Public trust in the Government to control the spread of COVID-19 in England after the first wave—a longitudinal analysis

Abstract Background To control the spread of coronavirus disease 2019 (COVID-19), governments are increasingly relying on the public to voluntarily manage risk. Effectiveness is likely to rely in part on how much the public trusts the Government’s response. We examined the English public’s trust in the Conservative Government to control the spread of COVID-19 after the initial ‘crisis’ period. Methods We analyzed eight rounds of a longitudinal survey of 1899 smartphone users aged 18–79 in England between October 2020 and December 2021. We fitted a random-effects logit model to identify personal characteristics and opinions associated with trust in the Conservative Government to control the spread of COVID-19. Results Trust was lowest in January 2021 (28%) and highest in March 2021 (44%). Being older, having lower educational attainment and aligning with the Conservative Party were predictors of higher levels of trust. Conversely, being less deprived, reporting that Government communications were not clear and considering that the measures taken by the Government went too far or not far enough were predictors of being less likely to report a great deal or a fair amount of trust in the Government to control the pandemic. Conclusion Trust in the Government’s response was found to be low throughout the study. Our findings suggest that there may be scope to avoid losing trust by aligning Government actions more closely with scientific advice and public opinion, and through clearer public health messaging. However, it remains unclear whether and how higher trust in the Government’s response would increase compliance with Government advice.


S
ince emerging in December 2019, the spread of SARS-CoV-2, resulting in the coronavirus disease 2019 (COVID-19) pandemic, has posed enormous challenges to public health specialists and governments around the globe.To slow the rate of transmission and avoid overwhelming health systems, many governments asked citizens to take unprecedented actions to control the spread of SARS-CoV-2 ranging from stay-at-home orders to mandatory mask wearing and vaccination.
The effectiveness of interventions largely relied on citizens to cooperate and follow public health guidance, even when mandatory since in most countries state capacity and willingness to put the population under surveillance is limited.Evidence from previous infectious disease outbreaks (SARS, H1N1, Ebola) suggests that individuals with higher trust in Government and/or Government institutions in general are more likely to adhere to public health measures. 1,2][5] Following the initial COVID-19 wave in spring 2020, many national governments sought to gradually ease restrictions.This marked a shift from regulations to guidance, with the public increasingly being asked to take greater personal responsibility to manage risk.There is considerable uncertainty about the direction the global pandemic will take.As such, governments need to be ready to respond to any resurgences and new variants, which may require the tightening of public health measures.In general, they will be relying on individuals to voluntarily follow advice to reduce their and other community members' risks.
In crises, levels of trust in Government often follow a typical pattern: initially increasing ('crisis effect' or 'rally-round-the-flag effect'), and then declining as the consequences of the crisis become clearer. 3,6The same phenomenon was observed during the first wave of the COVID-19 pandemic. 3,7,8In the context of shifting to increasing reliance on voluntary action as countries move beyond the initial crisis phase, efforts to sustain adherence to public health measures are likely to become increasingly reliant on the public's trust in Government and its institutions.
Using eight rounds of longitudinal survey data, this study examined changes in the English general public's trust in the Government specifically to control the spread of COVID-19 beyond the initial 'crisis' period.It sought to identify individual characteristics and opinions associated with higher levels of trust and to describe how changes in case numbers and Government policies affected the public's level of trust.

Study design
We undertook a panel study with eight rounds of survey data collected between October 2020 and December 2021, as part of a wider study of public attitudes towards, and use of, the National Health Service (NHS) COVID-19 contact tracing smart phone application.Ethics approval for this study was obtained from the London School of Hygiene & Tropical Medicine (Reference Numbers 22483 and 25819).

Setting
Figure 1 provides an overview of the Government's response to COVID-19 in England over the eight survey rounds.Most lockdown restrictions had been lifted by July 2020.However, by autumn 2020, restrictions once again began to tighten: on 14 September, people were legally prohibited from meeting more than six people socially, and on 14 October, the Government introduced a new three-tier system of restrictions. 9On 5 November 2020, England entered the second national lockdown, returning to the tier system at the beginning of December.On 19 December 2020, a fourth tier was introduced, before a third national lockdown was introduced on 6 January 2021 in response to the rapid spread of the Alpha variant.From March 2021, as vaccine rollout progressed, restrictions were lifted based on a four-step 'roadmap out of lockdown'. 10Step four of the roadmap was delayed by 4 weeks to allow more people to receive their first dose of the vaccine.By July 2021, most legal limits on social contacts were lifted and all sectors of the economy were reopened; England opened earlier than many comparable countries.The 'Autumn and Winter' plan for COVID-19 was released in September 2021, setting out a Plan A and Plan B for managing the virus. 11In response to the Omicron variant, England moved to Plan B on 8 December 2021, which included mandatory face masks in most indoor public spaces and a legal requirement to check in to some venues.

Participants
A representative sample of English smartphone users aged 18-79 was recruited from YouGov's volunteer online panel, with quotas set on age, gender, region and socioeconomic status.All respondents were invited to participate in each subsequent round of the survey.Participation was entirely voluntary, and participants' consent was obtained before completing the baseline survey.

Data collection
The online surveys were conducted roughly every 6-8 weeks.The survey questionnaire consisted of questions developed by the research team and adapted from previous COVID-19 tracker surveys. 12,13The survey was primarily designed to capture participants' views on and use of NHS England's COVID-19 app.It additionally included a module on participants' perceptions of the Government's response to, and handling of, the COVID-19 pandemic.Demographic information at baseline was provided by YouGov.

Statistical methods
The main analysis was restricted to participants who responded to all eight rounds of the survey because the balanced dataset gave a higher log likelihood coefficient than the unbalanced dataset.Our outcome of interest was the level of trust-based on the survey question 'To what extent, if at all, do you trust the Government to control the spread of COVID-19?' (options 'a great deal', 'a fair amount', 'not very much', 'not at all' and 'don't know').Explanatory variables were selected to capture participants' sociodemographic characteristics, and attitudes towards the pandemic and Government measures (see Supplementary table S1 for variables included in the model and their definitions).
The descriptive analysis used weighting within survey rounds (see Supplementary figure S1) so that the achieved sample was representative of the population of smartphone users in England using data from the 2019 Ofcom Technology Tracker Survey. 14e fitted a random-effects logit model to examine factors associated with trust (see Model specification in Supplementary file).The outcome of interest, trust, was binary, measured as 1 if a respondent reported 'a great deal' or 'a fair amount' of trust in the Government to control the spread of COVID-19 or 0 if a respondent reported 'not very much' or 'not at all'.Our categorization of the dependent variable was based on the assumption that at a given point during the pandemic, respondents either had trust or no trust in the Government to control the spread of COVID-19.To ensure that there was no loss of information results from our categorization of the dependent variable, we also investigated the data using randomeffects ordered logistic regression; however, the results were similar (see Supplementary tables S5 and S6).Respondents who reported 'don't know' (n ¼ 369 observations) and were missing data on education, index of multiple deprivation or political party affiliation (n ¼ 36 individuals) were excluded from the multivariate model.All variables associated with trust in the univariate analysis were included in the model (see Supplementary table S3).We estimated the multivariate model recursively, dropping each variable in turns to assess the performance of the model using the Likelihood-Ratio test, to arrive at the final model (see Supplementary table S4).

Results
In total, 1899 smartphone users aged 18-79 in England completed the baseline survey.Of these, 873 (46%) completed all eight survey rounds (see Supplementary table S2).Baseline characteristics are presented in table 1.Compared with the general population, there was overrepresentation of people with higher education (57% vs. 44%) and younger age groups in the sample. 15,16Since younger and more highly educated people are more likely to vote for parties other than the Conservative Party, Conservative voters are likely to be underrepresented in the sample. 17

Changes in trust in the Government to control the spread of COVID-19
Across all time points, most participants reported having no trust or not very much trust in the Government specifically to control the spread of COVID-19 (see figure 2).Trust in the Government was highest at survey round 5 (15 to 31 March 2021), when case numbers were lowest and restrictions started to be eased [44.0%(95% confidence interval [CI] 40.1-47.9)a great deal or fair amount vs. 54.1% (95% CI 50.2-58.0)not very much or not at all] and lowest at survey round 3 (28 December 2020 to 6 January 2021) when case numbers were highest just before the third national lockdown began (27.7% [95% CI 24.4-31.2]a great deal or fair amount vs. 70.0%[95% CI 65.3-72.5]not very much or not at all).The percentage of respondents reporting 'don't know' ranged from 1.9% (95% CI 1.1-3.4) at survey round 5 to 4.2% (95% CI 2.9-6.1) at round 8.

Factors associated with higher trust in the Government to control the spread of COVID-19
The results of the random-effects model are shown in table 2. Individuals were more likely to report a great deal or a fair amount of trust in the Government specifically to control COVID-19 at survey rounds 4, 5 and 7 compared with round 1. Older age groups [adjusted odds ratio (aOR) 3.75 (95% CI 1.82-7.71),P 0.001], participants with the lowest educational attainment [aOR 3.18 (95% CI 2.02-4.99),P 0.001], and those who reported that they generally supported the Conservative Party [aOR 9.17 (95% CI 5.6-15.12),P 0.001] were more likely to report having a great deal or a fair amount of trust.Conversely, individuals who were relatively less deprived [aOR 0.51 (95% CI 0.8-0.92),P ¼ 0.02] and those who found the Government's communications about what they should do in response to COVID-19 less clear were less likely to report a great deal or a fair amount of trust.Compared with those who thought that the measures taken by the Government to tackle COVID-19 were about right, both those that thought measures went too far [aOR 0.21 (95% CI 0.14-0.31),P 0.001] and those who thought they did not go far enough [aOR 0.12 (95% CI 0.09-0.16),P 0.001] were less likely to report trusting the Government to control COVID-19.

Discussion
This study examined the public's trust in the then Government to control the spread of COVID-19 in England after the first pandemic wave.Trust in the Conservative Government was found to be low throughout the period, with most respondents reporting they had no or not very much trust in the Government.Evidence from the first wave of the pandemic had shown a 'rally-round-the-flag' effect.However, by the time of the first survey round in autumn 2020, this had disappeared. 8,18A contributing factor was the revelation that in May 2020 Boris Johnson's then chief-advisor, Dominic Cummings, had broken lockdown rules by travelling to Durham with his family who had COVID-19 symptoms. 19However, while the levels of trust in the Government specifically to control the spread of COVID-19 during the period of this study never reached the highs reported during the first wave of the pandemic, our results suggest that there may be scope for governments to avoid losing trust. 8,13Notable changes in trust were observed following survey rounds 3, 5 and 6.
The significant increase in trust observed after round 3 coincided with the rollout of the first vaccine and the introduction of the third national lockdown.The Government was widely perceived to have handled the rollout of the vaccine well and the UK's vaccine programme was considered highly successful; it was the first country to start its mass vaccination programme and by round 4 of the survey, in February 2021, had administered more doses than any comparable country. 20Before the third national lockdown, introduced in January 2021, there had been rapidly rising infection rates and increasing divergence between Government actions and scientific advice. 21pposition parties had also been calling for the Government to take further actions during autumn 2021 and widely condemned the decision on 19 December not to go ahead with the planned relaxation of rules for Christmas as having been made too late. 22he decision was reported in the mainstream media as a 'major U-turn' and accompanied by negative headlines that 'Christmas had been cancelled'. 23These disparities between Government action and the scientific advice, and criticism from opposition parties, plus the negative portrayal of the Government's handling in the media, likely contributed to the perception that the Government was not to be trusted to control the spread of COVID-19. 24It might also have served to damage opinion of the clarity of the Government's communications on what to do in response to COVID-19, which we found to be a key determinant of trust.In contrast, the introduction of the third national lockdown, brought the Government's actions  more in line with public opinion; polling in January 2021 found that most of the public were supportive of the national lockdown (85%) and considered that it had come too late (77%). 25Our findings demonstrate that those who did not agree with the measures taken by the Government (either that they went too far or not far enough) had lower trust in the Government to control the spread of COVID-19.Survey round 5 in spring 2021 showed the highest level of trust during our study period.It took place in the context of low case numbers and ahead of the Government's four-step plan to lift the remaining restrictions. 10The decline in trust observed from survey round 6 potentially reflects dissatisfaction that the Government had pressed ahead with lifting all legal restrictions on 19 July despite rising case numbers as the Alpha and Delta variants took hold.By the last survey round, in winter 2021, the first allegations that parties took place at Downing Street and other Government buildings during the 2020 lockdowns ('Partygate') had emerged, which likely further corroded the public's confidence and trust in the Government's response, adding to the process of eroding trust which started with the Cummings affair in May 2020. 19,26Simultaneously, the Omicron variant started to spread and the Government reintroduced some restrictions as part of its Autumn and Winter Plan B. The mismatch between the Government's earlier rhetoric that vaccines 'will get us out of the pandemic' and the rising case numbers, hospitalizations and deaths, potentially reduced trust as perceptions of performance did not meet people's expectations. 27articipants who were younger, less deprived and more educated reported lower levels of trust in the Conservative Government to control the spread of COVID-19.The impact of age may have gained importance as a determinant of trust, given that the COVID-19 pandemic has been shown to have had a considerable impact on younger people and a negative impact on their confidence in political institutions and leaders. 28,29re-pandemic public trust in governmental institutions in general was found to increase with income-level and education. 30One potential reason why income and education were negatively associated with trust in the Government's response in this study is political party affiliation.We found that those participants who affiliated themselves with the Conservative Party were more than nine times more likely to report higher levels of trust in the Government's handling of the pandemic.Therefore, individuals who might more generally be expected to have had higher trust in Government, such as wealthier more highly educated individuals, may be influenced conversely when the Government is led by a party they do not support, in this case, Boris Johnson's Conservative Government.The interesting finding in this context is that people who affiliated themselves with the Conservative Party-who tend to oppose state regulations and restrictions more than other parties-trusted the Conservative Government in handling the pandemic with lockdowns and other restricting measures.In the extraordinary context of the pandemic, partisan trust seemed to overcome libertarian views.Political partisanship is part of a wider trend in the UK, which has become more evident since the Brexit referendum. 31,32It has also been observed elsewhere, with approval ratings for governments' responses to COVID-19 higher among those citizens who support the governing parties. 33

Strengths and limitations
This analysis took advantage of a longitudinal study conducted between October 2020 and December 2021, after the first pandemic wave in England.This design allowed us to examine changes in trust in the Government specifically to control the spread of COVID-19 over time and to capture the dynamic of the pandemic and the Government's response to changing situations.
The survey was conducted online, using quota-sampling to recruit a representative sample of smartphone users from YouGov's panel.Participants were younger and more highly educated than the general public and therefore less likely to vote conservative.This might have led to an overestimation of the level of distrust.Further, the study was conducted in English, thereby excluding a sub-group of the population with poor English skills.The question on the extent of trust in the Government to control the spread of COVID-19 could have been interpreted differently by the respondents, and responses may reflect conceptually distinct forms of trust such as general trust in politics, trust in national political leadership or COVID-19specific trust, though the question was worded in an attempt to focus the response more narrowly on the public's trust in the Government's control of the pandemic.Thus, comparisons between  studies asking questions about different aspects of trust in Government need to be drawn carefully.

Implications
5]34 Our findings that less than half of respondents trusted the Government to control the pandemic, potentially poses a significant barrier to the success of the UK's response to the current and future pandemics, if it undermines the public's adherence to Government recommendations.Evidence on the impact of trust on compliance during the current pandemic is complex because it appears to depend, at least in countries like the US and the UK, on people's party political affiliations and how these relate to the Government of the day. 35Thus, where citizens are more concerned about action to control the spread of the virus than a laissez-faire Government and believe that the Government is not taking sufficient action, citizens might even increase their adherence beyond the recommended measures, irrespective of their level of trust in the Government's response. 36n England during the period of the current study, it appears that to increase the public's trust in the Government's response, there needed to be greater alignment between the Government's actions, scientific advice and public opinion.The Johnson Government potentially overestimated the public's appetite to 'return to normal'.Given the higher levels of trust in scientists than the Government among the UK public, 37 greater transparency in how evidence and advice are being used might also help foster trust in the Government's ability to control the spread of COVID-19 and future pandemics. 38Tied to this is a need for clearer and more consistent communication of the Government's response, to overcome what has been described as 'alert fatigue' whereby the public struggle to understand constantly changing rules. 39

Figure 1
Figure1Overview of number of daily new cases and policy measures implemented in England over the course of the study.Round 1: 14 to 22 October 2020, took place after the introduction of the local three-tier system.Round 2: 12 to 23 November 2020, conducted during the second national lockdown, which was introduced on 5 November.Round 3: 28 December 2020 to 6 January 2021, conducted over the Christmas period when the country had returned to the tier system.The south east of England moved to newly created tier 4 ('stay at home').The first vaccine was administered in the UK on 8 December 2020.Round 4: 1 to 15 February 2021, took place during the third national lockdown, which had started on 6 January 2021.Round 5: 15 to 31 March 2021, took place at the start of the easing restrictions.The Government outlined a four-step plan, at the time of the survey schools had reopened.Round 6: 1 to 18 July 2021, took place ahead of the final step of the easing of restrictions; indoor venues had reopened, people were allowed to meet in groups of up to 30 outdoors and 6 indoors.Round 7: 31 August to 13 September 2021, took place after all legal limits on social contact had been lifted.Those who had been double vaccinated were also no longer required to self-isolate if they had come into contact with someone who tested positive for COVID-19, if they didn't have any symptoms.Round 8: 25 November to 13 December 2021, took place at the same time as the Omicron variant was first being reported; six Southern African countries were placed on the travel red list and the first cases were detected in the UK.On 8 December the Government announced that Plan B were to be reintroduced; face coverings mandatory in most indoor settings, work from home, use of the NHS COVID pass for entry into nightclubs and settings where large crowds gather.Number of daily new cases obtained from the website: https://coronavirus.data.gov.uk/details/casesTrust in the Government to control the spread of COVID-19

Figure 2
Figure 2 The extent of trust in the Government to control the spread of COVID-19, by survey round

Table 1
Baseline distribution of respondent's characteristics in the sample (survey round 1), N ¼ 873