OCD Prevalence: How Common Is OCD? Lifetime Risk, Age, Sex, and Global Estimates

Obsessive-compulsive disorder (OCD) is common enough to be a major public-health concern, but there is no single prevalence percentage that is correct for every population, age group, diagnostic system, or time window. A practical evidence-based orientation is that modern studies often place lifetime OCD prevalence in the low single digits: roughly 2% to 3% in several major estimates, with some recent cross-national surveys reporting higher values. Current or 12-month prevalence is generally lower than lifetime prevalence, although the exact gap varies substantially by study design.

The most important reason the numbers look inconsistent is methodological rather than mysterious. Epidemiologists may be measuring lifetime prevalence, 12-month prevalence, point prevalence, obsessive-compulsive symptoms, or a full clinical disorder; they may use DSM-III, DSM-IV, DSM-5, ICD criteria, or different structured interviews; and their samples may cover one country, selected regions, adults only, or all ages. The newest global modeling study makes this especially clear: estimated lifetime prevalence changed materially depending on which diagnostic criteria were modeled. Jeong et al. (2026).

This article separates those estimates instead of averaging incompatible numbers. It also distinguishes population prevalence from an individual diagnosis. Prevalence describes how often OCD occurs in a population; it cannot determine whether a particular person has OCD. For that question, see our guides to OCD diagnosis and diagnostic criteria.

How Common Is OCD? The Short Answer

If you need a concise answer, OCD affects a minority of the population but is not rare. In the United States, the National Institute of Mental Health reports a 12-month prevalence of 1.2% among adults and a lifetime prevalence of 2.3%, based on the National Comorbidity Survey Replication. NIMH. A 2020 meta-analysis of 34 representative adult community studies estimated overall lifetime prevalence at 1.3% and current prevalence at 1.1%. Fawcett et al. (2020).

More recent international work produces higher estimates in some settings. A 2026 systematic review and Bayesian modeling study incorporating 112 studies estimated global lifetime prevalence at 2.57% under ICD-10 criteria, 3.10% under DSM-III, 2.28% under DSM-IV, and 3.21% under DSM-5. Jeong et al. (2026). Meanwhile, a 2025 World Mental Health survey analysis of 26,136 adults across ten countries reported a combined DSM-IV lifetime prevalence of 4.1% and 12-month prevalence of 3.0%. Stein et al. (2025).

Those values should not be treated as competing guesses about one fixed number. They answer related but different epidemiological questions using different samples, diagnostic frameworks, eras, and statistical methods. The most defensible public summary is therefore a range plus context, not a single universal percentage.

What Does ‘Prevalence’ Mean?

Prevalence is the proportion of a defined population that meets the study's case definition during a specified period. The denominator matters, the time window matters, and the case definition matters. A prevalence estimate is therefore inseparable from the sentence that explains how it was measured.

Lifetime prevalence

Lifetime prevalence asks whether a person has met criteria for OCD at any time up to the assessment. It accumulates episodes across the years already lived, which is why it is usually higher than point or 12-month prevalence. ‘Lifetime risk’ is often used casually as a synonym, but the two ideas are not identical. A cross-sectional lifetime-prevalence survey measures the proportion who have already had OCD by interview; a true lifetime-risk estimate would attempt to quantify the probability of developing OCD at any point across the entire lifespan, including future years for younger respondents.

12-month or past-year prevalence

Twelve-month prevalence counts people who met the study's OCD definition during the preceding year. It is especially useful for estimating the number of people likely to need care or experience current impairment within a recent period. In the U.S. NCS-R, this estimate was 1.2% among adults; in the newer ten-country WMH analysis, the combined estimate was 3.0%. Ruscio et al. (2010); Stein et al. (2025).

Point or current prevalence

Point prevalence refers to cases present at a particular time or within a very narrow current window. Definitions of ‘current’ differ across studies, so current prevalence should not be compared mechanically with a 12-month rate. The 2020 adult meta-analysis estimated current OCD prevalence at 1.1%, but reported moderate heterogeneity, meaning individual studies varied meaningfully around that summary. Fawcett et al. (2020).

Incidence

Incidence is different from prevalence. It measures new cases arising during a period among people who were initially at risk. A disorder can have relatively modest incidence but substantial prevalence when it begins early and persists for years. This distinction matters for OCD because onset often occurs relatively early and symptoms can follow a chronic or fluctuating course. For developmental timing, see when OCD starts; for what happens over time, see the course of OCD.

The Best Current Global OCD Prevalence Estimates

The 2026 global systematic review and modeling study

The broadest recent attempt to estimate global prevalence is the American Journal of Psychiatry study by Jeong and colleagues, published online in July 2026. The authors systematically searched major databases through June 8, 2025 and included 112 studies representative of national or subnational general populations. Rather than forcing studies based on different diagnostic systems into one pooled percentage, they modeled prevalence by diagnostic criteria. Jeong et al. (2026).

The resulting lifetime estimates were 2.57% for ICD-10, 3.10% for DSM-III, 2.28% for DSM-IV, and 3.21% for DSM-5. The spread is epidemiologically important: changing the diagnostic framework shifts the estimated size of the population that meets criteria. The study also found that age-specific prevalence rose sharply during the teen years and peaked in the late twenties to early thirties. Jeong et al. (2026).

These are modeled global estimates, not the result of interviewing every region with one identical instrument in one year. Bayesian modeling is useful precisely because global data are uneven, but it does not erase gaps in primary epidemiological coverage. The values should therefore be read as evidence-informed estimates with uncertainty, not as census counts.

The 2025 World Mental Health surveys

A complementary picture comes from the World Mental Health surveys. Stein and colleagues analyzed face-to-face survey data collected between 2005 and 2019 from 26,136 adults across ten countries. Six surveys came from high-income countries and four from low- or middle-income countries; some were nationally representative and others represented specific regions. All used a coordinated DSM-IV assessment framework. Stein et al. (2025).

Across the combined sample, lifetime OCD prevalence was 4.1% and 12-month prevalence was 3.0%. Country or regional estimates varied widely: lifetime prevalence ranged from 0.4% in Murcia, Spain, to 5.5% in Shenzhen, China, while 12-month prevalence ranged from 0.3% to 4.5%. The authors also reported higher combined estimates in the included low- and middle-income settings than in the high-income settings. Stein et al. (2025).

The WMH result is valuable because it uses coordinated cross-national methods, but it is not synonymous with a worldwide prevalence rate. Ten countries cannot represent every country, and regional samples cannot automatically be generalized to entire nations. Its higher estimate also illustrates why publication year alone does not explain prevalence differences: sampling, interview algorithms, diagnostic implementation, case severity, and the populations included all matter.

The 2020 adult community meta-analysis

Fawcett and colleagues synthesized 34 representative adult community studies using DSM- or ICD-based diagnostic interviews. Their aggregate estimates were 1.3% for lifetime prevalence, 1.1% for current prevalence, and 0.8% for period prevalence. The authors found moderate heterogeneity, so the pooled numbers should be interpreted as central tendencies across diverse studies rather than fixed constants. Fawcett et al. (2020).

Placed beside the 2026 model and 2025 WMH analysis, the meta-analysis shows why a high-quality prevalence article must disclose the evidence source. ‘OCD affects 1.3% of people’ is accurate as a description of that meta-analysis. It is incomplete as a timeless global claim.

How Common Is OCD in the United States?

For U.S. adults, the most widely cited nationally representative estimates come from the National Comorbidity Survey Replication, conducted in 2001–2003. The National Institute of Mental Health currently summarizes the findings as 1.2% past-year prevalence and 2.3% lifetime prevalence among adults. In the past-year data, prevalence was 1.8% among females and 0.5% among males. NIMH.

The NCS-R OCD analysis by Ruscio and colleagues examined a subsample of 2,073 respondents assessed for lifetime DSM-IV OCD. It likewise estimated lifetime prevalence at 2.3% and 12-month prevalence at 1.2%. Crucially, 28.2% of respondents reported having experienced at least one assessed obsession or compulsion at some point in life, yet only a small fraction met the full disorder criteria. Ruscio et al. (2010).

That gap is one of the most useful findings in OCD epidemiology. Intrusive thoughts, checking, ordering, or other repetitive experiences can occur outside a clinical disorder. A population percentage for obsessive-compulsive experiences is not interchangeable with the prevalence of diagnosed OCD. The distinction depends on the full symptom pattern, distress, time consumption, impairment, exclusion of other explanations, and the diagnostic framework used.

The NCS-R remains important, but its fieldwork is more than two decades old and used DSM-IV criteria. It should be presented with its date and methodology rather than as a measurement taken from today's U.S. population. The newer global evidence does not invalidate it; it gives readers additional context for how estimates vary across time and design.

OCD Prevalence by Age

Age changes the way prevalence should be interpreted because OCD often begins before middle adulthood and lifetime prevalence accumulates across years lived. The 2026 global modeling study found age-specific prevalence increasing sharply during the teenage years and peaking in the late twenties to early thirties. Jeong et al. (2026). The WMH analysis likewise found early onset, with more than 80% of cases beginning by early adulthood. Stein et al. (2025).

Children and adolescents

OCD clearly occurs in childhood and adolescence, but pediatric prevalence cannot be reduced to one age-neutral number. Studies define youth age bands differently, use different interviews and informants, and may capture very different stages of the developmental risk curve. A contemporary review emphasizes that pediatric OCD is a clinically significant condition with developmental features that affect recognition and assessment. Stiede et al. (2024). For the clinical presentation in younger people, see OCD in children.

The age curve also helps explain an apparent paradox: early-onset OCD is clinically important even when prevalence in the youngest age bands is lower than in young adults. Prevalence at age ten and prevalence at age twenty-nine describe different cumulative exposure to the period of highest onset risk. They should not be read as evidence that childhood OCD is unimportant.

Young and middle-aged adults

U.S. NIMH data show past-year prevalence of 1.5% at ages 18–29, 1.4% at ages 30–44, 1.1% at ages 45–59, and 0.5% at age 60 or older. NIMH. These figures are specific to the NCS-R survey period and DSM-IV-based interview, but they align directionally with the newer global model's finding that prevalence peaks in early adulthood before declining at older ages.

For the way symptoms, work, relationships, and treatment appear in this life stage, see OCD in adults.

Older adults

Lower observed prevalence in older age groups does not mean OCD cannot persist into late life or first come to clinical attention there. Cohort effects, mortality, recall, changes in symptom expression, institutionalization, differential help-seeking, and survey participation can all affect age-specific estimates. A population curve also cannot tell whether an older individual's symptoms are longstanding OCD, a late-recognized presentation, or another condition requiring differential assessment. See OCD in older adults for the clinical distinctions.

Is OCD More Common in Women or Men?

Across adult community studies, women appear to have a higher prevalence on average. The 2020 meta-analysis found that women were approximately 1.6 times as likely as men to experience OCD; pooled lifetime prevalence was 1.5% in women and 1.0% in men. Fawcett et al. (2020). In the U.S. NCS-R past-year estimate summarized by NIMH, prevalence was 1.8% for females and 0.5% for males. NIMH.

Development complicates that adult pattern. A review of epidemiological and clinical sex/gender differences concluded that OCD may be more common among males in childhood, while females are more commonly affected in adolescence and adulthood. The authors also emphasized that findings across studies remain mixed. Mathes et al. (2019).

The terminology requires care. Many epidemiological datasets record a binary male/female variable and describe it as sex, gender, or both; those studies generally cannot answer detailed questions about gender identity. Their categories should not be stretched beyond what was actually measured. Similarly, a prevalence difference is not itself an explanation. It does not establish whether biological, developmental, social, diagnostic, cultural, or help-seeking mechanisms caused the observed pattern.

Does OCD Prevalence Differ Across Countries and Regions?

Yes, measured prevalence varies substantially across settings. Historical cross-national surveys using broadly similar DSM-III methods found annual OCD prevalence mostly around 1.1% to 1.8% across several countries, with Taiwan lower at 0.4%. Cross National Collaborative Group (1994). The newer WMH analysis found an even wider range, with lifetime estimates from 0.4% to 5.5% across its included surveys. Stein et al. (2025).

The 2026 global model also reported that lifetime prevalence was negatively correlated with the Socio-demographic Index, meaning modeled estimates tended to be lower in countries with higher SDI. Jeong et al. (2026). That correlation should not be converted into a causal story about wealth, modernization, or culture. Cross-country prevalence reflects a mixture of true population differences and measurement conditions: diagnostic instruments, translations, cultural interpretation of questions, willingness to disclose taboo thoughts, interviewer training, access to assessment, age structure, sampling frame, and case thresholds.

Global mental-health evidence is also geographically uneven. The 2025 WMH paper explicitly noted the shortage of representative OCD surveys in many non-Western and low- and middle-income settings. Stein et al. (2025). A country with fewer studies does not have less OCD by definition; it has less precise epidemiological knowledge.

Why Do OCD Prevalence Estimates Differ So Much?

The apparent disagreement is largely understandable once the study designs are separated. First, diagnostic criteria change. The 2026 model estimated substantially different lifetime prevalence depending on whether the case definition used ICD-10, DSM-III, DSM-IV, or DSM-5. Jeong et al. (2026). Even small changes in exclusion rules, insight requirements, symptom thresholds, and the handling of related disorders can alter who qualifies.

Second, interviews differ. A fully structured lay-administered interview is not identical to a clinician-administered diagnostic assessment. Skip logic can also matter: if a survey asks only a narrow set of gateway questions before deciding whether to administer the full OCD module, some presentations may be missed. Conversely, a highly sensitive symptom screen can classify many people above a cutoff without establishing a clinical diagnosis.

Third, the time window differs. Lifetime, 12-month, current, and point prevalence are distinct outcomes. Fourth, populations differ. Adult-only surveys cannot be compared directly with all-age estimates; national household samples differ from school surveys, regional samples, clinical records, online convenience samples, and specialty-clinic samples.

Fifth, disclosure matters. OCD can involve sexual, aggressive, religious, moral, or other taboo intrusive thoughts that people may hide because of shame or fear of being misunderstood. A structured anonymous or confidential epidemiological interview may identify cases that routine clinical records do not. At the same time, recall of past symptoms can be imperfect, especially in lifetime surveys. These opposing forces make prevalence estimation a measurement problem as well as a counting problem.

OCD Symptoms Are More Common Than OCD Disorder

One of the easiest epidemiological errors is to quote a study of obsessive-compulsive symptoms as if it measured OCD. The NCS-R illustrates the difference sharply: 28.2% of respondents in the OCD subsample reported at least one assessed lifetime obsession or compulsion, whereas 2.3% met lifetime DSM-IV criteria for OCD. Ruscio et al. (2010).

The 2025 WMH study found a similar conceptual gap. Across its ten-country sample, 13.6% reported lifetime obsessions or compulsions, while 4.1% met the study's lifetime DSM-IV OCD definition. Stein et al. (2025). The exact percentages differ, but the lesson is stable: an intrusive thought, ritual, checking tendency, or elevated questionnaire score is not automatically a clinical disorder.

This distinction is especially important when reading internet surveys or pandemic-era studies that report ‘OCD symptoms.’ A screening cutoff can be useful for identifying people who may benefit from further assessment, but screening prevalence is not diagnostic prevalence. The same rule applies at the individual level: a symptom can be real and distressing without telling us by itself which diagnosis, if any, best explains it.

Prevalence Is Not the Same as Cause, Heritability, or Individual Risk

Population prevalence describes frequency. It does not establish why OCD occurs. A higher observed rate in one demographic group or country cannot, by itself, identify the mechanism responsible. Etiological claims require different study designs, including genetic, longitudinal, family, neurobiological, and experimental research. For that evidence, see what causes OCD and whether OCD is hereditary.

Prevalence also cannot calculate one person's probability of developing OCD from a few traits. Family history, age, comorbidity, environmental exposures, and many other variables can alter risk, but population averages are not personalized predictions. Nor does being in a statistically lower-prevalence group rule out OCD when clinically significant obsessions and compulsions are present.

Is OCD Becoming More Common?

The current evidence does not support a simple conclusion that OCD itself has steadily become more common just because newer studies sometimes report higher prevalence. The methodological landscape has changed at the same time: diagnostic systems evolved, interview methods improved, awareness increased, surveys reached different populations, and newer analyses model previously sparse regions. Comparing a 1990s percentage with a 2026 modeled estimate as though both were identical thermometers would be misleading.

A genuine time-trend claim requires comparable repeated surveys using sufficiently similar sampling and measurement across calendar periods. Cross-study differences can suggest hypotheses, but they cannot cleanly separate a true increase in disorder prevalence from improved detection or changed definitions. The responsible formulation is therefore that modern estimates differ, sometimes substantially, and trend inference remains method-dependent.

How Epidemiologists Measure OCD

Representative sampling

A prevalence study aims to sample the population rather than only people who seek treatment. This is essential because treatment samples are shaped by access, affordability, recognition, referral, stigma, and willingness to seek care. Community surveys can therefore identify people who meet criteria but have never received an OCD diagnosis in routine health services.

Structured diagnostic interviews

Major epidemiological surveys use structured or semi-structured interviews that operationalize diagnostic criteria. The U.S. NCS-R used a modified World Health Organization Composite International Diagnostic Interview to generate DSM-IV diagnoses. NIMH. The WMH program likewise uses coordinated CIDI-based methods across countries. Stein et al. (2025).

Diagnostic threshold and impairment

OCD is more than the presence of a repetitive thought or behavior. Epidemiological case definitions attempt to determine whether obsessions and/or compulsions satisfy the relevant diagnostic threshold and whether the overall presentation meets the system's additional requirements. The details differ by diagnostic manual and survey implementation, which is why readers should follow the link from a prevalence number to its method rather than treating all ‘OCD’ labels as interchangeable.

Weighting and modeling

Representative surveys usually weight responses to account for sampling design and population structure. Global syntheses may go further, statistically modeling differences across criteria, age, geography, and data availability. Modeling can produce estimates for poorly observed populations, but the uncertainty of those estimates depends on the quality and distribution of the underlying studies.

How Common Is Clinically Significant OCD?

Prevalence alone does not show severity, but epidemiological studies demonstrate that OCD can be highly impairing. In NIMH's summary of U.S. adults with past-year OCD, 50.6% were classified as having serious impairment, 34.8% moderate impairment, and 14.6% mild impairment on the Sheehan Disability Scale. NIMH. Those severity proportions belong to people who met past-year OCD criteria in that survey; they should not be applied to everyone who experiences intrusive thoughts or rituals.

The public-health burden therefore has two dimensions: how many people meet criteria and how strongly the disorder affects functioning among those people. A condition can have a low-single-digit prevalence and still produce substantial disability because symptoms may be persistent, time-consuming, and disruptive across work, education, family life, relationships, and health. For the functional side of that question, see OCD and quality of life.

Which OCD Prevalence Number Should You Use?

Use the number that matches the question. For U.S. adults, the NIMH/NCS-R figures of 1.2% past-year and 2.3% lifetime prevalence remain the standard nationally representative reference, with the survey period and DSM-IV basis stated. NIMH. For a modern worldwide lifetime estimate, the 2026 systematic review is the most directly relevant source because it models global prevalence and explicitly separates diagnostic systems. Jeong et al. (2026).

For a cross-national adult survey estimate based on coordinated DSM-IV interviews, use the 2025 WMH figures of 4.1% lifetime and 3.0% 12-month prevalence, while stating that they come from ten countries and include both national and regional surveys. Stein et al. (2025). For a meta-analytic summary of representative adult community studies available through early 2017, use the 2020 estimate of 1.3% lifetime prevalence and 1.1% current prevalence. Fawcett et al. (2020).

A good article, clinical handout, or AI answer should rarely write only ‘OCD prevalence is X%.’ It should identify the population, timeframe, diagnostic framework, and source. That single habit prevents most of the confusion around OCD epidemiology.

What the Prevalence Evidence Does—and Does Not—Tell Us

The evidence is strong enough to conclude that OCD is a worldwide disorder found across ages and populations, that its measured lifetime prevalence is generally in the low single digits, that prevalence varies materially across studies and diagnostic systems, that symptoms are much more common than the full disorder, and that age and sex patterns exist at the population level. It is also strong enough to reject the older image of OCD as an exceptionally rare psychiatric condition.

The same evidence leaves important questions open. Many regions remain underrepresented in high-quality community surveys. Binary sex/gender reporting limits inference about gender-diverse populations. Cross-national differences remain difficult to separate into cultural, methodological, health-system, and etiological components. And rising awareness cannot be translated directly into a claim of rising disorder incidence without comparable longitudinal surveillance.

That uncertainty is not a weakness to hide. It is part of the answer. Epidemiology becomes more useful when readers can see which quantities are well measured, which are modeled, and which remain under-observed.

Frequently Asked Questions

What percentage of people have OCD?

There is no single universal percentage. Major modern estimates place lifetime prevalence from about 1.3% in an adult community meta-analysis to roughly 2.3%–3.2% in several U.S. and global diagnostic estimates, while a recent coordinated ten-country adult survey reported 4.1%. The correct number depends on population, timeframe, criteria, and method.

Is ‘1 in 40 people’ a reasonable estimate for OCD?

One in 40 equals 2.5%, which is a useful rough lifetime orientation and sits close to several modern estimates, including the U.S. NCS-R lifetime prevalence of 2.3% and the 2026 modeled ICD-10 estimate of 2.57%. It should still be presented as an approximation rather than a universal constant.

How common is OCD in a given year?

In the U.S. NCS-R, past-year adult prevalence was 1.2%. A newer WMH analysis across ten countries reported 3.0% 12-month prevalence. The difference reflects more than geography; the surveys also differ in period, samples, and implementation.

Is OCD more common in women?

Among adults, yes on average: the 2020 meta-analysis found women about 1.6 times as likely as men to experience OCD, with pooled lifetime prevalence of 1.5% versus 1.0%. Developmental patterns are more complex, with some evidence that males are more represented in childhood-onset OCD and females more represented from adolescence onward.

How common is OCD in children?

OCD occurs in children and adolescents, and the newest global modeling evidence shows prevalence rising sharply during the teen years. A single pediatric percentage can be misleading because estimates depend strongly on age band, informant, diagnostic method, and whether a study measures symptoms or a clinical disorder. OCD in children covers the developmental clinical picture in detail.

Does OCD become less common with age?

Observed past-year prevalence is lower in older adult groups in U.S. survey data, and the 2026 global model peaks in the late twenties to early thirties. That pattern does not mean symptoms necessarily disappear with age. Persistence, remission, cohort effects, mortality, recall, and measurement all contribute to age-specific prevalence.

Are intrusive thoughts as common as OCD?

No. Intrusive or obsessive-compulsive experiences are far more common than the full disorder. In the U.S. NCS-R, 28.2% of the assessed subsample reported at least one lifetime obsession or compulsion, compared with 2.3% meeting lifetime DSM-IV OCD criteria. A symptom is not a diagnosis.

Can a prevalence statistic tell me whether I have OCD?

No. Population prevalence cannot diagnose an individual. Diagnosis depends on the nature of obsessions and compulsions, their relationship to distress and functioning, time consumption, differential diagnosis, and the applicable clinical criteria. See how OCD is diagnosed for the assessment process.

References

Cross National Collaborative Group. (1994). The cross national epidemiology of obsessive compulsive disorder. Journal of Clinical Psychiatry, 55(Suppl), 5–10. PubMed.

Fawcett, E. J., Power, H., & Fawcett, J. M. (2020). Women Are at Greater Risk of OCD Than Men: A Meta-Analytic Review of OCD Prevalence Worldwide. Journal of Clinical Psychiatry, 81(4), 19r13085. https://doi.org/10.4088/JCP.19r13085.

Jeong, Y. D., Son, Y., Jeon, S., Cho, H., Ryuk, S. W., Jo, Y., Fond, G., Boyer, L., Smith, L., Cortese, S., Fusar-Poli, P., Yon, D. K., & Solmi, M. (2026). Global Prevalence of Obsessive-Compulsive and Related Disorders: A Systematic Review and Modeling Study. American Journal of Psychiatry. Advance online publication. https://doi.org/10.1176/appi.ajp.20250944.

Mathes, B. M., Morabito, D. M., & Schmidt, N. B. (2019). Epidemiological and Clinical Gender Differences in OCD. Current Psychiatry Reports, 21(5), 36. https://doi.org/10.1007/s11920-019-1015-2.

National Institute of Mental Health. (n.d.). Obsessive-Compulsive Disorder (OCD): Statistics. U.S. Department of Health and Human Services. https://www.nimh.nih.gov/health/statistics/obsessive-compulsive-disorder-ocd.

Ruscio, A. M., Stein, D. J., Chiu, W. T., & Kessler, R. C. (2010). The epidemiology of obsessive-compulsive disorder in the National Comorbidity Survey Replication. Molecular Psychiatry, 15(1), 53–63. https://doi.org/10.1038/mp.2008.94.

Stein, D. J., Ruscio, A. M., Altwaijri, Y., Chiu, W. T., Sampson, N. A., Aguilar-Gaxiola, S., Al-Hamzawi, A., Alonso, J., et al. (2025). Obsessive-compulsive disorder in the World Mental Health surveys. BMC Medicine, 23, 416. https://doi.org/10.1186/s12916-025-04209-5.

Stiede, J. T., Spencer, S. D., Onyeka, O., Mangen, K. H., Church, M. J., Goodman, W. K., & Storch, E. A. (2024). Obsessive–Compulsive Disorder in Children and Adolescents. Annual Review of Clinical Psychology, 20, 355–380. https://doi.org/10.1146/annurev-clinpsy-080822-043910.