UN for Men Too
Read
The evidence
The ask
Topics · 02 · The quietest number
Male suicide, laid out the way this campaign lays everything out: the scale, the highest-risk window, the signals services miss, and what actually works, each claim with its number and its primary source one tap away. Associations are never dressed up as causes.
The claims we refuse to make are at the bottom.
findahelpline.com finds your country's line. Norway: Mental Helse 116 123. Denmark: Livslinien 70 201 201. The rest of this page can wait.
While you have been on this page
0:00 elapsed
0men, at the rate the world already records
One mark every 63.9 seconds. WHO counted 493,456 male suicides in 2021, and a 365 day year holds 31,536,000 seconds, so the interval is 63.9084. The page rounds that to 64 in prose and the clock does not round at all. Nothing here is estimated live: the rate is fixed, the arithmetic is above, and the source is WHO Global Health Estimates. Every one of them was somebody's son.
Before the scale of suicide, the plainest number in this dossier. It is uncontested, it is published by every statistical agency in Europe, and it is not in the framework that decides what the world reports.
The gap is 5.2 years and has been the same order for decades. It is the largest single sex difference in any headline health statistic on this site, and unlike most figures here it is not disputed by anyone. The causes are mixed: biology contributes, and so do occupation, risk behaviour, help-seeking and smoking and drinking histories. Nobody serious attributes it to one of those alone, and this page does not either.
“Life expectancy by age and sex”
Denmark's is 3.7 years, at 80.2 against 83.9. Both campaign countries sit well below the EU average gap, which is worth stating plainly: the Nordic version of this argument is smaller than the European one, and a figure quoted from the EU average should not be presented as a Norwegian or Danish number.
“geo: NO and DK, 2024, life expectancy at birth”
Searched in the official framework after the 2026 refinement: the phrase occurs zero times. The most used summary measure of population health, and the largest sex difference in health, is not among the things the framework asks countries to report. This is a statement about one document, not about whether anyone anywhere measures it: WHO and Eurostat both publish it, which is where the figures above come from.
“life expectancy: 0 occurrences”
UN Statistics Division, Global indicator framework for the SDGs, after the 2026 refinementFACT 03
The numbers the world already collects and rarely says out loud. Every figure here is from WHO's own estimates or UNODC's, with the source one tap away.
WHO's Global Health Estimates for 2021 count 493,456 male suicides in the year. A 365 day year holds 31,536,000 seconds, so the interval is 63.9084 seconds, which the page rounds to 64 in prose. The arithmetic is here for anyone to redo, and the live tally on this page counts on the exact figure rather than the rounded one.
WHO counted 727,000 suicides worldwide in 2021; 493,456 were men, which is 68 percent. The share is stable across recent years and holds in nearly every country.
“An estimated 727 000 persons died by suicide in 2021.”
UNODC called 2021 an exceptionally lethal year for homicide, at about 458,000 victims of all sexes, 52 an hour. Male suicide alone ran at 56 an hour the same year. The quieter column is the larger one.
“2021 was an exceptionally lethal year with 458,000 deaths, a spike linked, in part, to the economic consequences of COVID-19 and a rise in organized crime and gang-related and sociopolitical violence.”
WHO GHE 2021 · UNODC, Global Study on Homicide 2023, press release 8 December 2023FACT 06
WHO GHO crude rates for 2021: 12.4 male vs 5.9 female per 100,000 globally, a 2.1 to one ratio. In high-income countries the crude ratio is about 3.1 to one, and in the US the CDC reports 3.8 to one for 2023.
WHO GHO · CDC NCHS brief 541FACT 07
Suicide is usually pictured as something that happens to the young. In the European data the male rate rises in every five-year band from adolescence to the sixties, and the widest gap between the sexes falls in the late thirties. This is the part of a life the rest of this site has least to say about.
Every five-year band is higher than the one before it. The female rate over the same span goes from 3.0 to 7.1 and stays comparatively flat. The years usually described as a man's most established are the years in which his rate is highest, which is the opposite of how the risk is usually pictured.
“Intentional self-harm, X60-X84 and Y87.0, crude death rate per 100 000 inhabitants”
16.5 per 100,000 against 3.9. The ratio is between three and four across most of adult life and is lowest in adolescence, at 1.9 for 15 to 19. It widens as people get older, then narrows slightly after 50 as the female rate rises too.
“men 16.46, women 3.90, ages 35-39, EU27”
The Norwegian male rate moves from 12.8 at 20 to 24 to 25.5 at 25 to 29 and back to 19.1 at 30 to 34, which is population size rather than epidemiology: single five-year bands in a country of five million turn on a handful of deaths. The shape across the whole adult range matches the European one, and that is as fine a grain as these numbers support.
“geo: NO, five-year bands, 2022”
Suicide risk is not evenly spread across a man's life. The strongest evidence points at one window: the collapse of family. This is where the campaign's family-law core and its quietest number meet.
Wilson and colleagues' 2025 review in Psychological Bulletin covers 75 studies across 30 countries, 29 of them meta-analyzed. Divorced men show pooled odds of 2.82 against married men; separated men 4.82. These are associations with real confounders, and the authors say so.
“Divorced men were 2.82 times more likely to die by suicide than married men.”
The same review finds separated men under 35 at pooled odds of 8.63 against married peers, and describes the risk as greatest in the immediate aftermath rather than naming a first-year peak, which is what the paper actually supports. The sharpest timing figure it carries is from Naess and colleagues 2021: a 13.20 times likelihood in the 30 days after a separation is lodged. Four of the studies reviewed found the elevated risk dissipating over time. That immediate window is where support should concentrate, and mostly does not.
“Separated men under 35 years of age had over eight times the odds of suicide than married men of the same age.”
The site's own derivation from UN sources: roughly 85,800 partner-and-family homicide deaths in 2021 against 493,456 male suicides gives a ratio of about 5.7 to one. One column has treaties, monitors and budgets. The other has this page.
WHO GHE 2021 · UNODC 2023, derivation in the registerFACT 13
Two separate sources, one calendar year. FHI's Dødsårsaksregisteret registered 658 suicides in Norway in 2021, of which 482 were men; the split by sex is published in FHI's statistics bank at https://statistikk.fhi.no/daar. Kripos registered 23 homicide cases with 29 victims in 2021, of whom 12 were killed by a partner, an ex-partner, a current or former girlfriend or boyfriend, a child, a parent or another family member, in Nasjonal drapsoversikt 2021 at https://www.politiet.no/globalassets/tall-og-fakta/drap/nasjonal-drapsoversikt-2021.pdf. 482 divided by 12 is 40.2, stated here as 40 to one, both figures for the calendar year 2021. Nothing here diminishes those twelve; it counts the column nobody is assigned to watch.
“Det ble registrert 658 selvmord i 2021 mot 639 i 2020 og 652 i 2019.”
FHI, Dødsårsaksregisteret, tall for 2021 · Kripos, Nasjonal drapsoversikt 2021FACT 14
Before the number comes the miss: depression that measurement misses, symptoms services do not read, and help that arrives shaped for someone else.
Martin, Neighbors and Griffith's 2013 JAMA Psychiatry analysis of a US national sample found that when externalizing symptoms such as anger, risk-taking and substance use are included alongside conventional criteria, men and women meet depression thresholds at equivalent rates (30.6% vs 33.3%, not significantly different).
“Analyses using the scale that included alternative and traditional depression symptoms found that men and women met criteria for depression in equal proportions: 30.6% of men and 33.3% of women (P = .57).”
Borowsky and colleagues analysed over 19,000 primary care visits and found physicians significantly less likely to recognise mental health problems in men, independent of case severity. The precise adjusted estimates are in the paper, one tap away.
“Strategies to improve detection of mental health problems among African Americans, Hispanics, and men should be explored and evaluated.”
Borowsky et al. 2000, Journal of General Internal MedicineFACT 16
The classic systematic review of 40 studies found that about three in four people who died by suicide had seen a primary care provider within the year of death, and about 45 percent within the final month, while only about one in three had any contact with mental health services in the year before death. These figures cover all suicide deaths, not men alone. The authors conclude that alternate approaches to suicide prevention may be needed for those less likely to be seen in primary care or mental health specialty care, and they name young men specifically.
“Given that this pattern is consistent with overall health-service-seeking, alternate approaches to suicide-prevention efforts may be needed for those less likely to be seen in primary care or mental health specialty care, specifically young men.”
A systematic review of 44 studies from 2000 to 2017 found an average 80 percent contact rate with primary health care in the year before suicide and 44 percent in the final month, against 31 percent contact with mental health care in the final 12 months. Sex differences run through the data: women had higher contact rates than men, which is the mirror image of who dies.
“In general, women and those over 50 years of age had the highest rates of contact with health care prior to suicide.”
The first meta-analysis of this literature (35 studies reviewed, 20 pooled) estimated that 25.7 percent of suicide decedents had inpatient or outpatient mental health service contact in the year before death, and 3.7 percent were inpatients at the time of death. Meta-regression across the pooled studies found that women had significantly higher levels of contact than men.
“Women had significantly higher levels of contact compared with men”
Meta-analyses across 78 samples and 19,453 participants found conformity to masculine norms modestly and unfavorably associated with mental health and moderately and unfavorably related to psychological help seeking. The norms of self-reliance, power over women and playboy were the most robustly related; the norm of primacy of work was not significantly related to any outcome. This is an association, not a demonstration of cause.
“moderately and unfavorably related to psychological help seeking”
A systematic review of 37 studies found conformity to traditional masculine norms affects the experience and expression of symptoms, attitudes and intention toward help seeking and actual help-seeking behaviour, and symptom management. The authors conclude that tailoring clinical interventions may increase men's service uptake and treatment efficacy. This is the peer-reviewed basis for saying services built mainly around verbal emotional disclosure may under-reach some men.
“conformity to traditional masculine norms has a threefold effect on men experiencing depression, impacting: i) their symptoms and expression of symptoms; ii) their attitudes to, intention, and, actual help-seeking behaviour; and, iii) their symptom management.”
A systematic review identified 6,598 records and found only nine eligible studies of interventions targeting men's psychological help seeking. Techniques used included role models to convey information, psychoeducation, help recognizing and managing symptoms, active problem-solving tasks, motivating behavior change, signposting services, and content built on positive male traits such as responsibility and strength. The authors are explicit that evidence-based solutions remain scarce.
“Despite considerable interest, a paucity in evidence-based solutions remains to solve this problem.”
In the Australian Man Up trial, 354 men were randomised to watch a documentary about masculinity and suicide or a control documentary; 337 completed the 4-week follow-up. The intervention group showed a significant increase in help-seeking intentions on the General Help Seeking Questionnaire. The outcome is intentions, not suicidal behaviour or deaths, and should be presented as such.
“Linear regression analysis showed a significant increase in intentions to seek help in the intervention group, but not in the control group (coef.=2.06, 95% CI 0.48 to 3.63, P=0.01).”
A mixed-methods systematic review of Men's Sheds in health promotion for older men (studies qualified if at least half the participants were aged 50 or over) included 52 studies: 35 qualitative, 9 quantitative and 8 mixed methods. It reports benefits of shed participation on self-rated health, social isolation and wellbeing, and identifies success factors (appropriate facilities, sufficient funding, participant-driven organisation). No randomised trials were included and the quantitative studies were mostly cross-sectional; the authors call for longitudinal studies to investigate causal relationships. No included study measured suicide outcomes, so the page should frame Sheds as a social connection approach for older men with promising but limited evidence, not as a proven suicide prevention measure.
“found evidence regarding benefits of shed participation on self-rated health, social isolation, and well-being”
A whole-population case-control study identified everyone who died by suicide in Wales from 2001 to 2017, 5,130 deaths in all, and fully linked GP, emergency department and hospital records for a representative cohort of 1,721 of them. Contact with health services rose toward death: in the final week, 31.4 percent of those who died had contacted services, double the rate of matched controls at 15.6 percent. The last contact was most often in general practice, 71.3 percent of the linked cohort. The authors conclude that help-seeking occurs in those at risk of suicide and escalates in the weeks before death. This counters the idea that people who die by suicide give no sign and cannot be reached.
“31.4% of cases and 15.6% of controls contacted health services”
Prevention is not a mystery. The strongest evidence is specific, cheap and mostly about narrowing the moment of crisis. This section states each effect with its number.
Systematic review and meta-analysis of 23 articles covering 18 studies of bridges, cliffs and similar sites. Interventions that restricted access to means were associated with an incidence rate ratio of 0.09 (95% CI 0.03 to 0.27, p<0.0001), about 91 percent fewer suicides per year; in studies where means restriction was assessed on its own the IRR was 0.07 (95% CI 0.02 to 0.19). Signs and crisis telephones encouraging help-seeking: IRR 0.49 (95% CI 0.29 to 0.83). Increasing likelihood of third-party intervention (surveillance, patrols): IRR 0.53 (95% CI 0.31 to 0.89). These estimates come from before and after comparisons at the sites, not randomized trials, so frame as association. The interpretation states the key approaches used at hotspots seem to be effective.
“interventions that restricted access to means were associated with a reduction in the number of suicides per year (incidence rate ratio 0.09, 95% CI 0.03-0.27; p<0.0001)”
Interrupted time series analysis (BMJ 2013, Hawton and colleagues). Average estimated reduction of 17 deaths per quarter (95% CI -25 to -9). A conservative analysis method still found a 35 percent decrease. Registrations for liver transplantation for paracetamol-induced liver damage fell 61 percent (482 fewer registrations), though actual transplantations showed no significant reduction. The 1998 legislation limited packs to 32 tablets in pharmacies and 16 elsewhere. Ecological design, so frame as association.
“This decrease represented a 43% reduction or an estimated 765 fewer deaths over the 11¼ years after the legislation.”
Study of the 2008-2011 phased bans of dimethoate, fenthion and paraquat. Pesticide suicides fell from 8.5 to 4.2 per 100,000 between 2011 and 2015, while non-pesticide suicides rose only 2 percent (9.9 to 10.1 per 100,000). In 2015 the ratio of observed to expected pesticide suicide rates was 0.49 (95% CI 0.40 to 0.62), corresponding to an estimated 937 (95% CI 574 to 1389) fewer pesticide suicides in that year than expected from pre-ban trends. The central point is that substitution was incomplete: the rise in other methods was much smaller than the fall in pesticide deaths. Ecological time series design, frame as association; the paper itself says the bans "were associated with" the reduction. Avoid stating a year count for the decline, since the paper describes 2011-2015 as five years; use the explicit span. A widely cited companion estimate of about 93,000 lives saved 1995-2015 could not be verified by fetching (see rejected).
“Overall suicide mortality dropped by 21% between 2011 and 2015, from 18.3 to 14.3 per 100,000.”
Cohort comparison study (not an RCT) of 1,640 suicidal patients in US Veterans Affairs emergency departments: 1,186 received SPI plus structured follow-up calls, 454 received usual care. Odds of suicidal behavior over 6 months: OR 0.56 (95% CI 0.33 to 0.95, p=0.03). Odds of attending at least one outpatient mental health visit: OR 2.06 (95% CI 1.57 to 2.71, p<0.001). Meta-analytic support: Nuij et al. 2021 (British Journal of Psychiatry) pooled safety planning-type interventions and found relative risk of suicidal behaviour 0.570 (95% CI 0.408 to 0.795, p=0.001; number needed to treat 16), with no significant effect on suicidal ideation (https://pubmed.ncbi.nlm.nih.gov/35048835/).
“The SPI+ was associated with 45% fewer suicidal behaviors, approximately halving the odds of suicidal behavior over 6 months (odds ratio, 0.56; 95% CI, 0.33-0.95, P = .03).”
WHO SUPRE-MISS randomized controlled trial (Fleischmann et al. 2008, Bulletin of the World Health Organization) of 1,867 suicide attempters at eight hospitals in Brazil, India, Sri Lanka, Iran and China; 91 percent completed follow-up. The intervention, called brief intervention and contact (BIC), was a one-hour information session near discharge plus nine follow-up contacts (phone calls or visits) over 18 months. One of very few interventions with RCT evidence of an effect on suicide deaths themselves, not only attempts. The paper concludes the low-cost brief intervention may be an important part of suicide prevention programmes for under-resourced low- and middle-income countries.
“Significantly fewer deaths from suicide occurred in the BIC than in the treatment-as-usual group (0.2% versus 2.2%, respectively; chi2 = 13.83, P < 0.001).”
LIVE LIFE is WHO's implementation guide for national suicide prevention (LIVE LIFE: An implementation guide for suicide prevention in countries, 2021). WHO describes the four as "key effective and evidence-based multisectoral interventions", supported by six cross-cutting pillars: situation analysis, multisectoral collaboration, awareness raising and advocacy, capacity building, financing, and surveillance, monitoring and evaluation. Useful as the framework anchor for the page: what a country-level response looks like.
“Limiting access to the means of suicide / Interacting with media for responsible reporting of suicide / Fostering socio-emotional life-skills in young people / Early identification and support to everyone affected by suicide and self-harm”
Zalsman et al. 2016 (Lancet Psychiatry), systematic review of suicide prevention strategies 2005-2014. The same review found means restriction evidence had strengthened (analgesics: overall decrease of 43 percent; jumping hotspots: 86 percent reduction) and that school-based awareness programmes reduced suicide attempts (OR 0.45, 95% CI 0.24 to 0.85). A 2025 BMC Public Health meta-analysis of gatekeeper RCTs (doi 10.1186/s12889-025-21736-1) similarly found moderate gains in knowledge and self-efficacy with effects that decay over time, and no definitive results on behavior or suicide outcomes. Honest framing for the page: worth doing as one layer, unproven as a standalone fix.
“Other approaches that need further investigation include gatekeeper training, education of physicians, and internet and helpline support.”
Department of Health and Social Care, announced 18 November 2025. The projects target areas of England where men are at most risk, including some of the most deprived areas, and are co-designed with experts and men with lived experience to break down stigma and lack of awareness of available help. The strategy also includes a partnership on the Premier League's Together Against Suicide initiative with the Samaritans to embed health messaging in the matchday experience. The press release states suicide is one of the biggest killers of men under 50 and three-quarters of all suicides are men.
“That's why the government is investing £3.6 million over the next 3 years in suicide prevention projects for middle-aged men in local communities across areas of England where men are at most risk of taking their own lives, including some of the most deprived areas in the country.”
Launched 27 May 2026 by the Irish government. The press release states the suicide rate has fallen by a third (32 percent) between 2000 and 2023, describing this as progress over the past two decades; it does not attribute the fall to any single strategy. What it funds: continued investment in Crisis Resolution Teams, Crisis Solace Cafes and Suicide Crisis Assessment Nurses, expanded access to alternatives to presenting to hospitals, enhanced suicide bereavement services, a new national postvention framework and improved inquest supports. Informed by nearly 1,900 public submissions. Separately, the HSE National Office for Suicide Prevention budget grew from about 5 million euro in 2012 to almost 15 million euro in 2025 (reported by healthmanager.ie; verify against the NOSP annual report before using that figure on the page).
“It includes an ambitious target to reduce the rate of suicide to 7 per 100,000 or below by 2035. The strategy builds on meaningful progress made over the past two decades, during which the suicide rate has fallen by a third (32%) between 2000 and 2023.”
What this page does not claim: suicide is multi-causal, the family-law window is an association with confounders, and no single policy fixes it. The refusal list below is the campaign's oldest discipline.
The global estimate above is one line. Underneath it sit national registers, each published by the country itself and each running its own length. Everything in this section is a count of deaths, not a rate, so nothing here may be compared with the WHO age-standardised figures above without population denominators. Where a register keeps two categories apart, this page keeps them apart too.
CBS StatLine table 7052_95, topic 17.2 Zelfdoding, read through the OData API on 28 August 2026. These are counts of deaths and not rates, so they are not comparable with WHO's age-standardised estimates without population denominators. The male share of the annual total was 66.7 per cent in 1950 and 67.6 per cent in 2024. Population change across seventy-five years means the counts alone do not describe risk. They describe what the register recorded.
CBS StatLine, table 7052_95, column k_172Zelfdoding_89, OData API, queried 28 August 2026FACT 35
FHI Dodsarsaksregisteret, table D10b_selvmord, deaths with suicide as the underlying cause, all methods, read through the open API on 27 August 2026. Counts, not rates. Men were 75.9 per cent of the 2025 total and 75.2 per cent of the 1996 total. The national register runs longer at both ends than the WHO series the machinery drew before it.
FHI Dodsarsaksregisteret, table D10b_selvmord, open API, queried 27 August 2026FACT 36
Danmarks Statistik StatBank table DODA1, cause group A-20, which Danmarks Statistik itself labels Suicide and attempted suicide. Read through the open API on 27 August 2026. Counts, not rates. Men were 72.5 per cent of the 2024 total. That the Danish series begins in 2007 while the Dutch begins in 1950 is a fact about what each register publishes, not about either country's history.
Danmarks Statistik StatBank, table DODA1, cause A-20, open API, queried 27 August 2026FACT 37
Socialstyrelsen's cause-of-death register, region Riket, read through its open statistics API on 28 August 2026. The register keeps X60 to X84, certain suicides, apart from Y10 to Y34, injury of undetermined intent. Swedish practice sometimes quotes the two together and sometimes only the first, so the same year can be reported with materially different totals in good faith. This project holds them as two series and never sums them. The certain-suicide series runs 1997 to 2024.
ONS publishes counts and its own age-standardised rates in the same table, so this ratio is computed inside one source and no measures are mixed. Age-standardised rates per 100,000: men 19.2 in 1981 and 17.6 in 2024; women 10.5 in 1981 and 5.7 in 2024. The male to female ratio therefore moved from about 1.8 to about 3.1. The underlying counts moved the other way, 3562 male deaths in 1981 against 4599 in 2024, because the population grew and aged. The rate is the comparable measure here and the count is not. Registration year basis, England and Wales only.
ONS, Suicides in England and Wales reference tables, Table 1, queried 27 August 2026FACT 39
CSO Ireland PxStat table VSD30, by year of occurrence rather than registration. The CSO's own note states the data is final up to 2023 and provisional afterwards, and that totals are revised upward as coroner-delayed registrations arrive. The published 2024 and 2025 figures, 351 and 370, are therefore undercounts that will rise, and neither may be quoted as final or compared with a final year. 2023 is the most recent final year.
CSO Ireland PxStat, tables VSD30 and VSD34, open API, queried 27 August 2026FACT 40
Statistics Canada table 13-10-0392-01, read through the Web Data Service on 27 August 2026. Counts, not rates, and not comparable with WHO age-standardised estimates without population denominators. Men were 74.8 per cent of the 2024 total and 77.7 per cent of the 2000 total.
Statistics Canada, table 13-10-0392-01, Web Data Service, queried 27 August 2026FACT 41
ABS Causes of Death, Australia, intentional self-harm cube, queried 27 August 2026. Counts, not rates. Men were 76.5 per cent of the 2024 total. The ABS release page carries revision markers, so earlier years in this cube can move between releases and a figure should be quoted with the release it came from.
ABS, Causes of Death, Australia, 2024, intentional self-harm cube, queried 27 August 2026FACT 42
CDC WONDER, Underlying Cause of Death 1999-2020 (D76), ICD-10 codes U03, X60 to X84 and Y87.0, by sex, read through the request_xml route on 27 August 2026. Counts with WONDER's own population denominators and crude rates, never age-standardised, so not comparable with the WHO figures above. Men were 79.5 per cent of the 2020 total and 80.3 per cent of the 1999 total. The 2020 total, 45979, reproduces CDC's published anchor exactly. The same file holds drug poisoning and alcoholic liver disease as separate series that must never be summed with suicide, because suicide and drug poisoning share the intentional self-poisoning codes X60 to X64. Final data for 2021 to 2023 live in CDC's single-race successor databases, which answer this register's scripts with errors; that is an open route, not a measured zero.
CDC WONDER, Underlying Cause of Death 1999-2020 (D76), request_xml API, queried 27 August 2026FACT 43
Same database and codes as the previous row, restricted to the ten-year age groups 25 to 34, 35 to 44 and 45 to 54 summed. Crude rates are per 100,000 of WONDER's own population for the group, not age-standardised. Men: 13067 deaths and 21.6 per 100,000 in 1999, 18119 and 28.1 in 2020. Women: 3586 and 5.9 in 1999, 4898 and 7.6 in 2020. Men were 78.5 per cent of the group's suicide deaths in 1999 and 78.7 per cent in 2020. Every crude rate in the file was recomputed from its own deaths and population before this row was written, with no mismatch.
CDC WONDER, Underlying Cause of Death 1999-2020 (D76), request_xml API, queried 27 August 2026FACT 44
Statistics Finland, StatFin tables 11by and 11ay, read at the office's own API on 2 September 2026 with the exact queries stored for replay. Counts by sex with full age structure, all 6,552 cells present, every band sum recomputed exactly. The office's age-standardised rate uses the European standard population as defined by Eurostat in 2012, stated in the office's own documentation; it fell from 42.4 per 100,000 for men in 1971 to 20.6 in 2024, with women from 10.1 to 6.5. Counts and rates are different quantities and are never drawn on one axis here. Figures for 2025 publish on 29 October 2026 by the office's written answer.
Statistics Finland, StatFin 11by and 11ay, API read 2 September 2026FACT 45
CBS StatLine table 7022eng (1950 to 2024, final) and the office's customised table of 19 June 2026 (1970 to 2025, with 2025 provisional and marked so by CBS), both read at CBS's own routes on 2 September 2026. The two agree on every one of their 55 shared years. Against the older StatLine table 7052 this register also holds, two cells differ by exactly one death, 1965 women and 1966 men; both tables were re-fetched live the same day and both still return their own values, so the disagreement is the office's, recorded here with both values, a question queued for the office. 2025 lands in StatLine in early October 2026 per the office's written answer.
CBS StatLine 7022eng and the CBS customised table of 19 June 2026, read 2 September 2026FACT 46
The national cause-of-death register at NIJZ, disseminated through SiStat, read at the API on 2 September 2026. Counts by sex and age from 1994, all internal sums exact across 1,488 cells; a both-sexes series from 1954 that the office's written reply called rates-only but the API serves as counts, stored as served. The long series' companion measure is suicides per 1,000 deaths, a share of deaths, never a population rate, and is labelled so. No sex split exists anywhere on the route before 1994; that absence is recorded as absence. 2025 publishes 30 September 2026 per the office's written answer.
The Office of Statistics' eTab table 471.001, read at the office's own API on 2 September 2026. Annual totals run 0 to 11; 2009 was a zero-death year and the office's published zeros are kept as its figures, while its suppression marks are never stored as numbers. The absence of any age-standardised rate is the office's own written answer, recorded as a documented absence. One overlap year disagrees with the held Eurostat rate series, 2011 women, office count 0 against a Eurostat rate of 4.45 per 100,000, about one death on this population; both values are held and neither is changed. 2025 publishes 6 November 2026 per the office.
Liechtenstein Office of Statistics, eTab 471.001, API read 2 September 2026FACT 48
Received from Statistics Austria by mail on 2 September 2026 in reply to a written inquiry, the national workbook parsed and validated: the cause row is the office's own, ICD-10 X60 to X84 plus Y87.0, men plus women equals the office's both-sexes sheet in all 56 years. The office's own age-standardised sheets, on the 2013 European standard population from 2002, run 0.10 to 0.29 per 100,000 below the held Eurostat series in every overlapping male year; both values are stored for every overlap year and the cause of the gap is not asserted. The nine Bundesland workbooks are held unparsed. The raw workbook and every validation are in the project archive.
Received from the Federal Statistical Office by mail on 2 September 2026: three workbooks spanning 1958 to 2024, male plus female matching the published total in all 67 years. Only 1998 onward, ICD-10, appears on this page: the sheets do not state their territorial coverage, the step from 13,167 deaths in 1979 to 18,451 in 1980 is far too large to be a change in suicide, and until the office answers which years cover the former federal territory only, no earlier figure is published and no trend is drawn across that boundary. The question is written and staged for the office contact form. Public home of the 1980-onward data is the Federal Health Reporting database; the pre-1980 sheets have no public URL we know of.
Statistics Denmark confirmed in writing on 2 September 2026 that its cause-of-death data comes from the Danish Health Data Authority. The authority's own route, read the same day, publishes cause group A-20, ICD-10 X60 to 84 plus Y87.0, from 2002, five years deeper than the office's table, but disclosure-controlled throughout: counts rounded to fives, small cells suppressed, no exact count of suicide for any year, none by sex, none by age. Its sex splits are a crude rate for 2015 to 2024 and, for 2024 only, sex-specific age-standardised rates by method; no sex split of counts exists for any year. Seventeen of eighteen overlap years match the office's exact counts rounded to five; 2008 cannot be reconciled even by rounding, office 607 against authority 610, register versions half a year apart, and both values are held. The authority also differs from itself by 0.1 on the national crude rate in five years between its own two workbooks; both are held.
Danish Health Data Authority, sundhedsdatabank.dk workbooks, read 2 September 2026FACT 51
Statistics Sweden answered a written inquiry on 2 September 2026 in three lines: suicide by sex is entirely Socialstyrelsen's, there is no cause-of-death series in Statistikdatabasen, and nothing exists beyond Socialstyrelsen's tables. This register already holds Socialstyrelsen's series, including age splits from 1997, so the answer changes no figure; the age-standardised rates this register draws for Sweden are its own computation on the 2013 European standard population and are labelled as such, since no office publishes one; it documents the routing in the system's own voice, dated, so that no reader wastes a search where the office itself says there is nothing.
Twenty claims below are ones this campaign found in circulation, checked, and will not make. Several would have made this page louder, and one of them is a figure this campaign itself published before checking it properly. Suicide is the subject where a wrong number does real damage beyond the argument, because people read these pages while unwell.
This page says where its evidence is weak, and it will keep doing that. But thin evidence and a settled question are not the same thing. The UN's own indicator on sexual violence in childhood is written to cover young women and men alike, and the comparable data actually collected covers about 68 low- and middle-income countries for women and about 12 for men, with high-income countries outside the count for both. The indicator excluded nobody. The measuring did.
So where you read here that the research is thin, read it as a description of what has been funded and counted, not as a verdict on what is happening to men and boys. Then ask the question this campaign exists to ask: who was assigned to collect it, and what would we know by now if somebody had been?
52 claims on this subject survived the check and are published above, each with its exact figure, its scope and its primary source.
20 more were checked and thrown out.
The discarded ones are not printed here, because repeating a false claim in order to reject it mostly just repeats it. They are not hidden either: every one travels with the evidence base in facts.json, with the reason it failed, so what we threw away can be audited as easily as what we kept.
If you have seen this campaign quoted with a number that is not on this page, that is the first place to look.
Worldwide, 493,456 men in 2021, out of 727,043 suicides in total. That is 67.9 per cent of them, or roughly one man every 64 seconds. The figures are WHO's own Global Health Estimates.
The honest answer is that the reasons are contested and this site will not pretend otherwise. What is not contested is the size of the gap or that it holds across almost every country measured. WHO Europe's men's health strategy states suicide is more than three times higher among males than females in all age groups over 15.
It is counted, which is not the same thing. WHO publishes the number and SDG indicator 3.4.2 tracks suicide mortality disaggregated by sex. What does not exist is any entity, target or funding marker whose object is male outcomes, so the number is recorded rather than owned.
Research on separated men finds suicide odds around 4.82 times those of married men, and over eight times for separated men under 35, with risk concentrated in the weeks immediately after separation. That is a window a service could be designed around, and this campaign has not found one designed around it.
One of WHO's six regions does. Europe adopted a Strategy on the health and well-being of men by resolution EUR/RC68/R4 in 2018, with progress reported jointly with the women's strategy in 2020 and 2023. The Americas published a report on masculinities in 2019, which is a document rather than a mandate.
Crisis lines for several countries are listed at the top and bottom of every page on this site. If you are in immediate danger, contact your local emergency number. Nothing on this page is a substitute for that.
The world counts these deaths beautifully and assigns nobody to them. Counting is where every fix in history started.
— voices so far · one tap, anonymous, no signup
Independent research by Mikal Steen Monslaup, supported by Manderådet and MannsForum · unformentoo.org