The study results show that 12% of boys and 8% of girls aged five to six watched more than two hours of TV on a weekday while 30% of fathers and mothers exceeded this threshold. On a weekend day 45% of boys and 42% of girls spent more than two hours watching TV, with 57% of fathers and 53% of mothers exceeding this threshold. A greater proportion of parents used a computer for more than 31 minutes on a weekend day than a weekday and a greater proportion of children used a games console on the weekend than a weekday, indicating differences in both parent and child SV on weekdays versus weekend days. If either mothers or fathers spent more than two hours per day watching a TV on a weekday, children were at least 3.4 times more likely to spend more than two hours watching TV. Estimates for mothers and fathers were similar. For weekend TV viewing there was an interaction between child gender and male parent TV viewing with boys 3.8 times more likely to exceed the two hour recommendation, rising to 7.9 times for girls. There was some evidence that parental weekend computer use was associated with child computer use but only in relation to girls and fathers. Collectively, these findings provide evidence of positive associations between parent and child SV time but suggest that patterns differ for weekend and weekend days. Associations are stronger for parental use at weekends than weekdays and fathers’ weekend TV and computer use may be more strongly associated with daughters’ than sons’ use.
The data reported here extends previous literature in this area which has largely focussed on mothers, young children and TV viewing. A study in Greece reported that 32% of children aged three to five spent more than two hours per weekday watching TV with both maternal (r = 0.27) and paternal (r = 0.20) hours of weekday TV correlated with child viewing time, with similar patterns for weekend days . In a sample of 750 UK families with a six to eight year old child, the child was 7.8 times more likely to watch more than two hours of TV per weekday if the parent did the same; this paper did not report separately for weekend days . The findings reported here are therefore comparable to previous literature but greatly extend the evidence by highlighting how patterns may differ for weekend versus weekdays. Interestingly, recent qualitative data from six European countries indicated that parents of children aged four to six reported that their children enjoyed watching TV and that most parents were not concerned about their child’s TV viewing . Efforts to change behaviour are therefore likely to require improving parental understanding of why high SV is a concern and how parents’ own SV behaviour may affect their child’s. Based on the results of this study it is not possible to determine whether strategies that focus on both children and the parents are needed to change children’s SV habits. However, intuitively it seems that integrated interventions which work with both children and parents have greater potential to change behaviour than parent only focussed interventions.
The results presented here show that SV levels are higher among children at the weekend, and children with parents who engage in high levels of SV at the weekend are more likely to engage in this behaviour. Stronger associations have also been found between fathers and their daughters, compared with sons, for weekend TV and computer use. In general, time spent in SV by both parents was positively and strongly associated with their child’s SV and the magnitudes of association with both genders combined were similar. This highlights the importance of engaging both parents in the development of any interventions to reduce child SV. Many studies have struggled to engage fathers in child-focussed research [23–25] and we are not aware of any study that has specifically attempted to engage fathers in reducing child SV.
The AAP guidance on youth SV was based on expert opinion in 2001  and at that time applied only to TV viewing. Although this guidance was amended in 2011 to take account of changes in technology and viewing patterns, it does not distinguish between weekday and weekend SV . Although the evidence presented might suggest that greater effects could be obtained by intervening on weekend days, there are more weekdays. Thus, a child with 5 hours SV each weekend day and 2 hours each weekday spends the same amount of time in total SV during the week as at the weekend (10 hours). If the time spent each day at the weekend was double that of weekdays (e.g. 6 vs. 3 or 5 vs. 2.5 hours), for most plausible examples the total time spent SV at weekends would be less than on weekdays. Furthermore, there is no evidence that interventions to reduce SV in parents (and consequently in children) would be more effective if they solely or largely targeted weekend SV. It should also be stressed that the evidence base for greater SV in children or adults being causally related to adverse health outcomes is weak and there is currently no evidence for risk increasing at any particular threshold . This is why the UK’s four Chief Medical Officers recently issued public health guidance that children and young people should limit SV but did not recommend a specific threshold . Thus, consistent strategies to reduce SV in children across the whole week are likely to be needed.
Strengths and limitations
The major strength of this study is the assessment of SV-time for both mothers and fathers, which has facilitated the examination of both maternal and paternal associations with child SV. The availability of information directly reported by fathers is a unique and important addition to the literature and addresses the current over-reliance on maternal reports of SV behaviour. Information on a key age group, for both weekdays and weekend days on a range of SV behaviours will also greatly inform the future development of targeted interventions. There are, however, limitations that should also be acknowledged. Firstly, the data is reliant on parental report of child SV. Although we are not aware of any alternative, reliable means of collecting this data, as the children are too young to self-report, a degree of error is likely in the parental reports. Of particular importance is the potential correlation of errors between the parent’s self-reported SV and their reported estimate of their child’s SV. An attempt was made to address this by adjusting for whichever parent had reported their child’s SV and it is somewhat reassuring that associations are similar for both parents. Secondly, we did not assess multi-SV in which children or parents simultaneously use multiple screen devices, which limits our ability to capture this modern pattern of viewing . Finally, the cross-sectional nature of our data precludes any interpretation of the direction of association between parent and child SV. However, due to the age of the children it seems unlikely that children are influencing parent behaviour.