AEGIS — The Human Algorithm Foundation

Global observatory · 2026 edition

The silent epidemic

One in six adults over 60 experiences abuse each year, yet the world writes almost no laws to see them. This is a comparative reading of who is harmed, how, and what a society builds to protect its final chapter.

Curated by AEGIS — The Human Algorithm
In collaboration with Fondazione Età Grande
42
Countries mapped
15.7%
WHO global past-year prevalence
62%
Countries where psychological abuse dominates
7
Nations with a dedicated law
6
Nations with a national hotline

The scale of the phenomenon

One hundred and forty-one million today. Three hundred and twenty million by 2050.

Applying the WHO past-year prevalence rate of 15.7% to the world's population aged 60 and over, current estimates place the number of older adults abused each year at around 141 million. As the global population of people aged 60+ doubles — from roughly 1 billion in 2020 to about 2.1 billion by 2050 — the same prevalence rate translates into approximately 320 million victims per year. Even a stable rate becomes an emergency when the denominator doubles.

2020 141M Older adults abused each year — one in six of the world's population aged 60+ (WHO, based on Yon et al., 2017)
2030 220M Projected victims when the 60+ population reaches ~1.4 billion, holding the current prevalence rate constant
2050 320M Projected victims as the 60+ population reaches ~2.1 billion — official WHO / UN DESA projection

Older adults affected by abuse each year, worldwide

Millions of people aged 60+ · past-year prevalence held constant at 15.7%

Sources: WHO fact sheet Abuse of older people (2024 update); UN DESA World Population Prospects 2024; Yon Y. et al., Lancet Global Health 2017.

The projection is arithmetic, not fatalistic: it assumes societies do nothing new. Every dedicated law, hotline, training programme or family-support policy bends the curve below the 320 million mark. That is the entire premise of this observatory.

Why an algorithm can be human

Four lenses on a hidden phenomenon

Elder abuse is what happens when the algorithm of a society — its rules, its habits, its care networks — stops recognising a person as fully human. AEGIS reads it through four dimensions.

A

Algorithm

The formal architecture of protection: laws, protocols, mandatory training. What a country has built to intercept harm before it repeats.

E

Ethics

How societies name the violence. Where language exists — psychological, financial, neglect — visibility follows. Silence is a policy choice.

G

Generations

Every case sits at a family or institutional interface. The perpetrator is almost always someone the elder trusted. The wound is relational before it is physical.

IS

Integrity & Sustainability

A society's dignity floor. Systems that fail their oldest members eventually fail everyone else — because the promise of care is what makes ageing possible.

The world in one map

Where the harm concentrates

Each circle represents a country. Colour indicates the dominant form of abuse in the most recent national study; size reflects past-year prevalence — how many older adults were harmed in the twelve months before the survey.

Dominant type Psychological Physical Financial Neglect Sexual Data limited

Typologies

Five forms of harm, unevenly counted

Psychological abuse — humiliation, isolation, verbal aggression — is the most reported form in nearly every country that measures. Financial exploitation follows in high-income nations; neglect prevails in fragile care systems.

Global averages across five typologies

Weighted mean of national studies (past-year prevalence, %)

Top ten highest prevalence

Countries with the highest overall past-year prevalence

Normative capacity

The protection gap

For each country AEGIS scores four building blocks of institutional protection: a dedicated law, a national hotline, official detection guidelines, and mandatory professional training. Most nations have none.

42 countries

Country Region Prevalence Dominant type Protection Instruments

Risk factors and ageism

What makes a life visible to violence

Prevalence maps say where. Risk factors say why. Six determinants recur in every meta-analysis — and they are all social before they are individual.

01

Social isolation

An elder without regular contact is up to three times more likely to experience psychological or financial abuse. Loneliness is not sentiment — it is exposure.

02

Cognitive dependency

Dementia and severe frailty multiply risk. The caregiver relationship, when unsupported, can invert into a corridor of harm.

03

Ageism

Where a society treats age as decline, harm becomes invisible. Ageist attitudes correlate with lower reporting even by victims themselves.

04

Institutional settings

Rates in long-term care rise sharply — up to 64% of staff self-report having witnessed abuse in the past year. Understaffing is the strongest predictor.

05

Financial dependence

Shared bank accounts, unpaid caregiving, and unclear inheritance rules open the door to what the literature calls "material appropriation".

06

Gender

Older women face disproportionate psychological and sexual abuse; older men face financial and physical harm. The mask changes; the wound is universal.

Methodology

How this observatory reads the data

Sources

  • Yon et al., Lancet Global Health 2017 — systematic review and meta-analysis of community prevalence
  • WHO — World Report on Ageing and Health; Elder abuse fact sheet (2024 update)
  • INPEA — International Network for the Prevention of Elder Abuse country profiles
  • National surveys: NEAIS (US), Age UK / Hourglass, ISTAT, Statistics Canada, Cabinet Office (Japan), NSO (Korea)
  • UN DESA — World Population Ageing reports for demographic denominators
  • Peer-reviewed national studies indexed in PubMed 2015–2025

Editorial notes

Prevalence figures refer to past-year experience of abuse among adults aged 60 and over in community settings, unless otherwise stated in the country card. Institutional prevalence is reported separately and is typically higher.

The protection score is an editorial synthesis by the AEGIS board of four objective criteria: dedicated national law, national hotline, official detection guidelines, mandatory professional training. Each present criterion adds one point.

Countries with data limited markers had no citable national prevalence study at the time of publication. Their normative infrastructure is still assessed.

The rights rating is an editorial synthesis by AEGIS and does not substitute the underlying primary sources.