Oral Health Determinants of Frailty in Older Adults: An Umbrella Review of Systematic Reviews and Meta-Analyses

Abstract

Oral health is an important yet often overlooked determinant of healthy ageing, with growing evidence linking it to physical frailty in older adults. Findings from systematic reviews and meta-analyses on the associations between oral health indicators and physical frailty among older adults are evaluated and summarised. Systematic reviews, with or without meta-analyses, assessing oral health indicators (exposures) and physical frailty (outcomes) in adults with mean age ≥60 years were included. The methodologies of included reviews were appraised using the Assessing the Methodological Quality of Systematic Reviews (AMSTAR-2) tool. The certainty of evidence associated with meta-analysis outcomes was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. PubMed (MEDLINE), Scopus, Web of Science and Cochrane Library, within the past 10 years, up to July 2025 were used. From the 654 records identified, 12 systematic reviews with 4 meta-analyses were included. All 12 reviews presented data from community-dwelling older adults, and 5 also included the institutionalised, hospitalised and home care populations. Findings suggested that physical frailty was significantly associated with (1) number of teeth, particularly ≤20 teeth and (2) impaired oral functions, including masticatory function assessed using various measures, tongue pressure, oral diadochokinesis and oral frailty. Evidence regarding the association between periodontal disease and physical frailty was insufficient and inconclusive. There is currently limited evidence on oral dryness, oral dysbiosis, oral pain and use/nonuse of denture. This umbrella review identified significant associations between poor oral health indicators and physical frailty, highlighting the importance of oral health in healthy ageing. Considerable heterogeneity in frailty definitions, oral health measures and study designs limits the ability to draw definitive conclusions or infer causality. Findings across systematic reviews are summarised, providing a comprehensive overview of associations between oral health and physical frailty. Research gaps and conflicting results were also identified to guide future studies.

Introduction

As global demographics shift towards an ageing society, the number of older adults aged 60 years and above is projected to rise by 34% worldwide, from 1.0 billion in 2019 to 1.4 billion in 2030.¹ This rapid increase presents significant challenges, highlighting the need for timely and effective preventive strategies to support healthy ageing, by preserving the functional, emotional and overall well-being of the geriatric population.

Among these challenges, frailty has been increasingly recognised as a major public health concern. It is characterised by progressive declines in physical, cognitive and physiological functions, along with decreased physiological reserves. Some of the key features include reduced grip strength, slower gait speed, unintentional weight loss, self-reported exhaustion and low physical activity.² It increases older adults’ vulnerability to adverse health outcomes, such as disability and hospitalisation, imposing great financial and caregiving burden on families, healthcare systems and the society.^3,4 Given the substantial implications, identifying modifiable risk factors of frailty is crucial. One important but yet often overlooked modifiable factor that may contribute to frailty is oral health.^5,6

Poor oral health conditions, such as tooth loss, periodontal disease and impaired oral functions, are prevalent among older adults.^7,8 They do not occur in isolation but have broader multifaceted links with general health. Declining oral conditions can affect individual’s chewing ability, food choices and nutritional intake, which may contribute to the development and progression of frailty through mechanisms such as malnutrition and systemic inflammation.^5,9,10 Furthermore, a recent meta-analysis has suggested a potential association between periodontitis and diabetic retinopathy, indicating shared inflammatory pathways that connect oral and systemic diseases. Such pathways may increase physiological burden and physical vulnerability, further contributing to frailty.¹¹

With growing evidence linking oral health and physical frailty, a number of systematic reviews have examined these associations in older adults. However, the reported findings are heterogeneous, and different reviews focused on distinct, and sometimes overlapping, oral health aspects or indicators. This variation has led to fragmented and inconsistent evidence, posing challenges for clinical translation. Therefore, this umbrella review aims to systematically identify, appraise and synthesise findings from existing systematic reviews and meta-analyses to provide a comprehensive overview of the associations between oral health and physical frailty among older adults, to inform clinical practice and future research.

Methods

An umbrella review approach was adopted to provide a higher-level summary of existing evidence on associations between oral health (as reflected by various indicators) and frailty. It offers a comprehensive perspective by identifying consistencies and discrepancies across reviews, gaps in the literature and assessing the quality of the evidence. This review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guideline.¹² The protocol was registered at the international prospective register of systematic reviews (PROSPERO) (CRD420251103800).

Eligibility criteria

This umbrella review included only systematic reviews (with or without meta-analysis), which were published within the past 10 years, up to 15 July 2025, to ensure the findings reflect contemporary evidence. The study population comprised of adults with mean age of 60 years and above. Systematic reviews investigating on any oral health status or conditions (exposure) such as tooth loss, periodontal disease, oral hygiene or impairments in oral function, were included. The outcomes reported were related to the concept of frailty, including presence of physical frailty, frailty phenotypes, or prefrailty stages.

Case reports, primary research studies, narrative reviews, commentaries, conference proceedings or editorials were excluded. Systematic reviews were excluded if the relationship between oral health and physical frailty was not reported. Articles in language other than English were also excluded.

Study selection and data extraction

A systematic literature search was conducted using SCOPUS, Web of Science (WOS), the Cochrane Library and PubMed databases with the following keyword combination: (‘oral health’ OR ‘dental health’ OR ‘tooth loss’ OR ‘edentulism’ OR ‘periodontal disease’) AND (‘frailty’ OR ‘frailty syndrome’ OR ‘pre-frailty’) AND (‘systematic review’ OR ‘meta-analysis’). The detailed search strategy is presented in Appendix 1.

The search yield from the 4 electronic databases was exported to the Endnote software version 20. After removal of duplicates, 2 reviewers (N.A. and N.L.) independently performed the citation screening, study selection and data extraction, ensuring rigor and objectivity throughout the process. Any disagreements between the reviewers were resolved through discussion with the third reviewer (E.C.). The reference lists of the included studies were also searched for further potentially eligible studies. The data extracted include year of publication, authors, study aim, details of systematic search (databases, date), primary studies included (total number, study design), population, risk of bias and quality assessment, outcome measures and main findings. Where multiple outcomes were reported, only those related to physical frailty were extracted.

Quality assessment of included systematic reviews

The Assessing the Methodological Quality of Systematic Reviews (AMSTAR2) was used by 2 independent reviewers (N.A. and N.L.) to evaluate the methodological quality of included studies in 16 methodological domains, including protocol registration, comprehensive literature search, risk of bias (RoB) assessment and reporting of conflicts of interest. Each domain was rated as ‘Yes’, ‘Partial Yes’ or ‘No’, with overall quality rated as High, Moderate, Low, or Critically Low.¹³

Additionally, 2 reviewers (E.C. and N.L.) independently assessed the certainty of evidence for each meta-analysis outcome, using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) system.¹⁴ The quality of evidence was classified into 1 of the 4 levels: high, moderate, low or very low certainty. Any discrepancies that arise were resolved through discussion with another reviewer (N.A.).

Data synthesis

The characteristics of the included systematic reviews were summarised descriptively. Thematic grouping of the review findings was undertaken to provide a structured overview of the relationship between various aspects of oral health and the outcome of physical frailty. The consistency of findings across included systematic reviews was also examined.

Additionally, meta-analytic findings (where available) were extracted. Forest plot was generated using the R software (version 4.4.2), to illustrate all the reported pooled estimates of associations between different oral health indicators and frailty outcomes. The effect measures, such as odds ratio (OR) and standardised mean difference (SMD), were extracted as reported in the original systematic review, without conversion or standardisation.

The data synthesis approach employed in this study aligns with established methodologies for umbrella reviews, which prioritise qualitative synthesis over numerical standardisation of effect sizes, to enable coherent presentation of complex and heterogeneous evidence.¹⁵

Results

The study selection process is presented in Figure 1. A total of 654 articles were initially identified from 4 electronic databases. After removing duplicates, 593 articles were screened, and following full-text review, 12 systematic reviews published between 2015 and 2024 met the inclusion criteria.16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 Of these, 4 also conducted meta-analyses.^16,17,19,21

Figure 1.

PRISMA flow diagram.

Characteristics of included systematic reviews

Table 1 depicts the characteristics of included systematic reviews (N = 12). All of these reviews presented data from the community settings, and some (n = 5) also included data from institutionalised, hospitalised and home care populations. The number of primary studies included in the reviews ranged from 5 to 68, with total sample sizes ranging from 3086 to 426,538. Data from cross-sectional studies dominated in most systematic reviews.

Table 1.

Summary of the included systematic reviews and meta-analyses studies.

Author(s), year	Study aim	Search strategy (databases, date)	Type of study (total studies and designs)	Population	Main findings related to impact of oral health on frailty	Thematic cluster
Huang et al, 2025	To explore the association between oral health status and frailty in older adults using comprehensive and objective oral health measurement indicators	PubMed, Embase, Cochrane Library and Web of Science (up to 1 July 2024)	Systematic review and meta-analysis (28 studies: 25 cross-sectional and 3 cohort from 2012 to 2024)	Older adults as defined by the original studies (n = 36,634) Settings: not specified	1) Dental-related indicators: Number of teeth, functional dentition (≥21 teeth), no false teeth and ≤20 teeth/with a denture were associated with frailty; (2) Oral hygiene care: Lower brushing frequency was related to frailty; (3) Oral function: The relationship between tongue pressure/decreased tongue pressure, occlusal force/occlusal force reduced ODK hypofunction and poor mixing ability were associated with frailty.	Theme 1: Tooth Loss and Edentulism Theme 2: Impaired oral Function
Zhang et al, 2025	To explore the relationship between tooth count and frailty in older adults	PubMed, Web of Science, Embase and Cochrane Library from their inception to June 2024	Systematic review and meta-analysis (25 studies: 8 longitudinal and 17 cross-sectional from 2012 to 2024	Older adults aged 60 y or above (n = 36,406) Community and hospital settings	Significant negative correlation between the number of teeth and frailty. Specifically, individuals with fewer than 20 teeth exhibited a higher risk of frailty compared to those with more than 20 teeth	Theme 1: Tooth Loss and Edentulism
de Smit et al, 2024	To clarify the predictive value of masticatory function for adverse health outcomes, such as frailty, sarcopenia and malnutrition, in older adults.	PubMed, Embase and CINAHL (year 2000-4 November 2022	Systematic review (34 studies: 5 prospective, 2 retrospective, 27 cross-sectional) *Note: only 8 studies reported frailty outcome	older adults aged ≥65 y from different countries. Settings: Community, institutionalised, hospital, outpatient clinic and convalescent rehabilitation centre. *Note: from the 3 relevant cohort and 5 cross-sectional studies: only community-dwelling older adults (n = 8382, range 135-4720)	Reduced masticatory function (indicated by maximum occlusal force and mixing ability) in older adults is associated with increased incidence of frailty, frailty progression, cognitive decline and mortality.	Theme 2: Impaired oral Function
Hunter et al, 2024	To explore the impact of edentulism on community-dwelling adults across different income levels globally.	MEDLINE, Embase, Scopus, Web of Science, Cochrane CENTRAL (Up to 21 March 2023)	Systematic review (32 studies: 9 cohort, 1 case-cohort, 22 cross-sectional from 2000 to 2022)	Community-dwelling adults from low-, middle- and high-income countries (n = 539 to 237,023)	Edentulism was negatively associated with frailty, mental health, general health, mortality, quality of life and nutrition.	Theme 1: Tooth Loss and Edentulism
Zhu et al, 2024	To determine the prevalence of oral frailty and its unfavourable outcomes among older adults, providing insights for healthcare professionals.	PubMed, Web of Science, Embase, Cochrane Library, Scopus, CINAHL (Up to March 2024)	Systematic review and meta-analysis (28 studies: 21 cross-sectional, 7 cohort from 2016 to 2024)	Older adults (≥60 y) from community, hospital and care home settings in 10 countries (n = 27,927)	Pooled prevalence of oral frailty was 32% (95% CI: 0.24, 0.41). Oral frailty was significantly associated with increased risks of physical frailty (OR = 1.67), malnutrition (OR = 2.27), low dietary variety (OR = 1.98) and social withdrawal (OR = 1.42).	Theme 2: Impaired Oral Function
Dibello et al, 2023	To systematically review oral health indicators related to adverse health outcomes in older adults, including mortality, frailty, disability, quality of life, hospitalisation and falls.	PubMed, MEDLINE, EMBASE, Scopus, Ovid, Google Scholar (Up to 10 September 2022)	Systematic review (68 studies: 50 cohort, 17 cross-sectional, 1 retrospective from 2011 to 2022) *Note: only 22 studies were relevant to physical frailty outcome	Older adults (≥60 y) from multiple countries (n = 426,538) Community, hospital, nursing home settings	Oral health deterioration (eg, number of teeth, periodontal disease) was the strongest predictor of frailty and mortality. Chewing/swallowing difficulties and oral motor function decline were also significantly linked to frailty and disability.	Theme 1: Tooth Loss and Edentulism Theme 3: Periodontal disease and/or other oral conditions
Kojima et al, 2022	To investigate the association between self-reported masticatory dysfunction and frailty among community-dwelling older adults through a systematic review and meta-analysis.	PubMed, CINAHL, AMED (Up to 12 September 2021)	Systematic review and meta-analysis (5 cross-sectional studies from 2012 to 2021)	Community-dwelling older adults (≥60 y) from Japan, China, Hong Kong, Thailand, Mexico (n = 7425)	Significant association between self-reported masticatory dysfunction and frailty (pooled OR = 1.83, 95% CI 1.55-2.18, P < .00001). Masticatory dysfunction may contribute to frailty through malnutrition and muscle weakness.	Theme 2: Impaired oral Function
Sakai et al, 2022	To investigate the association of oral function and dysphagia with frailty and sarcopenia in community-dwelling older adults	PubMed, EMBASE, Ichu-Shi Web, ClinicalTrials.gov and the WHO International Clinical Trials Platform (up to 1 February 2022)	Systematic review (24 studies: 20 cross-sectional, 4 prospective cohort) *Note: only 4 prospective cohort studies from 2018 to 2022 were related to frailty outcome (secondary outcome of the review). Meta-analysis reported was also not relevant (outcome is not frailty).	community-dwelling older adults with a mean age of ≥60 y (N = 17,634) *Note: from the 4 relevant cohort studies, n = 2848, range 97-2011	A prospective association between physical frailty and oral function (occlusal force, tongue pressure, tongue-lip motor function, dysphagia) could not be determined because of the lack of information and the limited number of studies.	Theme 2: Impaired oral Function
Dibello et al, 2021	To summarise studies on the relationship between oral health and frailty among older adults.	PubMed, MEDLINE, Embase, Scopus, Ovid, Google Scholar (Inception-20 March 2021)	Systematic review (39 studies: 26 cross-sectional, 13 longitudinal cohort from 2012 to 2021)	Older adults (>60 y) from various global settings (n = 164,499) Community, hospital, home care settings	Factors of few remaining teeth (29%), were most frequently associated with frailty, followed by poor oral health (15%), an impaired masticatory function (9%) and difficulty chewing (11%), oral diadochokinesis (5%), and occlusal force (7%). Patients with severe periodontitis had a 2·1-times higher risk of 3-y incidence of frailty than those without severe periodontitis	Theme 1: Tooth Loss and Edentulism Theme 2: Impaired oral Function Theme 3: Periodontal disease and/or other oral conditions
Slashcheva et al, 2021	To examine the association between oral health characteristics and frailty status, identifying gaps in translational dental research and clinical applications.	Ovid MEDLINE, Embase, Cochrane Central, Cochrane Database of Systematic Reviews, Scopus (Up to 2 January 2020)	Systematic review (26 studies: 17 cross-sectional, 9 longitudinal from 2012 to 2019)	Older adults (≥50 y) (n = 21,195, range 141-3635) Community setting	Significant association between frailty and oral health, with frail individuals more likely to have fewer teeth, worse chewing ability, higher prevalence of periodontitis and lower dental service utilisation.	Theme 1: Tooth Loss and Edentulism Theme 2: Impaired oral Function) Theme 3: Periodontal disease and/or other oral conditions
Hakeem et al, 2019	To systematically review longitudinal studies on the association between oral health and frailty.	MEDLINE, EMBASE, LILACS, OpenGrey, Google Scholar (Up to July 2018)	Systematic review (5 longitudinal studies from 2017 to 2018)	Older adults (≥60 y) from Mexico, Japan and the UK (n = 3086). Community-dwelling	Significant associations between number of teeth (2 studies), oral function (2 studies), accumulation of oral health problems, and dry mouth symptoms with frailty incidence, whereas periodontal disease shown inconsistent association	Theme 1: Tooth Loss and Edentulism Theme 2: Impaired oral Function Theme 3: Periodontal disease and/or other oral conditions
Tôrres et al, 2015	To assess the relationship between frailty or its components and poor oral health through a systematic review.	PubMed, EMBASE, Cochrane, LILACS, SciELO (1991-July 2013)	Systematic review (12 studies: 7 cross-sectional, 5 longitudinal from 2000 to 2013)	Older adults (≥60 y) from various countries (n = 7884, range from 193 to 1374) Community and institutionalised settings, some studies did not specify study settings.	Number of teeth, periodontal disease and dental prosthesis need were significantly associated with frailty. Periodontal disease was associated with weight loss over time. Self-reported poor oral health and lack of dental service use were linked to increased frailty risk.	Theme 1: Tooth Loss and Edentulism Theme 2: Impaired oral Function Theme 3: Periodontal disease and/or other oral conditions

Different risk of bias tools were used to assess the quality of the included studies: the Newcastle-Ottawa Scale (NOS) (n = 5), National Institutes of Health (NIH) Quality Assessment Toolkits (n = 4), Joanna Briggs Institute (JBI) critical appraisal tool (n = 2), QualSyst Tool Standardised Checklist for Quantitative Studies (n = 1), Agency for Healthcare Research and Quality (AHRQ) checklist for cross-sectional studies (n = 1), Appraisal tool for Cross-Sectional Studies (AXIS) (n = 1) and a custom critical appraisal checklist (n = 1). Three reviews reported using the NOS in combination with one other tool to cater for different study designs (Table 2).

Table 2.

Quality appraisal of systematic reviews using AMSTAR 2.

Author(s), year	PICO components	Protocol registered/justified deviations	Study design selection explanation	Comprehensive literature search	Duplicate study selection	Duplicate data extraction	Excluded studies justified	Detailed study descriptions	Risk of bias (RoB) assessment	Funding sources reported	Meta-analysis methods (if conducted)	RoB in meta-analysis (if conducted)	RoB Interpretation/ discussion	Heterogeneity explanation (if observed)	Investigation of publication bias	Conflict of interest reported	Quality rating
(Huang et al, 2025)	Yes	Yes (Registered in PROSPERO: CRD42024587687	Yes	Yes (Searched on PubMed, Embase, Cochrane Library and Web of Science)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used Joanna Briggs Institute critical appraisal tool and the Newcastle-Ottawa Scale)	Yes (Natural Science Foundation of Hubei Province (No. 2023AFB763).	Yes (random-effects, sensitivity analysis, heterogeneity)	Yes	Yes	Yes	No	Yes (No competing interests declared)	Low
(Zhang et al, 2025)	Yes	Yes (Registered in PROSPERO: CRD42023431843.	Yes	Yes (Searched on PubMed, Web of Science, Embase and Cochrane Library)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used Newcastle-Ottawa Scale)	Yes (Shenzhen’s Sanming Project of China (Grant number: SZSM202111014), Peking Union Medical College Hospital Research Fund (Grant number: ZC201900516) and the First Affiliated Hospital of Shenzhen University Research Fund (Grant number: 20223357018).	Yes (random-effects model, subgroup sensitivity analyses and heterogeneity	Yes	Yes	Yes	Yes	Yes (No competing interests declared)	High
de Smit et al, 2024	Yes	No	No (No PROSPERO registration mentioned)	Yes (Searched on PubMed, Embase and CINAHL)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions	Yes	Yes (Used NIH Quality Assessment Tool	Yes (Author’s institutions)	No meta-analysis conducted	No meta-analysis conducted	Yes	No	No	Yes (No competing interests declared)	Low
(Hunter et al, 2024)	Yes	Yes (Registered in PROSPERO: CRD42022320049)	Yes	Yes (Searched on MEDLINE, EMBASE, Scopus, Web of Science, Cochrane CENTRAL)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used QualSyst Tool for quality assessment)	Yes (Wellcome Trust funding acknowledged)	No meta-analysis conducted	No meta-analysis conducted	Yes	Yes	No	Yes (No competing interests declared)	Low
(Zhu et al, 2024)	Yes	Yes (Registered in PROSPERO: CRD42024537884)	Yes	Yes (Searched on PubMed, Web of Science, Embase, Cochrane Library, Scopus and CINAHL)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used Agency for Healthcare Research and Quality (AHRQ) for cross-sectional studies and Newcastle-Ottawa Scale for cohort studies)	Yes (Guangxi Medical and Health Appropriate Technology Development Project, Nanning Scientific Research and Technology Development Plan)	Yes (Random-effects model, subgroup analysis, sensitivity analysis, publication bias assessment using funnel plot and Egger’s test)	Yes	Yes	Yes	Yes	Yes (No competing interests declared)	High
(Dibello et al, 2023)	Yes	Yes (Registered in PROSPERO: CRD42021241075)	Yes	Yes (Searched on PubMed, MEDLINE, EMBASE, Scopus, Ovid, Google Scholar)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used National Institutes of Health Quality Assessment Toolkits)	Yes	No meta-analysis conducted	No meta-analysis conducted	Yes	No	No	Yes	Low
(Kojima et al, 2022)	Yes	Yes (Registered in PROSPERO: CRD42021277173)	Yes	Yes (Searched on PubMed, CINAHL, AMED)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used Joanna Briggs Institute Critical Appraisal Checklist)	No (Authors reported no specific funding)	Yes (Fixed-effects model for meta-analysis, publication bias assessed with funnel plot)	Yes	Yes	Yes	Yes	Yes (No competing interests declared)	High
(Sakai et al, 2022)	Yes	Yes (Registered in PROSPERO:CRD42019141277)	Yes	Yes (searched on PubMed, EMBASE, Ichu-Shi Web, ClinicalTrials.gov and the World Health Organisation International Clinical Trials Platform Search Portal	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (used NIH Quality Assessment Tool)	Yes (Japan Society for the Promotion of Science KAK- ENHI (grant number 21K21051) and a grant from the Dental Research Center)	No meta-analysis conducted	No meta-analysis conducted	Yes	Yes	No	Yes (No competing interests declared)	Low
(Dibello et al, 2021)	Yes	Yes (Registered in PROSPERO: CRD42021231450)	Yes	Yes (Searched on PubMed, MEDLINE, Embase, Scopus, Ovid, Google Scholar)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used NIH quality assessment toolkit)	No	No meta-analysis conducted	No meta-analysis conducted	Yes	No	No	Yes (No competing interests declared)	Low
(Slashcheva et al, 2021)	Yes	Yes (Registered in PROSPERO: CRD420160295)	Yes	Yes (Searched on Ovid MEDLINE, Embase, Cochrane CENTRAL, Cochrane Database of Systematic Reviews and Scopus)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used Newcastle-Ottawa Scale for longitudinal studies, Appraisal tool for Cross-Sectional Studies (AXIS))	Yes (American Dental Association National Elder Care Advisory Committee)	No meta-analysis conducted	No meta-analysis conducted	Yes	No	No	Yes (No competing interests declared)	Low
(Hakeem et al, 2019)	Yes	Yes (Registered in PROSPERO: CRD42018094004)	Yes	Yes (Searched on MEDLINE, EMBASE, LILACS, OpenGrey and Google Scholar)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used Newcastle-Ottawa Quality Assessment Scale)	No	No meta-analysis conducted	No meta-analysis conducted	Yes	Yes	No	Yes (No competing interests declared)	Low
(Tôrres et al, 2015)	Yes	No (No PROSPERO registration mentioned)	Yes	Yes (Searched on PubMed, EMBASE, Cochrane, LILACS and SciELO)	Yes	Yes	Partial Yes (List provided but no justification for all exclusions)	Yes	Yes (Used checklist based on previous methodological assessments)	No	No meta-analysis conducted	No meta-analysis conducted	Yes	No	No	Yes (No competing interests declared)	Low

A wide range of oral health conditions was assessed using different measures, such as number of teeth, edentulous status, masticatory function, maximum occlusal force, oral frailty, periodontal disease status, oral dryness and oral pain. In the present umbrella review, these indicators were broadly categorised into 3 domains: structural (tooth loss and edentulism), functional (impaired oral functions) and disease-related (periodontitis and/or other oral conditions). These thematic domains were established by grouping oral health indicators with similar characteristics through synthesis of the included reviews, and were refined through reviewer discussions to ensure appropriate and consistent classification across domains.

Different frailty outcome measures were considered in the included reviews, with Fried frailty criteria stated as being the most commonly used. Other frailty assessments employed include the Frailty index, Kihon Checklist, Physical Frailty Phenotype, Clinical Frailty Scale and others.

Figure 2 presents the pooled estimates extracted from 4 systematic reviews with meta-analyses,^16,17,19,21 depicting the magnitude of associations between different oral health indicators and frailty outcomes. Three reported using random-effects model^16,17,19 and one²¹ using fixed-effects model for the meta-analyses. Heterogeneity was assessed with I² statistics. No assessment of certainty in evidence was reported in these 4 systematic reviews with meta-analyses. However, 3 other systematic reviews without meta-analysis^20,22,26 reported using the GRADE rating system to assess the overall quality of evidence of the included studies.

Figure 2.

Summary of the pooled odds ratio of the associations between oral conditions and physical frailty.

Content-based thematic clustering

Theme 1: Tooth loss and edentulism

Most of the included systematic reviews (n = 8) reported findings related to this theme.16, 17, 18^,20,22, 23, 24, 25 The indicators assessed, in order of frequency, included number of teeth, presence of functional dentition (defined as ≥21 teeth, ≥20 teeth, or ≥9 occluding pairs), edentulism, use/nonuse of denture and need for denture.

Having more teeth was negatively associated with physical frailty among older adults based on consistent evidence^20,22, 23, 24, 25 with reviews reporting a pooled OR of 0.98 (95% CI: 0.97-0.99, 10,005 participants, 9 studies, I² = 67%, low-certainty evidence)¹⁷ and SMD of −0.59 (95% CI: −0.77 to −0.41, 8988 participants, 7 studies, I² = 85%, low-certainty evidence).¹⁶ Specifically, having a functional dentition of ≥21 teeth was associated with 76% lower odds of frailty (pooled OR = 0.24, 95% CI: 0.16-0.34, 9970 participants, 6 studies I² = 71%, low-certainty evidence),¹⁶ whereas having ≤20 teeth was associated with 99% increase in odds of frailty (pooled OR = 1.99, 95% CI: 1.57-2.53, participants = 26,004, 17 studies, I² = 89%, moderate-certainty evidence).¹⁷ The significant association between having ≤20 teeth and frailty remained consistent across subgroup analyses by country, setting, study design, frailty definition and model type (crude/adjusted).¹⁷ Additionally, being edentulous was reported to have a strong association with an increased chance of physical frailty.^18,20,22,23

In terms of denture use among partially edentulous older adults, the pooled results from 2 primary studies^28,29 showed that compared to those with ≥21 teeth, both ‘having ≤20 teeth and with denture’ (pooled OR = 2.32, 95% CI: 1.70-3.16, 5192 participants, 2 studies, I² = 0%, moderate-certainty evidence)¹⁶ and ‘having ≤20 teeth without denture’ (no pooled estimates available) were more likely to be frail. Meanwhile, among edentulous older adults, nonuse of dentures was associated with a higher probability of physical frailty compared to those with dentures.²³ Further, the need for insertion or replacement of dentures (based on normative assessed need) were associated with increased likelihood of being frail, when compared to those without such need, based on 2 systematic reviews^23,25 which synthesise findings from the same primary study.³⁰

Theme 2: Impaired oral function

Nine systematic reviews^16,20, 21, 22, 23, 24^,26,27 summarised findings related to oral functions, including mastication, swallowing, articulatory oral motor skills, tongue pressure and oral moisture, many of which are key parameters used to define oral frailty across the various available operational definitions.31, 32, 33, 34 While some reviews examined the association between physical frailty and individual oral functions, others assessed these oral functions collectively as ‘oral frailty’.

Masticatory function is among the most studied oral functions in relation to physical frailty, followed by oral frailty as a composite measure. Masticatory function was assessed using different measures, such as mixing ability test, glucose extraction test, maximum occlusal force and self-reported chewing ability. Across the included reviews, poor masticatory function, either assessed objectively or subjectively, showed consistent and strong association with frailty,20, 21, 22, 23^,27 except in one review,¹⁶ which found no significant difference between frail and nonfrail group when pooling results from 2 studies reporting glucose extraction from chewing gummy jelly (SMD = −0.44, 95% CI: −1.03 to 0.14, P = .138, 437 participants, I² = 55.5%, very low certainty). From pooled cross-sectional evidence involving 7425 community-dwelling older adults, it was concluded that those with self-reported masticatory dysfunction were 1.83 times as likely (95% CI: 1.55-2.18, 4 studies, I² = 0%, low-certainty evidence) to be frail than those without.²¹ On a similar note, Huang et al¹⁶ reported that older adults exhibiting poor mixing ability were more than twice as likely to be frail (pooled OR = 2.30, 95% CI: 1.75-3.57, 1765 participants, 2 studies, I² = 25.5%, moderate-certainty evidence). Additionally, a strong association between reduced occlusal force and physical frailty was consistently observed,^20,23,24,27 with one systematic review reporting a pooled OR of 1.85 (95% CI: 1.21-2.82, 1421 participants, 2 studies, I² = 0%, very low-certainty).¹⁶

On the other hand, reviews investigating the relationship between oral dryness and physical frailty reported inconsistent findings based on a small number of primary studies.^20,22, 23, 24 Huang et al¹⁶ conducted separate meta-analyses, one pooling result from 2 studies measuring oral dryness (oral moisture value < 27) (OR = 1.14, 95% CI: 0.59 to 2.19, I² = 65%, P = .693) and another pooling 3 studies assessing oral moisture (SMD = 0.21, 95% CI: −0.07 to 0.49, I² = 0%, P = .136), and reported no statistically significant associations with frailty (very low-certainty evidence).

Other oral functions, such as swallowing function (assessed through questionnaire), have also been shown to exhibit a significant inverse association with physical frailty^20,23 Similarly, tongue pressure, an objective indicator closely related to swallowing, was significantly associated with frailty,^16,20,23 wherein older adults with decreased tongue pressure had 1.62 times the odds (95% CI: 1.12-2.35, 1805 participants, 2 studies, I² = 0%, very low-certainty evidence) of being frail compared to those with normal tongue pressure.¹⁶ Besides that, hypofunction in oral diadochokinesis, which reflects articulatory oral motor skills, was associated with a higher likelihood of being frail.^20,22 The same finding was summarised in meta-analysis with pooled OR of 1.88 (95% CI: 1.33-2.64, 2915 participants, 4 studies, I² = 0%, very low-certainty evidence).¹⁶ Of note, in many studies, these oral functional measures came from the operational definition of oral frailty. Although various criteria were used to define oral frailty in the original studies, the current evidence consistently indicates a positive association between oral frailty and physical frailty.^19,22, 23, 24 A meta-analysis pooling data from 18 studies reported a combined OR of 1.67 (95% CI: 1.38-2.02, participants = 21,185, I² = 77%, low-certainty),¹⁹ suggesting that older adults with oral frailty are significantly more likely to be physically frail than those without oral frailty.

Theme 3: Periodontal disease and other oral conditions

Five systematic reviews without meta-analysis investigated the relationship between periodontal diseases and physical frailty.^20,22, 23, 24, 25 However, their findings were derived from only 1 to 4 primary studies, with considerable overlap among the original sources.35, 36, 37, 38 The association between periodontal disease and physical frailty remains inconclusive due to the discrepant findings across studies. Some evidence suggests the presence of a significant association, whereas other studies did not find any.

The included systematic reviews also identified a few studies on oral pain^20,22,23 and oral dysbiosis.²² However, the evidence was deemed to be of very low quality and their associations with physical frailty remain uncertain.

Quality assessment of included systematic reviews and meta-analyses

The AMSTAR 2 assessment revealed that 3 of the included reviews were of high confidence quality, and 9 were of low quality due to unmet critical domains, such as publication bias and protocol registration (Table 2). The high-quality reviews exhibited rigorous methodology, including preregistered protocols (such as PROSPERO registration), comprehensive literature searches, duplicate data extraction and detailed RoB assessments. Additionally, these reviews provided transparent evaluations of heterogeneity and clearly described the methods of meta-analyses.

The quality of evidence associated with the 14 meta-analysis outcomes was evaluated using the GRADE approach (Appendix 2). The overall ratings ranged from very low to moderate certainty. They were mainly downgraded due to risk of bias, inconsistency due to heterogeneity and imprecision in the effect size.

Discussion

This umbrella review summarised findings from 12 systematic reviews (with 4 meta-analyses) on the associations between oral health and physical frailty among older adults, using both qualitative and semi-quantitative analysis approach. Various oral health indicators associated with physical frailty were identified and systematically categorised into 3 overarching domains: (1) structural (tooth loss and edentulism), (2) functional (impaired oral function) and (3) disease-related (periodontal disease and/or other oral conditions).

Among all indicators across the domains, number of teeth, particularly having ≤20 teeth or edentulous, demonstrated the most consistent association with physical frailty, followed by deterioration in masticatory function and the composite measure of oral frailty. It is noteworthy that many studies used a 20-teeth cutoff, in line with the concept of functional dentition.39, 40, 41 Other oral functional measures, including decreased tongue pressure, impaired swallowing function and oral diadochokinesis hypofunction, were also associated with physical frailty, although the current evidence supporting these associations is considered weak. Of note, these measures were more commonly assessed as components of the composite ‘oral frailty’ construct rather than as independent factors. Meanwhile, evidence on the association of physical frailty with periodontal disease as well as oral dryness remains limited and inconclusive. For conditions such as oral pain, oral dysbiosis and use/nonuse of dentures, the available evidence is even more scarce, insufficient to credit or discredit any associations.

Several mechanisms have been postulated to explain the relationship between oral health and physical frailty, with one key pathway being nutrition.² Compromised dentition due to tooth loss can significantly impair masticatory function and occlusal force,⁴² limiting one’s ability to chew a variety of foods.⁴³ This often results in poor diet quality as meal preference shift towards softer food options and beverages.⁴⁴ Older adults with poor oral health have been shown to consume fewer nutrient-dense foods such as fruits and vegetables, and more processed, high-fat foods.^24,45 These dietary changes can lead to nutritional deficiencies which play a key role in the pathogenesis of frailty, as described in the Fried’s Frailty Cycle.^2,46

With ageing, older adults may experience diminished oral functions, such as swallowing difficulty, reduced tongue pressure and a decline in oral motor skills. These changes are frequently associated with underlying health conditions, and they often occur simultaneously rather than in isolation, leading to the emergence of the concept of oral frailty.³¹ It represents 1 of the 4 stages of oral function: healthy state, oral frailty, oral hypofunction and oral dysfunction,⁴⁷ and is defined as age-related gradual loss of oral function accompanied by a decreased interest in oral health, and decline in cognitive and physical functions.^48,49 These oral functional deteriorations can affect oral intake and has been identified as an important risk factor for malnutrition and sarcopenia,⁴⁸ linking oral health to physical frailty through the nutritional pathway.

Besides, decline in oral health may affect social participation, particularly in activities involving eating or speaking, leading to social isolation and poorer quality of life.⁵⁰ This social withdrawal can increase the risk of loneliness and late-life depression, which are associated to greater odds of physical frailty.51, 52, 53 In addition to this psychosocial pathway, a neurological mechanism has also been proposed, whereby dental occlusion and proprioception from the periodontal ligament contribute to body balance and coordination. Jaw position and poor dental occlusion may affect proprioceptive feedback, disrupting head posture stability and overall balance, potentially leading to fall, which is associated with frailty in older adults.^54,55

A relationship between inflammation and frailty has also been suggested,⁵⁶ potentially due to the effects of inflammatory cytokines on muscle tissue.^57,58 While periodontal disease is characterised by chronic inflammation, current evidence regarding its association with frailty remains inconclusive. Based on the included systematic reviews, only a limited number of original studies^24,36,38 investigated this relationship. Hakeem et al³⁵ revealed that periodontal disease has a weaker association with frailty compared to number of teeth. Among the 2 longitudinal studies identified, one found that severe periodontitis at baseline was associated with an increased risk of developing frailty over a 3-year follow-up, even after adjusting for confounders.³⁶ In contrast, the other study with a similar follow-up duration found no significant association between periodontal disease and frailty incidence.³⁸ The inconsistency in findings may be due to variations in periodontal assessment methods, such as pocket depth, clinical attachment loss, or number of teeth examined. It is also possible that the positive associations observed are mediated by tooth loss resulting from periodontitis. Further studies are needed to better understand the relationship.

Another notable finding is regarding denture use. While prosthetic rehabilitation is a common approach to restore function following tooth loss, the association between denture use and frailty is unclear. When compared to older adults with ≥21 teeth, having fewer teeth (≤20 teeth) has been associated with higher odds of frailty among nondenture wearers, but not among denture-wearers, suggesting a potential mitigating effect of denture use.^16,29 Nevertheless, others have found that this association between tooth loss and frailty persists regardless of denture use.^16,29 It is possible that the discrepancies in findings were due to confounding factors such as denture quality and fit, type of dentures and chewing ability. Further research is needed to investigate the role and effect of denture use.

Implications, strengths and limitations

Overall, the evidence supports a significant inter-relationship between oral health status and physical frailty in older adults, particularly in relation to the number of teeth and masticatory function. Although the causal pathways have yet to be firmly established, the findings summarised in this review underscore the importance of maintaining adequate dentition and good oral function. In line with this, the European College of Gerodontology and the European Geriatric Medicine Society⁷ have recommended providing oral health training to nondental healthcare providers and integrating oral health examination into routine geriatric assessments at the primary care level to emphasise prevention and early detection of oral health problems, as well as interdisciplinary care. Oral health deterioration may potentially serve as an early marker of physical frailty or vice versa. Most oral diseases and conditions are preventable and treatable, timely intervention is therefore crucial, especially before functional limitations, refusal behaviours, or diminished capacity to receive dental care arise.

The mixed findings on the association between periodontal disease and physical frailty indicate the need for further research to better understand the role of chronic oral inflammation, if any, in the development and progression of frailty in older adults. Future research may also explore interventions to improve oral functions among older adults and examine their impact on frailty outcomes. Although temporal associations between oral health deterioration and physical frailty have been observed in some studies, the potential influence of confounders and role of socioeconomic inequalities in contributing to both conditions warrant further investigation. Socioeconomic disadvantage may be a shared underlying determinant of poor oral health and frailty. Equitable health policies are therefore essential to address barriers to dental care access.

When interpreting the findings, it is important to note that although the electronic search was restricted to the past 10 years to ensure relevance and timeliness, the included systematic reviews contain primary studies published before this period. Additionally, this umbrella review does not attempt to standardise the different effect measures (odds ratio and standardised mean difference) reported in the 4 included meta-analyses, as the conversions could introduce methodological and statistical bias, leading to misrepresentation of the true effect.

This umbrella review has several notable strengths, including its methodological transparency and reproducibility. The systematic compilation and comparative synthesis of evidence from existing systematic reviews and meta-analyses provide a comprehensive and structured overview of the relationship between various oral health indicators and physical frailty. To the best of our knowledge, this is the first umbrella review in this topic of growing interest.

There are several limitations to this review. The inherent heterogeneity across included systematic reviews regarding definitions and measurements of oral health and frailty, as well as the use of different quality appraisal tools, limits the comparability of findings and ability to draw strong conclusions. Standardisation in future research is necessary to enhance the value and utility of findings. The synthesis in this umbrella review may be constrained by the methodological limitations and reporting of the included systematic reviews; where data were missing or incomplete, the individual primary studies were referred to. Additionally, the predominance of cross-sectional data within the reviewed literatures restricts the ability to infer causal relationships, highlighting the need for more rigorous study designs, such as cohort studies, to establish temporality and generate more robust evidence.

Some primary studies overlapped across the included systematic reviews, potentially influencing their conclusions. However, unlike meta-analyses, umbrella reviews do not pool effect sizes from the included reviews, reducing the possibility of duplication bias.⁵⁹ Differences in eligibility criteria and analytical methods in the reviews also led to distinct interpretations. Given that umbrella reviews focus on capturing the full spectrum of evidence on a topic and comparing the findings, the inclusion of all relevant systematic reviews, even with some overlapping primary studies, ensures a more comprehensive synthesis.⁶⁰ In contrast, selective exclusion may inadvertently omit crucial findings.

Conclusion

This umbrella review synthesised current evidence on the associations between oral health indicators and physical frailty, highlighting the importance of oral health in healthy ageing. While number of teeth and impaired masticatory function showed more consistent associations, the evidence regarding periodontal disease and other oral functional indicators remains limited. The considerable heterogeneity in frailty definitions, oral health measures and study designs limit the ability to draw definitive conclusions or infer causality. Further research is needed to elucidate potential causal pathways and mediators linking oral health and frailty to strengthen the evidence base.

Author contributions

Najihah Lokman: Conceptualisation, Funding acquisition, Investigation, Methodology, Software, Formal analysis, Writing – original draft, Writing – review and editing. Noor Aziella Mohd Nayan: Conceptualisation, Investigation, Methodology, Formal analysis Writing – original draft. Elaine Kar Man Choong: Investigation, Formal analysis, Validation, Writing – original draft, Writing – review and editing.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the author(s) used ChatGPT in order to improve readability and language. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Conflict of interest

None disclosed.