In this paper, the authors aim to build on current understandings on the impact of parent/child screen time on child development. While most research has generally focused on total screen time, the authors examined if the type of device (TV vs handheld) used by parents may also play a role.

The authors should be commended for taking a more nuanced approach to this emerging and important topic. Based on the identification of six parental screen use profiles, they have reported several associations with sociodemographic characteristics, child screen time and socioemotional problems. However, while the paper provides new knowledge, there are some major limitations that warrant a lot more caution from the authors about some of the conclusions drawn.

● Accuracy of reported screen use: I appreciate that the authors raise this as a limitation in their discussion. However, the issue remains as to how reliable the data are. As the authors noted, self-reported screen use has inherent biases; there are also likely to be difficulties in estimating screen use in a ‘typical’ day, and these may in turn be amplified when extrapolated to the whole week. Without a more objective means of obtaining the data (e.g. the Screen Time app on iOS devices), it is difficult to draw robust conclusions particularly from very small groups such as P6.

● In the authors’ discussion, they argue that the higher user profiles should be included despite the small numbers because of the potential social and clinical significance of problematic parental use. While I don’t disagree, I think the manuscript needs rewording to avoid overemphasising the significance of this particular association. Based solely on the data presented, and the problems with confounding factors, there does not appear to be a strong enough case to suggest adopting the US guidelines of a 2-hour limit (page 14, middle section), nor include the last sentence of the abstract.

● In relation to the findings on socioemotional problems, the paper would be greatly strengthened with information about parental style, including behaviours such as sensitivity and frequency of interactions with their child. This is because of the limitations of parental screen time as a proxy for the quality/quantity of parent-child interactions:

» Was time of day of parental screen exposure taken into account? E.g. could parents have been using screens when their children were at daycare or school?
» Some parents may be high users of screen and/or handheld devices, but could still engage in quality one-on-one time with their child.
» The study hypotheses are based on the assumption that handheld device use leads to fewer parent-child interactions (page 6 bottom para), but this is a very broad generalisation given the study doesn’t appear to have assessed how much of parents’ screen time may have involved their children too – e.g. playing gaming consoles together (which can be very interactive).
» Some types of screen use may not involve a long duration, e.g. quickly picking up a phone to check a notification, but could still be disruptive especially if it interrupts a parent-child interaction.

Other points:

● Computer/laptop use was measured, but was this restricted to non-work use (i.e. used during a time a parent could be – but was not – interacting with their child?)

● The authors should consider replacing the term ‘problematic’ with ‘excessive’, given that parents were not assessed for impacts on daily functioning.

● The discussion should note that the association between parent high use and child socioemotional difficulties was limited to the small group of ‘Extremely high users’, and not observed in the ‘High users’. Because if that association is real, there may be a usage threshold involved.

● The discussion mentions 6% of parents met or exceeded screen time guidelines. What proportion of children also did? And did they disproportionately belong to P6? It’s interesting that children of extremely high users had worse SDQ scores, but not significantly higher screen use themselves compared to P1 (and, likely, P4 and P5) children. This seems to warrant some discussion.

● (Page 13 lines 10-12) It is a little difficult to interpret the findings linking socioeconomic factors with screen use pattern in any meaningful way. E.g. P2 parents were more likely to be low-income, but this group has the second lowest total screen time.

I thank the authors for carefully considering all my comments and concerns. The revised manuscript now better reflects the study's limitations and is more circumspect about its immediate translational impact. The extra data and methodological elaborations are also useful. Overall, this is a much improved paper that should serve as a solid stepping stone for further research on parent screen use.