The Lancet: problematic digital media use is linked to mental health, sleep and inactivity harms in 9 million young people; experts say the evidence is ‘not conclusive but sufficient to act’, and age limits alone are not enough

The medical journal The Lancet published on 4 October a three-paper series on the digital determinants of health, written by 13 global experts. According to the authors, the way digital technologies are currently built and governed creates real risks to health worldwide alongside genuine benefits; the series calls for coordinated governance of digital systems and for the health and online safety of children and young people to be made an immediate global policy priority. According to series chair Professor Ilona Kickbusch of the Digital Transformations for Health Lab in Switzerland, governments have let technology companies write their own rules, leading to business models built to capture attention and collect data, with health treated as an afterthought.

At the core of the series is an umbrella review assessing 121 published reviews covering 9 million people under 25. It found consistent associations between problematic digital media use, defined as addiction-like patterns of use rather than time spent online, and a range of adverse outcomes in children, adolescents and young adults: common mental health problems, sleep problems and reduced physical activity. According to Fast Company, more than three-quarters of the reported exposure-outcome associations were adverse, though there was also evidence of beneficial use, such as children watching educational content with caregivers and adolescents playing active video games. Increased screen time for gaming, social media and video streaming was linked to reduced academic performance and delays in language development.

The authors do not hide the limits of their findings. According to Louise Holly of DTH-Lab UK, the associations identified in the review are modest in size and rest on low-certainty evidence that cannot establish cause and effect; the series says the figures should not be taken as an accurate measurement of harm. Even so, the authors argue that because the evidence is directionally consistent, the harms are biologically plausible and the risk is spread across large numbers of children, action is justified without waiting for perfect causal evidence. Features such as infinite scroll and autoplay, they stress, are deliberate design choices, not a matter of individual self-control.

The series recommends that governments require technology firms to evaluate new products or features for potential harm before release and to share data with independent researchers, that a dedicated international body monitor AI harms to children and young people, and that young people help shape the digital systems around them. Single measures such as age limits are not enough on their own, the authors say; Kickbusch adds that legal action can compel companies to disclose internal documents and do what regulation often cannot. Turkey is also debating a social media age limit for children, and after Australia’s ban for under-16s, similar rules are under discussion in France and Germany. Dr Samantha Marsh of the University of Auckland said the series makes a strong case for a precautionary, population-level response, and that for children under two the benefits appear negligible.