A 14-year-old picks up her phone before breakfast. In two minutes of scrolling she passes a clip about healing your gut in seven days, a before-and-after photo set, an ad for a greens powder, and a confident stranger explaining why her pediatrician is wrong about iron.
Nobody reviewed any of that. All of it arrived as entertainment.
The U.S. Surgeon General’s advisory on social media and youth mental health put real numbers on how much of this young people see. Up to 95% of young people aged 13 to 17 report using a social media platform. Nearly two thirds use one every day. One third describe their use as almost constant.
Those usage numbers are solid. What they mean for a given teenager is where the argument starts, and this article is going to be honest about that. The research is thinner than the headlines built on top of it.
The four numbers everyone quotes, and what each one measures
Start with the figures themselves, because they get mangled in retelling.
Up to 95% of 13- to 17-year-olds use a social media platform. The advisory draws this from survey work by Vogels and colleagues and by Rideout and colleagues, both published in 2022. Nearly two thirds of teenagers use social media daily. One third describe their use as almost constant.
Teenagers average 3.5 hours a day. The advisory cites Miech and colleagues, 2022, for that average. An average is not a typical day for your child. Some teens are at 30 minutes and some are at eight hours, and both are inside that mean.
More than three hours a day is associated with double the risk of mental health problems, including symptoms of depression and anxiety. The advisory cites Riehm and colleagues, 2019. Read that sentence slowly. It says associated, not caused. It reports a doubling of relative risk, which is a different thing from saying that half of heavy users become depressed.
46% of adolescents aged 13 to 17 said social media makes them feel worse about their body image, from Bickham and colleagues, 2022. That finding belongs to a companion article in this library on body image in young people, which handles what adults should do and avoid. It is worth knowing the figure exists. It is not the subject here.
Then there is the conclusion, which people quote loosely and should not. The Surgeon General’s advisory states that “we cannot conclude social media is sufficiently safe for children and adolescents,” while saying in the same breath that social media also provides benefits. That is a statement about missing evidence, not a verdict that social media has been proven harmful. The advisory also says plainly that we need to understand which content is most harmful and which factors protect young people, which is an admission that nobody knows yet.
Why this research is weaker than it sounds
Most of what exists is observational. Researchers measure how much time teenagers report spending online and compare it with how they report feeling. That design cannot tell you which came first.
Here is the problem underneath the problem. A 2020 study in the Journal of Affective Disorders used time use diaries instead of asking teenagers to estimate their own screen time, after finding that those self-estimates line up poorly with more objective measures. The researchers analyzed diary data from 4,032 British adolescents and did find associations between more social media time and self-harm, depressive symptoms and lower self-esteem in girls. The authors then said the cross-sectional design limits what you can conclude about causal direction. In other words, even a better-measured study still cannot tell you whether the app caused the low mood or the low mood caused the scrolling.
Newer US data complicates the simple story further. A 2026 analysis in the same journal used the 2023 Youth Risk Behavior Surveillance Survey, a nationally representative survey of American high school students. Bullying at school, cyberbullying and recent alcohol use were each significantly associated with four separate markers of suicide risk. Frequency of social media use was modestly associated with just one, feelings of sadness or hopelessness. The authors were explicit that causal and temporal relationships cannot be inferred from this design.
Effects also appear to be uneven rather than universal. A 2024 study in Social Science & Medicine set out to test an idea proposed by Orben and colleagues in 2022, that there are developmental windows when young people are especially sensitive to social media, with age and gender shaping the size of any effect. The Swiss researchers followed adolescents and young adults and found an effect of social media use on subjective well-being only among females from lower and middle parental education backgrounds, and only up to age 18. That is a narrow finding, not a blanket one. It is also the shape you would expect if social media acts as a stressor for some young people in some seasons of their lives rather than as a uniform toxin.
The broader mental health trend does not resolve cleanly either. The Centers for Disease Control and Prevention’s Youth Risk Behavior Survey trends report covering 2013 through 2023 found increases in signs of poor mental health and suicidal thoughts and behaviors across that decade. It also found what CDC describes as early signs of improvement between 2021 and 2023, a period when nobody’s phone use went down.
So where does that leave a parent? With a real association, an unproven mechanism, and a strong argument for paying attention to what a teenager is seeing rather than only how long they look at it.
The wellness feed is its own problem
This is the part the screen-time argument skips. A teenager who spends 90 minutes a day on social media is not spending it on a neutral surface. A large share of that time is spent on content about bodies, food, supplements and health, produced by people with no training and no accountability.
Researchers in Belgium asked emerging adults to collect food-related images from their own feeds and then talked through them. The 2026 study in the journal Appetite analyzed 559 images and found that 30% of the posts contradicted dietary guidelines, most often by showing very large portions of red meat. Most of the inaccurate content, 78.3% of it, came in recipe posts, and 90.7% of it came from creators with no relevant credentials. Participants tended to judge credibility by scientific-sounding language or by how healthy the creator looked, rather than by actual qualifications. The sample was European, so treat the exact percentages as indicative rather than as a US measurement.
A 2026 experiment published in the Journal of Health, Population and Nutrition got at the same weakness directly. Researchers showed 1,016 TikTok users aged 18 to 26 from the United States, United Kingdom, Australia and Canada one of six videos, varying whether the creator claimed a nutrition qualification and whether the information was accurate. Trust rose when the creator reported a qualification. It rose whether or not the underlying information was true.
That is the mechanism that matters. A stated credential, which nobody checks, buys trust that the content has not earned. A 16-year-old is not well placed to tell a registered dietitian from someone who bought a weekend certification, and the platform does not help.
Picture how this plays out. A 15-year-old who wants to put on muscle for a spring sport watches three videos about protein. By the fourth, the feed is showing creatine, then a pre-workout powder, then a “natural” testosterone booster. No single video told him to buy anything dangerous. The sequence did the work, and nobody along the way asked his age, his weight or what else he takes. Supplements for teenage athletes get their own article in this library, and the short version is that the products are not reviewed before they go on sale.
Supplement promotion rides on exactly this. A paid endorsement is supposed to be disclosed. The Federal Trade Commission tells influencers to disclose any financial, employment, personal or family relationship with a brand, to place that disclosure where it is hard to miss, and not to bury it in hashtags or behind a “more” link. How that rule works and how often it is ignored is covered in this library’s article on influencer disclosure. The practical point for a family is narrower: a sponsored post is an ad, and an ad for a supplement is not health advice.
Five questions worth asking about a 30-second health video
Teach these once and your teenager can run them in the time it takes a clip to load. They work better than a ban.
- Who is this person, and what would they have to have done to earn that title? “Nutritionist” is unregulated in many states. “Registered dietitian” requires a credential you can verify.
- What are they selling, and does the post say so? Look for a link in the bio, a discount code, or a product in the frame.
- Does the claim have a number attached to a study, or just a story? One person’s transformation is a story. Stories are not evidence, and this site says that about traditional remedies too.
- Is anything being called toxic, clean, or a cure? Those three words do more work in wellness marketing than in medicine.
- What would my doctor say if I showed them this? If the honest answer is “they would laugh,” you already have your answer.
What the advisory recommends
The Surgeon General’s advisory does not tell families to confiscate phones. Its advice to parents and caregivers is structural.
Make a family media plan. The advisory points to the American Academy of Pediatrics tool on HealthyChildren.org, which walks a household through building a personalized plan for how media gets used. The value is less in the document than in the fact that everyone agreed to it in advance, when nobody was angry.
Create tech-free zones and times. Bedrooms and meals are the two that families fight over and the two with the clearest payoff, because both protect sleep and conversation.
Encourage in-person friendship. Model the behavior you are asking for, which is the hardest item on the list and the one teenagers notice most.
The advisory’s guidance to young people themselves is worth reading out loud to a teenager. Reach out to a trusted friend or adult if social media is affecting you. Set boundaries around your own use. Be selective about what you post and with whom. And on cyberbullying, do not keep it a secret, do not forward harmful messages or images, and report offensive content.
That last instruction carries more weight than it looks like it does. A systematic review published in Worldviews on Evidence-Based Nursing identified 17 cyberbullying intervention programs described across 23 articles, and found that the components showing up most often were education about cyberbullying, coping skills, empathy training, communication and social skills, and digital citizenship. Programs with meaningful results also educated parents, not just the young person. The review’s authors noted that there were no interventions designed for the healthcare setting, which is a gap worth knowing about if you take a bullied teenager to a pediatrician and leave with less than you hoped for.
The trade-off nobody wants to state
Cutting a teenager off entirely is not a neutral act, and the advisory itself acknowledges that social media provides benefits alongside the risk.
For some young people, particularly those who are isolated at school, who are the only one like them in their town, or who manage a chronic illness, online community is not a distraction from support. It is the support. Removing it without replacing it can leave a kid lonelier than the scrolling did.
The honest version of the trade-off is this. Time spent online displaces sleep, movement and face-to-face contact, and those three things have far better evidence behind them than any claim about social media’s direct effect on mood. You do not have to win the causation argument to act on that.
Sleep is where the research is least ambiguous. A 2025 umbrella review in the International Journal of Environmental Research and Public Health pooled seven systematic reviews covering 127 primary studies and 867,003 participants. It found that digital device use is associated with shorter sleep, poorer sleep quality and later bedtimes in adolescents, that smartphones and computers show a stronger association than television, and that social media and internet use account for the largest disruptions. Late-night use was the worst offender. The authors still flagged heavy reliance on self-reported data and wide variation between studies, which is the same honest caveat that runs through this whole field.
That is enough to act on without settling the bigger argument. Protect the sleep and the in-person time, and the screen question largely answers itself.
When this stops being a screen-time conversation
At some point the question is not about the phone.
The National Institute of Mental Health gives parents a usable threshold for seeking an evaluation: behaviors or emotions that last weeks or longer, that cause distress for the young person or the family, or that interfere with school, home or friendships. For older children and teenagers, NIMH names signs including loss of interest in things they used to enjoy, low energy, sleep changes, pulling away from people, excessive dieting or exercise, self-harm, substance use and suicidal thoughts.
NIMH also says to seek help immediately if a young person’s behavior is unsafe or if they talk about hurting themselves or someone else. The next step there is a clinician, and in a crisis it is the 988 Suicide & Crisis Lifeline. The Surgeon General’s advisory tells young people directly: “If you or someone you know is experiencing a mental health crisis, call or text 988 for immediate help.” You can also chat at 988lifeline.org. The service is free, confidential and available every hour of every day, and it is for emotional distress and substance use concerns, not only for suicidal crisis. For a life-threatening emergency, call 911.
This library covers teen emotional well-being and the signs of a developing eating disorder in separate articles. If what you are seeing looks like either, start there and start with the pediatrician.
The most useful thing to take from all of this
Stop arguing about hours. The hour count is the weakest evidence in the whole field and the hardest number to enforce.
Argue about two things instead: what is in the feed, and what the feed is costing. Sit with your teenager and look at who is in their algorithm. Ask the five questions above about one video, not twenty. Then agree on where phones sleep at night.
That is a smaller intervention than banning an app, and it targets the part of the problem with the clearest evidence behind it. It also leaves your teenager with a skill that keeps working after they leave your house, which no screen-time limit does.
Sources
- U.S. Department of Health and Human Services, Office of the Surgeon General. “Social Media and Youth Mental Health: The U.S. Surgeon General's Advisory.” hhs.gov
- National Institute of Mental Health. “Children and Mental Health: Is This Just a Stage? (NIH Publication No. 24-MH-8085).” nimh.nih.gov
- Centers for Disease Control and Prevention. “Youth Risk Behavior Survey Data Summary & Trends Report: 2013-2023.” cdc.gov
- Federal Trade Commission. “Disclosures 101 for Social Media Influencers.” ftc.gov
- American Academy of Pediatrics (HealthyChildren.org). “Family Media Plan.” healthychildren.org
- 988 Suicide & Crisis Lifeline. “988 Suicide & Crisis Lifeline.” 988lifeline.org
- PubMed. “Bird K, et al. "Findings from the 2023 YRBSS: Examining the risk factors of social media use, alcohol use, bullying and adolescent suicide." Journal of Affective Disorders, 2026. PubMed. DOI: 10.1016/j.jad.2026.121211.” doi.org
- PubMed. “Barthorpe A, et al. "Is social media screen time really associated with poor adolescent mental health? A time use diary study." Journal of Affective Disorders, 2020. PubMed. DOI: 10.1016/j.jad.2020.05.106.” doi.org
- PubMed. “Mader S, et al. "The effect of social media use on adolescents' subjective well-being: Longitudinal evidence from Switzerland." Social Science & Medicine, 2024. PubMed. DOI: 10.1016/j.socscimed.2024.117595.” doi.org
- PubMed. “Glenisson L, et al. "What are influencers feed-ing us?: A photo-voice approach on emerging adults' social media food environment." Appetite, 2026. PubMed. DOI: 10.1016/j.appet.2026.108551.” doi.org
- PubMed. “Fox G, et al. "Examining the spread of nutrition-based misinformation by TikTok influencers: an experimental analysis of trust among young adults." Journal of Health, Population and Nutrition, 2026. PubMed. DOI: 10.1186/s41043-026-01321-x.” doi.org
- PubMed. “Fiore M, et al. "Digital Devices Use and Sleep in Adolescents: An Umbrella Review." International Journal of Environmental Research and Public Health, 2025. PubMed. DOI: 10.3390/ijerph22101517.” doi.org
- PubMed. “Hutson E, et al. "Systematic Review of Cyberbullying Interventions for Youth and Parents With Implications for Evidence-Based Practice." Worldviews on Evidence-Based Nursing, 2017. PubMed. DOI: 10.1111/wvn.12257.” doi.org