Intended for healthcare professionals


Institutional ageism in global health policy

BMJ 2016; 354 doi: (Published 31 August 2016) Cite this as: BMJ 2016;354:i4514
  1. Peter G Lloyd-Sherlock, professor1,
  2. Shah Ebrahim, professor2,
  3. Martin McKee, professor3,
  4. Martin James Prince, professor4
  1. 1School of Development Studies, University of East Anglia, Norwich NR4 7TJ, UK
  2. 2Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK
  3. 3ECOHOST, London School of Hygiene and Tropical Medicine
  4. 4Institute of Psychiatry, King’s College London, UK
  1. Correspondence to: P G Lloyd-Sherlock p.lloyd-sherlock{at}

Peter G Lloyd-Sherlock and colleagues argue that a focus on premature mortality is discriminating against the needs of a growing older population

The sustainable development goals agreed in March 2016 by the United Nations General Assembly set the global development agenda for the next 15 years. They include an ambitious target to reduce premature mortality from non-communicable diseases by a third by 2030. Premature mortality, defined by the World Health Organization as deaths occurring between the ages of 15 and 70, has gained broad acceptance in health research and policy over the past decade. We argue that it is explicitly ageist, reflecting institutional ageism in global health policy. Its inclusion in the sustainable development goals sends a strong signal in favour of discriminating against older people in the allocation of health resources and the collection of data. We consider the emergence of ageist approaches in global health policy and the potential effects of ageism in the sustainable development goals. We propose a less discriminatory approach.

Institutional ageism

Ageism is defined as: “a process of systematic stereotyping and discrimination against people because they are old.”1 Institutional ageism differs from, but is related to, interpersonal ageism. It involves the inclusion of ageist principles in formal rules and procedures and in wider institutional cultures. It is characterised by language consistently depicting older people in negative terms.2

Less research has been done on ageism than on other forms of discrimination, such as racism and sexism. Studies of institutional ageism have mainly explored discrimination in employment and the workplace,3 and few studies explicitly examine its role in health. Most of the health studies focus on specific groups, such as older people who are lesbian, gay, or bisexual or racial minorities.4

Several arguments have been used to justify age discrimination in health policy. The “fair innings” argument posits that everyone is entitled to a certain quality adjusted life expectancy and so policies should prioritise interventions that deliver this, even if they often favour younger people.5 Accordingly, ageism may be fundamentally different to discrimination based on gender or race since it redresses existing disadvantage. Of course, there is the vexed question of how many years might be considered a “fair innings”: something that would require informed debate and social consensus among people of all ages before being adopted. More generally, these justifications for age discrimination contradict the universal principle of health as a fundamental right for all.

Other arguments for age discrimination draw on concepts of human capital and efficiency. These claim that older people make fewer contributions to society than younger adults and thus have less social and economic value.6 This links to stereotypes equating old age with frailty, weakness, and dependency, which are contradicted by evidence that many older adults make substantial economic and social contributions.7 8 It is sometimes observed that older people’s health problems are relatively expensive to treat and that interventions generate few returns.9 10 Yet many conditions affecting older people can be treated cheaply, substantially extending healthy life expectancy.11

Epidemiological roots of ageism: from YLLs to DALYs

Simple measures of overall mortality are skewed towards conditions that disproportionately kill older people, since most deaths occur in later life.12 This could unfairly understate conditions affecting younger age groups. To take account of this, the years of potential life lost (YLL) approach to measuring mortality was developed in the 1980s.13 YLLs weight the burden of mortality according to the number of years between the age at which death occurs and an arbitrary age threshold in later life, typically somewhere between 65 and 80. Averting the death of a person over that age is given a value of zero.

YLLs have been widely applied in research and health policy, with a range of threshold ages. However, they raise ethical and practical problems, including where to set the age threshold beyond which any added life years are deemed worthless. Although proponents of YLLs may not advocate that the value of survival beyond a certain age should literally be understood as worthless, the approach frames thinking in this ageist direction.

In 1993 the World Bank and WHO launched disability adjusted life years (DALYs) as a tool for studying patterns of disease and assessing health priorities.14 DALYs apply a disability weight to remaining years of life, up to a fixed YLL age threshold. Originally DALYs applied an additional weighting according to the age at which each year of life was lost: a year lost in childhood or old age was worth much less than for other ages. For example, the value given to a year lived at age 70 was less than half one at age 25. It was argued: “The young, and often the elderly, depend on the rest of society for physical, emotional and financial support. Given different roles and changing levels of dependency with age, it may be appropriate to consider valuing the time lived at a particular age unequally.”6

The only evidence used to justify age weighting was an “informal polling of tuberculosis programme managers,” and no evidence was produced to justify the strength of the age weights. Several studies noted that DALY age weighting and the Global Burden of Disease unfairly undervalued conditions mainly affecting older people.15 16 17 Yet age weighting continued in the Global Burden of Disease studies until 2010.

Global priorities and older people

By 1960 roughly half the global population aged 60 or over lived in low and middle income countries, rising to two thirds by 2015. Despite this shift, until the 1990s research and policy discussion about older people’s health took place almost exclusively in high income countries.18 There has been some growth in geriatric research and practice in low and middle income countries since, but global policy priorities have remained heavily focused on other issues, such as infectious disease and reproductive health. In the case of HIV/AIDS, WHO and UNAIDS supported the collection of prevalence data only for people aged 15-49.

The slow response of the global health community to demographic change reflects several factors, including the influence of UN agencies and non-governmental organisations interested in specific issues and other population groups. At the same time, agencies interested in older people in low and middle income countries mainly focused on the extension of pension provision.19 Embryonic international networks for older people’s health struggled to influence the wider global discourse and sometimes lacked awareness of ageism in global policy. For example, the UN’s 2002 Madrid International Plan of Action on Ageing makes no reference to bias against older people in DALYs.20

The neglect of older people was paralleled by the low priority given to non-communicable diseases (NCDs). This resulted from a misperception that these diseases mainly affect wealthy people and sustained resistance from related industries.21 Ageism may have contributed to this neglect, inasmuch as NCDs are viewed as primarily “conditions of older age.” From 2000, when WHO published its first global strategy for prevention and control of NCDs, an influential global network began to promote them as a development concern.22 23 Potentially, this represented an opportunity to improve recognition of older people’s health, but the network took a different line. Many WHO documents and publications on NCDs make no reference to older people, or merely identify population ageing as a “driver” of NCD pandemics.24 They emphasise the effect of these diseases on younger adults to justify giving them a higher priority than conditions mainly affecting older people. For example, WHO’s 2008-13 Action Plan for the Prevention and Control of NCDs refers to gender, race, and people with disabilities but not to older people.24 Six of the 27 key NCD indicators that WHO advocates member states should prioritise, including tobacco use and raised blood pressure, are limited to people aged 25 to 64. No reasons are provided for these age limits.

Premature mortality: a new form of institutional ageism?

From 2008 the concept of premature mortality started to gain widespread acceptance among global agencies. In part, this was driven by concerns about the economic effects of disease, especially on people of “working age” and the view that scarce resources should focus on preventing death at younger ages. The foreword of WHO’s 2008-13 action plan claims that reducing premature mortality from NCDs could save millions of lives, but the term is not defined and is not mentioned in the rest of the plan.24 By contrast, premature mortality took centre stage in the 2010 global status report on NCDs. It is defined as mortality below the age of 70, lowering the threshold age of zero utility applied by DALYs by more than 10 years without any justification.25

The 2010 WHO report was the reference document for the 2011 UN high level meeting on NCDs, a political declaration, and a plan ratified by the 2013 World Health Assembly. The plan commits to reducing premature mortality from cardiovascular disease, diabetes, cancers, and respiratory disease by 25% between 2010 and 2025 and has already influenced national policies.26 A modified version of the premature mortality target was included in the sustainable development goals: “by 2030 reduce by one-third premature mortality from NCDs through prevention and treatment.” The goals do not define premature mortality but presumably refer to WHO’s under 70 threshold. While such global targets do not always translate directly into national policy, they are influential.27 These targets are therefore likely to promote institutional ageism in several ways, discouraging research and data collection, focusing resources on younger people, and exacerbating existing discrimination.

The potential effect on research and data collection is shown by the earlier exclusion of older people from HIV targets. In 2010, fewer than half UNAIDS national HIV progress reports included data on people aged over 50.28 A review of clinical trials on reducing the risk of sexually transmitted diseases found that over two thirds excluded people aged 50 or over.29 There is already evidence that older people are also excluded from studies of NCDs and from randomised controlled trials, making findings potentially inapplicable to them.30 A growing number of epidemiological studies apply the 70 and over premature mortality threshold.31 32 33

The challenges of reducing premature mortality from NCDs by a third are substantial, requiring a large reallocation of resources away from older people. Advocates of premature mortality argue early life interventions ultimately benefit people of all ages, but this approach excludes people currently aged 70 or over.34

Use of the term premature mortality exacerbates and justifies existing age discrimination in healthcare11 35 by implying that survival after the age of 70 is fundamentally less important than survival at younger ages. It also distracts attention from major challenges that especially affect older people, including multimorbidity and palliative care.36

Redressing the balance

Since the 1990s key aspects of health policy have been framed by ageist principles. The arguments used to justify age discrimination are evidently invalid. The prominent role given to premature mortality thresholds shows that ageism is becoming increasingly blatant. It is inconceivable that global targets would similarly discriminate against other groups, such as women or people with disabilities.

Engaging with ageism in health policy does not mean throwing the baby out with the bathwater. We still need mortality indicators such as YLLs, but they must be interpreted with care. Additionally, policy makers should explicitly assess how the choice of measure affects the priority given to each age group.

Highly cost effective health interventions exist for people of all ages. Improved surveillance increases the scope to capture the disease burden in older people. For example, promoting awareness and management of hypertension can substantially reduce preventable deaths and disability.37 In 2015 WHO published a major report on ageing and health,36 indicating that it is ready to jettison ageist concepts and champion a more ethical approach. Such progress must not be undermined by poorly considered global targets.

Key messages

  • Despite growing numbers of people aged over 60 in low and middle income countries, health priorities remain focused elsewhere

  • The target in the sustainable development goals to reduce premature mortality from non-communicable diseases reflects wider ageism in global health policy

  • Policy makers should consider how the choice of measure affects the priority given to each age group


  • Contributors and sources: PL-S is lead author and guarantor. He produced a first draft of the paper. This received substantial inputs from all coauthors. The paper is based on published research and other publicly available sources of information. All authors have participated in an informal campaign to alter the phrasing of SDG 3.4, which they view as explicitly ageist in its current form.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription