Mental health statistics: what the research actually shows

How common are intrusive thoughts?

More than 90% of people with no diagnosis at all report unwanted thoughts that feel completely at odds with who they are. Rachman and de Silva showed as far back as 1978 that people without OCD have obsessional thoughts much like those of patients, differing mainly in how often they come and how much they hurt, and David Clark's 2005 edited volume gathered the research that followed. Very few results in cognitive psychology have been replicated that often.

Radomsky et al. (2014) took the question to 777 people at 15 sites across 13 countries and 6 continents using the International Intrusive Thoughts Interview Schedule, and 93.6% reported at least one in the previous three months. Doubt and checking came up almost everywhere. Harm content, contamination worries and taboo sexual or blasphemous material all appeared too, with the cultural setting deciding which of them felt most shameful to admit rather than whether they happened at all.

What separates an ordinary mind from OCD is never the thought itself or how often it turns up. It's what happens next. Obsessive-compulsive disorder requires compulsive responding, real distress and functional impairment running past an hour a day, and none of that follows from simply having the thought. The intrusive thoughts calculator shows how your own pattern of unwanted thoughts sits against clinical population data, and the phobia baseline calculator covers the related question of how anxiety amplifies unwanted mental content.

How often do people cry?

Women cry roughly 3 to 5 times a month and men 1 to 2, according to the literature review Vingerhoets et al. published in 2000, which pulled together the diary and survey studies on adult crying. Diaries that American students kept for Szabo and Frey in 1981 came out at 5.3 episodes a month for women and 1.5 for men. Later studies in Britain and Israel found the same gap at lower counts.

The size of that gap moves depending on where you ask. Vingerhoets and colleagues ran the question across 37 countries and found that people in wealthier, more democratic and more individualistic countries reported crying more often, and that the gap between women and men was wider in the wealthier ones (Van Hemert et al., 2011). Cultural permission seems to be doing most of the work here, rather than anything fixed in the body.

No nationally representative dataset for crying exists in the way one does for sleep or blood pressure, so every figure here is an indicative range and nothing tighter. Individual variation swamps the group averages, and plenty of men out-cry the average woman. Age, hormonal transitions and personality all pull on the number. The crying frequency calculator puts your own frequency against survey data by age and gender, and for the people who cry most of all, the baby crying calculator maps the normal newborn curve.

Is it normal not to have an inner monologue?

Yes, and it's commoner than the question implies. Some people report almost no inner verbal thought, a pattern Nedergaard and Lupyan studied in 2024 under the name anendophasia, and Russell Hurlburt, whose sampling work at the University of Nevada has tracked inner experience for decades, wrote in 2026 that people with little or no inner speech probably turn up frequently. His method, Descriptive Experience Sampling, beeps participants at random through the day and asks what was in their head at that exact instant, which is a very different question from asking someone to describe their thinking in the abstract.

People with anendophasia are not thinking less. They think in images and spatial arrangement, or in a wordless knowing that arrives already complete. A wider group has inner speech as an occasional visitor rather than a permanent tenant, though nobody has yet put a reliable population percentage on either group. When the topic went round social media in 2020, the shock for most people was learning that something they had assumed was universal is nothing of the sort.

Inner experience runs along a continuum of frequency and vividness, and the two-camps framing that gets shared online does not survive contact with the data. Descriptive experience sampling keeps finding the same five kinds of inner experience, inner speaking, inner seeing, unsymbolized thinking, feelings and sensory awareness, and in 2019 the Nevada team built a questionnaire to estimate how often each one turns up in a given person (Heavey et al., 2019). The inner monologue calculator shows how your thinking style compares with the population.

How common is anxiety?

Generalised Anxiety Disorder affects an estimated 3 to 6% of the population at any given moment and 5 to 9% at some point across a lifetime, which puts it among the most common diagnosable conditions in medicine, and the GAD-7 is the seven-item questionnaire clinicians and researchers use to measure it. It was built to be quick enough that a GP can hand it over in a waiting room.

Most people score well below the clinical line. In a German national sample of 5,030 adults the average GAD-7 score was about 3 out of 21, and only about 5% scored 10 or more, the usual screening cut-off (Löwe et al., 2008). The 2017 Global Burden of Disease study put anxiety disorders as a group at about 284 million people worldwide. Phobias, social anxiety disorder and panic disorder each carry their own prevalence profile, and the phobia figures are set out further down this page.

Anxiety and intrusive thoughts feed each other in both directions. A high baseline of anxiety makes unwanted mental content turn up more often and land harder when it does, and the content then sets off another round of anxiety. Clinical GAD is marked out by persistence, reach and cost rather than by the presence of worry, which nearly everybody carries. For the unwanted thought content that so often travels with it, see the intrusive thoughts calculator, and the childhood trauma test covers early adversity, which shows a strong association with anxiety in adulthood.

What are the most common phobias?

Specific phobias affect roughly 7 to 9% of the population in any 12-month period, which puts them ahead of every other anxiety condition, and the two datasets behind that figure are the National Comorbidity Survey Replication published by Kessler et al. in Archives of General Psychiatry in 2005 and the WHO World Mental Health Surveys that Wardenaar et al. analysed in 2017. The NCS-R drew on a nationally representative US sample; the WHO pooling across many countries came out lower, at 7.4% lifetime and 5.5% over 12 months.

Phobias start earlier than any other anxiety condition, with a median onset at 8 years old. Only 23% of people carrying a 12-month diagnosis ever get treatment for it, and the reason is mostly practical. You can arrange a life around never going near a spider or a needle, and most people quietly do, so the phobia never presents itself as a problem that needs solving.

The gender split here is one of the steadiest findings in anxiety epidemiology. Lifetime prevalence runs at about 9.8% in women against 4.9% in men, a 2:1 ratio that holds across cultures and phobia subtypes, widest for animal phobias and narrowest for blood-injection-injury. Whether that reflects a real difference in fear or a difference in who will admit to a fear is not settled. The phobia baseline calculator shows how common your particular fear is against the population data.

How common is ADHD in adults?

About 2.8% of adults meet DSM-IV criteria for current ADHD, according to Fayyad et al. (2017), who pooled 26,744 interviews from 20 WHO World Mental Health surveys. Averages hide a wide spread, from 3.6% in high-income countries to 1.4% in low and lower-middle income ones. The WHO Adult ADHD Self-Report Scale those surveys lean on was developed with researchers at Harvard Medical School and validated across several countries.

Country-to-country differences say as much about diagnostic practice as about brains. The same WHO surveys found 7.3% in France, 5.2% in the United States and 0.8% in Poland, and Kessler et al.'s clinical follow-up of the US NCS-R sample put current adult ADHD at 4.4%. Diagnosis rates in the UK and Australia are climbing fast, which reads as an epidemic if you squint at the wrong chart. Surveys like these measure symptoms directly rather than counting diagnoses, and it's the diagnosis count that has been climbing.

Women are the group this system has served worst. The criteria were built from studies of hyperactive boys, and an inattentive presentation that disrupts nobody's classroom trips no alarms at all, so women are routinely diagnosed a decade or more after men with the same condition. A first diagnosis in someone's 30s or 40s is common, and it often follows their own child's. The ADHD screening quiz gives an evidence-based screening result, and the autism screening quiz covers the condition that most often travels alongside it.

How common is autism?

Around 1 in 31 children in the United States, or 3.2%, are identified with autism spectrum disorder in the CDC's 2025 report, which draws on 2022 surveillance of 8-year-olds from its Autism and Developmental Disabilities Monitoring Network. That is up from 1 in 150 in the year 2000, 1 in 54 in 2016 and 1 in 36 in 2020, and the shape of that climb is where the whole public argument sits.

Adult prevalence comes out lower, at roughly 1 to 2%, for reasons that are historical rather than biological. Nobody was looking for autism in the people now in their 50s and 60s, so it went unrecorded. The male-to-female diagnostic ratio has run at about 3 to 4 to 1 and is narrowing as clinicians get better at spotting it in women and girls, who are likelier to have built masking strategies that keep autistic behaviours out of sight and diagnosis out of reach.

Most of the rise in the headline figure is the criteria and the counting shifting under it. DSM-5 folded several separate diagnoses into one spectrum in 2013, clinical awareness has climbed steeply, and screening programs have widened their reach. The autism screening quiz shows how your answers to the validated AQ-10 questions compare with the population, and since autism and ADHD co-occur in 50 to 70% of autistic people, the ADHD screening quiz is the natural companion to it.

How common is loneliness?

About 1 in 4 adults in Great Britain (27%) feel lonely at least some of the time, and 7% often or always, according to Office for National Statistics data from December 2023, and AARP's 2025 survey of Americans aged 45 and over found 42% of men and 37% of women lonely. In June 2025 the WHO Commission on Social Connection put the global figure at 1 in 6 people in its report From loneliness to social connection.

Loneliness is more common in younger adults than older ones, which is the reverse of what nearly everybody assumes. The ONS reports that pattern in Britain, and AARP found the same tilt within the over-45s, with 46% lonely at 45 to 59 against 35% at 60 and over. Younger adults live alone more and move more often for work or study, and they have not yet accumulated the decades of steady acquaintance that hold an older person's week together.

Holt-Lunstad et al. (2015) pooled 70 studies covering 3.4 million participants and found that loneliness raised the risk of early death by 26%, and social isolation by 29%. Those are the sort of numbers usually attached to obesity. The UCLA Loneliness Scale is the standard validated measure, and the loneliness scale puts your score against the population distribution from the Health and Retirement Study. Loneliness and friend count are separate measures, and the friendship count calculator and adult friendship statistics cover the second.

How common is childhood trauma?

63.9% of US adults report at least one adverse childhood experience and 17.3% report four or more, according to the CDC's analysis of Behavioral Risk Factor Surveillance System data from all 50 states for 2011 to 2020, covering 264,882 people (Swedo et al., 2023). The original ACE study, a CDC and Kaiser Permanente collaboration led by Vincent Felitti and Robert Anda and published in the American Journal of Preventive Medicine in 1998, asked 9,508 adults about seven categories of childhood adversity and found a graded relationship between the score and adult health.

That dose-response curve has held up better than almost anything else in public health research. Each extra category of adversity lifts the statistical risk of depression, substance misuse, heart disease and a shortened life. What gets lost whenever the score goes viral is that these are population-level associations and a high number predicts nothing about any one person. Stable relationships and access to mental health care both measurably blunt the effect.

ACE scores are not spread evenly across the population. In the CDC data, four or more ACEs were most common among American Indian and Alaska Native adults (32.4%) and multiracial adults (31.5%), against 17.3% overall, and those gaps track structural conditions rather than anything about the people inside them. The childhood trauma test compares your ACE score with the US adult population, and the intrusive thoughts calculator covers one of the ways early adversity shows up in adult mental life.

How common is narcissism?

6.2% of US adults meet lifetime criteria for Narcissistic Personality Disorder, according to Stinson et al. (2008), the largest epidemiological study of personality disorders ever run in the country, and the split by sex is 7.7% of men against 4.8% of women. Narrower estimates that insist on demonstrated functional impairment put the figure closer to 1 to 3%. The diagnosis needs a pervasive pattern of grandiosity, hunger for admiration and absent empathy, all of it doing real damage.

The NPI-16, developed by Ames, Rose and Anderson in 2006, measures the trait rather than the disorder, and the trait runs through everybody. An average adult scores about 7 to 8 out of 16, so half the population sits above the midpoint with nothing wrong with any of them. Confidence, ambition and a working amount of self-regard all load onto this scale. Scores drift downward across the adult lifespan, and 18 to 25 year olds score highest of any age band.

Almost everyone who types "am I a narcissist" into a search box lands in the normal range. The disorder involves very little insight into one's own effect on other people, which means the people with the most severe presentation are the least likely to be asking. Put your own answers through the narcissism test to see where an NPI-16 result sits against the general population, and the childhood trauma test covers how early experience shapes adult personality.