Mental health statistics: what the research actually shows

How common are intrusive thoughts?

The foundational study in this area is Rachman and de Silva (1978), which found that more than 90% of non-clinical participants reported experiencing unwanted intrusive thoughts that felt out of character with their values. David Clark (2005) synthesised subsequent research and confirmed this figure holds across clinical and community samples. The near-universality of intrusive thoughts is one of the most replicated findings in cognitive psychology.

Radomsky et al. (2014), using the International Intrusive Thoughts Interview Schedule across 777 participants in 15 sites, 13 countries, and 6 continents, found that 93.6% of non-clinical adults reported at least one intrusive thought in the preceding three months. Thematic categories include doubt and checking (nearly universal), harm-related content, contamination concerns, and taboo sexual or blasphemous content. Cultural context shapes which themes feel most taboo, but the intrusive thought mechanism itself is universal.

The critical clinical distinction is not the presence, frequency, or content of intrusive thoughts, but the response to them. The presence of intrusive thoughts is not diagnostic of OCD. Obsessive-compulsive disorder requires compulsive responding, significant distress, and functional impairment exceeding one hour per day. Use the intrusive thoughts calculator to see how your pattern of unwanted thoughts compares to clinical population data. For related context on how anxiety amplifies unwanted mental content, see the phobia baseline calculator.

How often do people cry?

The most widely cited population-level figures on crying frequency come from William H. Frey II, whose research from the 1980s estimated that women cry approximately 5.3 times per month and men approximately 1.3 times per month. Vingerhoets et al. (2000) published a comprehensive review and cross-cultural analysis that produced slightly lower but similarly structured estimates: women cry 3 to 5 times per month on average, men 1 to 2 times. Both bodies of work confirm a substantial gender difference in frequency.

A cross-cultural study by Vingerhoets and colleagues spanning 37 countries found that the gender gap in crying frequency was not constant. Wealthier, more individualistic societies with higher norms around emotional expressiveness reported higher crying rates for both sexes and a wider gender gap. This suggests that cultural permission to express emotion, rather than fundamental physiological differences, drives much of the variation.

Individual variation is enormous, and no large-scale nationally representative dataset exists for crying frequency in the way that such datasets exist for sleep or physical health. The figures above are indicative ranges rather than precise percentiles. Age, hormonal transitions, and personality traits all contribute to individual differences. Use the crying frequency calculator to compare your own frequency to survey data by age and gender.

Is it normal not to have an inner monologue?

Russell Hurlburt's Descriptive Experience Sampling (DES) research at the University of Nevada provides the most systematic data on inner speech. DES uses random beeper prompts throughout the day, asking participants to report their exact mental experience at the moment of the beep. The studies consistently find that inner speech occupies a median of only 20 to 25% of waking moments even in people who report having a frequent inner voice. The inner monologue is far less continuous than most people believe.

Approximately 15 to 16% of the population experience anendophasia: the near-total absence of inner verbal thought. These individuals do not hear a narrating voice in their head. They may think in images, spatial patterns, abstract knowing, or wordless feeling. A broader group, estimated at 25 to 30% in Hurlburt's studies, reports that inner speech is a rare or minor component of their waking experience rather than a regular feature. The 2020 viral discussion of inner monologue absence revealed how many people had never realised their experience differed from the assumed norm.

Inner experience exists on a continuum of frequency and vividness, not in two discrete categories. Deamer et al. (2021), published in Frontiers in Psychology, found enormous variation even among participants who reported having inner speech, with some experiencing it in nearly every sampled moment and others only occasionally. Use the inner monologue calculator to find out how your thinking style compares to the population.

How common is anxiety?

The GAD-7 (Generalised Anxiety Disorder 7-item scale) is the most widely used validated screening instrument for anxiety in clinical and population research. Generalised Anxiety Disorder (GAD) is estimated to affect 3 to 6% of the population at any given point, making it one of the most common diagnosable mental health conditions. Lifetime prevalence is higher: approximately 5 to 9% of adults will meet criteria for GAD at some point in their lives.

Subclinical anxiety symptoms, those that do not meet the full clinical threshold for GAD but are significant and persistent, are far more common. The 2017 Global Burden of Disease study estimated that anxiety disorders as a group affected approximately 284 million people worldwide. Specific anxiety conditions including phobias, social anxiety disorder, and panic disorder each have distinct prevalence profiles covered elsewhere on this page.

Anxiety and intrusive thoughts have a bidirectional relationship. High baseline anxiety increases the frequency and salience of unwanted intrusive mental content, and intrusive thoughts in turn activate anxiety responses. Population data shows that subclinical anxiety is a near-universal life experience; clinical GAD is characterised by its persistence, pervasiveness, and functional impact. For related data on unwanted thought content that commonly co-occurs with anxiety, see the intrusive thoughts calculator and the childhood trauma test, as adverse early experiences are significantly associated with adult anxiety.

What are the most common phobias?

The most comprehensive epidemiological data on specific phobias comes from two sources: the National Comorbidity Survey Replication (NCS-R), published by Kessler et al. in Archives of General Psychiatry (2005), and the WHO World Mental Health Surveys analysed by Wardenaar et al. (2017). The NCS-R, based on a nationally representative US sample, found that specific phobias affect approximately 7 to 9% of the population in any given 12-month period, making them among the most prevalent of all anxiety conditions.

Wardenaar et al. (2017) using WHO data across multiple countries found a lifetime prevalence of 7.4% and a 12-month prevalence of 5.5%. The median age of onset is 8 years old, the earliest of any anxiety condition. Only 23% of those with a 12-month specific phobia diagnosis ever receive treatment, largely because functional avoidance allows most people to manage around the phobia for much of daily life.

Gender differences are among the most consistent findings in anxiety epidemiology. Lifetime prevalence is approximately 9.8% in women versus 4.9% in men, a 2:1 ratio that holds across cultures and phobia subtypes. Animal phobias show the largest gender gap; blood-injection-injury phobias show the smallest. Use the phobia baseline calculator to see how common your specific fear is compared to population data.

How common is ADHD in adults?

The WHO Adult ADHD Self-Report Scale (ASRS) was developed in collaboration with researchers at Harvard Medical School and validated across multiple countries. Fayyad et al. (2017), using WHO World Mental Health Survey data from 20 countries, found an overall adult ADHD prevalence of approximately 2.8% using strict DSM-IV criteria. When broader diagnostic criteria are applied, estimates rise to approximately 5% globally. These figures represent the most rigorous cross-national data available.

Prevalence estimates vary considerably by country and diagnostic criteria. The United States, which has broader diagnostic and prescribing practices, reports higher rates (4.2 to 4.4% by NCS-R data, Kessler et al. 2005) than many European countries. Diagnosis rates are also rising rapidly in the UK and Australia, primarily reflecting improved recognition rather than true prevalence change. The population prevalence has remained relatively stable at 2.8 to 4.4% across studies spanning two decades; what has changed is the proportion receiving diagnosis.

ADHD is significantly underdiagnosed in women. Diagnostic criteria were historically developed from studies of hyperactive boys, and women with predominantly inattentive presentations are often diagnosed a decade or more later than men with the same condition. Many women receive a first diagnosis in their 30s or 40s, often following their children's diagnosis. For an evidence-based screening result, use the ADHD screening quiz. For related context on conditions that commonly co-occur with ADHD, see the autism screening quiz.

How common is autism?

The CDC's Autism and Developmental Disabilities Monitoring (ADDM) Network provides the most comprehensive and consistently collected data on autism prevalence in the United States. The 2023 update reports that approximately 1 in 36 children (2.8%) are identified with autism spectrum disorder, based on data from 2020. This figure represents a marked increase from 1 in 150 in the year 2000 and 1 in 54 in 2016.

Adult prevalence estimates are lower, at approximately 1 to 2%, primarily because diagnosis rates were much lower historically. The gap between childhood and adult prevalence reflects underdiagnosis in older cohorts rather than a genuine difference in the conditions those cohorts experience. The diagnostic ratio of males to females has historically been approximately 3 to 4 to 1, but this is narrowing as recognition of autism in women improves. Women and girls on the autism spectrum are more likely to develop masking strategies that suppress visible autistic behaviours, leading to later and less frequent diagnosis.

The rising prevalence figures largely reflect broadened diagnostic criteria (DSM-5 unified the spectrum under a single diagnosis in 2013), improved clinical awareness, and expansion of screening programs, rather than a genuine epidemic of new cases. Use the autism screening quiz to see how your pattern of responses to validated AQ-10 questions compares to the population. For context on the overlap between autism and ADHD, which co-occur in 50 to 70% of autistic individuals, see the ADHD screening quiz.

How common is loneliness?

The UK's Office for National Statistics (ONS) 2023 data indicates that approximately 1 in 4 adults in the UK report feeling lonely often or always. The Survey Center on American Life reported in 2025 that 42% of American men and 37% of American women report feeling lonely. These figures align with the WHO's 2025 Global Status Report on Social Connection, which designated loneliness a global public health priority.

One of the most counterintuitive findings in loneliness research is the age distribution. Multiple large datasets, including the Health and Retirement Study and the ONS surveys, consistently show that adults aged 18 to 34 report higher rates of loneliness than adults aged 65 and over. The popular assumption that older adults are the most lonely demographic is not supported by the data. Young adults are more likely to live alone, move frequently for education or work, and lack the stable community structures that older adults have built over decades.

Holt-Lunstad et al. (2015), in a meta-analysis of 70 studies covering 3.4 million participants, found that loneliness increases the risk of premature death by 26% and social isolation by 29%. These figures place loneliness in the same risk category as obesity for long-term health outcomes. The UCLA Loneliness Scale is the most widely used validated measure: use the loneliness scale to compare your score to the population distribution from the Health and Retirement Study.

How common is childhood trauma?

The Adverse Childhood Experiences (ACE) Study was a landmark collaboration between the CDC and Kaiser Permanente, led by Vincent Felitti and Robert Anda and published in the American Journal of Preventive Medicine in 1998. The original study surveyed 17,337 adults about 10 categories of childhood adversity and found a strong, graded relationship between ACE score and adult health outcomes. More recent data from the CDC's Behavioral Risk Factor Surveillance System (BRFSS) survey, with approximately 444,000 respondents, shows that 57.8% of US adults report at least one ACE.

The dose-response relationship is one of the most robust findings in public health research. As ACE scores increase, so do statistical risks for depression, substance misuse, heart disease, and reduced life expectancy. Adults with an ACE score of 4 or more, approximately 15.6% of the population, show significantly elevated risk across multiple health domains compared to those with a score of zero. Critically, these are population-level risk associations, not individual predictions: protective factors including stable relationships, access to mental health care, and community belonging significantly buffer the effects of high ACE scores.

ACE prevalence varies substantially by demographic group. It is higher among adults with lower household incomes, among Black adults compared to white non-Hispanic adults, and among LGBTQ adults. These disparities reflect structural inequalities rather than individual characteristics. Use the childhood trauma test to compare your ACE score to the US adult population. For context on how early adversity shapes adult anxiety patterns, see the intrusive thoughts calculator.

How common is narcissism?

Narcissistic traits exist on a continuum running through the entire population. Clinical Narcissistic Personality Disorder (NPD) is a distinct diagnosis requiring a pervasive pattern of grandiosity, need for admiration, and lack of empathy causing significant functional impairment. Stinson et al. (2008), in the largest epidemiological study of personality disorders in the United States, found that 6.2% of the population meets lifetime criteria for NPD, with a gender difference: 7.7% of men versus 4.8% of women. Some narrower estimates using impairment criteria place the figure at 1 to 3%.

The NPI-16 (Narcissistic Personality Inventory), developed by Ames, Rose, and Anderson (2006), measures subclinical narcissistic personality traits across the full population. The average adult scores approximately 7 to 8 out of 16, meaning half the population scores above the midpoint. Moderate narcissistic traits, including healthy self-esteem, confidence, and ambition, are normal and not predictive of clinical problems. Problems arise when traits become extreme and rigid. NPI scores decline gradually across the adult lifespan, with young adults (18 to 25) typically scoring highest.

Most people who search "am I a narcissist" score in the normal range. Ironically, clinical NPD is characterised by limited self-reflection about interpersonal impact, which means those with the most severe presentation are least likely to be asking the question. Use the narcissism test to see where your NPI-16 result sits compared to the general population. For context on how childhood experiences shape adult personality traits, see the childhood trauma test.