Studying Satisfaction with the Restriction Measures Implemented in Greece during the First COVID-19 Pandemic Wave

There is an ongoing debate regarding the effectiveness of policy responses to the coronavirus disease 2019 (COVID-19) pandemic. As satisfaction with such measures is a function of their effectiveness, the main objective of this study was to identify the factors driving satisfaction with the measures implemented in Greece during the first pandemic wave. The study used data from the first round of the “Public Opinion in the European Union (EU) in Time of Coronavirus Crisis” survey. The sample consisted of 1050 individuals aged between 16 and 54 years. The results of the analysis, which was based on an adjacent category logistic model, indicate that materially deprived individuals were less likely to be satisfied with the restriction measures, while those considering that the health benefits were greater than the economic damage, those concerned for their own health and those concerned for the health of their family and friends were more likely to be satisfied with the restriction measures. The results confirm the existence of uncertainty regarding the effectiveness and impacts of the restriction measures.


Introduction
The containment of virus spread and reduction in public health impacts requires adaptive behavioral and policy responses to the coronavirus disease 2019 (COVID- 19) pandemic [1]. In general, governments around the world have adopted a "mix" of tools in a variety of policy areas, i.e., health, medical, economic and social, rather than a single tool [2]. However, the extent to which such responses are appropriate or not has become a subject of debate [3]. For instance, there is major debate on whether these measures are effective [4]. On the other hand, one debate regarding the appropriate economic policy response to a pandemic such as COVID-19 questions how aggressively economic activity should be restricted in order to limit the virus spread and how quickly these restrictions should be lifted as the epidemiologic picture improves [5]. For example, although quarantine has been characterized as an effective strategy to mitigate the spread of the virus, its cost-effectiveness is questionable, as it results in heavy economic losses [6]. The same is also true for lockdowns and social distancing, which result in productivity loss and decline in demand for goods and services, leading to a collapse in economic activity [7]. In addition, lockdowns and social distancing measures are likely to have negative effects on physical and mental health and health inequalities; such effects have numerous mechanisms, including economic, social and health-related behaviors and disruptions to services and education [8,9].
Thus, according to Appleby (2020) [10], the characteristics of the COVID-19 pandemic suggest that we should not base any decisions about doing something versus doing nothing on the results of inevitably imperfect and premature analysis of the costs and benefits.
In addition, the dichotomy between the economic costs of the COVID-19 crisis and the health impacts is not a valid distinction. As the health-risk losses associated with the pandemic can be monetized to be in the same units as the cost of disruptions to the economy, the COVID-19-related morbidity and mortality rates imply real and considerable economic losses [11]. Furthermore, it should be highlighted that, although the economic recession has been framed as a dipole of the form "the economy versus saving lives from COVID-19", such a dichotomy is also false; the economic recession, through austerity in government expenditure on the social determinants of health, can be translated to a far greater loss in lives and well-being over the long term than COVID-19 can [12].
What is mentioned above becomes even more complicated if one takes into consideration the existence of uncertainty in several aspects of the COVID-19 crisis. That is, in parallel with health-related uncertainties-namely, uncertainties in infectiousness, viral lethality [13], mutations [14] and prevention and treatment [15]-the pandemic has created considerable uncertainty regarding both personal finances [16] and the overall state of the economy [17]. In addition, there is substantial uncertainty regarding the effectiveness [18], economic impacts [19] and the duration of containment measures [20]. Thus, the COVID-19 pandemic affects the economy not only through supply and demand disruption, but also through uncertainty [21].
According to the limited literature regarding satisfaction with policy responses to the COVID-19 pandemic, it is evident that the results of such responses, i.e., the number of COVID-19-related cases and deaths per million population, are those that individuals pay attention to [22]. In addition, satisfaction with both the information dissemination during the pandemic and the delivery of daily necessities and protection materials decreases significantly as the number of confirmed cases increases [23]. Thus, because satisfaction with policy responses is a function of the responses' effectiveness, it is not surprising that opinions regarding the costs and benefits of the restriction measures are also a predictor of satisfaction with such measures [24].
In Greece, the first COVID-19 case was diagnosed on 26 February 2020, and the first death was reported on 12 March 2020. As of 6 June 2021, 408,789 confirmed cases and 12,253 deaths have been reported in Greece [25].
Greece was the country with the earliest adoption of a lockdown in response to the COVID-19 outbreak [26]. The business and workplace suspensions implemented in Greece between 10 March and 18 March 2020, i.e., during the first wave of the pandemic on which this study focused, included the following: However, although one may argue that the timely application of preventive measures and strict compliance to guidelines limited the spread of COVID-19 in Greece during the first wave [28], such measures were expected to negatively affect almost all sectors of economic activity [29].
Based on the previous points, the objective of this study was to investigate the debate regarding the costs and benefits of the restriction measures implemented in Greece during the first pandemic wave, but also to identify the factors driving satisfaction with such measures.

Material and Methods
For the purpose of this study, data from the first round of the "Public Opinion in the EU in Time of Coronavirus Crisis" [30] survey were used. The survey was conducted using Kantar's online access panel between 23 April and 1 May 2020, with among 21,804 respondents in 21 EU Member States. The survey was limited to respondents aged between 16 and 64 years. In some countries, including Greece, the sample was limited to respondents aged between 16 and 54 years. Representativeness at the national level was ensured by quotas concerning gender, age and region. The sample size was n = 1050 in Greece. The data collection took place between 23 April and 27 April 2020.
The respondents were asked the question, "How satisfied or not are you with the measures your government has taken so far against the coronavirus pandemic?" The potential answers were as follows: 1-very satisfied, 2-fairly satisfied, 3-not very satisfied, 4-not at all satisfied, 5-don't know/not applicable. As the response was ordinal, an adjacent category logistic model was used. As potential predictors in the analysis, the following variables were considered: (a) region (1: Attica; 2: Macedonia and Thrace; 3: Epirus and Western Macedonia; 4: Thessaly and Central Greece; 5: Peloponnese, West Greece and Ionian Islands; 6: Aegean Islands and Crete), (b) gender (0: female, 1: male); (c) age, (d) marital status (1: married/living with partner, 2: never married (single), 3: divorced/widowed, 4: living with parents, 5: domestic partner/living with other adults), (e) presence of children (0: no, 1: yes), (f) social class (1: low-/semi-skilled or unskilled manual workers, students, retired and living on state pension only, unemployed (for over six months) or not working due to long-term sickness; 2: middle-skilled manual workers, supervisory or clerical/junior managerial/professional/administrator; 3: high-intermediate managerial/professional/administrative, higher managerial/professional/administrative-this variable was based on the occupation of the main earner of the household), (g) experiencing loss of income since the beginning of the COVID-19 pandemic (0: no, 1: yes), (h) experiencing unemployment or partial unemployment since the beginning of the COVID-19 pandemic (0: no, 1: yes), (i) experiencing difficulties paying rent/bills/bank loans since the beginning of the COVID-19 pandemic (0: no, 1: yes), (j) experiencing use of personal savings sooner than planned since the beginning of the COVID-19 pandemic (0: no, 1: yes), (k) experiencing asking for financial help from family or friends since the beginning of the COVID-19 pandemic (0: no, 1: yes), (l) experiencing difficulties having proper and decent quality meals since the beginning of the COVID-19 pandemic (0: no, 1: yes), (m) experiencing bankruptcy (0: no, 1: yes), (n) experiencing other financial issues (0: no, 1: yes), (o) current employment status (1: employed (full-time), 2: employed (part-time), 3: self-employed, 4: retired/unable to work/disabled, 5: still at school, 6: in full-time higher education, 7: unemployed and seeking work, 8: not working and not seeking work, 9: prefer not to say), (p) COVID-19-related concern for own health (1: very concerned, 2: fairly concerned, 3: not very concerned, 4: not at all concerned, 5: don't know/not applicable), (q) COVID-19-related concern for the health of family and friends (1: very concerned, 2: fairly concerned, 3: not very concerned, 4: not at all concerned, 5: don't know/not applicable), (r) opinion regarding the costs and benefits of restriction measures (1: the health benefits are greater than the economic damage; 2, 3, 4, 5, 6: the economic damage is greater than the health benefits and 7: don't know/not applicable).
When respondents answered "don't know/not applicable" or "prefer not to say" in certain questions, their answer was recoded as missing due to the lack of information on the specific item. More specifically, the response category "don't know/not applicable" was recoded as missing in the following variables: (a) satisfaction with restriction measures, (b) COVID-19-related concern for one's own health, (c) COVID-19-related concern for the health of one's family and friends and (d) opinion regarding the costs and benefits of the restriction measures; the frequencies of missing values were found to be equal to n = 10, n = 10, n = 15 and n = 27, respectively. In addition, the response category "prefer not to say" was recoded as missing in the current employment status variable; the frequency of missing values was found to be equal to n = 21.
Helmert coding was applied to the following ordinal variables: (a) social class, (b) COVID-19-related concern for one's own health, (c) COVID-19-related concern for the health of one's family and friends health and (d) opinion regarding the costs and benefits of the restriction measures. The Helmert contrast measure compares each category of an ordinal variable (except the last) with the mean of the subsequent levels. Indicator coding was applied to the following nominal variables: (a) region, (b) marital status and (c) current employment status. Indicator contrast was used to compare the reference category of a nominal variable with the remaining categories. The binary variables expressing whether each financial issue had been experienced since the beginning of the COVID-19 pandemic, gender and presence of children were treated as such.
The model's goodness of fit was tested using the test of Lipsitz et al. (1996) [31]. In addition, the model was tested for specification error using the link test.
Furthermore, to measure the consensus in both satisfaction with policy responses and public opinion regarding the costs and benefits of such responses, the Cns measure of consensus of Tastle and Wierman (2007) [32] was calculated.

Results
Regarding gender, 50% of the respondents were female and 50% were male. The mean age of the respondents was 37.14 years (±10.47), while 17.05% of the respondents were aged between 16 and 24 years, 22.29% were between 25 and 34 years, 29.90% were between 35 and 44 years and 30.76% were between 45 and 54 years. The respondents' characteristics are presented in Table 1. Regarding the satisfaction with the restriction measures, 30.19% of the respondents were very satisfied, while only 6.06% of the respondents were not at all satisfied (Table 2, Figure 1).
The Cns measure regarding satisfaction with the restriction measures was found to be equal to 0.627, indicating a weak consensus. The percentage of those considering that the health benefits were greater than the economic damage (21.60%) was more than three times higher than the percentage of those considering that the economic damage was greater than the health benefits (6.55%).
In addition, the percentage of category 2 (25.90%) was more than two times higher than the percentage of category 5 (11.93%). Finally, the percentages of categories 3 and 4 slightly differed (17.50% and 16.52%, respectively) (Table 3, Figure 2).  Table 3. Opinion regarding the costs and benefits of the restriction measures.

Category % (n)
The health benefits are greater than the economic damage 21.60 (221) 2 25.90 (265) 3 17.50 (179) 4 16.52 (169) 5 11.93 (122) The economic damage is greater than the health benefits 6.55 (67) The Cns measure regarding the costs and benefits of the restriction measures was found to be equal to 0.528, indicating neither a complete lack of consensus nor a complete consensus.
Furthermore, it is worth noting that only 11.35% of the respondents were very concerned for their own health, while 23.29% of the respondents were very concerned for the health of their family and friends (Table 4, Figures 3 and 4).    According to the adjacent category logistic model, satisfaction with the restriction measures depends on whether the respondents experienced difficulties in having proper and decent quality meals, on whether the respondents experienced other financial issues, on respondents' opinion regarding the costs and benefits of the restriction measures, on the concern for respondents' own health and on the concern for the health of the respondents' family and friends.

Conclusions
As mentioned in the Introduction, a few weeks after the first cases of COVID-19, strict containment measures came into effect in Greece, including closures of schools, universities, non-essential shops, cafes, restaurants and public spaces, as well as movement restrictions, including a ban on gatherings and travel [37].
The descriptive measures of the study, i.e., the higher percentage of respondents that were satisfied with the restriction measures, may be explained by the relatively low number of COVID-19-related cases and deaths during the first pandemic wave. That is, as of 27 April 2020, i.e., the last day of data collection, 2534 confirmed cases and 136 deaths had been reported in Greece [38]. Thus, the results reflect the fact that the perceived severity of the COVID-19-related situation is positively associated with satisfaction with measures [39]. Afterall, according to Fouda et al. (2020) [40], the strict early policies implemented in Greece may be correlated to the relatively low number of positive cases and deceased cases at the early stage of COVID-19 pandemic.
However, although, to a large degree, the respondents declared that they were satisfied with the restriction measures, the consensus measure Cns was equal to 0.627, indicating a weak consensus. In addition, the Cns regarding the costs and benefits of the restriction measures was equal to 0.528, indicating neither a complete lack of consensus nor a complete consensus. Thus, the results of the study confirm the findings in the literature but also the existence of a considerable uncertainty regarding (a) the effectiveness and impacts of the restriction measures [41,42] and (b) the nature of the disease and the dynamics of transmission [43]. In other terms, the findings of the study captured the existence of a public debate about whether or not the restriction measures are appropriate-i.e., a debate regarding the relative benefits of various measures aimed to control the spread of COVID-19 [44], but also a debate regarding the economic consequences of such measures [45]. The debate regarding lockdowns has focused on the potential inefficiency of such a policy, and also on the estimation of a reasonable trade-off between protecting individuals' health and avoiding damage to the economy [46]. The debate regarding restrictions such as physical distancing, community containment measures for reducing public transport use and public gatherings, school closures and working from home where possible concerns the type and magnitude of such measures that could be appropriate for controlling the COVID-19 outbreak. The optimal balance between the potential positive effects of such restrictions on public health and the adverse effects on freedom of movement, the economy and society is unclear [47]; however, it should be noted that the COVID-19 pandemic constitutes an economic crisis [48], which has yielded adverse effects for the global economy [49]. In addition, the lack of complete consensus in satisfaction with measures may also be explained by the pandemic management at the level of daily information, since it is evident that news regarding the virus spread increases uncertainty about the health risk and the economy [50]. That is, the high degree of multidimensional uncertainty makes it difficult to interpret COVID-19-related news and official recommendations [51].
On the other hand, according to the adjacent category logistic model, materially deprived individuals were less likely to be satisfied with the restriction measures, while those considering that the health benefits were greater than the economic damage, those concerned for their own health and those concerned for the health of family and friends were more likely to be satisfied with the restriction measures. The influence of the abovementioned variables, and the respective sign of their association with satisfaction with measures, reflects what is evident: (a) that perceived economic risks are negatively associated with acceptance of the measures, (b) that people who consider that the economic and social aspects of the prevention measures are not sufficiently taken into consideration are expected to report lower acceptance of such measures and (c) that perceived health risks are positively associated with acceptance of the measures [52]. These results indicate the existence of the public debate highlighted above.
From an economic perspective, due to the COVID-19 pandemic, Greece is facing a new economic crisis [53] that has caused severe negative effects on the demand and supply of goods and services, employment and disposable income [54,55].
From an epidemiologic perspective, as of mid-September 2020, Greece was not defined as a country with a trend of concern according to the European Centre for Disease Prevention and Control (ECDC) criteria, but had observed a strong increasing trend in intensive care unit (ICU) admissions. However, the ECDC model also highlighted Greece as having the potential for a large resurgence [56].
To summarize, the results of this study confirmed the existence of considerable uncertainty regarding the effectiveness and impact of the restriction measures [57]. On the one hand, the Greek economy has been severely affected by the COVID-19 pandemic [58], but on the other hand, the number of COVID-19-related cases and deaths was relatively low as of late April [59]. Thus, the existence of a public debate regarding the measures implemented is not surprising.

Conflicts of Interest:
The author declares no conflict of interest.