Beyond What Meets the Eye: Unveiling Dynamics of Compliance with Preventive Measures in the COVID-19 Era

Beyond What Meets the Eye: Unveiling Dynamics of Compliance with Preventive Measures in the COVID-19 Era | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Beyond What Meets the Eye: Unveiling Dynamics of Compliance with Preventive Measures in the COVID-19 Era Sahar Ramazan Ali, Eric Lacourse, Mathieu Pelletier-Dumas, Jean-Marc Lina, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4391822/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 Mar, 2026 Read the published version in BMC Public Health → Version 1 posted 4 You are reading this latest preprint version Abstract Background Previous longitudinal studies have identified variability in compliance with COVID-19 preventive measures, noting the heightened sensitivity of the least compliant groups to situational factors like easing restrictions. However, they overlooked other forms of variability inherent in compliance behaviour. Hence, we investigated compliance with social distancing and staying-at-home measures, and its dynamic nature, along with its association with social and individual factors. Methods Data from a longitudinal study involving 3617 Canadians across twelve measurement times from April 2020 to April 2022 were analysed. Compliance levels were assessed through self-reported items, alongside social and individual factors like trust in science and government, perception of social norms, and health literacy. Results Joint-Trajectory analysis revealed three compliance trajectories during mandatory and lifting measures periods: "Low and constant" (shifting to "Low and decreasing" during lifting measures), "High and decreasing," and "High and constant." Transition movements between trajectories mostly showed stability (between 85.83% and 90.44% for social distancing; between 80.68% and 87.71% for staying-at-home), with health literacy consistently predicting trajectory membership (social distancing: mandatory measures period- Odds = 1.67; 95% CI [1.20, 2.32], lifting measures period- Odds = 1.43; 95% CI [0.93, 2.37]; staying-at-home: mandatory measures period- Odds = 1.75; CI [1.24, 2.56], lifting measures period- Odds = 1.90; 95% CI [0.26, 1.01]). Perception of provincial norms (social distancing: comparison 1- Odds = 4.10; 95% CI [1.30, 17.54], comparison 2- Odds = 0.38; 95% CI [0.12, 1.04]; staying-at-home: comparison 1- Odds = 2.22; 95% CI [1.13, 4.51], comparison 2- Odds = 0.74; 95% CI [0.55, 1.00]) and trust in science (social distancing: Odds = 3.57; 95% CI [1.29, 10.91], comparison 2- Odds = 3.00; 95% CI [1.26, 7.96]; staying-at-home: comparison 3- Odds = 0.77; 95% CI [0.58, 1.02]) moderately predicted trajectory change. Conclusions Our study not only reaffirms variability within the lowest compliance group but also unveils variability among higher compliance groups, notably in trajectory transition movements. Health literacy consistently emerged as a strong indicator of trajectory membership, while perception of provincial norms and trust in science moderately influenced trajectory change. COVID-19 compliance preventive public health measure factors latent trajectories Figures Figure 1 Figure 2 Introduction Research related to COVID-19 saw an exponential growth following the World Health Organisation’s announcement on the severity of the virus outbreak, upgrading its classification to a pandemic. Researchers from diverse fields were mandated by government agencies to answer society’s pressing questions in response to the COVID-19 pandemic, whether they were related to the virus’ transmission modes or to the communities’ adaptive capacities. One of the more prominent themes in the novel COVID-19 literature is compliance to preventive measures, namely hand washing, mask wearing, social distancing, and staying at home. Early on, during the first wave of the pandemic, studies reported that a small minority (between 5% and 12%) did not comply with governmental guidelines, furthering the need to explore the reasons behind this transgressive behaviour [ 73 ][ 47 ][ 51 ]. These studies evaluated compliance levels to preventive measures as a stable phenomenon without considering transition movements between levels of compliance as the pandemic situation evolved. Furthermore, several social and sociodemographic factors were identified as probable reasons behind non-compliance to preventive measures, but these factors were only examined in regard to their relationship with compliance levels and not change in compliance levels ([ 54 ]; [ 5 ]; [ 64 ]; [ 3 ]). Hence, this study represents a continuation of a longitudinal investigation initiated by Courdi et al. (2023) with a representative Canadian sample (N = 3617). The primary aim of the earlier study was to identify trajectories of compliance with social distancing, contacts limitation, and mask wearing throughout the first year of the pandemic and its relation with factors. Here, we aim to identify and describe individuals’ compliance behaviours with only social distancing and staying-at-home measures, whether they are stable or varying in terms of compliance patterns and trajectory change, and determine their association with factors during the first wave of the pandemic. The global situation during the early months of the COVID-19 pandemic was critical, with the virus spreading rapidly across all countries, disregarding geographical boundaries. Within two months of reporting the first case, the World Health Organization documented approximately 20,000 cases in March 2020 ([ 70 ]). As a result, governments worldwide implemented preventive measures to mitigate the spread of the virus, as vaccines were still in the development phase. Essential preventive measures were classified into two categories: individual and community-based measures ([ 49 ]). Individual measures encompassed contact limitation (staying-at-home and social distancing), respiratory and hygiene etiquette, indoor ventilation, and surface disinfection ([ 22 ].; [ 49 ]). Meanwhile, community-based measures comprised the closure of non-essential businesses and restrictions on the number of people in a setting ([ 22 ]; [ 49 ]). Given that individual measures relied on voluntary compliance from the population, it became more pertinent to delve into these compliance behaviours rather than focusing on forced directives, such as business closures [ 53 ]. However, the effectiveness of each individual preventive measure was not equivalent but hinged on its ability to limit the spread of the virus [ 22 ]. For instance, limiting contacts by staying at home was regarded as the most effective measure as it prevented individuals from coming into physical contact with others, thereby significantly reducing the risk of transmission [ 22 ]. If limiting contacts by staying-at-home was impossible, it was recommended to implement physical barriers, such as practising social distancing and wearing masks. Notably, social distancing took precedence as a physical barrier measure over mask wearing, because it was specifically designed to restrict interactions between individuals, thus also decreasing the likelihood of infection [ 22 ]. Hence, studies suggested focusing on staying-at-home and social distancing as measures of interest when studying compliance during the pandemic, as they significantly influenced transmission rates [ 32 ][ 61 ]. Longitudinal studies on compliance behaviour with social distancing and staying-at-home measures revealed notable variations in compliance levels during the COVID-19 pandemic. Indeed, [ 73 ] conducted a study in the United Kingdom with 50,000 participants during the first year of the pandemic and found four distinct longitudinal patterns of compliance: Class 1 (32.8%) with high and constant compliance, Class 2 (28.66%) with high and fluctuating compliance, Class 3 (23.98%) with medium high and fluctuating compliance, Class 4 (14.56%) with low and decreasing compliance. Other longitudinal studies examining trajectory compliance have consistently identified a similar number of compliance groups and reported a decline in compliance, particularly evident in the lowest compliance group during the initial months of the pandemic, prompting further investigation [ 33 ] [ 13 ] [ 69 ]. The decrease in compliance with preventive measures, specifically to social distancing and staying-at-home measures, was first observed at the end of the first wave of the pandemic (Spring 2020) as highlighted in previous literature [ 73 ] [ 33 ] [ 69 ]. In the Canadian pandemic context, this period spanned from March 2020 to July 2020 [ 7 ] [ 57 ]. During that first wave, Canadians experienced tremendous changes, such as being required to comply with governmental recommendations under the threat of sanctions [ 66 ]. As the COVID-19 situation improved with fewer reported cases, provincial governments gradually lifted mandatory preventive measures, permitting individuals from different households to gather at home, for instance [ 66 ] [ 7 ]. Hence, two noticeable periods emerged during the first wave of the pandemic in Canada: mandatory measures period and lifting measures period. With slight variations among the provinces, the mandatory measures period commenced in March 2020, and the lifting measures period began toward the end of May 2020 [ 66 ] [ 7 ] [ 57 ]. Whereas the previously mentioned longitudinal studies successfully identified variations in compliance levels during that first pandemic wave [ 73 ] [ 33 ] [ 13 ] [ 69 ], they were unable to explore the effect of the easing of measures on membership in compliance trajectories. Therefore, two questions arise in relation to this gap: (1) Are there different trajectories of compliance with regards to social distancing and staying-at-home measures during both periods, and (2) did individuals persist in similar compliance trajectories during the first wave of the pandemic? To examine the presence of different trajectories and shifts in trajectory membership between the mandatory and lifting measures periods, it is recommended to conduct a Joint-Trajectory Analysis [ 40 ] [ 41 ] [ 23 ]. An extension of the Latent Class Growth Analysis (LCGA), the Joint-Trajectory not only identifies distinguishable patterns of response over time but also enables the estimation of probabilities of co-occurrence or continuity in two distinct yet related behaviours [ 40 ] [ 41 ] [ 12 ]. It has been widely employed in developmental research to investigate how trajectories of a behaviour at an early age can serve as precursors for maladaptive behaviour later in life [ 27 ] [ 12 ] [ 67 ]. For example, in Côté et al. (2007), the researchers aimed to examine the relation between trajectories of physical aggression at an early age and trajectories of indirect aggression later. They reported that most children exhibiting a low trajectory of physical aggression were consistently paired with a low trajectory of indirect aggression later on, indicating stability in aggression levels. On the other hand, among children initially showing higher levels of physical aggression, a shift toward a trajectory of high indirect aggression was observed. A comparable study design in the context of compliance to preventive measures has not been identified in the literature, leaving a gap in our understanding. Such a study could aid governments in targeting behaviours susceptible to change as situations evolve and identifying associated risk factors. During the COVID-19 pandemic, researchers sought to identify factors of compliance with preventive measures, aiming to offer valuable insights for government agencies in refining their intervention strategies. Numerous social (e.g., social norms, trust, risk communication) and individual characteristics (e.g., health literacy and perceived self-efficacy) were identified as consistent factors of compliance with preventive measures like social distancing and staying-at-home [ 54 ] [ 5 ] [ 64 ] [ 3 ]. However, these factors were examined in the context of adopting a compliant behaviour rather than sustaining or reinforcing it as the situation evolves. This distinction is significant, as different processes and interventions come into play for both behaviours, as highlighted in the Kwasnicka et al. (2016) review. Hence, outlined below are the factors anticipated not only to forecast trajectory membership but also to predict changes in trajectories between mandatory and lifting measures periods. Between January 1st and June 30th, 2020, a total of 23,634 scientific articles on the topic of COVID-19 were documented globally [ 65 ]. Indeed, media outlets and the scientific community generated new information at an unprecedented rate, prompting the World Health Organization to label this phenomenon as an “infodemic” [ 71 ]. The term "infodemic" is specifically employed in the context of a disease outbreak, referring to an excessive amount of information that can be harmful if misleading [ 71 ]. This excess of information can contribute to the adoption of risk-taking behaviours that directly impact health and may exacerbate the course of an outbreak [ 71 ] [ 46 ]. Therefore, the ability to comprehend, utilise, and critically assess new health information, which translates to the three dimensions of health literacy, becomes crucial in the face of an information overload [ 62 ]. In Canada, the lack of health literacy is a well-documented concern for public health officials [ 6 ]. Reports indicate that nine million people have limited literacy skills [ 63 ], with over half of Canadians reading below a high school level [ 63 ]. These statistics are alarming considering the prejudicial impacts of low health literacy on health outcomes, like higher mortality rates and more hospitalisations [ 62 ]. On the other hand, high health literacy promotes better awareness of health issues and their consequences, facilitating the adoption of adaptive health behaviours, such as preventive measures during the pandemic [ 46 ]. A team of researchers from France investigated the impact of health literacy on compliance with social distancing measures and discovered a positive association, indicating that higher health literacy was linked to increased compliance with social distancing [ 54 ]. Other studies also confirmed the pivotal role of health literacy in fostering the adoption of preventive behaviours in the context of the COVID-19 pandemic [ 21 ] [ 15 ]. In addition to facilitating the adoption of new health behaviours, health literacy also acts as a facilitator when individuals need to better adapt to evolving situations, like the gradual easing of governmental restrictions [ 46 ]. As Canadians transitioned to the lifting measures period, individuals with high health literacy were better adjusted to the changing environment, hence expected to more likely maintain high compliance levels or transition to higher compliance responses due to their critical assessment of the still-present virus's dangerousness. As the pandemic unfolded, practices once considered normal in social settings, such as shaking hands, were strongly discouraged and replaced by behaviours like social distancing and staying-at-home as much as possible to limit the spread of the virus. Therefore, adopting these preventive behaviours to protect oneself and others became the prevailing social norms during the pandemic, establishing a set of expectations for the new normalcy [ 42 ]. These expectations were based on others ’ behaviours (descriptive norms) or others ’ attitudes and beliefs (injunctive norms), others being the ingroup. In the existing literature, there is a consensus among researchers that descriptive norms are more influential in prompting changes in health behaviours compared to injunctive norms due to the high conformity pressure associated with them [ 17 ] [ 64 ] [ 76 ]. A comparable conclusion was reached in a COVID-19 compliance study conducted in Spain [ 5 ]. The authors noted that the perception of descriptive norms was the most important factor of compliance with social distancing measures, followed by trust in science, perceived effectiveness of measures, and finally, perceived risk of infection. Furthermore, a high perception of descriptive norms is not only more likely to induce the desired behavioural change but also to foster sustainable changes in behaviour, possibly due to the norm internalisation process that is immune to situational factors like the lifting of preventive measures [ 20 ]. Thus, individuals who initially reported a strong perception of descriptive norms should be expected not only to comply strongly with preventive measures, as indicated in Cabrera-Álvarez et al. (2022), but also to continue exhibiting high levels of compliance as the pandemic situation evolves. Finally, when referring to descriptive norms, the others become crucial in navigating new expectations [ 3 ], since they represent individuals from the same group. In the Canadian political landscape, the others could be Canadians in general or the residents of the same province, two distinct referent-points [ 42 ]. To our knowledge, no study has made this distinction in regard to compliance with preventive measures, which would help governments target relevant national and provincial interventions. During the pandemic, some individuals were confined in small living arrangements, while others were unable to work remotely. In both instances, situational factors posed obstacles to compliance with preventive measures, specifically social distancing and staying-at-home, which can hinder perception of control and ultimately perceived self-efficacy. Perceived self-efficacy reflects the belief in one's capability to modify health behaviours by taking personal actions [ 59 ]. It consistently emerges as a significant individual factor in various health behaviour theories, such as the Health Belief Model [ 56 ], Reasoned Action approach [ 16 ], and Protection Motivation Theory [ 55 ]. In the Health Belief Model, for example, the likelihood of adopting a preventive health behaviour is dependent on having greater confidence in one’s ability to apply that health behaviour, which amounts to higher perceived self-efficacy [ 56 ]. Additionally, individuals with a high sense of self-efficacy tend to invest more effort and demonstrate greater persistence compared to those with low self-efficacy [ 59 ]. In the face of obstacles, like living in small arrangements and the impossibility to work remotely, the former rebound more quickly and maintain health behaviours. On the other hand, studies report that a low sense of self-efficacy can have damaging effects beyond decreasing compliance to health behaviours, like higher risk of depression, anxiety and helplessness [ 38 ]. In the context of the COVID-19 pandemic, this individual factor was studied in relation to compliance with preventive measures such as social distancing and staying-at-home. In a longitudinal study conducted in China [ 65 ], the importance of perceived self-efficacy and perceived barriers was highlighted as crucial factors in adopting and sustaining compliance with physical distancing measures over a span of three months, even in the presence of stressors such as the lifting of preventive measures. Given these results, individuals with high perceived self-efficacy are expected not only to initially exhibit high compliance behaviour but also to sustain or transit to these high levels of compliance, even during the transition between mandatory and lifting measures periods. During the initial months of the virus outbreak, many individuals questioned the government's ability to effectively manage the pandemic situation, while others challenged the scientific community to provide reliable information. The lack of trust at both levels, government and science, has been widely documented as a strong factor of defiant health behaviour [ 60 ] [ 4 ] [ 28 ]. Trust in government is akin to supporting and being confident about public directives [ 43 ]. To preserve a functional democracy during an economic or social crisis, a minimum level of trust in government is necessary, as it would be easier to implement directives in the population if they believed in the competence of public institutions to deal with the crisis [ 75 ]. Trust in science involves being confident in the results produced by the scientific community and their ability to elaborate effective social and economic policies [ 51 ]. Indeed, when new issues or problems emerge, individuals, lacking the resources to understand these complex problems, must come to trust the scientific community to make the most appropriate decisions [ 51 ]. The literature indicates that compliance behaviour with preventive measures was predicted by both trust in government and trust in science; however, the influence of trust in science outweighed that of trust in government [ 3 ]. Indeed, individuals with higher trust in science were found to be more open to acquiring new knowledge related to COVID, facilitating their adaptation to novel information, and ultimately aiding in navigating an evolving situation such as the easing of measures [ 2 ]. In light of previous findings, Canadians who expressed high levels of trust in both government and science are expected to exhibit greater compliance with preventive measures. However, it is anticipated that only those individuals with a high level of trust in science will demonstrate better adaptation in maintaining or transitioning to a high compliance behaviour amid changing situations. At the onset of the health crisis, Canadian provincial governments implemented region-specific public policies, particularly concerning preventive measures. This decentralised approach offers advantages, such as increased flexibility and innovation. However, solely relying on this governance model does not ensure satisfactory outcomes, as evidenced by varying death and case numbers across provinces during the pandemic [ 67 ]. According to a report from the Organisation for Economic Co-operation and Development [ 45 ], the effectiveness of crisis management lies in fostering coordination between different government levels and clearly defining their respective roles. Such coordination can enhance coherent communication between provincial and federal authorities and result in clearer dissemination of government messages at all levels [ 45 ]. Effective communication, grounded in principles of coherence and clarity during a social crisis, is more likely to motivate individuals to support public policies, including preventive measures [ 44 ]. As validated by Pelletier-Dumas et al. (2020), a heightened perception of clarity and coherence in the government's communication strategy was linked to increased compliance with preventive measures. Given the dynamic nature of the pandemic, clear and coherent communication became increasingly crucial as Canadians navigated the lifting measures period, marked by the uncertainty and novelty of the situation. Hence, individuals perceiving coherent and clear messages from both governance levels (federal and provincial) will be more likely to comply with preventive measures during the mandatory and lifting measures periods and transition or maintain high compliance behaviour during the transition period. In the literature, some studies considered the effect of socio-economic characteristics such as age, gender and level of education. It has been shown that older people are more likely to comply with preventive measures [ 73 ]. Women [ 18 ] and university-level graduates [ 14 ] seem to follow the same trend, i.e. a higher level of compliance with preventive measures. The current research aims to identify patterns of compliance with preventive measures, like social distancing and staying-at-home, during the mandatory measures and lifting measures periods in Canada. This project also investigates the dynamic nature of trajectory change as Canadians transitioned between periods. Finally, we examine the factors’ association with trajectory and trajectory change to further our understanding of the dynamic nature of compliance during a sanitary crisis. Hypothesis The first objective attempts to identify trajectories of compliance to preventive measures during the first wave, which can be divided into 2 distinct periods. The first period is associated with the mandatory imposition of measures (April 2020 - May 2020), while the second period represents the lifting of measures (May 2020 - July 2020). In response to this objective, we suppose that different trajectories of compliance will be identified during both periods (March 2020 - May 2020/ May 2020 - July 2020). Also, based on previous studies [ 74 ] [ 13 ] [ 33 ] [ 70 ], we suggest that the majority of Canadians will be grouped into trajectories of high levels of compliance during both periods of the first wave. The second objective relates to describing the transitions between both periods of the first wave in terms of group membership in compliance. It entails quantifying the percentage of Canadians transitioning from one group of compliance to another between periods (April 2020 - May 2020/ May 2020 - July 2020). We expect to find some Canadians transitioning from one group of compliance to another between periods. We also expect that the majority of Canadians will preserve similar levels of compliance between both periods. The last objective aims to identify factors of compliance at baseline in April 2020 and during the transition period, which refers to the transition between the first and second periods of the first wave. At baseline, we expect that there will be differences in compliance to preventive measures due to variables like perception of social norms, level of health literacy, perceived self-efficacy, level of institutional and scientific trust. Trajectories of high compliance at baseline will be associated with strong perception of social norms, high levels of health literacy, social support, perceived self-efficacy and trust in institutions and in science compared to the modal trajectory of compliance. During the transition period, Canadians who initially recorded strong perceptions of social norms, high levels of health literacy, perceived self-efficacy and trust in science will preserve high and constant levels of compliance to preventive measures. Canadians who perceive low levels of social norms, present low levels of health literacy and low social support, higher perceived self-efficacy, and more trust in science will be more at risk of fluctuating between patterns of compliance during the transition period. Institutional trust will not be associated with transition movements between trajectories of compliance, as its impact on compliance to preventive measures is not consistent in time [ 3 ]. Methods Sample This project is part of a larger initiative titled "COVID-19 Canada: The end of the world as we know it?". The underlying longitudinal survey, central to this project, was executed in partnership with the polling firm Delvinia, utilising the AskingCanadians survey panel, which comprises an extensive database of over one million Canadians. This longitudinal survey encompassed twelve time points spanning a duration of 2 years, from April 2020 to April 2022. However, for the purposes of this study, we focus exclusively on the initial six time points, extending from April 2020 to July 2020. Time was encoded by weeks to accommodate the varying intervals between each time point (see Supplementary Materials). In terms of sampling, the project “COVID-19: the end of the world as we know it?” accounted for 3617 participants in the first measurement time. This sample is non-probabilistic, employing the weighted quota method for participant selection. Upon further analysis of the representativeness of the sample, the researchers found that the sample was representative in terms of household composition, employment rate, immigration status. However, when the sample is compared with other socio-demographic characteristics, it is less representative for certain groups, such as Francophones, Canadians with lower levels of education and First Nations [ 15 ]. Attrition was observed from the second wave of data collection onward, with an average attrition rate of 43.35% (standard deviation = +/- 6.12%) for measurement time points two to twelve, ranging from a minimum of 34.50% to a maximum of 53.77%. For young people, the attrition rate seems to have been greater over time. To limit the effect of differential attrition according to certain individual characteristics, we incorporated weights based on demographic characteristics (e.g. age, gender, province of residence, household income, etc.) in the modelling of trajectories. Finally, in this project, we used full information maximum likelihood estimation in modelling trajectories of compliance with preventive measures [ 8 ] [ 9 ]. The Latent Class Growth Analysis (LCGA) analysis typically necessitates a minimum of two measurement time points for each period (mandatory and lifting measures periods). Therefore, participants who did not provide a minimum of two responses during each period were excluded. This resulted in a final sample size of 1984 participants. A preregistration of this study can be found here: https://doi.org/10.17605/OSF.IO/R276S . An initial analysis was conducted using the provided computation of variables, but collinearity issues were identified, necessitating additional attention. As a result, the format in which factors are presented in the subsequent sections reflects the final, resolved format. For more details, see supplementary materials. Compliance to preventive measures Participants self-reported their compliance levels to social distancing and staying-at-home measures from April 2020 to July 2020 at six measurement times. They answered the following statements on a scale of 1 to 10 (never to always): "Currently, how often do you do the following? (1) Maintain a distance of at least two metres (about two arm's lengths) from others when I am not at home; (2) stay home as much as I can. Since this study is the continuation of the Courdi et al. study (2023), we used the same items to assess compliance behaviour with social distancing and staying-at-home. For the multinomial logistic regression analysis, the trajectories were transformed as binary outcomes (0 = stayed in a low trajectory or shifted to a lower trajectory; 1 = remained in a high trajectory or shifted to a higher trajectory). Factors of trajectories and joint trajectories Health literacy This variable refers to the ability of extracting and utilising information in a critical manner in a medical context [ 65 ]. We evaluated this ability in the context of the COVID-19 pandemic by using true and false statements from the survey’s first measurement time: “We would like to assess what Canadians know about COVID-19. Mark whether the following statements are true or false. (1) COVID-19 is a hoax. (2) I know what “maintaining a social distance” means. (3) Avoiding gatherings with large numbers of people helps prevent the spread of COVID-19. (4) Only the elderly can be seriously affected by COVID-19. (5) Healthy people should wear facemasks in public to avoid infection. (6) Staying home if you feel sick is important to prevent the spread of COVID-19. (7) COVID-19 is no more dangerous than the seasonal flu.”. Participants received one point if they provided the correct answer to the statement; otherwise, they were assigned zero points. We summed the points to evaluate the participants’ level of health literacy. Due to the application of multiple imputation for this set of questions, not all participants were able to respond to every true and false statement. As a result, participants' health literacy levels were evaluated using a proportion (number of correct answers/number of statements answered), where a higher proportion score indicated a better health literacy level. The calculation of our Health Literacy variable was derived from the methodology employed in the study by Montagni et al. (2021). In their study, the authors aggregated participants' scores to determine the final level of health literacy. Social norms We interrogated Canadians on their perception of descriptive federal social norms by asking the following question at the first measurement time: “Most Canadians are following governments’ measures recommendations concerning COVID-19.”. We also evaluated their perception of descriptive provincial social norms with this statement at the first measurement time: “In general, [Your province citizens] are complying with the governmental measures.” Both items are on a scale of 1–10, 1 being completely disagree and 10 completely agreeing. Several COVID-19 studies have employed single items to assess perception of descriptive norms [ 5 ] [ 34 ] [ 61 ]. Trust To evaluate Canadians’ levels of trust towards the scientific community and the government during the COVID-19 pandemic, we used the following question from the survey at the first measurement time: “ How much do you trust each of the following actors to address the COVID-19 crisis?” (1) Canadian hospitals and health services (2) Prime Minister [Name] [Pipe in the name of the Prime Minister]”. The first item refers to trust in science, while the second relates to trust in government. Participants answered on a scale of 1–10, 1 distrusting completely and 10 trusting completely. Numerous studies used a single item to evaluate trust in science and in government [ 5 ] [ 61 ]. Self-efficiency This variable refers to the perception of self-efficiency in protecting oneself and loved ones against the COVID-19 virus. It was evaluated with a single item at the first measurement time: “I do not know what is the best strategy to fight COVID-19.”. The item is on a scale of 1–10, 1 being completely disagree and 10 being completely agree. We recoded the variable to establish a positive gradient, wherein a higher score reflects a greater sense of self-efficacy. A similar single item was used in another study [ 36 ], where the authors studied the influence of Health Belief factors on compliance levels. In their study, they evaluated the participants’ perception of self-efficacy in regards to the specific behaviour of social distancing. However, in our study, we asked the participants a more general statement since we focused on more than one behaviour of compliance to preventive measures. Clarity Canadians’ clear understanding of measures was assessed with a single item at the first measurement time with a 1 to 10 scale (1 = Strongly disagree, 10 = Strongly agree): “ In general, I have a clear understanding of the various measures established by Canada’s public health agency.”. A similar item was already used in a previous article to evaluate perception of clarity in the message about measures [ 61 ]. Coherence The participants' perception of self-reported coherence towards government messages was assessed with two items at the first measurement time with a 1 to 10 scale (1 = Strongly disagree, 10 = Strongly agree): “ I am confused about the different recommendations coming from the federal and provincial governments and public health agencies. (2) The measures established by Canadian and provincial public health agencies are similar.”. We averaged both items to obtain a final coherence variable. In line with our study, other research endeavours have also employed a self-constructed coherence variable [ 50 ]. For determining factors’ association to trajectory and joint-trajectory membership, factors, assessed initially as continuous variables, were coded as binary variables. Indeed, responses falling within the first lowest quartile represented a low score, while responses outside the first quartile were considered as indicating a high score. For instance, Canadians who obtained a result within the first lower quartile that fell or was equivalent to 0.08 were considered to have a low level of health literacy, while others a high level of health literacy. The cutoffs varied depending on the variable, as some factors displayed less or more variation on the distribution. For the analyses, participants were assigned a value of 1 if they self-reported a high level of factors and 0 if they reported a low level. A similar approach was observed in the Kil and al. study (2023) to determine the high and low cutoffs. For more details on the threshold for every variable, see supplementary materials. Analyses We conducted Latent Class Growth Analysis (LCGA), or commonly referred as Group-based trajectory analysis, to discern longitudinal patterns of compliance behaviour to preventive measures, hence regrouping participants with similar trajectories together [ 43 ] [ 26 ]. Since compliance to preventive measures was assessed as a continuous variable and displayed a normal distribution, a censored normal distribution (CNORM) was employed [ 27 ]. In contrast to many other longitudinal methods, LCGA does not necessitate the assumption of homogeneity of variance between time points to be upheld. As suggested by Nagin (1999), the Bayesian Information Criterion (BIC) serves as our initial guide to determine the most suitable number of trajectory groups. Once we establish the number of trajectory groups for the final model, we proceed to choose the polynomial function for each trajectory—whether constant, linear, quadratic, or cubic. The model with the best fit, as indicated by the BIC, is retained. During model selection, we also consider the participant distribution across trajectory groups, for it is crucial that the trajectories exhibit visually distinguishable patterns. The determination of the polynomial function for each trajectory is further refined based on a significance level ( p-value < 0.05). After identifying longitudinal patterns of compliance to preventive measures (social distancing and staying at home) during each period (mandatory and lifting periods), we sought to identify transition movements between trajectories by conducting a Joint-Trajectory analysis, an extension of LCGA [ 44 ] [ 26 ]. Similarly to the LCGA, the Joint-Trajectory procedure allows for modelization of trajectories over time. However, the latter distinguishes itself by linking trajectories of distinct but related behaviours. There are two approaches to Joint-trajectory analysis: comorbidity and heterotypic continuity. The first method allows for the estimation of probabilities for simultaneous membership in two distinct but related longitudinal patterns of behaviours, while the second method focuses on predicting the probabilities of the occurrence of trajectories for two behaviours expressed at different periods. In this study, we solely focused on the second approach, as it enabled us to identify transitional movements for trajectories of compliance to preventive measures between the mandatory and lifting measures periods. Finally, we used the RISK function from the PROC TRAJ package in the SAS software to study factors’ association to trajectory membership for both preventive measures during the mandatory and lifting measures periods, while controlling for age, gender and level of education [ 43 ] [ 26 ]. The modal trajectory was chosen as the reference group, enabling us to discern high and low compliance behaviours from the majority. We also evaluated the factors’ association to transition movements while performing a multinomial logistic regression analysis. The factors were entered as binary variables, and their reference level was low. Given the exploratory nature of this study and limited samples for trajectory movements, we incorporated a lower threshold ( p-value < 0.10) to identify trends for potential exploration in future research for this section of the study. All visualisations were generated using the ggplot2 package in R. Results Compliance trajectories to preventive measures In observance of our first objective, we aimed to discern longitudinal patterns of compliance with preventive measures, specifically social distancing and staying at home. Upon conducting group-based trajectory analysis, we determined that three trajectories most accurately characterised our sample's compliance to both social distancing and staying at home during both periods (mandatory measures and lifting measures). This determination was based on the evaluation of BIC values and the significance of the parameters (see Supplementary Materials). Figure 1 depicts the trajectories of the best-fitting models for the social distancing measure during both periods. For the mandatory measures period, we observed three trajectories characterised by constant, linear, and linear parameters (BIC=-7501.67), representing: a Low and constant trajectory (n = 116, 6.5%); a High and decreasing trajectory (n = 1060, 53.3%); and a High and constant trajectory (n = 722, 40.2%). Similarly, for the lifting measures period, three distinct trajectories were identified with linear, linear, and constant parameters (BIC= -8414.28): a Low and decreasing trajectory (n = 172, 9.3%); a High and decreasing trajectory (n = 1090, 54.4%); and a High and constant trajectory (n = 722, 36.3%). Figure 2 illustrates the trajectories of the best-fitting models for the staying-at-home measure during both periods. For the mandatory measures period, we observed three trajectories characterised by constant, linear, and constant parameters (BIC= -7594.57), representing: a Low and stable trajectory (n = 167, 8.8%); a High and decreasing trajectory (n = 1026, 51.8%); and a High and constant trajectory (n = 791, 39.4%). Similarly, for the lifting measures period, three distinct trajectories were identified with linear, linear, and constant parameters (BIC=-9182.35): a Low and decreasing trajectory (n = 221, 11.7%); a High and decreasing trajectory (n = 1061, 53.2%); and a High and constant trajectory (n = 702, 35.1%). Probabilities of heterotypic continuity Given membership to a trajectory during the mandatory measures period, we aim to estimate the probabilities of belonging to a trajectory during the lifting measures period, contingent upon the initial membership. Table 1 displays the conditional probabilities of Canadians' compliance levels to social distancing during the lifting measures period based on their initial membership during the mandatory measures period. Generally, individuals exhibit stability from one period to another, as indicated by high probabilities along the diagonal. Specifically, 89.28% (n = 106) of the sample remained in the lowest compliance trajectories for both periods; 90.44% (n = 970) remained in the high and decreasing trajectories, and 85.85% (n = 696) remained in the high and constant trajectories. The off-diagonal elements indicate that a small percentage of the population transitioned between periods. For example, 10.72% (n = 10) of individuals initially in the "Low and stable" trajectory during the mandatory measures period shifted to the "High and decreasing" trajectory in the lifting measures period. In contrast, 3.38% (n = 26) of Canadians transitioned from a "High and decreasing" to a "High and constant" trajectory. A substantial proportion of Canadians (13.70%, n = 110) experienced a decline in compliance between periods, moving from a "High and constant" to a "High and decreasing" trajectory. Notably, only two individuals exhibited a substantial decrease in compliance levels, transitioning from a "High and constant" to a "Low and decreasing" trajectory. No transition movements were identified from the "Low and stable" to the "High and constant" trajectories. Table 2 also presents the conditional probabilities of Canadians’ compliance levels to staying-at-home during the lifting measures period based on their initial membership during the mandatory measures period. Similarly to social distancing probabilities, Canadians remain stable for the staying-at-home measure during both periods as indicated by the probabilities along the diagonal: 80.68% (n = 134) of the population remained in the lowest compliance trajectory; 90.44% (n = 970) in the “High and decreasing” trajectory; 85.83% (n = 696) in the “High and constant” trajectory. In comparison, few individuals exhibited transition movements between trajectories: 19.32% (n = 10) of Canadians previously in the “Low and stable” trajectory are now in the “High and decreasing” trajectory; 8.90% (n = 87) of Canadians transitioned from “High and decreasing” to “Low and decreasing”; 3.39% (n = 29) shifted from “High and decreasing to “High and constant”; finally a portion of Canadians (15.28%, n = 118) displayed a decrease in compliance levels by shifting from “High and constant” to “ High and decreasing”. Interestingly, we did not find any transition movements between “Low and stable”, “Low and decreasing” and “High and constant”. Table 1 Percentage (and Number) of individuals in each trajectory for social distancing measure, by period Lifting measures period Mandatory measures period Low and decreasing High and decreasing High and constant Low and constant 89.28 (106) 10.72 (10) .000 (0) High and decreasing 6.18 (64) 90.44 (970) 3.38 (26) High and constant .472 (2) 13.70 (110) 85.83 (696) Table 2 Percentage (and Number) of individuals in each trajectory for staying-at-home measure, by period Lifting measures period Mandatory measures period Low and decreasing High and decreasing High and constant Low and constant 80.68 (134) 19.32 (33) .000 (0) High and decreasing 8.90 (87) 87.71 (910) 3.39 (29) High and constant .000 (0) 15.28 (118) 84.71 (673) Predicting trajectory membership We then used the RISK function from PROC TRAJ to investigate the predictive value of social and individual characteristics in predicting trajectory membership. This analysis included controls for age, gender, and level of education, and independent variables were coded as high and low (reference group). Tables 3 and 4 present the odds of factors associated with membership in trajectories for social distancing and staying-at-home measures. Social distancing As shown in Table 3 , Canadians exhibited greater likelihood of being in the "High and constant" social distancing group compared to the "High and decreasing" social distancing group when possessing high health literacy ( Odds = 1.67; 95% CI [1.20, 2.32]; p = .005) during the mandatory measures period. Likewise, they were more likely to have higher levels of perceptions of provincial social norms ( Odds = 1.86; 95% CI [1.44, 2.39]; p < .001), trust in science ( Odds = 1.86; 95% CI [1.44, 2.39]; p < .001) and clarity ( Odds = 2.34; 95% CI [1.06, 5.17]; p < .001). The “Low and constant” group, in comparison to the “High and decreasing” group, was associated with lesser likelihood of high perception of provincial social norms ( Odds = .27; 95% CI [.10, .70] ; p = .027) and trust in science ( Odds = .19; 95% CI [.09, .42]; p < .001). During the lifting period, as outlined in Table 3 , Canadians who registered high levels of health literacy ( Odds = 1.43; 95% CI [.93, 2.37]; p = .048), perceived self-efficacy ( Odds = 1.51; 95% CI [.89, 2.35]; p = .020), perceptions of provincial social norms ( Odds = 1.80; 95% CI [1.00, 3.67]; p = .001), trust in science ( Odds = 1.54;95% CI [1.08, 3.69]; p = .014) and clarity ( Odds = 2.16; 95% CI [1.58, 4.92]; p < .001) were more likely associated with the “High and constant” social distancing trajectory when compared to the modal reference group. In comparison to the “High and decreasing group, individuals with high health literacy levels ( Odds = .33; 95% CI [.15, .54]; p < .001), high trust in science ( Odds = .51; 95% CI [.32, .91]; p = .040) and high trust in government ( Odds = .45; 95% CI [.35, 1.02]; p = .007) are less likely to be found in the “Low and decreasing” trajectory. Staying-at-home As depicted in Table 4 , individuals who indicated elevated levels of health literacy ( Odds = 1.75; CI [1.24, 2.56]; p = .031), trust in science ( Odds = 1.67; 95% CI [1.13, 2.36]; p = .006), clarity ( Odds = 2.23; 95% CI [1.53, 4.46]; p < .001), and coherence ( Odds = 1.70; 95% CI [.22, .97]; p = .008) were more inclined to be associated with the "High and constant" staying-at-home trajectory compared to the "High and decreasing" group during the mandatory measures period. In contrast, Canadians found in the “Low and constant” group were less likely to possess high levels of health literacy ( Odds = 0.51; 95% CI [0.21, 1.01]; p = 0.031), trust in government ( Odds = 0.39; 95% CI [0.26, 0.94]; p = 0.003). During the lifting period, as presented by Table 4 , belonging to the "High and constant" group, as opposed to the "High and decreasing" group, is associated with high levels of health literacy ( Odds = 1.90; 95% CI [0.26, 1.01]; p = 0.008), perception of provincial social norms ( Odds = 1.63; 95% CI [1.06, 3.82]; p = 0.020), trust in science ( Odds = 1.80; 95% CI [1.11, 2.57]; p = 0.002) and clarity ( Odds = 2.05; 95% CI [1.41, 4.59] ; p = 0.003 ). On the other hand, Canadians who were associated with the “Low and decreasing” group when compared to the “High and decreasing” group were less likely to record high levels of health literacy ( Odds = 0.46; 95% CI [0.26, 1.01] ; p = 0.008 ) and trust in government ( Odds = 0.38; 95% CI [0.35, 0.69] ; p = 0.004 ). Sociodemographic variables Age, gender and level of education were reported as inconsistent factors of membership to trajectories of social distancing and staying-at-home, as displayed in Tables 3 and 4 . Table 3. Factors of trajectory membership for social distancing during the mandatory and lifting measures periods Table 4. Factors of trajectory membership for staying-at-home during the mandatory and lifting measures periods Predicting trajectory change We employed multinomial logistic regression to investigate the predictive role of social and individual characteristics in determining trajectory change membership. Three contrasts were deemed feasible for the analysis due to the rare occurrence of trajectory change: “Low and constant” to “Low and decreasing” (0) vs “Low and decreasing” to “High and decreasing” (1); “High and decreasing” to “Low and constant” (0) vs “High and decreasing” to “High and constant” (1); and “High and constant” to “High and decreasing” (0) vs remained in “High and constant” (1). Tables 5 and 6 illustrate the results of the multinomial logistic regression in relation to trajectory changes. Social distancing As shown in Table 5 , only a limited number of factors appear to forecast trajectory changes across all contrasts. For instance, Canadians who exhibited high levels of perception of provincial social norms ( Odds = 4.10; 95% CI [1.30, 17.54] ; p = 0.027 ) and trust in science ( Odds = 3.57; 95% CI [1.29, 10.91] ; p =0.013 ) were more likely to transition from “Low and constant” to “High and decreasing” trajectories during the lifting measures period, unlike those who reported high clarity of messages ( Odds = 0.23; 95% CI [0.04, 0.87] ; p= 0.056 ). On the other hand, Canadians were more prone to transition from the "High and decreasing" to the "High and constant" pattern during the lifting period, as opposed to the transition from "High and decreasing" to "Low and decreasing," if they indicated higher levels of trust in science ( Odds =3.00; 95% CI [1.26, 7.96] ; p= 0.014) and perception of federal social norms ( Odds = 2.48; 95% CI [0.98, 6.94]; p= 0.061 ). However, greater perception of social norms reduces the likelihood of transitioning from the "High and decreasing" to the "High and constant" pattern during the transition period ( Odds = 0.38; 95% CI [0.12, 1.04] ; p = 0.063). Staying-at-home Table 6 reveals that shifting from a “Low and constant” to a “High and decreasing” trajectory is more likely when individuals report high trust in government ( Odds = 2.08; 95% CI [1.10, 4.11] ; p = 0.028). Furthermore, an increase in compliance (“High and decreasing” to “High and constant”) as opposed to a decline in compliance ("High and decreasing" to "High and constant") is more likely when reporting high perception of social norms ( Odds = 2.22; 95% CI [1.13, 4.51]; p =0.022). Finally, maintaining membership in "High and constant" trajectories during both periods is more probable when possessing high perception of self-efficacy ( Odds = 1.30; 95% CI [0.98, 1.74]; p =0.070), as opposed to those reporting high perception of provincial social norms ( Odds = 0.74; 95% CI [0.55, 1.00]; p =0,051) and trust in science ( Odds = 0.77; 95% CI [0.58, 1.02]; p =0,071). Sociodemographic variables Transitioning from a "High and decreasing" trajectory to a "High and constant" trajectory was significantly associated with being over 65 years of age ( Odds = 4.00; 95% CI [0.55, 1.00]; p = 0.051) and having a university-level education ( Odds = 2.58; 95% CI [1.18, 5.94]; p = 0.022). Table 5. Factors of trajectory change for social distancing measure during the first wave Table 6. Factors of trajectory change for staying-at-home measure during the first wave Discussion This study aimed to identify and understand compliance trajectories with preventive measures during the mandatory and lifting measures periods of the COVID-19 pandemic, along with their factors. In both periods, three distinct compliance trajectories were discerned for social distancing and staying-at-home measures. Notably, a large proportion of Canadians exhibited high compliance, with 93.5% for social distancing and 91.2% for staying-at-home during mandatory measures, and 90.7% for social distancing and 88.3% for staying-at-home during lifting measures. The majority of Canadians maintained their compliance patterns across trajectories during both periods, with 89.28% staying in the lowest trajectory for social distancing and 80.68% for staying-at-home. Health literacy, perception of provincial social norms, trust in science, trust in government, and perception of clarity consistently emerged as significant factors for trajectory membership in both preventive measures. Perceived self-efficacy exclusively predicted membership in social distancing trajectories, while inconsistent predictions were observed for perception of coherence and federal social norms. In terms of trajectory changes, perception of provincial social norms and trust in science frequently emerged as significant factors of increasing or maintaining high compliance behavior. Overall, the study provides valuable insights into the dynamics of compliance with preventive measures, emphasising the role of various factors in shaping individual trajectories and changes in behaviour. In line with our first objective, we posited that distinct compliance patterns would emerge for both social distancing and staying-at-home measures during the mandatory and lifting periods, with the expectation that a significant majority of Canadians would report high levels of compliance. These initial hypotheses were validated, as our findings unveiled three distinct trajectories during each period. During the mandatory measures period, the lowest compliance trajectory for both social distancing and staying-at-home measures, labelled as Low and constant , exhibited stability with an initially lower compliance level. As the measures were lifted, this lowest compliance trajectory underwent a transformation, adopting a declining linear trend. This shift in trend implies that situational factors, such as the easing of measures, might negatively impact the motivation to comply for individuals already at risk of non-compliance. In contrast to findings from other longitudinal studies [ 15 ] [ 36 ], where the lowest compliance trajectory exhibited a decline from the onset of the pandemic, our trajectory modelling reveals a distinct pattern—a decline in compliance commencing as the measures were lifted for this group. This differentiation is crucial, as it enhances the precision of our understanding of longitudinal compliance patterns by scrutinising specific periods more susceptible to influencing trajectory trends. During both mandatory and lifting measures periods, we observed a group, called High and decreasing , that reported a decrease in compliance with an initial high level of compliance with preventive measures. Interestingly, this predominant group, encompassing the majority of Canadians, appeared unaffected by the imposition or lifting of measures; they simply adopted the necessary behaviours and gradually exhibited a waning interest in maintaining these practices. A similar characterization was noted in previous LCGA studies [ 77 ] [ 36 ] [ 15 ] [ 73 ]. Existing literature has proposed the concept of pandemic fatigue as a potential explanation for this decline in compliance levels, positing that sustaining beneficial yet demanding behaviours, such as social distancing and staying-at-home, becomes progressively challenging compared to initially adopting them [ 51 ]. Given the unconventional nature of the novel preventive measures, it seems justifiable that a single wave of a pandemic may not suffice for complete adoption and perpetuation, as integrating a new behaviour into routine requires time. This aligns with observations in other health-related behaviours, like dietary regimens, fitness programs, or medication plans, where the motivation to uphold costly and unnatural behaviours tends to diminish over time [ 39 ]. Lastly, the High and constant group exhibited constant patterns of high compliance during both the mandatory and lifting measures periods, seemingly impervious to the changing situation. Overall, our findings align with previous Canadian longitudinal studies on compliance with preventive measures. However, we propose that the initial decline observed in these studies may not solely be attributed to a changing situation, as hypothesised, but also to psychological fatigue, depending on the group. While executing Latent Class Growth Analysis, we observed variations in the proportion of Canadians in each trajectory between the mandatory and lifting measures periods. For example, in regard to the social distancing measure, we identified that 40.2% of individuals were in the High and constant trajectory during the mandatory period, but this number decreased to 36.3% during the lifting period. Likewise, in relation to the staying-at-home measure, we noted a decrease in membership for the highest compliance trajectory ( High and constant ), declining from 39.4% during the mandatory period to 35.1% during the lifting period. Given these variations in membership proportions, our second objective sought to describe the transition movements between both periods of the first pandemic wave in terms of group compliance membership. We found that the majority of Canadians preserved a similar pattern of compliance during both periods, as suggested by the probabilities along the diagonals (between 85.83% − 90.44% for social distancing and between 80.68% − 87.71% for staying-at-home). In addition to confirming our initial hypotheses, these findings highlight the importance of the initial compliance level in predicting future compliance patterns, especially in the dynamic context of an evolving pandemic. Indeed, to our knowledge, this is the first study to examine stability within a group of compliance, emphasising the need for governments to implement their interventions in the initial weeks of a health crisis to improve compliance with measures in the long term. Furthermore, this noticeable stability in compliance membership between periods informs us that a change in behaviour is not as affected by situational factors like the easing of preventive measures, but is rather influenced by already present inter-individual differences. Furthermore, we noticed a general trend where more Canadians transitioned to a lower compliance trajectory rather than high, as the measures were lifted. More specifically, we noted that 13.70% (n = 110) of Canadians transitioned from a High and constant to a High and decreasing social distancing trajectory, and a similar transition was observed for staying-at-home, accounting for 15.28% (n = 118). This movement, characterised by a decrease in compliance, represented the most substantial transition between trajectories in our study in terms of number, followed by the transition from High and decreasing to Low and constant. Therefore, our results suggest that transition movements are more likely to be found within the higher compliance groups, such as High and constant and High and decreasing . In prior LCGA studies examining compliance with preventive measures [ 77 ] [ 36 ] [ 15 ], the researchers documented fluctuating compliance behaviour during the COVID-19 pandemic across nearly all trajectories, except for the trajectory associated with the highest levels of compliance. This group, identified as Class 1 in Wright et al. (2022), High and constant in Courdi et al. (2023), and High adherers in MacNeil et al. (2022), exhibited a consistent compliance level, remaining unaffected by the various waves of the virus. In contrast, the lowest compliance trajectory, designated as Class 4 in Wright et al. (2021), Fast Decliners in MacNeil et al. (2022), and Low and fluctuating in Courdi et al. (2023), was characterised by significant variability dependent on the prevailing pandemic situation. However, according to our findings, we identified two distinct forms of variability: one pertains to fluctuations in the trend of a pattern, as exemplified by the change in trend of the lowest compliance trajectory between periods, while the other involves variations in membership within compliance groups, such as the High and constant group in our study. This differentiation has not been addressed in studies on compliance behaviour with preventive measures, challenging our conventional perception of stability associated with higher compliance groups. Hence, situational factors, such as the lifting of measures, appear to impact groups disparately, emphasising the importance of elaborating intervention plans tailored to the specific characteristics of each group and not just the lowest trajectory group. Every individual transitioning from a high compliance group to a lower one poses as much of a risk as individuals in the lowest compliance group, as they too are susceptible to finding themselves in these low trajectories, subsequently affecting infection rates. As mentioned, the proportions along the diagonals in terms of trajectory membership suppose a stability in compliance behaviour throughout the first wave of the pandemic, suggesting that inter-individual differences might be at cause and not situational factors. Our findings consistently demonstrate that health literacy alone serves as a reliable factor of trajectory membership for both social distancing and staying-at-home behaviours across both observed periods. Aligned with existing COVID-19 literature, health literacy—encompassing the ability to comprehend and critically assess health-related information—stands validated as a pivotal individual factor of compliance with preventive measures, particularly amid the challenges posed by the Infodemic [ 24 ] [ 18 ] [ 57 ]. However, contrary to our expectations, health literacy emerged as a factor solely for trajectory membership and not for change in membership. This surprising outcome suggests a certain resilience and stability in the concept of health literacy, impervious to dynamic situations such as the easing of preventive measures. When delving into the factors contributing to a high level of health literacy, it becomes evident that it is largely anticipated by constant sociodemographic attributes such as education, income and migration status, as highlighted in the work of Martin et al. (2009). Given the relatively stable nature of these factors and their gradual evolution over an extended period, it could elucidate why health literacy proves to be a more effective factor of the stability of trajectory membership [ 38 ]. Our results further imply that health literacy plays a role in shaping the overall compliance pattern within a group characterised by shared stable attributes, as seen in the lowest trajectory groups in our study ( Low and constant to Low and decreasing ). Indeed, the probability of belonging in the lowest trajectory significantly decreased when displaying high health literacy levels between the observed periods. With the government initiating the easing of measures, individuals who were less inclined to exhibit high health literacy tended to persist within the same trajectory, but the pattern of their trajectory underwent a shift, adopting a decreasing trend. Therefore, individual factors, such as health literacy, appear to be more adept at forecasting stability within a group while also capturing variations in trends. We observed another type of variability in our study, that challenged the conventional perception of stability associated with high compliance groups. Trajectory change, observed mostly within the High and constant and High and decreasing groups, was moderately predicted by social factors, like perception of provincial social norms and trust in science. Social norms, defined by a set of rules and expectations in terms of socially acceptable behaviours and attitudes, were often studied in relation to compliance with preventive measures, as they elicit long term changes in behaviour and target large groups [ 45 ]. In affirming prior research emphasising the pivotal role of social norms in predicting compliance behaviour [ 5 ] [ 61 ], our study goes further and proposes that social norms may also serve as a factor for changes in trajectory. In accordance with Reynolds et al. (2015)’s comprehensive framework on behaviour change, individuals from the same province, acting as referent points, play a crucial role in shaping attitudes and behaviour toward compliance. The salience of this provincial identity may have diminished as the situation evolved with the easing of measures. Notably, during the initial months of the pandemic, most provincial governments conveyed messages of encouragement, emphasising the imperative of collective action within the province to limit the virus's spread [ 33 ] [ 10 ]. As the measures were lifted, the salience of this provincial identity waned in the public discourse, potentially explaining the behavioural shifts observed among some Canadians. In contrast, individuals whose provincial identity was initially salient were more likely to comply with the highest compliance trajectory or transition to it even amid the easing of measures. Noteworthy, our results suggest the relevance of the provincial identity in predicting compliance behaviour, compared to national identity, positing that the government should prioritise norm-based interventions at state levels. Furthermore, our study revealed that trust in science, rather than trust in the government, emerged as a significant factor of trajectory change, thus affirming our initial hypothesis. Trust in science encompasses confidence in the policies implemented and decisions made by the scientific community to safeguard against emerging threats [ 54 ]. Throughout the pandemic, scientists were entrusted by governments to devise optimal strategies for curbing the virus's spread. However, as new information surfaced regarding the virus's contagiousness, policies underwent adjustments, such as the introduction of mask mandates [ 70 ]. Our findings suggest that individuals with higher levels of trust in science were not only more likely to comply with recommended behaviour but also demonstrated a propensity to either maintain or transition to higher compliance patterns. This adaptability could stem from their capacity to assimilate novel information about the virus communicated by the scientific community and understand the reasoning behind these policies. Strengths and limitations To our knowledge, this study stands as the first to explore the dynamic nature of behaviour compliance with preventive measures during the COVID-19 pandemic and its associated factors, through Joint-Trajectory analysis. Apart from its innovative methodology, the study boasts several strengths, including its longitudinal design, a representative sample drawn from the Canadian population, short intervals between survey periods during the initial wave, and measurement times spanning both mandatory and lifting measure periods. These attributes collectively enhance the study's robustness and reliability. However, it is crucial to interpret our results in consideration of several limitations. First and foremost, compliance behaviours such as social distancing and staying-at-home were evaluated through self-reported scores, introducing a potential impact on the reproducibility of our findings. While it is noteworthy that many studies on the COVID-19 pandemic have relied on self-reported compliance measures [ 15 ] [ 77 ] [ 36 ], we advocate for the incorporation of external measures, such as mobile-phone mobility data [ 51 ], to mitigate desirability bias. Additionally, our factors were predominantly assessed using single homemade self-reported items, limiting the comprehensive evaluation of the various dimensions inherent in a complex concept like social norms and challenging the validity of our measures. Lastly, it is essential to exercise caution when interpreting our findings related to factors of trajectory change, as some comparisons involved relatively small sample sizes (e.g., 106 vs. 10). Conclusion Latent Class Growth Analysis proves invaluable in assessing trends in compliance behaviour with preventive measures, allowing for the identification of inter-individual differences and variability within a group. Through this analysis, we discerned three distinct trajectories during both mandatory and lifting measures periods: Low and constant ( decreasing during the lifting measures period), High and decreasing , and High and constant . Expanding our comprehension of compliance behaviour, we delved into transitions between these trajectories and found that the majority of individuals tended to remain within similar groups. Contrary to the conventional belief in the stability of high compliance groups, our findings challenge this assumption, revealing that these groups were the most susceptible to transitioning to lower compliance categories. This insight bears significant implications for policies related to compliance behaviour, suggesting that vulnerability to decreased compliance is universal, albeit manifesting differently across various groups. Declarations Ethics approval and consent to participate The authors declare that all experimental protocols were approved by Comité d'éthique de la recherche en éducation et en psychologie (CEREP) of University of Montreal (Certificat no CEREP-20-038-D). Consent for publication Not applicable. Availability of data and material The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This study is part of a larger national research project financed by the Canadian Institutes of Health Research (CIHR) that examines the social consequences of the COVID-19 pandemic on Canadians (grant number = 170633, for more information, see https://csdc-cecd.wixsite.com/covid19csi?lang=en). This study is also part of the Mathematics for Public Health program funded by the joint Natural Sciences and Engineering Research Council (NSERC) and Public Health Agency of Canada (PHAC) Emerging Infectious Disease Modeling initiative. Authors’ contribution S.R.A. performed the data analysis and wrote the first drafts of the manuscript. E.L., M.P.D., J.M.L. and J.B. assisted in the operationalization of the research question, reviewed the methodology and the manuscript. R.d.l.S provided the research question and assisted at all stages of the study. Acknowledgements Not applicable. References Aragão DP, dos Santos DH, Mondini A, Gonçalves LMG. National Holidays and Social Mobility Behaviors: Alternatives for Forecasting COVID-19 Deaths in Brazil. Int J Environ Res Public Health. 2021;18(21):11595. https://doi.org/10.3390/ijerph182111595 . Battiston P, Kashyap R, Rotondi V. Reliance on scientists and experts during an epidemic: Evidence from the COVID-19 outbreak in Italy. SSM - Popul Health. 2021;13:100721. https://doi.org/10.1016/j.ssmph.2020.100721 . Bicchieri C, Fatas E, Aldama A, Casas A, Deshpande I, Lauro M, Parilli C, Spohn M, Pereira P, Wen R. In science we (should) trust: Expectations and compliance across nine countries during the COVID-19 pandemic. PLoS ONE. 2021;16(6):e0252892. https://doi.org/10.1371/journal.pone.0252892 . Birkhäuer J, Gaab J, Kossowsky J, Hasler S, Krummenacher P, Werner C, Gerger H. Trust in the health care professional and health outcome: A meta-analysis. PLoS ONE. 2017;12(2):e0170988. https://doi.org/10.1371/journal.pone.0170988 . Cabrera-Álvarez P, Hornsey MJ, Lobera J. Determinants of self-reported adherence to COVID-19 regulations in Spain: social norms, trust and risk perception. Health Promot Int. 2022;37(6):daac138. https://doi.org/10.1093/heapro/daac138 . Canadian Council on Learning. (2008). Health Literacy in Canada: A Healthy Understanding (p. 38). https://www.vec.ca/pdfs/2012-39.pdf . Canadian Institute for Health Information. (2022). Canadian Data Set of COVID-19 Interventions — Data Tables . Canadian Institute for Health Information. https://www.cihi.ca/en/canadian-covid-19-intervention-timeline . Caron-Diotte M, Dorfman A, Pelletier-Dumas M, Lacourse É, Lina JM, Stolle D, Taylor DM, de la Sablonnière R. (2021). COVID-19 Canada: The end of the world as we know it? Technical Report No. 2. Handling planned and unplanned missing data. Université de Montréal. Caron-Diotte M, de la Sablonnière P-D. Roxane (2023) Handling Planned and Unplanned Missing Data in a Longitudinal Study, The Quantitative Methods for Psychology, 19(2), 123–135. 10.20982/tqmp.19.2.p123 . CBC. (2023, November 14). Billboard campaign aims to dispel mistrust of vaccines in African Nova Scotian community | CBC News . CBC. https://www.cbc.ca/news/canada/nova-scotia/billboard-campaign-vaccine-mistrust-african-nova-scotia-1.7027975 . CDC. (2023, March 15). CDC Museum COVID-19 Timeline . Centers for Disease Control and Prevention. https://www.cdc.gov/museum/timeline/covid19.html . Chen AMH, Armbruster AL, Buckley B, Campbell JA, Dang DK, Devraj R, Drame I, Edwards A, Haack SL, Ma Q, Petry N, Planas LG, Sadowski CA, Santee J, Wade L, Borja-Hart N. Inclusion of Health Disparities, Cultural Competence, and Health Literacy Content in US and Canadian Pharmacy Curriculums. Am J Pharm Educ. 2021;85(1):8200. https://doi.org/10.5688/ajpe8200 . Chen X, Zhang A, Wang H, Gallaher A, Zhu X. Compliance and containment in social distancing: mathematical modeling of COVID-19 across townships. Int J Geogr Inf Sci. 2021;35(3):446–65. https://doi.org/10.1080/13658816.2021.1873999 . Côté SM, Vaillancourt T, Barker ED, Nagin D, Tremblay RE. The joint development of physical and indirect aggression: Predictors of continuity and change during childhood. Dev Psychopathol. 2007;19(1):37–55. https://doi.org/10.1017/S0954579407070034 . Courdi C, Ali R, Pelletier-Dumas S, Stolle M, Dorfman D, Lina A, Lacourse J-M, É., de la Sablonnière R. How level of understanding and type of used sources relate to adherence to COVID-19 public health measures in Canada. Sci Rep. 2023;13(1):13065. https://doi.org/10.1038/s41598-023-38824-0 . Dekeyser S, Schmits E, Glowacz F, Klein O, Schmitz M, Wollast R, Yzerbyt V, Luminet O. Predicting Compliance with Sanitary Behaviors among Students in Higher Education During the Second COVID-19 Wave: The Role of Health Anxiety and Risk Perception. Physiol Belgica. 2023;63(1):1–15. https://doi.org/10.5334/pb.1171 . de la Sablonnière R, Dorfman A, Pelletier-Dumas M, Lacourse É, Lina JM, Stolle D, Taylor DM, Benoit Z, Boulanger A, Caron-Diotte M, Mérineau S, Nadeau A. (2020). COVID-19 Canada: The end of the world as we know it? (Technical report No. 1). Presenting the COVID-19 Survey. Université de Montréal. Duplaga M. The Roles of Health and e-Health Literacy, Conspiracy Beliefs and Political Sympathy in the Adherence to Preventive Measures Recommended during the Pandemic. Int J Environ Res Public Health. 2022;19(14):8346. https://doi.org/10.3390/ijerph19148346 . Fishbein M, Ajzen I. Predicting and Changing Behavior: The Reasoned Action Approach. Taylor & Francis; 2011. Forward SE. The theory of planned behaviour: The role of descriptive norms and past behaviour in the prediction of drivers’ intentions to violate. Transp Res Part F: Traffic Psychol Behav. 2009;12(3):198–207. https://doi.org/10.1016/j.trf.2008.12.002 . Galasso V, Pons V, Profeta P, Becher M, Brouard S, Foucault M. (2020). Gender differences in COVID-19 attitudes and behavior: Panel evidence from eight countries. Proceedings of the National Academy of Sciences , 117 (44), 27285–27291. https://doi.org/10.1073/pnas.2012520117 . Gao S, Binod P, Chukwu CW, Kwofie T, Safdar S, Newman L, Choe S, Datta BK, Attipoe WK, Zhang W, van den Driessche P. A mathematical model to assess the impact of testing and isolation compliance on the transmission of COVID-19. Infect Disease Modelling. 2023;8(2):427–44. https://doi.org/10.1016/j.idm.2023.04.005 . Gelfand MJ, Gavrilets S, Nunn N. Norm Dynamics: Interdisciplinary Perspectives on Social Norm Emergence, Persistence, and Change. Ann Rev Psychol. 2024;75(1). https://doi.org/10.1146/annurev-psych-033020-013319 . annurev-psych-033020-013319. Hermans L, Van den Broucke S, Gisle L, Demarest S, Charafeddine R. Mental health, compliance with measures and health prospects during the COVID-19 epidemic: the role of health literacy. BMC Public Health. 2021;21(1):1365. https://doi.org/10.1186/s12889-021-11437-w . INSPQ. (2020, May 22). COVID-19: Mesures sanitaires recommandées pour la population générale | INSPQ . Institut national de santé publique du Québec. https://www.inspq.qc.ca/publications/3008-mesures-sanitaires-population-generale-covid19 . Jones BL, Nagin DS. Advances in Group-Based Trajectory Modeling and an SAS Procedure for Estimating Them. Sociol Methods Res. 2007;35(4):542–71. https://doi.org/10.1177/0049124106292364 . Jones BL, Nagin DS, Roeder K. A SAS Procedure Based on Mixture Models for Estimating Developmental Trajectories. Sociol Methods Res. 2001;29(3):374–93. https://doi.org/10.1177/0049124101029003005 . Jørgensen F, Bor A, Rasmussen MS, Lindholt MF, Petersen MB. (2022). Pandemic fatigue fueled political discontent during the COVID-19 pandemic. Proceedings of the National Academy of Sciences , 119 (48), e2201266119. https://doi.org/10.1073/pnas.2201266119 . Kil H, Lacourse E, Mageau GA, Pelletier-Dumas M, Dorfman A, Stolle D, Lina J-M, de la Sablonnière R. (2023). Initial risk factors, self-compassion trajectories, and well-being outcomes during the COVID-19 pandemic: A person-centered approach. Frontiers in Psychology , 13 . https://www.frontiersin.org/articles/ 10.3389/fpsyg.2022.1016397 . Kokko K, Tremblay RE, Lacourse E, Nagin DS, Vitaro F. Trajectories of Prosocial Behavior and Physical Aggression in Middle Childhood: Links to Adolescent School Dropout and Physical Violence. J Res Adolescence. 2006;16(3):403–28. https://doi.org/10.1111/j.1532-7795.2006.00500.x . Krot K, Sousa JP. Factors Impacting On Patient Compliance with Medical Advice: Empirical Study. Eng Manage Prod Serv. 2017;9(2):73–81. https://doi.org/10.1515/emj-2017-0016 . Kwasnicka D, Dombrowski SU, White M, Sniehotta F. Theoretical explanations for maintenance of behaviour change: a systematic review of behaviour theories. Health Psychol Rev. 2016;10(3):277–96. https://doi.org/10.1080/17437199.2016.1151372 . La Presse. (2020, September 22). Une campagne publicitaire gouvernementale « plus percutante » sur la COVID-19. La Presse . https://www.lapresse.ca/actualites/covid-19/2020-09-22/une-campagne-publicitaire-gouvernementale-plus-percutante-sur-la-covid-19.php . Leviston Z, Stanley SK, Walker I. Australians underestimate social compliance with coronavirus restrictions: findings from a national survey. Aust N Z J Public Health. 2022;46(3):304–6. https://doi.org/10.1111/1753-6405.13195 . Lyu W, Wehby GL. Comparison of Estimated Rates of Coronavirus Disease 2019 (COVID-19) in Border Counties in Iowa Without a Stay-at-Home Order and Border Counties in Illinois With a Stay-at-Home Order. JAMA Netw Open. 2020;3(5):e2011102. https://doi.org/10.1001/jamanetworkopen.2020.11102 . MacNeil S, Deschênes S, Knäuper B, Carrese-Chacra E, Dialahy IZ, Suh S, Durif F, Gouin J-P. Group-based trajectories and predictors of adherence to physical distancing during the COVID-19 pandemic. Psychol Health. 2022;37(12):1492–510. https://doi.org/10.1080/08870446.2021.2014486 . Martin LT, Ruder T, Escarce JJ, Ghosh-Dastidar B, Sherman D, Elliott M, Bird CE, Fremont A, Gasper C, Culbert A, Lurie N. Developing predictive models of health literacy. J Gen Intern Med. 2009;24(11):1211–6. https://doi.org/10.1007/s11606-009-1105-7 . McCaffery KJ, Ayre J, Dodd R, Pickles K, Copp T, Muscat DM, Nickel B, Cvejic E, Zhang M, Mac O, Isautier J, Cornell S, Wolf MS, Bonner C. Disparities in public understanding, attitudes, and intentions during the Covid-19 pandemic: The role of health literacy. Inform Serv Use. 2023;43(2):101–13. https://doi.org/10.3233/ISU-230185 . Middleton KR, Anton SD, Perri MG. Long-Term Adherence to Health Behavior Change. Am J Lifestyle Med. 2013;7(6):395–404. https://doi.org/10.1177/1559827613488867 . Montagni I, Ouazzani-Touhami K, Mebarki A, Texier N, Schück S, Tzourio C, the CONFINS group. Acceptance of a Covid-19 vaccine is associated with ability to detect fake news and health literacy. J Public Health. 2021;43(4):695–702. https://doi.org/10.1093/pubmed/fdab028 . Morales-Rodríguez FM, Pérez-Mármol JM. (2019). The Role of Anxiety, Coping Strategies, and Emotional Intelligence on General Perceived Self-Efficacy in University Students. Frontiers in Psychology , 10 . https://www.frontiersin.org/articles/10.3389/fpsyg.2019.01689 . Mukerjee S, Chow CM, Li M. Mitigation strategies and compliance in the COVID-19 fight; how much compliance is enough? PLoS ONE. 2021;16(8):e0239352. https://doi.org/10.1371/journal.pone.0239352 . Nagin DS. Analyzing developmental trajectories: A semiparametric, group-based approach. Psychol Methods. 1999;4(2):139–57. https://doi.org/10.1037/1082-989X.4.2.139 . Nagin DS, Tremblay RE. Analyzing developmental trajectories of distinct but related behaviors: A group-based method. Psychol Methods. 2001;6(1):18–34. https://doi.org/10.1037/1082-989X.6.1.18 . Neville FG, Templeton A, Smith JR, Louis WR. Social norms, social identities and the COVID-19 pandemic: Theory and recommendations. Soc Pers Psychol Compass. 2021;15(5):e12596. https://doi.org/10.1111/spc3.12596 . Nielsen JH, Lindvall J. Trust in government in Sweden and Denmark during the COVID-19 epidemic. West Eur Politics. 2021;44(5–6):1180–204. https://doi.org/10.1080/01402382.2021.1909964 . OECD. (2020, July 3). Transparency, communication and trust: The role of public communication in responding to the wave of disinformation about the new coronavirus . OECD. https://read.oecd-ilibrary.org/view/?ref=135_135220-cvba4lq3ru&title=Transparency-communication-and-trust-The-role-of-public-communication-in-responding-to-the-wave-of-disinformation-about-the-new-coronavirus OECD. (2021, May 21). The Territorial Impact of COVID-19: Managing the Crisis and Recovery across Levels of Government . OECD. https://read.oecd-ilibrary.org/view/?ref=1095_1095253-immbk05xb7&title=The-territorial-impact-of-COVID-19-Managing-the-crisis-and-recovery-across-levels-of-government Okan O, Messer M, Levin-Zamir D, Paakkari L, Sørensen K. Health literacy as a social vaccine in the COVID-19 pandemic. Health Promot Int. 2022;daab197. https://doi.org/10.1093/heapro/daab197 . Pelletier-Dumas M, Dorfman A, Jean-Marc L, Dietling S, Taylor D, de la Sablonnière R. (2020). The role of clear and coherent government measures to reduce the spread of COVID-19 (No. 1). https://a5eaedf4-8d49-4b28-9cae-77db17679ab5.filesusr.com/ugd/68376a_689097aa55924fcdab5e7c6e513440ab.pdf . Petherick A, Goldszmidt R, Andrade EB, Furst R, Hale T, Pott A, Wood A. A worldwide assessment of changes in adherence to COVID-19 protective behaviours and hypothesized pandemic fatigue. Nat Hum Behav. 2021;5(9):1145–60. https://doi.org/10.1038/s41562-021-01181-x . Public Health Agency of Canada. (n.d.). COVID-19: Prevention and risks . Public Health Agency of Canada. Retrieved 10 December 2023, from https://www.canada.ca/en/public-health/services/diseases/2019-novel-coronavirus-infection/prevention-risks.html#p . Rashed EA, Hirata A. One-Year Lesson: Machine Learning Prediction of COVID-19 Positive Cases with Meteorological Data and Mobility Estimate in Japan. Int J Environ Res Public Health. 2021;18(11):5736. https://doi.org/10.3390/ijerph18115736 . Resnik DB. Scientific Research and the Public Trust. Sci Eng Ethics. 2011;17(3):399–409. https://doi.org/10.1007/s11948-010-9210-x . Reynolds KJ, Subašić E, Tindall K. The Problem of Behaviour Change: From Social Norms to an Ingroup Focus. Soc Pers Psychol Compass. 2015;9(1):45–56. https://doi.org/10.1111/spc3.12155 . Rigotti T, De Cuyper N, Sekiguchi T. The Corona Crisis: What Can We Learn from Earlier Studies in Applied Psychology? Appl Psychol. 2020;69(3):1–6. https://doi.org/10.1111/apps.12265 . Rodon C, Chin J, Chevalier A. Assessing COVID-19 Health Literacy (CoHL) and its relationships with sociodemographic features, locus of control and compliance with social distancing rules during the first lockdown in France. Health Educ Res. 2022;37(3):143–54. https://doi.org/10.1093/her/cyac009 . Rogers RW. A Protection Motivation Theory of Fear Appeals and Attitude Change1. J Psychol. 1975;91(1):93–114. https://doi.org/10.1080/00223980.1975.9915803 . Rosenstock IM. The Health Belief Model and Preventive Health Behavior. Health Educ Monogr. 1974;2(4):354–86. https://doi.org/10.1177/109019817400200405 . Rutty C. (2022, March 2). COVID-19 Pandemic in Canada . Can Encyclopedia. https://www.thecanadianencyclopedia.ca/en/article/covid-19-pandemic . Schumpe, B. M., Van Lissa, C. J., Bélanger, J. J., Ruggeri, K., Mierau, J., Nisa,C. F., Molinario, E., Gelfand, M. J., Stroebe, W., Agostini, M., Gützkow, B., Jeronimus,B. F., Kreienkamp, J., Kutlaca, M., Lemay, E. P., Reitsema, A. M., vanDellen, M. R.,Abakoumkin, G., Abdul Khaiyom, J. H., … Leander, N. P. (2022). Predictors of adherence to public health behaviors for fighting COVID-19 derived from longitudinal data. Scientific Reports , 12 (1), 3824. https://doi.org/10.1038/s41598-021-04703-9. Schwarzer R, Warner LM. (2013). Perceived Self-Efficacy and its Relationship to Resilience. In S. Prince-Embury & D. H. Saklofske, editors, Resilience in Children, Adolescents, and Adults: Translating Research into Practice (pp. 139–150). Springer. https://doi.org/10.1007/978-1-4614-4939-3_10 . Sheppard VB. Providing health care to low-income women: a matter of trust. Fam Pract. 2004;21(5):484–91. https://doi.org/10.1093/fampra/cmh503 . Sjödin H, Wilder-Smith A, Osman S, Farooq Z, Rocklöv J. Only strict quarantine measures can curb the coronavirus disease (COVID-19) outbreak in Italy, 2020. Eurosurveillance. 2020;25(13):2000280. https://doi.org/10.2807/1560-7917.ES.2020.25.13.2000280 . Sørensen K, Van den Broucke S, Fullam J, Doyle G, Pelikan J, Slonska Z et al. Health literacy and public health: A systematic review and integration of definitions and models. BMC Public Health [Internet]. 2012 Jan 25 [cited 2024 May 8];12(1):80. https://doi.org/10.1186/1471-2458-12-80 . Statistics Canada. (2013, October 8). The Daily — Skills in Canada: First results from the Programme for the International Assessment of Adult Competencies, 2012 (final) . https://www150.statcan.gc.ca/n1/daily-quotidien/131008/dq131008b-eng.htm . Stok FM, De Ridder DTD, De Vet E, De Wit JBF. Don’t tell me what I should do, but what others do: The influence of descriptive and injunctive peer norms on fruit consumption in adolescents. Br J Health Psychol. 2014;19(1):52–64. https://doi.org/10.1111/bjhp.12030 . Tam CC, Li X, Li X, Wang Y, Lin D. Adherence to preventive behaviors among college students during COVID-19 pandemic in China: The role of health beliefs and COVID-19 stressors. Curr Psychol. 2023;42(20):17280–90. https://doi.org/10.1007/s12144-021-01942-x . Teixeira da Silva JA, Tsigaris P, Erfanmanesh M. Publishing volumes in major databases related to Covid-19. Scientometrics. 2021;126(1):831–42. https://doi.org/10.1007/s11192-020-03675-3 . The Canadian Press. (2021). From first cases to first vaccines: A timeline of COVID-19 in Canada - National | Globalnews.ca . Global News. https://globalnews.ca/news/7597228/covid-canada-timeline/ . Trentacosta CJ, Criss MM, Shaw DS, Lacourse E, Hyde LW, Dishion TJ. Antecedents and Outcomes of Joint Trajectories of Mother–Son Conflict and Warmth During Middle Childhood and Adolescence. Child Dev. 2011;82(5):1676–90. https://doi.org/10.1111/j.1467-8624.2011.01626.x . Usherwood T, LaJoie Z, Srivastava V. A model and predictions for COVID-19 considering population behavior and vaccination. Sci Rep. 2021;11(1):12051. https://doi.org/10.1038/s41598-021-91514-7 . Wollast R, Schmitz M, Bigot A, Speybroeck N, Lacourse É, De La Sablonnière R, Luminet O. Trajectories of health behaviors during the COVID-19 pandemic: a longitudinal analysis of handwashing, mask wearing, social contact limitations, and physical distancing. Psychol Health. 2023;1–28. https://doi.org/10.1080/08870446.2023.2278706 . World Health Organisation. (n.d.). WHO Coronavirus (COVID-19) Dashboard. Retrieved 10. December 2023, from https://covid19.who.int . World Health Organization. (n.d.). Let’s flatten the infodemic curve. World Health Organization. Retrieved 10. December 2023, from https://www.who.int/news-room/spotlight/let-s-flatten-the-infodemic-curve . Wright L, Fancourt D. Do predictors of adherence to pandemic guidelines change over time? A panel study of 22,000 UK adults during the COVID-19 pandemic. Prev Med. 2021;153:106713. https://doi.org/10.1016/j.ypmed.2021.106713 . Wright L, Steptoe A, Fancourt D. Trajectories of Compliance With COVID-19 Related Guidelines: Longitudinal Analyses of 50,000 UK Adults. Ann Behav Med. 2022;56(8):781–90. https://doi.org/10.1093/abm/kaac023 . Zmerli S, van Meer TWG. der. (2017). Handbook on Political Trust . Edward Elgar Publishing. Zou X, Savani K. Descriptive norms for me, injunctive norms for you: Using norms to explain the risk gap. Judgm Decis Mak. 2019;14(6):644–8. https://doi.org/10.1017/S1930297500005362 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterialsBCM.docx Cite Share Download PDF Status: Published Journal Publication published 04 Mar, 2026 Read the published version in BMC Public Health → Version 1 posted Editorial decision: Revision requested 04 Jun, 2024 Submission checks completed at journal 13 May, 2024 Editor assigned by journal 13 May, 2024 First submitted to journal 08 May, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4391822","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":304964370,"identity":"939e8459-f493-405a-ab88-c95e588b6c36","order_by":0,"name":"Sahar Ramazan Ali","email":"","orcid":"","institution":"University of Montreal","correspondingAuthor":false,"prefix":"","firstName":"Sahar","middleName":"Ramazan","lastName":"Ali","suffix":""},{"id":304964371,"identity":"a073a6c0-071e-4361-8632-1f89e96e1b96","order_by":1,"name":"Eric Lacourse","email":"","orcid":"","institution":"University of Montreal","correspondingAuthor":false,"prefix":"","firstName":"Eric","middleName":"","lastName":"Lacourse","suffix":""},{"id":304964372,"identity":"f2c50400-013d-430c-b573-1519a0758fbd","order_by":2,"name":"Mathieu Pelletier-Dumas","email":"","orcid":"","institution":"University of Montreal","correspondingAuthor":false,"prefix":"","firstName":"Mathieu","middleName":"","lastName":"Pelletier-Dumas","suffix":""},{"id":304964373,"identity":"15ae2eaf-3b01-4fd4-a140-29c8ee69c0eb","order_by":3,"name":"Jean-Marc Lina","email":"","orcid":"","institution":"École de Technologie Supérieure","correspondingAuthor":false,"prefix":"","firstName":"Jean-Marc","middleName":"","lastName":"Lina","suffix":""},{"id":304964374,"identity":"c7019547-492f-4d62-8e05-8da0cd2876f1","order_by":4,"name":"Jacques Belair","email":"","orcid":"","institution":"University of Montreal","correspondingAuthor":false,"prefix":"","firstName":"Jacques","middleName":"","lastName":"Belair","suffix":""},{"id":304964375,"identity":"2ff767a5-01d6-41e7-b8b4-ee8665cfc9b1","order_by":5,"name":"Roxane de la Sablonniere","email":"data:image/png;base64,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","orcid":"","institution":"University of Montreal","correspondingAuthor":true,"prefix":"","firstName":"Roxane","middleName":"de la","lastName":"Sablonniere","suffix":""}],"badges":[],"createdAt":"2024-05-09 01:14:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4391822/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4391822/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12889-026-26347-y","type":"published","date":"2026-03-04T15:57:36+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":57298926,"identity":"51ab0f87-b956-4a4d-b5df-9a2ffd098149","added_by":"auto","created_at":"2024-05-28 20:47:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":71257,"visible":true,"origin":"","legend":"\u003cp\u003eCompliance trajectories to social distancing measure during the first wave of the pandemic (April 2020- May 2020 and May 2020- July 2020)\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-4391822/v1/ab822994a7c2f89c800671a6.png"},{"id":57298927,"identity":"e1c90fe1-6ab0-46c2-b164-983bfa08ef77","added_by":"auto","created_at":"2024-05-28 20:47:12","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":272291,"visible":true,"origin":"","legend":"\u003cp\u003eCompliance trajectories to stay at home measure during the first wave of the pandemic (April 2020- May 2020 and May 2020- July 2020)\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-4391822/v1/ef514e0516ef443e85e7a679.jpeg"},{"id":104250745,"identity":"48b39329-26ee-466a-9e15-83dbdff0b9ed","added_by":"auto","created_at":"2026-03-09 16:07:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2889495,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4391822/v1/51c750c0-7746-42d0-8887-9da8771f3856.pdf"},{"id":57298925,"identity":"54cebc0e-5ca9-4b27-adc1-12a477d7f3c9","added_by":"auto","created_at":"2024-05-28 20:47:12","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":41962,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialsBCM.docx","url":"https://assets-eu.researchsquare.com/files/rs-4391822/v1/97482634d941d278f37cae2c.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Beyond What Meets the Eye: Unveiling Dynamics of Compliance with Preventive Measures in the COVID-19 Era","fulltext":[{"header":"Introduction","content":"\u003cp\u003eResearch related to COVID-19 saw an exponential growth following the World Health Organisation\u0026rsquo;s announcement on the severity of the virus outbreak, upgrading its classification to a pandemic. Researchers from diverse fields were mandated by government agencies to answer society\u0026rsquo;s pressing questions in response to the COVID-19 pandemic, whether they were related to the virus\u0026rsquo; transmission modes or to the communities\u0026rsquo; adaptive capacities. One of the more prominent themes in the novel COVID-19 literature is compliance to preventive measures, namely hand washing, mask wearing, social distancing, and staying at home. Early on, during the first wave of the pandemic, studies reported that a small minority (between 5% and 12%) did not comply with governmental guidelines, furthering the need to explore the reasons behind this transgressive behaviour [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e][\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e][\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. These studies evaluated compliance levels to preventive measures as a stable phenomenon without considering transition movements between levels of compliance as the pandemic situation evolved. Furthermore, several social and sociodemographic factors were identified as probable reasons behind non-compliance to preventive measures, but these factors were only examined in regard to their relationship with compliance levels and not change in compliance levels ([\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]; [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]; [\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e]; [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]). Hence, this study represents a continuation of a longitudinal investigation initiated by Courdi et al. (2023) with a representative Canadian sample (N\u0026thinsp;=\u0026thinsp;3617). The primary aim of the earlier study was to identify trajectories of compliance with social distancing, contacts limitation, and mask wearing throughout the first year of the pandemic and its relation with factors. Here, we aim to identify and describe individuals\u0026rsquo; compliance behaviours with only social distancing and staying-at-home measures, whether they are stable or varying in terms of compliance patterns and trajectory change, and determine their association with factors during the first wave of the pandemic.\u003c/p\u003e \u003cp\u003eThe global situation during the early months of the COVID-19 pandemic was critical, with the virus spreading rapidly across all countries, disregarding geographical boundaries. Within two months of reporting the first case, the World Health Organization documented approximately 20,000 cases in March 2020 ([\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e]). As a result, governments worldwide implemented preventive measures to mitigate the spread of the virus, as vaccines were still in the development phase. Essential preventive measures were classified into two categories: individual and community-based measures ([\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]). Individual measures encompassed contact limitation (staying-at-home and social distancing), respiratory and hygiene etiquette, indoor ventilation, and surface disinfection ([\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].; [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]). Meanwhile, community-based measures comprised the closure of non-essential businesses and restrictions on the number of people in a setting ([\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]; [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]). Given that individual measures relied on voluntary compliance from the population, it became more pertinent to delve into these compliance behaviours rather than focusing on forced directives, such as business closures [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. However, the effectiveness of each individual preventive measure was not equivalent but hinged on its ability to limit the spread of the virus [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. For instance, limiting contacts by staying at home was regarded as the most effective measure as it prevented individuals from coming into physical contact with others, thereby significantly reducing the risk of transmission [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. If limiting contacts by staying-at-home was impossible, it was recommended to implement physical barriers, such as practising social distancing and wearing masks. Notably, social distancing took precedence as a physical barrier measure over mask wearing, because it was specifically designed to restrict interactions between individuals, thus also decreasing the likelihood of infection [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Hence, studies suggested focusing on staying-at-home and social distancing as measures of interest when studying compliance during the pandemic, as they significantly influenced transmission rates [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e][\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eLongitudinal studies on compliance behaviour with social distancing and staying-at-home measures revealed notable variations in compliance levels during the COVID-19 pandemic. Indeed, [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e] conducted a study in the United Kingdom with 50,000 participants during the first year of the pandemic and found four distinct longitudinal patterns of compliance: \u003cem\u003eClass 1\u003c/em\u003e (32.8%) with high and constant compliance, \u003cem\u003eClass 2\u003c/em\u003e (28.66%) with high and fluctuating compliance, \u003cem\u003eClass 3\u003c/em\u003e (23.98%) with medium high and fluctuating compliance, \u003cem\u003eClass 4\u003c/em\u003e (14.56%) with low and decreasing compliance. Other longitudinal studies examining trajectory compliance have consistently identified a similar number of compliance groups and reported a decline in compliance, particularly evident in the lowest compliance group during the initial months of the pandemic, prompting further investigation [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] [\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe decrease in compliance with preventive measures, specifically to social distancing and staying-at-home measures, was first observed at the end of the first wave of the pandemic (Spring 2020) as highlighted in previous literature [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e] [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] [\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e]. In the Canadian pandemic context, this period spanned from March 2020 to July 2020 [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. During that first wave, Canadians experienced tremendous changes, such as being required to comply with governmental recommendations under the threat of sanctions [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e]. As the COVID-19 situation improved with fewer reported cases, provincial governments gradually lifted mandatory preventive measures, permitting individuals from different households to gather at home, for instance [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e] [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Hence, two noticeable periods emerged during the first wave of the pandemic in Canada: mandatory measures period and lifting measures period. With slight variations among the provinces, the mandatory measures period commenced in March 2020, and the lifting measures period began toward the end of May 2020 [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e] [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. Whereas the previously mentioned longitudinal studies successfully identified variations in compliance levels during that first pandemic wave [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e] [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] [\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e], they were unable to explore the effect of the easing of measures on membership in compliance trajectories. Therefore, two questions arise in relation to this gap: (1) Are there different trajectories of compliance with regards to social distancing and staying-at-home measures during both periods, and (2) did individuals persist in similar compliance trajectories during the first wave of the pandemic?\u003c/p\u003e \u003cp\u003eTo examine the presence of different trajectories and shifts in trajectory membership between the mandatory and lifting measures periods, it is recommended to conduct a Joint-Trajectory Analysis [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e] [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e] [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. An extension of the Latent Class Growth Analysis (LCGA), the Joint-Trajectory not only identifies distinguishable patterns of response over time but also enables the estimation of probabilities of co-occurrence or continuity in two distinct yet related behaviours [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e] [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e] [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. It has been widely employed in developmental research to investigate how trajectories of a behaviour at an early age can serve as precursors for maladaptive behaviour later in life [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] [\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e]. For example, in C\u0026ocirc;t\u0026eacute; et al. (2007), the researchers aimed to examine the relation between trajectories of physical aggression at an early age and trajectories of indirect aggression later. They reported that most children exhibiting a low trajectory of physical aggression were consistently paired with a low trajectory of indirect aggression later on, indicating stability in aggression levels. On the other hand, among children initially showing higher levels of physical aggression, a shift toward a trajectory of high indirect aggression was observed. A comparable study design in the context of compliance to preventive measures has not been identified in the literature, leaving a gap in our understanding. Such a study could aid governments in targeting behaviours susceptible to change as situations evolve and identifying associated risk factors.\u003c/p\u003e \u003cp\u003eDuring the COVID-19 pandemic, researchers sought to identify factors of compliance with preventive measures, aiming to offer valuable insights for government agencies in refining their intervention strategies. Numerous social (e.g., social norms, trust, risk communication) and individual characteristics (e.g., health literacy and perceived self-efficacy) were identified as consistent factors of compliance with preventive measures like social distancing and staying-at-home [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e] [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] [\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e] [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. However, these factors were examined in the context of adopting a compliant behaviour rather than sustaining or reinforcing it as the situation evolves. This distinction is significant, as different processes and interventions come into play for both behaviours, as highlighted in the Kwasnicka et al. (2016) review. Hence, outlined below are the factors anticipated not only to forecast trajectory membership but also to predict changes in trajectories between mandatory and lifting measures periods.\u003c/p\u003e \u003cp\u003eBetween January 1st and June 30th, 2020, a total of 23,634 scientific articles on the topic of COVID-19 were documented globally [\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. Indeed, media outlets and the scientific community generated new information at an unprecedented rate, prompting the World Health Organization to label this phenomenon as an \u0026ldquo;infodemic\u0026rdquo; [\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e]. The term \"infodemic\" is specifically employed in the context of a disease outbreak, referring to an excessive amount of information that can be harmful if misleading [\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e]. This excess of information can contribute to the adoption of risk-taking behaviours that directly impact health and may exacerbate the course of an outbreak [\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e] [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Therefore, the ability to comprehend, utilise, and critically assess new health information, which translates to the three dimensions of health literacy, becomes crucial in the face of an information overload [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. In Canada, the lack of health literacy is a well-documented concern for public health officials [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Reports indicate that nine million people have limited literacy skills [\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e], with over half of Canadians reading below a high school level [\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e]. These statistics are alarming considering the prejudicial impacts of low health literacy on health outcomes, like higher mortality rates and more hospitalisations [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. On the other hand, high health literacy promotes better awareness of health issues and their consequences, facilitating the adoption of adaptive health behaviours, such as preventive measures during the pandemic [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. A team of researchers from France investigated the impact of health literacy on compliance with social distancing measures and discovered a positive association, indicating that higher health literacy was linked to increased compliance with social distancing [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. Other studies also confirmed the pivotal role of health literacy in fostering the adoption of preventive behaviours in the context of the COVID-19 pandemic [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. In addition to facilitating the adoption of new health behaviours, health literacy also acts as a facilitator when individuals need to better adapt to evolving situations, like the gradual easing of governmental restrictions [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. As Canadians transitioned to the lifting measures period, individuals with high health literacy were better adjusted to the changing environment, hence expected to more likely maintain high compliance levels or transition to higher compliance responses due to their critical assessment of the still-present virus's dangerousness.\u003c/p\u003e \u003cp\u003eAs the pandemic unfolded, practices once considered normal in social settings, such as shaking hands, were strongly discouraged and replaced by behaviours like social distancing and staying-at-home as much as possible to limit the spread of the virus. Therefore, adopting these preventive behaviours to protect oneself and others became the prevailing social norms during the pandemic, establishing a set of expectations for the new normalcy [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. These expectations were based on \u003cem\u003eothers\u003c/em\u003e\u0026rsquo; behaviours (descriptive norms) or \u003cem\u003eothers\u003c/em\u003e\u0026rsquo; attitudes and beliefs (injunctive norms), \u003cem\u003eothers\u003c/em\u003e being the ingroup. In the existing literature, there is a consensus among researchers that descriptive norms are more influential in prompting changes in health behaviours compared to injunctive norms due to the high conformity pressure associated with them [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] [\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e] [\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e]. A comparable conclusion was reached in a COVID-19 compliance study conducted in Spain [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The authors noted that the perception of descriptive norms was the most important factor of compliance with social distancing measures, followed by trust in science, perceived effectiveness of measures, and finally, perceived risk of infection. Furthermore, a high perception of descriptive norms is not only more likely to induce the desired behavioural change but also to foster sustainable changes in behaviour, possibly due to the norm internalisation process that is immune to situational factors like the lifting of preventive measures [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Thus, individuals who initially reported a strong perception of descriptive norms should be expected not only to comply strongly with preventive measures, as indicated in Cabrera-\u0026Aacute;lvarez et al. (2022), but also to continue exhibiting high levels of compliance as the pandemic situation evolves. Finally, when referring to descriptive norms, the \u003cem\u003eothers\u003c/em\u003e become crucial in navigating new expectations [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], since they represent individuals from the same group. In the Canadian political landscape, the \u003cem\u003eothers\u003c/em\u003e could be Canadians in general or the residents of the same province, two distinct referent-points [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. To our knowledge, no study has made this distinction in regard to compliance with preventive measures, which would help governments target relevant national and provincial interventions.\u003c/p\u003e \u003cp\u003eDuring the pandemic, some individuals were confined in small living arrangements, while others were unable to work remotely. In both instances, situational factors posed obstacles to compliance with preventive measures, specifically social distancing and staying-at-home, which can hinder perception of control and ultimately perceived self-efficacy. Perceived self-efficacy reflects the belief in one's capability to modify health behaviours by taking personal actions [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. It consistently emerges as a significant individual factor in various health behaviour theories, such as the Health Belief Model [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], Reasoned Action approach [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], and Protection Motivation Theory [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. In the Health Belief Model, for example, the likelihood of adopting a preventive health behaviour is dependent on having greater confidence in one\u0026rsquo;s ability to apply that health behaviour, which amounts to higher perceived self-efficacy [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. Additionally, individuals with a high sense of self-efficacy tend to invest more effort and demonstrate greater persistence compared to those with low self-efficacy [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. In the face of obstacles, like living in small arrangements and the impossibility to work remotely, the former rebound more quickly and maintain health behaviours. On the other hand, studies report that a low sense of self-efficacy can have damaging effects beyond decreasing compliance to health behaviours, like higher risk of depression, anxiety and helplessness [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. In the context of the COVID-19 pandemic, this individual factor was studied in relation to compliance with preventive measures such as social distancing and staying-at-home. In a longitudinal study conducted in China [\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e], the importance of perceived self-efficacy and perceived barriers was highlighted as crucial factors in adopting and sustaining compliance with physical distancing measures over a span of three months, even in the presence of stressors such as the lifting of preventive measures. Given these results, individuals with high perceived self-efficacy are expected not only to initially exhibit high compliance behaviour but also to sustain or transit to these high levels of compliance, even during the transition between mandatory and lifting measures periods.\u003c/p\u003e \u003cp\u003eDuring the initial months of the virus outbreak, many individuals questioned the government's ability to effectively manage the pandemic situation, while others challenged the scientific community to provide reliable information. The lack of trust at both levels, government and science, has been widely documented as a strong factor of defiant health behaviour [\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e] [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Trust in government is akin to supporting and being confident about public directives [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. To preserve a functional democracy during an economic or social crisis, a minimum level of trust in government is necessary, as it would be easier to implement directives in the population if they believed in the competence of public institutions to deal with the crisis [\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e]. Trust in science involves being confident in the results produced by the scientific community and their ability to elaborate effective social and economic policies [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Indeed, when new issues or problems emerge, individuals, lacking the resources to understand these complex problems, must come to trust the scientific community to make the most appropriate decisions [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. The literature indicates that compliance behaviour with preventive measures was predicted by both trust in government and trust in science; however, the influence of trust in science outweighed that of trust in government [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Indeed, individuals with higher trust in science were found to be more open to acquiring new knowledge related to COVID, facilitating their adaptation to novel information, and ultimately aiding in navigating an evolving situation such as the easing of measures [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. In light of previous findings, Canadians who expressed high levels of trust in both government and science are expected to exhibit greater compliance with preventive measures. However, it is anticipated that only those individuals with a high level of trust in science will demonstrate better adaptation in maintaining or transitioning to a high compliance behaviour amid changing situations.\u003c/p\u003e \u003cp\u003eAt the onset of the health crisis, Canadian provincial governments implemented region-specific public policies, particularly concerning preventive measures. This decentralised approach offers advantages, such as increased flexibility and innovation. However, solely relying on this governance model does not ensure satisfactory outcomes, as evidenced by varying death and case numbers across provinces during the pandemic [\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e]. According to a report from the Organisation for Economic Co-operation and Development [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e], the effectiveness of crisis management lies in fostering coordination between different government levels and clearly defining their respective roles. Such coordination can enhance coherent communication between provincial and federal authorities and result in clearer dissemination of government messages at all levels [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Effective communication, grounded in principles of coherence and clarity during a social crisis, is more likely to motivate individuals to support public policies, including preventive measures [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. As validated by Pelletier-Dumas et al. (2020), a heightened perception of clarity and coherence in the government's communication strategy was linked to increased compliance with preventive measures. Given the dynamic nature of the pandemic, clear and coherent communication became increasingly crucial as Canadians navigated the lifting measures period, marked by the uncertainty and novelty of the situation. Hence, individuals perceiving coherent and clear messages from both governance levels (federal and provincial) will be more likely to comply with preventive measures during the mandatory and lifting measures periods and transition or maintain high compliance behaviour during the transition period.\u003c/p\u003e \u003cp\u003eIn the literature, some studies considered the effect of socio-economic characteristics such as age, gender and level of education. It has been shown that older people are more likely to comply with preventive measures [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e]. Women [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] and university-level graduates [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] seem to follow the same trend, i.e. a higher level of compliance with preventive measures.\u003c/p\u003e \u003cp\u003e The current research aims to identify patterns of compliance with preventive measures, like social distancing and staying-at-home, during the mandatory measures and lifting measures periods in Canada. This project also investigates the dynamic nature of trajectory change as Canadians transitioned between periods. Finally, we examine the factors\u0026rsquo; association with trajectory and trajectory change to further our understanding of the dynamic nature of compliance during a sanitary crisis.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis\u003c/strong\u003e \u003cp\u003eThe first objective attempts to identify trajectories of compliance to preventive measures during the first wave, which can be divided into 2 distinct periods. The first period is associated with the mandatory imposition of measures (April 2020 - May 2020), while the second period represents the lifting of measures (May 2020 - July 2020).\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eIn response to this objective, we suppose that different trajectories of compliance will be identified during both periods (March 2020 - May 2020/ May 2020 - July 2020).\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAlso, based on previous studies [\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e] [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] [\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e], we suggest that the majority of Canadians will be grouped into trajectories of high levels of compliance during both periods of the first wave.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThe second objective relates to describing the transitions between both periods of the first wave in terms of group membership in compliance. It entails quantifying the percentage of Canadians transitioning from one group of compliance to another between periods (April 2020 - May 2020/ May 2020 - July 2020).\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWe expect to find some Canadians transitioning from one group of compliance to another between periods.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWe also expect that the majority of Canadians will preserve similar levels of compliance between both periods.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThe last objective aims to identify factors of compliance at baseline in April 2020 and during the transition period, which refers to the transition between the first and second periods of the first wave.\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAt baseline, we expect that there will be differences in compliance to preventive measures due to variables like perception of social norms, level of health literacy, perceived self-efficacy, level of institutional and scientific trust. Trajectories of high compliance at baseline will be associated with strong perception of social norms, high levels of health literacy, social support, perceived self-efficacy and trust in institutions and in science compared to the modal trajectory of compliance.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDuring the transition period, Canadians who initially recorded strong perceptions of social norms, high levels of health literacy, perceived self-efficacy and trust in science will preserve high and constant levels of compliance to preventive measures. Canadians who perceive low levels of social norms, present low levels of health literacy and low social support, higher perceived self-efficacy, and more trust in science will be more at risk of fluctuating between patterns of compliance during the transition period. Institutional trust will not be associated with transition movements between trajectories of compliance, as its impact on compliance to preventive measures is not consistent in time [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSample\u003c/h2\u003e \u003cp\u003eThis project is part of a larger initiative titled \"COVID-19 Canada: The end of the world as we know it?\". The underlying longitudinal survey, central to this project, was executed in partnership with the polling firm Delvinia, utilising the AskingCanadians survey panel, which comprises an extensive database of over one million Canadians. This longitudinal survey encompassed twelve time points spanning a duration of 2 years, from April 2020 to April 2022. However, for the purposes of this study, we focus exclusively on the initial six time points, extending from April 2020 to July 2020. Time was encoded by weeks to accommodate the varying intervals between each time point (see Supplementary Materials). In terms of sampling, the project \u0026ldquo;COVID-19: the end of the world as we know it?\u0026rdquo; accounted for 3617 participants in the first measurement time. This sample is non-probabilistic, employing the weighted quota method for participant selection. Upon further analysis of the representativeness of the sample, the researchers found that the sample was representative in terms of household composition, employment rate, immigration status. However, when the sample is compared with other socio-demographic characteristics, it is less representative for certain groups, such as Francophones, Canadians with lower levels of education and First Nations [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Attrition was observed from the second wave of data collection onward, with an average attrition rate of 43.35% (standard deviation = +/- 6.12%) for measurement time points two to twelve, ranging from a minimum of 34.50% to a maximum of 53.77%. For young people, the attrition rate seems to have been greater over time. To limit the effect of differential attrition according to certain individual characteristics, we incorporated weights based on demographic characteristics (e.g. age, gender, province of residence, household income, etc.) in the modelling of trajectories. Finally, in this project, we used full information maximum likelihood estimation in modelling trajectories of compliance with preventive measures [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe Latent Class Growth Analysis (LCGA) analysis typically necessitates a minimum of two measurement time points for each period (mandatory and lifting measures periods). Therefore, participants who did not provide a minimum of two responses during each period were excluded. This resulted in a final sample size of 1984 participants.\u003c/p\u003e \u003cp\u003eA preregistration of this study can be found here: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.17605/OSF.IO/R276S\u003c/span\u003e\u003cspan address=\"10.17605/OSF.IO/R276S\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. An initial analysis was conducted using the provided computation of variables, but collinearity issues were identified, necessitating additional attention. As a result, the format in which factors are presented in the subsequent sections reflects the final, resolved format. For more details, see supplementary materials.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eCompliance to preventive measures\u003c/h2\u003e \u003cp\u003eParticipants self-reported their compliance levels to social distancing and staying-at-home measures from April 2020 to July 2020 at six measurement times. They answered the following statements on a scale of 1 to 10 (never to always): \"Currently, how often do you do the following? (1) Maintain a distance of at least two metres (about two arm's lengths) from others when I am not at home; (2) stay home as much as I can. Since this study is the continuation of the Courdi et al. study (2023), we used the same items to assess compliance behaviour with social distancing and staying-at-home. For the multinomial logistic regression analysis, the trajectories were transformed as binary outcomes (0\u0026thinsp;=\u0026thinsp;stayed in a low trajectory or shifted to a lower trajectory; 1\u0026thinsp;=\u0026thinsp;remained in a high trajectory or shifted to a higher trajectory).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eFactors of trajectories and joint trajectories\u003c/h2\u003e \u003cp\u003eHealth literacy\u003c/p\u003e \u003cp\u003eThis variable refers to the ability of extracting and utilising information in a critical manner in a medical context [\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. We evaluated this ability in the context of the COVID-19 pandemic by using true and false statements from the survey\u0026rsquo;s first measurement time: \u0026ldquo;We would like to assess what Canadians know about COVID-19. Mark whether the following statements are true or false. (1) COVID-19 is a hoax. (2) I know what \u0026ldquo;maintaining a social distance\u0026rdquo; means. (3) Avoiding gatherings with large numbers of people helps prevent the spread of COVID-19. (4) Only the elderly can be seriously affected by COVID-19. (5) Healthy people should wear facemasks in public to avoid infection. (6) Staying home if you feel sick is important to prevent the spread of COVID-19. (7) COVID-19 is no more dangerous than the seasonal flu.\u0026rdquo;. Participants received one point if they provided the correct answer to the statement; otherwise, they were assigned zero points. We summed the points to evaluate the participants\u0026rsquo; level of health literacy. Due to the application of multiple imputation for this set of questions, not all participants were able to respond to every true and false statement. As a result, participants' health literacy levels were evaluated using a proportion (number of correct answers/number of statements answered), where a higher proportion score indicated a better health literacy level. The calculation of our Health Literacy variable was derived from the methodology employed in the study by Montagni et al. (2021). In their study, the authors aggregated participants' scores to determine the final level of health literacy.\u003c/p\u003e \u003cp\u003eSocial norms\u003c/p\u003e \u003cp\u003eWe interrogated Canadians on their perception of descriptive federal social norms by asking the following question at the first measurement time: \u0026ldquo;Most Canadians are following governments\u0026rsquo; measures recommendations concerning COVID-19.\u0026rdquo;. We also evaluated their perception of descriptive provincial social norms with this statement at the first measurement time: \u0026ldquo;In general, [Your province citizens] are complying with the governmental measures.\u0026rdquo; Both items are on a scale of 1\u0026ndash;10, 1 being completely disagree and 10 completely agreeing. Several COVID-19 studies have employed single items to assess perception of descriptive norms [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e] [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTrust\u003c/p\u003e \u003cp\u003eTo evaluate Canadians\u0026rsquo; levels of trust towards the scientific community and the government during the COVID-19 pandemic, we used the following question from the survey at the first measurement time: \u0026ldquo; How much do you trust each of the following actors to address the COVID-19 crisis?\u0026rdquo; (1) Canadian hospitals and health services (2) Prime Minister [Name] [Pipe in the name of the Prime Minister]\u0026rdquo;. The first item refers to trust in science, while the second relates to trust in government. Participants answered on a scale of 1\u0026ndash;10, 1 distrusting completely and 10 trusting completely. Numerous studies used a single item to evaluate trust in science and in government [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSelf-efficiency\u003c/p\u003e \u003cp\u003eThis variable refers to the perception of self-efficiency in protecting oneself and loved ones against the COVID-19 virus. It was evaluated with a single item at the first measurement time: \u0026ldquo;I do not know what is the best strategy to fight COVID-19.\u0026rdquo;. The item is on a scale of 1\u0026ndash;10, 1 being completely disagree and 10 being completely agree. We recoded the variable to establish a positive gradient, wherein a higher score reflects a greater sense of self-efficacy. A similar single item was used in another study [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e], where the authors studied the influence of Health Belief factors on compliance levels. In their study, they evaluated the participants\u0026rsquo; perception of self-efficacy in regards to the specific behaviour of social distancing. However, in our study, we asked the participants a more general statement since we focused on more than one behaviour of compliance to preventive measures.\u003c/p\u003e \u003cp\u003eClarity\u003c/p\u003e \u003cp\u003eCanadians\u0026rsquo; clear understanding of measures was assessed with a single item at the first measurement time with a 1 to 10 scale (1\u0026thinsp;=\u0026thinsp;Strongly disagree, 10\u0026thinsp;=\u0026thinsp;Strongly agree): \u0026ldquo; In general, I have a clear understanding of the various measures established by Canada\u0026rsquo;s public health agency.\u0026rdquo;. A similar item was already used in a previous article to evaluate perception of clarity in the message about measures [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eCoherence\u003c/p\u003e \u003cp\u003eThe participants' perception of self-reported coherence towards government messages was assessed with two items at the first measurement time with a 1 to 10 scale (1\u0026thinsp;=\u0026thinsp;Strongly disagree, 10\u0026thinsp;=\u0026thinsp;Strongly agree): \u0026ldquo; I am confused about the different recommendations coming from the federal and provincial governments and public health agencies. (2) The measures established by Canadian and provincial public health agencies are similar.\u0026rdquo;. We averaged both items to obtain a final coherence variable. In line with our study, other research endeavours have also employed a self-constructed coherence variable [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFor determining factors\u0026rsquo; association to trajectory and joint-trajectory membership, factors, assessed initially as continuous variables, were coded as binary variables. Indeed, responses falling within the first lowest quartile represented a low score, while responses outside the first quartile were considered as indicating a high score. For instance, Canadians who obtained a result within the first lower quartile that fell or was equivalent to 0.08 were considered to have a low level of health literacy, while others a high level of health literacy. The cutoffs varied depending on the variable, as some factors displayed less or more variation on the distribution. For the analyses, participants were assigned a value of 1 if they self-reported a high level of factors and 0 if they reported a low level. A similar approach was observed in the Kil and al. study (2023) to determine the high and low cutoffs. For more details on the threshold for every variable, see supplementary materials.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eAnalyses\u003c/h2\u003e \u003cp\u003eWe conducted Latent Class Growth Analysis (LCGA), or commonly referred as Group-based trajectory analysis, to discern longitudinal patterns of compliance behaviour to preventive measures, hence regrouping participants with similar trajectories together [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e] [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Since compliance to preventive measures was assessed as a continuous variable and displayed a normal distribution, a censored normal distribution (CNORM) was employed [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. In contrast to many other longitudinal methods, LCGA does not necessitate the assumption of homogeneity of variance between time points to be upheld. As suggested by Nagin (1999), the Bayesian Information Criterion (BIC) serves as our initial guide to determine the most suitable number of trajectory groups. Once we establish the number of trajectory groups for the final model, we proceed to choose the polynomial function for each trajectory\u0026mdash;whether constant, linear, quadratic, or cubic. The model with the best fit, as indicated by the BIC, is retained. During model selection, we also consider the participant distribution across trajectory groups, for it is crucial that the trajectories exhibit visually distinguishable patterns. The determination of the polynomial function for each trajectory is further refined based on a significance level (\u003cem\u003ep-value\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). After identifying longitudinal patterns of compliance to preventive measures (social distancing and staying at home) during each period (mandatory and lifting periods), we sought to identify transition movements between trajectories by conducting a Joint-Trajectory analysis, an extension of LCGA [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e] [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Similarly to the LCGA, the Joint-Trajectory procedure allows for modelization of trajectories over time. However, the latter distinguishes itself by linking trajectories of distinct but related behaviours. There are two approaches to Joint-trajectory analysis: comorbidity and heterotypic continuity. The first method allows for the estimation of probabilities for simultaneous membership in two distinct but related longitudinal patterns of behaviours, while the second method focuses on predicting the probabilities of the occurrence of trajectories for two behaviours expressed at different periods. In this study, we solely focused on the second approach, as it enabled us to identify transitional movements for trajectories of compliance to preventive measures between the mandatory and lifting measures periods. Finally, we used the RISK function from the PROC TRAJ package in the SAS software to study factors\u0026rsquo; association to trajectory membership for both preventive measures during the mandatory and lifting measures periods, while controlling for age, gender and level of education [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e] [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. The modal trajectory was chosen as the reference group, enabling us to discern high and low compliance behaviours from the majority. We also evaluated the factors\u0026rsquo; association to transition movements while performing a multinomial logistic regression analysis. The factors were entered as binary variables, and their reference level was low. Given the exploratory nature of this study and limited samples for trajectory movements, we incorporated a lower threshold (\u003cem\u003ep-value\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.10) to identify trends for potential exploration in future research for this section of the study. All visualisations were generated using the ggplot2 package in R.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCompliance trajectories to preventive measures\u003c/h2\u003e \u003cp\u003eIn observance of our first objective, we aimed to discern longitudinal patterns of compliance with preventive measures, specifically social distancing and staying at home. Upon conducting group-based trajectory analysis, we determined that three trajectories most accurately characterised our sample's compliance to both social distancing and staying at home during both periods (mandatory measures and lifting measures). This determination was based on the evaluation of BIC values and the significance of the parameters (see Supplementary Materials).\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e depicts the trajectories of the best-fitting models for the social distancing measure during both periods. For the mandatory measures period, we observed three trajectories characterised by constant, linear, and linear parameters (BIC=-7501.67), representing: a \u003cem\u003eLow and constant\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;116, 6.5%); a \u003cem\u003eHigh and decreasing\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;1060, 53.3%); and a \u003cem\u003eHigh and constant\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;722, 40.2%). Similarly, for the lifting measures period, three distinct trajectories were identified with linear, linear, and constant parameters (BIC= -8414.28): a \u003cem\u003eLow and decreasing\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;172, 9.3%); a \u003cem\u003eHigh and decreasing\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;1090, 54.4%); and a \u003cem\u003eHigh and constant\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;722, 36.3%).\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e illustrates the trajectories of the best-fitting models for the staying-at-home measure during both periods. For the mandatory measures period, we observed three trajectories characterised by constant, linear, and constant parameters (BIC= -7594.57), representing: a \u003cem\u003eLow and stable\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;167, 8.8%); a \u003cem\u003eHigh and decreasing\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;1026, 51.8%); and a \u003cem\u003eHigh and constant\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;791, 39.4%). Similarly, for the lifting measures period, three distinct trajectories were identified with linear, linear, and constant parameters (BIC=-9182.35): a \u003cem\u003eLow and decreasing\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;221, 11.7%); a \u003cem\u003eHigh and decreasing\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;1061, 53.2%); and a \u003cem\u003eHigh and constant\u003c/em\u003e trajectory (n\u0026thinsp;=\u0026thinsp;702, 35.1%).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eProbabilities of heterotypic continuity\u003c/h2\u003e \u003cp\u003eGiven membership to a trajectory during the mandatory measures period, we aim to estimate the probabilities of belonging to a trajectory during the lifting measures period, contingent upon the initial membership.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e displays the conditional probabilities of Canadians' compliance levels to social distancing during the lifting measures period based on their initial membership during the mandatory measures period. Generally, individuals exhibit stability from one period to another, as indicated by high probabilities along the diagonal. Specifically, 89.28% (n\u0026thinsp;=\u0026thinsp;106) of the sample remained in the lowest compliance trajectories for both periods; 90.44% (n\u0026thinsp;=\u0026thinsp;970) remained in the high and decreasing trajectories, and 85.85% (n\u0026thinsp;=\u0026thinsp;696) remained in the high and constant trajectories. The off-diagonal elements indicate that a small percentage of the population transitioned between periods. For example, 10.72% (n\u0026thinsp;=\u0026thinsp;10) of individuals initially in the \"Low and stable\" trajectory during the mandatory measures period shifted to the \"High and decreasing\" trajectory in the lifting measures period. In contrast, 3.38% (n\u0026thinsp;=\u0026thinsp;26) of Canadians transitioned from a \"High and decreasing\" to a \"High and constant\" trajectory. A substantial proportion of Canadians (13.70%, n\u0026thinsp;=\u0026thinsp;110) experienced a decline in compliance between periods, moving from a \"High and constant\" to a \"High and decreasing\" trajectory. Notably, only two individuals exhibited a substantial decrease in compliance levels, transitioning from a \"High and constant\" to a \"Low and decreasing\" trajectory. No transition movements were identified from the \"Low and stable\" to the \"High and constant\" trajectories.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e also presents the conditional probabilities of Canadians\u0026rsquo; compliance levels to staying-at-home during the lifting measures period based on their initial membership during the mandatory measures period. Similarly to social distancing probabilities, Canadians remain stable for the staying-at-home measure during both periods as indicated by the probabilities along the diagonal: 80.68% (n\u0026thinsp;=\u0026thinsp;134) of the population remained in the lowest compliance trajectory; 90.44% (n\u0026thinsp;=\u0026thinsp;970) in the \u0026ldquo;High and decreasing\u0026rdquo; trajectory; 85.83% (n\u0026thinsp;=\u0026thinsp;696) in the \u0026ldquo;High and constant\u0026rdquo; trajectory. In comparison, few individuals exhibited transition movements between trajectories: 19.32% (n\u0026thinsp;=\u0026thinsp;10) of Canadians previously in the \u0026ldquo;Low and stable\u0026rdquo; trajectory are now in the \u0026ldquo;High and decreasing\u0026rdquo; trajectory; 8.90% (n\u0026thinsp;=\u0026thinsp;87) of Canadians transitioned from \u0026ldquo;High and decreasing\u0026rdquo; to \u0026ldquo;Low and decreasing\u0026rdquo;; 3.39% (n\u0026thinsp;=\u0026thinsp;29) shifted from \u0026ldquo;High and decreasing to \u0026ldquo;High and constant\u0026rdquo;; finally a portion of Canadians (15.28%, n\u0026thinsp;=\u0026thinsp;118) displayed a decrease in compliance levels by shifting from \u0026ldquo;High and constant\u0026rdquo; to \u0026ldquo; High and decreasing\u0026rdquo;. Interestingly, we did not find any transition movements between \u0026ldquo;Low and stable\u0026rdquo;, \u0026ldquo;Low and decreasing\u0026rdquo; and \u0026ldquo;High and constant\u0026rdquo;.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePercentage (and Number) of individuals in each trajectory for social distancing measure, by period\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003e\u003cem\u003eLifting measures period\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eMandatory measures period\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eLow and decreasing\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eHigh and decreasing\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eHigh and constant\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eLow and constant\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e89.28 (106)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e10.72 (10)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.000 (0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eHigh and decreasing\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e6.18 (64)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e90.44 (970)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e3.38 (26)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eHigh and constant\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.472 (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e13.70 (110)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e85.83 (696)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePercentage (and Number) of individuals in each trajectory for staying-at-home measure, by period\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003e\u003cem\u003eLifting measures period\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eMandatory measures period\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eLow and decreasing\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eHigh and decreasing\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eHigh and constant\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eLow and constant\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e80.68 (134)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e19.32 (33)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.000 (0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eHigh and decreasing\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e8.90 (87)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e87.71 (910)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e3.39 (29)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eHigh and constant\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.000 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e15.28 (118)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e84.71 (673)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003ePredicting trajectory membership\u003c/h2\u003e \u003cp\u003eWe then used the RISK function from PROC TRAJ to investigate the predictive value of social and individual characteristics in predicting trajectory membership. This analysis included controls for age, gender, and level of education, and independent variables were coded as high and low (reference group). Tables\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e present the odds of factors associated with membership in trajectories for social distancing and staying-at-home measures.\u003c/p\u003e \u003cp\u003eSocial distancing\u003c/p\u003e \u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Canadians exhibited greater likelihood of being in the \"High and constant\" social distancing group compared to the \"High and decreasing\" social distancing group when possessing high health literacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.67; 95% \u003cem\u003eCI\u003c/em\u003e [1.20, 2.32]; \u003cem\u003ep\u0026thinsp;=\u003c/em\u003e\u0026thinsp;.005) during the mandatory measures period. Likewise, they were more likely to have higher levels of perceptions of provincial social norms (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.86; 95% \u003cem\u003eCI\u003c/em\u003e [1.44, 2.39]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), trust in science (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.86; 95% \u003cem\u003eCI\u003c/em\u003e [1.44, 2.39]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001) and clarity (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.34; 95% \u003cem\u003eCI\u003c/em\u003e [1.06, 5.17]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). The \u0026ldquo;Low and constant\u0026rdquo; group, in comparison to the \u0026ldquo;High and decreasing\u0026rdquo; group, was associated with lesser likelihood of high perception of provincial social norms (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.27; 95% \u003cem\u003eCI\u003c/em\u003e [.10, .70] ; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.027) and trust in science (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.19; 95% \u003cem\u003eCI\u003c/em\u003e [.09, .42]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001).\u003c/p\u003e \u003cp\u003eDuring the lifting period, as outlined in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Canadians who registered high levels of health literacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.43; 95% \u003cem\u003eCI\u003c/em\u003e [.93, 2.37]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.048), perceived self-efficacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.51; 95% \u003cem\u003eCI\u003c/em\u003e [.89, 2.35]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.020), perceptions of provincial social norms (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.80; 95% \u003cem\u003eCI\u003c/em\u003e [1.00, 3.67]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.001), trust in science (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.54;95% \u003cem\u003eCI\u003c/em\u003e [1.08, 3.69]; \u003cem\u003ep\u0026thinsp;=\u003c/em\u003e\u0026thinsp;.014) and clarity (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.16; 95% \u003cem\u003eCI\u003c/em\u003e [1.58, 4.92]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001) were more likely associated with the \u0026ldquo;High and constant\u0026rdquo; social distancing trajectory when compared to the modal reference group. In comparison to the \u0026ldquo;High and decreasing group, individuals with high health literacy levels (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.33; 95% \u003cem\u003eCI\u003c/em\u003e [.15, .54]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), high trust in science (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.51; 95% \u003cem\u003eCI\u003c/em\u003e [.32, .91]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.040) and high trust in government (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.45; 95% \u003cem\u003eCI\u003c/em\u003e [.35, 1.02]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.007) are less likely to be found in the \u0026ldquo;Low and decreasing\u0026rdquo; trajectory.\u003c/p\u003e \u003cp\u003eStaying-at-home\u003c/p\u003e \u003cp\u003eAs depicted in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, individuals who indicated elevated levels of health literacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.75; \u003cem\u003eCI\u003c/em\u003e [1.24, 2.56]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.031), trust in science (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.67; 95% \u003cem\u003eCI\u003c/em\u003e [1.13, 2.36]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.006), clarity (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.23; 95% \u003cem\u003eCI\u003c/em\u003e [1.53, 4.46]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), and coherence (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.70; 95% \u003cem\u003eCI\u003c/em\u003e [.22, .97]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.008) were more inclined to be associated with the \"High and constant\" staying-at-home trajectory compared to the \"High and decreasing\" group during the mandatory measures period. In contrast, Canadians found in the \u0026ldquo;Low and constant\u0026rdquo; group were less likely to possess high levels of health literacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.51; 95% \u003cem\u003eCI\u003c/em\u003e [0.21, 1.01]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.031), trust in government (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.39; 95% \u003cem\u003eCI\u003c/em\u003e [0.26, 0.94]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003).\u003c/p\u003e \u003cp\u003eDuring the lifting period, as presented by Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, belonging to the \"High and constant\" group, as opposed to the \"High and decreasing\" group, is associated with high levels of health literacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.90; 95% \u003cem\u003eCI\u003c/em\u003e [0.26, 1.01]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.008), perception of provincial social norms (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.63; 95% \u003cem\u003eCI\u003c/em\u003e [1.06, 3.82]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.020), trust in science (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.80; 95% \u003cem\u003eCI\u003c/em\u003e [1.11, 2.57]; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002) and clarity (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.05; 95% \u003cem\u003eCI\u003c/em\u003e [1.41, 4.59] ; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003 ). On the other hand, Canadians who were associated with the \u0026ldquo;Low and decreasing\u0026rdquo; group when compared to the \u0026ldquo;High and decreasing\u0026rdquo; group were less likely to record high levels of health literacy (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.46; 95% \u003cem\u003eCI\u003c/em\u003e [0.26, 1.01] ; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.008 ) and trust in government (\u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.38; 95% \u003cem\u003eCI\u003c/em\u003e [0.35, 0.69] ; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.004 ).\u003c/p\u003e \u003cp\u003eSociodemographic variables\u003c/p\u003e \u003cp\u003eAge, gender and level of education were reported as inconsistent factors of membership to trajectories of social distancing and staying-at-home, as displayed in Tables\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003ch4\u003e\u003cstrong\u003eTable 3.\u003c/strong\u003e Factors of trajectory membership for social distancing during the mandatory and lifting measures periods\u003c/h4\u003e\n\u003cp\u003e\u003cimg width=\"865\" src=\"https://myfiles.space/user_files/127393_c7e80a1c9bb65875/127393_custom_files/img1716360035.png\" alt=\"image\"\u003e\u003c/p\u003e\n\u003ch4\u003e\u003cstrong\u003eTable 4.\u003c/strong\u003e Factors of trajectory membership for staying-at-home during the mandatory and lifting measures periods\u003c/h4\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cimg width=\"864\" src=\"https://myfiles.space/user_files/127393_c7e80a1c9bb65875/127393_custom_files/img1716360036.png\" alt=\"image\"\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003ch3\u003ePredicting trajectory change\u003c/h3\u003e\n\u003cp\u003eWe employed multinomial logistic regression to investigate the predictive role of social and individual characteristics in determining trajectory change membership. Three contrasts were deemed feasible for the analysis due to the rare occurrence of trajectory change: \u0026ldquo;Low and constant\u0026rdquo; to \u0026ldquo;Low and decreasing\u0026rdquo; (0) vs \u0026ldquo;Low and decreasing\u0026rdquo; to \u0026ldquo;High and decreasing\u0026rdquo; (1); \u0026ldquo;High and decreasing\u0026rdquo; to \u0026ldquo;Low and constant\u0026rdquo; (0) vs \u0026ldquo;High and decreasing\u0026rdquo; to \u0026ldquo;High and constant\u0026rdquo; (1); and \u0026ldquo;High and constant\u0026rdquo; to \u0026ldquo;High and decreasing\u0026rdquo; (0) vs remained in \u0026ldquo;High and constant\u0026rdquo; (1). \u003cstrong\u003eTables 5\u003c/strong\u003e and \u003cstrong\u003e6\u0026nbsp;\u003c/strong\u003eillustrate the results of the multinomial logistic regression in relation to trajectory changes.\u0026nbsp;\u003c/p\u003e\n\u003ch4\u003eSocial distancing\u003c/h4\u003e\n\u003cp\u003eAs shown in \u003cstrong\u003eTable 5\u003c/strong\u003e, only a limited number of factors appear to forecast trajectory changes across all contrasts. For instance, Canadians who exhibited high levels of perception of provincial social norms (\u003cem\u003eOdds\u003c/em\u003e= 4.10; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[1.30, 17.54]\u003cem\u003e\u0026nbsp;\u003c/em\u003e; \u003cem\u003ep\u003c/em\u003e= 0.027 ) and trust in science (\u003cem\u003eOdds\u003c/em\u003e= 3.57; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[1.29, 10.91]\u003cem\u003e\u0026nbsp;\u003c/em\u003e; \u003cem\u003ep\u003c/em\u003e=0.013 ) were more likely to transition from \u0026ldquo;Low and constant\u0026rdquo; to \u0026ldquo;High and decreasing\u0026rdquo; trajectories during the lifting measures period, unlike those who reported high clarity of messages (\u003cem\u003eOdds\u003c/em\u003e= 0.23; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[0.04, 0.87]\u003cem\u003e\u0026nbsp;\u003c/em\u003e; \u003cem\u003ep=\u003c/em\u003e0.056 ). On the other hand, Canadians were more prone to transition from the \u0026quot;High and decreasing\u0026quot; to the \u0026quot;High and constant\u0026quot; pattern during the lifting period, as opposed to the transition from \u0026quot;High and decreasing\u0026quot; to \u0026quot;Low and decreasing,\u0026quot; if they indicated higher levels of trust in science (\u003cem\u003eOdds\u003c/em\u003e=3.00; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[1.26, 7.96]\u003cem\u003e\u0026nbsp;\u003c/em\u003e; \u003cem\u003ep=\u003c/em\u003e 0.014) and perception of federal social norms (\u003cem\u003eOdds\u003c/em\u003e= 2.48; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[0.98, 6.94]; \u003cem\u003ep=\u003c/em\u003e0.061 ). However, greater perception of social norms reduces the likelihood of transitioning from the \u0026quot;High and decreasing\u0026quot; to the \u0026quot;High and constant\u0026quot; pattern during the transition period (\u003cem\u003eOdds\u003c/em\u003e= 0.38; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[0.12, 1.04]\u003cem\u003e\u0026nbsp;\u003c/em\u003e; \u003cem\u003ep\u003c/em\u003e= 0.063).\u0026nbsp;\u003c/p\u003e\n\u003ch4\u003eStaying-at-home\u003c/h4\u003e\n\u003cp\u003e\u003cstrong\u003eTable 6\u0026nbsp;\u003c/strong\u003ereveals that shifting from a \u0026ldquo;Low and constant\u0026rdquo; to a \u0026ldquo;High and decreasing\u0026rdquo; trajectory is more likely when individuals report high trust in government (\u003cem\u003eOdds\u003c/em\u003e= 2.08; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[1.10, 4.11]\u003cem\u003e\u0026nbsp;\u003c/em\u003e; \u003cem\u003ep\u003c/em\u003e= 0.028). Furthermore, an increase in compliance (\u0026ldquo;High and decreasing\u0026rdquo; to \u0026ldquo;High and constant\u0026rdquo;) as opposed to a decline in compliance (\u0026quot;High and decreasing\u0026quot; to \u0026quot;High and constant\u0026quot;) is more likely when reporting high perception of social norms (\u003cem\u003eOdds\u003c/em\u003e= 2.22; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[1.13, 4.51]; \u003cem\u003ep\u003c/em\u003e=0.022). Finally, maintaining membership in \u0026quot;High and constant\u0026quot; trajectories during both periods is more probable when possessing high perception of self-efficacy (\u003cem\u003eOdds\u003c/em\u003e= 1.30;\u003cem\u003e\u0026nbsp;\u003c/em\u003e95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[0.98, 1.74]; \u003cem\u003ep\u003c/em\u003e=0.070), as opposed to those reporting high perception of provincial social norms (\u003cem\u003eOdds\u003c/em\u003e= 0.74; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[0.55, 1.00]; \u003cem\u003ep\u003c/em\u003e=0,051) and trust in science (\u003cem\u003eOdds\u003c/em\u003e= 0.77; 95% \u003cem\u003eCI\u0026nbsp;\u003c/em\u003e[0.58, 1.02]; \u003cem\u003ep\u003c/em\u003e=0,071).\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eSociodemographic variables\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eTransitioning from a \u0026quot;High and decreasing\u0026quot; trajectory to a \u0026quot;High and constant\u0026quot; trajectory was significantly associated with being over 65 years of age (\u003cem\u003eOdds\u003c/em\u003e = 4.00; \u003cem\u003e95% CI\u003c/em\u003e [0.55, 1.00]; \u003cem\u003ep\u003c/em\u003e = 0.051) and having a university-level education (\u003cem\u003eOdds\u003c/em\u003e = 2.58; \u003cem\u003e95% CI\u003c/em\u003e [1.18, 5.94]; \u003cem\u003ep\u003c/em\u003e = 0.022).\u003c/p\u003e\n\u003ch4\u003e\u003cstrong\u003eTable 5.\u003c/strong\u003e Factors of trajectory change for social distancing measure during the first wave\u0026nbsp;\u003c/h4\u003e\n\u003cp\u003e\u003cimg width=\"864\" src=\"https://myfiles.space/user_files/127393_c7e80a1c9bb65875/127393_custom_files/img171636003668.png\" alt=\"image\"\u003e\u003c/p\u003e\n\u003ch4\u003e\u003cstrong\u003eTable 6.\u0026nbsp;\u003c/strong\u003eFactors of trajectory change for staying-at-home measure during the first wave\u0026nbsp;\u003c/h4\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cimg width=\"864\" src=\"https://myfiles.space/user_files/127393_c7e80a1c9bb65875/127393_custom_files/img171636003611.png\" alt=\"image\"\u003e\u003c/strong\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study aimed to identify and understand compliance trajectories with preventive measures during the mandatory and lifting measures periods of the COVID-19 pandemic, along with their factors. In both periods, three distinct compliance trajectories were discerned for social distancing and staying-at-home measures. Notably, a large proportion of Canadians exhibited high compliance, with 93.5% for social distancing and 91.2% for staying-at-home during mandatory measures, and 90.7% for social distancing and 88.3% for staying-at-home during lifting measures. The majority of Canadians maintained their compliance patterns across trajectories during both periods, with 89.28% staying in the lowest trajectory for social distancing and 80.68% for staying-at-home. Health literacy, perception of provincial social norms, trust in science, trust in government, and perception of clarity consistently emerged as significant factors for trajectory membership in both preventive measures. Perceived self-efficacy exclusively predicted membership in social distancing trajectories, while inconsistent predictions were observed for perception of coherence and federal social norms. In terms of trajectory changes, perception of provincial social norms and trust in science frequently emerged as significant factors of increasing or maintaining high compliance behavior. Overall, the study provides valuable insights into the dynamics of compliance with preventive measures, emphasising the role of various factors in shaping individual trajectories and changes in behaviour.\u003c/p\u003e \u003cp\u003eIn line with our first objective, we posited that distinct compliance patterns would emerge for both social distancing and staying-at-home measures during the mandatory and lifting periods, with the expectation that a significant majority of Canadians would report high levels of compliance. These initial hypotheses were validated, as our findings unveiled three distinct trajectories during each period. During the mandatory measures period, the lowest compliance trajectory for both social distancing and staying-at-home measures, labelled as \u003cem\u003eLow and constant\u003c/em\u003e, exhibited stability with an initially lower compliance level. As the measures were lifted, this lowest compliance trajectory underwent a transformation, adopting a declining linear trend. This shift in trend implies that situational factors, such as the easing of measures, might negatively impact the motivation to comply for individuals already at risk of non-compliance. In contrast to findings from other longitudinal studies [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e], where the lowest compliance trajectory exhibited a decline from the onset of the pandemic, our trajectory modelling reveals a distinct pattern\u0026mdash;a decline in compliance commencing as the measures were lifted for this group. This differentiation is crucial, as it enhances the precision of our understanding of longitudinal compliance patterns by scrutinising specific periods more susceptible to influencing trajectory trends. During both mandatory and lifting measures periods, we observed a group, called \u003cem\u003eHigh and decreasing\u003c/em\u003e, that reported a decrease in compliance with an initial high level of compliance with preventive measures. Interestingly, this predominant group, encompassing the majority of Canadians, appeared unaffected by the imposition or lifting of measures; they simply adopted the necessary behaviours and gradually exhibited a waning interest in maintaining these practices. A similar characterization was noted in previous LCGA studies [\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e] [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e] [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e]. Existing literature has proposed the concept of pandemic fatigue as a potential explanation for this decline in compliance levels, positing that sustaining beneficial yet demanding behaviours, such as social distancing and staying-at-home, becomes progressively challenging compared to initially adopting them [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Given the unconventional nature of the novel preventive measures, it seems justifiable that a single wave of a pandemic may not suffice for complete adoption and perpetuation, as integrating a new behaviour into routine requires time. This aligns with observations in other health-related behaviours, like dietary regimens, fitness programs, or medication plans, where the motivation to uphold costly and unnatural behaviours tends to diminish over time [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Lastly, the \u003cem\u003eHigh and constant\u003c/em\u003e group exhibited constant patterns of high compliance during both the mandatory and lifting measures periods, seemingly impervious to the changing situation. Overall, our findings align with previous Canadian longitudinal studies on compliance with preventive measures. However, we propose that the initial decline observed in these studies may not solely be attributed to a changing situation, as hypothesised, but also to psychological fatigue, depending on the group.\u003c/p\u003e \u003cp\u003eWhile executing Latent Class Growth Analysis, we observed variations in the proportion of Canadians in each trajectory between the mandatory and lifting measures periods. For example, in regard to the social distancing measure, we identified that 40.2% of individuals were in the \u003cem\u003eHigh and constant\u003c/em\u003e trajectory during the mandatory period, but this number decreased to 36.3% during the lifting period. Likewise, in relation to the staying-at-home measure, we noted a decrease in membership for the highest compliance trajectory (\u003cem\u003eHigh and constant\u003c/em\u003e), declining from 39.4% during the mandatory period to 35.1% during the lifting period. Given these variations in membership proportions, our second objective sought to describe the transition movements between both periods of the first pandemic wave in terms of group compliance membership. We found that the majority of Canadians preserved a similar pattern of compliance during both periods, as suggested by the probabilities along the diagonals (between 85.83% \u0026minus;\u0026thinsp;90.44% for social distancing and between 80.68% \u0026minus;\u0026thinsp;87.71% for staying-at-home). In addition to confirming our initial hypotheses, these findings highlight the importance of the initial compliance level in predicting future compliance patterns, especially in the dynamic context of an evolving pandemic. Indeed, to our knowledge, this is the first study to examine stability within a group of compliance, emphasising the need for governments to implement their interventions in the initial weeks of a health crisis to improve compliance with measures in the long term. Furthermore, this noticeable stability in compliance membership between periods informs us that a change in behaviour is not as affected by situational factors like the easing of preventive measures, but is rather influenced by already present inter-individual differences.\u003c/p\u003e \u003cp\u003eFurthermore, we noticed a general trend where more Canadians transitioned to a lower compliance trajectory rather than high, as the measures were lifted. More specifically, we noted that 13.70% (n\u0026thinsp;=\u0026thinsp;110) of Canadians transitioned from a \u003cem\u003eHigh and constant\u003c/em\u003e to a \u003cem\u003eHigh and decreasing\u003c/em\u003e social distancing trajectory, and a similar transition was observed for staying-at-home, accounting for 15.28% (n\u0026thinsp;=\u0026thinsp;118). This movement, characterised by a decrease in compliance, represented the most substantial transition between trajectories in our study in terms of number, followed by the transition from \u003cem\u003eHigh and decreasing\u003c/em\u003e to \u003cem\u003eLow and constant.\u003c/em\u003e Therefore, our results suggest that transition movements are more likely to be found within the higher compliance groups, such as \u003cem\u003eHigh and constant\u003c/em\u003e and \u003cem\u003eHigh and decreasing\u003c/em\u003e. In prior LCGA studies examining compliance with preventive measures [\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e] [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e] [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], the researchers documented fluctuating compliance behaviour during the COVID-19 pandemic across nearly all trajectories, except for the trajectory associated with the highest levels of compliance. This group, identified as \u003cem\u003eClass 1\u003c/em\u003e in Wright et al. (2022), \u003cem\u003eHigh and constant\u003c/em\u003e in Courdi et al. (2023), and \u003cem\u003eHigh adherers\u003c/em\u003e in MacNeil et al. (2022), exhibited a consistent compliance level, remaining unaffected by the various waves of the virus. In contrast, the lowest compliance trajectory, designated as \u003cem\u003eClass 4\u003c/em\u003e in Wright et al. (2021), \u003cem\u003eFast Decliners\u003c/em\u003e in MacNeil et al. (2022), and \u003cem\u003eLow and fluctuating\u003c/em\u003e in Courdi et al. (2023), was characterised by significant variability dependent on the prevailing pandemic situation. However, according to our findings, we identified two distinct forms of variability: one pertains to fluctuations in the trend of a pattern, as exemplified by the change in trend of the lowest compliance trajectory between periods, while the other involves variations in membership within compliance groups, such as the \u003cem\u003eHigh and constant\u003c/em\u003e group in our study. This differentiation has not been addressed in studies on compliance behaviour with preventive measures, challenging our conventional perception of stability associated with higher compliance groups. Hence, situational factors, such as the lifting of measures, appear to impact groups disparately, emphasising the importance of elaborating intervention plans tailored to the specific characteristics of each group and not just the lowest trajectory group. Every individual transitioning from a high compliance group to a lower one poses as much of a risk as individuals in the lowest compliance group, as they too are susceptible to finding themselves in these low trajectories, subsequently affecting infection rates.\u003c/p\u003e \u003cp\u003eAs mentioned, the proportions along the diagonals in terms of trajectory membership suppose a stability in compliance behaviour throughout the first wave of the pandemic, suggesting that inter-individual differences might be at cause and not situational factors. Our findings consistently demonstrate that health literacy alone serves as a reliable factor of trajectory membership for both social distancing and staying-at-home behaviours across both observed periods. Aligned with existing COVID-19 literature, health literacy\u0026mdash;encompassing the ability to comprehend and critically assess health-related information\u0026mdash;stands validated as a pivotal individual factor of compliance with preventive measures, particularly amid the challenges posed by the Infodemic [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. However, contrary to our expectations, health literacy emerged as a factor solely for trajectory membership and not for change in membership. This surprising outcome suggests a certain resilience and stability in the concept of health literacy, impervious to dynamic situations such as the easing of preventive measures. When delving into the factors contributing to a high level of health literacy, it becomes evident that it is largely anticipated by constant sociodemographic attributes such as education, income and migration status, as highlighted in the work of Martin et al. (2009). Given the relatively stable nature of these factors and their gradual evolution over an extended period, it could elucidate why health literacy proves to be a more effective factor of the stability of trajectory membership [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Our results further imply that health literacy plays a role in shaping the overall compliance pattern within a group characterised by shared stable attributes, as seen in the lowest trajectory groups in our study (\u003cem\u003eLow and constant\u003c/em\u003e to \u003cem\u003eLow and decreasing\u003c/em\u003e). Indeed, the probability of belonging in the lowest trajectory significantly decreased when displaying high health literacy levels between the observed periods. With the government initiating the easing of measures, individuals who were less inclined to exhibit high health literacy tended to persist within the same trajectory, but the pattern of their trajectory underwent a shift, adopting a decreasing trend. Therefore, individual factors, such as health literacy, appear to be more adept at forecasting stability within a group while also capturing variations in trends.\u003c/p\u003e \u003cp\u003e We observed another type of variability in our study, that challenged the conventional perception of stability associated with high compliance groups. Trajectory change, observed mostly within the \u003cem\u003eHigh and constant\u003c/em\u003e and \u003cem\u003eHigh and decreasing\u003c/em\u003e groups, was moderately predicted by social factors, like perception of provincial social norms and trust in science. Social norms, defined by a set of rules and expectations in terms of socially acceptable behaviours and attitudes, were often studied in relation to compliance with preventive measures, as they elicit long term changes in behaviour and target large groups [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. In affirming prior research emphasising the pivotal role of social norms in predicting compliance behaviour [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e], our study goes further and proposes that social norms may also serve as a factor for changes in trajectory. In accordance with Reynolds et al. (2015)\u0026rsquo;s comprehensive framework on behaviour change, individuals from the same province, acting as referent points, play a crucial role in shaping attitudes and behaviour toward compliance. The salience of this provincial identity may have diminished as the situation evolved with the easing of measures. Notably, during the initial months of the pandemic, most provincial governments conveyed messages of encouragement, emphasising the imperative of collective action within the province to limit the virus's spread [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. As the measures were lifted, the salience of this provincial identity waned in the public discourse, potentially explaining the behavioural shifts observed among some Canadians. In contrast, individuals whose provincial identity was initially salient were more likely to comply with the highest compliance trajectory or transition to it even amid the easing of measures. Noteworthy, our results suggest the relevance of the provincial identity in predicting compliance behaviour, compared to national identity, positing that the government should prioritise norm-based interventions at state levels. Furthermore, our study revealed that trust in science, rather than trust in the government, emerged as a significant factor of trajectory change, thus affirming our initial hypothesis. Trust in science encompasses confidence in the policies implemented and decisions made by the scientific community to safeguard against emerging threats [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. Throughout the pandemic, scientists were entrusted by governments to devise optimal strategies for curbing the virus's spread. However, as new information surfaced regarding the virus's contagiousness, policies underwent adjustments, such as the introduction of mask mandates [\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e]. Our findings suggest that individuals with higher levels of trust in science were not only more likely to comply with recommended behaviour but also demonstrated a propensity to either maintain or transition to higher compliance patterns. This adaptability could stem from their capacity to assimilate novel information about the virus communicated by the scientific community and understand the reasoning behind these policies.\u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eStrengths and limitations\u003c/h2\u003e \u003cp\u003eTo our knowledge, this study stands as the first to explore the dynamic nature of behaviour compliance with preventive measures during the COVID-19 pandemic and its associated factors, through Joint-Trajectory analysis. Apart from its innovative methodology, the study boasts several strengths, including its longitudinal design, a representative sample drawn from the Canadian population, short intervals between survey periods during the initial wave, and measurement times spanning both mandatory and lifting measure periods. These attributes collectively enhance the study's robustness and reliability. However, it is crucial to interpret our results in consideration of several limitations. First and foremost, compliance behaviours such as social distancing and staying-at-home were evaluated through self-reported scores, introducing a potential impact on the reproducibility of our findings. While it is noteworthy that many studies on the COVID-19 pandemic have relied on self-reported compliance measures [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] [\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e] [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e], we advocate for the incorporation of external measures, such as mobile-phone mobility data [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e], to mitigate desirability bias. Additionally, our factors were predominantly assessed using single homemade self-reported items, limiting the comprehensive evaluation of the various dimensions inherent in a complex concept like social norms and challenging the validity of our measures. Lastly, it is essential to exercise caution when interpreting our findings related to factors of trajectory change, as some comparisons involved relatively small sample sizes (e.g., 106 vs. 10).\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eLatent Class Growth Analysis proves invaluable in assessing trends in compliance behaviour with preventive measures, allowing for the identification of inter-individual differences and variability within a group. Through this analysis, we discerned three distinct trajectories during both mandatory and lifting measures periods: \u003cem\u003eLow and constant\u003c/em\u003e (\u003cem\u003edecreasing\u003c/em\u003e during the lifting measures period), \u003cem\u003eHigh and decreasing\u003c/em\u003e, and \u003cem\u003eHigh and constant\u003c/em\u003e. Expanding our comprehension of compliance behaviour, we delved into transitions between these trajectories and found that the majority of individuals tended to remain within similar groups. Contrary to the conventional belief in the stability of high compliance groups, our findings challenge this assumption, revealing that these groups were the most susceptible to transitioning to lower compliance categories. This insight bears significant implications for policies related to compliance behaviour, suggesting that vulnerability to decreased compliance is universal, albeit manifesting differently across various groups.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics approval and consent to participate\u003c/p\u003e\n\u003cp\u003eThe authors declare that all experimental protocols were approved by Comit\u0026eacute; d\u0026apos;\u0026eacute;thique de la recherche en \u0026eacute;ducation et en psychologie (CEREP) of University of Montreal (Certificat no CEREP-20-038-D).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConsent for publication\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAvailability of data and material\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis study is part of a larger national research project financed by the Canadian Institutes of Health Research (CIHR) that examines the social consequences of the COVID-19 pandemic on Canadians (grant number = 170633, for more information, see https://csdc-cecd.wixsite.com/covid19csi?lang=en). This study is also part of the Mathematics for Public Health program funded by the joint Natural Sciences and Engineering Research Council (NSERC) and \u0026nbsp; Public Health Agency of Canada (PHAC) Emerging Infectious Disease Modeling initiative.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAuthors\u0026rsquo; contribution\u003c/p\u003e\n\u003cp\u003eS.R.A. performed the data analysis and wrote the first drafts of the manuscript. E.L., M.P.D., J.M.L. and J.B. assisted in the operationalization of the research question, reviewed the methodology and the manuscript. R.d.l.S provided the research question and assisted at all stages of the study.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eArag\u0026atilde;o DP, dos Santos DH, Mondini A, Gon\u0026ccedil;alves LMG. National Holidays and Social Mobility Behaviors: Alternatives for Forecasting COVID-19 Deaths in Brazil. Int J Environ Res Public Health. 2021;18(21):11595. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph182111595\u003c/span\u003e\u003cspan address=\"10.3390/ijerph182111595\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBattiston P, Kashyap R, Rotondi V. Reliance on scientists and experts during an epidemic: Evidence from the COVID-19 outbreak in Italy. SSM - Popul Health. 2021;13:100721. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ssmph.2020.100721\u003c/span\u003e\u003cspan address=\"10.1016/j.ssmph.2020.100721\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBicchieri C, Fatas E, Aldama A, Casas A, Deshpande I, Lauro M, Parilli C, Spohn M, Pereira P, Wen R. In science we (should) trust: Expectations and compliance across nine countries during the COVID-19 pandemic. PLoS ONE. 2021;16(6):e0252892. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0252892\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0252892\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBirkh\u0026auml;uer J, Gaab J, Kossowsky J, Hasler S, Krummenacher P, Werner C, Gerger H. Trust in the health care professional and health outcome: A meta-analysis. PLoS ONE. 2017;12(2):e0170988. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0170988\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0170988\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCabrera-\u0026Aacute;lvarez P, Hornsey MJ, Lobera J. Determinants of self-reported adherence to COVID-19 regulations in Spain: social norms, trust and risk perception. Health Promot Int. 2022;37(6):daac138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/heapro/daac138\u003c/span\u003e\u003cspan address=\"10.1093/heapro/daac138\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCanadian Council on Learning. (2008). \u003cem\u003eHealth Literacy in Canada: A Healthy Understanding\u003c/em\u003e (p. 38). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.vec.ca/pdfs/2012-39.pdf\u003c/span\u003e\u003cspan address=\"https://www.vec.ca/pdfs/2012-39.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCanadian Institute for Health Information. (2022). \u003cem\u003eCanadian Data Set of COVID-19 Interventions \u0026mdash; Data Tables\u003c/em\u003e. Canadian Institute for Health Information. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cihi.ca/en/canadian-covid-19-intervention-timeline\u003c/span\u003e\u003cspan address=\"https://www.cihi.ca/en/canadian-covid-19-intervention-timeline\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaron-Diotte M, Dorfman A, Pelletier-Dumas M, Lacourse \u0026Eacute;, Lina JM, Stolle D, Taylor DM, de la Sablonni\u0026egrave;re R. (2021). COVID-19 Canada: The end of the world as we know it? Technical Report No. 2. Handling planned and unplanned missing data. Universit\u0026eacute; de Montr\u0026eacute;al.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaron-Diotte M, de la Sablonni\u0026egrave;re P-D. Roxane (2023) Handling Planned and Unplanned Missing Data in a Longitudinal Study, The Quantitative Methods for Psychology, 19(2), 123\u0026ndash;135. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.20982/tqmp.19.2.p123\u003c/span\u003e\u003cspan address=\"10.20982/tqmp.19.2.p123\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCBC. (2023, November 14). \u003cem\u003eBillboard campaign aims to dispel mistrust of vaccines in African Nova Scotian community | CBC News\u003c/em\u003e. CBC. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cbc.ca/news/canada/nova-scotia/billboard-campaign-vaccine-mistrust-african-nova-scotia-1.7027975\u003c/span\u003e\u003cspan address=\"https://www.cbc.ca/news/canada/nova-scotia/billboard-campaign-vaccine-mistrust-african-nova-scotia-1.7027975\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCDC. (2023, March 15). \u003cem\u003eCDC Museum COVID-19 Timeline\u003c/em\u003e. Centers for Disease Control and Prevention. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cdc.gov/museum/timeline/covid19.html\u003c/span\u003e\u003cspan address=\"https://www.cdc.gov/museum/timeline/covid19.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen AMH, Armbruster AL, Buckley B, Campbell JA, Dang DK, Devraj R, Drame I, Edwards A, Haack SL, Ma Q, Petry N, Planas LG, Sadowski CA, Santee J, Wade L, Borja-Hart N. Inclusion of Health Disparities, Cultural Competence, and Health Literacy Content in US and Canadian Pharmacy Curriculums. Am J Pharm Educ. 2021;85(1):8200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5688/ajpe8200\u003c/span\u003e\u003cspan address=\"10.5688/ajpe8200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen X, Zhang A, Wang H, Gallaher A, Zhu X. Compliance and containment in social distancing: mathematical modeling of COVID-19 across townships. Int J Geogr Inf Sci. 2021;35(3):446\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13658816.2021.1873999\u003c/span\u003e\u003cspan address=\"10.1080/13658816.2021.1873999\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eC\u0026ocirc;t\u0026eacute; SM, Vaillancourt T, Barker ED, Nagin D, Tremblay RE. The joint development of physical and indirect aggression: Predictors of continuity and change during childhood. Dev Psychopathol. 2007;19(1):37\u0026ndash;55. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0954579407070034\u003c/span\u003e\u003cspan address=\"10.1017/S0954579407070034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCourdi C, Ali R, Pelletier-Dumas S, Stolle M, Dorfman D, Lina A, Lacourse J-M, \u0026Eacute;., de la Sablonni\u0026egrave;re R. How level of understanding and type of used sources relate to adherence to COVID-19 public health measures in Canada. Sci Rep. 2023;13(1):13065. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-023-38824-0\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-38824-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDekeyser S, Schmits E, Glowacz F, Klein O, Schmitz M, Wollast R, Yzerbyt V, Luminet O. Predicting Compliance with Sanitary Behaviors among Students in Higher Education During the Second COVID-19 Wave: The Role of Health Anxiety and Risk Perception. Physiol Belgica. 2023;63(1):1\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5334/pb.1171\u003c/span\u003e\u003cspan address=\"10.5334/pb.1171\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede la Sablonni\u0026egrave;re R, Dorfman A, Pelletier-Dumas M, Lacourse \u0026Eacute;, Lina JM, Stolle D, Taylor DM, Benoit Z, Boulanger A, Caron-Diotte M, M\u0026eacute;rineau S, Nadeau A. (2020). COVID-19 Canada: The end of the world as we know it? (Technical report No. 1). Presenting the COVID-19 Survey. Universit\u0026eacute; de Montr\u0026eacute;al.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuplaga M. The Roles of Health and e-Health Literacy, Conspiracy Beliefs and Political Sympathy in the Adherence to Preventive Measures Recommended during the Pandemic. Int J Environ Res Public Health. 2022;19(14):8346. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph19148346\u003c/span\u003e\u003cspan address=\"10.3390/ijerph19148346\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFishbein M, Ajzen I. Predicting and Changing Behavior: The Reasoned Action Approach. Taylor \u0026amp; Francis; 2011.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForward SE. The theory of planned behaviour: The role of descriptive norms and past behaviour in the prediction of drivers\u0026rsquo; intentions to violate. Transp Res Part F: Traffic Psychol Behav. 2009;12(3):198\u0026ndash;207. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.trf.2008.12.002\u003c/span\u003e\u003cspan address=\"10.1016/j.trf.2008.12.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalasso V, Pons V, Profeta P, Becher M, Brouard S, Foucault M. (2020). Gender differences in COVID-19 attitudes and behavior: Panel evidence from eight countries. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e117\u003c/em\u003e(44), 27285\u0026ndash;27291. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.2012520117\u003c/span\u003e\u003cspan address=\"10.1073/pnas.2012520117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao S, Binod P, Chukwu CW, Kwofie T, Safdar S, Newman L, Choe S, Datta BK, Attipoe WK, Zhang W, van den Driessche P. A mathematical model to assess the impact of testing and isolation compliance on the transmission of COVID-19. Infect Disease Modelling. 2023;8(2):427\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.idm.2023.04.005\u003c/span\u003e\u003cspan address=\"10.1016/j.idm.2023.04.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGelfand MJ, Gavrilets S, Nunn N. Norm Dynamics: Interdisciplinary Perspectives on Social Norm Emergence, Persistence, and Change. Ann Rev Psychol. 2024;75(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev-psych-033020-013319\u003c/span\u003e\u003cspan address=\"10.1146/annurev-psych-033020-013319\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. annurev-psych-033020-013319.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHermans L, Van den Broucke S, Gisle L, Demarest S, Charafeddine R. Mental health, compliance with measures and health prospects during the COVID-19 epidemic: the role of health literacy. BMC Public Health. 2021;21(1):1365. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12889-021-11437-w\u003c/span\u003e\u003cspan address=\"10.1186/s12889-021-11437-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eINSPQ. (2020, May 22). \u003cem\u003eCOVID-19: Mesures sanitaires recommand\u0026eacute;es pour la population g\u0026eacute;n\u0026eacute;rale | INSPQ\u003c/em\u003e. Institut national de sant\u0026eacute; publique du Qu\u0026eacute;bec. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.inspq.qc.ca/publications/3008-mesures-sanitaires-population-generale-covid19\u003c/span\u003e\u003cspan address=\"https://www.inspq.qc.ca/publications/3008-mesures-sanitaires-population-generale-covid19\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones BL, Nagin DS. Advances in Group-Based Trajectory Modeling and an SAS Procedure for Estimating Them. Sociol Methods Res. 2007;35(4):542\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0049124106292364\u003c/span\u003e\u003cspan address=\"10.1177/0049124106292364\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones BL, Nagin DS, Roeder K. A SAS Procedure Based on Mixture Models for Estimating Developmental Trajectories. Sociol Methods Res. 2001;29(3):374\u0026ndash;93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0049124101029003005\u003c/span\u003e\u003cspan address=\"10.1177/0049124101029003005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ\u0026oslash;rgensen F, Bor A, Rasmussen MS, Lindholt MF, Petersen MB. (2022). Pandemic fatigue fueled political discontent during the COVID-19 pandemic. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e119\u003c/em\u003e(48), e2201266119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.2201266119\u003c/span\u003e\u003cspan address=\"10.1073/pnas.2201266119\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKil H, Lacourse E, Mageau GA, Pelletier-Dumas M, Dorfman A, Stolle D, Lina J-M, de la Sablonni\u0026egrave;re R. (2023). Initial risk factors, self-compassion trajectories, and well-being outcomes during the COVID-19 pandemic: A person-centered approach. \u003cem\u003eFrontiers in Psychology\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frontiersin.org/articles/\u003c/span\u003e\u003cspan address=\"https://www.frontiersin.org/articles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpsyg.2022.1016397\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2022.1016397\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKokko K, Tremblay RE, Lacourse E, Nagin DS, Vitaro F. Trajectories of Prosocial Behavior and Physical Aggression in Middle Childhood: Links to Adolescent School Dropout and Physical Violence. J Res Adolescence. 2006;16(3):403\u0026ndash;28. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1532-7795.2006.00500.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1532-7795.2006.00500.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrot K, Sousa JP. Factors Impacting On Patient Compliance with Medical Advice: Empirical Study. Eng Manage Prod Serv. 2017;9(2):73\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1515/emj-2017-0016\u003c/span\u003e\u003cspan address=\"10.1515/emj-2017-0016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKwasnicka D, Dombrowski SU, White M, Sniehotta F. Theoretical explanations for maintenance of behaviour change: a systematic review of behaviour theories. Health Psychol Rev. 2016;10(3):277\u0026ndash;96. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17437199.2016.1151372\u003c/span\u003e\u003cspan address=\"10.1080/17437199.2016.1151372\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLa Presse. (2020, September 22). Une campagne publicitaire gouvernementale \u0026laquo; plus percutante \u0026raquo; sur la COVID-19. \u003cem\u003eLa Presse\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.lapresse.ca/actualites/covid-19/2020-09-22/une-campagne-publicitaire-gouvernementale-plus-percutante-sur-la-covid-19.php\u003c/span\u003e\u003cspan address=\"https://www.lapresse.ca/actualites/covid-19/2020-09-22/une-campagne-publicitaire-gouvernementale-plus-percutante-sur-la-covid-19.php\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeviston Z, Stanley SK, Walker I. Australians underestimate social compliance with coronavirus restrictions: findings from a national survey. Aust N Z J Public Health. 2022;46(3):304\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1753-6405.13195\u003c/span\u003e\u003cspan address=\"10.1111/1753-6405.13195\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLyu W, Wehby GL. Comparison of Estimated Rates of Coronavirus Disease 2019 (COVID-19) in Border Counties in Iowa Without a Stay-at-Home Order and Border Counties in Illinois With a Stay-at-Home Order. JAMA Netw Open. 2020;3(5):e2011102. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1001/jamanetworkopen.2020.11102\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2020.11102\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacNeil S, Desch\u0026ecirc;nes S, Kn\u0026auml;uper B, Carrese-Chacra E, Dialahy IZ, Suh S, Durif F, Gouin J-P. Group-based trajectories and predictors of adherence to physical distancing during the COVID-19 pandemic. Psychol Health. 2022;37(12):1492\u0026ndash;510. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/08870446.2021.2014486\u003c/span\u003e\u003cspan address=\"10.1080/08870446.2021.2014486\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartin LT, Ruder T, Escarce JJ, Ghosh-Dastidar B, Sherman D, Elliott M, Bird CE, Fremont A, Gasper C, Culbert A, Lurie N. Developing predictive models of health literacy. J Gen Intern Med. 2009;24(11):1211\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11606-009-1105-7\u003c/span\u003e\u003cspan address=\"10.1007/s11606-009-1105-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCaffery KJ, Ayre J, Dodd R, Pickles K, Copp T, Muscat DM, Nickel B, Cvejic E, Zhang M, Mac O, Isautier J, Cornell S, Wolf MS, Bonner C. Disparities in public understanding, attitudes, and intentions during the Covid-19 pandemic: The role of health literacy. Inform Serv Use. 2023;43(2):101\u0026ndash;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3233/ISU-230185\u003c/span\u003e\u003cspan address=\"10.3233/ISU-230185\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiddleton KR, Anton SD, Perri MG. Long-Term Adherence to Health Behavior Change. Am J Lifestyle Med. 2013;7(6):395\u0026ndash;404. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/1559827613488867\u003c/span\u003e\u003cspan address=\"10.1177/1559827613488867\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMontagni I, Ouazzani-Touhami K, Mebarki A, Texier N, Sch\u0026uuml;ck S, Tzourio C, the CONFINS group. Acceptance of a Covid-19 vaccine is associated with ability to detect fake news and health literacy. J Public Health. 2021;43(4):695\u0026ndash;702. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/pubmed/fdab028\u003c/span\u003e\u003cspan address=\"10.1093/pubmed/fdab028\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorales-Rodr\u0026iacute;guez FM, P\u0026eacute;rez-M\u0026aacute;rmol JM. (2019). The Role of Anxiety, Coping Strategies, and Emotional Intelligence on General Perceived Self-Efficacy in University Students. \u003cem\u003eFrontiers in Psychology\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frontiersin.org/articles/10.3389/fpsyg.2019.01689\u003c/span\u003e\u003cspan address=\"https://www.frontiersin.articles/10.3389/fpsyg.2019.01689\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMukerjee S, Chow CM, Li M. Mitigation strategies and compliance in the COVID-19 fight; how much compliance is enough? PLoS ONE. 2021;16(8):e0239352. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0239352\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0239352\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagin DS. Analyzing developmental trajectories: A semiparametric, group-based approach. Psychol Methods. 1999;4(2):139\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/1082-989X.4.2.139\u003c/span\u003e\u003cspan address=\"10.1037/1082-989X.4.2.139\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagin DS, Tremblay RE. Analyzing developmental trajectories of distinct but related behaviors: A group-based method. Psychol Methods. 2001;6(1):18\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/1082-989X.6.1.18\u003c/span\u003e\u003cspan address=\"10.1037/1082-989X.6.1.18\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeville FG, Templeton A, Smith JR, Louis WR. Social norms, social identities and the COVID-19 pandemic: Theory and recommendations. Soc Pers Psychol Compass. 2021;15(5):e12596. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/spc3.12596\u003c/span\u003e\u003cspan address=\"10.1111/spc3.12596\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNielsen JH, Lindvall J. Trust in government in Sweden and Denmark during the COVID-19 epidemic. West Eur Politics. 2021;44(5\u0026ndash;6):1180\u0026ndash;204. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/01402382.2021.1909964\u003c/span\u003e\u003cspan address=\"10.1080/01402382.2021.1909964\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOECD. (2020, July 3). \u003cem\u003eTransparency, communication and trust: The role of public communication in responding to the wave of disinformation about the new coronavirus\u003c/em\u003e. OECD. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://read.oecd-ilibrary.org/view/?ref=135_135220-cvba4lq3ru\u0026amp;title=Transparency-communication-and-trust-The-role-of-public-communication-in-responding-to-the-wave-of-disinformation-about-the-new-coronavirus\u003c/span\u003e\u003cspan address=\"https://read.oecd-ilibrary.org/view/?ref=135_135220-cvba4lq3ru\u0026amp;title=Transparency-communication-and-trust-The-role-of-public-communication-in-responding-to-the-wave-of-disinformation-about-the-new-coronavirus\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOECD. (2021, May 21). \u003cem\u003eThe Territorial Impact of COVID-19: Managing the Crisis and Recovery across Levels of Government\u003c/em\u003e. OECD. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://read.oecd-ilibrary.org/view/?ref=1095_1095253-immbk05xb7\u0026amp;title=The-territorial-impact-of-COVID-19-Managing-the-crisis-and-recovery-across-levels-of-government\u003c/span\u003e\u003cspan address=\"https://read.oecd-ilibrary.org/view/?ref=1095_1095253-immbk05xb7\u0026amp;title=The-territorial-impact-of-COVID-19-Managing-the-crisis-and-recovery-across-levels-of-government\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOkan O, Messer M, Levin-Zamir D, Paakkari L, S\u0026oslash;rensen K. Health literacy as a social vaccine in the COVID-19 pandemic. Health Promot Int. 2022;daab197. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/heapro/daab197\u003c/span\u003e\u003cspan address=\"10.1093/heapro/daab197\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePelletier-Dumas M, Dorfman A, Jean-Marc L, Dietling S, Taylor D, de la Sablonni\u0026egrave;re R. (2020). \u003cem\u003eThe role of clear and coherent government measures to reduce the spread of COVID-19\u003c/em\u003e (No. 1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://a5eaedf4-8d49-4b28-9cae-77db17679ab5.filesusr.com/ugd/68376a_689097aa55924fcdab5e7c6e513440ab.pdf\u003c/span\u003e\u003cspan address=\"https://a5eaedf4-8d49-4b28-9cae-77db17679ab5.filesusr.com/ugd/68376a_689097aa55924fcdab5e7c6e513440ab.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePetherick A, Goldszmidt R, Andrade EB, Furst R, Hale T, Pott A, Wood A. A worldwide assessment of changes in adherence to COVID-19 protective behaviours and hypothesized pandemic fatigue. Nat Hum Behav. 2021;5(9):1145\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41562-021-01181-x\u003c/span\u003e\u003cspan address=\"10.1038/s41562-021-01181-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePublic Health Agency of Canada. (n.d.). \u003cem\u003eCOVID-19: Prevention and risks\u003c/em\u003e. Public Health Agency of Canada. Retrieved 10 December 2023, from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.canada.ca/en/public-health/services/diseases/2019-novel-coronavirus-infection/prevention-risks.html#p\u003c/span\u003e\u003cspan address=\"https://www.canada.ca/en/public-health/services/diseases/2019-novel-coronavirus-infection/prevention-risks.html#p\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRashed EA, Hirata A. One-Year Lesson: Machine Learning Prediction of COVID-19 Positive Cases with Meteorological Data and Mobility Estimate in Japan. Int J Environ Res Public Health. 2021;18(11):5736. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph18115736\u003c/span\u003e\u003cspan address=\"10.3390/ijerph18115736\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eResnik DB. Scientific Research and the Public Trust. Sci Eng Ethics. 2011;17(3):399\u0026ndash;409. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11948-010-9210-x\u003c/span\u003e\u003cspan address=\"10.1007/s11948-010-9210-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReynolds KJ, Subašić E, Tindall K. The Problem of Behaviour Change: From Social Norms to an Ingroup Focus. Soc Pers Psychol Compass. 2015;9(1):45\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/spc3.12155\u003c/span\u003e\u003cspan address=\"10.1111/spc3.12155\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRigotti T, De Cuyper N, Sekiguchi T. The Corona Crisis: What Can We Learn from Earlier Studies in Applied Psychology? Appl Psychol. 2020;69(3):1\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/apps.12265\u003c/span\u003e\u003cspan address=\"10.1111/apps.12265\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodon C, Chin J, Chevalier A. Assessing COVID-19 Health Literacy (CoHL) and its relationships with sociodemographic features, locus of control and compliance with social distancing rules during the first lockdown in France. Health Educ Res. 2022;37(3):143\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/her/cyac009\u003c/span\u003e\u003cspan address=\"10.1093/her/cyac009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRogers RW. A Protection Motivation Theory of Fear Appeals and Attitude Change1. J Psychol. 1975;91(1):93\u0026ndash;114. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00223980.1975.9915803\u003c/span\u003e\u003cspan address=\"10.1080/00223980.1975.9915803\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosenstock IM. The Health Belief Model and Preventive Health Behavior. Health Educ Monogr. 1974;2(4):354\u0026ndash;86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/109019817400200405\u003c/span\u003e\u003cspan address=\"10.1177/109019817400200405\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRutty C. (2022, March 2). \u003cem\u003eCOVID-19 Pandemic in Canada\u003c/em\u003e. Can Encyclopedia. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.thecanadianencyclopedia.ca/en/article/covid-19-pandemic\u003c/span\u003e\u003cspan address=\"https://www.thecanadianencyclopedia.ca/en/article/covid-19-pandemic\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchumpe, B. M., Van Lissa, C. J., B\u0026eacute;langer, J. J., Ruggeri, K., Mierau, J., Nisa,C. F., Molinario, E., Gelfand, M. J., Stroebe, W., Agostini, M., G\u0026uuml;tzkow, B., Jeronimus,B. F., Kreienkamp, J., Kutlaca, M., Lemay, E. P., Reitsema, A. M., vanDellen, M. R.,Abakoumkin, G., Abdul Khaiyom, J. H., \u0026hellip; Leander, N. P. (2022). Predictors of adherence to public health behaviors for fighting COVID-19 derived from longitudinal data. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(1), 3824. https://doi.org/10.1038/s41598-021-04703-9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwarzer R, Warner LM. (2013). Perceived Self-Efficacy and its Relationship to Resilience. In S. Prince-Embury \u0026amp; D. H. Saklofske, editors, \u003cem\u003eResilience in Children, Adolescents, and Adults: Translating Research into Practice\u003c/em\u003e (pp. 139\u0026ndash;150). Springer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-1-4614-4939-3_10\u003c/span\u003e\u003cspan address=\"10.1007/978-1-4614-4939-3_10\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSheppard VB. Providing health care to low-income women: a matter of trust. Fam Pract. 2004;21(5):484\u0026ndash;91. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/fampra/cmh503\u003c/span\u003e\u003cspan address=\"10.1093/fampra/cmh503\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSj\u0026ouml;din H, Wilder-Smith A, Osman S, Farooq Z, Rockl\u0026ouml;v J. Only strict quarantine measures can curb the coronavirus disease (COVID-19) outbreak in Italy, 2020. Eurosurveillance. 2020;25(13):2000280. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2807/1560-7917.ES.2020.25.13.2000280\u003c/span\u003e\u003cspan address=\"10.2807/1560-7917.ES.2020.25.13.2000280\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS\u0026oslash;rensen K, Van den Broucke S, Fullam J, Doyle G, Pelikan J, Slonska Z et al. Health literacy and public health: A systematic review and integration of definitions and models. BMC Public Health [Internet]. 2012 Jan 25 [cited 2024 May 8];12(1):80. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1471-2458-12-80\u003c/span\u003e\u003cspan address=\"10.1186/1471-2458-12-80\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStatistics Canada. (2013, October 8). \u003cem\u003eThe Daily \u0026mdash; Skills in Canada: First results from the Programme for the International Assessment of Adult Competencies, 2012 (final)\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www150.statcan.gc.ca/n1/daily-quotidien/131008/dq131008b-eng.htm\u003c/span\u003e\u003cspan address=\"https://www150.statcan.gc.ca/n1/daily-quotidien/131008/dq131008b-eng.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStok FM, De Ridder DTD, De Vet E, De Wit JBF. Don\u0026rsquo;t tell me what I should do, but what others do: The influence of descriptive and injunctive peer norms on fruit consumption in adolescents. Br J Health Psychol. 2014;19(1):52\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/bjhp.12030\u003c/span\u003e\u003cspan address=\"10.1111/bjhp.12030\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTam CC, Li X, Li X, Wang Y, Lin D. Adherence to preventive behaviors among college students during COVID-19 pandemic in China: The role of health beliefs and COVID-19 stressors. Curr Psychol. 2023;42(20):17280\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12144-021-01942-x\u003c/span\u003e\u003cspan address=\"10.1007/s12144-021-01942-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeixeira da Silva JA, Tsigaris P, Erfanmanesh M. Publishing volumes in major databases related to Covid-19. Scientometrics. 2021;126(1):831\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11192-020-03675-3\u003c/span\u003e\u003cspan address=\"10.1007/s11192-020-03675-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Canadian Press. (2021). \u003cem\u003eFrom first cases to first vaccines: A timeline of COVID-19 in Canada - National | Globalnews.ca\u003c/em\u003e. Global News. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://globalnews.ca/news/7597228/covid-canada-timeline/\u003c/span\u003e\u003cspan address=\"https://globalnews.ca/news/7597228/covid-canada-timeline/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrentacosta CJ, Criss MM, Shaw DS, Lacourse E, Hyde LW, Dishion TJ. Antecedents and Outcomes of Joint Trajectories of Mother\u0026ndash;Son Conflict and Warmth During Middle Childhood and Adolescence. Child Dev. 2011;82(5):1676\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1467-8624.2011.01626.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1467-8624.2011.01626.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUsherwood T, LaJoie Z, Srivastava V. A model and predictions for COVID-19 considering population behavior and vaccination. Sci Rep. 2021;11(1):12051. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-91514-7\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-91514-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWollast R, Schmitz M, Bigot A, Speybroeck N, Lacourse \u0026Eacute;, De La Sablonni\u0026egrave;re R, Luminet O. Trajectories of health behaviors during the COVID-19 pandemic: a longitudinal analysis of handwashing, mask wearing, social contact limitations, and physical distancing. Psychol Health. 2023;1\u0026ndash;28. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/08870446.2023.2278706\u003c/span\u003e\u003cspan address=\"10.1080/08870446.2023.2278706\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organisation. (n.d.). WHO Coronavirus (COVID-19) Dashboard. Retrieved 10. December 2023, from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://covid19.who.int\u003c/span\u003e\u003cspan address=\"https://covid19.who.int\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organization. (n.d.). Let\u0026rsquo;s flatten the infodemic curve. World Health Organization. Retrieved 10. December 2023, from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/news-room/spotlight/let-s-flatten-the-infodemic-curve\u003c/span\u003e\u003cspan address=\"https://www.who.int/news-room/spotlight/let-s-flatten-the-infodemic-curve\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWright L, Fancourt D. Do predictors of adherence to pandemic guidelines change over time? A panel study of 22,000 UK adults during the COVID-19 pandemic. Prev Med. 2021;153:106713. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ypmed.2021.106713\u003c/span\u003e\u003cspan address=\"10.1016/j.ypmed.2021.106713\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWright L, Steptoe A, Fancourt D. Trajectories of Compliance With COVID-19 Related Guidelines: Longitudinal Analyses of 50,000 UK Adults. Ann Behav Med. 2022;56(8):781\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/abm/kaac023\u003c/span\u003e\u003cspan address=\"10.1093/abm/kaac023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZmerli S, van Meer TWG. der. (2017). \u003cem\u003eHandbook on Political Trust\u003c/em\u003e. Edward Elgar Publishing.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZou X, Savani K. Descriptive norms for me, injunctive norms for you: Using norms to explain the risk gap. Judgm Decis Mak. 2019;14(6):644\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S1930297500005362\u003c/span\u003e\u003cspan address=\"10.1017/S1930297500005362\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"COVID-19, compliance, preventive public health measure, factors, latent trajectories","lastPublishedDoi":"10.21203/rs.3.rs-4391822/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4391822/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePrevious longitudinal studies have identified variability in compliance with COVID-19 preventive measures, noting the heightened sensitivity of the least compliant groups to situational factors like easing restrictions. However, they overlooked other forms of variability inherent in compliance behaviour. Hence, we investigated compliance with social distancing and staying-at-home measures, and its dynamic nature, along with its association with social and individual factors.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eData from a longitudinal study involving 3617 Canadians across twelve measurement times from April 2020 to April 2022 were analysed. Compliance levels were assessed through self-reported items, alongside social and individual factors like trust in science and government, perception of social norms, and health literacy.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eJoint-Trajectory analysis revealed three compliance trajectories during mandatory and lifting measures periods: \"Low and constant\" (shifting to \"Low and decreasing\" during lifting measures), \"High and decreasing,\" and \"High and constant.\" Transition movements between trajectories mostly showed stability (between 85.83% and 90.44% for social distancing; between 80.68% and 87.71% for staying-at-home), with health literacy consistently predicting trajectory membership (social distancing: mandatory measures period- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.67; 95% \u003cem\u003eCI\u003c/em\u003e [1.20, 2.32], lifting measures period- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.43; 95% \u003cem\u003eCI\u003c/em\u003e [0.93, 2.37]; staying-at-home: mandatory measures period- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.75; \u003cem\u003eCI\u003c/em\u003e [1.24, 2.56], lifting measures period- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.90; 95% \u003cem\u003eCI\u003c/em\u003e [0.26, 1.01]). Perception of provincial norms (social distancing: comparison 1- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.10; 95% \u003cem\u003eCI\u003c/em\u003e [1.30, 17.54], comparison 2- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.38; 95% \u003cem\u003eCI\u003c/em\u003e [0.12, 1.04]; staying-at-home: comparison 1- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.22; 95% \u003cem\u003eCI\u003c/em\u003e [1.13, 4.51], comparison 2- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.74; 95% \u003cem\u003eCI\u003c/em\u003e [0.55, 1.00]) and trust in science (social distancing: \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.57; 95% \u003cem\u003eCI\u003c/em\u003e [1.29, 10.91], comparison 2- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.00; 95% \u003cem\u003eCI\u003c/em\u003e [1.26, 7.96]; staying-at-home: comparison 3- \u003cem\u003eOdds\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.77; 95% \u003cem\u003eCI\u003c/em\u003e [0.58, 1.02]) moderately predicted trajectory change.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003e Our study not only reaffirms variability within the lowest compliance group but also unveils variability among higher compliance groups, notably in trajectory transition movements. Health literacy consistently emerged as a strong indicator of trajectory membership, while perception of provincial norms and trust in science moderately influenced trajectory change.\u003c/p\u003e","manuscriptTitle":"Beyond What Meets the Eye: Unveiling Dynamics of Compliance with Preventive Measures in the COVID-19 Era","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-28 20:47:07","doi":"10.21203/rs.3.rs-4391822/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-06-04T08:08:53+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-05-13T10:17:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-05-13T10:17:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Public Health","date":"2024-05-09T01:11:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"be47e775-9576-46cc-93ac-378767a9302e","owner":[],"postedDate":"May 28th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-09T16:04:16+00:00","versionOfRecord":{"articleIdentity":"rs-4391822","link":"https://doi.org/10.1186/s12889-026-26347-y","journal":{"identity":"bmc-public-health","isVorOnly":false,"title":"BMC Public Health"},"publishedOn":"2026-03-04 15:57:36","publishedOnDateReadable":"March 4th, 2026"},"versionCreatedAt":"2024-05-28 20:47:07","video":"","vorDoi":"10.1186/s12889-026-26347-y","vorDoiUrl":"https://doi.org/10.1186/s12889-026-26347-y","workflowStages":[]},"version":"v1","identity":"rs-4391822","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4391822","identity":"rs-4391822","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}