Transparent by design
Show the goal.
Show the evidence.
Our ambition is shared. Our assumptions should be inspectable. This page explains what the dashboard measures, what we choose, and what the numbers cannot establish.
Every country, honest coverage
The atlas contains all 193 United Nations member states plus its two permanent observer states, the Holy See and the State of Palestine. Inclusion follows this explicit country list rather than assuming that a provider’s countries and territories are identical to sovereign states. World Bank West Bank and Gaza data are mapped to the Palestine profile and identified in source notes. Small states with missing observations remain in the atlas.
Snapshot generated 2026-10-06T20:04:04.179127+00:00. Indicators use their latest published years, often earlier than the retrieval date. WHO estimates are modeled and can revise historical series. Country means can conceal inequality and underreporting. Rankings do not control for age structures, conflict, wealth, or reporting quality.
A demanding north-star score · version 2
Version 2 is a deliberately demanding editorial measure of distance to the north stars, not an official WHO/UN index or the percentage of people who are healthy. Strong services cannot cancel out ongoing suicide, health loss, or extreme poverty. At least 9 of 12 core measures and all three recent zero-harm inputs are required. Missing inputs suppress the final score, rather than implying success. Health loss is a proxy broader than avoidable suffering; the $3.00/day measure covers extreme poverty, not every form of deprivation. Lower scores under this new method are a recalibration, not evidence of deterioration.
The former version-1 arithmetic mean rewarded strong service and infrastructure measures even when serious harm continued. Version 2 retains that mean as a foundation, then multiplies it by three separate factors. It measures distance to an ambition, not the percentage of healthy people, a probability, or the estimated effect of a policy.
Core mean × 1/(1 + suicide deaths per 100,000 / 5) × 1/(1 + DALYs per person / 0.1) × 1/(1 + extreme poverty percent / 5)
Each zero-harm factor is 1 / (1 + observed burden / editorial anchor). A zero observation gives a factor of 1; the anchor halves that factor. These deliberately demanding half-penalty anchors are 5 suicide deaths per 100,000 people annually, 0.1 healthy life-years lost per person annually, and 5% of the population in extreme poverty. These are disclosed value choices, not WHO standards or medically safe thresholds. Every amount of measured harm continues to reduce the score; there is no artificial ceiling imposed on countries.
The suicide rate is derived from WHO GHE self-harm deaths and the same-release, same-year population. Health loss is all-cause DALYs divided by that population. Both use 2023. Extreme poverty uses the latest eligible World Bank $3.00/day (2021 PPP) observation. A country needs at least 9 of 12 foundation observations, plus all three zero-harm inputs, within 8 years. Missing inputs suppress the final score. 100 requires all foundation horizons and zero measured values for the three factors; no country meets these conditions.
The proxy limits matter. DALYs include health loss that cannot currently be prevented; they do not isolate needless suffering or measure all distress, pain, exclusion, or deprivation. The extreme-poverty threshold does not cover every form of poverty. The aspirations are broader than these available comparable measures. Repeated emphasis on mortality and suicide in the foundation and factors is an intentional editorial weighting, not an addition of death counts. Multiplication makes the result sensitive to each factor and to the disclosed anchors; it is not a causal or probabilistic model. Lower version-2 scores are a recalibration, not evidence that a country suddenly deteriorated.
Foundation component anchors
Foundation components still use 100 × clamp((value − bad) / (good − bad), 0, 1), averaged with equal weights. Partial foundation coverage, age structure, source years, underreporting, and uncertainty limit cross-country comparison.
| Core measure | 0-point anchor | 100-point horizon | Unit |
|---|---|---|---|
| Life expectancy | 50 | 90 | years |
| Under-five mortality | 100 | 0 | per 1,000 live births |
| Maternal mortality | 500 | 0 | per 100,000 live births |
| Suicide mortality | 25 | 0 | per 100,000 people |
| Premature chronic-disease mortality | 35 | 0 | % probability ages 30–70 |
| Tuberculosis incidence | 500 | 0 | per 100,000 people |
| Child stunting | 50 | 0 | % children under 5 |
| Essential health-service coverage | 0 | 100 | index 0–100 |
| DTP immunization | 0 | 100 | % children ages 12–23 months |
| Safely managed drinking water | 0 | 100 | % of population |
| Safely managed sanitation | 0 | 100 | % of population |
| Fine-particle air pollution | 60 | 5 | µg/m³ |
A north star on every chart
A north star describes the desired direction of progress. Zero suicide deaths is the explicit ambition. Zero preventable child and maternal deaths are aspirations; plotted zero-rate horizons are not forecasts that all biological mortality can be eliminated. Life expectancy of 90 and healthy life expectancy of 80 are chosen horizons, not official international mandates.
Where an authoritative numerical standard exists, it is linked: the WHO annual PM2.5 guideline is 5 µg/m³. UN Sustainable Development Goals provide interim child and maternal mortality thresholds, distinct from the aspirational zero horizon. Spending, workforce density, employment, income, and other contextual measures have qualitative goals because there is no universally optimal number.
Average-rate death and suicide clocks
These repeating countdowns illustrate published annual totals. They do not observe an event, predict who will die, or claim a death occurred when a cycle restarts. The interval is seconds in the reference calendar year / annual deaths; the 2023 source year has 31,536,000 seconds. A country clock uses its own WHO estimates. The world dashboard sums the distinct country rows covered by WHO, 184 of 195 countries, and explicitly excludes missing countries.
Both clocks use WHO Global Health Estimates 2023, both sexes and all ages. The death clock uses the all-cause row, and the suicide clock uses the intentional self-harm row (GHE cause 1610). Suicide is included in all-cause deaths; the clocks must not be added. Sources estimate 60,803,808 all-cause deaths and 740,938 suicide deaths in the covered population. These are modeled past-year counts, not 2026 live totals. WHO mortality release and downloadable workbooks ↗
The page starts a visual cycle when opened. At each completed cycle, the countdown briefly shows zero and a human silhouette fades away, with a caption identifying an average-rate illustration. This is a symbolic animation, not evidence that a person just died. Pause stops both clocks and illustrations; reduced-motion preference pauses them by default and keeps silhouettes still even if the clocks are resumed. Returning to a hidden tab starts a fresh cycle without replaying missed animations. Missing estimates show no countdown or loss animation. Every all-cause death is a life lost, but not every death can be prevented. The clocks provide urgency without treating every death as avoidable.
Lives, healthy years, and suffering
The annual burden baseline comes from WHO Global Health Estimates 2023, released in October 2026. It includes all-cause deaths, years of life lost to premature mortality (YLL), years of nonfatal health loss weighted by severity (YLD), and disability-adjusted life years (DALYs).
- Deaths: a count of people who died in the covered year; not every death is preventable.
- YLL: remaining life-years lost relative to the WHO reference life table.
- YLD: severity-weighted years of health loss. This is not a judgment about the worth of a person with a disability.
- DALYs: YLL + YLD, already a combined total. Do not add DALYs to YLL, YLD, or deaths.
For a selected percentage r, every scenario output is baseline × r / 100. Equivalent 80-year healthy lifetimes are DALYs reduced / 80, a unit conversion. They are not a count of persons saved. A 100% reduction is an arithmetic ceiling and is not presented as achievable.
The global view sums same-year WHO all-cause country estimates over the covered countries and reports the coverage count. It is a covered-country sum, not a complete official world estimate. People and burdens in missing states are not zero. The source already adjusts YLD for comorbidity; we do not sum separate risk factors or overlapping cause categories.
The checklist does not establish these effects. Estimating the effect of a programme requires intervention-specific effectiveness evidence, local baseline risks, reach, uptake, displacement, time horizon, uncertainty, and adjustment for overlap. Completing action items cannot be assumed to deliver the scenario reduction. DALYs capture health loss, not every form of suffering: distress, grief, insecurity, loneliness, discrimination, and lost agency require additional evidence.
Sources and reuse
World Bank World Development Indicators — international compiled series, with indicator-specific primary producers and definitions.
WHO Global Health Estimates — deaths, YLL, YLD and DALYs; country coverage and release-specific source files are recorded in the download.
WHO Global Health Observatory — healthy life expectancy and reviewed indicator-specific gap supplements. Each supplemental chart identifies its source and method; model vintages may differ from the burden release.
UN Sustainable Development Goals and WHO air quality guidelines — international milestones and standards.
The data-gap audit
Reviewed 2026-10-06. 9,368 of 10,140 country–indicator series contain numerical observations. 772 still lack a matching numerical series in the sources successfully retrieved. Availability does not mean a recent year or complete national measurement.
This audit recovered 64 previously absent numerical series across 48 countries, adding 957 observations. Historical records retain their original dates. A recent model explicitly based on inputs older than the scoring window is excluded from the score, even when its modeled reference year is recent.
WHO register information now accompanies 85 missing malaria-rate charts. Certification and supplementary-list status are shown separately, with the register’s publication date and geographic scope. They do not become numerical zero rates. Japan’s FAO hunger series is published as a range below 2.5%; no exact number is inferred.
WHO homicide estimates use a different modeling method from UNODC administrative observations. Added charts identify that difference. The WHO GHE mortality and health-loss totals remain drawn from one consistent release, covering 184 countries.
Sources checked and access limits
- World Bank WDI: All registered WDI measures with gaps queried for absent countries over 1960–2026; pagination validated.
- WHO GHO: 21 direct series checked. Subnational water/hygiene rows, sex-specific values, all-age HIV rates, smoking-only prevalence and different malaria denominators are not substitutes.
- World Bank Poverty and Inequality Platform: 2021 PPP, $3/day, no interpolation/extrapolation. No additional missing-country coverage beyond the historical WDI observations.
- UNSD SDG database: Release 2026.Q3.G.01 queried for 13 matching measures. Non-relevant zero rows, subnational measures, wrong age/sex and censored values are excluded from numeric fills. No applicable national sanitation additions were found. HIV retrieval returned HTTP 500 and timeouts, including narrowed country requests; this feed could not be audited completely.
- UNICEF JME bulk downloads: Direct XLSX retrieval returned HTTP 403; official browser downloads succeeded. Both-sex stunting models and national wasting surveys were checked. No additional numeric series were found beyond the reviewed WHO/WDI additions. Missing-value hyphens, subgroup rows and excluded surveys are not numeric fills. Bahrain and Mauritius model estimates retain the warning that their latest input precedes 2000.
- UNAIDS 2026 bulk estimates: Direct ZIP retrieval returned HTTP 403; the official browser download succeeded. National age-15–49, both-sex Spectrum estimates were reviewed and recovered for 11 missing country series. Blank cells and wrong age/sex groups are excluded.
- WHO GHE / IHME: WHO GHE 2023 covers 184 atlas countries. Other model vintages are not inserted into the same-release burden totals, clocks or score factors.
For water, sanitation, hygiene and literacy, some sources publish only a subgroup. Urban-only, rural-only, youth-only and female-only figures are not passed off as whole-country measures. The same rule applies to different HIV age groups, diagnoses, poverty definitions and malaria denominators.
Download every remaining gap and its review notes ↗ · Download source URLs, retrieval dates and file hashes ↗
What is still missing
This is a reviewed set of 52 comparable measures, not a claim that all meaningful national health has been measured. Comparable mental-health prevalence, pain, loneliness, treatment quality, catastrophic care costs, healthy housing, conflict injury, displacement, and within-country inequality deserve their own evidence. A future addition must declare the definition, source, years, coverage, goal, and overlap before affecting the score or scenario totals.
Independence and stewardship
Unified Nations is an independent civic project and is not affiliated with the United Nations, WHO, or the World Bank. Data providers do not endorse this score or agenda. Method version 2.0. The source refresh script retains published years, source URLs, missingness, and a retrieval manifest; the release gate checks country coverage, metadata, unique share cards, and public serving.