Massive analysis of longitudinal data links social media to poorer youth mental health

Spending time on digital media, particularly social media, tends to be consistently linked with poorer developmental outcomes in children and teens. A massive review of existing research suggests that while video games might offer a slight boost to certain mental skills, heavy digital media use generally corresponds with increased depression, behavioral problems, and substance use. The findings were published in JAMA Pediatrics.

Digital media has become deeply embedded in the daily lives of children, with use rising steadily across all age groups from infancy through adolescence. Much of the debate surrounding this topic relies on cross-sectional research. A cross-sectional study looks at data from a single point in time, which makes it impossible to tell if screen time causes problems or is merely associated with them.

To get a clearer picture, the authors of the new study wanted to focus entirely on longitudinal studies. Longitudinal research tracks the same individuals over a period of time, observing their habits before any health outcomes are measured.

“The debate around children’s digital media use has been fierce, but most of the evidence fuelling it comes from studies that only capture a single snapshot in time — they can’t tell us whether screen time is actually causing problems, or just associated with them. We wanted to change that,” said study author Sam Teague, a senior research fellow at James Cook University and head of the JCU Digital Wellbeing Group.

“By focusing exclusively on longitudinal studies, where children’s media use was tracked before any health outcomes were measured, we could start to map the direction of these relationships. We also deliberately cast a wide net — looking across all types of digital media use and all outcomes, positive and negative — to give a comprehensive picture rather than a narrow slice of the story.”

Free daily newsletter

Previous research reviews have often been highly fragmented. Many past studies aggregated all digital engagement into broad measures of screen time, failing to distinguish between watching television and interacting on social platforms. Other reviews focused too narrowly on specific outcomes, like academic performance, without considering a child’s broader physical or emotional development. The researchers hoped to address these gaps by looking at a wide variety of specific media types and outcomes.

The researchers conducted a systematic review and meta-analysis, which is a method of combining and analyzing data from multiple independent studies to find common trends. They searched major academic databases for longitudinal observational research published between January 2000 and August 2024. The final analysis included 153 studies representing 115 unique groups of people.

Altogether, the data encompassed roughly 360,000 participants from across the globe, with individual study sample sizes ranging from 26 to 54,908 children. The ages of the participants spanned from two to 19 years old. The gender breakdown was nearly even, with 53.8 percent girls and 46.2 percent boys. Most of the included studies were conducted in Europe and North America, with smaller numbers from Asia, Australia, and Latin America.

The researchers extracted data on how often children used different types of digital media. They categorized the exposures into social media, video games, and other media like messaging platforms, educational applications, or general internet use. They then compared these habits against 26 developmental subdomains. These subdomains were grouped into four broad categories that included social-emotional, cognitive, physical, and motor development. Study durations ranged from a single day to 22 years, with an average follow-up period of about two and a half years.

The researchers found that social media use was consistently associated with multiple social-emotional difficulties. Higher social media engagement provided evidence of increased depression, anxiety, behavioral problems, self-injurious thoughts, and problematic internet use. It was also linked to lower academic achievement, less positive development, and a poorer sense of self-perception.

In the physical health domain, social media use tended to correspond with higher rates of substance use. This association included alcohol, tobacco, and cannabis, with no major differences across the type of substance.

Video game use showed a slightly different pattern in the data. Playing video games was associated with increased aggression and externalizing behaviors, which are outward actions like rule-breaking or hostility.

At the same time, the data showed a small positive association between gaming and better attention and executive functioning. Executive functioning refers to mental skills that include working memory, flexible thinking, and self-control. Other media use, such as messaging applications and general digital device use, was associated with depression and poorer general health.

“The clearest finding is that digital media use — and social media in particular — is consistently linked to poorer outcomes for children and young people,” Teague told PsyPost. “Social media stood out: it was associated with worse outcomes across every domain we looked at, from mental health and social development through to physical health. The one bright spot was video gaming, which showed a small link to better executive functioning — things like attention and mental organization.”

The researchers also ran moderation analyses to see if factors like age, gender, or the year of the study influenced the results. They found that associations between social media and depression were stronger in early adolescence than in school-aged children.

Additionally, studies conducted after 2012, a time when smartphones became widespread, showed stronger links between social media and substance use. The researchers noted that effects may continue to change as digital platforms evolve and become more immersive.

Across the board, the negative associations between digital media use and child development were small to moderate in size. In practical terms, the researchers explain that these statistical effects are similar to the impacts of other common lifestyle factors, such as eating a poor diet or getting too little physical activity. Yet, even modest effects can add up over time. Teague said that the sheer volume of data showing the exact same results elevates the importance of these subtle shifts.

“The key word is consistent: across more than 100 studies involving around 360,000 children worldwide, the pattern kept showing up,” Teague told PsyPost. “This isn’t a problem that individual families can solve on their own by simply setting better rules around screen time. The scale of the evidence points to a need for action from policymakers and technology companies to make online environments genuinely safer for children.”

Teague was particularly surprised by the distinct mismatch between the positive experiences families frequently describe and the lack of supporting evidence found in the data.

“Families often describe digital media as a source of creativity and stronger friendships,” Teague explained. “But those benefits simply don’t appear in the longitudinal literature,” she said. “We looked hard for them. For example, we analyzed whether social media or gaming were linked to improvements in peer relationships, and found no such link.”

“And areas like creativity, empathy, and imagination haven’t been rigorously studied in this way at all. That absence of evidence matters. It doesn’t mean the benefits aren’t real, but it does mean we can’t confirm them. More research tracking these potential positives is genuinely needed.”

A primary limitation of the research is that it cannot definitively prove causation. While longitudinal studies provide stronger evidence than snapshots in time, other underlying factors could still explain the associations. Additionally, the vast majority of the studies were from upper-middle-income countries, limiting the ability to apply these findings to more diverse global populations.

“The most important one is that we can’t prove causation,” Teague noted. “Longitudinal studies give us stronger grounds than a single snapshot — we’re looking at media use before outcomes emerge — but other explanations remain possible. For instance, children already struggling with their mental health may engage with social media differently.”

“Or there may be underlying factors, like a difficult home environment, that drive both higher social media use and poorer mental health simultaneously. We pooled studies that controlled for some of these factors, which helps, but it wasn’t consistent across all the research. Future studies really need to control for baseline health to give us a firmer causal story.”

Moving forward, the researchers hope to understand how children’s digital habits will shift as countries begin implementing age-based restrictions on social media platforms. “With social media bans for children now in place in Australia and spreading globally, I want to understand how children’s digital habits shift in response — and in particular, what role AI will play as other platforms become harder to access,” Teague said.

The researchers also plan to look closer at early childhood habits. “We see children exceeding screen time recommendations at every age, which suggests the window for intervention may be earlier than we think,” Teague explained. “If we can understand how media habits formed in the early years track through to adolescence, we can better support families in building healthier habits from the start.”

The study, “Digital Media Use and Child Health and Development: A Systematic Review and Meta-Analysis,” was authored by Samantha Teague, Klaire Somoray, Adrian Shatte, and colleagues.