Available via license: CC BY 4.0
Content may be subject to copyright.
Page 1/18
A randomized trial of the effects of limiting digital screen
use on parent-child synchrony in physical behaviours and
family cohesion
Anders Grøntved
University of Southern Denmark
Sarah Sørensen
Sports Science and Clinical Biomechanics, University of Southern Denmark
Jesper Schmidt-Persson
University College Copenhagen
Peter Kristensen
Sports Science and Clinical Biomechanics, University of Southern Denmark
Martin Rasmussen
Steno Diabetes Center Odense, Odense University Hospital
Soe Mortensen
Research and Implemention Unit PROgrez, Department of Physiotherapy and Occupational Therapy, Naestved-
Slagelse-Ringsted Hospitals, Region Zealand
Anne Gejl
University of Southern Denmark
Lauren Arundell
Deakin University
Jo Salmon
Deakin University
Brendan Halpin
University of Limerick Limerick
Article
Keywords:
Posted Date: July 24th, 2024
DOI: https://doi.org/10.21203/rs.3.rs-4593710/v1
License: This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full
License
Additional Declarations: No competing interests reported.
Page 2/18
Page 3/18
Abstract
Engagement in shared activities between parents and children is potentially compromised by the pervasive use of
digital screens in familial contexts. In this randomized trial in 326 parent-child dyads nested in 87 families, we
investigated the effects of limiting screen use in parents and children on the amount of synchrony in physical
behaviors and family cohesion. Families were randomly assigned to wither undergo an extensive screen media
reduction intervention or to control. For seven days at baseline and follow-up, parents and children each wore two
accelerometers, positioned on the thigh and trunk, 24 hours/day, enabling the second-by-second classication of
their physical behaviors. Time-series sequence analysis of physical behavior revealed signicant enhancements in
dyadic synchrony for the screen reduction group. In shared leisure time, the between-group mean difference in
change favored the screen reduction group, with a -0.18 point (95%CI -0.27 to -0.10) decrease in time-warp edit
distance dissimilarity score and a 32.9 min/day (95%CI 16.0 to 49.9) of more direct matched activity. Additionally,
parents in the screen reduction group reported enhanced family communication, more collaborative tasks, and
engagement in new shared activities. Our ndings highlight the potential benets of reducing screen time for
improving parent-child behavioral synchrony and familial cohesion.
Main
The family is widely recognized as a social context that provides emotional and physical support, stability, and a
sense of belonging. Therefore, the time parents and children spend time together is considered vital for fostering
cohesion and improve family and interpersonal relationships [1], and is important for long-term health and well-
being [2, 3]. The parent-child interaction encompasses various forms of engagement, including behavioral
synchrony, which specically refers to the coordination of biological and behavioral processes that occur between
individuals as they interact with each other over time. This synchrony is considered a prosocial mechanism that is
deeply rooted in our evolutionary history, which serves to enhance group cohesion [4] and is important for
promoting healthy parent-child relationships and children’s future self-regulation and resilience [5–7]. Today, use of
digital screen devices (such as smartphones) is a common leisure activity among children and their parents. The
widespread use of screen devices, particularly for entertainment purposes, in families raises concerns about
whether it substitutes time spent on shared activities and impacts behavioral synchrony and family cohesion. Two
recent reviews concluded that individual screen media use, which refers to screen media use by an individual in
isolation, can displace meaningful face-to-face interaction and create conicts in families [1, 8]. Yet, most studies
conducted in families with children have been observational, leaving the causal impact of screen use on behavioral
synchrony and family cohesion open to further investigation.
Only a few experimental studies have investigated the effect of changing screen use behaviors on activities
engaged between parent and child and family cohesion. One randomized experiment investigated the effect of
social media abstinence for 7–28 days on the allocation of other daily activities in individuals aged 18–69 years
(31% of whom had children). The authors reported that a reduction in social media usage led to an increase in self-
reported time spent caring for children or other family members [9]. Another experimental study, which examined
the effects of either limiting or maximizing parental smartphone use during a single museum visit on parent-child
interaction, found that increased smartphone use diminished feelings of social connection with the child and
reduced meaningful interactions [10]. While these studies offer valuable insight into the effects of reducing social
media and smartphone use, their focus was primarily on individual screen usage behaviors, rather than on the
wider impact of limiting general screen media use within a family. Furthermore, these studies relied solely on self-
reported outcomes and adherence to the screen media use experiment, which could introduce a bias due to
Page 4/18
potential deviations from the intended intervention. It is important to further examine how reductions in leisure
time screen use in the family context can inuence objectively assessed behavioral synchrony and family
cohesion (determined by collaboration on everyday activities, talking together and spending time together in the
family). Objective measures are needed for providing more denitive and unbiased insights into the impact of
screen media use in modern family life. Consequently, the primary aim of this study was to investigate the effects
of limiting screen media use in families on the amount of parent-child physical behavioral synchrony during leisure
time on weekdays and weekend days. A secondary aim was to investigate parent reported responses of family
cohesion at the end of the digital screen use intervention.
Results
The current study included data from 87 randomized families, including 160 adults and 175 children. The analysis
includes 326 parent-child dyads consisting of one parent and one child from the same family (Supplementary Fig.
1). In families with two parents, every child was matched with both parents, and likewise, each parent was
matched with every child in the family. The mean age of children and parents were 8.8 (2.5) years and 40.8 (4.8)
years in the intervention group, and 9.7 (2.4) years and 42.4 (4.8) years in the control group, respectively. In both
groups 50% of the children were girls, while 54% of the parents were females. The mean amount of shared awake
leisure time in parent and child dyads was not signicantly different between the intervention and control group at
either baseline or follow-up, and no signicant within-group changes from baseline to follow-up in shared awake
leisure time were observed (Supplementary Table 1). As illustrated in Supplementary Figure 2A and 2B, the
intervention and control group were similar according to proportion of daily time children and parents had direct
second-by-second matching of active behaviors at baseline. Also, hourly mean time-warp edit distance (TWED)
dissimilarity scores across week- and weekend days were similar at baseline between the groups (Figure 2).
Combined for both groups the mean amount of daily aggregated time with directly matching active behaviors were
105.2 (SD 4.1), 82.0 (SD 43.4), and 174.0 min/day (SD 7.4) at baseline for total leisure time, leisure time on
weekdays, and weekends, respectively.
Momentary synchrony in physical behaviors
As described above the groups were similar at baseline regarding the amount of daily time children and parents
were momentary synchron in active behaviors. At follow-up dyads in the intervention group had increased the time
where children and parents were momentary synchron in their movement behaviors during total leisure time,
weekdays, and weekend days (Table 1). Mean differences in change of active behaviors were 32.94 (95% CI 16.00
to 49.89) min/day during leisure time in the screen reduction group as compared to the control group, which
corresponds to a 30% increase from baseline. The greatest mean difference in change was observed on weekend
days. No interaction effect was found for child age or gender.
Figure 1 provides a graphical overview of the crude mean proportion of time during each 5-minute interval across a
24-hour period in which parents and children showed momentary movement synchrony during their shared leisure
time in the nal week of the experiment. Generally, there appeared to be a trend indicating that dyads in the screen
reduction group had more time in momentary synchron in their active behaviors throughout the day, except for
weekday mornings.
The Time Warp Edit Distance (TWED) and optimal matching (OM) sequence analyses provided similar results to
the direct matching approach for total-, weekday and weekend day leisure time awake hours where children and
Page 5/18
parent share the possibility of being together. Dyads of parents and children allocated to the screen reduction
group had a -0.15 mean reduction in the TWED dissimilarity score as compared to a 0.04 increase in the control
group (between group mean difference in change -0.18 (95% CI -0.27 to -0.10)) for the total potential shared awake
leisure time (Table 1). Similar results were obtained for the OM sequence analysis. Estimated marginal means of
hourly mean TWED dissimilarity score in the screen reduction group and control group at baseline and follow-up
are provided in Figure 2 for weekdays and weekend days. Hourly mean TWED dissimilarity scores on both
weekdays and weekends were lower in the screen reduction group at follow-up as compared to baseline, while
scores in the control group were unchanged (Figure 2).
Page 6/18
Table 1:Parent-child dyadic synchrony in physical behaviors during total shared awake leisure time and on
weekdays and weekend days
Control (crude) Screen use restriction (crude) Between group mean
difference in change*
Mean
(SD)
baseline
Mean
(SD)
follow-
up
Mean
(SD)
change
Mean
(SD)
baseline
Mean
(SD)
follow-
up
Mean
(SD)
change
β95%
CI p-
value
TWED
dissimilarity
score
(points)
Total awake
leisure time 1.42
(0.29) 1.45
(0.30) 0.04
(0.31) 1.41
(0.30) 1.27
(0.28) -0.15
(0.39) -0.18 -0.27;
-0.10 0.000
Weekdays 1.41
(0.30) 1.44
(0.32) 0.05
(0.32) 1.41
(0.30) 1.26
(0.29) -0.15
(0.40) -0.19 -0.27;
-0.10 0.000
Weekend
days 1.43
(0.30) 1.44
(0.33) 0.01
(0.35) 1.41
(0.32) 1.29
(0.30) -0.14
(0.42) -0.15 -0.24;
-0.05 0.002
OM
dissimilarity
score
(points)
Total awake
leisure time 0.68
(0.14) 0.69
(0.14) 0.02
(0.14) 0.68
(0.14) 0.61
(0.13) -0.08
(0.18) -0.09 -0.13;
-0.05 0.000
Weekdays 0.67
(0.14) 0.69
(0.15) 0.03
(0.15) 0.68
(0.14) 0.61
(0.14) -0.08
(0.19) -0.10 -0.14;
-0.06 0.000
Weekend
days 0.69
(0.15) 0.69
(0.15) 0.00
(0.16) 0.68
(0.15) 0.62
(0.14) -0.07
(0.19) -0.07 -0.11;
-0.03 0.002
Direct
matching
(min/day)
Total awake
leisure time 101.58
(61.23) 101.18
(66.71) -3.00
(66.32) 93.37
(8.47) 119.78
(72.74) 29.26
(79.56) 32.94 16.00;
49.89 0.000
Weekdays 86.47
(47.29) 84.69
(56.57) -6.45
(58.86) 80.46
(52.50) 104.61
(60.25) 24.77
(70.26) 27.15 13.09;
41.20 0.000
Weekend
days 171.16
(100.87) 173.79
(107.18) -0.37
(115.00) 181.64
(119.25) 224.30
(101.08) 48.09
(144.46) 50.16 21.24;
79.07 0.001
TWED = Time Warp Edit Distance, OM = Optimal Matching
* Between group mean difference in change was estimated using mixed linear regression adjusting for age and
sex of parents and children and the time spent together in each parent-child dyad and accounting for repeated
observations in each dyad being clustered within families
Page 7/18
Family cohesion
As shown in Figure 3A, 82.9% of the parents in the screen reduction group reported in the end of the intervention
that it increased communication within the family, while four out of ve reported spending more time together
overall. Increased collaboration on everyday activities after reducing screen media was reported by 65.9% of the
parents, while 61.0% reported engaging in new activities with the family.
Figure 3B illustrates that during the rst week of the intervention, more than half (56.1%) of the families in the
intervention group reported that reduced screen media use increased the family’s time spent together on things
other than using screen media “to a very big extent”. In the second intervention week, 34.2% and 36.6% parents
responded with “to a very big extent” and “to a big extent”, respectively. In addition, we examined whether screen
media use was replaced with other individual activities, without increasing the family’s time together during the
intervention. In the rst week the most frequent answer was “to some extent” (46.3%), while in the second week
the most frequent answer was “to hardly any extent” (39.0%).
Discussion
In a cluster randomized controlled trial, we examined the effect of limiting parents’ and children’s screen media
use during leisure time on the amount of behavioral synchrony in physical behaviors and family cohesion.
Compared with the control group, the intervention signicantly increased instances where parent-child dyads
exhibited behavioral synchrony in objectively assessed physical behaviors during their shared awake leisure time,
as evidenced by both direct matching analysis and the TWED and OM sequence analyses. These analyses
collectively demonstrated an increased similarity in physical behaviors within the screen reduction group, whereas
the control group exhibited no signicant change. The direct matching effect size corresponds to a notable 30%
mean increase in time where children and parents displayed concurrent active behaviors. Parent-reported
perceptions of changes in the family's activities and time spent together among participants in the screen
reduction group further substantiate the idea of increased family time following a reduction in screen media use.
Activities such as collaborating on everyday tasks and conversations within the family were reported to have risen
in most participating families. To the best of our knowledge this is the rst randomized trial to investigate whether
limiting screen media use in families with children affect both the behavioral synchrony in objectively measured
activities among children and their parents, and the cohesion within the family.
In a previous study examining joint physical activity and sedentary behavior among 291 pairs of parent and
children aged 8–14 years who both wore an accelerometer and global positioning systems (GPS) device over the
same 7-day period, the results showed that parents and children spent a mean of 92.9 min/day in sedentary
behavior and 2.4 min/day of moderate-to-vigorous physical activity together during non-school waking hours [11].
When aggregating seconds over waking leisure hours, during which parents and children are engaged in the same
activity, each dyad in our study spent an average of 108 min/day on sedentary activities and 7.7 min/day on
running or walking at baseline, closely aligning with ndings from previous research. This may indicate that our
approach for estimating the amount of time spent on activity types where parents and children engage together in
shared leisure time is satisfactory. However, despite the presence of objective data indicating synchrony in the
physical activity behavior of parents and children, we cannot conclusively determine if it accurately reects their
social engagement. Nevertheless, it is reasonable to presume that the analysis in the current study captures
certain patterns in the engagement between parent-child dyads.
Page 8/18
Overall, the results suggest that limiting screen media use results in a greater behavioral synchrony in physical
activity between parents and their children. This increased synchrony could serve as an important mechanism for
reinforcing social bonds within the family. To further investigate the effect of the intervention on family
relationship parents completed questionnaires concerning this issue. The analyses showed that limiting screen
media use in families with children enhanced family cohesion by increasing time spent together as a family. A
study in 2023 found that the time parents spend with children is crucial for children’s growth and wellbeing [12].
This could be attributed to the fact that children require emotional support from their parents, and parents play a
crucial role in shaping the emotional and psychological wellbeing of their children [12]. Nevertheless, the wellbeing
of children is not solely determined by the amount of time parents spend with them, but also by the quality of the
interaction. As mentioned previously, two recently conducted systematic reviews showed that an individual’s use
of screen media devices is related to less meaningful face-to-face interactions and more conicts in the family [1,
8], which is in accordance with the results from the current study. Meaningful face-to-face interactions and social
connections in families are crucial as positive family relationships are associated with better mental health later in
life [2, 3]. Furthermore, previous studies have shown that adolescents perceiving that parents are using or being
distracted by their phone when interacting with their child, have a higher level of e.g. depressive symptoms and
anxiety [13, 14]. This is supported by our recent cross-sectional analysis in a population-based sample of Danish 7-
year-old children, which revealed a dose-dependent relationship between higher maternal smartphone addiction
scores and increased behavioral problems in children, including externalizing and internalizing diculties [15],
underscoring the broader implications of digital screen use on child development and family dynamics.
A strength of the current study is the experimental design used to investigate if changing screen media use habits
in families with children will impact the amount of behavioral synchrony within parent-child dyads and family
cohesion. In this experimental design, known and unknown confounding factors are expected to be equally
distributed between the intervention and control group. Furthermore, a strength of the current study is the use of
objectively measured physical activity behaviors and successful compliance with the screen reduction
intervention, as evidenced by objectively measured screen use. Finally, the included participants and those who
were non-eligible had similar background characteristics [16], which suggests that our ndings may be broadly
applicable across diverse family contexts and demographics within the types of families sampled for the
population-based survey described in the methods.
Despite the strengths, the current study has some limitations. One limitation is that blinding of participants to the
intervention was not possible in the current study. As previously reported it was found that compliance to the
intervention group was high, however those in the control group slightly reduced their screen media use during the
two week period even though they were instructed to continue with their usual screen media behaviors [16]. A
limitation of the direct matching analysis, where seconds are matched 1:1 between parent and child, may be the
potential underestimation of behavioral synchrony in active behaviors due to the time-lag to synchrony, i.e. the lag
of seconds between change in one person’s behavior and parallel change in another person’s behaviors [6]. Also,
the initial offset and drift in the real time clock between the accelerometers attached to the parent and child may
also result in underestimation of behavioral synchrony [17]. However, the sequence analysis using TWED, which
allows incorporating variations in behavior timing between the child and parent yielded similar results. In addition,
while the accelerometry data indicate whether parents and children exhibited similar classied behaviors at a
specic times of the day, they do not conrm whether they were physically together during those times. Future
studies may benet from collecting this information e.g., by using self-reported diaries, GPS, or video. Moreover,
the present study has limitations in terms of measuring family cohesion through self-reports and solely within the
Page 9/18
intervention group. The parents were aware of their randomization into the intervention group, which may
potentially inuence their reports of family cohesion. The absence of family cohesion reports in the control group,
prevents us from comparing the two groups. As a result, we cannot determine whether the observed changes in
family cohesion are a direct result of the intervention.
In conclusion, the results from this cluster randomized controlled trial showed that limiting screen media use in
families with children led to a signicant increase in the behavioral synchrony in physical activity behaviors
between parents and children during waking leisure time where they have the possibility to spend time together.
Furthermore, parents in the screen reduction group reported perceived improvements in family cohesion, primarily
attributed to spending more time together as a family. While direct comparisons with the control group were not
feasible due to the nature of the questions, self-reported observations in the intervention group highlight the
potential positive impact of reducing screen time on family relationships. Our study underscores the importance of
balancing screen use within the family setting, pointing to its signicant role in enhancing parent-child interactions
and fostering a cohesive family environment, which are key factors affecting children’s development and long-term
well-being.
Methods
Study design
The current study is a secondary analysis of The Short-term Ecacy of Reducing Screen-Based Media Use
(SCREENS) trial. The SCREENS trial is a parallel cluster randomized controlled trial that investigated the effect of
limiting household recreational screen media use for two weeks on physical behaviors in children and parents. The
enrolment of participants began on the 6th of June 2019, and the last family completed the follow-up
measurements on the 30th of March 2021. The study was approved by the Ethical Committee of Southern
Denmark (S-20170213) and performed in accordance with the Declaration of Helsinki. The trial was registered at
ClinicalTrials.gov (NCT04098913) and written informed consent was obtained before baseline assessments. A
detailed description of the SCREENS trial can be found in the study protocol [18]. The current study was reported in
compliance with the CONSORT 2010 statement for cluster randomized trials [19].
Participants
Families from 10 municipalities in the region of Southern Denmark were recruited through a population-based
survey inquiring about multiple aspects of individual and family screen media use [20] sent through a Danish
mandatory digital mailbox (e-Boks). The Danish Health Data Authority randomly selected a parent within families
with a least one child aged 6–10 years of age based on data from the Danish Civil Registration System. At the end
of the survey the adult was asked whether they were interested in participating in the SCREENS trial. The families
were eligible if: 1) the responding parent had reported more than 2.4 hours/day of screen media use (> 40th
percentile of screen media use based on the rst 1000 responders); 2) the family only had children 4 years or
older; and 3) at least one adult were full-time students or employed full-time, but not in a profession including
regular night shifts. If the family was eligible a member of the research team phoned them and asked if at least
one adult and one child were interested in participating in the study and that at least one participating adult and all
children were able to hand over their smartphone(s) and/or tablet(s) for the intervention period of two weeks.
Individual members were excluded if they: 1) were not able to engage in simple everyday activities; 2) had been
diagnosed with a sleep, neuropsychiatric or developmental disorder; or 3) had been on sick leave related to stress
Page 10/18
within the last three months. One thousand, four hundred and twenty (1,420) families were assessed for eligibility
and 89 families were included in the study (Supplementary File A).
Randomization and intervention
Families were randomized to an intervention or control group using a 1:1 allocation rate. The researcher service
organization Odense Patient Data Explorative Network, which did not participate in the intervention delivery or data
collection, generated the randomization sequence, and managed the online randomization platform that
performed a block randomization (permuted blocks of 2–4 families). After baseline assessments a member of the
research team randomized the families using the online randomization platform during a visit at the family’s
house. The intervention and control group had similar baseline characteristics [16].
The control group was instructed to continue with their usual screen media behaviors while the intervention group
was instructed to reduce their recreational screen media use to no more than three hours/week for the
intervention period. All children and at least one participating adult from each family in the intervention group
handed over their smartphone(s) and tablet(s) and received a non-smart cell phone (Nokia 130) which could only
be used for calling and sending text messages. During the intervention, adults and children above 10 years were
allowed a maximum of 30 min/day of necessary screen media use for activities such as planning daily
appointments, checking online banking or doing homework. All families completing the SCREENS trial received a
reimbursement of 70 Euros.
Measurements
All measurements conducted in the SCREENS trial are described in detail in the study protocol [18]. The specic
measurements, which are the basis for the analysis in this study, are described below.
Physical behaviors
Children’s and adults’ physical activity was measured using Axivity AX3 (Axivity Ltd) triaxial accelerometers. The
accelerometers were worn using elastic belts at both the thigh and waist for 24 hours per day through seven
consecutive days at baseline and follow-up with time annotation of school or work according to parent diaries
throughout the baseline and follow-up periods [16]. Accelerometry data was classied into activity types in a 1-
second resolution. Specically, we classied accelerometry data into 1-sec epochs of lying, sitting, standing,
standing with minor movement, walking, running, and cycling. The activity types standing, standing with minor
movement, walking, running, and cycling were considered physically active behaviors. Algorithms were used to
determine activity types with high accuracy in children and adults [21, 22]. Details on the identication of non-wear
periods are described in the study protocol [18]. Participants with physical activity data for at least three weekdays
and one weekend day with less than 10% nonwear time during leisure were included in the analysis.
Screen media use
A non-commercial monitoring device called SDU Device Tracker was used to assess the smartphone, tablet and
computer usage of both parents and children. The SDU Device Tracker was installed on the participants’ screen
devices and used to register screen time activity on a second-by-second basis, as well as if (and when) the
application has been closed by the user [23]. If the application was closed by a participant, the time period was
classied as missing. Television usage was tracked minute-to-minute using a TV-monitor (i.e. a separate device)
by assessing overall (not specic to an individual) power cord current. During the assessment periods, families
kept track of their own TV usage in personal diaries. Furthermore, during each day of the intervention all
Page 11/18
participants in the intervention group had to report their daily screen media use in a paper version diary. A detailed
description of combining the different screen media use measurements, including rules applied to assign TV-
monitor data (TV usage) to individual participants can be found in Pedersen et al. [16].
Family cohesion
Family cohesion was assessed with parents in the intervention group by questionnaires at follow-up (n = 41
families). Parents were asked to report their experiences relating to the period of screen use reduction.
Specically, they were asked to report on whether they noticed the following changes: ‘More time together in the
family’; ‘Increased collaboration on everyday activities (cooking, laying the table etc.)’; ‘Spoke more together in the
family’; ‘Made new activities with the family’; and ‘Completed tasks that you didn't have time to before’.
Furthermore, after both the rst- and second week of the intervention the parents were asked to report their level
of agreement with the following statements: i) reduced screen media use has increased the family’s time together
on other activities than screen media devices; and ii) whether screen media use was replaced with other solitary
activities, without increasing the families time together. The possible response categories were: ‘To a very large
extent’, ‘To a large extent’, ‘To some extent’, ‘To hardly any extent’, and ‘Not at all’.
Statistical analysis
Statistical analyses were performed using STATA MP version 18 and data pre-processing and wrangling were
performed in Python 3.10, MATLAB R2021b (v.9.11), and STATA. Adult children (17–18 years) were excluded from
the analysis (n = 2). Also, adults and children with less than three weekdays and one weekend day of
accelerometry data at baseline and follow-up (n = 8) were excluded from the analyses. Furthermore, the analyses
included only measurements from awake leisure time during weekdays and weekend days identied based on self-
reports, i.e. during the time when parents and children could engage in activities together. Based on accelerometry
data, we classied the data into types of physical activities at a 1-second resolution and counted the number of 1-
second intervals during which each parent-child dyad displayed momentary synchrony in physical activity types,
using the principle of direct matching. Therefore, a moment of synchrony was dened as a 1-second interval
during which both the parent and the child were engaged in the same type of activity (for instance, both were
walking, both were standing, etc.). We then aggregated the time-series data into 5-minute intervals and tallied the
total number of 'synchrony seconds' during shared awake leisure time. For a descriptive overview of the amount of
momentary synchrony in active behaviors, we constructed time-series line plots. These plots represent the mean
proportion of momentary synchrony in parent-child dyads in 5-minute intervals across a 24-hour period, separately
for weekdays and weekends, and at both baseline and follow-up.
We also applied both TWED and OM algorithms to assess momentary behavioral synchrony between children and
parents. These algorithms are specically developed to compare categorical sequence data [24], offering a
nuanced analysis of the extent of synchrony in the long time-series sequences of activity behavior patterns
between each parent-child dyad. TWED, adjusting for time dimensions, is particularly apt for our study as it can
accommodate slight temporal shifts, which is common when parents and children engage in the same activity,
and for inherent drifts in accelerometer internal clocks. OM, used for comparison, evaluates the transformation
cost between sequences through insertions, deletions, and substitutions. Both algorithms utilize a substitution
matrix here encompassing six states: sitting/lying, standing still/with minor movement, walking, running, cycling,
and missing data. This matrix is tailored to the specics of our study: a low cost is assigned to simultaneous
sedentary states, to reect our assumption that identical sedentary states are likely but not guaranteed to
represent shared activities. Moderate costs are attributed to combinations like one participant being sedentary
Page 12/18
and the other standing, reecting plausible shared activities but with greater uncertainty compared to active
states. For active states that differ (excluding cycling), we apply low costs, recognizing their potential for being
synchronized behaviors despite not being identical. When active states are identical between the parent and child,
they are assigned zero cost to indicate a perfect match. Given the minimal amount of missing data and the
inability of TWED and OM algorithms to process sequences with missing data for either dyad member, we
assigned a moderate cost for missing data scenarios. Mean TWED and OM dissimilarity scores in each parent-
child dyad were calculated by hour and across weekday, weekend day, and total awake leisure time where the
parent-child dyad had the possibility to spend time together. The SADI package in STATA were used for the
sequence analyses [25].
The primary analysis estimated the mean difference in daily behavioral synchrony changes in physical activity
behaviors between the screen reduction group and control groups. This was done using direct matching (mean
total daily time accumulated with matched behaviors) and the mean TWED and OM dissimilarity scores in parent-
child dyads as outcomes. Mixed-model linear regression analysis including two random intercepts (one for family-
level and one for the dyad-level) was used to estimate the mean difference and was restricted to shared awake
leisure time within each dyad. Analyses were carried out with data aggregated to total leisure time data and
according to weekdays and weekends. An interaction between group allocation (screen reduction vs. control) and
period of assessment (baseline vs. follow-up) was included as a xed effect to estimate the mean difference in
change between groups. The analysis was adjusted for age and gender of the child and parent and total leisure
time where parent-child dyads have the possibility to spent time together by including these covariates as xed
effects in the model. The analyses were repeated with the mean TWED and OM dissimilarity scores in parent-child
dyads as outcomes. Then, we conducted the analyses in specic time domains, separately; total leisure awake
time, weekdays, and weekend days. Furthermore, in families in the intervention group we calculated the
percentage reporting that they perceived changes in the family during the trial, and percentage reporting each
response on the two ve-point Likert scale questions asked after the rst and second week of the screen reduction
intervention.
Declarations
Author Contributions
SOS participated in the data collection, conducted the statistical analyses, interpreted data and led the writing of
the paper. AG designed the SCREENS study, received funding, supervised data analysis, interpreted data, and
reviewed and commented all drafts of the paper. JSP and MGR designed the SCREENS study, participarted in the
data collection, and reviewed and commented on the nal draft of the paper. PLK designed the SCREENS study,
processed accelerometry data, and reviewed and commented on the nal draft of the paper. SRM participated in
the data collection, reviewed and commented on the nal draft of the paper. AKG, LA and JS reviewed and
commented on the nal draft of the paper. BH supervised data analysis and reviewed and commented on the nal
draft of the paper. All authors approved the nal manuscript and agreed to be personally accountable for the
author’s own contributions. LA is supported by an Australian Research Council Discovery Early Career Researcher
Award (DE220100847). JS is supported by a Level 2 Investigator Grant, National Health and Medical Research
Council (1176885).
Competing interests
Page 13/18
The authors declare no competing interests.
Materials & Correspondence
Correspondence and materials request should be addressed to Anders Grøntved.
Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding
author on reasonable request which align with Danish Data Protection law.
References
1. Tammisalo, K. and A. Rotkirch,
Effects of information and communication technology on the quality of family
relationships: A systematic review
. Journal of Social and Personal Relationships, 2022. 39(9): p. 2724–2765.
2. Chen, P. and K.M. Harris,
Association of Positive Family Relationships With Mental Health Trajectories From
Adolescence to Midlife
. JAMA Pediatrics, 2019. 173(12): p. e193336-e193336.
3. Ramos, M.C., et al.,
Positive family relationships across 30 years: Predicting adult health and happiness
. J
Fam Psychol, 2022. 36(7): p. 1216–1228.
4. Launay, J., B. Tarr, and R.I.M. Dunbar,
Synchrony as an Adaptive Mechanism for Large-Scale Human Social
Bonding
. Ethology, 2016. 122(10): p. 779–789.
5. Bell, M.A.,
Chapter Six - Mother-child behavioral and physiological synchrony
, in
Advances in Child
Development and Behavior
, J.B. Benson, Editor. 2020, JAI. p. 163–188.
. Feldman, R.,
Parent–infant synchrony and the construction of shared timing; physiological precursors,
developmental outcomes, and risk conditions
. Journal of Child Psychology and Psychiatry, 2007. 48(3–4): p.
329–354.
7. Birk, S.L., L. Stewart, and T.M. Olino,
Parent-Child Synchrony After Early Childhood: A Systematic Review
. Clin
Child Fam Psychol Rev, 2022. 25(3): p. 529–551.
. Sbarra, D.A., J.L. Briskin, and R.B. Slatcher,
Smartphones and Close Relationships: The Case for an
Evolutionary Mismatch
. Perspectives on Psychological Science, 2019. 14(4): p. 596–618.
9. Hall, J.A., R.M. Johnson, and E.M. Ross,
Where does the time go? An experimental test of what social media
displaces and displaced activities’ associations with affective well-being and quality of day
. New Media &
Society, 2019. 21(3): p. 674–692.
10. Kushlev, K. and E.W. Dunn,
Smartphones distract parents from cultivating feelings of connection when
spending time with their children
. Journal of Social and Personal Relationships, 2019. 36(6): p. 1619–1639.
11. Dunton, G.F., et al.,
Joint physical activity and sedentary behavior in parent-child pairs
. Med Sci Sports Exerc,
2012. 44(8): p. 1473–80.
12. Li, D. and X. Guo,
The effect of the time parents spend with children on children's well-being
. Frontiers in
Psychology, 2023. 14.
13. Stockdale, L.A., S.M. Coyne, and L.M. Padilla-Walker,
Parent and Child Technoference and socioemotional
behavioral outcomes: A nationally representative study of 10- to 20-year-Old adolescents
. Computers in
Human Behavior, 2018. 88: p. 219–226.
Page 14/18
14. Wang, X., et al.,
Parental Phubbing and Adolescents' Depressive Symptoms: Self-Esteem and Perceived Social
Support as Moderators
. J Youth Adolesc, 2020. 49(2): p. 427–437.
15. Mortensen, S.R., et al.,
Parental Recreational Screen Media Practices and Behavioral Diculties Among
Danish 7-Year-Old Children
. Acad Pediatr, 2023. 23(3): p. 667–674.
1. Pedersen, J., et al.,
Effects of Limiting Recreational Screen Media Use on Physical Activity and Sleep in
Families With Children: A Cluster Randomized Clinical Trial
. JAMA Pediatrics, 2022.
17. Brønd, J.C., et al.,
Temporal Alignment of Dual Monitor Accelerometry Recordings
. Sensors, 2021. 21(14): p.
4777.
1. Rasmussen, M., et al.,
Short-term ecacy of reducing screen media use on physical activity, sleep, and
physiological stress in families with children aged 4–14: study protocol for the SCREENS randomized
controlled trial
. BMC Public Health, 2020. 20.
19. Campbell, M.K., et al.,
Consort 2010 statement: extension to cluster randomised trials
. BMJ: British Medical
Journal, 2012. 345: p. e5661.
20. Klakk, H., et al.,
The development of a questionnaire to assess leisure time screen-based media use and its
proximal correlates in children (SCREENS-Q)
. BMC Public Health, 2020. 20(1): p. 664.
21. Brønd, J.C., et al.,
Simple method for the objective activity type assessment with preschoolers, children and
adolescents
. Children, 2020. 7(7): p. 72.
22. Skotte, J., et al.,
Detection of physical activity types using triaxial accelerometers
. Journal of physical activity
and health, 2014. 11(1): p. 76–84.
23. Kristensen, P.L., et al.,
Criterion validity of a research-based application for tracking screen time on android
and iOS smartphones and tablets
. Computers in Human Behavior Reports, 2022. 5: p. 100164.
24. Halpin, B.,
Three Narratives of Sequence Analysis.
In Life Course Research and Social Policies Advances in
Sequence Analysis: Theory, Method, Applications (eds P.Blanchard, F.Bühlmannand J.-A.Gauthier), vol., pp.
75–103. Berlin: Springer., 2014.
25. Halpin, B.,
SADI: Sequence Analysis Tools for Stata
. The Stata Journal, 2017. 17(3): p. 546–572.
Figures
Page 15/18
Figure 1
Line plots (24h) of crude mean proportions of time with second-by-second momentary synchrony in active
behaviors by group on weekdays (A) and weekend days (B) during the last week of the experiment. Data
aggregated over 5-minute intervals.
Page 16/18
Figure 2
Estimated marginal means of hourly TWED dissimilarity scores of behavioral synchrony of physical behaviors in
parent-child dyads in the screen reduction group and control group on weekdays (A) and weekend days (B) at
baseline and follow-up
Page 17/18
Figure 3
Perceived positive changes (A) and perceived effects on time spent (B) among participating in the intervention
group during the two weeks of the SCREENS trial (n=41 parents).
Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.
