Concepts and definitions of healthy ageing: a systematic review and synthesis of theoretical models.

Abstract

Healthy ageing (HA) has been defined using multiple approaches. We aim to produce a comprehensive overview and analysis of the theoretical models underpinning this concept and its associated normative terms and definitions.

We conducted a systematic review of peer-reviewed HA models in Embase.com , Medline (Ovid), Cochrane CENTRAL, CINAHL, PsycINFO, and Web of Science until August 2022. Original theoretical papers, concept analyses, and reviews that proposed new models were included. Operational models/definitions, development psychology theories and mechanisms of ageing were excluded. We followed an iterative approach to extract the models’ characteristics and thematically analyze them based on the approach of Walker and Avant. The protocol was registered in PROSPERO (CRD42021238796).

Out of 10,741 records, we included 59 papers comprising 65 models/definitions, published in English (1960–2022) from 16 countries in Europe, Asia, and America. Human ageing was described using 12 normative terms, mainly (models (%)): successful (34 (52%)), healthy (eight (12%)), well (five (8%)), and active (four (6%)). We identified intrinsic/extrinsic factors interacting throughout the life course, adaptive processes as attributes, and outcomes describing ageing patterns across objective and subjective dimensions (number of models/definitions): cognitive (62), psychological (53), physical (49), social (49), environmental (19), spiritual (16), economic (13), cultural (eight), political (six), and demographic (four) dimensions. Three types of models emerged: health-state outcomes (three), adaptations across the life course (31), or a combination of both (31). Two additional sub-classifications emphasized person-environment congruence and health promotion.

HA conceptualizations highlight its multidimensionality and complexity that renders a monistic model/definition challenging. It has become evident that life long person-environment interactions, adaptations, environments, and health promotion/empowerment are essential for HA. Our model classification provides a basis for harmonizing terms and dimensions that can guide research and comparisons of empirical findings, and inform social and health policies enabling HA for various populations and contexts.

MM, ZMRD, and OI are supported by the European Union’s Horizon 2020 Marie Skłodowska-Curie grant No 801076, and MM is also supported by the Swiss National Foundation grant No 189235.

Keywords: Healthy ageing, Concept, Definition, Model, Dimension, Concept analysis

Evidence before this study

We conducted an umbrella review of healthy ageing concepts and definitions in pubmed and google scholar between October and December 2020 using the terms: healthy, successful, active, and optimal ageing combined with definition, model, concept, domain, construct, determinant, mechanism, or dimension. We included reviews and concept analyses of theoretical and operational models and definitions in social and medical gerontology, mechanisms of ageing, and determinants of healthy ageing until December 31, 2020. The findings showed a great heterogeneity in defining and operationalizing this concept and a gap on systematic reviews and analyses that comprehensively map out the theoretical literature on healthy ageing. This gap underlies the lack of clarity on the theoretical underpinnings from which operationalizations of healthy ageing emerged.

Added value of this study

By combining systematic review and concept analysis approaches in synthesizing theoretical models, this study comprehensively delineates and clarifies the concept of HA and its dimensions and explores whether a consensus is possible on a definition or model valid and applicable in multiple contexts.

Implications of all the available evidence

Bringing convergence and conceptual clarity to HA has wide implications in identifying gaps and improving the clinical and research applications of the concept in the context of existing operationalizations, in exploring the validity of existing indices and measures across the different dimensions of ageing, and in informing policies to enable HA measures and strategies for everyone.

Introduction

The increase in life expectancy at birth and global demographic shift into an older age present unprecedented social and economic challenges to the modern world. 1 This is of particular concern as this increase in life expectancy is often accompanied by increased years spent in ill health. 2 This raises the question of how to promote healthy ageing (HA) in the population.

The WHO defines HA as maintaining a functional ability that enables individuals to meet their needs and contribute to society within their environment. However, many other definitions also exist. Past efforts in exploring HA show an explosion of normative terms, including active, resilient, and successful ageing, among many others, and a great heterogeneity in operationalizing conceptual models and definitions. 3 , 4 , 5 , 6 , 7 Perhaps the difficulty in achieving conceptual clarity in these terms lies in the complexities of the network of biological mechanisms underlying the ageing process, the different meanings of HA for different populations and contexts, and the ongoing debate on the concept of health. 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 Aligning concepts constitutes a priority action toward shaping policies and optimizing HA with targets set by WHO for 2030. 2

Understanding the theoretical grounds underpinning the operationalizations of HA is a requirement for two reasons: to advance empirical research on clear conceptual dimensions and outcomes across various populations and contexts and to consequently enable the implementation of evidence-based strategies targeting biological, demographic, social, psychological, and behavioural determinants of HA in these settings. 16 , 17 This clarification has not proved to be simple. Several researchers have taken on the task of reviewing and analyzing theoretical models of HA and its associated terms. 18 , 19 , 20 , 21 Chapman for e.g. reflects on six frameworks of ageing well to construct a narrative around self-development that accompany life changes. In contrast, more recent reviews are focused on critiques of existing models, exploring researcher vs. older adult definitions and questions around the feasibility and desirability of HA.

To the best of our knowledge, none of the available reviews takes on the basic task of rethinking the foundations of HA theoretical models and aim to set a clear basis for future studies by combining systematic review and conceptual analysis methods. There is an ensuing need for a homogenized approach in understanding the use of terms referring to HA as well as mapping the characteristics of this concept. We aim to systematically review the literature on HA, analyze the theoretical models and definitions, explore the related normative terms and concepts, and produce a comprehensive thematic overview of what constitutes HA's dimensions, attributes, antecedents, and consequences.

Methods

We conducted a systematic review and conceptual thematic analysis of HA models and definitions following the 24-step guide and PRISMA guidelines. 22 The protocol is registered in PROSPERO under CRD42021238796, which was edited to accommodate the change in the reporting of findings to two separate papers because the results were too broad to be combined in one paper. The current paper includes theoretical models and subsequent efforts will consider operational and empirical models.

Search strategy and selection criteria

A search strategy was developed with the help of experts, adapted and completed across Embase.com , Medline (Ovid), Cochrane CENTRAL, CINAHL, PsycINFO, and Web of Science until August 17, 2022. The main search terms were healthy, successful, active, robust, positive, optimal, well, and productive ageing combined with definition, construct, model, theory, concept, and dimension. The full search strategy per database is available in ( Supplement 1 ).

We included published peer-reviewed original conceptual articles, concept analyses, and reviews only when these proposed new theoretical definitions or models with a normative description of ageing. We excluded empirical models from qualitative and quantitative studies, operational definitions and validations, studies on HA determinants and risk factors, and animal/experimental studies on mechanisms of ageing. Lifespan development and psychology theories that tackle ageing outside the context of HA were excluded. No limit on publication year or language was added.

We also conducted backward search of the reference lists of included papers and a forward citation search for full-text review to identify other eligible articles not identified in our initial search. When original papers of newly identified models from the reviews could not be found or were published in books, we included the earliest reference we could access that allowed the extraction of the model and noted this as a forward search paper.

Results were exported into EndnoteX8 for screening. Two reviewers independently screened the articles by title and abstract. When consensus could not be reached, a third independent reviewer was consulted. The full text was retrieved for all included articles available online and by contacting authors. Similarly, two reviewers screened full texts, and reasons for exclusion were noted. Non-english full-text papers were reviewed by team members who were fluent in the language.

Data analysis

Data were extracted using a predesigned excel sheet piloted with key theoretical papers. Two reviewers independently extracted the main characteristics, including author, title, country, year, term, theoretical framework, definitions, dimensions, descriptions, antecedents, and consequences of the models. Information on empirical support was noted.

We followed an iterative approach to extract the characteristics of the models and thematically analyze them. This process was informed by the strategies of Walker and Avant for analyzing concepts According to this approach, a conceptual model is separated into its different constituting elements: the antecedents or what causes the concept, the consequences or the outcomes of the concept, and the attributes or the describing characteristics of the concept. We developed a glossary for attributes, antecedents, and consequences, and defined dimensions according to the Merriam-Webster dictionary 23 , 24 ( Table 1 ). The models and their descriptions were examined to determine the dimensions, attributes, antecedents, and consequences. To account for the heterogeneity in the use of terms, we adopted a comprehensive approach of overarching domains to specify dimensions (sometimes referred to in the models as constructs or domains), which was also informed by Ebert's multidimensional model and WHO's definition of health. 25 A database of variables for the dimensions determined from the definitions/descriptions and outcomes was then collapsed into summary variables and classified as subjective or objective. The initially extracted variables for attributes, antecedents, and consequences were organized into inventories of descriptions/processes, determinants, and outcomes, respectively. These were then examined for relations and collapsed into broader summary themes, informing the types of the models based on commonalities/differences and dominant approach. Models with more than one approach were classified into more than one type. Because of the conceptual nature of the review, an evaluation of the risk of bias and quality assessment was not fit for purpose. Our rationale stems from the processes of theory and concept development described by Walker and Avant. The evaluation of a concept or theory occurs through its empirical validation in different populations/contexts, sometimes necessitating multiple revisions and validations to become practically useful. Furthermore, and to the best of our knowledge, an objective tool for theory and concept assessment in health outside empirical validation is not available.

Glossary of terms and definitions.

Role of the funding source

No funding was sought for this paper. Marilyne Menassa, Zayne Milena Roa Díaz, and Oche Itodo are GlobalP3HS PhD Fellows whose projects have received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801076, through the SSPH + Global PhD Fellowship Programme in Public Health Sciences (GlobalP3HS) of the Swiss School of Public Health. Marilyne Menassa is also co-funded by the Swiss National Foundation under grant number 189235 for LYRICA (Lifestyle Prevention of Cardiovascular Ageing) project. All authors had full access to the full data in the study and accept responsibility to submit for publication.

Results

The search yielded 24,621 records, from which 10,727 were included for screening by title and abstract after de-duplication. After careful independent screening by two reviewers, 129 papers in English, Spanish, German, Farsi, and Dutch were selected for full-text screening in addition to 14 papers identified from citation tracking. Fifty-nine articles describing 65 models and definitions in English were eligible for inclusion ( Fig. 1 , Supplement 2 ). The papers were published between 1960 and 2022 in 16 countries in Europe, Asia, and America. Most originated from the USA (31), followed by Germany (six), the Netherlands (five), Mexico (three), Sweden (two), and Switzerland (two). Brazil, China, Canada, Italy, Jamaica, New Zealand, Philippines, Spain, Taiwan, and the UK contributed each with one article.

PRISMA flowchart selection of studies .

All 65 models adopted a theoretical approach in defining HA, while some included additional approaches. Twelve (18%) included practical outlook for guiding policy or interventions. 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 Six (9%) were empirically validated with original data, 38 , 39 , 40 , 41 , 42 , 43 and four definitions were derived from one common theory of successful ageing in one article. 44 The authors described the models as theories or concepts which were either original or derived from concept analyses. Eight (12%) described theories that included the concept of successful and active ageing, 38 , 40 , 44 , 45 , 46 , 47 , 48 , 49 three of which were middle-range theories. 44 , 47 , 48 Seven (11%) were generated from concept analyses: three on successful ageing, two on HA, and one on each of active and resilient ageing. 3 , 4 , 5 , 36 , 50 , 51 , 52 All except 20 models (69%) were based on clear theoretical frameworks or were adapted from previous ones. 3 , 4 , 37 , 41 , 43 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64

In the 65 models, there were 12 terms used to refer to or evaluate human ageing: active, adjustment, well, compression of morbidity, graceful, healthy, sustainable, positive, productive, resilient, optimal, and successful. Successful ageing was the most widely used (number of models (%)): (34 (52%)), followed by HA (eight (12%)), ageing well (five (8%)), and active ageing (four (6%)), while the remaining 14 models used the other terms.

Most (62 models (95%)) included alife course approach to ageing, except for three (5%) that focused on specific outcomes in a single period. 41 , 57 , 65 Fifteen (23%) were developed for specific populations or contexts, out of which: five for productive ageing or ageing at work (33%), 29 , 31 , 64 , 66 , 67 three for people who are Lesbian, Gay or Bisexual (20%), 46 , 52 , 65 two for people with HIV (13%), 33 , 35 two for Chinese and/or American populations (13%), 29 , 68 one for people with disabilities (7%), 30 one for Low- Middle Income Countries (LMICs) (7%), 69 and one for child health in HA (7%). 63

Conceptual dimensions : The number of dimensions ranged between two (eight models (12%)), three (11 models (17%)), four (19 models (29%)), five (16 models (25%)), six (five models (8%)), seven (three models (5%)), eight (one model (2%)) and nine (two models (3%)). Among these, the most frequently included dimension (number of models) was cognitive (62), followed by psychological (53), physical (49), social (49), environmental (19), spiritual (16), economic (13), cultural (eight), political (six), and demographic (four). Most models included both objective and subjective dimensions (57), with cognitive dimension common to all except two models and one of the definitions derived from the activity theory, 44 , 66 , 70 while eight models included only objective dimensions having physical and cognitive dimensions in common ( Table 2 , Supplement 3 ).

Characteristics of included studies, models, and definitions.

Antecedents , attributes , and consequences : Our thematic concept analysis of the models yielded a list of antecedents, attributes, and consequences, summarized below ( Fig. 2 , Supplement 4 ):

Healthy ageing concept: antecedents, attributes, and consequences .

Antecedents were classified into 11 themes of which five were intrinsic factors of individuals and six were extrinsic factors of the built and social environments and systems in which individuals age. The models represented dynamic interactions of these two groups of factors impacted by age and changing perceptions of the remaining time along the life course. Intrinsic factors included socio-demography and biography, personal goals and values, psychological and behavioural, physical, and cognitive capacities that define reserves, emotions, personalities, health, and life narratives. Extrinsic factors included the contextual provisions and resources reflected in the built and natural environments, and social, economic, political, health, cultural, and educational systems.

Attributes in the models were described by adaptative and development processes to lifelong changes based on dynamic interactions between individual and contextual life dimensions (physical, cognitive, psychological, environmental, political, cultural, social, economic, spiritual) and systems (macro, meso, micro, and chronosystems). These attributes include: selection, optimization, and compensation, coping, transformative adaptations, self-regulation, and transcendence processes across dimensions.

Consequences are the outcomes of the concept that are determined by the antecedents. They reflect the level of person-environment fit or congruence that is manifested by the individual's health and achieved level of engagement, activity, and/or disengagement across different life dimensions. These consequences manifest themselves as patterns of ageing and behaviour reflected in c oping outcomes , trajectories of ageing, and meaning of life and death. They are the result of interacting objective (physical and cognitive well-being) and subjective (psychological and spiritual well-being) outcomes of health within one's contextual environment across physical, social, economic, and health systems.

Types of HA models and definitions : Based on the identified antecedents, attributes, and consequences describing the processes and outcomes, three types and two subtypes of models and definitions emerged ( Fig. 3 ):

Types of healthy ageing models and definitions .

Type 1 Health outcomes Cognitive, physical, and psychological health are the core elements of these models and lifelong processes are not explicitly described. 41 , 57 , 65

Type 2 Adaptations throughout the lifecourse These models define the concept as a pathway of adaptative, compensatory, and/or development processes to changing environments and goals, and to gains and losses with ageing across multiple life dimensions. Health problems are not explicit (e.g. maintaining cognitive functioning is omitted but can be implied as cognition underlies biological processes etc.). Instead, adaptation processes are central and determine the ageing pattern based on the adjusted life in old age. This adjustment is described through a set of objective and subjective outcomes such as positive functioning and well-being, meaningful relationships and life satisfaction, optimal quality of life and autonomy, productivity, and identity development. 3 , 28 , 31 , 38 , 39 , 40 , 42 , 44 , 45 , 46 , 49 , 50 , 51 , 52 , 53 , 55 , 58 , 61 , 64 , 66 , 67 , 68 , 70 , 72 , 79 , 80 , 81

Sub-type 2.1 Person-environment congruence The person-environment fit, also referred to as congruence or mastery, is an explicit contextual factor of adaptation that depends on life-acquired resourcefulness, resilience, and proactivity. Individuals create adjusted spaces, ecologies, or subcultures to age well in adequate environments. This adequacy involves demographic, social, cultural, political, economic, psychological, cognitive, physical, and spiritual dimensions. 31 , 39 , 40 , 42 , 44 , 45 , 46 , 52 , 53 , 55 , 58 , 64 , 66 , 67 , 68 , 70 , 72 , 80 Most models build on clear theoretical frameworks: activity, disengagement, aged subculture, lifespan development, social construction, value orientations, motivational, and ecology theories. The outcomes describe belonging and ageing well in the environment and development of identity, self-conception, life satisfaction, meaning, spirituality, goal attainment, quality of life, positive functioning, activity, and productivity.

Type 3 Adaptations throughout the life course and health outcomes They combine types 1 and 2 approaches by focusing on adaptation and optimization to strengthen resilience while emphasizing the importance of maintaining health across physical and cognitive dimensions. Subjective well-being is equally highlighted to include life satisfaction, goal attainment, generativity, and engagement across personal, professional, and social dimensions. 4 , 5 , 26 , 27 , 29 , 30 , 32 , 33 , 34 , 35 , 36 , 37 , 43 , 47 , 48 , 54 , 56 , 57 , 59 , 60 , 62 , 63 , 69 , 71 , 73 , 74 , 75 , 76 , 77 , 78 Rowe and Kahn's model constitutes the base for some of these models: proposing additional subjective dimensions such as spirituality and personal ageing experience, 76 adapting it to LMICs by adding an objective environmental dimension, 69 or proposing a comprehensive concept for practical strategies to ageing with disabilities. 30

Sub-type 3.1 Health promotion Health promotion and empowerment activities are central to strengthening resilience and improving reserve capacities to limitations with ageing across cognitive, physical, psychological, cultural, demographic, economic, social, and spiritual dimensions. 4 , 5 , 26 , 27 , 32 , 33 , 34 , 35 , 36 , 37 , 43 , 47 , 59 , 62 , 63 , 69 , 71 , 76 These activities were framed in different ways across models. Some focused on community gerontology through a practical capacity-building framework for health promotion and engagement of older adults to maintain physical, mental, social, and subjective well-being and life satisfaction. 43 Others adopted a practical health behaviour change approach for achieving self-efficacy and personal goals, tackled physical activity and personal predispositions in ageing to achieve subjective well-being, or highlighted the importance of adequate opportunities for optimal health and quality of life to strengthen resilience from childhood. 27 , 59 , 63 Moreover, certain models built on theories of activity, disengagement, gero-transcendence, or the WHO's active ageing framework for health promotion for HA. 5 , 26 , 34

Discussion

Our study shows that HA has been recurrently and heterogeneously conceptualized in the literature, mostly from developed countries. Notwithstanding, there is consensus that HA depends on personal characteristics, resources, goals, and context-specific factors across subjective and objective dimensions: cognitive, physical, psychological, social, environmental, political, cultural, economic, demographic, and spiritual. HA can be defined based on two dominant approaches, separately or combined: (1) through health outcomes across cognitive, physical, social, and psychological dimensions, mainly depicted as the absence of disease and disability at the individual level and compression of morbidity and mortality at the population level, and (2) through developmental adaptation processes of lifelong, dynamic person-environment interactions to changes accompanying ageing across many dimensions. HA can be further defined based on congruence with the environment or health promotion and empowerment.

Similar to our findings, Wahl et al. classified the HA models according to objective and subjective criteria and adaptive mechanisms. 19 Other reviews built the case for broader environmental, developmental, and adaptive approaches, some highlighting disability and the continuous reconstruction of the self given changes with ageing. 20 , 21 , 82 , 83 , 84 More than three decades ago, Ryff compared HA conceptualizations and called for integrating lifespan development theories in psychology to strengthen the theoretical frameworks for each dimension. 83 Based on our review, models from the past 20 years build upon earlier ageing theories, lifespan, and psychological development, mainly as adaptations of previously proposed models to certain contexts and populations.

Our findings indicate that one comprehensive theoretical model or monistic definition of HA, even if desired, is practically challenging. The emphasis on person-context interactions and the constantly evolving understanding of subjective and objective dimensions in HA underlies this impracticality. For instance, certain psychological/behavioural constructs are today widely accepted as sociocultural responses to life events rather than mere biological individual manifestations. 85 This point becomes more evident when our findings are compared with the WHO HA definition. Despite the overall alignment as a lifelong process that builds on intrinsic capacities, functional abilities, and the interaction with the environment, there seems to be an overlap between domains in the WHO definition which requires further distinction to enable harmonization of operationalizations. For instance, the contribution to society that WHO classifies as a functional ability dimension overlaps with the environment dimension, and would be considered a social dimension based on our synthesis. Furthermore, there seems to be an elusiveness and complexity intrinsic to the conceptualization of health, which is foundational to HA. Haverkamp et al. argue that health is a family of diverse “thick concepts” that, although interrelated, cannot be unified under one single concept. 14 In analyzing the social gerontology critiques of Rowe and Kahn's model, Martinson et al. found that when many criteria are used to generate an inclusive model, the result was sometimes a more exclusionary one. 18 Consequently, the conceptual variety and multidimensionality we found can, to a considerable extent, explain the heterogeneity found in HA operational definitions. It renders the comparison of findings, their validity, applicability, and impact on healthcare decisions and policies quite challenging. 6 , 7 , 86

This review provides a rationale for a dimension, population, and context-specific conceptual guidance that can bridge the gap between theory and practice and simplify HA operationalizations across dimensions. Of note are evident gaps for conceptualizations which are specific to gender, disability, and ethnicity as well as to LMICs. Our findings also provide a basis to critically reflect and rethink existing social and healthcare systems, which can enable HA by transitioning from a negative focus on ageism and disease management to a more positive, adaptive, and supportive context-specific approach. This could be achieved, for example, through priority-guided interventions and equitable opportunities for health promotion throughout the life course to strengthen baseline reserve capacities. Furthermore, providing adequate resources and supportive environments for individuals to prosper and age healthily based on their goals, network, and priorities are quintessential contextual factors.

It also is important to note that concept, model, definition, and theory are terms used interchangeably in these conceptualizations. If we intend to develop operational definitions that are theoretically solid and practically useful, there is an inevitable need to differentiate between terms in theorization, research, and practice, and adopt theory development methods to this field. Furthermore, the evolution of conceptualizations of HA is evident in the theories and the choice of terms across models, most of which were used in our search strategy and fulfilled our inclusion criteria for the models they describe. These reflect the contexts and times in which they have been produced and carry a normative approach to the concept of healthy ageing that evolves with the evolution of terms. 20 Disengagement, activity, adjustment, and successful ageing have dominated most of the earlier literature originating from the USA, paralleling a predominant concern of youth unemployment and retirement policies and convey a certain judgment of ageing, which is considerably based on the level of social engagement and productivity. At the turn of the century, active, resilient, and HA became more frequent in Europe. 12 This evolution coincides with the emerging concerns around the demographical changes and the WHO frameworks of active and HA that shifted the discourse to personality, agency, health promotion, and the context in developing resilience and maintaining a person-environment fit. The models as reflected by their describing terms represent an evolution from adaptation only or disease-specific to a more balanced, holistic, and context-specific approach combining objective and subjective dimensions. Furthermore, the more recent HA models and their normative terms seem primarily geared toward public utilitarian functions such as sustainability, productivity, and success. In other words, HA is somehow linked to economic improvement, which is essential. Thus, it's equally important to understand what ageing populations most value in life. Beyond cognitive and physical functional outcomes and self-caring abilities, key aspects of wellbeing such as friendship, socialization, sexuality, and love shall be better understood and integrated into the literature.

The findings of this review should be interpreted in light of certain limitations. We excluded book chapters and non-peer-reviewed articles which means that certain concepts and governmental/non-governmental reports on theoretical frameworks have not been included. Furthermore, our adopted definitions for dimensions and their classifications as subjective or objective might differ from other definitions used in papers and might appear more numerous in terms of dimensions than those initially described by authors. This adaptation was necessary to harmonize the heterogeneity in dimensions for comparative purposes. Furthermore, the conceptual nature of this systematic review might have introduced subjectivity, mitigated in the discussions of the thematic analysis and types with the authors. By describing the process in detail and developing a glossary of terms for dimensions, we aimed to produce a reproducible classification. Finally, our objective was not to conduct a review of mechanisms of ageing or developmental psychology and stage theories across the life course. We also did not aim to do a concept analysis per se to identify all possible empirical referents and model cases of HA as such approaches have been adopted before. Our focus was on the theory rather than the empirical use of HA concepts.

This paper provides a comprehensive overview of theoretical HA models. Our analysis identified the terms, dimensions, characteristics, antecedents, and consequences of HA conceptualizations. Our classification of models is based on the emphasis on health outcomes and/or adaptation processes, also highlighting person-environment congruence or health promotion across the life course approaches. It has become clear that a monistic model or definition of HA cannot practically accommodate the heterogeneous concept across different populations, dimensions, and contexts. Our type classification provides a basis for further harmonizing the use of conceptual terms and dimensions as well as theoretical grounds specific to HA dimensions that can guide research and comparisons of empirical findings. This would inform social and health policies enabling HA for different contexts and populations.

Contributors

OHF and MM contributed to the conceptualization and design of the review protocol and registration. MM, OHF, and BM designed the search strategy run across all databases by BM. MM screened all articles by title, abstract, and full text in pairs with FK, ZMRD, OPE, MG, OI, JB, and FW. OHF and KS supervised MM, who conducted the data extraction, analysis, and classification of models by types, which were iteratively revised and discussed until agreement was reached with OHF and KS. MM wrote the original draft of the manuscript. All authors reviewed, edited, and approved the manuscript.

Data sharing statement

Data inventories are provided as supplementary material.

Declaration of interests

The authors declare no competing interests.

Acknowledgements

No funding was sought for this paper. Marilyne Menassa, Zayne Milena Roa Díaz, and Oche Itodo are GlobalP3HS PhD Fellows whose projects have received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801076, through the SSPH + Global PhD Fellowship Programme in Public Health Sciences (GlobalP3HS) of the Swiss School of Public Health. Marilyne Menassa is also co-funded by the Swiss National Foundation under grant number 189235 for LYRICA (Lifestyle Prevention of Cardiovascular Ageing) project. All authors had full access to the full data in the study and accept responsibility to submit for publication.

Footnotes

Supplementary data

References

Associated Data

Supplementary Materials