Skip to main content
Download PDF

Trends in educational disparities in preventive behaviours, risk perception, perceived effectiveness and trust in the first year of the COVID-19 pandemic in Germany

BMC Public Health volume 22, Article number: 903 (2022) Cite this article

Abstract

Background

Educational disparities in health and health behaviours have always been relevant in public health research and are particularly challenging in the context of the COVID-19 pandemic. First studies suggest that factors important for the containment of the COVID-19 pandemic, such as prevention behaviour, risk perception, perceived effectiveness of containment measures, and trust in authorities handling the pandemic, vary by educational status. This study builds on recent debate by examining trends in absolute and relative educational disparities in these factors in the first year of the COVID-19 pandemic in Germany.

Methods

Data stem from four waves of the GESIS Panel surveyed between March and October 2020 in Germany (15,902 observations from 4,690 individuals). Trends in absolute and relative disparities were examined for preventive behaviour, risk perception, perceived effectiveness of COVID-19 containment measures, and trust in individuals and institutions handling the COVID-19 pandemic by educational status using sex, age, residence, nationality, children under 16 living in household, family status, household size, the Big Five Inventory, and income class as control factors. Descriptive statistics as well as unadjusted and adjusted linear regression models and random effects models were performed.

Results

We observed an initially rising and then falling trend in preventive behaviour with consistent and significant absolute and relative disparities with a lower preventive behaviour among low educated individuals. Indication of a U-shaped trend with consistent significantly lower values among lower educated individuals was found for risk perception, whereas perceived effectiveness and trust decreased significantly over time but did not significantly vary by educational status.

Conclusions

Results indicate persistent educational disparities in preventive behaviour and risk perception and a general decline in perceived effectiveness and trust in the first year of the COVID-19 pandemic in Germany. To address this overall downward trend and existing disparities, comprehensive and strategic management is needed to communicate the risks of the pandemic and the benefits of COVID-19 containment measures. Both must be adapted to the different needs of educational groups in particular in order to overcome gaps in preventive behaviour and risk perception by educational status.

Peer Review reports

Introduction

On March 11, 2020, the COVID-19 virus outbreak was officially classified as a pandemic. While more than 118,000 cases from 114 countries and a total of 4,291 deaths had been reported by that date, now more than 350 million cases of infection have been reported, with more than 5 million deaths [1]. The lessons learned from countries such as Italy, France or Spain [2,3,4], which were particularly hard hit at the beginning of the COVID-19 pandemic in Western countries, led many governments around the world to introduce various public health measures to contain the COVID-19 pandemic. These COVID-19 containment measures ranged from simple recommendations (such as keeping a minimum distance or stay-at-home recommendations) to severe restrictions (such as curfews or closures of educational institutions or of public spaces) [5, 6].

The effectiveness and public acceptance of measures to contain the course, duration, and consequences of a pandemic essentially require large-scale collective action of all citizens within a country. They have a vital role in the successful containment of the pandemic by appropriate preventive behaviours (such as staying at home) and a public support of public containment measures [7,8,9]. In addition to demographic factors such as age and gender, risk perception, perceived effectiveness of containment measures, and trust in individuals and institutions handling the pandemic play a key role in the effectiveness of public health measures to contain the COVID-19 pandemic [7, 10]. Risk perception is considered a core feature of psychological models of behaviour change, such as the Health Belief Model [11] or the Protection Motivation Theory [12], and is thus an important determinant of cooperation and adoption of preventive behaviours during pandemics [9, 13,14,15]. Perceived effectiveness and trust are considered crucial factors against detrimental psychological effects of governmental restrictions, and are known to support one's own preventive behaviour and to foster a positive social climate [7, 9, 10, 16, 17]. Recent studies suggest that individuals with low educational status not only have an increased incidence and severity of a COVID-19 infection, but also show less COVID-19 preventive behaviours than others [18]. According to single studies, these educational disparities are also demonstrated in risk perceptions [19], trust [20], and perceived effectiveness of COVID-19 containment measures [21].

Describing and explaining educational disparities in health and health behaviours have always been relevant in public health research and is particularly challenging in the context of the COVID-19 pandemic [22,23,24]. This study builds on this debate by examining trends in absolute and relative disparities in preventive behaviour, risk perception, perceived effectiveness of COVID-19 containment measures, and trust in individuals and institutions by educational status between March and October 2020 in Germany. To better contextualize this study, Fig. 1 provides the reported number of confirmed SARS-CoV-2 infections, COVID-19 related deaths, and COVID-19 patients in intensive care units per day and per 100.000 inhabitants as well as the degree of restrictions in Germany. These factors might correspond to trends in preventive behaviour, risk perception, perceived effectiveness, and trust by educational status, and should be considered when interpreting our findings.

Fig. 1
figure 1

Restriction stringency index, relative confirmed cases of SARS-CoV-2 infections and COVID-19 related deaths, and number of COVID-19 patients in intensive care units in Germany for the data collection period, based on ourworldindata.org. The stringency index is a composite measure based on nine indicators, including school closures, workplace closures, cancellation of public events, restrictions on public gatherings, closures of public transport, stay-at-home requirements, public information campaigns, restrictions on internal movements, and international travel controls, which range from 0 to 100 (100 = most severe) (see: ourworldindata.org)

Full size image

Methods

Study design and participants

We used data from the GESIS Panel established in 2013 at the GESIS-Leibniz-Institute for the Social Sciences in Mannheim, Germany. The GESIS Panel is a probability-based mixed-mode access panel that includes a representative sample of the German-speaking population between 18 and 70 years of age with permanent residence in Germany. The initial recruitment was based on random samples of individuals from population registries stratified by regions [25]. Since 2014, the GESIS Panel has been conducted every two months for a duration of 20 min as an online or offline questionnaire (n = 4,854 at baseline in 2014, response rate 86.4%) [25]. In 2016 and 2018, the GESIS Panel sample was supplemented by refresher samples taken from samples of the General Population Survey of the Social Sciences (ALLBUS). In 2020, the GESIS Panel comprised approximately 5,000 panellists (a detailed overview of the number of panel participants in the waves included in this study is available in Supplementary Table 1).

The GESIS Panel questionnaires are structured in modules. These modules have different thematic orientations in each of the six annual surveys. The COVID-19 module was introduced initially in the special survey on the Coronavirus SARS-CoV-2 Outbreak in Germany in March 2020 (baseline survey) and were subsequently implemented in three further waves of the survey in May to June 2020 (survey wave 2), July to August 2020 (survey wave 3), and August to October 2020 (survey wave 4). In order to be able to collect data on the pandemic in timely manner, only the subsample of online participants was considered in the special (baseline) survey. In the further survey waves, both online and offline participants were considered. The analytical sample consists of all individuals for which information on the variables considered was available (15,902 observations from 4,690 individuals) (a detailed overview of the total non-response rate and the item unit-nonresponse is available in Supplementary Table 2). The median participation rate was 3.39 observations per person, with 4.6% participating in one (n = 217), 8.3% in two (n = 389), 30.5% in three (n = 1,429), and 56.6% in all survey waves (n = 2,655). The characteristics of the analytical sample per survey wave are shown in Table 1.

Table 1 Sample characteristics of the analytical sample per survey wave
Full size table

Measurements

Dependent variables

Preventive behaviour was measured by asking individuals which measures they have taken in the last seven days. To answer this question, the respondents were given a multiple choice of nine different measures (response category: no vs. yes.) (the exact wording of all items is available in Supplementary Table 3). All items were combined into an additive scale that ranges from 0 to 100, where higher values indicate higher preventive behaviour. Internal consistency measured by the Kuder-Ricardson 20 formula proved to be adequate in each survey wave (KR 20 > 0.5).

Risk perception was measured by a sum score of five items. These capture the respondents' assessment of the likelihood that they or someone in their immediate environment would become infected with SARS-CoV-2, would need hospitalization due to a SARS-CoV-2 infection, would need to be in quarantine, or would infect other persons in the next 24 months (ranging from 1: “not at all likely” to 7: “absolutely likely”) (the exact wording of all items is available in Supplementary Table 4). All items were combined into an additive scale that ranges from 0 to 100, where higher values indicate higher perceptions of risks. The internal consistency measured by Cronbach’s alpha proved to be high in each survey wave (α > 0.8).

Perceived effectiveness of containment measures against SARS-CoV-2 infections was measured by a sum score of seven items. Respondents were asked to rate the effectiveness of measures such as the closure of various facilities and businesses, the ban on visiting hospitals and long-term care facilities, and exit restrictions in containing the COVID-19 pandemic (ranging from: 1 “not effective at all” to 5: “very effective”) (the exact wording of all items is available in Supplementary Table 5). All items were combined into an additive scale that ranges from 0 to 100, where higher values indicate higher perceived effectiveness of COVID-19 containment measures. The internal consistency measured by Cronbach’s alpha proved to be high in each survey wave (α > 0.8).

Trust in people and institutions to handle the COVID-19 pandemic was measured by a sum score of nine items. Respondents were asked to evaluate how much they trust general practitioners, local health authority, municipal and city administration, the Robert Koch Institute (German federal government agency and research institute for disease control and prevention), the federal chancellor, the federal government, the federal ministry of health, the World Health Organization, and scientists handling coronavirus (ranging from 1 “do not trust at all” to 5 “trust completely”) (the exact wording of all items is available in Supplementary Table 6). All items were combined into an additive scale that ranges from 0 to 100, where higher values indicate higher levels of trust in people and institutions to handle the COVID-19 pandemic. The internal consistency measured by Cronbach’s alpha proved to be high in each survey wave (α > 0.8).

Independent variable and covariates

Educational status was measured using the ISCED-97 scale [26]. We distinguish between low (ISCED 0–2), intermediate (ISCED 3), and high educational status (ISCED 4–6) in accordance with previous studies [27]. The information on educational status was taken from the last available survey wave, in which the educational status of the panellists was collected. The analyses controlled individual differences in factors that were known to relate to the dependent variables and educational status: sex, age, residence (East or West Germany), nationality (German, others), children under 16 living in household (no, yes), family status (unmarried, married or in partnership or widowed/divorced), household size (one, two, three or four or more persons), the Big Five Inventory (extraversion, agreeableness, conscientiousness, Neuroticism, and Open-Mindedness), and income class [15, 17, 18, 28].

Statistical analyses

The statistical analyses begin with a descriptive presentation of raw mean values of preventive behaviour, risk perception, perceived effectiveness, and trust for each survey wave stratified by educational status. raw mean values with 95% CIs are presented in Table 2. We tested for educational disparities in preventive behaviour, risk perception, perceived effectiveness, and trust for each survey wave by bivariate linear regression model. To identify the significance of linear or curvilinear trends of preventive behaviour, risk perceptions, perceived effectiveness, and trust, we fitted a random effects panel model with a linear and quadratic trend variable for the total sample and for each educational status group separately. The benefit of random effect modelling is the ability to control for serial correlation of the unobserved characteristics of each individual via time [29]. Therefore, the estimated standard errors will be corrected for the panel structure of the data.

Table 2 Raw mean level of preventive behaviour, risk perception, perceived effectiveness, and trust for the total sample and different educational groups by survey wave
Full size table

In a second step, marginal mean values of preventive behaviour, risk perception, perceived effectiveness, and trust were estimated by educational status and for each survey wave separately adjusted for sex, age, residence, nationality, children under 16 living in household, family status, household size, the Big Five Inventory, and income class (Fig. 2). The estimation of the adjusted marginal mean values was based on linear regression models. We also estimated random effects model for each educational status group including all covariates and a linear and quadratic trend variable to test for linear and quadratic trends of preventive behaviour, risk perception, perceived effectiveness, and trust.

Fig. 2
figure 2

Adjusted marginal means of preventive behaviour (A), risk perception (B), perceived effectiveness (C), and trust (D) by survey wave and educational status with 95%-Confidence Interval. Adjusted marginal means were based on linear regression models per survey wave including age, sex, region, nationality, children under 16 years in household, family status, household size, the big five inventory, and income class as control variables. Ref.: reference category. *p < 0.05; **p < 0.01; ***p < 0.001

Full size image

In a third step, we quantified the absolute and relative disparities in preventive behaviour, risk perception, perceived effectiveness, and trust by educational status and their trend over time using the Slope Index of Inequality (SII) and Relative Index of Inequality (RII) (Table 3 & 4). The SII represents an absolute difference in outcomes between the lowest and highest educational groups. The RII represents relative disparities in terms of the percentage difference from the population mean in the outcomes between the highest and lowest educational groups. To calculate the SII and RII, educational status was transformed into ridit scores, ranging from 0 (lowest educational status) to 1 (highest educational status). Two-sided 95% CIs for the RII and SII were estimated based on random effects panel modelling. Stepwise regression was used to examine factors affecting relative and absolute disparities in preventive behaviour, risk perception, perceived effectiveness, and trust by educational status. First, educational status, a linear and quadratic trend variable, and an interaction term between educational status and the trend variables was fitted to the data in regression model 1 (M1). Model 2 (M2) was based on M1, and also included the Big Five Inventory to test whether personal traits influence the association between educational status and the outcome measures. Model 3 (M3) was based on M2 and also included sex, age, residence, nationality, children under 16 living in household, family status, and household size, which are assumed to interact with educational status and the outcome measures. Finally, model 4 (M4) was based on M3 and also included income class, which strongly relates to educational status and might also influence educational disparities in preventive behaviour, risk perception, perceived effectiveness, and trust. The degree of model fit was assessed with McFadden’s pseudo R2. McFadden’s R2 ranges from 0 to 1, with higher values indicating a better model fit. Values between 0.2 and 0.4 are considered to be indicative of excellent model fit. Analyses were performed with Stata 16.1 (StataCorp, Texas, USA).

Table 3 Random effects model for absolute educational disparities (SII) in preventive behaviours, risk perception, perceived effectiveness, and trust (Number of observations: 15,902; number of individuals: 4,690)
Full size table
Table 4 Random effects model for relative educational disparities (RII) in preventive behaviours, risk perception, perceived effectiveness, and trust (Number of observations: 15,902; number of individuals: 4,690)
Full size table

Results

Raw mean levels and trends in preventive behaviour, risk perception, perceived effectiveness, and trust by educational status were shown in Table 2. For preventive behaviour, we observed a significant inverted U-shape trend for all educational groups with initially rising and then falling mean values from survey wave three onward. In each wave of the survey, individuals with low educational status are significantly less likely to engage in preventive behaviours than individuals with high educational status (p < 0.05). A significant non-linear trend with an indication of a U-shape trend can be identified in risk perception for all education groups. Raw mean values in risk perception fell sharply after the baseline survey and slightly increased at survey wave four. Over the entire observation period, persons with a low educational status had a significantly lower risk perception than persons with a high educational status (p ≤ 0.001). For raw mean values in perceived effectiveness of COVID-19 containment measures, we observed a significant steady decline over the course of the study across all educational groups. This trend was strongest for groups with intermediate educational status and resulted in significantly lower perceived effectiveness compared to low educational status groups in survey wave three (p = 0.006) and four (p = 0.001). Raw mean values in trust in persons and institutions handling the COVID-19 pandemic also significantly declined across all educational groups. Over the course of the study, raw mean values in trust did not significantly vary by educational status.

Figure 2 shows the marginal mean levels in preventive behaviour, risk perception, perceived effectiveness, and trust by educational status adjusted for sex, age, residence, nationality, children under 16 living in household, family status, household size, the Big Five Inventory, and income class (the full results are available in Supplementary Tables 7,8,9,10). Results shown in Fig. 2 did not substantially change after adjusting for the control variables compared to the results in Table 2. The different trend courses in preventive behaviour, risk perception, perceived effectiveness, and trust proved to be significant (p < 0.05) in all educational groups. Significant differences (p < 0.05) in preventive behaviour emerge in all survey waves to the disadvantage of low educational status groups. These educational differences were significant for risk perception from wave two onward and for trust only in wave four. Perceived effectiveness significantly varied by educational status only in survey wave four with significantly lower perceptions of effectiveness among intermediate educational groups compared to low educational groups.

Table 3 and Table 4 present the results for the multivariate analysis of trends in absolute and relative educational disparities in preventive behaviour, risk perception, perceived effectiveness, and trust (the full results are available in Supplementary Table 11,12,13,14,15,16,17,18). In all model specifications, people with the highest educational status show significantly higher preventive behaviour in comparison to the lowest educational status group not only in absolute terms but also in relative terms, as a percentage of the average in the population. Trends in SII were not significant indicating that absolute disparities in preventive behaviour did not vary over time (p > 0.05), whereas we noted a small and significant downward trend in RIIs in preventive behaviour (p < 0.05). For risk perception, we observed significant disparities by educational status in absolute and relative term that were not influenced by the control variables. Moreover, the non-significant trends for the SII and RII indicated that the absolute und relative disparities in risk perception did not vary over time. Finally, the results reveal consistent and non-significant absolute and relative disparities by educational status over time for perceived effectiveness and trust.

Discussion

The one-year longitudinal study found various trends and educational disparities in preventive behaviour, risk perception, perceived effectiveness, and trust. We observed an initially rising and then falling trend in preventive behaviour with consistent absolute and relative disparities with a lower preventive behaviour among low educated individuals. Indication of a U-shaped trend with consistent significantly lower values among lower educated individuals was found for risk perception, whereas perceived effectiveness and trust decreased significantly over time but did not vary by educational status.

The general trends in preventive behaviour, risk perception, perceived effectiveness, and trust coincide with the general pandemic pattern and the measures introduced to contain the COVID-19 pandemic as shown in Fig. 1. The initially increasing and subsequently decreasing general preventive behaviour can be explained on the one hand by the introduction of more restrictive containment measures, such as the obligation to wear a protective mask, and on the other hand by the low SARS-CoV-2 incidence between May and August 2020. This mild course of the pandemic, which gained pace again from September 2020 (Fig. 1), is possibly responsible for the indicated U-sharp trend in risk perception and illustrates the sensitivity of the respondents to infection patterns during the first year of the COVID-19 pandemic in Germany. Respondents appear to react far more sensitive to the perceived effectiveness of COVID-19 containment measures and trust in individuals and institutions handling the COVID-19 pandemic. Both the perceived effectiveness and trust decreased significantly in the first year of the COVID-19 pandemic in Germany, and this despite again increasing infection rates in the last wave of observation in August/October 2020 (Fig. 1).

The relatively high level of perceived effectiveness and trust at the beginning of the pandemic in March 2020 might be explained by an anxiety effect and a cognitive need for security [30, 31]. Accordingly, the uncertainty and severity of the pandemic might lead to the so-called 'rally round the flag' effect, which results in short-term support for the entire government or political leaders of a country by the population. As the intensity of the pandemic decreased from May 2020 onwards, as evidenced by the falling numbers of infections, emotional responses might again be replaced by rational evaluation of policies and measures indicated by decreases in perceived effectiveness and trust levels. According to Smith, in the context of pandemics, risk perception in particular influences public support for people and institutions handling the pandemic and evaluation of policy measures [32]. Risk perception is influenced by a variety of social and individual factors, such as age and gender [33], professional knowledge [34], individual impact of the pandemic and the implemented measures [33], media coverage [35] or the political coordination and communication of information [36], and can therefore deviate significantly from a political-scientific risk perception [37]. One indication of difference in risk perception might be that during the first year of the COVID-19 pandemic in Germany, the restrictiveness of COVID-19 containment measures was only slightly reduced despite a mild progression of the pandemic, while the risk perception, perceived effectiveness, and trust of respondents declined sharply. This difference in risk perception must therefore be more fully considered and aligned with the needs of different social groups when communicating political decisions regarding containment measures [38].

The finding that, contrary to other studies [19,20,21], both trust and perceived effectiveness did not differ by educational status might indicate how extensive and across all social strata the uncertainty regarding COVID-19 containing measures and authorities handling COVID-19 was during the mild course of the pandemic in Germany. A recent study for Germany showed that moderate opponents and supporters of COVID-19 containment measures overlapped in their criticism of anxiety-inducing media coverage and fuzzy governmental communication, which might explain the general downward trend in trust and perceived effectiveness observed in our study [39]. Wegwarth et al. further indicate that the political communication of pandemic threat scenarios was devoid of uncertainty in Germany, although the communication of uncertainties seems to be particularly effective in increasing compliance of those who are currently sceptical towards COVID-19 containment measures [40]. Our findings of consistent educational disparities in preventive behaviour and risk perception contradicts with a previous trend study for Germany that only found small and varying disparities in both outcomes during the first year of the pandemic; however, methodological differences in operationalizing educational status limit the comparability of the results [19]. We consider the stability of educational disparities in preventive behaviour and risk perception as an indication of the deep underlying disparities in health and health behaviours that were already existent before the COVID-19 pandemic in Germany [41]. These socially structured disparities and its underlying factors might interact with the COVID-19 pandemic, which is why the pandemic can also be understood as a kind of magnifying glass on societal structures and conditions associated with health and health behaviours [42]. In addition to a number of different material and psychosocial factors, such as inadequate financial resources or low self-efficacy, low health literacy in particular may explain lower levels of preventive behaviour and risk perception among individuals with a low educational status [43]. COVID-19-related studies found associations of educational status with preventive behaviours and COVID-19-related health literacy such as accessing and understanding of COVID-19-related information [44, 45]. Moreover, a recent trend study for Germany showed constant disparities in knowledge of COVID-19-related information by educational status over the course of the first year of the COVID-19 pandemic [19]. A further contribution factor to the consistent educational disparities in preventive behaviours and risk perceptions might be an inadequate communication by policymakers about the risks of a SARS-CoV-2 infection and the benefits of COVID-19 containment measures. In order to reach individuals with a low educational status as well as other social groups, planned risk management by leadership in times of pandemic is necessary [38].

Strengths and limitations

Strengths of the study include its large sample of adults, consistency in sampling and measurement over the first year of the pandemic in Germany that allow detailed descriptions of trends and differentiated analyses. Also, the use of the RII and SII provided a more complete account of trends in educational disparities in preventive behaviour, risk perception, perceived effectiveness, and trust than that either could index alone.

Our study also has limitations. First, some of the used outcomes may depend on individuals’ health status, which has not been surveyed in the GESIS Panel [18]. The results may therefore be slightly biased, as risk perception and health behaviour in particular may be influenced by individual health risk. Furthermore, in the GESIS Panel, the measurement of the perceived effectiveness of the COVID-19 containment measures referred exclusively to restrictive measures, such as the closure of recreational facilities or stores and curfews. Because these measures involve significant restrictions on individual lives, results could vary when the effectiveness of less restrictive measures, such as wearing a protective mask, is surveyed. The comparison between the first and subsequent surveys may be slightly biased because only online panellists were invited to participate in the first survey. Finally, the representativeness of the results for Germany is slightly limited because no design weight was available for the first survey wave. Sensitivity analyses of the other waves, however, have shown that the descriptive results only slightly change when a design weight was applied.

Conclusions

This study highlights the relevance of a comprehensive and strategic management in communicating the risks of the pandemic and the benefits of COVID-19 containment measures by politics and public health. Risk and benefit communication must be adapted to the different needs of social groups in order to overcome disparities in preventive behaviour and risk perception by educational status. Further trend studies need to show whether and to what extent the trends in disparities in preventive behaviour, risk perception, perceived effectiveness, and trust by educational status change as pandemic severity increases.

Availability of data and materials

All data are available for scientific use on the website: https://www.gesis.org/gesis-panel/gesis-panel-home. The analytic scripts in Stata are available from the corresponding author on reasonable request.

Abbreviations

ISCED:

International Standard Classification of Education

CI:

Confidence Interval

RII:

Relative Index of Inequality

SII:

Slope Index of Inequality

References

  1. Dong E, Du H, Gardner L. An interactive web-based dashboard to track COVID-19 in real time. Lancet Infect Dis. 2020;20:533–4. https://doi.org/10.1016/S1473-3099(20)30120-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Health TLP. COVID-19 in France: challenges and opportunities. The Lancet Public Health. 2021;6: e192. https://doi.org/10.1016/S2468-2667(21)00054-2.

    Article  Google Scholar 

  3. Remuzzi A, Remuzzi G. COVID-19 and Italy: what next? The Lancet. 2020;395:1225–8. https://doi.org/10.1016/S0140-6736(20)30627-9.

    Article  CAS  Google Scholar 

  4. Legido-Quigley H, Mateos-García JT, Campos VR, Gea-Sánchez M, Muntaner C, McKee M. The resilience of the Spanish health system against the COVID-19 pandemic. The Lancet Public Health. 2020;5:e251–2. https://doi.org/10.1016/S2468-2667(20)30060-8.

    Article  PubMed  PubMed Central  Google Scholar 

  5. ECDC & JRC. The Response Measure Database (RMD); 2020.

  6. Hale T, Angrist N, Goldszmidt R, Kira B, Petherick A, Phillips T, et al. A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker). Nat Hum Behav. 2021;5:529–38. https://doi.org/10.1038/s41562-021-01079-8 .

    Article  PubMed  Google Scholar 

  7. Romano A, Spadaro G, Balliet D, Joireman J, van Lissa C, Jin S, et al. Cooperation and Trust Across Societies During the COVID-19 Pandemic. J Cross-Cult Psychol. 2021;52:622–42. https://doi.org/10.1177/0022022120988913.

    Article  Google Scholar 

  8. Lang R, Benham JL, Atabati O, Hollis A, Tombe T, Shaffer B, et al. Attitudes, behaviours and barriers to public health measures for COVID-19: a survey to inform public health messaging. BMC Public Health. 2021;21:765. https://doi.org/10.1186/s12889-021-10790-0.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Schumpe BM, van Lissa CJ, Bélanger JJ, Ruggeri K, Mierau J, Nisa CF, et al. Predictors of adherence to public health behaviors for fighting COVID-19 derived from longitudinal data. Sci Rep. 2022;12:3824. https://doi.org/10.1038/s41598-021-04703-9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Scandurra C, Bochicchio V, Dolce P, Valerio P, Muzii B, Maldonato NM. Why people were less compliant with public health regulations during the second wave of the Covid-19 outbreak: The role of trust in governmental organizations, future anxiety, fatigue, and Covid-19 risk perception. Curr Psychol. 2021:1–11. https://doi.org/10.1007/s12144-021-02059-x.

  11. Champion VL, Skinner CS. The Health Belief Model. In: Kasprzyk D, editor. Health Behavior and Health Education. San Francisco: John Wiley & Sons; 2008. p. 45–66.

    Google Scholar 

  12. Rogers RW. Cognitive and Physiological Processes in Fear Appeals and Attitude Change: A Revised Theory of Protection Motivation. In: Cacioppo J, Petty R, editors. Social Psychophysiology. New York: Guilford Press; 1983. p. 153–77.

    Google Scholar 

  13. Nisa CF, Bélanger JJ, Faller DG, Buttrick NR, Mierau JO, Austin MMK, et al. Lives versus Livelihoods? Perceived economic risk has a stronger association with support for COVID-19 preventive measures than perceived health risk. Sci Rep. 2021;11:9669. https://doi.org/10.1038/s41598-021-88314-4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Dryhurst S, Schneider CR, Kerr J, Freeman ALJ, Recchia G, van der Bles AM, et al. Risk perceptions of COVID-19 around the world. J Risk Res. 2020;23:994–1006. https://doi.org/10.1080/13669877.2020.1758193.

    Article  Google Scholar 

  15. Schneider CR, Dryhurst S, Kerr J, Freeman ALJ, Recchia G, Spiegelhalter D, van der Linden S. COVID-19 risk perception: a longitudinal analysis of its predictors and associations with health protective behaviours in the United Kingdom. J Risk Res. 2021;24:294–313. https://doi.org/10.1080/13669877.2021.1890637.

    Article  Google Scholar 

  16. Dohle S, Wingen T, Schreiber M. Acceptance and adoption of protective measures during the COVID-19 pandemic: The role of trust in politics and trust in science. Soc Psychol Bull. 2020. https://doi.org/10.32872/spb.4315.

    Article  Google Scholar 

  17. Mækelæ MJ, Reggev N, Dutra N, Tamayo RM, Silva-Sobrinho RA, Klevjer K, Pfuhl G. Perceived efficacy of COVID-19 restrictions, reactions and their impact on mental health during the early phase of the outbreak in six countries. R Soc Open Sci. 2020;7: 200644. https://doi.org/10.1098/rsos.200644.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Lüdecke D, von dem Knesebeck O. Protective Behavior in Course of the COVID-19 Outbreak-Survey Results from Germany. Front Public Health. 2020;8: 572561. https://doi.org/10.3389/fpubh.2020.572561.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Rattay P, Michalski N, Domanska OM, Kaltwasser A, de Bock F, Wieler LH, Jordan S. Differences in risk perception, knowledge and protective behaviour regarding COVID-19 by education level among women and men in Germany. Results from the COVID-19 Snapshot Monitoring (COSMO) study. PLoS One. 2021;16:e0251694. https://doi.org/10.1371/journal.pone.0251694.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Price D, Bonsaksen T, Ruffolo M, Leung J, Chiu V, Thygesen H, et al. Perceived Trust in Public Authorities Nine Months after the COVID-19 Outbreak: A Cross-National Study. Soc Sci. 2021;10:349. https://doi.org/10.3390/socsci10090349.

    Article  Google Scholar 

  21. Meier K, Glatz T, Guijt MC, Piccininni M, van der Meulen M, Atmar K, et al. Public perspectives on protective measures during the COVID-19 pandemic in the Netherlands, Germany and Italy: A survey study. PLoS ONE. 2020;15: e0236917. https://doi.org/10.1371/journal.pone.0236917.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Bambra C, Riordan R, Ford J, Matthews F. The COVID-19 pandemic and health inequalities. J Epidemiol Community Health. 2020;74:964–8. https://doi.org/10.1136/jech-2020-214401.

    Article  PubMed  Google Scholar 

  23. Burström B, Tao W. Social determinants of health and inequalities in COVID-19. Eur J Public Health. 2020;30:617–8. https://doi.org/10.1093/eurpub/ckaa095.

    Article  PubMed  Google Scholar 

  24. Marmot M, Allen J. COVID-19: exposing and amplifying inequalities. J Epidemiol Community Health. 2020;74:681–2. https://doi.org/10.1136/jech-2020-214720.

    Article  PubMed  Google Scholar 

  25. Bosnjak M, Dannwolf T, Enderle T, Schaurer I, Struminskaya B, Tanner A, Weyandt KW. Establishing an Open Probability-Based Mixed-Mode Panel of the General Population in Germany. Soc Sci Comput Rev. 2018;36:103–15. https://doi.org/10.1177/0894439317697949.

    Article  Google Scholar 

  26. Schneider SL. Applying the ISCED-97 to the German educational qualifications. In: Schneider SL, editor. The international standard classification of education. Mannheim: MZES; 2008. p. 77–102.

    Google Scholar 

  27. Präg P, Subramanian SV. Educational inequalities in self-rated health across US states and European countries. Int J Public Health. 2017;62:709–16. https://doi.org/10.1007/s00038-017-0981-6.

    Article  PubMed  Google Scholar 

  28. Gozgor G. Global Evidence on the Determinants of Public Trust in Governments during the COVID-19. Appl Res Qual Life. 2021:1–20. doi:https://doi.org/10.1007/s11482-020-09902-6.

  29. Petersen T. Analyzing Panel Data: Fixed- and Random-Effects Models. In: Hardy M, Bryman A, editors. Handbook of Data Analysis. London. London: SAGE Publications, Ltd; 2004. p. 332–45.

    Chapter  Google Scholar 

  30. Schraff D. Political trust during the Covid-19 pandemic: Rally around the flag or lockdown effects? Eur J Polit Res. 2020. https://doi.org/10.1111/1475-6765.12425.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Altiparmakis A, Bojar A, Brouard S, Foucault M, Kriesi H, Nadeau R. Pandemic politics: policy evaluations of government responses to COVID-19. West Eur Polit. 2021;44:1159–79. https://doi.org/10.1080/01402382.2021.1930754.

    Article  Google Scholar 

  32. Smith RD. Responding to global infectious disease outbreaks: lessons from SARS on the role of risk perception, communication and management. Soc Sci Med. 2006;63:3113–23. https://doi.org/10.1016/j.socscimed.2006.08.004.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Chen Y, Feng J, An C, Lee JE, An L. Risk perception of COVID-19: A comparative analysis of China and South Korea. Int J Disaster Risk Reduct. 2021;61:102373. https://doi.org/10.1016/j.ijdrr.2021.102373.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Slovic P, Finucane ML, Peters E, MacGregor DG. Risk as analysis and risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk Anal. 2004;24:311–22. https://doi.org/10.1111/j.0272-4332.2004.00433.x.

    Article  PubMed  Google Scholar 

  35. Cinelli M, Quattrociocchi W, Galeazzi A, Valensise CM, Brugnoli E, Schmidt AL, et al. The COVID-19 social media infodemic. Sci Rep. 2020;10:16598. https://doi.org/10.1038/s41598-020-73510-5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Aleinikov AV, Kurochkin AV, Maltseva DA. Information Management under the Conditions of Global Pandemic. Sci Tech Inf Proc. 2021;48:38–44. https://doi.org/10.3103/S0147688221010068.

    Article  Google Scholar 

  37. Zinn JO. Introduction: Towards a sociology of pandemics. Curr Sociol. 2021;69:435–52. https://doi.org/10.1177/00113921211020771.

    Article  Google Scholar 

  38. Wardman JK. Recalibrating pandemic risk leadership: Thirteen crisis ready strategies for COVID-19. J Risk Res. 2020;23:1092–120. https://doi.org/10.1080/13669877.2020.1842989.

    Article  Google Scholar 

  39. Schieferdecker D. Beliefs, Attitudes, and Communicative Practices of Opponents and Supporters of COVID-19 Containment Policies: A Qualitative Case Study from Germany. Javnost - The Public. 2021;28:306–22. https://doi.org/10.1080/13183222.2021.1969620.

    Article  Google Scholar 

  40. Wegwarth O, Wagner GG, Spies C, Hertwig R. Assessment of German Public Attitudes Toward Health Communications With Varying Degrees of Scientific Uncertainty Regarding COVID-19. JAMA Netw Open. 2020;3: e2032335. https://doi.org/10.1001/jamanetworkopen.2020.32335.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Lampert T, Kroll LE, Kuntz B, Hoebel J. Health inequalities in Germany and in international comparison: trends and developments over time. Journal of Health Monitoring. 2018;3:1–24. https://doi.org/10.17886/RKI-GBE-2018-036.

    Article  Google Scholar 

  42. Blundell R, Costa Dias M, Joyce R, Xu X. COVID-19 and Inequalities. Fisc Stud. 2020;41:291–319. https://doi.org/10.1111/1475-5890.12232.

    Article  Google Scholar 

  43. Stormacq C, van den Broucke S, Wosinski J. Does health literacy mediate the relationship between socioeconomic status and health disparities? Integrative review Health Promot Int. 2019;34:e1–17. https://doi.org/10.1093/heapro/day062.

    Article  PubMed  Google Scholar 

  44. Okan O, Bollweg TM, Berens E-M, Hurrelmann K, Bauer U, Schaeffer D. Coronavirus-Related Health Literacy: A Cross-Sectional Study in Adults during the COVID-19 Infodemic in Germany. Int J Environ Res Public Health. 2020. https://doi.org/10.3390/ijerph17155503.

    Article  PubMed  PubMed Central  Google Scholar 

  45. McCaffery KJ, Dodd RH, Cvejic E, Ayrek J, Batcup C, Isautier JM, et al. Health literacy and disparities in COVID-19-related knowledge, attitudes, beliefs and behaviours in Australia. Public Health Res Pract. 2020. https://doi.org/10.17061/phrp30342012.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

None.

Funding

Open Access funding enabled and organized by Projekt DEAL. This study was funded by the German Research Foundation for the project ‘Socioeconomic inequalities in health during the COVID-19 pandemic (INHECOV): empirical analyses and implications for pandemic preparedness’ (project number 458531028).

Author information

Authors and Affiliations

  1. Research Methods Division, Faculty of Human Sciences, University of Cologne, Cologne, Germany

    Timo-Kolja Pförtner

  2. Institute of Medical Sociology, Health Services Research and Rehabilitation Science, Faculty of Human Sciences and Faculty of Medicine, University Hospital Cologne, University of Cologne, Cologne, Germany

    Timo-Kolja Pförtner & Kira Isabel Hower

  3. Institute of General Practice and Family Medicine, University Hospital Bonn, University of Bonn, Bonn, Germany

    Simone Dohle

Authors
  1. Timo-Kolja Pförtner
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Simone Dohle
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Kira Isabel Hower
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

TKP conceived and designed the study with advice from KIH and SD. TKP analysed and interpreted the data and drafted the manuscript. SD and KIH critically revised the draft of the paper. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Timo-Kolja Pförtner.

Ethics declarations

Ethics approval and consent to participate

Ethical approval was not required for the study as the data set is provided by GESIS in anonymized form for scientific use. The data were collected in accordance with the Declaration of Helsinki. Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

No potential conflict of interest was reported by the authors.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table 1.

Panel Study Description. Table 2. Unit-non response rate and total response rate per survey wave. Table 3. Question wording, mean level and internal consistency of items measuring preventive behaviour. Table 4. Question wording, mean level and internal consistency of items measuring risk perception. Table 5. Question wording, mean level and internal consistency of items measuring perceived effectiveness of containing measures. Table 6. Question wording, mean level and internal consistency of items measuring trust in people and institutions. Table 7. Linear regression of risk perception for each survey wave separately. Table 8. Linear regression of risk perception for each survey wave separately. Table 9. Linear regression of perceived effectiveness for each survey wave separately. Table 10. Linear regression of trust for each survey wave separately. Table 11. Random effects model for absolute educational inequalities (SII) in preventive behaviours (Number of observations: 15,902; number of individuals: 4,690). Table 12. Random effects model for absolute educational inequalities (SII) in risk perception (Number of observations: 15,902; number of individuals: 4,690). Table 13. Random effects model for absolute educational inequalities (SII) in perceived effectiveness (Number of observations: 15,902; number of individuals: 4,690). Table 14. Random effects model for absolute educational inequalities (SII) in trust (Number of observations: 15,902; number of individuals: 4,690). Table 15. Random effects model for relative educational inequalities (RII) in preventive behaviours (Number of observations: 15,902; number of individuals: 4,690). Table 16. Random effects model for relative educational inequalities (RII) in risk perception (Number of observations: 15,902; number of individuals: 4,690). Table 17. Random effects model for relative educational inequalities (RII) in perceived effectiveness (Number of observations: 15,902; number of individuals: 4,690). Table 18. Random effects model for relative educational inequalities (RII) in trust (Number of observations: 15,902; number of individuals: 4,690).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Pförtner, TK., Dohle, S. & Hower, K.I. Trends in educational disparities in preventive behaviours, risk perception, perceived effectiveness and trust in the first year of the COVID-19 pandemic in Germany. BMC Public Health 22, 903 (2022). https://doi.org/10.1186/s12889-022-13341-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12889-022-13341-3

Keywords

BMC Public Health

ISSN: 1471-2458

Contact us