The claim that smartphones are driving adolescent mental health decline has become near-consensus in public discourse. The methodological quality of the studies underpinning that claim varies considerably. I'd like to see a systematic review of the strongest evidence on both sides.
Exactly. The intervention literature is thin and the effect sizes in the observational studies are small enough that confounding is a real concern. That's not to say there's nothing there. It's to say "put down your phone" may be better parenting advice than it is public health policy.
I keep thinking about the access dimension here. Restricting screen time is much easier for some families than others. If the intervention works, the families most likely to implement it are the ones whose kids are probably already at lower risk. Would be worth knowing whether the effect sizes hold across socioeconomic groups.
The Haidt and Twenge work gets a lot of traction precisely because it fits a story people already want to believe. What I'd want to know is how the intervention studies actually hold up. Correlational data on screen time and depression is everywhere. RCTs where you actually take phones away from teenagers and measure outcomes are much rarer and the results are less clean.