What is doomscrolling and how do you stop it?

Summary: Doomscrolling is reading bad news on a feed long past the point where it tells you anything, and the studies put a measurable cost on it: Lower mood within minutes, higher anxiety and depression scores, and worse sleep. A read with a last line fixes it better than quitting the news.

The scroll rarely starts as a decision. You open the phone to check one thing, and an hour later you know a little more about several disasters and feel worse about all of them. That hour now has a name, a scale that measures it, and enough research to say what it costs and what stops it.

What is doomscrolling?

Merriam-Webster's entry defines the verb as spending excessive time scrolling through "online content (especially news) that makes one feel sad, anxious, angry" and so on. The research draws it tighter. A 2022 paper in Technology, Mind, and Behavior, which built the first scale for it, describes a habit where "social media users persistently attend to negative information in their newsfeeds about crises, disasters, and tragedies." By that definition, a long evening on a ranked feed of other people's emergencies counts, and a long evening with one newspaper falls outside it.

A 2021 paper in Journalism Studies, built on 550 written replies from Norway's March 2020 lockdown, found three ingredients: Dark news, a habit of monitoring it on the phone, and feeds built for the attention economy, together "creating emotional drain through a flow which users find hard to get out of." Only the last ingredient was designed by someone.

Where the word came from

The word is younger than it feels. A finance reporter spotted it in a social media post from October 2018, per Merriam-Webster's article on the term, and it spread in the spring of 2020, when lockdowns put a lot of people at home with a phone and a pandemic to follow. The Macquarie Dictionary named it word of the year for 2020, per its announcement. Merriam-Webster added it in September 2023.

Three items from the 2022 scale separate doomscrolling from reading:

• You refresh for bad news. "I constantly refresh my newsfeeds to see if something bad happened."

• You stay up for it. "I stay up late at night trying to find more negative news."

• You feel braced. "I feel tense as if something bad is going to happen soon."

Scored 1 to 7, the 502 people in the original study averaged 2.10, which the authors called "less prevalent than anticipated, but variable." Fewer people doomscroll than the word's popularity suggests, and scores varied widely from person to person.

Is doomscrolling bad?

Yes, on every measure the studies have taken. The effects are small to moderate, and most of the evidence can't prove cause.

Mood has the strongest evidence. A pair of randomized experiments in PLOS ONE in 2021 gave people 2 to 4 minutes of pandemic news, 299 of them on a live social feed and 603 on video. Compared with people shown nothing, that dose lowered positive mood right away, explaining about 3% to 6% of the variation in how people felt. A few minutes was enough.

A 2022 diary study in Psychological Trauma followed 61 young adults in Vermont for 30 days from April 2020. On days with social media, depression scores ran about 0.44 of a standard deviation higher and post-traumatic stress scores about 0.36 higher. Traditional news showed no significant link. The feed, not the news, looks like the active ingredient.

Anxiety and depression rest on one-time surveys. The 2022 scale paper found a correlation of 0.31 with anxiety across 27 countries, and a 2024 study in BMC Psychiatry of 2,885 Chinese students found 0.35 with depression and 0.37 with anxiety. Those links share roughly 9% to 14% of the variation, and a survey taken once can't tell doomscrolling that lowers mood from low mood that leads to doomscrolling.

Sleep is the cost people notice in themselves. In a February 2026 American Academy of Sleep Medicine survey of 2,007 US adults, 38% said news on a phone or tablet before bed made them sleep worse, and 46% of 18 to 24 year olds said so. A December 2025 study of 663 adults in Türkiye found a small, statistically reliable link to poor sleep quality.

Two 2026 reviews, of 17 studies and 50 studies, found the same links with anxiety, depression, stress, and lost productivity. Only the 2021 experiments earn a causal verb, and only for a few minutes of mood. None of the effects is large. All of them recur every evening, and a small cost paid nightly for years adds up.

Does quitting the news work?

The evidence says no. A 2022 experiment in Humanities and Social Sciences Communications paid some of its 803 US participants to avoid news for a week. Their well-being didn't improve, and they knew no less than the people who kept reading. News made up roughly 3% of their online information diet, so removing it removed almost nothing.

Cutting back works better. A 2025 meta-analysis from the University of New England in Australia pooled 10 randomized trials with 1,491 people. Less social media eased depression symptoms, an effect the authors call small, and reducing use showed twice the effect of stopping entirely, though the gap wasn't statistically significant. The urge to check doesn't take a week off. A smaller, fixed dose gives it somewhere to go that ends.

What to do instead

A read with a last line:

• One newspaper front page, print or digital, read top to bottom and closed.

• One morning brief from a source you chose the night before. A feed chooses for you based on what you clicked, and a 2023 study of 22,743 headline tests found each additional negative word raised the click-through rate by 2.3%.

• One weekly longread on the story you care most about, instead of a trickle of updates.

• Anything with pages. A 2006 University of Pennsylvania study on portion size found people treat one unit as the right amount, whatever the unit is. A feed has no unit, so the stop has to come from fatigue, a dead battery, or a mood bad enough to make you look up.

How to stop doomscrolling

We call the method the fixed read, and our how-to post lays out its four steps.

1. Set the stop before you start. Pick a point you can see coming, the last line of one article or the end of one brief, and pick it the night before while you're calm. A 2025 Ulm University study of 72 people found infinite scroll is designed for long sessions, and prompts that interrupt it help at first and then wear off as people get used to them. A decision made beforehand holds.

2. Keep the news and shrink the dose. One read a day, the same size every time.

3. Choose the source before you see a headline. Open it directly. The feed has learned which headlines you can't resist, and they're the negative ones. Our post on curated news vs algorithmic feeds explains why.

4. Protect the stop in your low moments. The Ulm study tracked people for seven days and found that when they felt low and were at home, they took longer to stop after a prompt. If your worst scrolls happen on the couch after a hard day, read somewhere you're usually steady: The train, a lunch bench, or your desk before work. When you can't name the last thing you read, close it there. One item on the 2022 scale reads, "I lose track of time when I read bad news on social media." Noticing is the cue.

For some people the checking itself is the problem. A 2022 Texas Tech study put 16.5% of a national US sample in a severely problematic news consumption group, and the team's 2024 replication put 7.5% there, with worse mental health both times. If you can't bring the checking down, or it's costing you sleep, talk to a doctor or therapist.

We built informed.now as one version of the fixed read: A handful of fact-only summaries each weekday morning, each linked to its primary source, and then it ends. Any format with a last line does the same job.

What's left when the scroll ends

In a Pew Research Center survey of 3,560 US adults, published in February 2026, 52% said they're worn out by the amount of news. Doomscrolling is what happens when the wish to stay informed meets a format with no bottom. Tonight something bad will have happened somewhere, and the phone will be within reach. A read that ends takes the decision out of the moment you're least equipped to make it.