Search strategy

A sensitive search strategy was conducted using PubMed, Scopus, and Web of Science. Search terms were: (adherence OR acceptance OR agreement OR trust OR distrust OR compliance OR willing*) OR (perceived coerc* OR percept* coerc* OR pressure OR force OR influence OR control OR threat OR justice) AND (lockdown) AND (COVID OR SARS-CoV-2 OR COVID-19). Search terms such as acceptance, agreement and willingness were included as the presence of these may imply lower perceived coercion whilst the absence is likely to signify the presence of perceived coercion. Other search terms were also tested but excluded because of the limited relevance of the resulting studies. Articles were included if they pertained to COVID-19-related lockdowns (i.e. where individuals were legally mandated to stay at home) and assessed attitudes and behaviours relevant to perceived coercion (i.e. acceptance, agreement, trust, compliance or willingness) or a key component of the main measure of perceived coercion in healthcare settings (the MacArthur Admission Experience Survey (AES), i.e. perceived pressures, coercion, force, influence, control, threat, (in)justice) [1]. Articles were excluded if they did not refer to the COVID-19 pandemic or lockdown, if they did not pertain to community samples of adult participants (≥18 years) or if they pertained to participant groups outside of the remit of our review (e.g. surgery or asthma patients etc). Examples of excluded search terms were: (adherence OR acceptance OR agreement OR trust OR distrust OR compliance OR willing*) OR (perceived coerc* OR percept* coerc* OR pressure OR force OR influence OR control OR threat OR justice) AND (quarantine OR lockdown OR isolation), as many of the search results were non-specific to the COVID-19 pandemic; and (perceived coerc* OR percept* coerc* OR pressure OR force OR influence OR control OR threat OR justice) AND (COVID-19 or coronavirus), as most results were in relation to mental health admissions.

The search was completed between April-May, 2022. All titles, abstracts, and full-text articles were screened by the first author (VR). The other members of the research team (S.E. and S.K.K.) independently screened 12% of all titles and abstracts (n = 5000), and remaining full texts to ensure that these met the inclusion criteria. Discussion regarding the included and excluded articles between the three researchers also took place to ensure that only relevant articles were included in the review.

Data extraction

Extracted details included article authors, country in which the research was performed, year of publication, journal title, article type (e.g., editorial/commentary or research), sample population, study design, and key findings. For a copy of this, please see Table 1. Prevalent similarities or differences found across the literature were grouped into themes. Each theme was categorized by VR and reviewed by all authors. A description of these is presented below.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Publication details of all articles included in the scoping review after full-text screening.

https://doi.org/10.1371/journal.pgph.0001250.t001

Types of literature

The majority of articles originated from European countries (52.5%, n = 21). The remaining articles originated from Asia (20%, n = 8), the Americas (7.5%, n = 3), Australasia/Oceania (10%, n = 4), Africa (5%, n = 2) and the Middle East (2.5%, n = 1). One further study was conducted internationally and included data from 79 countries. Most articles reported novel findings from primary data (80%, n = 32). The remaining articles consisted of five commentaries (13%), one systematic literature review (3%), one letter (3%), and one policy document (3%). Out of all reviewed articles, 70% were quantitative (n = 28), 5% were qualitative (n = 2), and ~3% (n = 1) used mixed methods. Further information on the included articles is outlined in Table 1.

Identified themes

Five themes were identified from the studies: perceived acceptability and willingness to adhere to lockdown; perceived control during lockdown; perceived pressures arising from lockdown; perceived threat of sanction from others and procedural (in)justice of lockdown, as presented below.

Theme 1: Perceived acceptability and willingness to adhere to lockdown

The available studies examining individuals’ willingness to comply with lockdown reported that participants based in high-income countries such as Saudi Arabia and some European countries generally expressed a willingness to restrict their right to freedom of movement for the protection and health of others [10–13]. Such willingness decreased as individuals experienced frustration and anger over their continued restrictive circumstances, yet increased again when rates of COVID-19 and perceived risk of contracting SARS-CoV-2 rose with subsequent surges as seen in one longitudinal study based in South Africa [14]. A further study, undertaken in Germany, revealed that when presented with differing potential scenarios for lockdown, acceptance of restrictive measures was greatest for the strictest short-term lockdown scenario (e.g. only being allowed to leave the home with official consent and severe penalties for violations) and lowest for lengthier though less restrictive lockdown scenarios (e.g. where citizens could leave their home at certain times, with no potential severe sanctions outside those times [15].

As expected, individuals from North America with right-wing political leanings were less willing to comply with lockdown restrictions [16, 17]. Moreover, willingness to follow restrictions, measured by the absence of oppositional attitudes to lockdown and compliance with such restrictions was positively correlated with beliefs that political leaders were competent and that the aim of such restrictions was to consolidate social solidarity, according to a South African study [14]. Though opposition to lockdowns was generally low across studies, some studies suggested higher rates in participants who expressed scepticism regarding the funding received by their governments from international organisations. Those sceptical about taxpayer funded relief initiatives during the pandemic were also more likely to resist lockdown restrictions, as seen in both low and high-income countries such as Sierra Leone and Slovenia [18, 19]. Opposition to lockdown amongst Europeans from high-income countries tended to be associated with higher COVID-19-related stress (e.g. feelings of intolerability, boredom, anger, fear and pessimism), lower perceived risk of infection, less clarity regarding restrictions and conspiracy beliefs [19–21].

Willingness to live with restrictions also differed according to the which aspect of freedom the restriction curtailed. In the sole South African study, approximately half of participants stated that they were willing to concede their right to religious assembly and freedom to travel, whilst a third were willing to suspend the right to attend school or protest [14]. However, even temporary restrictions that impacted individuals’ ability to travel to work, and their privacy, were viewed as much less acceptable [14]. Socioeconomic characteristics were, in part, linked to perceptions regarding acceptability. Indeed, higher income predicted willingness to sacrifice a broad range of rights apart from the right to work across nations [10, 13, 14]. Similarly, French participants who were financially poorer were less in favour of lockdown, viewing it as coercive and disproportionate to the risk posed by the disease [13]. Of note, nonetheless, are the inconsistent findings within the limited available studies in relation to the impact of other demographic factors on perceived acceptability of lockdowns. For instance, in a South African sample, willingness to adhere to lockdown restrictions was lower in White adults and not linked to gender, level of education or age [14]. Additionally, older age and higher education levels were linked to willingness to accept lockdown restrictions in Eastern European participants [21, 22].

Theme 2: Perceived control during lockdown

The included studies on perceived control focused on three main areas: 1) the extent to which individuals perceived themselves or others to be in control of their circumstances and their attitudes towards coercive control during lockdown; 2) the impact of perceived control on their psychological wellbeing; and 3) perceived control as a predictor of adherence to restrictions. In a qualitative study, some individuals spoke of not having control over their day-to-day lives whilst others reported feeling indifferent to, or accepting of, restrictions [19]. Nonetheless, included studies highlighted a change in the extent to which people felt in control over their circumstances as lockdowns continued, with individuals’ initial sense of tolerance for restrictions and personal perceived control decreasing, and a sense of intrusiveness by authorities increasing as lockdown continued, for instance in Spain [23].

The mental health impact of low perceived control was recorded in two studies. One of these noted that low perceived control predicted depressive and anxious symptomatology in participants spanning 79 countries [24]. Furthermore, feelings of ‘entrapment’ arising during lockdown and the negative impact of these on individuals’ mental health were noted in a prior systematic review [25]. One study noted that belief in conspiracy theories may have acted as a form of coping with distress and satisfied a need for greater control [26].

Both greater perceived control and greater internal locus of control, accompanied by fear of contracting COVID-19 or perceived risk of COVID-19, acted as determinants of willingness to adhere to lockdown in some of the included studies undertaken with European, Turkish and Northern American participants [22, 27–30]. One further study concluded that external locus of control was predictive of adherence to lockdown restrictions [20]. Those who did not feel they had the decision-making power to leave their house were less likely to adhere to restrictions [29]. For some of those who lived alone, ‘bending’ the rules by creating unsanctioned bubbles or meeting outside with others during lockdown was done in an effort to counteract isolation [19]. There is some disagreement within the literature as to whether perceived behavioural control predicted adherence to lockdown, with some studies linking it to intent to adhere to restrictions [31, 32] and others to non-compliance [33].

Theme 3: Perceived pressures

Social influences and pressure from friends and family was highlighted as an influential factor in how individuals viewed and responded to lockdown regulations. Those close to family members who held favourable views regarding lockdown were positively influenced to comply with regulations to protect themselves, their families and vulnerable others [34]. Conversely, those whose family members did not adhere to lockdown regulations felt lower perceived pressure to follow such regulations themselves [35]. Pressures to ‘belong’ or conform to a group identity were also indicative of individuals’ attitudes to lockdown, with lower perceptions of normative pressure from friends being predictive of non-compliance in central European countries [29, 34–36]. Individuals who conveyed fears of losing touch with friends and relatives if they followed restrictions tended to oppose to lockdown regulations [37].

Societal norms were also reported to play a role in individuals’ perceptions regarding lockdown in a minority of studies. In two qualitative studies examining attitudes to movement control/stay-at-home orders in Malaysia and Indonesia, participants reported that “collectivistic societal norms” pressured them to comply with restrictions and to feel that respecting such regulations was every citizen’s duty or responsibility [31, 38]. A sense of social responsibility and civic duty was not exclusive to these South-East Asian studies and was also noted in Australia and some European countries [32, 35, 39]. In one study, individuals who linked the spread of COVID-19 to insufficient compliance with restrictive measures by others tended to favour greater pandemic-related social control [40]. Yet, where members of the general public and government officials were seen to not obey those restrictions, for instance in the UK, the public too felt less pressured and inclined to do so [35].

Theme 4: Perceived threat from others

Three studies examined how the general population responds to perceiving a threat from others in relation to lockdowns, with inconsistent findings. Two of these studies, both European, indicated that individuals were less likely to respond to commands to stay home if these were perceived as threatening to their autonomy [41, 42]. Another study from Japan, suggested that individuals who perceive a threat of imprisonment or heavy penalty would be more likely to stay at home due to potentially feeling shame associated with such punishments, in addition to fear of financial risk [43]. The extent to which such experiences of shame are culturally determined, is not clear. Both findings suggest that restrictive strategies and their messaging should ideally be tailored to different people. For instance, in Japan, these could either promote respect for authority or provide clearer information on the risks of COVID-19 to those who place less trust in authorities, as threats and sanctions may counterproductively lead to less compliant outcomes [43].

Theme 5: Procedural Justice of lockdown

The screened literature on procedural justice purports to the ethical justification and fairness of lockdown [44]. According to an Indian policy framework by Zadey, Dharmadhikari & Mukuntharaj (2021), where the extent of harm that a potential pathogen poses is unknown, decision-making and guidelines regarding restrictions of human rights must be clearly communicated, equitable and reciprocal. Such decision-making must uphold the use of least restrictive means and, as more information unfolds, decisions must be guided by the principles of preventing harm, justifiability and proportionality [45]. Other authors in Western cultures focused on such restrictions being justified in light of the risk of negative outcomes to others, particularly in the absence of a vaccine [12, 19, 46]. Nonetheless, examples of discriminative implementation of lockdown and unfair burdens to some of the general population were observed in some studies. For instance, the implementation of lockdown was unequal in India, with authorities adopting stringent measures with the least powerful vulnerable whilst the wealthier were able to conduct and attend marriages and other ceremonies [47]. It also forced those without reliable access to livelihoods, sanitation, transport, and food to stay at home, resulting in deaths that were not related to COVID-19 infection [47].

Summary of findings

This review provides an initial synthesis of studies relating to the concept of perceived coercion in the context of the COVID-19 pandemic. To reiterate, this concept originally derived from mental health practices that were perceived (by patients) as being coercive and consists of three interrelated constructs: perceived coercion per se, perceived control and procedural justice. The reviewed studies suggest that these constructs indeed have relevance to and implications for pandemic-related public health messaging and efforts to promote adherence to restrictive measures (lockdowns). They also highlight differences across geopolitical contexts.

The reviewed studies suggest that participant groups from different countries (with different socioeconomic contexts, cultural norms etc.) were initially accepting of lockdown measures. Acceptance of such measures increased with higher rates of infection and perceived risk of infection. Some of those opposed to lockdown tended to express greater distrust in authorities, held more conspiracy beliefs, viewed the risk of infection as low and the guidance regarding restrictions as unclear. The extent to which individuals felt a sense of control over their circumstances during the pandemic differed across and within studies. Low perceived control was linked to greater depressive and anxious symptomatology [24], and those who felt less in control over their circumstances were less likely to adhere to lockdown [22, 27–30]. Nonetheless, perceived control and tolerance for restrictions lessened over time as a sense of intrusiveness by authorities emerged in one study [23].

In some of the studies, adherence to lockdown was influenced by the views and behaviours of those around an individual. Those with close family members or friends who held favourable views regarding lockdown were influenced to comply with regulations [34], whilst those whose social circles did not adhere to lockdown regulations felt lower perceived pressure to follow such regulations [35]. Messaging too impacted how individuals perceived and responded to lockdown measures. The limited evidence suggests that those who place less trust in authorities may be less likely to respond positively to commanding messages if these are perceived as threatening to their autonomy [41]. Others who are more focused on harm avoidance or who hold greater respect for authority figures may be more likely to stay at home when these perceive a threat that could impose both emotive and financial consequences upon them [43]. However, as the studies pertaining to social norms are few in number, their findings may not be generalisable to other countries. Finally, there is some limited debate regarding the ethicality and fairness of lockdown within the literature, particularly in those with fewer economic means and less reliable access to food, water and sanitation [47]. Some authors argued that decision-making regarding lockdown must adopt the least restrictive means possible until clear information on a pathogen and the risks it poses emerge, whilst others argue that lockdown is justified where there is substantial risk of loss of life [12, 19, 45, 46].

Implications

As suggested within the review’s findings, the adult participants sampled from the general population were more accepting of lockdown where guidance and information regarding risk of illness from a pathogen and resulting restrictions was clear and cohesive and where these were articulated by authorities whom they trusted. Therefore, preparedness for the possibility of future widespread infectious diseases should focus on identifying and incorporating respected members of communities who can clearly convey public health messages and the rationale for restrictive measures. This is important as clear public messaging delivered by trusted and credible figures influences the attitudes both of individuals who receive the messaging and those around them, with a consequent snowballing influence at a community level [42]. Therefore, messaging from respected, trusted and credible community members may be less likely to be experienced as coercive.

The findings also suggest that those who felt less in control over their circumstances experienced greater anxiety, depression and feelings of entrapment. This has important implications for mental health services as an increase in psychological symptoms may result in greater demand on such services. In countries where psychological distress is more stigmatised, this may result in individuals not having a source of support and containment. One potential solution may be to provide a forum for the general public’s voice to feel heard and included when designing public health measures. Another, perhaps more idealistic, option may be to create brief low-intensity psychological intervention referral pathways designed to help individuals with COVID-related anxiety or depression who have less complex psychological presentations (as seen in some Improving Access to Psychological Therapies (IAPT) services in the UK, that provide mental health first aid to healthcare professionals working with patients diagnosed with COVID-19) whilst scaffolding secondary care services [48].

Finally, as highlighted within the included studies, a uniform lockdown can heighten a sense of discrimination among those less privileged and/or historically discriminated against. Under the umbrella of the harm prevention principle, we remain unclear about what level of restriction is justified for what level of risk of harm and, whether the risk of contracting COVID-19, disease burden and cost to individuals is equal for all and proportional to the enforcement of lockdown for all [46]. An assessment of the costs and benefits of lockdown would therefore be warranted to ensure that some individuals are not disproportionately affected by costs and to prevent discrimination [44, 46]. Such an assessment and future policy should aim to provide equitable, rather than equal, support to those at risk of loss of income or access to essential goods.

Strengths and limitations

In line with the aims of a scoping review, which are to provide a broad overview of the current state of knowledge in a rapidly developing field, this review included various types of literature, ranging from empirical papers with both quantitative and qualitative methodologies, policy frameworks, systematic reviews and commentaries that allowed for the broad mapping of a lesser-known area. Most of the empirical articles employed quantitative online survey methodology. This method ensured that researchers could reach the general population during lockdowns. However, the absence of representative sampling and a consistent measure of perceived coercion, pressures and procedural justice within the general population has serious implications for generalisability. In particular, the samples were biased towards high income countries for whom technology was not a barrier. It also meant that there was less space for participants to speak of their experiences and the meaning they attributed to these in their own voice. The inclusion of commentaries, though biased towards the writer’s opinion, also provided some useful philosophical debate on the topic. Missing from this picture is also an account of the grey literature on the topic. Though authors have, more recently, called for the inclusion of a quality assessment in relation to scoping reviews, there is yet to be a comprehensive tool that can uniformly assess a range of methodologies. In the absence of such a tool, we urge caution in interpreting the findings above.