Social media breaks can genuinely help your mental health, but the picture isn’t simple
If you’ve ever considered taking a pause from social media for a mental clarity boost, emerging evidence suggests you might be onto something. In a study of young adults published in JAMA Network Open, participants who undertook a one-week social media detox reported meaningful improvements: anxiety symptoms dropped by 16.1%, depression by 24.8%, and insomnia by 14.5%. These results represent the initial phase of a larger research program, says lead author John Torous, an associate professor at Harvard Medical School and the director of the digital psychiatry division at Beth Israel Deaconess Medical Center. In this edited discussion, he shares what surprised him about the early results and what the team plans next.
What does prior research tell us about social media and mental health?
Much of the existing work in this area relies on self-reports. Researchers often ask young people to estimate how many hours they spent on various platforms over weeks or months, and to gauge how those screen habits affected their relationships, sleep, exercise, and routines. If you asked me to recall my own screen time and sleep patterns over the past two weeks, I wouldn’t have precise numbers. A lot of the key findings in this field have depended on these self-reported measures.
That constraint partly inspired this study. It’s important to note: this wasn’t designed as a treatment trial. Rather, it’s a methodological exploration to show that we can capture and interpret data in novel ways using individuals’ phone data, with the aim of advancing the field.
What did the researchers hope to learn?
The main aim was to use a voluntary social media detox to observe real-time shifts in how people use social platforms and how they felt as a result. The smartphone automatically records usage, enabling researchers to track changes as they happen. The study monitored typical usage for two weeks, followed by a one-week detox.
What did they discover?
Over the initial two weeks, participants used social media about two hours per day. During the detox, overall social-media time plummeted from about 1.9 hours per day to roughly 30 minutes—an appreciable drop. Interestingly, total screen time remained roughly constant, indicating that people weren’t reducing all screen activity; they were simply substituting non-social-screen activities for social media. Across five platforms, Instagram and Snapchat emerged as the toughest to resist.
Were there any surprises?
The most striking finding was the wide variation in individual responses. Some participants who started with higher levels of depressive symptoms saw noticeable improvements, while others experienced little to no change. Some individuals turned to exercise, increased their daily steps, or left home more often. This pronounced heterogeneity caught the researchers off guard and underscored that a one-size-fits-all approach may miss important nuances.
These results point to the need for a more personalized view of social media and mental health. The path forward likely involves tailoring strategies to each person, based on objective data collected from their own devices.
What comes next for the research?
Phase 1 established a baseline. In Phase 2, the plan is to pursue more targeted interventions. For instance, if a surge in social-media use correlates with sleep disruption, researchers could implement a focused sleep-improvement strategy. By identifying distinct patterns or clusters of behavior, they aim to pinpoint digital signals that predict how someone will respond to a given intervention and then deliver a customized mini-detox.
The goal isn’t simply to ban social media; it’s to address specific vulnerabilities, such as sleep. If sleep is a weak point, the intervention would prioritize improving sleep hygiene while allowing for healthy online engagement. That approach could be a breakthrough in personalizing digital well-being.
Why is this research timely?
Some states are considering rules to ban phones in schools, and while those policy moves have merit in certain contexts, new measurement methods could enable smarter, more targeted solutions. Like many aspects of healthcare, precise measurement opens doors to personalized care.
A key takeaway is that average effects often hide how individuals respond. A blanket detox can be blunt and ineffective for some people, especially since social media can alleviate loneliness for others. Rather than removing access entirely, the aim should be to learn how to manage it so it works for each person. This research invites policymakers and the public to support a new generation of personalized digital-health work, focused on practical, individualized strategies.
