Sociodemographic predictors of changes in physical activity, screen time, and sleep among toddlers and pre-schoolers in Chile during the COVID-19 pandemic

The aim was to examine the sociodemographic predictors associated with changes in movement behaviours (physical activity, screen time and sleep) among toddlers and pre-schoolers during the early stages of the coronavirus disease 2019 pandemic in Chile. Caregivers of 1to 5-yearold children completed an online survey between March 30th and April 27th, 2020. Information about the child's movement behaviours before (retrospectively) and during the pandemic, as well as family characteristics were reported. In total, 3,157 participants provided complete data (mean children age: 3.1±1.38 years). During early stages of the pandemic, time spent in physical activity decreased, recreational screen time and sleep duration increased, and sleep quality declined. Toddlers and preschoolers with space to play at home and living in rural areas experienced an attenuated impact of the pandemic restrictions on their physical activity levels, screen time, and sleep quality. Older children, those whose caregivers had a higher educational level, and children living in apartments had greater changes, mainly a decrease in total physical activity and increase in screen time. This study has shown the significant impact of the pandemic restrictions on movement behaviours in toddlers and pre-schoolers in Chile.

The coronavirus disease 2019 (COVID-19) pandemic has had a significant impact on everyday life worldwide. Since the declaration of the global pandemic and the first national cases, Chile has had a total of 462,991 infected and suffered 12,741 deaths (from March through September 2020). [1] During the pandemic, Chile has been in the top 10 countries with the highest ratio of deaths per 100k population (67.9 deaths/100k inhabitants by September 30th 2020). [2] In response to this pandemic, the Chilean government implemented several approaches to reduce the health and economic impact of COVID-19. On March 16th, 2020, all schools were closed, and, in the same week, workplaces implemented work-from-home strategies. On March 17th, all national parks were closed, and on March 26th some districts enforced lockdowns and curfews. [3] These restrictions likely had a negative impact on young children's movement behaviours, (physical activity, sedentary behaviour -including screen time, and sleep). [4,5] Global guidelines recommend that children aged 1 to 5 should accumulate at least 180 min of physical activity, engage in no more than 1 h sedentary screen time (none for 1-year olds), and have between 11-14 h (1-to 2-year olds) and 10 to 13 h (3-to 5-year olds) good-quality sleep per night. [6,7] Limiting young children's ability to meet these recommendations may have long-term health consequences. [4,8] Studies conducted during the COVID-19 pandemic have shown that children and adolescents have decreased physical activity, increased screen time, and slightly increased sleep duration. [5] However, there is a lack of data regarding the impact of COVID-19 on movement behaviours in children under 5 years of age and from Latin-American countries.
Understanding the association between COVID-19 factors and young children's movement behaviours is important in guiding decision-making among policymakers, and in educating parents and health professionals. This information would be relevant not only during the pandemic but also when returning to a "new normal". The purpose of this study was to examine the early impact of COVID-19 restrictions implemented in Chile on levels of physical activity, screen time, and sleep among toddlers and pre-schoolers. In particular, we aimed to explore the sociodemographic factors associated with changes in these movement behaviours in this population group.

Materials and Methods
Main caregivers of 1-to 5-year old children living in Chile were invited to participate in an online survey from March 30th to April 27th, 2020. The study was promoted through social media (Facebook, Twitter, and Instagram) and emails from educational institutions. Potential participants accessed a personalized link to read more detailed information about the study and gave their online informed consent to participate in the study. The inclusion criteria were: 1) living in Chile, 2) being the main caregiver of a 1-to 5-year old child, and 3) living with the child most of the time before and during the COVID-19 pandemic. The study was approved by the Scientific Ethics Committee at Universidad de La Frontera, Chile (ORD.: 009-2020).
Data were collected and managed using REDCap (Research Electronic Data Capture) [9] hosted at Universidad de La Frontera. Data collection started two weeks after the government enforced educational centres to close in Chile (March 16th, 2020) due to COVID-19. At the end of the data collection, the education centres were still closed at a national level (April 27th, 2020).

Sociodemographic variables
The online survey included questions about the caregiver, family, and the child participating in the study. The sociodemographic and home characteristics section of the survey included questions about educational level, family income, occupational situation, dwelling type and size, space to play at home, inhabitants per home, area of residence (urban/rural), city, and region. If there was more than one child at home between 1 to 5 years of age, the caregiver was asked to answer for only one of the children (freely chosen). Also, caregivers were asked about child sex, age, disability (yes/no and type), and enrolment in an early childhood education centre (ECEC; yes/no). Caregivers were also asked if they were in lockdown at home at the moment of the survey, and the number of days spent in lockdown.

Movement behaviours
A proxy-report from the main caregiver was used to assess children's movement behaviours (physical activity, screen time, and sleeping). Caregivers were asked to provide the time (in hours and minutes) spent by the child on each of the movement behaviours before and during the COVID-19 pandemic. Sleep quality was assessed with a Likert scale (1 to 7; a higher score indicated better quality). The survey was piloted in a small sample before the official launch to optimize readability, length, and pertinence of the questions. The questions regarding movement behaviours are included as a supplementary file 1.

Statistical analysis
Only participants with complete data for all sections were included in the analysis. Mean (standard deviation, SD), median (interquartile range, IQR), and proportions were used to describe participants on key characteristics and outcomes based on data distribution. Comparisons between sociodemographic characteristics were performed using parametric and non-parametric tests to compare behaviours before and during COVID-19. We used multiple linear regressions with a residualized change score approach [10,11] to investigate the sociodemographic predictors of changes in physical activity, recreational screen time, sleep duration, and sleep quality during the early stages of the pandemic in Chile. This approach provides robust estimates by eliminating autocorrelated errors and regression towards the mean, which often makes it preferable to the simple change score approach. [10] First, we regressed the standardized score during the COVID-19 pandemic on the standardized scores before the COVID-19 pandemic for each of the behaviours assessed in this study. The residualized change score (i.e., trend) for each behaviour was then estimated as the average of each participant's residual score (i.e., the difference between the estimated value and the observed value). A positive residualized change score indicates an increase in the specific behaviour from the time before COVID-19 and a negative score indicates a decrease. For each residualized change score, we explored the predictive role of a series of sociodemographic factors a priori thought to have influenced changes in the behaviours herein assessed after ECEC and school closures due to COVID-19. We adjusted all models for age and sex of the child, family income, geographic region, and presence of lockdown. Data preparation and validation was conducted with The level of significance was set at usual p<0.05, two-tailed.

Results
In total, 5,505 individuals accessed the survey and 5,266 accepted participation and signed the consent form. We had participants from each of the 16 regions in Chile. A total of 2,109 (40.0%) participants were excluded from the analysis because they provided incomplete data (726 in demographics and COVID-19 related questions and 1,383 for the movement behaviours). The final sample consisted of 3,157 participants with complete data. Participants that provided complete data were more likely to be older ( with university studies or more, p<0.001) than those with incomplete data. The caregivers' characteristics from those with complete data were comparable with those observed for the respective age group in the last National Census in terms of dwelling (81.6% vs 79.7% living in a house) and living area (11.4% vs 13.5% living in a rural area), but the current sample was more educated (65.7% vs 39.7% with more than 12 years of education).[12] The mean age of children was 3.1 (1.38) years.
Children enrolled in ECEC or schools were 28.1% of 1-to 2-year olds and 71.9% of 3-to 5-year olds.
Most participants lived in houses (81.6%). About 9 out of 10 caregivers reported space to play at home. A large proportion of participants resided in urban areas (88.6%), and 76.3% s were under lockdown at the moment of the survey (mean duration 23.5 days [7.26]). No differences were observed between boys and girls in sociodemographic characteristics, except for the number of people (4.2 vs 4.1, p=0.013) and children (1.6 vs 1.7, p=0.004) per home (Table 1).    Table 3 shows the sociodemographic predictors of changes in movement behaviours due to the COVID-19 pandemic. Older children had greater reductions in physical activity and sleep duration, while they had more marked increase in their screen time. Children whose family had higher income had smaller decrease in sleep quality. Children whose main caregiver was more educated and those   Significance: *** = p < 0.001; ** = p < 0.01; * = p < 0.05

Discussion
To our knowledge, this is the first study examining the effects of COVID-19 on movement behaviours among children aged 1 to 5 years. We found that all movement behaviours changed across all ages, reflecting important secondary detrimental effects from this pandemic. The most common sociodemographic predictors of movement behaviours change during the COVID-19 pandemic were age, family income, previous enrolment in an ECEC, the main caregiver's education, dwelling type, available space to play, and type of residence area (urban/rural). This is also the first known study in Chile that examined all three components of the 24-hour movement behaviours in toddlers and pre-schoolers. A novel finding was that children from poorer families were more physically active than those in more affluent or educated families during the pandemic. This finding is different from that commonly observed in Chilean children and adolescents. [13,14] It is plausible that those caregivers were less likely to work from home under lockdown and kept more "normal" routines with their families, at the same time that those children may have had more freedom to play indoors and less demand to stay quiet during working hours.
Also, having space to play at home was consistently related to healthier levels in all movement behaviours both before and during the pandemic. This finding reinforces the need for ensuring spaces for children at home and surrounding areas to play as this is likely to promote these behaviours not only during the pandemic but also in the return to a "new normal". In Chile, specific content was developed for promoting physical activity through social media and national TV (since the April 27th 2020) as a response to the pandemic [15,16], and an expert committee on physical activity was installed by the Ministries of Sports, Health and Education to inform, adapt and add physical activity content into the curriculum of the classes delivered remotely in this context [17].
Nevertheless, the lockdown measures did not allow people to participate in physical activity outdoors and most public parks were still closed during the period included in our study, [3] with few councils prioritizing pedestrian zones to favour physical distancing. [18] These opposing strategies where on one hand PA is promoted indoors while at the same time it is severely restricted outdoors need to be periodically revised and aligned with the latest evidence so health impairing over-restrictions are prevented.
Pre-schoolers were the most affected by the restrictions during the pandemic, particularly for physical activity and recreational screen time. This may be explained by pre-schoolers needing more space to play and having greater access to screen-based devices than toddlers. The reduction in physical activity and the increase in screen time have particularly affected those with previous enrolment in an ECEC. Schools would be in a better position to respond with specific actions to prevent physical inactivity during this period than ECECs, as children and adolescents are more likely to be connected to participate in their virtual school activities than toddlers and pre-schoolers.
Therefore, governments, policymakers and professionals should devote particular attention to this age group.
Our results suggest that more educated parents tend to restrict their children's physical activity more, while at the same time may provide more opportunities to engage in screen-based behaviours.
This is an interesting finding as, before the pandemic, children with more educated caregivers tended to engage less with screen-based devices than children with less-educated parents. This may be partially explained as more educated caregivers may have to work from home, and this, in turn, may require the caregiver to use screens to entertain their child while working from home. This may be even more complicated in families living in apartments, one of the strongest predictors of declines in physical activity and increases of screen time observed in this study during the pandemic. This finding has implications for urban planning. As the full impact of COVID-19 is uncertain and other pandemics may occur, upgrading substandard areas that are currently plenty of small social apartments with insufficient recreational public spaces should be considered, together with stronger regulation, as part of a plan to build healthier and more resilient cities. Undoubtedly, families and caregivers play a key role in facilitating movement behaviours of toddlers and pre-schoolers, but some political actions such as allowing specific time for those residing in apartments to go outside with their family may boost the process. The family composition (number of people and children per home) was also associated with changes in movement behaviours during the pandemic, so strategies should consider messages and actions for the entire family group, particularly small families. [19] Overall, sleep quality worsened, but children with higher family income and less crowded homes showed less marked declines. This aspect may reflect other social issues that are not captured in this study that are related to the economic situation in Chile. [20] The decrease in sleep quality has also been described in a study conducted in Italy in children aged 3 to 6 during the early stages of the pandemic, however, after two weeks of follow up, this decline plateaued. [21] Some key messages to promote healthy sleep duration and quality, such as creating a bedtime routine and avoiding screens before sleep should be regularly disseminated and reinforced; and particularly tailored to lowincome families. [4,6,22] Having space to play at home was the most consistent factor in the home environment predicting changes in movement behaviours. A novel finding from our study is that the play space was not only important for physical activity and screen time but also had an impact on sleep quality. This is particularly important as this association reinforce the call from the World Health Organization for focusing on the interaction of all three behaviours as they benefit each other. [6] Therefore, strategies originally thought for promoting active playing at home or surroundings may impact sleep, and this, in turn, may benefit the whole home environment. Another relevant factor that resulted in healthier changes in physical activity and screen time was living in a rural area. Recent studies conducted during the COVID-19 pandemic in children and adolescents have also reported that rurality is an important correlate of movement behaviours. [23,24] A study conducted in Croatia reported that the decrease in physical activity was more evident in adolescents living in urban areas than it was in those from rural areas. [23] For children living in Canada, factors such as living in an apartment and the proximity to major roads were barriers to engage in outdoor activities. [24] Also, adolescents spent more time outdoor if they lived in a low-density area and had access to parks in high-density neighbourhoods. [25] We acknowledge that one of the primary measures taken by countries to control the pandemic was the imposition of mobility restrictions. However, decision-makers should seek options to facilitate outdoor recreational activities for the population while preserving safety and physical distancing instructions. Outdoor time is not only relevant for all movement behaviours but also the lack of it may impact other areas such as mental health, vitamin D deficiency and myopia. [4] In our study, the region of residence and lockdown situation were not predictors of the movement behaviour changes suggesting that regardless of location and containment measures, movement behaviours, in both toddlers and pre-schoolers, were similarly affected during the pandemic in Chile.
The provision of safe spaces for physical activity is critical when physical distancing is required.
As parks may be limited in space, availability and accessibility, programs such as open streets (e.g., Ciclovía) or play streets can be adapted and implemented while maintaining the physical distancing and other COVID-19-safe measures. [26][27][28] Thus, councils could provide additional safe spaces by closing lanes to cars, implementing complete streets schemes and transforming them into spaces for physical activity, play and recreation. Colombia has been highly innovative during the pandemic as they modified their current programs, usually delivered in parks or closed areas to face the needs of the population during the pandemic. [29] For example, in Bogotá, one of the first measures taken during the pandemic was the closure of lanes to private cars and the instalment of daily open streets. [30] Also, professionals from the Ministry of Sports Colombia have delivered their physical activity programs in the neighbourhoods while maintaining an appropriate physical distance. [29] These low-cost initiatives have been positively received by the general public and policymakers, [31,32] so once adapted to local needs they could be implemented in other contexts and countries, including Chile, and could help to alleviate the inactivity crisis. [3] The detrimental effects of the COVID-19 pandemic in toddlers and pre-schoolers are still uncertain. [

Strengths and limitations
Strengths of this study include a large sample of participants from each of the Chilean regions.
As recruitment was conducted online, voluntary response bias may have affected the sample composition with participants being more concerned regarding their children's health and having therefore higher chances to notice impairments in their movement behaviours. Besides this limitation, the respondents' characteristics were comparable from those observed in the last National Census for the respective age group in terms of dwelling type and living area, but the sample in our study was more educated.
[12] Although our instrument was piloted before it was officially launched, we had a large percentage of incomplete questionnaires (40.0%). This could be explained since some people may be more reluctant to provide personal information through online platforms than in face-to-face modes. We acknowledge that self-report measures used in this study may have biased the results (e.g., social desirability and recall). The use of accelerometers was not possible as physical distancing was mandated at the national level, including for those studies that were not strictly related to COVID-19 in clinical settings during the early stages of the pandemic. [33] Considering this, we collected data only during four weeks to minimize the risk of recall bias regarding the retrospective nature of some assessments.

Conclusions
The COVID-19 lockdown measures have shown to have a serious impact on children's behaviours. This study has shown that all movement behaviours changed during the early stages of the COVID-19 pandemic in toddlers and pre-schoolers in Chile. The most common sociodemographic factors associated with these changes were age, family income, main caregiver's education, previous enrolment in an ECEC, dwelling type, space to play at home and type of residence area. Toddlers and pre-schoolers with space to play at home and living in rural areas had less marked impacts on physical activity, screen time and sleep quality due to the pandemic. In contrast, older children, those whose caregivers had higher educational level and children living in apartments had greater changes, decreasing the total physical activity and increasing screen time.
Together, the information provided in this study may help professionals and decision-makers to balance more precisely the health risks and benefits of confinements, inform future strategies and focus resources to reduce the potential adverse effects of the pandemic, immediately, and in the longterm.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Table S1: title

Funding:
The study was partially supported by the Research Office at the Universidad de La Frontera (DIUFRO DFP19-0012, Programa para grupos de alta productividad, apoyo a la inserción de estudiantes pre-y postgrado a la investigación DI20-2009, y apoyo a postulantes FONDECYT DI20-1002). ADO is supported by a NHMRC Investigator Grant.