A map for giving · judged blind
Survive, then thrive. A map of where giving can help.
Philanthropy has one job: less suffering and more flourishing, for every living being it can reach. This page lays the whole field on one map, so that nothing falls through the cracks and nothing is counted twice. Then it asks each corner the only question we ever ask: did it work?
What we believe
Giving is for the one receiving it.
Outcomes, not intentions
A gift is good to the degree that it changes something for a person, or an animal, at the other end: one less death, less pain, more of a life worth living. The gala, the plaque and the moving story are all label. Intentions are the label on the bottle; outcomes are what is inside.
One map, no overlaps, no holes
We use a map that is mutually exclusive and collectively exhaustive. Every harm sits in exactly one place, so no death or dollar is counted twice and no charity can claim the same life under two headings. Every harm sits somewhere, so the quiet corners (injuries, chronic pain, loneliness, animals) stay on the page even when almost nobody pays for them.
Two questions, kept apart
Survive asks how many people die, and of what. Thrive asks how well the living live. They are measured in different units, and turning one into the other is a moral judgment. That judgment is yours, so we never fold them into one number.
Size is not priority
We show where deaths and money don’t line up. A mismatch is a question, not a verdict: some causes are cheap to fix, some are paid for by governments, and some have no proven fix yet. Which cause matters most depends on your values; arithmetic can’t settle it, and we won’t pretend it can.
Honest gaps
Where nobody tracks the money, we write “not tracked”. Where no evaluator has a top pick, we say so. A gap you can see is a gap someone can fill.
Evidence, then joy
When something plainly worked, we say so with delight. Every figure here names its source and year, and every cost estimate is the evaluator’s own: modeled, not guaranteed.
Survive
Every death has one place on the map.
The Global Burden of Disease study, led by IHME, files every death in the world under one cause, in a tree. At the top are three branches; below them, 22 groups. Each death is counted once, in one place, so the branches add up to the total. In 2023 that total was 60.0 million deaths.1
The 22 groups, with the money and the best options
How to read it. Health aid is development assistance for health to low- and middle-income countries, as tracked by IHME for 2023, prevention and treatment together; nobody we know of tracks prevention-only giving by cause. Dollars are 2022 dollars. Aid share ÷ death share is worked out only from the figures shown: above 1 means a cause gets a bigger share of aid than of deaths, below 1 a smaller one. WHO best buys are policies WHO rates at no more than I$100 per healthy year of life gained in poorer countries6: a government’s tool, so a donor can only fund the push for them, and no evaluator we follow has rated that.
| Cause group | Deaths in 2023 | Health aid tracked by IHME | Aid share ÷ death share | Strongest evidence-backed option |
|---|---|---|---|---|
| Infections, mothers, newborns and nutrition 10.3 million deaths · 17.2% | ||||
| HIV/AIDS and sexually transmitted infectionsGBD: HIV/AIDS and sexually transmitted infections | 917,2001 1.5% | HIV/AIDS: $14.1 billion in 2023. Other sexually transmitted infections are not tracked separately.2 | 15.7×$14.1 billion ÷ $64.6 billion = 21.8% of aid) ÷ (833,400 ÷ 60,043,100 = 1.39% of deaths) = 15.7 | Gap No evaluator we follow has a top pick here yet. |
| Respiratory infections and tuberculosis (includes COVID-19)GBD: Respiratory infections and tuberculosis | 4.3 million1 7.2% | COVID-19: $12.4 billion in 2023. Tuberculosis: $2.6 billion in 2022 (IHME gives no 2023 figure in words). Pneumonia and other lower respiratory infections: not tracked separately.2 | 14.4×$12.4 billion ÷ $64.6 billion = 19.2% of aid) ÷ (797,600 ÷ 60,043,100 = 1.33% of deaths) = 14.4 | Gap No evaluator we follow has a top pick here yet. New Incentives (listed under other infections) pays for routine childhood vaccines, including ones against tuberculosis and pneumococcal disease.3 |
| Diarrhoea, typhoid and other gut infectionsGBD: Enteric infections | 1.3 million1 2.1% | Not tracked Not tracked as its own area. IHME counts this money inside child health and other infections. | Not computed | Gap No evaluator we follow has a top pick here yet. See our safe water page for what evaluators have funded. |
| Malaria and neglected tropical diseasesGBD: Neglected tropical diseases and malaria | 804,9001 1.3% | Malaria: $3.3 billion in 2023. Neglected tropical diseases: not stated separately.2 | 4.6×$3.3 billion ÷ $64.6 billion = 5.1% of aid) ÷ (670,000 ÷ 60,043,100 = 1.12% of deaths) = 4.6 | Against Malaria Foundation GiveWell Top Charity Malaria Consortium GiveWell Top Charity |
| Other infections (measles, meningitis, whooping cough and more)GBD: Other infectious diseases | 841,7001 1.4% | Not yet sourced IHME tracks “other infectious diseases”, but states no 2023 figure in words. | Not computed | New Incentives GiveWell Top Charity |
| Mothers and newbornsGBD: Maternal and neonatal disorders | 1.9 million1 3.1% | Not yet sourced IHME tracks reproductive, maternal, newborn and child health together, but states no 2023 figure in words. | Not computed | Gap No evaluator we follow has a top pick here yet. An honest gap, and an opening. |
| Nutritional deficienciesGBD: Nutritional deficiencies | 288,6001 0.5% | Not yet sourced Nutrition is a programme area inside child health. No figure stated. | Not computed | Helen Keller Intl GiveWell Top Charity Taimaka HLI Promising Charity |
| Non-communicable diseases 44.8 million deaths · 74.7%Health aid, whole group: All non-communicable diseases together: less than 4.0% of health aid each year. Not split by group.2 Aid share ÷ death share: under 0.054 (working) | ||||
| CancersGBD: Neoplasms | 10.6 million1 17.6% | Counted with the group (see above) | Group only (see above) | WHO best buy “Vaccination against human papillomavirus (1or 2 doses) of girls aged 9–14 years”; “Prevention of liver cancer through hepatitis B immunization”6. |
| Heart disease and strokeGBD: Cardiovascular diseases | 19.2 million1 31.9% | Counted with the group (see above) | Group only (see above) | WHO best buy “Secondary prevention of rheumatic fever and rheumatic heart disease by developing a register of patients who receive regular prophylactic penicillin”6. WHO puts drug treatment of high blood pressure in its next category (above I$100 per healthy year). Tobacco and salt best buys also act here. |
| Chronic lung disease (COPD, asthma)GBD: Chronic respiratory diseases | 4.2 million1 6.9% | Counted with the group (see above) | Group only (see above) | WHO best buy “Acute treatment of asthma exacerbations with inhaled bronchodilators and oral steroids”; “Long-term management of COPD with inhaled bronchodilator”6. |
| Digestive diseases (including cirrhosis)GBD: Digestive diseases | 2.4 million1 4.0% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
| Neurological disorders (including dementia)GBD: Neurological disorders | 3.0 million1 5.0% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
| Mental disordersGBD: Mental disorders | 2001 under 0.1% | Counted with the group (see above) | Group only (see above) | Almost no deaths are counted here: GBD records suicide under self-harm. The suffering is real and large; see Mind, under Thrive. |
| Alcohol and drug use disordersGBD: Substance use disorders | 344,1001 0.6% | Counted with the group (see above) | Group only (see above) | WHO best buy “Increase excise taxes on alcoholic beverages”6. |
| Diabetes and kidney diseaseGBD: Diabetes and kidney diseases | 3.5 million1 5.9% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. WHO lists no best buy for managing diabetes; its listed options cost more than I$100 per healthy year gained. |
| Skin diseasesGBD: Skin and subcutaneous diseases | 161,1001 0.3% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
| Vision and hearing lossGBD: Sense organ diseases | No deaths counted | Counted with the group (see above) | Group only (see above) | GBD counts no deaths in this group: vision and hearing loss are counted as years lived with disability. See Body, under Thrive. |
| Musculoskeletal disordersGBD: Musculoskeletal disorders | 131,7001 0.2% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
| Other non-communicable diseases (including birth defects)GBD: Other non-communicable diseases | 1.3 million1 2.2% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
| Injuries 4.9 million deaths · 8.1%Health aid, whole group: Not tracked Injuries are not one of IHME’s health-aid focus areas. | ||||
| Road and other transport injuriesGBD: Transport injuries | 1.4 million1 2.4% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. WHO puts drink-driving laws enforced with sobriety checkpoints in its next category (above I$100 per healthy year). |
| Falls, drowning, fire, poisoning and other accidentsGBD: Unintentional injuries | 2.1 million1 3.5% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
| Suicide, homicide and conflictGBD: Self-harm and interpersonal violence | 1.4 million1 2.3% | Counted with the group (see above) | Group only (see above) | Gap No evaluator we follow has a top pick here yet. |
Money and deaths
Where the money goes is not where the deaths are.
In 2023, health aid to low- and middle-income countries was about $64.6 billion2, of which $12.3 billion came from the Gates Foundation, companies and other private philanthropy.2 Set its shares beside the shares of deaths and four contrasts stand out. Each is worked out in full, from the cited figures only.
HIV/AIDS
15.7×
Working: ($14.1 billion ÷ $64.6 billion = 21.8% of aid) ÷ (833,400 ÷ 60,043,100 = 1.39% of deaths) = 15.7.21
COVID-19
14.4×
Working: ($12.4 billion ÷ $64.6 billion = 19.2% of aid) ÷ (797,600 ÷ 60,043,100 = 1.33% of deaths) = 14.4.21
IHME puts the last one in its own words: “Non-communicable diseases account for 52.0% of all healthy years lost in low- and middle-income countries but constituted less than 4.0% of development assistance for health each year.”2 Injuries, 8.1% of the world’s deaths, are not one of IHME’s aid categories at all, so we can’t put a number on their share.
The other axis
Future deaths: the risks that haven’t happened yet.
Pandemics, nuclear war, misuse of advanced technology such as AI, and other rare events that could kill very many people at once. These are rare and enormous, and the table above can’t show them, because it counts deaths that already happened in one year.
Why it isn’t counted twice. The Survive table counts deaths that happened in one year. When a catastrophe happens, GBD files its deaths in the same tree (COVID-19 sits under respiratory infections; war under self-harm and violence). The tail-risk axis is about deaths that have not happened yet, so it is never added to the table.
How big a catastrophe can be
14.9 million
excess deaths associated with the COVID-19 pandemic in 2020 and 2021, WHO’s estimate (range 13.3 to 16.6 million).13
Money
Pandemic preparedness aid peaked in 2020 at over $1 billion.2 Other catastrophic risks: Not tracked
Strongest evidence-backed option
Gap No evaluator we follow has a top pick here yet. We show no probability for any catastrophe, because none comes from a source we follow.
Thrive
More life in the living.
Counting deaths misses most of what we mean by a good life. GBD already splits health loss into two parts that never overlap: years lost to early death, and years lived with disability. In 2023 the second part was 990 million years, about 35.4% of all health lost.7 That is the seam between our two halves: deaths on one side, suffering among the living on the other.
Health is only part of flourishing, so Thrive has six domains, plus one readout.
The rule that keeps it exclusive
Every measure belongs to exactly one domain, and a domain is judged only by its own measures. Money that buys medicine shows up once as income (Means) and once as health (Body), but we never add the two together. A programme is listed under the domain its evaluator actually measured. The readout, how people say their life is going, sums up the rest, so it is never added to them either.
The check that keeps it exhaustive
The OECD’s well-being framework covers 11 dimensions of current well-being.14 Each falls into exactly one of ours (our mapping):
- Body: health (physical), environmental quality
- Mind: health (mental)
- Means: income and wealth, work and job quality, housing, work-life balance
- Learning: knowledge and skills
- Connection: social connections
- Safety and voice: safety, civic engagement
- The readout: subjective well-being
| Domain | How big, in its own measure | Money tracked | Strongest evidence-backed option |
|---|---|---|---|
| BodyPhysical health while alive: pain, injury, illness and disability that do not kill (everything GBD counts as years lived with disability, except mental disorders). | 990 million years lived with disability in 2023, about 35.4% of all health lost7; low back pain alone: 70.4 million7; falls: 53.1 million7; hearing loss: 53.1 million7. | Not tracked Not tracked apart from the health-aid areas in the Survive table. | Gap No evaluator we follow has a top pick here yet. The picks on this page are costed per life saved or per WELLBY, never per year of pain relieved: a real hole in the map. |
| MindMental health: depression, anxiety and other mental disorders (GBD’s mental-disorder share of years lived with disability). | nearly 1 in 7 people (1.2 billion) lived with a mental disorder in 20238; depression: 56.0 million years lived with disability7; anxiety: 55.4 million7. | Not yet sourced IHME mentions mental health programmes inside non-communicable disease aid, but states no figure. | StrongMinds HLI Top Charity Or let the evaluator choose: Happier Lives Fund. |
| MeansMoney, work, a home and time: income and wealth, jobs, housing, and the hours left after work. | 811 million people live on less than $3 a day (2026)9. | Not yet sourced | GiveDirectly HLI’s benchmark |
| LearningWhat people know and can do: reading, numbers and skills. | an estimated 70% of 10-year-olds in low- and middle-income countries could not understand a simple written text (2022)10. | Not yet sourced | Gap No evaluator we follow has a top pick here yet. An open field for careful donors. |
| ConnectionRelationships: family, friends, community, and not being lonely. | 1 in 6 people worldwide is affected by loneliness (2025)11.WHO links loneliness to more than 871,000 deaths a year. Those deaths are already inside the cause-of-death count (as heart disease, stroke and so on), so we never add them on top. | Not tracked | Gap No evaluator we follow has a top pick here yet. |
| Safety and voiceFreedom from fear and violence, and a say in how you are governed. | global freedom declined for the 20th year in a row in 202512.Deaths from violence and conflict are counted once, under Survive. Here we count the fear and the lost freedom of the living. | Not tracked | Gap No evaluator we follow has a top pick here yet. |
The readout. Not a seventh domain: the summary readout. The Happier Lives Institute measures it in WELLBYs. We never add it to the domains, because it already reflects them. “One WELLBY is equivalent to a 1-point increase on a 0-10 wellbeing scale for one year.”5 We never turn WELLBYs into lives, or lives into WELLBYs.
Beyond humans
Suffering counts wherever it is felt.
The circle of concern does not stop at our species. The scale is hard to hold in mind: in 2024, about 86.3 billion land animals of these seven kinds were slaughtered for meat, by FAO’s count15 (our sum of FAO’s figures for each kind, which leave out dairy and egg production).
- Chickens 78.5 billion
- Ducks 4.2 billion
- Pigs 1.5 billion
- Sheep 703.9 million
- Goats 563.7 million
- Turkeys 504.1 million
- Cattle 304.7 million
Fish and other sea animals: FAO reports them by weight, not by number. Not yet sourced as a count. Money for animal welfare: Not tracked by any source we follow.
Strongest evidence-backed option
Animal Charity Evaluators’ 2025 Recommended Charities: Animal Welfare Observatory, Aquatic Life Institute, Çiftlik Hayvanlarını Koruma Derneği, Dansk Vegetarisk Forening, Good Food Fund, Shrimp Welfare Project, Sinergia Animal, Sociedade Vegetariana Brasileira, The Humane League, Wild Animal Initiative.16
In ACE’s words, “These charities work in ways that are most likely to produce the greatest gains for animals, actively evaluate and improve their programs, and have a demonstrated need for more funding.” ACE also runs a Recommended Charity Fund that splits gifts among them.
We weigh no animal against a person, and no species against another.
Principles
Seven rules for giving on this map.
- Judge the result, not the label. Ask what changed, and compared with what.
- Count every life once. If two charities claim the same death averted, one of them is wrong.
- Keep the units apart. Lives, years of health, wellbeing and animals are different goods. Your values decide between them.
- Put the money next to the need. A neglected cause with a proven fix is the best news in giving.
- Name the gap. “Not tracked” and “no top pick yet” are findings, not failures.
- Check again later. A gift is a prediction. Write it down and see whether it came true.
- Nobody pays to be on the map. No placement here is for sale, and none ever will be.
Planning a gift across causes? Plan a gift shows the evaluators’ picks for the causes you choose. Every figure, quote and source on this page is also in a downloadable copy of the map.
Sources
Every figure above links here. We read each source on 28 October 2026 and kept a dated copy. Figures are as published; estimates are the publisher’s, not ours. Shares and ratios are our arithmetic on those figures, shown in full.
- Global burden of 292 causes of death in 204 countries and territories and 660 subnational locations, 1990–2023: a systematic analysis for the Global Burden of Disease Study 2023 · IHME / GBD 2023 Causes of Death Collaborators, The Lancet, 2025
- Financing Global Health 2023: The Future of Health Financing in the Post-Pandemic Era · Institute for Health Metrics and Evaluation (IHME), 2024
- New Incentives (programme review) · GiveWell, 2025
- Our Top Charities · GiveWell, 2025
- Our recommended charities · Happier Lives Institute, 2026
- Tackling NCDs: best buys and other recommended interventions for the prevention and control of noncommunicable diseases, 2nd ed. · World Health Organization, 2024
- Burden of 375 diseases and injuries, risk-attributable burden of 88 risk factors, and healthy life expectancy in 204 countries and territories, including 660 subnational locations, 1990–2023: a systematic analysis for the Global Burden of Disease Study 2023 · IHME / GBD 2023 Disease and Injury and Risk Factor Collaborators, The Lancet, 2025
- Mental disorders (fact sheet) · World Health Organization, 2025
- Poverty: overview · World Bank, 2026
- 70% of 10-Year-Olds now in Learning Poverty, Unable to Read and Understand a Simple Text · World Bank, 2022
- Social connection linked to improved health and reduced risk of early death · World Health Organization, 2025
- Freedom in the World 2026: The Growing Shadow of Autocracy · Freedom House, 2026
- 14.9 million excess deaths associated with the COVID-19 pandemic in 2020 and 2021 · World Health Organization, 2022
- How's Life? (series page) and the OECD Well-being Data Monitor · OECD, 2024
- Yearly number of animals slaughtered for meat (FAO, Production: Crops and livestock products, with processing by Our World in Data) · Food and Agriculture Organization of the United Nations, via Our World in Data, 2025
- 2025 Recommended Charities · Animal Charity Evaluators, 2025