Does One-Week Social Media Detox Really Improve Mental Health by 25%?
- The White Hatter
- 13 minutes ago
- 5 min read
"Depression and anxiety decreased by 25 percent after a one-week social media break." A new headline, referencing a new research study (1-4). It's a compelling title: take a short break from social media, feel dramatically better. But once you look into the actual study, the results become far more nuanced.
The Big Number (25% Improvement) Is a Misleading Average
The large statistical improvement originates from the overall average decline in depression scores among participants. However, averages can obscure significant individual variations. Participants who began the study with moderate to severe depression exhibited the most substantial improvements in:
Self-reported signs of depression (based on a 9-point questionnaire)
Self-reported signs of anxiety (based on a 7-point questionnaire)
Self-reported signs of insomnia (based on a 7-point questionnaire)
The study authors specifically noted:
“Improvements were more pronounced among participants with higher baseline symptom severity.”
For participants with no depression signs at baseline:
There was a reduction in depression, but it was minimal compared to others, as there was little to reduce.
Participants with minimal depression still showed statistically significant improvements in anxiety and insomnia.
While all participants reported feeling better after the study, those who were in a worse mental state before the study experienced the most significant improvements. The detox did not universally increase happiness by 25%; it was most beneficial for those already facing challenges.
This Was Not a Randomized Controlled Trial
Participants self-selected into the detox group after reviewing their baseline data.
The authors are clear:
“Participants then self-selected into a 1-week social media detox intervention.”
That means people looking for relief may have chosen to detox, and motivated participants may have invested more effort.
Out of 398 eligible participants, 373 completed the baseline two-week assessments. Of these, only 295 opted into the detox program, meaning that approximately 20% of the study's participants chose to participate in the detox portion.
Plus, the sample was narrow and not representative. The participant group looked like this:
• Mean age: 21
• 74.3% female
• 90.3% iOS users
• 76.9% undergraduate students
It does not represent: Teens or younger adolescents
Most Participants Were Not Fully Off Social Media
The detox targeted only five platforms:
• Snapchat
• TikTok
• X (Twitter)
What it didn’t target, everything else.
• YouTube
• Discord
• Messenger
• Group chats
• News apps
• Gaming communities
And even within the five targeted platforms, many people still used the targeted platforms while on the detox.
• Instagram: 67.8% still used it
• Snapchat: 48.8% still used
• TikTok: 36.3% still used
This was not a week without social media. It was a partial reduction on five apps, with a great deal of continued screen usage.
No "Detox" Therapy
The term "detox" often suggests a therapeutic approach, expert-led mindfulness, and comprehensive professional mental health support. However, in this study, participants were simply asked if they wished to partake in a detox. According to the published study, this was not a structured therapeutic program. It consisted merely of a brief instruction and an option to participate, provided during a virtual visit, without any specified guide, curriculum, coaching, or guidance from a mental health practitioner.
Overall Screen Usage Actually Increased
One of the most confounding findings is that daily screen duration (all phone activity) increased by 4.5%. So people reduced time on the targeted platforms but increased time elsewhere.
This can be attributed to classic displacement behaviour where perhaps:
Quit TikTok and watch more YouTube.
Quit Instagram and scroll through Discord or other apps.
Participants Were Paid Up to $150 USD
Payment was for completing surveys and participating fully, not directly for quitting social media. So being paid to take part in a structured, time-limited experiment can itself:
• Elevate mood
• Boost motivation
• Reduce financial stress
• Frame the detox as a “reset week”
This situation differs from individuals who independently choose to take a break. Many could benefit from a week away from social media, temporarily distancing themselves from distressing world news, economic instability, violence, and inappropriate actions by world leaders, while also earning an additional $150 by answering some questions. It also contrasts with scenarios where one might incur costs for medical treatment.
Loneliness Did Not Improve
Although time spent on major platforms was reduced, there was no significant improvement in loneliness scores among individuals with depression or those without.
Real-Time Mood Did Not Meaningfully Change
The study used real-time check-ins. These showed no significant improvements in:
• Momentary mood
• Momentary anxiety
• Real-time functioning
Although the end-of-week surveys indicated reported improvements during reflection, the participants' real-time, day-to-day experiences did not demonstrate similar emotional enhancement.
The majority of improvements were observed in longer retrospective surveys, which may be affected by:
• Expectations
• Bias
• Placebo effects
• A desire to report improvement
What about the Anxiety and Insomnia improvements, even with those who didn't reduce depression?
Even participants who did not begin the study with high levels of depression still reported notable reductions in insomnia and anxiety. Why might this be?
First, it’s important to remember that the study relies entirely on self-reported measures, drawn from 14 questionnaire items assessing anxiety and insomnia. While this data is useful, it offers no insight into participants’ personal contexts, daily stressors, or the underlying causes of their symptoms. Without this background, interpreting why those improvements occurred becomes speculative.
One plausible explanation is that stepping away from the constant stream of online content, such as news about economic uncertainty, geopolitical tension, climate issues, or polarizing social commentary, may naturally reduce stress and cognitive load. In this sense, “ignorance is bliss” isn’t far off: reducing exposure to distressing information can improve subjective well-being, even if nothing external in a person’s life actually changes.
Another possibility relates to the classic “don’t think of a pink elephant” effect. When participants are asked to assess their sleep, anxiety, and mood, this draws conscious attention to issues they may not typically monitor. Ironically, the act of thinking about their habits may shift behaviour, yet the study includes no qualitative interviews, no follow-up conversations, and no participant narratives to help interpret these patterns. The data points are just numbers without context.
It’s also conceivable that the intake questions unintentionally served as a mirror; by being asked about their sleep routines at baseline, participants may have recognized poor habits (e.g., doomscrolling at night, inconsistent bedtimes). During the detox period, simply by knowing their sleep was being monitored, they may have made small improvements, even without being explicitly instructed to.
Summary
The detox didn’t uniformly boost mental health.
It helped some, did little for others, and didn’t change loneliness.
The biggest improvements were among those already struggling.
Those with moderate to severe symptoms saw the most benefit.
Many participants simply shifted to other online content.
Five social media apps' screen time was down a bit; overall, screen time was up.
The study does not show that “social media is bad.”
Rather, it shows:
• Problematic usage patterns matter
• Emotional context matters
• Self-selection matters
• Content exposure matters
• Temporary breaks can relieve stress
The viral “25% improvement” headline oversimplifies a much more nuanced study.
It’s evidence that context, intention, behaviour, and baseline mental health are the real drivers.
Candice L. Odgers, a professor of psychological science in informatics at the University of California, said in a New York Times article in response to this study:
“if we constantly tell people social media is bad for them and that taking a break is good, and then ask them to take a break and report how they feel after, this is what we see” (5).
Digital Food For Thought
The White Hatter
Facts Not Fear, Facts Not Emotions, Enlighten Not Frighten, Know Tech Not No Tech
