland China, the original epicenter of the COVID-19 pandemic, and were hardly hit by the pandemic since late January 2020. In late March 2020, they all entered a second wave of the pandemic as more imported cases from Europe and the United States were detected. They are the Tiger economies characterized by relatively high economic development and governing capacity in Asia. Yet, they also vary in regime types, media development, the stringency of public health measures, and effectiveness in containing the pandemic. Thus, these cases provide an ideal mixture of similarities and differences to examine the impacts of information-seeking behavior, perceived information overload, and psychological well-being during the COVID-19 pandemic.
The surveys were completed using online panels provided by a globally recognized professional survey company. The company’s online panels consist of an opt-in list of 56,000 to 1,440,000 individuals relative to the population size in the 6 jurisdictions surveyed in this study. Online panels have been used increasingly among psychological, social, and medical research [
For this study, we requested nationally representative samples of around 2000 adults aged 18 years or older in each of the 6 jurisdictions. Age and gender sampling quotas were set to match the latest available census estimates for age and gender in each jurisdiction. Participants were invited through email messages with an embedded link. The panel provider continuously invited participants until the predetermined quota was met. To increase the response rate, participants would get modest monetary rewards upon completion of the survey. Participation was voluntary, and all responses were anonymous. Details of the survey method of this project can be found elsewhere [
We developed a questionnaire that includes questions on perceived information overload about COVID-19, information-seeking behavior, psychological well-being, and protective behavior during the pandemic. We conducted a pre-test of the survey questions and modified wordings based on the feedback from the pre-testers. The questionnaire was available in English, Chinese, Korean, Japanese, and Thai for participants from different jurisdictions. A total of 12,062 representative respondents was collected, with approximately 2000 individuals in each jurisdiction. Cases with incomplete information were excluded from the analysis. The final sample size was 10,063 (see
Sample characteristics.
Variable | Full sample (N=10,063), n (%) | Hong Kong (n=1813), n (%) | Japan (n=1372), n (%) | Singapore (n=1681), n (%) | South Korea (n=1749), n (%) | Taiwan (n=1695), n (%) | Thailand (n=1753), n (%) | |
|
||||||||
|
18-29 | 2444 (24.29) | 413 (22.78) | 254 (18.51) | 400 (23.80) | 431 (24.64) | 411 (24.25) | 535 (30.52) |
|
30-39 | 2441 (24.26) | 425 (23.44) | 304 (22.16) | 389 (23.14) | 441 (25.21) | 408 (24.07) | 474 (27.04) |
|
40-49 | 2379 (23.64) | 470 (25.92) | 263 (19.17) | 385 (22.90) | 451 (25.79) | 394 (23.24) | 416 (23.73) |
|
50-59 | 1970 (19.58) | 392 (21.62) | 348 (25.36) | 346 (20.58) | 273 (15.61) | 357 (21.06) | 254 (14.49) |
|
≥60 | 829 (8.24) | 113 (6.23) | 203 (14.80) | 161 (9.58) | 153 (8.75) | 125 (7.37) | 74 (4.22) |
|
||||||||
|
Male | 5258 (52.25) | 865 (47.71) | 761 (55.47) | 867 (51.58) | 927 (53.00) | 955 (56.34) | 883 (50.37) |
|
Female | 4805 (47.75) | 948 (52.29) | 611 (44.53) | 814 (48.42) | 822 (47.00) | 740 (43.66) | 870 (49.63) |
|
||||||||
|
Secondary school or below | 1690 (16.79) | 518 (28.57) | 316 (23.03) | 232 (13.80) | 169 (9.66) | 194 (11.45) | 389 (22.19) |
|
College or above | 8373 (83.21) | 1295 (71.43) | 1056 (76.97) | 1449 (86.20) | 1580 (90.34) | 1501 (88.55) | 1364 (77.81) |
|
||||||||
|
Urban | 8147 (80.96) | 1673 (92.28) | 802 (58.45) | 1648 (87.33) | 1580 (90.34) | 1475 (87.02) | 1149 (65.54) |
|
Rural | 1916 (19.04) | 140 (7.72) | 570 (41.55) | 213 (12.67) | 169 (9.66) | 220 (12.98) | 604 (34.46) |
|
||||||||
|
Lower or lower-middle class | 4034 (40.09) | 947 (52.23) | 658 (47.96) | 523 (31.11) | 867 (49.57) | 632 (37.29) | 407 (23.22) |
|
Middle class | 4915 (48.84) | 628 (34.64) | 523 (38.12) | 958 (56.99) | 719 (41.11) | 827 (48.79) | 1260 (71.88) |
|
Upper or upper-middle class | 1114 (11.07) | 238 (13.13) | 191 (13.92) | 200 (11.9) | 163 (9.32) | 236 (13.92) | 86 (4.91) |
|
||||||||
|
Working full time | 6278 (62.39) | 1345 (74.19) | 725 (52.84) | 1155 (68.71) | 886 (50.66) | 1253 (73.92) | 914 (52.14) |
|
Other | 3785 (37.61) | 468 (25.81) | 647 (47.16) | 526 (31.29) | 863 (49.34) | 442 (26.08) | 839 (47.86) |
|
||||||||
|
Privately owned housing | 6043 (60.05) | 1017 (56.09) | 893 (65.09) | 1210 (71.98) | 1152 (65.87) | 1200 (70.80) | 1310 (74.73) |
|
Other | 4020 (39.95) | 796 (43.91) | 479 (34.91) | 471 (28.02) | 597 (34.13) | 495 (29.20) | 443 (25.27) |
|
||||||||
|
Yes | 1378 (13.69) | 198 (10.92) | 244 (17.78) | 195 (11.60) | 343 (19.61) | 161 (9.50) | 237 (13.52) |
|
No | 8685 (86.31) | 1615 (89.08) | 1128 (82.22) | 1486 (88.40) | 1406 (80.39) | 1534 (90.50) | 1516 (86.48) |
|
||||||||
|
Yes | 2958 (29.39) | 535 (29.51) | 383 (27.92) | 378 (22.49) | 377 (21.56) | 620 (36.58) | 665 (37.93) |
|
No | 7105 (70.61) | 1278 (70.49) | 989 (72.08) | 1303 (77.51) | 1372 (78.44) | 1075 (63.42) | 1088 (62.07) |
|
||||||||
|
Yes | 2975 (29.56) | 533 (29.4) | 254 (18.51) | 489 (29.09) | 376 (21.50) | 490 (28.91) | 833 (47.52) |
|
No | 7088 (70.44) | 1280 (70.6) | 1118 (81.49) | 1192 (70.91) | 1373 (78.50) | 1205 (71.09) | 920 (52.48) |
The study was approved by the Human Subject Ethics Committee of the City University of Hong Kong (Ref No: 8-2020-04-E295-18). All necessary participant consent was obtained.
Psychological distress was gauged by the Posttraumatic Stress Disorder Checklist for Diagnostic and Statistical Manual of Mental Disorders (DSM-5; PCL-5) [
Protective behavior was assessed by asking the respondents to rate on a 7-point Likert scale the level of compliance with 6 protective behaviors suggested by the government (eg, “kept a distance of at least 2 meters to other people” and “wore a mask in public space”) in the past week. Higher scores indicated higher levels of compliance.
Perceived information overload about COVID-19 was measured by the question, “To what extent do you feel overwhelmed by news related to the COVID-19 pandemic?” (0 = “Not at all” to 6 = “Very much”). This item was selected and adapted from the Perceived Information Overload scale [
Information-seeking behavior was assessed by the frequency of accessing COVID-19 information from the following platforms: newspapers, television, online news outlets, and social media (jurisdiction-specific examples were included in blanket statements based on the most popular social media tools in each jurisdiction; 0 = “Never” to 6 = “Very frequently”). We also controlled for perceived source credibility and time spent searching for COVID-19 information. Perceived source credibility was gauged by asking the respondents to rate how trustworthy they considered news about COVID-19 from the sources mentioned above on a 7-point Likert scale (1 = “Not at all trustworthy” to 7 = “Very trustworthy”). We also controlled for the average time (in hours) the respondents spent viewing information about COVID-19 each day.
We adjusted for demographic variables, such as age (18-29, 30-39, 40-49, 50-59, ≥60 years) and sex (male vs female). We also included a series of socioeconomic factors, including education (secondary school or below vs college or above), rural or urban residence, employment status (working full-time vs other), and housing type (privately owned housing vs other). We also assessed perceived social status by asking the respondents to declare their perceived social position (lower or lower-middle class, middle class, upper or upper-middle class). In addition, we controlled for the perceived threat of COVID-19 and level of worry about contracting COVID-19. To assess the impact of having a vulnerable household member, we asked whether the respondent lived with a pregnant woman or an adult older than 65 years and whether they had a child aged 12 years or below.
Descriptive statistics are reported as mean and SD for normally distributed continuous variables or median and IQR in the case of skewed distributions (
Descriptive statistics of key covariates (N=10,063).
Key covariates | Results | Range | |
Psychological distress, mean (SD) | 14.27 (8.00) | 0-30 | |
Preventative behavior, median (IQR) | 6 (6.33-6.67) | 1-7 | |
Perceived information overload, median (IQR) | 3 (3-5) | 0-6 | |
Perceived susceptibility to COVID-19, median (IQR) | 5 (4-7) | 1-7 | |
Perceived severity of COVID-19, median (IQR) | 6 (5-7) | 1-7 | |
|
|||
|
Newspaper | 3 (1-5) | 0-6 |
|
TV | 5 (3-6) | 0-6 |
|
Online news outlet | 5 (4-6) | 0-6 |
|
Social media | 4 (3-6) | 0-6 |
|
Family and friends | 4 (3-5) | 0-6 |
|
|||
|
Newspaper | 5 (4-6) | 1-7 |
|
TV | 5 (4-6) | 1-7 |
|
Online news outlet | 5 (4-6) | 1-7 |
|
Social media | 4 (4-5) | 1-7 |
|
Family and friends | 5 (4-6) | 1-7 |
Association between sociodemographic variables and perceived information overload.
Sociodemographic variables | ba | 95% CI | |||||
|
|||||||
|
18-29 | Reference | N/Ab | N/A | |||
|
30-39 | 0.02 | –0.07 to 0.11 | .70 | |||
|
40-49 | –0.11 | –0.20 to –0.01 | .02 | |||
|
50-59 | –0.13 | –0.23 to –0.04 | .008 | |||
|
≥60 | –0.21 | –0.33 to –0.08 | .002 | |||
|
|||||||
|
Male | Reference | N/A | N/A | |||
|
Female | 0.20 | 0.14 to 0.26 | <.001 | |||
|
|||||||
|
Secondary or below | Reference | N/A | N/A | |||
|
College or above | 0.08 | –0.01 to 0.17 | .08 | |||
|
|||||||
|
Rural | Reference | N/A | N/A | |||
|
Urban | 0.05 | –0.04 to 0.13 | .28 | |||
|
|||||||
|
Lower or lower-middle class | Reference | N/A | N/A | |||
Middle class | –0.02 | –0.08 to 0.05 | .67 | ||||
|
Upper and upper-middle class | 0.19 | 0.09 to 0.30 | .001 | |||
|
|||||||
|
Other | Reference | N/A | N/A | |||
|
Working full time | 0.11 | 0.04 to 0.17 | .003 | |||
|
|||||||
|
Other | Reference | N/A | N/A | |||
|
Privately owned housing | 0.01 | –0.06 to 0.08 | .81 | |||
|
|||||||
|
No | Reference | N/A | N/A | |||
|
Yes | 0.18 | 0.09 to 0.27 | <.001 | |||
|
|||||||
|
No | Reference | N/A | N/A | |||
|
Yes | 0.16 | 0.09 to 0.23 | <.001 | |||
|
|||||||
|
No | Reference | N/A | N/A | |||
|
Yes | 0.35 | 0.28 to 0.43 | <.001 | |||
|
|||||||
|
Hong Kong | Reference | N/A | N/A | |||
|
Japan | 0.05 | –0.06 to 0.17 | .37 | |||
|
Singapore | 0.57 | 0.47 to 0.68 | <.001 | |||
|
South Korea | 0.94 | 0.84 to 1.05 | <.001 | |||
|
Taiwan | –0.41 | –0.52 to –0.31 | <.001 | |||
|
Thailand | 0.66 | 0.55 to 0.77 | <.001 | |||
Constant | 2.64 | 2.50 to 2.78 | <.001 |
aUnstandardized coefficient.
bN/A: not applicable
We then examined the effects of using different sources for COVID-19 information on perceived information overload after adjusting for perceived source credibility, average time spent on viewing COVID-19 information, and various sociodemographic variables (
The association between information seeking behavior and perceived information overload.
Information-seeking behaviors | ba,b | 95% CI | |||||
|
|||||||
|
Newspaper | 0.12 | 0.11 to 0.14 | <.001 | |||
|
TV | 0.07 | 0.05 to 0.09 | <.001 | |||
|
Online news outlet | 0.00 | –0.03 to 0.03 | .98 | |||
|
Social media | 0.02 | –0.00 to 0.04 | .13 | |||
|
Family and friends | 0.11 | 0.09 to 0.14 | <.001 | |||
|
|||||||
|
Newspaper | –0.00 | –0.03 to 0.03 | .94 | |||
|
TV | –0.05 | –0.08 to –0.01 | .01 | |||
|
Online news outlet | 0.05 | 0.01 to 0.08 | .01 | |||
|
Social media | 0.10 | 0.07 to 0.13 | <.001 | |||
|
Family and friends | 0.06 | 0.03 to 0.10 | <.001 |
aThe model was adjusted for sociodemographic variables, including age, sex, education, area, perceived social status, employment, housing types, a history of chronic diseases, having pregnant women or older adults (>65 years old) in the household, having children aged under 12 years in the household, and survey locations.
bUnstandardized coefficient.
The relationships between perceived source credibility and perceived information overload were mixed. Specifically, people who trusted COVID-19 information on television more experienced lower levels of information overload (b=–0.05, 95% CI –0.08 to –0.01;
Next, we investigated the effect of perceived information overload and information-seeking behavior on psychological distress. Model 1 in
Association between perceived information overload and psychological distress and preventive behaviors.
Variables | Model 1a: psychological distress | Model 2a: preventive behavior | ||||||
|
bb | 95% CI | bb | 95% CI | ||||
Perceived information overload | 2.18 | 2.09 to 2.26 | <.001 | 0.01 | –0.09 to 0.07 | .85 | ||
|
||||||||
|
Newspaper | 0.58 | 0.51 to 0.65 | <.001 | –0.06 | –0.13 to 0.00 | .06 | |
|
TV | –0.03 | –0.13 to 0.06 | .55 | 0.19 | 0.10 to 0.29 | <.001 | |
|
Online news outlet | –0.32 | –0.44 to –0.21 | <.001 | 0.37 | 0.26 to 0.47 | <.001 | |
|
Social media | 0.05 | –0.05 to 0.15 | .38 | 0.18 | 0.09 to 0.28 | <.001 | |
|
Family and friends | 0.29 | 0.18 to 0.40 | <.001 | 0.01 | –0.10 to 0.11 | .86 | |
|
||||||||
|
Newspaper | –0.27 | –0.39 to –0.15 | <.001 | 0.08 | –0.03 to 0.19 | .17 | |
|
TV | –0.26 | –0.39 to –0.12 | <.001 | 0.14 | 0.01 to 0.27 | .03 | |
|
Online news outlet | 0.05 | –0.09 to 0.19 | .55 | –0.00 | –0.13 to 0.13 | .98 | |
|
Social media | 0.56 | 0.43 to 0.68 | <.001 | –0.16 | –0.27 to –0.03 | .01 | |
|
Family and friends | 0.24 | 0.11 to 0.37 | <.001 | 0.18 | 0.06 to 0.31 | .004 | |
Perceived susceptibility | 0.80 | 0.71 to 0.88 | <.001 | 0.41 | 0.33 to 0.49 | <.001 | ||
Perceived severity | 0.30 | 0.20 to 0.39 | <.001 | 0.55 | 0.46 to 0.63 | <.001 |
aThe model was adjusted for sociodemographic variables, including age, sex, education, area, perceived social status, employment, housing types, a history of chronic diseases, having pregnant women or older adults (>65 years old) in the household, having children aged under 12 years in the household, and survey locations.
b Unstandardized coefficient.
Finally, we assessed whether the impacts of COVID-19 information overload on psychological distress varied by different sociodemographic characteristics. A series of 2-way interaction terms between perceived information overload and sociodemographic variables were computed.
As for SES, there was a positive interaction between COVID-19 information overload and an urban residence (b=0.23, 95% CI 0.04 to 0.41;
Associations between perceived information overload and psychological distress among Asian populations with different sociodemographic backgrounds.
Models and variables | ba | 95% CI | |||||
|
|||||||
|
Perceived IO | 2.08 | 1.93 to 2.24 | <.001 | |||
|
Age: 18-29 years | Reference | N/Ad | N/A | |||
|
Age: 30-39 years | –0.67 | –1.48 to 0.13 | .10 | |||
|
Age: 40-49 years | –1.26 | –2.07 to –0.46 | .002 | |||
|
Age: 50-59 years | –1.35 | –2.18 to –0.55 | .001 | |||
|
Age: ≥60 years | –3.23 | –4.28 to –2.18 | <.001 | |||
|
Perceived IO × 30-39 years | 0.18 | –0.03 to 0.39 | .10 | |||
|
Perceived IO × 40-49 years | 0.18 | –0.03 to 0.40 | .09 | |||
|
Perceived IO × 50-59 years | –0.01 | –0.23 to 0.21 | .94 | |||
|
Perceived IO × ≥60 years | 0.13 | –0.16 to 0.41 | .39 | |||
|
|||||||
|
Perceived IO | 2.07 | 1.96 to 2.17 | <.001 | |||
|
Sex: male | Reference | N/A | N/A | |||
|
Sex: female | –1.23 | –1.77 to –0.68 | <.001 | |||
|
Perceived IO × female | 0.24 | 0.10 to 0.39 | .001 | |||
|
|||||||
|
Perceived IO | 2.16 | 1.98 to 2.34 | <.001 | |||
|
Education: secondary or below | Reference | N/A | N/A | |||
|
Education: college or above | –0.05 | –0.75 to 0.65 | .90 | |||
|
Perceived IO × college or above | 0.02 | –0.17 to 0.21 | .83 | |||
|
|||||||
|
Perceived IO | 2.00 | 1.82 to 2.17 | <.001 | |||
|
Residential area: rural | Reference | N/A | N/A | |||
|
Residential area: urban | –0.53 | –1.22 to 0.16 | .13 | |||
Perceived IO × urban | 0.23 | 0.04 to 0.41 | .02 | ||||
|
|||||||
|
Perceived IO | 2.10 | 1.98 to 2.23 | <.001 | |||
Perceived status: lower or lower-middle class | Reference | N/A | N/A | ||||
Perceived status: middle class | –0.79 | –1.37 to –0.20 | .008 | ||||
Perceived status: upper and upper-middle class | 0.62 | –0.27 to 1.52 | .17 | ||||
|
Perceived IO × middle class | 0.18 | 0.02 to 0.33 | .03 | |||
|
Perceived IO × upper or upper-middle class | –0.07 | –0.31 to 0.16 | .53 | |||
|
|||||||
|
Perceived IO | 2.03 | 1.91 to 2.16 | <.001 | |||
|
Employment status: non-full time work | Reference | N/A | N/A | |||
|
Employment status: working full time | –0.76 | –1.33 to –0.20 | .008 | |||
|
Perceived IO × working fulltime | 0.24 | 0.09 to 0.39 | .002 | |||
|
|||||||
|
Perceived IO | 2.06 | 1.93 to 2.18 | <.001 | |||
|
Housing type: non-privately owned housing | Reference | N/A | N/A | |||
|
Housing type: privately owned housing | –0.77 | –1.34 to –0.21 | .007 | |||
|
Perceived IO × privately-owned housing | 0.20 | 0.05 to 0.35 | .01 |
aUnstandardized coefficient.
bAll the models adjusted for sociodemographic variables, including age, sex, education, area, perceived social status, employment, housing types, a history of chronic diseases, having pregnant women or older adults (>65 years old) in the household, having children aged under 12 years in the household, and survey locations. The model also controlled for perceived susceptibility to and perceived severity of COVID-19.
cIO: information overload.
dN/A: not applicable.
Model 2 in
This study is among the first to investigate the antecedents and consequences of information overload about COVID-19 among Asian populations. Using data from a cross-sectional survey of 10,063 residents of 6 jurisdictions in East and Southeast Asia, our study showed a high level of perceived information overload during the pandemic. Regression results further revealed that young people, women, people with a higher SES (ie, full-time workers, self-perception as being upper or upper-middle class), and those with vulnerable populations in the household were more likely to experience COVID-19 information overload. As for the behavioral consequence of information overload, the results showed no significant relationship between perceived information overload and protective behaviors during the pandemic. Consistent with previous studies, we found a positive relationship between perceived information overload and psychological distress. Notably, the association between perceived information overload and psychological distress was more substantial among women and people with a higher SES (urban residents, self-perceived as middle class, full-time workers, and people living in privately owned housing). The findings of this study contribute to a better understanding of the level and correlates of information overload during the COVID-19 pandemic and help identify subpopulations that are particularly susceptible to information overload and its potential downstream consequences. We discuss the main findings in the following sections.
Although the occurrence of information overload has been documented in other disease outbreaks, the level and consequences of information overload during the current global pandemic are unparalleled. On the one hand, the unprecedented scale and impacts of COVID-19 on public health and individual lives have led to intensive media coverage. Other epidemics such as Zika, Ebola, Middle East Respiratory Syndrome (MERS), and H1N1 (swine flu) have caused great damage in daily life. Still, the level of panic caused by COVID-19 is much more severe and has resulted in a large volume of news attention to COVID-19. Also, the high scientific uncertainty and rapidly evolving settings of COVID-19 create opportunities for generation of ambiguous, inaccurate, and conflicting information, which may amplify information overload. On the other hand, the media environment and spreading of information have significantly changed over the past several decades. When another deadly viral disease—Severe Acute Respiratory Syndrome (SARS)—broke out in 2003, none of the major social media outlets was present. In pandemics before the social media era, a multilayered process involving careful review by experts, editing by journals, and releasing essential information under embargo to journalists who seek third-party comments would be performed before releasing scientific knowledge to the public [
As for the first research question about which segments of the population perceived higher levels of information overload during the COVID-19 pandemic, our results showed that younger people (18-39 years old), women, and respondents with higher SES (having a college education or above, having a full-time job, and self-perceived as upper or upper-middle class) expressed higher levels of perceived information overload about COVID-19. The finding that women tended to experience more information overload about COVID-19 than men is consistent with findings of previous studies [
Though not consistent with previous studies [
Although several studies have suggested that people with higher SES may experience less information overload [
Moreover, the socioeconomic disparities in perceived information overload may gradually emerge as the pandemic unfolds. A study of South Korean residents found no sociodemographic differences in perceived information overload about COVID-19 during the early stage of the pandemic [
In addition, respondents with vulnerable family members were more sensitive to the threats from COVID-19 and, accordingly, were more attentive to relevant information. Our results support that argument by revealing that respondents with vulnerable significant others at home (eg, pregnant women, young children, or people over 65 years old) were more likely to feel overwhelmed by COVID-19 information. However, inconsistent with previous studies showing that people who perceived themselves to have less-than-excellent health were more susceptible to information overload [
As for the effect of information-seeking behavior on perceived information overload (second research question), we found that the use of traditional media (eg, newspaper, television) for seeking COVID-19 information was positively associated with perceived information overload. In contrast, the effect of getting COVID-19 information from online news outlets was not significant. Given that the internet and online media contribute to the over-proliferation of health information available to the lay public [
Our results underscore the critical role of trust in information sources in crisis management. Interestingly, the perceived reliability of COVID-19 information from various media channels exerted differential effects on perceived information overload in our study. Perceived trustworthiness of COVID-19 information from television was negatively associated with perceived information overload. It was consistent with prior research that higher trust in information sources predicted less distress by information and benefited crisis management [
As for the third research question about the psychological and behavioral consequences of information overload, we found that an overabundance of COVID-19 information can harm mental well-being by increasing the likelihood of experiencing posttraumatic stress disorder. Such a finding was consistent with previous studies about the negative psychological impact of information overload [
However, our results did not show a significant effect of perceived information overload on protective behaviors during the COVID-19 pandemic, which was consistent with a study conducted during the COVID-19 pandemic in Korea [
The fourth research question focused on who was more vulnerable to the psychological and behavioral consequences of information overload. Since there was no significant effect of perceived information overload on protective behaviors, we did not further examine whether such an effect may be different across various sociodemographic characteristics. As for psychological distress, we found that women and people with a higher SES were more vulnerable to the adverse effects of perceived information overload on psychological well-being. Female respondents were more likely to experience psychological distress when exposed to information overload, possibly because women are more likely than men to perceive the pandemic as a severe health problem and to agree and comply with restraining measures [
Meanwhile, individuals with a lower SES tend to experience more everyday stressors even when there is no global pandemic. Thus, they may not undergo a significant change in their subjective well-being when reading pessimistic news about COVID-19. Besides, people with higher education care more about the quality of information and may feel frustrated if they cannot find trustworthy sources during the pandemic. Given the ambiguous information and inconsistent guidance about COVID-19, especially during the early stage of the outbreak, it is hard to identify valuable and reliable information even for people with a higher SES who are believed to have more access to health information and a higher ability to process such information.
Since information overload can harm mental well-being and potentially reduce compliance during the pandemic, it is urgent to manage the overload of information that exceeds people’s cognitive ability to process. From the information provider side, government and media should disseminate evidence-based and transparent information swiftly and widely among the public. Social media technology companies must constantly review content shared on their platforms and closely verify the reliability of information related to the pandemic. Since there are stratified mechanisms and consequences of information overload as shown in this study, information policies and management should be accordingly “stratified” as well. It is essential to develop efficient health communication strategies targeting people with different sociodemographic characteristics. Certain subgroups may be more frustrated with the uncertainty caused by the pandemic and eagerly look for sources to fill their information needs. It is necessary to formulate different information dissemination strategies in terms of information content and channels for different groups. It is also essential to establish interventions to help people vulnerable to information overload to cope with information anxiety and mental health concerns.
On the receiver side, individuals should carry out an “information diet” by controlling the extent and type of information they consume. Researchers have suggested that people visit authentic and official websites for COVID-19 information and try to verify suspicious news on one of the many fact-checking websites dedicated to debunking myths [
Despite the significant findings, this study is not without limitations. First, the data were obtained from a cross-sectional survey, and it is hard to ascertain the causal relationship between variables. For example, our results showed a negative effect of perceived information overload on psychological well-being. However, trait anxiety was significantly associated with higher levels of perceived information overload [
Perceptions of information overload are prevalent during the COVID-19 pandemic and have caused significant psychological and behavioral consequences. This study is among the first to examine how the antecedents and consequences of perceived information overload vary between different sociodemographic groups among the Asian population. A cross-sectional survey with representative data of 10,063 residents in 6 jurisdictions in Asia was conducted in May 2020. Regression results confirmed a positive relationship between perceived information overload and psychological distress. Also, people with a higher SES were more exposed to information overload about COVID-19 and suffered more psychological distress because of perceived information overload. Our findings suggested that the provision of trustworthy information and reduction of the perceived information overload can significantly ameliorate psychological distress during the pandemic. Effective policies and interventions should be promoted to target higher-SES populations who are more susceptible to the occurrence and adverse psychological influence of perceived information overload.
Tests of Ordinal Least Squares (OLS) assumptions.
Checklist for Reporting Results of Internet E-Surveys
Diagnostic and Statistical Manual of Mental Disorders 5
Middle East Respiratory Syndrome
ordinary least squares
Posttraumatic Stress Disorder Checklist for DSM-5
Severe Acute Respiratory Syndrome
socioeconomic status
World Health Organization
This study was supported by a Knowledge Transfer Earmarked Fund from Hong Kong University Grants Commission (6354048). The funder had no role in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.
None declared.