Frailty: a global health challenge in need of local action

Search strategy and evidence synthesis

To understand the geographical representativeness of frailty-related research, we searched PubMed, Google Scholar and Web of Science for different types of reviews that pooled evidence to understand frailty prevalence, incidence, screening, management and guidelines in the last decade. These reviews helped us understand ongoing frailty debates and served as starting points to explore region-specific and country-specific research through citation searches. We excluded reviews that focused on frailty in specific disease populations. We also reviewed reports published by the WHO to understand global healthy ageing goals and strategies. This helped us formulate our arguments for context-specific considerations in the global frailty debate. We primarily used Web of Science and Google Scholar to retrieve frailty-related policy reports and documents for specific countries.

The subsequent sections present an overview of frailty from a global perspective and considerations for contextualising frailty in research, practice and policy.

Frailty in research and practice from a global perspective

In the frailty discourse, commonly reported challenges are the absence of an internationally recognised frailty definition, lack of standardised frailty assessment tools and interventions, and uncertainties about the true impacts of frailty. These issues are partly attributed to how frailty research is skewed towards specific geographies and economic areas. Most frailty research is from HICs, and the evidence base from lower-income countries is limited and has only been recently growing. Regarding research on the burden of frailty, systematic reviews highlight the disproportionate representation of research from HICs, as exemplified in a systematic review that determined the pooled prevalence of frailty from datasets from 62 countries. Only 2.6% (11/425) of the datasets came from populations in Africa while 36.2% (154/425) datasets came from populations in Europe, followed by the Americas (29.6%), Asia (27.8%) and Oceania (3.8%).12 The review also highlighted large in-region and in-country heterogeneities in the population samples. There are wide variations in the frailty research landscape even within continents. For example, in Asia, frailty research is predominantly conducted in Eastern Asia, particularly in Japan and South Korea. Japan is a super-aged society (>20% of the population is aged ≥65)15 while South Korea is projected to become one by 2025.16 In contrast, other countries in the region have different population structures; for example, 12% of the population in China is aged >6517 while the estimate is 5% in Mongolia.18 This research trend in Japan and South Korea reflects the national priorities to address the super-ageing populations in these countries. One systematic review estimated the prevalence of frailty and prefrailty in lower-income countries and noted that most studies came from upper-middle-income countries such as Brazil, China, Mexico and Colombia while representation from low-income countries and low-income and middle-income countries was comparatively lower.13

Research on frailty management also lacks geographical and population representativeness. Systematic reviews evaluating the effectiveness of frailty interventions have been predominantly conducted in HICs in Europe (such as the UK, the Netherlands, Spain and Sweden) and Asia (such as Japan, South Korea, Hong Kong and Singapore).19–24 Few trials have been conducted in China, Mexico and Brazil.19 22 24 This disproportionately skewed frailty research landscape, compounded by variations in frailty definitions and assessment tools, makes it challenging to establish the true global burden of frailty and to compare the effectiveness of frailty screening and management tools used worldwide.

The characteristics of overall frailty research emerging from lower-income countries and HICs are evidently heterogeneous. Although there are exceptions, frailty research conducted in lower-income countries has predominantly included smaller samples and focused on specific regions and populations within the country to quantify the frailty burden and to adapt different frailty assessment tools. In contrast, frailty research in HICs is exploring large administrative and survey databases to further understand frailty,25 determine the prognostic value of frailty for treatment and medical procedures26 27 and investigate the biological and genetic characteristics of frailty.28 29 Notably, the two predominant conceptualisations of frailty in the literature, the phenotypic and deficit accumulation models, were developed from two population surveys in HICs in North America: the Cardiovascular Health Study in the USA30 and the Canadian Study of Health and Aging in Canada.31 Population surveys from HICs, such as the Health and Retirement Study in the USA,32 English Longitudinal Study of Ageing in England,33 Korean Longitudinal Study on Health and Aging,34 and Japanese Study of Aging and Retirement,35 have been collecting information from as far back as the 1980s, making them data-rich sources for large-scale cross-sectional and longitudinal studies on frailty. In lower-income countries, frailty-specific research has recently shown an increasing trend, aided by the roll-out of population surveys on ageing. Examples include the Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa36 (baseline survey conducted in 2014–2015), Longitudinal Ageing Study in India (baseline survey conducted in 2017–2018)37 and Brazilian Longitudinal Study of Aging (ELSI-Brazil) (baseline survey conducted in 2015–2016).38 These surveys enable cross-country comparisons while ensuring the capturing of context-specific data. As these surveys mature, we can expect a more enriched frailty research landscape from lower-income countries that leverages more data points, enables generalisability and sheds light on context-specific nuances in frailty prevalence, screening and management.

Since the two seminal frailty conceptualisation publications in 2001,30 31 a plethora of frailty assessment tools has emerged (table 1), although developed and validated in populations in high-income contexts. More recently, evidence is emerging on adapting and validating existing frailty screening tools, such as the Edmonton Frailty Scale,39 cumulative deficit frailty index40 and Clinical Frailty Scale,41 in different populations. Efforts are also being directed towards understanding ageing on a global scale. The WHO Study on Global AGEing and Adult Health is an ongoing programme collecting longitudinal data on ageing and health from the following countries: China, Ghana, India, Mexico, the Russian Federation and South Africa.42 This data harmonisation exercise will further promote cross-country comparisons of ageing-related metrics, with more frailty-specific peer-reviewed publications expected in the future.

In clinical and social care practice, frailty screening and management have been the focus only in HICs owing to the large ageing populations and associated healthcare needs, driving their prioritisation in public health and policy. A recent systematic review retrieved clinical guidelines on frailty prevention and management and found that most originated from the UK, the USA and Canada.43 The remaining two guideline documents originated from the Asia-Pacific region44 and the International Conference on Frailty and Sarcopenia Research.45 Regarding real-world practice, the National Health Service in the UK became the first health system globally to integrate mandatory frailty assessment in routine primary care practice using the electronic frailty index, which automatically calculates frailty scores based on diagnostic codes in a patient’s electronic health record.46 47 However, variations persist in its uptake and utilisation in practice, with uncertainties in the tool’s accuracy and clinical utility.48 49 In Japan, the Long-Term Care Insurance system is a form of mandatory social insurance system that provides preventive and supportive care services to frail and prefrail individuals.50 Eligibility is decided by evaluating frailty status using the Kihon Checklist, followed by a multistep process to determine the dependency levels and care needs of potential beneficiaries.50 51 Evidence on frailty in clinical practice in lower-income countries is sparse; hence, how frailty is assessed, recorded and addressed by practitioners in these contexts remains unclear. As the manifestations of frailty overlap with those of chronic diseases and disability, identifying and diagnosing frailty as a distinct clinical condition may be limited due to overburdened health workforces and lack of appropriate training, equipment and frailty tools validated in local populations.52 For example, Nigeria reportedly has less than 100 geriatricians for 4 million older people.53 Therefore, the clinical opportunities to address frailty once it develops are limited in lower-income countries.

Considerations for contextualising frailty in research and practice

Evaluating the global narrative around frailty and ageing

There is a call to shift the rhetoric of ageing from a unidirectional path to inevitable morbidity to a positive transition where older people can still thrive, contribute to society and maintain meaningful interactions with their community. Regression in frailty states has been reported,64–67 but whether frailty progresses or regresses depends on various sociodemographic and clinical factors.7 68 69 In efforts to establish shared goals for public health response to population ageing, the WHO advocate moving away from a disease-focused model of ageing, including going beyond the traditional view on frailty.70 71 Frailty remains associated with negative connotations, being perceived as identity loss, deterioration of status, a synonym for weakness and dependency, and an inevitable part of the ageing process.72–74 The WHO is drawing focus on the intrinsic capacity of an individual, which in turn determines their functional ability.75 Intrinsic capacity is described as the loss in the functioning of multiple health domains (cognition, locomotion, sensorium, psychology and vitality),76 overlapping with the clinical characteristics of frailty. Belloni and Cesari pointed out how intrinsic capacity and frailty can be considered as ‘two sides of the same coin’ (Belloni G, Cesari M, p356).77 Intrinsic capacity encompasses a person’s physical and mental reserves throughout their life while frailty correlates with greater vulnerabilities to adverse outcomes from losses in multisystemic physiological reserves in later life.78 Thus, intrinsic capacity is a measure of what a person has (reserves) rather than what they lack (deficits),79 lending the construct a more positive connotation. Intrinsic capacity may also be beneficial in understanding longitudinal changes in residual capacities throughout the life course, especially as most frailty instruments are only validated in older populations.80 Research on operationalising intrinsic capacity81 82 and the contributing factors,83 84 associations with adverse outcomes85 86 and role in evaluating interventions81 87 is emerging from both HICs and lower-income countries. However, inconsistencies in the measurement models and operationalisation methods prevail,88 impeding translations from theory to real-world practice.77 One may argue that the shift from frailty to intrinsic capacity may diminish the wide-ranging impacts of frailty on individuals, communities and health systems. Vernon described how the narrative of frailty as a long-term, clinical, identifiable and potentially reversible condition fueled policy implementation in the UK.89 Some frailty tools, such as the frailty index, also capture characteristics not included in the intrinsic capacity composite score, such as comorbidities and social vulnerability. Frailty is challenging to distinguish from manifestations of multiple chronic conditions and disability; hence, screening for frailty specifically using validated tools is important to ensure people receive timely intervention. It should be acknowledged that the construct of intrinsic capacity was developed from the need to provide personalised and integrated care plans, and this has also been recognised for frailty. As research on intrinsic capacity is still in its infancy, there are opportunities to develop a standardised and feasible measurement tool. A composite score that integrates both intrinsic capacity and frailty components may be considered. Although such a standardised tool may have global research and policy benefits by facilitating generalisability and consistency, we recommend that this tool rather serves as a blueprint for more context-specific adaptations that would account for national policies, population health priorities and feasibility in community settings. Recent studies have reported associations between intrinsic capacity and different dimensions of frailty.90–92 Further research on the potential of embedding intrinsic capacity measurement into current frailty screening and chronic disease management pathways, and public perceptions of intrinsic capacity as an ageing-related concept is warranted.

Considering the social determinants of health in the frailty discourse

Frailty involves a complex interplay of non-modifiable and modifiable risk factors, which may contribute to differing incidence rates, manifestations and trajectories. The WHO’s social determinants of health93 have gained a strong foothold in the global health landscape. Although there is a non-modifiable, age-related component in frailty development,2 addressing the social determinants of health can help lower the risk of frailty onset and worsening across populations and contexts, especially given how frailty is preventable and reversible. Drawing from this conceptual framework, Sadana et al94 proposed a preliminary model for healthy ageing and health equity in a global context, consisting of four components: (1) the physical–socio–economic–political or overall context; (2) genetic inheritance and socioeconomic position; (3) intermediary determinants (such as health and social care systems, intrinsic capacity and the built environment) and (4) the outcome of interest (healthy ageing).94 Evidence already shows associations between elements of these components (such as the built environment, sociodemographic characteristics and comorbidities) and frailty.54 95 Hence, this model for healthy ageing can also be applied to frailty. Given the limited availability of specialised geriatric workforces to identify and treat frailty in lower-income countries, the model’s components could guide research, policy and actionable strategies to achieve the outcome of interest, that is, frailty prevention and regression.

For example, the social determinants of health and the healthy ageing model can be used to address chronic diseases, which are risk factors for frailty and occur at younger ages in lower-income countries and in global majority populations in HICs.96 97 Chronic diseases are related to the intermediary determinants of the healthy ageing model. Living with multiple chronic diseases (multimorbidity) is a growing challenge in lower-income countries that demands context-specific guidelines for its diagnosis and management.98 Thus, given the correlation between chronic diseases and frailty, frailty incidence rates may be greater in these populations and geographies, and frailty trajectories may begin earlier and progress more rapidly. Integrating opportunistic frailty screening and management into existing models of chronic disease care may be feasible and cost-effective in such contexts, and determinants such as the built environment, lifestyle behaviours and exposures can be considered while designing and implementing interventions for both chronic diseases and frailty. This demonstrates how intrinsic and extrinsic determinants at multiple levels, such as coverage of services to manage chronic diseases, chronological age, ethnicity, comorbidities and lifestyle, are worth considering for frailty prevention and management in health and social care practice.

Potential socioeconomic benefits of addressing frailty

Whether the benefits of integrating frailty assessment in routine care outweigh the costs to already strained, resource-limited health and social care systems requires consideration. The frailty research landscape is saturated with studies on new assessment tools rather than management strategies.99 Thus, actions necessary following frailty identification may still be unclear to healthcare practitioners in lower-income countries, especially given the lack of evidence on frailty management in these contexts. We argue that frailty should still be considered in healthcare practice and policy in lower-income countries for several reasons. First, evidence shows that the onset of frailty is associated with greater healthcare use and subsequent healthcare costs,100–102 given the multidisciplinary care needs for the condition.103 Second, although evidence is mixed, frailty has been shown to correlate with socioeconomic factors such as education levels, income levels, residence status (rural/urban), neighbourhood deprivation levels and ethnicity.104–109 Thus, large disparities in socioeconomic status in lower-income countries may drive the disproportionate incidence and progression of frailty, further widening health and social inequities. Third, frailty poses threats to the economic productivity of older people aged ≥55, who make up a larger share of the labour workforce globally.110 If there is a risk of earlier frailty onset in populations in lower-income countries, prime economically productive years may be spent being frail, with serious implications on public health expenditure and pension contributions. This can have adverse consequences on the long-term economic health of a country. Evidence suggests that an increasing ageing population could drive long-term economic growth in lower-income countries due to accumulation of capital and assets, consumption patterns and contributions to household incomes if they continue to work.111 Finally, frailty is a dynamic condition, and research has shown that the likelihood of improving from prefrailty to robust is greater than improving from frailty to robust.65–67 Identifying early stages of frailty may facilitate timely intervention, thus optimising scope for frailty reversal.112 This is also important in hospital settings, as deconditioning associated with hospitalisation and prolonged bedrest are linked with frailty onset and progression. Frailty should be considered in postdischarge pathways to prevent long-term consequences on individuals and their carers; loss of work productivity and subsequent strains on secondary care, community care and economic resources. Thus, national policies and guidelines that support frailty screening and interventions are in the best interests of healthcare systems and labour markets. These policies and guidelines should account for the heterogeneous risk factors and population dynamics, feasibility of frailty screening in specific healthcare settings and cultural considerations for frailty interventions related to exercise and nutrition. Preventing and optimally managing frailty can help healthcare systems steer away from catastrophic costs and ensure older people remain active contributors to national economies and their communities.

Strengths and limitations of this review

Owing to the large body of literature, our review is limited by its lack of a systematic approach to extract relevant literature and in-depth analyses. However, we drew from evidence from multiple countries and contexts to provide a narrative overview of the breadth of the frailty literature through a global health lens. Our review is also coauthored by clinical and research experts with broad knowledge of frailty and healthy ageing in research, practice and policy, thus providing balanced arguments on frailty and healthy ageing debates.