Breaking Down Stigma Around Mental Illness: 2025 Statistics

There is no single trustworthy figure for the share of people with a mental health condition who experience stigma. What a study reports depends on the instrument it used, who it asked, in which country, and in which year. A page offering one global percentage is usually stacking studies that were never built to be compared.

So this page does something more useful than quoting a headline number. It sets out the figures that carry their own denominators and dates, names where each came from, and shows how to test any stigma statistic you meet elsewhere.

Why there is no single stigma percentage

Stigma figures sit on top of prevalence figures, so the base population matters first. Bhugra and colleagues report that the World Health Organization estimates more than 25 percent of people worldwide develop one or more mental disorders during their lifetime. Those World Mental Health Surveys covered 85,052 respondents in 17 countries, and lifetime prevalence of any mental disorder ran above 25 percent in 11 of them.¹

Reporting is part of what those surveys measure. In the same surveys, the low rates reported in China and Nigeria were judged likely to be downwardly biased by respondents' reluctance to admit mental health problems.¹ A country can appear to have less illness, and less stigma, precisely because people there are less willing to say so.

Comparison across borders is harder still. Pescosolido and colleagues also note that differences in samples and instrumentation make mental illness stigma findings difficult, if not impossible, to compare across countries. Writing in 2008, Pescosolido and colleagues put mental illness at 11% of the global burden of disease.² That figure is old, and it is a burden estimate rather than a stigma estimate, which is exactly the kind of distinction stat pages tend to lose.

Even within one research group, two scales of the same thing disagree. Stuart and colleagues built one scale for how often stigma is experienced and another for the intensity of its psychosocial impact on life domains such as quality of life. They found the frequency and the impact of stigma experiences to be only modestly correlated.³ A survey asking how often, and a survey asking how much it hurt, will not return the same percentage.

How to test a stigma statistic

Five questions separate a citable figure from a decorative one. If a page cannot answer them, the number cannot be checked, and an uncheckable number should not be cited.

The last question catches most people out. Global Burden of Disease estimates are produced by a Bayesian meta-regression tool, DisMod-MR 2.1, which also estimates prevalence for locations that are missing raw epidemiological data.⁴ A country with no survey of its own can still appear in a global table, carrying a number modelled from its neighbours.

The measures themselves are also easy to confuse. The Global Burden of Disease Study 2019 assessed 12 mental disorders across 204 countries and territories between 1990 and 2019. It reported prevalence, disability-adjusted life-years, years lived with disability, and years of life lost as separate measures for mental disorders.⁵ A percentage of disability is not a percentage of all disease burden, and swapping one for the other inflates the claim.

Sex differences have the same trap. A separate analysis of the 2019 study reported incidence, prevalence and DALYs for anxiety disorders by gender and age group in 204 countries.⁶ Read those, rather than one country's household survey quoted as a world figure.

The kinds of stigma that get measured

Stigma is not one variable. A systematic review of stigma and help-seeking separates anticipated, experienced, internalised, perceived and treatment stigma, along with people's own stigmatising attitudes towards others.⁷ Two studies can both be about stigma and still be measuring different things.

• Anticipated stigma: expecting to be treated unfairly, before anything has happened.
• Experienced stigma: having actually been treated unfairly.
• Internalised stigma: holding the stigmatising view about yourself.
• Perceived stigma: a view about what people in general think.
• Treatment stigma: stigma attached to seeking or receiving care, rather than to the condition.
• Structural stigma: the rules, laws and services that disadvantage people whatever any individual believes.

The structural kind is the one aggregation pages usually leave out, and it carries weight. Yang, Link and Phelan reviewed 123 articles on the stigma of mental illness published between 1995 and June 2003 and examined the measures used in them. They cite the 1999 United States Surgeon General's report on the central place stigma holds in unequal treatment, reducing access to resources and opportunities such as housing and jobs.⁸

Country figures, and what they actually mean

Country numbers are not a ranking of prejudice. They reflect what was asked, of whom, and what people in that setting felt able to say. They are informative about differences in measurement and in local explanations of illness. Used as a league table, they mislead.

Treatment gaps are the most concrete thing measured at country level. An analysis of a national mental health survey across twelve Indian states found the treatment gap for mental disorders ranged between 70% and 92% depending on the disorder. In that survey the treatment gap was 85.0% for common mental disorders and 73.6% for severe mental disorders. The treatment gap reached 91.8% for tobacco use and 86.3% for alcohol use disorder, against 75.5% for psychosis.⁹

Stigma is one driver among several here, and service capacity is another. A review of Pakistan's mental health system describes a large treatment gap alongside severe limits in system capacity, and points to integrating services into primary care.¹⁰

How illness is explained locally shapes what stigma looks like. A 2020 WHO country report describes stigma towards mental illness and its treatment as widespread in Bangladesh, where symptoms are often attributed to possession by evil spirits. That report notes families may hide a relative's mental health condition, because stigma attaches to the whole family and not only to the person.¹¹ Anti-stigma work is public policy in many countries: the UAE government describes programmes to reduce stigma and end discrimination against people with mental illness, including a journalism fellowship awarded to three Emirati journalists.¹²

Stigma also damages the data itself, which is the deepest reason to distrust tidy percentages. WHO reports that stigma surrounding suicide, and the illegality of suicidal behaviour in some countries, make under-reporting and misclassification more likely. WHO also reports that only some 80 Member States hold good-quality vital registration data that can be used to estimate suicide rates.¹³ The thing being counted is the thing people are under pressure to conceal.

What stigma costs

The honest summary of the help-seeking evidence is that the direction is clear and the size is not. Earlier reviews each reported some evidence that stigma impedes help-seeking, and proposed mechanisms rather than settling them.⁷ In consulting rooms this shows up as delay: people arrive years after their difficulties began, often with a first sentence about what they were afraid someone would think.

Diagnosis is not protection from it. WHO's 2025 fact sheet on bipolar disorder states that many people with the condition are misdiagnosed or untreated, and meet discrimination and stigma. WHO estimates that in 2021 about 37 million people, or 0.5% of the global population, were living with bipolar disorder.¹⁴ Alongside delayed care sit the losses Yang, Link and Phelan describe: housing, work and other everyday opportunities.

What is known to help

Public knowledge is one lever, with a realistic ceiling. A review in the British Journal of Psychiatry (2000) found that attitudes which hinder recognition and appropriate help-seeking are common. It also found much of the mental health information easily available to the public is misleading, while attitudes and mental health literacy can be improved.¹⁵ Better information helps, and it does not on its own dismantle discrimination.

Policy is the second lever, and it targets the structural kind. WHO's action plan for mental health names reducing stigma and discrimination, together with increasing access to evidence-based care, among its implementation strategies. For workplaces, WHO points to non-discrimination policies and to employment laws aligned with international human rights instruments, alongside reducing stigma and discrimination.¹⁶ Bringing care closer to people does similar work, which is part of the argument for integrating mental health into primary care.

One thing this page will not do is quote effect sizes for anti-stigma programmes that we could not verify against a primary source. If you are assembling evidence on what works, treat any intervention percentage without a named trial count, population and follow-up period as unusable.

If stigma is part of your own experience

If you have been putting off contacting someone because of what it would mean about you, that is internalised stigma. It is common enough to be unremarkable in clinical practice, and it is workable. Therapy treats the belief that you are your diagnosis as something to examine rather than accept. Your therapist will usually ask what you expect other people to think, and where that expectation came from.

If a family conversation is the obstacle, it helps to separate the two things being asked: telling someone you are struggling, and asking them to do something about it. Those can be different conversations, held on different days. If you would like help thinking through the first contact, how to find the right therapist sets out what to look for.

Written and reviewed by Team TherapyRoute.

References

1. [1] Kaplan & Sadock's Comprehensive Textbook of Psychiatry, 10th ed (2017, vols 1+2): Kaplan & Sadock's Comprehensive Textbook of Psychiatry, 10th ed (2017, vols 1+2), 62. World Aspects of Psychiatry (pt 1) (pp 11488-11495).
2. [2] Arboleda-Florez & Sartorius - Understanding the Stigma of Mental Illness (2008): Arboleda-Florez & Sartorius - Understanding the Stigma of Mental Illness (2008), 2 Cross-cultural aspects of the stigma of mental illness (pt 1) (pp 34-41).
3. [3] Arboleda-Florez & Sartorius - Understanding the Stigma of Mental Illness (2008): Arboleda-Florez & Sartorius - Understanding the Stigma of Mental Illness (2008), Appendix Inventories to measure the scope and impact of stigma experiences from the perspective of those who are stigmatized, consumer and family versions (pt 1) (pp 208-215).
4. [4] NCBI Bookshelf. ncbi.nlm.nih.gov/pmc/articles/PMC8500697.
5. [5] PubMed Central. pmc.ncbi.nlm.nih.gov/articles/PMC8776563.
6. [6] Source hosted at www.cambridge.org. cambridge.org/core/journals/epidemiology-and-psychiatric-....
7. [7] DOI record. doi.org/10.1017/S0033291714000129.
8. [8] Arboleda-Florez & Sartorius - Understanding the Stigma of Mental Illness (2008): Arboleda-Florez & Sartorius - Understanding the Stigma of Mental Illness (2008), 11 Stigma measurement approaches: conceptual origins and current applications (pt 1) (pp 190-197).
9. [9] PubMed Central. pmc.ncbi.nlm.nih.gov/articles/PMC5419008.
10. [10] PubMed Central. pmc.ncbi.nlm.nih.gov/articles/PMC11835998.
11. [11] World Health Organization. who.int/docs/default-source/mental-health/special-initiat....
12. [12] Source hosted at u.ae. u.ae/en/information-and-services/health-and-fitness/menta....
13. [13] World Health Organization. who.int/news-room/fact-sheets/detail/suicide.
14. [14] World Health Organization. who.int/news-room/fact-sheets/detail/bipolar-disorder.
15. [15] Source hosted at www.cambridge.org. cambridge.org/core/journals/the-british-journal-of-psychi....
16. [16] World Health Organization. who.int/news-room/fact-sheets/detail/mental-health-at-work.