jfa journal

AND option

OR option



M. Ong1, K. Pek1, C.N. Tan1, J. Chew1,2, J.P. Lim1,2, S. Yew1, A. Yeo1, W.S. Lim1,2


1. Institute of Geriatrics and Active Ageing, Tan Tock Seng Hospital, Singapore; 2. Department of Geriatric Medicine, Tan Tock Seng Hospital (TTSH), Singapore

Corresponding Author: Ms. Melissa Ong, TTSH Annex 2, Level 3, 11 Jalan Tan Tock Seng, Institute of Geriatrics and Active Ageing, Tan Tock Seng Hospital, Singapore 308433, Telephone: +65 6359 6327, Email: Melissa_HM_ONG@ttsh.com.sg

J Frailty Aging 2021;in press
Published online October 26, 2021, http://dx.doi.org/10.14283/jfa.2021.43



Background: Despite emerging evidence about the association between social frailty and cognitive impairment, little is known about the role of executive function in this interplay, and whether the co-existence of social frailty and cognitive impairment predisposes to adverse health outcomes in healthy community-dwelling older adults.
Objectives: We aim to examine independent associations between social frailty with the MMSE and FAB, and to determine if having both social frailty and cognitive impairment is associated with worse health outcomes than either or neither condition.
Methods: We studied 229 cognitively intact and functionally independent community-dwelling older adults (mean age= 67.2±7.43). Outcome measures comprise physical activity; physical performance and frailty; geriatric syndromes; life space and quality of life. We compared Chinese Mini Mental State Examination (CMMSE) and Chinese Frontal Assessment Battery (FAB) scores across the socially non-frail, socially pre-frail and socially frail. Participants were further recategorized into three subgroups (neither, either or both) based on presence of social frailty and cognitive impairment. Cognitive impairment was defined as a score below the educational adjusted cut-offs in either CMMSE or FAB. We performed logistic regression adjusted for significant covariates and mood to examine association with outcomes across the three subgroups.
Results: Compared with CMMSE, Chinese FAB scores significantly decreased across the social frailty spectrum (p<0.001), suggesting strong association between executive function with social frailty. We derived three subgroups relative to relationship with socially frailty and executive dysfunction: (i) Neither, N=140(61.1%), (ii) Either, N=79(34.5%), and (iii) Both, N=10(4.4%). Compared with neither or either subgroups, having both social frailty and executive dysfunction was associated with anorexia (OR=4.79, 95% CI= 1.04-22.02), near falls and falls (OR= 5.23, 95% CI= 1.10-24.90), lower life-space mobility (odds ratio, OR=9.80, 95% CI=2.07-46.31) and poorer quality of life (OR= 13.2, 95% CI= 2.38-73.4).
Conclusion: Our results explicated the association of executive dysfunction with social frailty, and their synergistic relationship independent of mood with geriatric syndromes, decreased life space and poorer quality of life. In light of the current COVID-19 pandemic, the association between social frailty and executive dysfunction merits further study as a possible target for early intervention in relatively healthy older adults.

Key words: Social frailty, cognitive performance, executive dysfunction, Frontal Assessment Battery, older adults.



Frailty refers to a geriatric syndrome whereby there is increased vulnerability to adverse outcomes after a stressor event due to diminishing homeostatic reserves, leading to increased risk of functional decline, dependency and/or mortality (1, 2). A holistic approach targeted at addressing the multi-dimensional determinants of frailty is needed to prevent and reverse frailty in older adults (1, 2). Amongst these dimensions, social frailty is the least understood, but has gained traction over time for its importance in contributing to the trajectory of frailty in older adults (1-3). Based on Bunt’s conceptual framework premised upon the Theory of Social Production Function, social frailty is defined as a continuum of being at risk of losing, or having lost, social resources, general resources and social activities or abilities that are important for fulfilling one or more basic social needs during their lifetime (3). Previous studies have reported associations between social frailty with increased risk of disability, depressive symptoms, malnutrition, lower physical activity and performance, and cognitive dysfunction amongst community-dwelling older adults (3-9). Social frailty is particularly germane in light of the ongoing COVID-19 pandemic, with emerging evidence that pandemic control measures can exacerbate social frailty with concomitant impact on mood and lifestyle activity in relatively healthy older adults (10).
There is a growing body of evidence which substantiates the relationship between social frailty and cognition (6, 11-13). A previous study of community-dwelling older adults in China reported a high prevalence of social frailty amongst participants who had dementia, subjective memory decline, and cognitive impairment as measured by the Mini-Mental State Examination (MMSE) (13). This was corroborated by a Japanese study which showed an independent association between those socially pre-frail and frail presenting with deficits in at least two tests in a neuropsychological battery assessing cognitive domains of memory, attention, executive function and processing speed (6). In tandem with these findings, depression is a key determinant often associated with social frailty and cognitive impairment in older adults (5, 13, 14).
However, gaps remain in our understanding of the relationship between social frailty and cognition. The majority of earlier studies did not adopt a theory-grounded definition of social frailty (5-8, 13). There exists uncertainty about which cognitive domain is associated with social frailty, with cognitive impairment largely defined by general memory-based evaluations such as the MMSE or loosely based on the observation of poor scores in components of a neuropsychological battery (6, 13). With recent evidence suggesting that declines in executive function predicted onset of physical frailty and preceded declines in memory-biased domains in relatively healthy older adults (15, 16), it will be important to ascertain the relative contribution of executive function vis-à-vis amnestic domains. Specifically, the Frontal Assessment Battery (FAB) assesses executive functioning and is able to discern early cognitive impairment in older adults (17-19). Furthermore, the combined effect of both social frailty and cognitive impairment on daily activities and overall health in older adults remains largely unexplored. The confounding effect of depressive symptoms on social frailty and cognition is often unaccounted for in earlier studies.
This provided the impetus for the current study to examine the relationship between social frailty and cognition in a representative cohort of relatively healthy community-dwelling older adults. We aim to: (i) examine independent associations of social frailty with the MMSE and FAB; (ii) describe the prevalence of subgroups (neither, either or both) of social frailty and cognitive impairment; and (iii) determine if physical activity, physical frailty and performance, geriatric syndromes, life space mobility and quality of life are more adversely affected in ‘either’ or ‘both’ groups compared with ‘neither’. Better understanding of the relationship between social frailty and cognition will shed light on whether community screening programs should aim to detect both conditions as opposed to either alone.



Study Population

The “Longitudinal Assessment of Biomarkers for characterization of early Sarcopenia and Osteosarcopenic Obesity in predicting frailty and functional decline in community-dwelling Asian older adults Study” (GeriLABS-2) is a prospective cohort study involving functionally independent community-dwelling adults aged 50 and older. Participants independent in both basic and instrumental activities of daily living (ADL) and non-frail as defined by the FRAIL criteria (FRAIL ≤3)(20) were recruited from December 2017 to March 2019. Participants were excluded if they had prior diagnosis of dementia; scored ≤21 on the modified Chinese Mini-Mental State Examination (CMMSE)(21); were unable to walk 8m independently; or resided in a long-term institutional care facility. The study was protocol was approved by the Domain Specific Review Board of the National Health Group. Written consent was obtained from all participants prior to study participation.

8-item Social Frailty Scale (SFS-8)

Social frailty was assessed with the locally validated eight-item Social Frailty Scale (SFS-8) guided by Bunt’s framework on social frailty (3, 22), with items summed to yield a total score (range: 0-8 points)(4). A score of 0-1 indicates social non-frailty (SNF), 2-3 indicates social pre-frailty (SPF), and ≥4 indicates social frailty (SF). SFS-8 measures the three domains of social resources, social activities and financial resources, and social need fulfilment.

Cognitive assessments

Cognitive function was assessed with the CMMSE and the locally validated Chinese FAB (17, 18, 21). The CMMSE consists of 28 questions that assesses six specific cognitive function domains: orientation to time, orientation to place, registration, attention and calculation, recall and language and praxis (23). Impairment in cognitive function was determined by education-adjusted cut-offs (CMMSE ≤21 for ≤6 years education and ≥24 for >6 years education) as previously described (21). The locally validated Chinese FAB assesses executive functioning in two domains across six different subtests. The first domain of cognitive control measures conceptualization, mental flexibility, and motor programming. The second domain of behavioural control measures sensitivity to interference, mental flexibility, and environmental autonomy (18, 19). Impairment in executive functioning was determined by education-adjusted cut-offs (Chinese FAB score ≤13 for ≤6 years education, and ≤14 for >6 years education) (18).


We collected demographic data such as age, gender, medical history and assessed body mass index (BMI). Functional status was assessed by the Barthel’s index for basic activities of daily living (ADL) and the Lawton and Brody’s index for instrumental ADL (24, 25). Mood was evaluated using the 15-item Geriatric Depression Scale (26).

Outcome measures

We collected data on four main groups of outcome measures: (i) physical activity, (ii) physical frailty and performance, (iii) geriatric syndromes, and (iv) life space and quality of life. Physical activity was determined by both the International Physical Activity Questionnaire (IPAQ) which converts responses to Metabolic Equivalent Tasks (METs) and the Frenchay Activity Index (FAI) (27, 28). Low physical activity was measured by IPAQ score of <2826 METS and FAI score ≤29 using cohort quintile cut-offs (4). Physical frailty was measured by the modified Fried criteria where a score of 1-2 denotes pre-frailty and ≥3 denotes physical frailty (29).We assessed physical performance via maximal hand grip strength using the North Coast Exacta™ hydraulic hand dynamometer, the Short Physical Performance Battery (SPPB), the three-meter walk comfortable gait speed test, and the five-time sit-to-stand chair test (30). The following cut-offs determined poor physical performance: SPPB scores <11 (31); maximal handgrip strength <28kg for males and <18kg for females; gait speed <1.0m/s; and five-time-sit-to-stand chair test ≥12s, based on the Asian Working Group for Sarcopenia (AWGS) 2019 consensus (32). For geriatric syndromes, we studied near falls or falls, risk of malnutrition, and mood. The occurrence of near falls or falls during the past 12 months was recorded as a single self-reported event, risk of malnutrition due to anorexia of aging was evaluated by the simplified nutritional appetite questionnaire (SNAQ) with a locally validated cut-off at ≤15 (22).
Lastly, life-space mobility was measured by the life-space assessment (LSA), which comprises five life-space levels corresponding to mobility outside the bedroom, home, neighbourhood, outside the neighbourhood and beyond (33). LSA score <76 denotes low life-space mobility (33). Quality of life (QoL) was assessed using index scores of the five-level version of the EuroQol five-dimensional (EQ-5D-5L) questionnaire based on Singapore preference weights derived using an indirect interim mapping method (34, 35). Poor QoL was denoted by the cohort quintile cut-off of EQ-5D-5L index scores <0.881.

Statistical analysis

We performed statistical analyses using IBM SPSS version 23.0 (IBM Corporation, Armonk, NY, USA). All statistical tests were two-tailed with p<.05 considered statistically significant. Continuous variables were expressed as mean (standard deviation) or as median (interquartile range). Categorical variables were expressed as counts and percentages.
To ascertain the cognitive domain (amnestic versus non-amnestic) that is more strongly associated with social frailty, we compared CMMSE and Chinese FAB total and factor scores across SFS-8 categories of socially non-frail, pre-frail and frail. We classified participants as cognitively impaired if their total scores on CMMSE and Chinese FAB were below age and education-adjusted cut-offs (17, 21). We conducted Shapiro-Wilk test to check for assumption of normality. Parametric continuous variables were analysed using one-way analysis of variance with Bonferroni correction for post-hoc comparisons, and non-parametric continuous variables were analysed with the Kruskal-Wallis test. We conducted chi-squared test to analyse categorical variables.
Next, we constructed a 2×2 table for social frailty (non-frail vs pre-frail/frail) and cognitive domain (normal vs impaired). We based the choice of cognitive domain (either CMMSE or Chinese FAB) on which test showed a stronger relationship with social frailty. We then categorized participants into three subgroups: (1) Neither socially frail or cognitively impaired, (2) Either socially frail or cognitively impaired, (3) Both socially frail and cognitively impaired (Figure 1). We compared baseline demographics, functional and frailty status, geriatric syndromes, and outcome measures (physical performance, physical activity, life-space mobility, and quality of life) across the three subgroups.

Figure 1. Classification of social frailty vs executive dysfunction groups


To determine the independent association of social frailty and cognition with the pre-specified outcomes, logistic regression was performed for significant variables (p <.05) to determine the odds ratio (OR) and their 95% confidence intervals (CI). We performed unadjusted analysis, followed by model 1 adjusting for age, gender, education, and hypertension, and finally model 2 which additionally adjusted for GDS, as depressive symptoms have been shown to be a significant determinant of both social frailty and cognition (4, 5, 14, 36).



Amongst 229 participants in this study, the mean age was 67.2±7.4 years, with an average education of 10.7±4.4 years and of predominantly Chinese (92.6%) ethnicity. Comorbidities included hypertension (35.8%), hyperlipidemia (56.8%) and type II diabetes mellitus (14.4%). The median (IQR) for the Barthel basic ADL index, Lawton’s instrumental ADL index, and SPPB were 100 (95.0-100), 23.0 (22.0-23.0) and 12.0 (11.0-12.0) respectively, attesting to the fairly robust health state of the participants. Correspondingly, 51.5% were classified as physically robust, 44.5% as physically pre-frail and only 3.9% physically frail based on the Modified Fried criteria (Table 1).

Table 1. Social frailty and executive dysfunction: Comparison of neither, either and both groups

Values are expressed as mean±SD; median (interquartile range); or N (%).; ADL: Activities of Daily Living; EQ-5D-5L: five-level dimension EuroQoL questionnaire; FAI: Frenchay Activity Index; GDS: Geriatrics Depression Scale; IPAQ: International Physical Activity Questionnaire; SNAQ: Short Nutritional Assessment Questionnaire; SPPB: Short Physical Performance Battery.; *N (%) of individuals with poor self-reported health status based on an EQ-5D utility value less than 0.881 (33); †p<0.05 compared with ‘neither’ group in post-hoc test.; ‡p<0.05 compared with ‘either’ group in post-hoc test.


Relationship between CMMSE vs Chinese FAB with Social Frailty

Total Chinese FAB scores significantly decreased moving from SNF through to SPF and SF (mean±SD: 16.9±1.7 vs 15.7±2.1 vs 15.7±2.4, p<.001), with SPF/SF significantly lower than SNF in post-hoc comparison (p<.05, Bonferroni correction). In contrast, total CMMSE scores did not differ across social frailty (Table 2). Correspondingly, there was an increase in proportion with cognitive impairment for Chinese FAB (p=.077) but not CMMSE (p=.151) across the spectrum of social frailty. In terms of Chinese FAB factor scores, cognitive control (p<.001) but not behavioural control domain was significantly different. For CMMSE, only the non-amnestic domains of orientation to time and place as well as language and praxis were significant (p≤.001 and p=.012 respectively); the recall domain was not significant (p=.154).

Table 2. CMMSE and CFAB: Comparison of total and factor scores by social frailty subgroups

Values are expressed in mean±SD (continuous variables) and N(%) (categorical variables); CMMSE: Chinese Mini-Mental State Examination; FAB: Frontal Assessment Battery; †p<0.05 compared with Social Non-Frail in post-hoc test; ‡p<0.05 compared with Social Pre-Frail in post-hoc test.


Derivation of Social Frailty-Executive Dysfunction subgroups

Based on the above-mentioned results, we chose executive dysfunction (measured using Chinese FAB) as the cognitive domain to be analysed with social frailty. We derived three subgroups based on the relationship between social frailty and executive dysfunction, namely: (1) Neither, N=140 (61.1%); (2) Either, comprising either SPF/SF, N=73 (34.5%), or executive dysfunction, N=6 (2.6%); and (3) Both, N=10 (4.4%) (Figure 1).

Comparison across Social Frailty-Executive Dysfunction subgroups

There was a significant trend towards increase in age, fewer years of education, higher prevalence of hypertension and osteopenia/osteoporosis, lower BADL scores, lower iADLs scores and higher GDS (all p<.05) moving across ‘neither’ through to ‘both’ subgroups (Table 1). For outcomes, physical activity as measured by FAI and IPAQ were significantly different amongst the three subgroups, with ‘both’ performing the worst in FAI and ‘either’ the worst in IPAQ. Modified Fried scores were highest in the ‘both’ subgroup (20% classified as physically frail, compared with 7.59% and 0.07% in ‘either’ and ‘neither’ subgroups respectively), whilst ‘both’ performed the worst in SPPB, gait speed and repeated chair stand (p<.01, 1-way ANOVA). Compared with ‘neither’, there was a significant trend for near falls/falls, lower SNAQ, lower life space level 2 and total scores, and lower EQ-5D index scores in the ‘either’ and ‘both’ subgroups (all p<.05, 1-way ANOVA).

Logistic regression analysis for outcome measures

We performed logistic regression to determine independent associations with outcome measures (Table 3). In model 1, adjusting for age, gender, education, and hypertension, ‘both’ subgroup was significantly associated with risk of malnutrition, near falls/falls and life space mobility as compared with ‘either’ (SNAQ, OR= 5.67, 95%CI= 1.26-25.58 vs 2.09, 95%CI= 1.15-3.78; Near falls/falls, OR= 5.13, 95%CI=1.09-21.17 vs 1.52, 95%CI= 0.71-3.38 and LSA, OR= 10.56, 95%CI 2.26-49.36 vs 1.56, 95%CI= 0.75-3.26). These associations for the ‘both’ subgroup remained significant even with adjustment for mood in model 2 (SNAQ, OR= 4.79, 95%CI= 1.04-22.02; Near falls/falls, OR= 5.23, 95%CI= 1.10-24.90; LSA, OR= 9.80, 95%CI= 2.07-46.31). Similarly, quality of life had a much larger association with ‘both’ compared with ‘either’ subgroup, which remained significant after adjustment for mood (EQ-5D, OR= 13.2, 95%CI= 2.38-73.4 vs 2.89, 95%CI= 1.23-6.80). In contrast, for the outcomes of IPAQ, physical frailty and SPPB, there was a significant association with ‘either’ subgroup, even after adjusting for mood (all p<.05). FAI was not significant for ‘either’ or ‘both’ subgroups.

Table 3. Social frailty and/or executive dysfunction: Logistic regression against different health outcomes

Binary logistic regression was used to determine the odds ratios (OR) and 95% Confidence Intervals (95% CI) for the association between health outcomes and social frailty dysfunction (SF-ED); Model 1: adjusted for age, gender, years of education, and hypertension; Model 2: adjusted for age, gender, years of education, hypertension and GDS; EQ-5D-5L: five-level dimension EuroQoL questionnaire; FAI: Frenchay Activity Index; IPAQ: International Physical Activity Questionnaire; LSA: Life-Space Assessment; SNAQ: Short Nutritional Assessment Questionnaire; SPPB: Short Physical Performance Battery; †p<0.05; ‡p<0.01



Our paper corroborates the growing body of evidence about the relationship between social frailty and cognition by explicating the deleterious role of concomitant executive dysfunction amongst fairly robust community-dwelling older adults. Using a locally validated social frailty scale built on Bunt’s framework, our results explicated the significant association of executive dysfunction with social frailty in around 5% of our study cohort, and a possible synergistic relationship independent of mood which is associated with near-falls/falls, risk of malnutrition due to anorexia of aging, decreased life space mobility, and lowered quality of life. The strengths of our study include the use of a validated instrument to assess executive function; the comprehensive range of outcomes; statistical adjustment for depressive symptoms in regression analysis; and the relatively robust health status of our study participants which facilitates the exploration of socio-cognitive constructs in an older adult population. Taken together, our study highlights the potential deleterious impact of the co-existence of both social frailty and executive dysfunction, and the importance for community screening programs to detect both conditions as opposed to either alone.
The strong association between executive dysfunction and social frailty in our study, as opposed to the memory and non-memory cognitive domains in the CMMSE, suggests that executive function might be more sensitive to initial dysregulation in community-dwelling older adults. This corroborates reports from previous longitudinal studies that executive dysfunction may occur preclinically before the onset of physical frailty and disease-related memory changes in relatively robust older adults (15, 16). Notably, our results showed a strong association with the cognitive control rather than the behavioural control domain in the Chinese FAB. The conceptualization and mental flexibility items within the cognitive control domain showed discriminative ability to discern between normal and MCI groups (18), thereby supporting their utility to detect early impairment in executive function amongst relatively robust older adults with social frailty.
Our results also suggest a synergistic relationship between social frailty and executive dysfunction with poorer health outcomes. Even after adjusting for mood, our results showed independent associations of having ‘both’ social frailty and executive dysfunction with malnutritional risk due to anorexia of aging, near falls/falls, poorer life-space mobility, and a poorer quality of life. Previous studies did not consistently demonstrate these associations because social frailty and executive dysfunction were examined in isolation, which is akin to the ‘either’ group in our study. Notably, the odds ratio for lower quality of life was 4.5 times higher in the ‘both’ subgroup compared with the ‘either’ subgroup, and corroborates the synergistic relationship above and beyond the independent contribution of executive function and social support on quality of life (37, 38). Significantly, social frailty rather than executive dysfunction is more prevalent in the ‘either’ subgroup in our study (Figure 1) and is likely the main driver in the observed associations with physical frailty, physical performance and physical activity, findings which are synonymous with evidence from earlier studies (4, 22).
Executive functioning plays an important role in enabling higher-level functions such as planning, making decisions and sourcing for new information, thus maintaining independence in daily life (16, 39). Earlier studies in healthy community-dwelling populations have reported that the inability to plan is one of the first signs of cognitive decline (18). Coupled with social frailty, these individuals will have increasing difficulty in maintaining their independence, leading to eventual disability (7, 11, 40). A longitudinal study recently demonstrated temporal associations between restricted life-space mobility with executive dysfunction at baseline (41). Recently, a novel concept described as motoric cognitive risk (MCR), characterised by slower gait and an increase in subjective memory complaints, despite being independent in daily activities of living, was associated with falls, disability and death in older adults (42). The association of the MCR syndrome with higher cognitive motor dual tasks costs was corroborated by recent evidence, suggesting impairment in executive functioning of these older adults leading to lower cognitive performance and poorer motor function (43). Taken together, the association with restricted life-space mobility, near falls/ falls and anorexia of aging in our study could be explained by the interplay between social frailty and executive dysfunction. Because executive dysfunction is primarily mediated by alterations in the frontal lobes (43, 44), it is integral that neuro-imaging studies be conducted to ascertain and elucidate the association between neuro-pathological mechanisms with the observed changes in outcomes.
Taken together, our findings suggest that under the broad umbrella of frailty, a specific phenotype characterised by concomitant social and cognitive issues might exist even amongst relatively healthy older persons. The synergistic relationship of the social frailty-dysexecutive phenotype with geriatric syndromes, decreased life space and poorer quality of life, highlights the importance of early identification of at-risk older persons via a comprehensive assessment which includes both social domains and cognitive function. This can in turn facilitate early intervention via community support programs and innovative platforms to provide a safe yet effective means to engage older persons in social and cognitive dimensions. The implications are especially salient for early identification of frail populations for intervention planning in the COVID-19 era (45). The implementation of strict safe distancing measures to curb the COVID-19 contagion has resulted in social estrangement and a sharp decrease in physical activity due to restrictions on movement (10, 46, 47). This disruption to daily life poses a threat to health even amongst non-frail community dwelling older adults. Due to the lack of social interaction and cognitive stimulation, older adults who fulfil the social frailty-dysexecutive phenotype may be most vulnerable to the resultant adverse effects. Studies have shown that older adults who were socially isolated during the COVID_19 pandemic had 2.74 times higher likelihood of cognitive decline compared to those who did not (48). Earlier studies which have examined the social and cognitive impact of COVID-19 and the accompanying public health control measures on older persons, tend to examine these two constructs separately and in isolation (49). There is thus a need for well-designed studies with longitudinal outcomes to examine both social frailty and executive dysfunction in tandem in order to further understand their combined impact during the COVID-19 pandemic.

Limitations of the study and future work

The results of our study may not be generalizable to non-Asian populations or a frailer spectrum of older adults. Due to the cross-sectional analysis, reverse causality cannot be excluded and the association of social frailty and executive dysfunction with adverse health outcomes needs to be confirmed in well-conducted longitudinal studies with a larger sample size to establish causality. Lastly, executive function is a broad construct that captures various aspects including basic functions such as attention, inhibitory control, working memory, set switching, and higher order functions including planning, decision making, and problem solving. However, assessment of executive function is limited to domains within the Chinese FAB in our study, and the sample size does not permit further analysis to determine which aspects of executive function drive the synergistic association of social frailty with adverse outcomes. Future studies should address the issue of whether this association of both social frailty and executive dysfunction holds true across different cultural populations, and to deconstruct the executive function domain to determine specific cognitive factors that contribute to the synergistic relationship we see between social frailty and executive dysfunction.


Our study highlights the association of social frailty with executive dysfunction, particularly the cognitive control domain, in relatively healthy older adult community dwellers. This synergistic relationship, independent of mood, is associated with higher risk for near-falls/falls, anorexia of aging, decreased life space mobility, and quality of life. Our results support the role of community screening programs for both cognitive impairment and social frailty, with a focus on those with co-existent executive dysfunction and social frailty as a possible target for early intervention, as opposed to social frailty or executive dysfunction in isolation. In light of the current COVID-19 pandemic, the association between social frailty and executive dysfunction merits further study as a possible target for early intervention in relatively healthy older adults.


Funding: This research was funded by the Lee Foundation Grant 2019. The funder had no role in the design of the study; in the collection, analyses, or interpretation of data; in the preparation of the manuscript, or in the review or approval of the manuscript and in the decision to publish the results.

Acknowledgements: We would like to thank all participants who contributed to this study.

Conflicts of Interest: The authors declare no conflict of interest.



1. Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. The lancet. 2013;381(9868):752-62.
2. Wleklik M, Uchmanowicz I, Jankowska EA, Vitale C, Lisiak M, Drozd M, et al. Multidimensional approach to frailty. Front Psychol. 2020;11:564.
3. Bunt S, Steverink N, Olthof J, Van Der Schans C, Hobbelen J. Social frailty in older adults: a scoping review. Eur J Ageing. 2017;14(3):323-34.
4. Pek K, Chew J, Lim JP, Yew S, Tan CN, Yeo A, et al. Social frailty is independently associated with mood, nutrition, physical performance, and physical activity: Insights from a theory-guided approach. Int J Environ Res Public Health. 2020;17(12):4239.
5. Tsutsumimoto K, Doi T, Makizako H, Hotta R, Nakakubo S, Kim M, et al. Social frailty has a stronger impact on the onset of depressive symptoms than physical frailty or cognitive impairment: a 4-year follow-up longitudinal cohort study. J Am Med Dir Assoc. 2018;19(6):504-10.
6. Tsutsumimoto K, Doi T, Makizako H, Hotta R, Nakakubo S, Makino K, et al. Association of social frailty with both cognitive and physical deficits among older people. J Am Med Dir Assoc. 2017;18(7):603-7.
7. Makizako H, Shimada H, Tsutsumimoto K, Lee S, Doi T, Nakakubo S, et al. Social frailty in community-dwelling older adults as a risk factor for disability. J Am Med Dir Assoc. 2015;16(11):1003. e7-. e11.
8. Makizako H, Shimada H, Tsutsumimoto K, Hotta R, Nakakubo S, Makino K, et al. Social frailty leads to the development of physical frailty among physically non-frail adults: A four-year follow-up longitudinal cohort study. Int J Environ Res Public Health. 2018;15(3):490.
9. Suthutvoravut U, Tanaka T, Takahashi K, Akishita M, Iijima K. Living with family yet eating alone is associated with frailty in community-dwelling older adults: the Kashiwa study. J Frailty Aging. 2019;8(4):198-204.
10. Pek K, Tan CN, Yew S, Yeo A, Lim JP, Chew J, et al. COVID-19 Pandemic Control Measures: Impact on Social Frailty and Health Outcomes in Non-Frail Community-Dwelling Older Adults. J Nutr Health Aging. 2021.
11. Teo N, Gao Q, Nyunt MSZ, Wee SL, Ng T-P. Social frailty and functional disability: Findings from the Singapore Longitudinal Ageing Studies. J Am Med Dir Assoc. 2017;18(7):637. e13-. e19.
12. Teo N, Yeo PS, Gao Q, Nyunt MSZ, Foo JJ, Wee SL, et al. A bio-psycho-social approach for frailty amongst Singaporean Chinese community-dwelling older adults–evidence from the Singapore longitudinal aging study. BMC Geriatr. 2019;19(1):1-14.
13. Ma L, Sun F, Tang Z. Social frailty is associated with physical functioning, cognition, and depression, and predicts mortality. J Nutr Health Aging. 2018;22(8):989-95.
14. Santos NC, Costa PS, Cunha P, Cotter J, Sampaio A, Zihl J, et al. Mood is a key determinant of cognitive performance in community-dwelling older adults: a cross-sectional analysis. Age. 2013;35(5):1983-93.
15. Carlson MC, Xue Q-L, Zhou J, Fried LP. Executive decline and dysfunction precedes declines in memory: the Women’s Health and Aging Study II. J Gerontol A Biol Sci Med Sci. 2009;64(1):110-7.
16. Gross AL, Xue Q-L, Bandeen-Roche K, Fried LP, Varadhan R, McAdams-DeMarco MA, et al. Declines and impairment in executive function predict onset of physical frailty. J Gerontol A Biol Sci Med Sci. 2016;71(12):1624-30.
17. Chong MS, Lim W, Chan SP, Feng L, Niti M, Yap P, et al. Diagnostic performance of the Chinese Frontal Assessment Battery in early cognitive impairment in an Asian population. Dement Geriatr Cogn Disord. 2010;30(6):525-32.
18. Goh WY, Chan D, Ali N, Chew A, Chuo A, Chan M, et al. Frontal Assessment Battery in Early Cognitive Impairment: Psychometric Property and Factor Structure. J Nutr Health Aging. 2019;23(10):966-72.
19. Wang T-L, Hung Y-H, Yang C-C. Psychometric properties of the Taiwanese (traditional Chinese) version of the Frontal Assessment Battery: A preliminary study. Appl Neuropsychol Adult. 2016;23(1):11-20.
20. Van Kan GA, Rolland Y, Bergman H, Morley J, Kritchevsky S, Vellas B. The IANA Task Force on frailty assessment of older people in clinical practice. J Nutr Health Aging. 2008;12(1):29-37.
21. Sahadevan S, Lim PiPJ, Tan NJL, Chan SP. Diagnostic performance of two mental status tests in the older Chinese: influence of education and age on cut-off values. Int J Geriatr Psychiatry. 2000;15(3):234-41.
22. Lau S, Pek K, Chew J, Lim JP, Ismail NH, Ding YY, et al. The Simplified Nutritional Appetite Questionnaire (SNAQ) as a Screening Tool for Risk of Malnutrition: Optimal Cutoff, Factor Structure, and Validation in Healthy Community-Dwelling Older Adults. Nutrients. 2020;12(9):2885.
23. Yu R, Morley JE, Kwok T, Leung J, Cheung O, Woo J. The effects of combinations of cognitive impairment and pre-frailty on adverse outcomes from a prospective community-based cohort study of older Chinese people. Front Med (Lausanne). 2018;5:50.
24. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. The gerontologist. 1969;9(3_Part_1):179-86.
25. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index: a simple index of independence useful in scoring improvement in the rehabilitation of the chronically ill. Md State Med J. 1965.
26. Lim KK, Chan A. Association of loneliness and healthcare utilization among older adults in Singapore. Geriatrics & gerontology international. 2017;17(11):1789-98.
27. Hurtig-Wennlöf A, Hagströmer M, Olsson LA. The International Physical Activity Questionnaire modified for the elderly: aspects of validity and feasibility. Public Health Nutr. 2010;13(11):1847-54.
28. Schuling J, De Haan R, Limburg Mt, Groenier K. The Frenchay Activities Index. Assessment of functional status in stroke patients. Stroke. 1993;24(8):1173-7.
29. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-M57.
30. Guralnik JM, Ferrucci L, Pieper CF, Leveille SG, Markides KS, Ostir GV, et al. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci. 2000;55(4):M221-M31.
31. Oh B, Cho B, Choi H-C, Son K-Y, Park SM, Chun S, et al. The influence of lower-extremity function in elderly individuals’ quality of life (QOL): an analysis of the correlation between SPPB and EQ-5D. Arch Gerontol Geriatr. 2014;58(2):278-82.
32. Chen L-K, Woo J, Assantachai P, Auyeung T-W, Chou M-Y, Iijima K, et al. Asian Working Group for Sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment. J Am Med Dir Assoc. 2020;21(3):300-7. e2.
33. Peel C, Baker PS, Roth DL, Brown CJ, Bodner EV, Allman RM. Assessing mobility in older adults: the UAB Study of Aging Life-Space Assessment. Phys Ther. 2005;85(10):1008-19.
34. Luo N, Wang P, Thumboo J, Lim Y-W, Vrijhoef HJ. Valuation of EQ-5D-3L health states in Singapore: modeling of time trade-off values for 80 empirically observed health states. Pharmacoeconomics. 2014;32(5):495-507.
35. Van Hout B, Janssen M, Feng Y-S, Kohlmann T, Busschbach J, Golicki D, et al. Interim scoring for the EQ-5D-5L: mapping the EQ-5D-5L to EQ-5D-3L value sets. Value Health. 2012;15(5):708-15.
36. Lockwood KA, Alexopoulos GS, van Gorp WG. Executive dysfunction in geriatric depression. Am J Psychiatry. 2002;159(7):1119-26.
37. Davis JC, Marra CA, Najafzadeh M, Liu-Ambrose T. The independent contribution of executive functions to health related quality of life in older women. BMC Geriatr. 2010;10(1):1-8.
38. de Labra C, Maseda A, Lorenzo-López L, López-López R, Buján A, Rodríguez-Villamil JL, et al. Social factors and quality of life aspects on frailty syndrome in community-dwelling older adults: the VERISAÚDE study. BMC Geriatr. 2018;18(1):1-9.
39. De Silva NA, Gregory MA, Venkateshan SS, Verschoor CP, Kuspinar A. Examining the association between life-space mobility and cognitive function in older adults: a systematic review. J Aging Res. 2019;2019.
40. Yamada M, Arai H. Social frailty predicts incident disability and mortality among community-dwelling Japanese older adults. J Am Med Dir Assoc. 2018;19(12):1099-103.
41. Poranen-Clark T, von Bonsdorff MB, Rantakokko M, Portegijs E, Eronen J, Pynnönen K, et al. The temporal association between executive function and life-space mobility in old age. J Gerontol A Biol Sci Med Sci. 2018;73(6):835-9.
42. Chhetri JK, Chan P, Vellas B, Cesari M. Motoric cognitive risk syndrome: predictor of dementia and age-related negative outcomes. Front Med (Lausanne). 2017;4:166.
43. Ward N, Menta A, Peach S, White S, Jaffe S, Kowaleski C, et al. Cognitive Motor Dual Task Costs in Older Adults with Motoric Cognitive Risk Syndrome. J Frailty Aging. 2021:1-6.
44. Harrington MG, Chiang J, Pogoda JM, Gomez M, Thomas K, Marion SD, et al. Executive function changes before memory in preclinical Alzheimer’s pathology: a prospective, cross-sectional, case control study. PLoS One. 2013;8(11):e79378.
45. Cerami C, Canevelli M, Santi GC, Galandra C, Dodich A, Cappa SF, et al. Identifying Frail Populations for Disease Risk Prediction and Intervention Planning in the Covid-19 Era: A Focus on Social Isolation and Vulnerability. Frontiers in Psychiatry. 2021;12.
46. Steinman MA, Perry L, Perissinotto CM. Meeting the care needs of older adults isolated at home during the COVID-19 pandemic. JAMA Intern Med. 2020;180(6):819-20.
47. Launay CP, Cooper-Brown L, Ivensky V, Beauchet O. Frailty and Home Confinement during the COVID-19 Pandemic: Results of a Pre-Post Intervention, Single Arm, Prospective and Longitudinal Pilot Study. J Frailty Aging. 2021:1-2.
48. Noguchi T, Kubo Y, Hayashi T, Tomiyama N, Ochi A, Hayashi H. Social Isolation and Self-Reported Cognitive Decline Among Older Adults in Japan: A Longitudinal Study in the COVID-19 Pandemic. J Am Med Dir Assoc. 2021.
49. Manca R, De Marco M, Venneri A. The impact of COVID-19 infection and enforced prolonged social isolation on neuropsychiatric symptoms in older adults with and without dementia: a review. Front Psychiatry. 2020;11:1086.




R. Otsuka1, S. Zhang1, C. Tange1, Y. Nishita1, M. Tomida1, K. Kinoshita2, Y. Kato1,3, F. Ando1,3, H. Shimokata1,4, H. Arai5


1. Department of Epidemiology of Aging, Research Institute, National Center for Geriatrics and Gerontology, Aichi, Japan; 2. Department of Frailty Research, Research Institute, National Center for Geriatrics and Gerontology, Aichi, Japan; 3. Faculty of Health and Medical Sciences, Aichi Shukutoku University, Aichi, Japan; 4. Graduate School of Nutritional Sciences, Nagoya University of Arts and Sciences, Aichi, Japan; 5. National Center for Geriatrics and Gerontology, Aichi, Japan.

Corresponding Author: Rei Otsuka, Department of Epidemiology of Aging, Research Institute, National Center for Geriatrics and Gerontology, 7-430 Morioka-cho, Obu, Aichi 474-8511, Japan, E-mail: otsuka@ncgg.go.jp, Tel: +81-562-46-2311

J Frailty Aging 2021;in press
Published online October 7, 2021, http://dx.doi.org/10.14283/jfa.2021.42



Background: Frailty is a dynamic process, with frequent transitions between frailty, prefrailty, and robust statuses over time. The effect of dietary intake on frailty transitions is unknown.
Objective: To examine the association between dietary intake and frailty transitions.
Design: Survey-based retrospective analysis of the National Institute for Longevity Sciences-Longitudinal Study of Aging data.
Setting: Areas neighboring the National Center for Geriatrics and Gerontology in Aichi Prefecture, Japan.
Participants: We included 469 prefrail community dwellers aged 60–87 years who participated both in the baseline (2008–2010) and 2-year follow-up (2010–2012) surveys of the National Institute for Longevity Sciences-Longitudinal Study of Aging.
Measurements: Transitions of frailty were categorized by changes in status from baseline to follow-up: “deterioration (prefrail to frail),” “persistence (persistent prefrail),” and “reversal (prefrail to robust).” Estimated dietary (nutrients and food) intakes assessed by 3-day dietary records in each frailty transition were analyzed with a multivariate-adjusted general linear model after adjusting for sex, age, education, family income, smoking, and chronic disease.
Results: At the 2-year follow-up, 28%, 7%, and 65% of participants had robust, frail, and pre-frail status, respectively. Among 13 food groups, only milk and dairy product intake was positively associated with frailty reversal even after adjusting for all frailty criteria at baseline. Despite insignificant differences in the estimated mean intakes, the baseline intake of saturated fatty acids, potassium, and vitamin B1 tended to be the highest in the reversal group. The estimated mean (standard error) for milk and dairy product intake (g/day) was 79.1 (28.6), 129.3 (19.9), and 161.7 (21.7) for the deterioration, persistence, and reversal groups, respectively (P=0.0036, P-trend=0.0019).
Conclusions: Daily consumption of dairy products may contribute to frailty reversal and frailty prevention among older community dwellers who consume small amounts of dairy products. Other food groups showed no association with frailty status transitions.

Key words: Frailty, reversibility, diet, retrospective study.



Frailty is an emerging global burden among the older population (1) and is characterized by a decline in functioning across multiple physiological systems that is accompanied by an increased vulnerability to stressors. Frailty is a dynamic process, with frequent transitions between frailty, prefrailty, and robust statuses over time. As an intermediate state between robust and frailty, those with prefrailty have a high risk of progressing to frailty (2, 3). Pre-frailty naturally tends to rapidly worsen and transit to frailty. However, previous studies have shown that prefrail individuals may reverse toward a robust state if appropriate interventions are taken (for example, improving nutritional intake or increasing physical activity) (4-7). Considering the high prevalence (approximately 50%) of prefrail individuals in the older population (8, 9), identifying potentially modifiable risk factors for prefrail transitions is needed.
Review studies have shown that dietary factors, such as macronutrients (e.g., protein), micronutrients [e.g., carotenoids and 25(OH)D], food groups (e.g., low-fat dairy products), and dietary patterns (e.g., the Mediterranean diet), are related to frailty syndrome or frail criteria (10-12). However, the subjects of these studies were predominantly non-frail individuals. Meanwhile, interventional studies on frail participants suggested similar promising effects of dietary factors on frailty improvement. Nevertheless, the duration in most of the interventional studies was too short (from a few weeks to 6 months) to yield long-term outcomes (10, 12). Therefore, the effect of dietary intake on frail transitions is unknown.
In this retrospective study, we focused on prefrail individuals at baseline and examined changes in their status at the 2-year follow-up. We aimed to evaluate the association between baseline dietary intake and 2-year frailty transitions to clarify the dietary/nutritional factors that are related to frailty deterioration and reversal among older community dwellers.



Study design and participants

Data for this survey were obtained from the National Institute for Longevity Sciences-Longitudinal Study of Aging (NILS-LSA), which used detailed questionnaires, medical checkups, anthropometric measurements, physical fitness tests, and nutritional examinations to assess the normal aging process over time. NILS-LSA participants comprised randomly selected age- and sex-stratified individuals from a pool of community-dwelling residents of Obu City and Higashiura Town, which are in the neighborhood of the National Center for Geriatrics and Gerontology in Aichi Prefecture, Japan. The initial survey of the NILS-LSA involved 2,267 men and women aged between 40 and 79 years, including approximately 280 men and 280 women in each decade of age. These subjects were followed-up every 2 years from the first wave (November 1997–April 2000) to the second wave (April 2000–May 2002), third wave (May 2002–May 2004), fourth wave (June 2004–July 2006), fifth wave (July 2006–July 2008), sixth wave (July 2008–July 2010), and seventh wave (July 2010–July 2012). When participants could not be followed up (e.g., moved to another area, dropped out for personal reasons, or died), except for those aged 80 years or older, new decade- and sex-matched participants were randomly recruited from the second to seventh waves. Moreover, participants aged ≥40 years were newly recruited every year. Each wave included approximately 2300 men and women. The Committee on the Ethics of Human Research of the National Center for Geriatrics and Gerontology approved the study protocol (No. 899-6), and written informed consent was obtained from all participants. Details of the NILS-LSA have been reported previously (13).
Given the small number of older adults in the first to fifth waves of the NILS-LSA, the sixth wave was used as the baseline, and the seventh wave was used as the follow-up survey (Figure 1). The main study of NILS-LSA sixth wave included participants aged 40 to 89 years, but this study focused on older adults aged ≥60 years. We considered individuals aged ≥60 years as older adults.
The baseline included 1,173 men and 1,129 women (age range: 40–89 years), and the exclusion criteria (shown in Figure 1) were as follows: 1) age <60 years at baseline (n = 1,042), 2) non-participation in the follow-up (n = 205), 3) lack of body weight measurement at the fifth wave (n = 53; this was important because shrinking at baseline could not be assessed otherwise), 4) incomplete data of any frailty criterion at baseline (n = 11), 5) definition as frail (n = 57) or robust (n = 416) at baseline, and 6) incomplete data on nutritional assessments (n = 29) or self-reported questionnaires (n = 20) at baseline.
Ultimately, the data of 469 Japanese older adults (232 men and 237 women, aged 60–87 years) were available for analysis. The mean (standard deviation) interval between baseline and follow-up for all participants was 2.0 (0.1) years.

Figure 1. Participant flow chart and variables used in the analyses

Frailty assessment and transitions

Frailty was assessed using five criteria—slowness, weakness, exhaustion, low activity, and shrinking—based on a modified set of criteria established by the Cardiovascular Health Study (14). Details of the modified criteria have been reported previously (15).
In brief, slowness was defined as gait disturbance or a gait speed <1.0 m/s in a 10-m walk test using a comfortable gait. Weakness was defined as a maximum grip strength <26 and <18 kg for men and women, respectively. Exhaustion was assessed through self-reports based on responses to two statements in the Center for Epidemiologic Studies Depression Scale (16): the statements included “I felt that everything I did was an effort,” and “I could not get ‘going.’” The responses included “Rarely or none of the time (less than 1 day during the past week),” “Some or a little of the time (1–2 days),” “Occasionally or a moderate amount of time (3–4 days),” and “Most or all of the time (5–7 days).” Participants who did not answer “Rarely or none of the time” for both statements were identified as showing exhaustion. Low activity was defined as the lowest 20% metabolic equivalents of leisure-time physical activity based on sex, assessed using the modified Minnesota Leisure-time Physical Activity Questionnaire (17). Shrinking was defined as a ≥5% weight loss within the preceding 2 years (during the fifth to sixth [baseline] and sixth to seventh [follow-up] waves).
Frailty was defined as meeting 3 or more criteria, prefrailty was defined as meeting 1–2 criteria, and robustness was defined as not meeting any of the prespecified frailty criteria. Thus, transitions of frailty among prefrail participants at baseline (n = 469) were categorized into three groups according to changes in status from baseline to follow-up: “deterioration (prefrail to frail),” “persistence (persistent prefrail),” and “reversal (prefrail to robust).”
Supplemental Table 1 shows the prevalence of each frailty criterion at baseline and follow-up according to frailty transitions.

Nutritional assessments

On the day they participated in the baseline survey, trained dietitian staff explained the purpose and methods of the dietary record survey during lunchtime. The participants were instructed to take the survey on days when they ate normal meals as much as possible, avoiding special days such as anniversaries, because the dietary survey assessed their normal dietary habits. During this time, a 1-kg kitchen scale (Sekisui Jushi, Tokyo, Japan) and one and/or more disposable camera (27 shots; Fuji Film, Tokyo, Japan) were provided to all participants, and a practice session on taking pictures with the disposable camera was also conducted (Appendix Figure 1).
After participation in the baseline study, subjects completed a 3-day dietary record to assess dietary intake, including supplement use. The dietary record was completed over 3 consecutive days (2 weekdays and 1 day in the weekend) since food habits differed on weekdays and weekends; 3 consecutive days were selected based on the same method used by the National Health and Nutrition Examination Survey in Japan (18, 19).
All food items, including spices and seasonings, were weighed/measured separately on the 1-kg kitchen scale with a lightweight cup or spoon before being cooked, or the portion sizes were estimated. During the 3 days, all activities of eating and drinking, including consuming snacks, were noted in detail. In addition, subjects used the disposable camera to take photos of their meals before and after eating. When taking the photos, they were asked to place a scale paper on the dining table so that we could estimate the size of the plates and foods. If it was difficult for the subjects to write detailed records (i.e., if they were not in charge of cooking), the cooks conducted the recording. Subjects completed the dietary record at home, and most returned it within 1 month.
The films in the returned disposable cameras were developed. Dietitians then used these photos to complete the missing information in each subject’s dietary record and assigned a code number from the Standard Tables of Food Composition in Japan 2010 (20) for every food (the number of codes in each subject’s records was around a few hundreds). Based on the code number and weight records, we used the Statistical Analysis System software version 9.3 (SAS Institute, Cary, NC, USA) to calculate the average food and nutrient intake (including alcohol intake) over 3 days from the nutrient intakes included in the Standard Tables of Food Composition in Japan 2010 (20), without using any dietary assessment software. There are 17 food groups in the Japanese food composition table (14): cereals, potatoes, sugars and sweeteners, beans, nuts and seeds, vegetables (non-green-yellow/green-yellow), fruits, mushrooms, seaweed, fish and shellfish, meats, eggs, milk and dairy products, fats and oils, confectionaries, beverages, and seasoning and spices. In this study, we included 13 food groups: cereals, potatoes, beans, nuts and seeds, non-green-yellow vegetables, green-yellow vegetables, fruits, mushrooms, seaweed, fish and shellfish, meats, eggs, milk and dairy products, and excluded sugars and sweeteners, fats and oils, confectionaries, beverages, and seasoning and spices. Each food group contains dozens or hundreds of foods, for example, the milk and dairy products group contains the nutritional values of 52 different foods, including natural milk, low-fat milk, yogurt, cheese, and butter.
Calculating the nutrient values for each subject based on the coding chart took several hours even for a trained dietitian, and the values were checked by two or more other trained dietitians to prevent coding errors. Information on any discrepancies and any requisite additional information was obtained via telephone calls to the subjects. However, if the records were still insufficient, the dietary survey data were considered missing.

Other measurements

Weight and height were measured in the fasting state (around 9–10 am) to the nearest 0.1 kg and 0.1 cm, respectively, with participants wearing light clothing and no shoes. The body mass index was calculated as the body weight in kilograms divided by the square of the height in meters. Data on years of education (≤12 or ≥13 years of school), annual family income (<5.5 or ≥5.5 million yen per year), smoking status (current, never/former), and chronic disease history (past and present stroke, heart disease, hypertension, dyslipidemia, and diabetes mellitus; yes/no, for each) were collected using a self-administered questionnaire and confirmed by medical doctors or trained staff.

Statistical analysis

According to frailty transitions, differences in proportions and means of baseline characteristics were assessed using the chi-squared test and general linear model, respectively. As for dietary variables, we selected the parametric test (general linear model) because each variable is a continuous quantity and the distribution was close to a normal distribution. A general linear model adjusted for sex, baseline age, years of education, annual family income, smoking status, and history of chronic disease was applied to compare baseline nutrient intakes according to the frailty transitions. Moreover, to compare baseline food intakes according to frailty transitions, general linear regression was performed again. Model 1 was adjusted for sex, baseline age, years of education, annual family income, smoking status, and chronic disease history. Model 2 was further adjusted for the total number of frailty criteria at baseline. BMI and physical activity were not treated as an adjustment factor because 1) the weight used in the BMI calculation was used to determine the weight loss (shrinking) of the frailty component, 2) the leisure-time physical activity was used to determine the low activity of the frailty component.
To test any trend in the association between dietary intakes and frailty transitions, the trend test of the general linear model was applied with ascending ordinal values −1, 0, and 1 assigned to “deterioration (prefrail to frail),” “persistence (persistent prefrail),” and “reversal (prefrail to robust),” respectively. In sub-analyses, we focused on dairy intakes and estimated the daily mean dairy intakes (including milk, yogurt, and cheese) according to frailty transitions using the general linear model. Adjustments were the same as those in Model 2.
All statistical analyses were performed using the Statistical Analysis System software version 9.3 (SAS Institute, Cary, NC, USA). All reported P-values were two-sided, and a P-value <0.05 was considered significant.



At the 2-year follow-up, 28% (n = 130) of participants had reversed to robust status, whereas 7% (n = 32) and 65% (n = 307) had deteriorated to frail and remained at prefrail statuses, respectively. Participants in the reversal group were younger than those in the other two groups. Although the prevalence of chronic disease (except dyslipidemia) tended to be high in the deterioration group, no significant differences were observed (Table 1).

Table 1. Baseline characteristics of subjects according to the transitions of frailty

*General linear model used for continuous variables; χ2 test used for the categorical variables; †Trend test of the general linear model used with ascending ordinal values −1, 0, and 1 assigned to “deterioration,” “persistence,” and “reversal,” respectively.


Table 2 shows the multivariate-adjusted baseline nutrient intakes according to transitions of frailty. After adjusting for sex, baseline age, years of education, annual family income, smoking status, and chronic disease history, the baseline intake of saturated fatty acids, potassium, and vitamin B1 tended to be high in the reversal group (P-trend: 0.03, 0.04, and 0.02, respectively), although the differences in estimated mean intakes were not significant among the three groups.

Table 2. Baseline nutrient intakes according to the transitions of frailty *

*General linear model used. Adjusted for sex, baseline age, years of education, annual family income, smoking status, and chronic disease history; †Trend test of the general linear model used with ascending ordinal values −1, 0, and 1 assigned to “deterioration,” “persistence,” and “reversal,” respectively; ‡Estimated mean (standard error), all such values.


Table 3 shows the baseline food intakes according to the transitions of frailty. After multivariate adjustment, milk and dairy product intake were positively associated with frailty reversal. The corresponding estimated mean (standard error) values for milk and dairy products intake (g/day) for “frailty deterioration,” “frailty persistence,” and “frailty reversal” were 79.1 (28.6), 129.3 (19.9), and 161.7 (21.7), respectively (P-value 0.0036, P-trend 0.0019; Model 1). Even after adjusting for the total number of frailty criteria at baseline, the positive association remained (P-value 0.0028, P-trend 0.0014; Model 2). No other food intake was associated with frailty reversal.
In sub-analyses (Supplemental Table 2), milk and yogurt intake, but not cheese intake, were positively associated with frailty reversal.

Table 3. Baseline food intakes according to transitions of frailty *

*General linear model was used; †Adjusted for sex, baseline age, years of education, annual family income, smoking status, and chronic disease history. ‡Adjusted for Model 1 + total number of frailty criteria at baseline; §Trend test of the general linear model used with ascending ordinal values −1, 0, 1 assigned to “deterioration,” “persistence,” and “reversal,” respectively; ||Estimated mean (standard error), for all such values.



This retrospective study indicated that dairy product intake was positively associated with frailty reversal among older community dwellers. To our best knowledge, this is the first study to investigate the association between dietary intake and frailty transitions.
Cuesta-Triana et al. (21) reviewed the effect of milk and other dairy products on the risk of frailty, sarcopenia, and cognitive performance decline in older adults. They concluded that dairy product consumption might reduce the risk of frailty, especially high consumption of low-fat milk and yogurt. In our study, milk and yogurt intake, but not low-fat milk or cheese intake, were associated with frailty reversal. This may be because our participants consumed relatively low amounts of low-fat dairy products (mean 21.1 g/day) and cheese (mean 2.5 g/day); therefore, the association between the intake of these products and frailty transitions was difficult to determine. Although the mean value of dairy product intake in our study was smaller than reports from previous (Spain) studies (306 g vs. 141 g per day) among robust community-dwelling older adults (21, 22), both studies observed benefits of high dairy product intake on frailty prevention. A recent study indicated that dairy consumption was associated with a lower risk of mortality and major cardiovascular disease events in a diverse multinational cohort in 21 countries (23). This suggests that daily dairy product consumption may protect frailty development both in robust and prefrail older adults.
Dairy products are good sources of protein, vitamins, and minerals (24), and these nutrients are considered to play an important role in the prevention of frailty. Although the differences in the estimated mean intakes were not significant, the intake of saturated fatty acids, potassium, and vitamin B1 tended to be highest in the reversal group among the frailty transition groups. A higher percentage of saturated fatty acid intake was associated with both higher frailty and mortality (25), even after considering the degree of nutritional deficits. However, the mean total fatty acid and saturated fatty acid intakes were higher in that study (25) compared to our sample (total fatty acid: 75.65 g vs. 46.3–50.8 g per day, saturated fatty acid: 24.77 g vs. 12.2–14.4 g per day; Table 2). Therefore, while excessive intake of saturated fatty acids is detrimental to frailty, the moderate intake of saturated fatty acids among older adults who consume small amounts of saturated fatty acid may be effective in preventing frailty.
Potassium exerts beneficial effects on cardiovascular diseases (26). The salt intake of the Japanese people is extremely high compared to the rest of the world, and in our sample, the sodium intake was 3809.2, 3976.7, and 3891.4 mg, the potassium intake was 2415.8, 2615.3, and 2676.7 mg, and the sodium/potassium ratio was 1.58, 1.52, and 1.54 for the deterioration, persistence, and reversal groups, respectively. Therefore, in a population that has a high salt intake, the high potassium intake may partly modify the frailty status through a reduction in the risk of cardiovascular disease because high sodium intake is a strong risk factor for cardiovascular disease (27). The results of this study indicate frailty reversal in older adults with high potassium intake. Among Japanese cohorts, vitamin B1, potassium, and other micronutrient intake were cross-sectionally lower in the prefrailty group (26, 28). These nutrients, including saturated fatty acids, potassium, and vitamin B1, are found abundantly in vegetables, fruits, meat, and/or fish; thus, dairy product intake may not be solely responsible for the improved nutritional status. As the nutritional status of our subjects was relatively good compared to the results of the National Health and Nutrition Examination Survey in Japan (29), the difference in the dairy intake may have affected their prognosis in terms of the frailty status.
Furthermore, the differences in the estimated mean total protein intakes were not significant among the transitions of frailty groups. Therefore, health awareness and accessibility to dairy products may play a role in improving prefrail status by promoting physical and mental health rather than simply relying on protein intake. In our previous study, the intake of dairy products and fruits decreased with age (30). These foods need to be fresh and are difficult for Japanese families and communities to procure and grow at home. Furthermore, these foods are usually heavy and difficult to carry home from the store, which may cause prefrail older adults to refrain from purchasing them. Therefore, our findings indicate that health awareness and actions may be key factors for frailty reversal.
The proportion of transitions among our prefrail participants was similar to that of a previous study. Corresponding proportions for “reversal,” “persistence,” and “deterioration” were 35%, 55%, and 7% over 1.6 years (31) (28%, 65%, and 7% over 2 years in our study). Considering that there are many prefrail older adults in the community and that frailty reversal is highly achievable, advocating dairy intake may be an important nutritional strategy to promote health and prevent frailty.
The strengths of our study are as follows: (1) this was the first study that investigated the association between dietary intake and frailty transitions among prefrail older participants; (2) dietary and nutritional intake was assessed by 3-day dietary records and photographs, which may introduce less recall bias than other methods (e.g., dietary recall and self-reported questionnaires); (3) the estimated mean consumptions of milk and yogurt were 96.5 g/day and 36.4 g/day in the “frailty reversal” group, which are almost half a cup of milk and yogurt. Therefore, if accessed sustainably, even older adults could easily consume these volumes. In other words, at the level of public health policy formulation and advocacy, our findings provide more feasible evidence for frailty prevention and health promotion.
Our study has a few limitations that warrant consideration. First, dietary and nutritional intake was only assessed at baseline. Diet naturally changes over time, and it is affected by various factors related to aging. Second, a 3-day dietary record was not enough to accurately assess long-term eating habits. In addition, the Japanese intake of dairy products is relatively lower than the average global intake (29, 32). Therefore, our findings may not be generalizable to Western populations who consume larger amounts of dairy products. However, even among Western populations, the consumption of dairy products would decline with age; thus, our findings may be valuable for those populations.
In conclusion, consumption of dairy products, including milk and yogurt, might be beneficial for promoting frailty reversal and frailty prevention among older community dwellers who consume small amounts of dairy products. Our findings might be applicable for robust and frail individuals. Future studies are needed to address this hypothesis.


Funding: This work was supported in part by grants from the Food Science Institute Foundation and Research Funding for Longevity Sciences from the National Center for Geriatrics and Gerontology, Japan (grant number 19-10,21-18). The sponsors had no role in the design and conduct of the study; in the data collection, analysis, and interpretation; in the preparation of the manuscript; or in the review or approval of the manuscript.

Acknowledgments: We wish to express our sincere appreciation to the study participants and our colleagues in the NILS-LSA for completing the surveys in this study.

Conflicts of interest: All authors declare no conflicts of interest.

Ethical Standards: This study was carried out in accordance with the ethical standards.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.






1. Hoogendijk EO, Afilalo J, Ensrud KE, Kowal P, Onder G, Fried LP. Frailty: implications for clinical practice and public health. Lancet 2019;394:1365-1375. doi: 10.1016/S0140-6736(19)31786-6.
2. Xue QL. The frailty syndrome: definition and natural history. Clin Geriatr Med 2011;27:1-15. doi: 10.1016/j.cger.2010.08.009.
3. Fallah N, Mitnitski A, Searle SD, Gahbauer EA, Gill TM, Rockwood K. Transitions in frailty status in older adults in relation to mobility: a multistate modeling approach employing a deficit count. J Am Geriatr Soc 2011;59:524-529. doi: 10.1111/j.1532-5415.2011.03300.x.
4. Michel JP, Cruz-Jentoft AJ, Cederholm T. Frailty, Exercise and Nutrition. Clin Geriatr Med 2015;31:375-387. doi: 10.1016/j.cger.2015.04.006.
5. Kojima G, Taniguchi Y, Iliffe S, Jivraj S, Walters K. Transitions between frailty states among community-dwelling older people: A systematic review and meta-analysis. Ageing Res Rev 2019;50:81-88. doi: 10.1016/j.arr.2019.01.010.
6. Gill TM, Gahbauer EA, Allore H, Han L. Transitions between frailty states among community-living older persons. Arch Intern Med 2006;166:418-423. doi: 10.1001/archinte.166.4.418.
7. Mendonça N, Kingston A, Yadegarfar M, et al. Transitions between frailty states in the very old: the influence of socioeconomic status and multi-morbidity in the Newcastle 85+ cohort study. Age Ageing 2020;49:974-981. doi: 10.1093/ageing/afaa054.
8. Hanlon P, Nicholl BI, Jani BD, Lee D, McQueenie R, Mair FS. Frailty and pre-frailty in middle-aged and older adults and its association with multimorbidity and mortality: a prospective analysis of 493 737 UK Biobank participants. Lancet Public Health 2018;3:e323-e332. doi: 10.1016/S2468-2667(18)30091-4.
9. Siriwardhana DD, Hardoon S, Rait G, Weerasinghe MC, Walters KR. Prevalence of frailty and prefrailty among community-dwelling older adults in low-income and middle-income countries: a systematic review and meta-analysis. BMJ Open 2018;8:e018195. doi: 10.1136/bmjopen-2017-018195.
10. Yannakoulia M, Ntanasi E, Anastasiou CA, Scarmeas N. Frailty and nutrition: From epidemiological and clinical evidence to potential mechanisms. Metabolism 2017;68:64-76. doi: 10.1016/j.metabol.2016.12.005.
11. Lorenzo-López L, Maseda A, de Labra C, Regueiro-Folgueira L, Rodríguez-Villamil JL, Millán-Calenti JC. Nutritional determinants of frailty in older adults: A systematic review. BMC Geriatr 2017;17:108. doi: 10.1186/s12877-017-0496-2.
12. Morante JJH, Martínez CG, Morillas-Ruiz JM. Dietary factors associated with frailty in old adults: a review of nutritional interventions to prevent frailty development. Nutrients 2019;11:102. doi: 10.3390/nu11010102.
13. Shimokata H, Ando F, Niino N. A new comprehensive study on aging–the National Institute for Longevity Sciences, Longitudinal Study of Aging (NILS-LSA). J Epidemiol 2000;10:S1-S9. doi: 10.2188/jea.10.1sup_1.
14. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56:M146-M156. doi: 10.1093/gerona/56.3.m146.
15. Yuki A, Otsuka R, Tange C, et al. Epidemiology of frailty in elderly Japanese. J Sports Med Phys Fitness 2016;5:301-307.
16. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385-401.
17. Kozakai R, Doyo W, Tsuzuku S, et al. Relationships of muscle strength and power with leisure-time physical activity and adolescent exercise in middle-aged and elderly Japanese women. Geriatr Gerontol Int 2005;5:182-188.
18. Imai T, Sakai S, Mori K, Ando F, Niino N, Shimokata H. Nutritional assessments of 3-day dietary records in National Institute for Longevity Sciences–Longitudinal Study of Aging (NILS-LSA). J Epidemiol 2000;10:S70-S76. doi: 10.2188/jea.10.1sup_70.
19. Yoshiike N, Matsumura Y, Iwaya M, Sugiyama M, Yamaguchi M. National Nutrition Survey in Japan. J Epidemiol 1996;6:S189-200. doi: 10.2188/jea.6.3sup_189.
20. Ministry of Education Culture, Sports, Science and Technology. Standard Tables of Foods Composition in Japan 2010. Ishiyaku-syuppan.
21. Cuesta-Triana F, Verdejo-Bravo C, Fernández-Pérez C, Martín-Sánchez FJ. Effect of milk and other dairy products on the risk of frailty, sarcopenia, and cognitive performance decline in the elderly: a systematic review. Adv Nutr 2019;10:S105-S119. doi: 10.1093/advances/nmy105.
22. Lana A, Rodriguez-Artalejo F, Lopez-Garcia E. Dairy consumption and risk of frailty in older adults: a prospective cohort study. J Am Geriatr Soc 2015;63:1852-1860. doi: 10.1111/jgs.13626. doi: 10.1111/jgs.13626.
23. Dehghan M, Mente A, Rangarajan S, et al. Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. Lancet 2018;392:2288-2297. doi: 10.1016/S0140-6736(18)31812-9.
24. Thorning TK, Raben A, Tholstrup T, Soedamah-Muthu SS, Givens I, Astrup A. Milk and dairy products: good or bad for human health? An assessment of the totality of scientific evidence. Food Nutr Res 2016;60:32527. doi: 10.3402/fnr.v60.32527.
25. Jayanama K, Theou O, Godin J, Cahill L, Rockwood K. Association of fatty acid consumption with frailty and mortality among middle-aged and older adults. Nutrition 2020;70:110610. doi: 10.1016/j.nut.2019.110610.
26. Liu F, Zhang R, Zhang W, et al. Potassium supplementation blunts the effects of high salt intake on serum retinol-binding protein 4 levels in healthy individuals. J Diabetes Investig 2021;12:658-663. doi: 10.1111/jdi.13376.
27. Cook NR, Appel LJ, Whelton PK. Lower levels of sodium intake and reduced cardiovascular risk. Circulation 2014;129:981-989. doi: 10.1161/CIRCULATIONAHA.
28. Kaimoto K, Yamashita M, Suzuki T, et al. Association of protein and magnesium intake with prevalence of prefrailty and frailty in community-dwelling older Japanese women. J Nutr Sci Vitaminol (Tokyo) 2021;67:39-47. doi: 10.3177/jnsv.67.39.
29. Ministry of Health, Labour and Welfare. The National Health and Nutrition Survey in Japan. 2009. http://www.mhlw.go.jp/bunya/kenkou/eiyou/h21-houkoku.html. Accessed [17 April 2021].
30. Otsuka R, Nishita Y, Tange C, et al. Dietary diversity decreases the risk of cognitive decline among Japanese older adults. Geriatr Gerontol Int 2017;17:937-944. doi: 10.1111/ggi.12817.
31. Gill TM, Gahbauer EA, Allore HG, Han L. Transitions between frailty states among community-living older persons. Arch Intern Med 2006;166:418-423. doi: 10.1001/archinte.166.4.418.
32. Micha R, Khatibzadeh S, Shi P, et al. Global, regional, and national consumption levels of dietary fats and oils in 1990 and 2010: a systematic analysis including 266 country-specific nutrition surveys. BMJ 2014;348:g2272. doi: 10.1136/bmj.g2272.



R. McGrath


Department of Health, Nutrition, and Exercise Sciences, North Dakota State University, Fargo, ND, USA; Fargo VA Healthcare System, Fargo, ND, USA

Corresponding Author: Ryan McGrath, Department of Health, Nutrition and Exercise Sciences, North Dakota State University, NDSU Dept. 2620, PO Box 6050, Fargo, ND 58108, Phone: 701-231-7474, Fax: 701-231-8872, Email: ryan.mcgrath@ndsu.edu

J Frailty Aging 2021;in press
Published online October 6, 2021, http://dx.doi.org/10.14283/jfa.2021.41


Dear Editor,

Screening for age-related disease and disability is critical for proper diagnosis and management of a condition (1). The disabling process, for example, often consists of successive deficits in 1) muscle function, 2) physical performance, and 3) physical functioning during aging. Muscle function is generally characterized by strength, endurance, and power. Objectively measured whole body tasks related to mobility such as gait speed and the short physical performance battery are used to examine physical performance. Persons with poor physical functioning often have problems completing basic self-care tasks and are also at an elevated risk for age-related morbidities such as sarcopenia (2). Given that muscle dysfunction represents the onset and potential progression of the disabling process (2, 3), routinely examining muscle function in clinical settings will allow for the early intervention of age-related disease and disability, which in turn, may increase intervention efficacy and more quickly decelerate disablement.
Strength, endurance, and power have been constantly identified as hallmark characteristics of musculoskeletal fitness that have widespread health benefits across the lifespan (4). Handgrip strength is a well utilized assessment of muscle function in clinical settings, but handgrip strength measurements are inconsistently included in routine geriatric clinical practice and have known limitations (5). For example, the generalizability of handgrip strength alone for being representative of overall muscle function is lacking because handgrip strength does not directly account for lower extremity functioning. Moreover, only strength capacity is being evaluated during handgrip strength assessments, and other aspects of muscle function such as endurance and power are not ascertained (5). However, alternative muscle function assessments that account for both lower extremity function and additional muscle function aspects are not feasible. Procedures for examining muscle endurance with handgrip dynamometry are not yet well established (5). Further, knee extension power measurements require expensive equipment and complex protocols that are mostly invasive for older adults (2). Therefore, continuing to examine how we can feasibly and accurately assess muscle function in clinical practice is crucial for precision in geriatric screenings and appropriate referrals.
The 30-second chair stand test is a measure of lower extremity power and endurance (2). Similar to the gait speed test, the chair stand test can be both conducted as a stand-alone assessment or as part of the short physical performance battery (2). Equipment for conducting the chair stand test as a stand-alone assessment requires a straight back chair without arm rests and a stopwatch, and the number of total sit-stand-sit cycles are counted over 30-seconds (2). The 30-second chair stand protocol differs from that used in the short physical performance battery, such that the time to complete five sit-stand-sit cycles is instead recorded (2). Additional evaluations such as coronal plane angle can also be adopted in chair stand tests (6). Although gait speed is a commonly used physical performance assessment (2), chair stands as an individual assessment are likewise predictive of adverse health outcomes (7-9).
Indeed, standing-and-sitting from a chair includes mobility-related characteristics, and the 30-second chair stand test is accordingly considered a physical performance assessment. However, the test could be more suitable for assessing muscle function because the movements involved in the test are repetitive and fixed. The 30-second chair stand test, as a muscle function assessment, presents a feasible protocol for older adult patients and healthcare providers, and does not require expensive equipment. Moreover, the European Working Group on Sarcopenia in Older People classifies the chair stand test at the assessment category alongside handgrip strength, not in the physical performance category which is instead used to evaluate sarcopenia severity (10). Both power and endurance are also evaluated in the 30-second chair stand test, and these muscle function characteristics are absent when handgrip strength alone is used as an overall assessment of muscle function. As such, the 30-second chair stand test fulfills the limitations of only examining handgrip strength, and should be considered a muscle function assessment.
Given that the 30-second chair stand test could be considered an assessment of muscle function, the timed-up-and-go test should be emphasized as a replacement to chair stands for evaluating physical performance. Similar to chair stands, the timed-up-and-go test not only includes sitting-and-standing as part of the exam, but better encompasses mobility because walking is also involved. Figure 1 proposes a conceptual model of clinical assessments for examining each stage of the disabling cascade (2, 5). Additional handgrip strength measurements are included as an assessment of muscle function (5), but are in need of more research before being recommended in routine geriatric health assessments. Nevertheless, utilizing the 30-second chair stand test as a muscle function assessment, instead of a physical performance assessment, will improve how the disabling process is evaluated in clinical and translational research settings.

Figure 1. Conceptual Model of Clinical Assessments for Examining Each Stage of the Disabling Process

*May include the following handgrip assessments: rate of force development, asymmetry, bilateral strength, force steadiness, fatigability, and task-specific tremoring.


Conflicts of Interest: None declared.

Funding Sources: None declared.



1. Crosignani S, Sedini C, Calvani R, Marzetti E, Cesari M. Sarcopenia in Primary Care: Screening, Diagnosis, Management. J Frailty Aging. 2021;10(3):226-232. doi:10.14283/jfa.2020.63.
2. Beaudart C, Rolland Y, Cruz-Jentoft AJ, et al. Assessment of Muscle Function and Physical Performance in Daily Clinical Practice : A position paper endorsed by the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO). Calcif Tissue Int. 2019;105(1):1-14. doi:10.1007/s00223-019-00545-w.
3. McGrath R, Erlandson KM, Vincent BM, Hackney KJ, Herrmann SD, Clark BC. Decreased Handgrip Strength is Associated With Impairments in Each Autonomous Living Task for Aging Adults in the United States. J Frailty Aging. 2019;8(3):141-145. doi:10.14283/jfa.2018.47.
4. Fraser BJ, Rollo S, Sampson M, et al. Health-Related Criterion-Referenced Cut-Points for Musculoskeletal Fitness Among Youth: A Systematic Review [published online ahead of print, 2021 Aug 2]. Sports Med. 2021;10.1007/s40279-021-01524-8. doi:10.1007/s40279-021-01524-8.
5. McGrath R, Tomkinson GR, Clark BC, et al. Assessing Additional Characteristics of Muscle Function With Digital Handgrip Dynamometry and Accelerometry: Framework for a Novel Handgrip Strength Protocol [published online ahead of print, 2021 Jun 21]. J Am Med Dir Assoc. 2021;S1525-8610(21)00517-X. doi:10.1016/j.jamda.2021.05.033.
6. Takeshima N, Kohama T, Kusunoki M, et al. Development of Simple, Objective Chair-Standing Assessment of Physical Function in Older Individuals Using a KinectTM Sensor. J Frailty Aging. 2019;8(4):186-191. doi:10.14283/jfa.2019.23.
7. Bahat G, Kilic C, Eris S, Karan MA. Power Versus Sarcopenia: Associations with Functionality and Physical Performance Measures. J Nutr Health Aging. 2021;25(1):13-17. doi:10.1007/s12603-020-1544-8.
8. Looijaard SMLM, Oudbier SJ, Reijnierse EM, Blauw GJ, Meskers CGM, Maier AB. Single Physical Performance Measures Cannot Identify Geriatric Outpatients with Sarcopenia. J Frailty Aging. 2018;7(4):262-267. doi:10.14283/jfa.2018.19.
9. Patrizio E, Calvani R, Marzetti E, Cesari M. Physical Functional Assessment in Older Adults. J Frailty Aging. 2021;10(2):141-149. doi:10.14283/jfa.2020.61.
10. Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: revised European consensus on definition and diagnosis [published correction appears in Age Ageing. 2019 Jul 1;48(4):601]. Age Ageing. 2019;48(1):16-31. doi:10.1093/ageing/afy169.




I. Pilati1, A. Slee1, R. Frost2


1. UCL Division of Medicine, Faculty of Medical Sciences, University College London, London, UK; 2. Department of Primary Care and Population Health, University College London, London, UK.

Corresponding Author: Ioanna Pilati, UCL Division of Medicine, Faculty of Medical Sciences, University College London, London, UK, i.pilati213@gmail.com

J Frailty Aging 2021;in press
Published online September 30, 2021, http://dx.doi.org/10.14283/jfa.2021.39



Background: Sarcopenic obesity (SO) is a condition combining two important public health issues commonly seen amongst older individuals, obesity and sarcopenia. Depressive symptoms are common among older people, whose population is increasing worldwide. Obesity and sarcopenia alone, are clearly associated with depression while the coexistence of these two conditions (SO) upon depressive disorders is currently unclear. We aimed to systematically review the association between primary SO and depressive disorders.
Methods: Searches were run on MEDLINE, EMBASE, PsycINFO, and CINAHL (inception to June 2019). One reviewer screened titles, abstracts, and full-texts, with 10% checked independently by a second reviewer. Cohort and cross-sectional studies were included. Two reviewers independently assessed risk of bias using the Mixed Methods Appraisal Tool. Results were narratively synthesised.
Results: Out of the 7 studies eligible for inclusion, evidence of sarcopenic obesity as a predictor of depressive symptoms was found in two studies. The main observed trend was that diagnosing sarcopenia using muscle strength led to significant associations between sarcopenic obesity and depressive symptoms. Two cross-sectional studies found a significant association between SO and depressive symptoms, whilst three others found no statistically significant associations. All possessed some methodological limitations.
Discussion: This is the first review to systematically examine a potential relationship between sarcopenic obesity and depressive disorders. Currently, the results are heterogeneous due to the large variability in assessment methods and outcome measurements. Future longitudinal studies would achieve greater confidence in the provisional conclusion that sarcopenic obesity, when measured using muscle strength, is associated with depressive symptoms.

Key words: Sarcopenic obesity, sarcopenia, obesity, depression, ageing, older adults.



In 2015, it was estimated that 39% of the world’s population were overweight or obese (1) and this is progressively increasing. Obesity is associated with a range of adverse outcomes, including a greater risk of diabetes, cardiovascular disease and mortality (2). Among obese populations, depression’s prevalence is elevated and a bidirectional effect is usually implied in studies. It has been observed that in extreme weight measurements (underweight and obese), the prevalence of depression is higher, by 23% among obese participants (3). Concurrently, depressed individuals are significantly more likely to develop obesity than those who are not depressed (4). Additionally, obesity increases the risk of depression (5). A meta-analysis including cross-sectional studies revealed a strong positive association between obesity and depression in the general population, confirming the previous findings (6).
Approximately 20% of people aged over 60 experience a mental disorder while amongst them 7% are diagnosed with depression (7). Depression is associated with greater functional decline, poorer quality of life and increased use of healthcare services. Common characteristics include feelings of sadness, emptiness, irritability and mood dysregulations along with cognitive and physical impairments which disrupt the person’s functionality (8). It has also been argued that additional changes in physiological function that should be measured such as sleep disturbances, loss of appetite, fatigue, hopelessness and cognitive impairments should be taken into account when diagnosing depression in older people (9).
There is a global increase in the older population and by the year 2050, the proportion of the world’s population aged over 65 years old will reach 22%, compared to the 12% documented in 2015 (7). Ageing is accompanied by losses in muscle mass and muscle strength leading to ‘primary sarcopenia’ (10). There are different working groups on sarcopenia (European Working Group on Sarcopenia in Older People (EWGSOP), the Asian Working Group (AWGS), International Working Group on Sarcopenia) which all exhibit differences in standards and cut-off points for defining sarcopenia. The Asian Working Group (AWGS) considers muscle mass as a primary indicator but uses different cut-off values due to morphological differences between the Caucasians and Asians (11). The International Working Group on Sarcopenia based the diagnosis on low fat-free mass combined with inadequate physical functioning (12). A recent meta-analysis concluded that sarcopenia is positively associated with depression rates among the elderly (13). Whilst sarcopenia is recognised as a disease primarily affecting older people, it may also be a comorbidity of a health condition, which is defined as secondary sarcopenia (14). Chronic inflammation from a health condition is paired with an increased circulation of pro-inflammatory cytokines which shift the balance of protein turnover promoting muscles’ catabolism and thus, secondary sarcopenia (15).
Obesity can exacerbate sarcopenia (10) through further loss of muscle mass due to its infiltration by the adipose tissue (16). Sarcopenic obesity (SO) is therefore a combination of excessive adipose tissue and decreased skeletal muscle mass and strength (17). Factors involved in the pathogenesis of sarcopenic obesity include ageing, sedentary lifestyle, increased energy intake, insulin resistance, inflammation, and oxidative stress (18).
A plethora of complications are attributed to sarcopenic obesity. In a meta-analysis of prospective studies, there was a 24% increase in the mortality rate attributed to all-cause mortality in sarcopenic obese adults compared to the non-sarcopenic obese adults (19). The disability burden is exacerbated; weakness due to sarcopenia is combined with the need to support added weight from obesity (20). Reduced muscle strength increases the risk of falling which in combination with age-related declines in bone density, poses a major risk of fractures (21). There is a negative correlation between body mass index (BMI) and quality of life, which increases in the presence of sarcopenia (22). There are various ways in which SO appears to influence the quality of life, including reducing exercise ability, daily activity and self- care (23).
In a longitudinal study, low grip strength, used as the measurement for the diagnosis of sarcopenia, was associated with depressive symptoms and interestingly, the relationship was present only between those participants classified as obese (24), posing an intriguing scenario of a potential synergistic effect of increased fat mass and sarcopenia on the onset of depression. There is some evidence stemming from individual studies suggesting that increased fat mass and decreased muscle mass, alone, are associated with a worse psychological condition (21).
Although previous reviews have assessed the relationship between sarcopenia and depression (13), no previous review has assessed the relationship between sarcopenic obesity and depression. Therefore, the main aim of this review is to investigate the association between sarcopenic obesity and depression. In parallel, it may trigger future research on a possible ability of sarcopenic obesity to predict the development of depressive disorders. The review focuses on primary sarcopenic obesity as secondary sarcopenic obesity may have other confounding factors in relation to the health condition itself and depression.



We carried out a systematic review of observational studies, assessing the association between sarcopenic obesity (SO) and depression.

Data sources and search strategy

We searched MEDLINE, EMBASE, PsycINFO, and CINAHL (inception to June 2019), using a combination of subject heading and free text terms relating to obesity (including “high fat mass”, “adiposity”, “increased BMI”), sarcopenia (including “reduced muscle mass”, “reduced muscle strength”, “muscle atrophy”), and depressive disorders (including “depressive symptoms”, “low mood”, “dysthymia”). All relevant papers written in any language were included to avoid language bias. We did not search for or include grey literature as this was felt to be an unlikely source of cohort and cross-sectional studies.

Study selection

Titles and abstracts were screened by one reviewer (IP) for eligibility, with 10% checked by a second reviewer (RF). Agreement was 97.46%. Full texts were screened by one reviewer (IP), and two additional reviewers (RF and AS) checked those assessed as eligible for inclusion and those considered unclear.
Inclusion criteria: cohort and cross-sectional studies; adults (>18 years of age, as although typically age-related declines in muscle strength and function are only seen after age 50, there are individual variations affecting this (25)), with primary (age-related) sarcopenia and co-existing obesity, located either in the community or in a specific care setting; assessing depression as an outcome using a questionnaire or self-reported or clinician diagnosis, including major depression or depressive symptomatology. As the European Working Group definition is relatively recent and is not used worldwide, in order to find the full range of relevant studies, primary sarcopenia in this review could be measured as muscle mass, muscle strength or both (although currently, low handgrip strength is the main determinant for the diagnosis of the condition). As the relationship between factors can be bi-directional, it was also regarded as appropriate to include and discuss studies assessing obesity status and depressive symptoms as predictors for a measured decline in muscle strength (sarcopenia), to investigate the direction of relationships between these variables.
Exclusion criteria: Studies only assessing outcomes such as quality of life, wellbeing or other psychiatric conditions were excluded. Studies focusing on patients with a particular health condition (e.g. cancer, diabetes) were excluded, as features of individual diseases may independently influence the risk of depression.

Quality assessment

The risk of bias within eligible studies was assessed with section 3 of the Mixed Methods Appraisal Tool (MMAT) (26). A second reviewer (RF) independently rated each study as well, with disagreements resolved through email discussion.

Data extraction and data synthesis

A data extraction form was designed to extract all relevant and necessary data including each study’s design, the number and age of participants, the diagnostic tools used to assess sarcopenic obesity and outcome measured (major depression, depressive symptomatology, onset of depressive symptoms, etc.) and tool(s) used to assess it.
Meta-analysis was intended to be undertaken where possible. However, in many cases, this was precluded due to heterogeneity in methods and in measures used between studies. In two studies which were eligible for meta-analysis, there was substantial statistical heterogeneity (I2 = 86%) due to large variation in sample sizes, and so we have not reported this in our results. The remaining studies were tabulated and narratively synthesised. Due to the wide diversity in definitions and measures used for both predictors and outcomes, studies are critically compared and contrasted, with considerations for quality and methodological approaches.



Out of the 757 initial citations identified, of which 556 remained after removing duplicate data, the final screening identified seven studies fulfilling the inclusion criteria (Figure 1.). Out of the nineteen studies which were screened full- text, twelve were excluded. A common justification for exclusion was that some of the studies were not assessing the predictor of interest, which was the co-existence of sarcopenia and obesity defined as sarcopenic obesity (n=6), or they did not measure for the outcome of interest, which was depressive symptoms (n=2), or both (n=2). Two studies were excluded due to their design (they were reviews).

Figure 1. PRISMA Flow Diagram (37)


Of the seven included studies, four were cross- sectional, two used a longitudinal design while one of them (27) incorporated both design methods. Five assessed the relationship between sarcopenic obesity and depression, and two assessed whether obesity and depressive symptoms predicted declines on handgrip strength. Two were carried out in the USA, one in Japan, one in England, two in Korea and one in the Netherlands. Participants’ mean ages varied from 43.3 years to 77.1 years, and sample sizes ranged from 506 to 11521.
Sarcopenic obesity was assessed with a combination of measuring sarcopenia and obesity. Obesity was measured either using BMI, (24, 28, 29, 30), waist circumference (23, 27) or body fat percentage (31); whilst sarcopenia was assessed using handgrip strength in five studies (24, 27, 28, 30, 31) and appendicular skeletal mass in two studies (23, 29).
Depressive symptoms were measured using: a positive response to the question ‘In the past year, have you felt sadness or despair continuously for two or more weeks that was severe enough to interfere with daily life?’ alone (23) or also asking for self-reported diagnosis (29), the Centre of Epidemiological Studies Depression Scale (24, 32) or a shorter adaptation (28, 33); 15- item self-reported Geriatric Depression Scale (GDS), (31) validated for the Japanese population (34); the Patient Health Questionnaire (PHQ-9) (30), which is 9-item depression screener (35); or the Composite International Diagnostic Interview (CIDI version 2.1) (36) to assess the presence of depression or dysthymia according to the (DSM-IV-TR) (27).

Risk of bias within included studies

Table 1, summarises the quality assessment process.
Overall, the majority of studies met most quality criteria. The purpose of this study was to detect adults with age-related sarcopenia so the population included in each study was judged as appropriately selected. Since there is no consensus respecting the definition of sarcopenic obesity no strict judgment could be made regarding the appropriateness of the measurement of the predictor, so this was divided into appropriateness of sarcopenia measure and appropriateness of obesity measure. Moreover, given the small amount of literature available, it was thought as most appropriate to be more lenient in quality assessment regarding variability between the studies’ definitions.
This fact, along with the differences in population characteristics across the studies (different ethnicities, races, ages included) increases the heterogeneity. All studies adjusted their models for covariates (demographic, socioeconomic or clinical factors), but different kinds and numbers were used across them.

Table 1. Results of the methodological quality assessment


Longitudinal associations between sarcopenic obesity and depression

Within the two longitudinal studies (Table 2.), there was high quality evidence that sarcopenic obesity was associated with higher risk of depressive symptoms at 6 years follow up, which weakened slightly but was still significant after adjusting for covariates (adjusted OR 1.79 (1.10, 2.89), reference group non-obese, non-sarcopenic adults) (24). In a smaller depressed Dutch cohort, there was a significant interaction between sarcopenia and obesity (when measured continuously as low handgrip strength and waist circumference) in predicting non-remission of depressive symptoms after 2 years; in the sarcopenic group an increase in waist circumference was associated with an adjusted OR 1.06 [1.01 – 1.11] (27). When assessed dichotomously, this association was not found (OR=1.54 [95% CI: 0.75 – 3.16], p=.241).

Table 2. Summary of results from longitudinal studies. Mean (SD) is reported for age


Cross-sectional associations between sarcopenic obesity and depression

Evidence for cross-sectional associations was conflicting (Table 3). Associations were mostly observed in studies of better quality. Sarcopenic obesity was significantly associated with an increased risk of depressive symptoms in one high quality large Japanese study in older adults (adjusted OR 2.79 (1.43–5.43), p=0.003, compared to the reference group of non-obese, non-sarcopenic adults) (31). In the adjusted model, there was also a significant interaction between sarcopenia and obesity as two independent variables. However, the increased risk of depressive symptoms was seen most strongly in those aged 65-74 years old (adjusted OR 6.05 (1.89–19.38), p=0.003) and became non-significant in those aged >75years old (adjusted OR (1.77 (0.75–4.18), p=0.2) (31).

Table 3. Summary of results of the selected studies with a cross-sectional design. Mean (SD) is reported for age


Within the Korean National Health Study, sarcopenic obesity was examined in relation to the umbrella term of “psychological health”, separately in two age groups; >=60 years old and those of a younger age. Focusing on subgroup analysis, associations with depressive symptoms, suicidal ideation and perceived stress for those aged >=60 were non- significant (adjusted OR 0.95 (CI 0.66-1.38)) compared to the younger age group who exhibited significant associations with both perceived stress and suicidal ideation, but not depressive symptoms. Interestingly, in the <60 age group associations of SO with perceived stress were stronger than in the general obesity group, implying an important additional effect of sarcopenia on obesity (23).
Two other studies did not find an association. In one large Korean study, a self-reported depression diagnosis and depressive symptoms had a higher prevalence within sarcopenic obese participants, but this was not statistically significant in an unadjusted Chi-squared test across any age group (29). Similarly, the NESDO found no cross-sectional association between sarcopenic obesity and depression (adjusted OR 1.09 [0.54 – 2.23]), whether measured continuously or dichotomously (27). However, NESDO recruited participants on the basis of having depression, in addition to a smaller number of healthier controls, and the representativeness of this cohort is questionable.

Studies of obesity and depression as a predictor of sarcopenia

In a moderate quality longitudinal study (28), older adults with the coexistence of an increased BMI and depressive mood did not exhibit significant declines in handgrip strength compared to the reference group of those with increased BMI and no depressive symptoms (adjusted OR not reported, no difference when adjusted for antidepressant use). However, within this study, an overweight BMI cutoff (>25kg/m2) rather than an obese cutoff, was used and participant’s age range may be considered high (71-92 years). In another high-quality cross-sectional study, among obese adults aged >60, both men (adjusted OR −3.72 (−7.00 to −0.43)) and women (adjusted OR −1.83 (−2.87 to −0.78)) with moderate to severe depressive symptoms had lower handgrip strength compared to those with no depressive symptomatology (30). Associations were not significant for those with mild depressive symptoms (Table 4.).


Table 4. Summary of results of obesity and depression as a predictor of sarcopenia



The original motive of the present review was the examination of the effect of sarcopenic obesity on depressive disorders. The included studies found limited evidence that sarcopenic obesity increases the risk of depressive symptoms and non-remission of depression. Cross-sectional studies were heterogenous, but suggested associations may exist in studies where handgrip strength is used as the primary definition for sarcopenia, with stronger associations in younger populations. These heterogeneous results may be explained in a number of ways.
Firstly, studies using muscle strength to define sarcopenia were more likely to report significant associations with depressive symptoms. The two studies using muscle mass to define sarcopenia without considering muscle strength did not find an association. The revised guidelines for sarcopenia (10), suggest that handgrip strength is a better indicator of muscle function and is preferred over measures of muscle mass alone, although this derives from a European Working Group, and other groups (e.g. the Asian Working Group) may use measures more relevant to their populations. Although low muscle mass explains 13% of the variance in muscle strength (38), other parameters such as age and fat mass are equally recognised as crucial determinants (39). As obesity can incorporate an increase in muscle as well as fat mass, in obese individuals quantity of muscle mass is commonly found to be normal while its quality is insufficient (40). While in normal-weight individuals mood is not correlated with muscle mass, studies using hand-grip strength as a useful measurement of general muscle function, report a significant association between low hand- grip strength and depressive symptomatology (41, 42). In obese individuals, anxiety and depression levels are influenced by lean mass reductions paired with its infiltration by the adipose tissue (43). Other studies have also concluded that it is only in the presence of obesity when sarcopenia is related to some health complications (44, 45), nominating muscle tissue’s condition as a reason for variations in health between obese and non-obese individuals. Another study suggested a synergistic effect of low grip strength and depression on all-cause mortality in older adults (46).
Another explanation may be the obesity definitions used. In an earlier study examining the relationship between body composition and depression amongst older people no association between central obesity and depression can be found (47). At the same time, what is already known is that abdominal fat is more pathogenic than subcutaneous. In a large studied sample from NHANES, participants classified as overweight using BMI, with an increased waist circumference, had a greater prevalence of depression (48). On the current review, studies using waist circumference (23, 27), were less likely to find a cross sectional association than those using BMI as a diagnostic criterion.
Similarly, use of lower obesity cutoffs for BMI (25 rather than 30kg/m2), may also impact upon outcomes. The cut off point for obesity diagnosis in Caucasians is 30 kg/m2, while for Asian populations BMI should be 27.5 kg/m2 or higher (49, 50). The earliest study included found no association between elevated BMI (>25kg/m2) and depression based on baseline measurements (28). This is in contrast with current knowledge since the relationship between depression and obesity is confirmed by a plethora of evidence (4, 5). A meta-analysis including cross-sectional studies revealed a strong positive association between obesity and depression in the general population confirming the previous findings (6). In those papers finding a significant association, these were much stronger in younger than older populations, which may be related to better mortality outcomes in obese and overweight older adults (51). One paper included in the current review found a significant association between depressive symptoms and SO in adults <65years old (23). This could be an insight suggesting that it is equally important for younger obese adults to be screened for primary sarcopenia since deteriorations on muscle mass may vary depending on general health and physical condition.
Measurement of depressive symptoms may also influence results. Self-reported mental health is a methodologically vulnerable tool since the way people perceive their health is controversial and could be easily influenced by many factors or even vary from day to day and rely on experiences during the past days (52). Studies using a validated questionnaire were more likely to report associations with sarcopenic obesity than those using a non-validated single question (23, 29). This may relate to the broadness of the population captured – one study (30) found that an association was only present in those with more severe symptoms, although one study using self-reported depression diagnoses found no association (28). No studies measured the association between sarcopenic obesity and clinician-diagnosed Major Depressive Disorder alone.

Overall completeness, applicability, and quality of evidence

There was a limited number of papers eligible for inclusion in this review, suggesting conclusions may be open to change in the future. Most of these were conducted recently, reflecting that sarcopenic obesity is a condition which is gaining attention and being more explicitly defined and understood. The results of included studies are supported by the existing knowledge of biological and pathophysiological mechanisms and previous work on the field, but currently are very heterogeneous in the measures used. However, findings do stem from different countries all over the world.

Strengths and limitations

To the best of our knowledge, up to date, this is the first systematic review investigating the association between sarcopenic obesity and depressive disorders. Two reviewers assessed the quality of included studies.
However, limitations exist in the heterogeneity of measurement methods used. Although currently sarcopenic obesity is identified as a condition on its own, its assessment is still based on individual measurements of obesity and sarcopenia, with no universal definition for combining the two. Some of the papers adjusted the results for some comorbidities but possibly, results cannot be generalised to populations with greater comorbidity. Only a few studies adjusted for anti-depressant use, which may be a strong predictor of depressive symptoms. Another limitation is the observational design of the studies which restricts the ability to draw conclusions regarding causal relationships. Likewise, many of the included studies were cross-sectional, limiting conclusions that can be drawn regarding the direction of the relationship.

Implications for practice and policy

The evidence suggests that there may be a link between sarcopenic obesity and depressive symptoms, but that this is likely to be stronger in younger populations and when sarcopenia is defined using handgrip strength. There is some evidence for an interaction between sarcopenia and obesity, and that the link between sarcopenic obesity and depression may be bi-directional, but this requires further investigation. Health professionals should be aware that for those who are obese, low muscle strength may further increase the risk of depressive symptoms in addition to other health issues. In that case, they should be screened regularly for depressive symptoms so that the effects of sarcopenic obesity may be combated with appropriate treatment strategies. Current evidence suggests that the most effective strategies are those incorporating hypocaloric diet (no more than 200-700 Kcals deficit to minimize muscle mass loss) along with aerobic training for fat loss, resistance training and increased protein intake to preserve the quality and quantity of muscles (53). The type of protein is important which should be high quality, evenly distributed through a day aiming to achieve an intake every 3 to 4 hours (53).

Implications for future research

Further longitudinal studies need to be undertaken to provide definitive evidence for a relationship between sarcopenic obesity and depressive symptoms. There needs to be an international consensus regarding the tools, measurements and cut-off points used in the diagnosis of sarcopenic obesity. The absence of a clear, universally established definition introduces a big variation in the rates of sarcopenic obesity depending on the used criteria (54), however this may be challenging given the necessary differences in cut-offs between different populations. More attention should be drawn towards the importance of the in-depth study of the condition and recognition of potentially modifiable risk factors could trigger prevention strategies on a population basis. In a society in which 39% of the population are overweight, sarcopenic obesity and its resulting effects is likely to become a larger issue in future.
Furthermore, so far there is not an identified, accepted mechanism by which sarcopenic obesity manifests the disturbances in depressive disorders. Gaining an understanding of the underlying pathophysiological mechanisms would help design appropriate tackle strategies.

Author’s conclusions

This review found limited evidence that sarcopenic obesity increases the risk of depressive symptoms, with some conflicting findings. Reductions in obesity and sarcopenia levels may have the potential to reduce the risk of depressive symptoms in addition to improving functioning and reducing the risk of comorbidities such as disability, stress and anxiety disorders, metabolic impairments such as insulin resistance and quality of life in general.
Funding statement: The authors received no financial support for the authorship of this article.


Conflict of Interest: The authors declare that they have no conflict of interest.





1. Chooi, Y. C., Ding, C., Magkos, F. The epidemiology of obesity. Metabolism, 2018;92, pp. 6-10. https://doi.org/10.1016/j.metabol.2018.09.005
2. Djalalinia, S., Qorbani, M., Peykari, N., & Kelishadi, R. Health impacts of Obesity. Pakistan journal of medical sciences, 2015;31(1), pp. 239–242. https://doi.org/10.12669/pjms.311.7033
3. Carey, M., Small, H., Yoong, S. L., Boyes, A., Bisquera, A., & Sanson-Fisher, R. Prevalence of comorbid depression and obesity in general practice: a cross-sectional survey. The British journal of general practice: the journal of the Royal College of General Practitioners, 2014;64(620), e122–e127. https://doi.org/10.3399/bjgp14X677482
4. Blaine, B. Does Depression Cause Obesity? A Meta-analysis of Longitudinal Studies of Depression and Weight Control. Journal of Health Psychology, 2008;13(8), pp. 1190–1197. https://doi.org/10.1177/1359105308095977.
5. Luppino, F.S., de Wit, L.M., Bouvy, P.F., Stijnen, T., Cuijpers, P., Penninx, B.W., Zitman, F.G. Overweight, Obesity, and Depression: A Systematic Review and Meta-analysis of Longitudinal Studies. Arch Gen Psychiatry, 2010;67(3), pp. 220–229. https://doi.org/10.1001/archgenpsychiatry.2010.2
6. Wit, L., Luppino, F., Straten, A.,Penninx, B.W., Zitman, F., Cuijpers, P. Depression and Obesity: A Meta-Analysis of Community-Based Studies. Psychiatry Res, 2010;178(2), pp. 230-5. https://doi.org/10.1016/j.psychres.2009.04.015
7. World Health Organization, 2017. Mental health of older adults [Fact sheet]. Available at: https://www.who.int/news-room/fact-sheets/detail/mental-health-of-older-adults
8. American Psychiatric Association, 2013. Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596
9. Luppa, M., Sikorski, C., Luck, T., Ehreke, L., Konnopka, A., Wiese, B., Weyerer, S., König, H.H., Riedel-Heller, S.G. Age- and gender-specific prevalence of depression in latest-life – Systematic review and meta-analysis. Journal of Affective Disorders, 2012;136(3), 212–221. https://doi.org/10.1016/j.jad.2010.11.033
10. Cruz-Jentoft, A. J., Bahat, G., Bauer, J., Boirie, Y., Bruyère, O., Cederholm, T., Cooper, C., Landi, F., Rolland, Y., Aihie Sayer, A., Schneider, S., C Sieber, C., Topinková, E., Vandewoude, M., Visser, M., Zamboni, M. Sarcopenia: revised European consensus on definition and diagnosis. Age and Ageing. 2019;48(1), pp.16-31. https://doi.org/10.1093/ageing/afy169
11. Chen, L.-K., Lee, W.-J., Peng, L.-N., Liu, L.-K., Arai, H., & Akishita, M. Recent Advances in Sarcopenia Research in Asia: 2016 Update From the Asian Working Group for Sarcopenia. Journal of the American Medical Directors Association, 2016;17(8), 767.e1–767.e7. https://doi.org/10.1016/j.jamda.2016.05.016
12. Fielding, R., Vellas, B., Evans, W., et al. Sarcopenia: an undiagnosed condition in older adults. Current consensus definition: prevalence, aetiology, and consequences. J Am Med Dir Assoc., 2011;12(4), pp.249–56. https://doi.org/10.1016/j.jamda.2011.01.003
13. Chang, K.V., Hsu, T.H., Wu, W.Y., Huang, K.C., Han, DS. Is sarcopenia associated with depression? A systematic review and meta-analysis of observational studies. Age and Ageing, 2017;46(5), pp. 738–746. https://doi.org/10.1093/ageing/afx094
14. Fan, J., Kou, X., Yang, Y., Chen, N. MicroRNA-Regulated Proinflammatory Cytokines in Sarcopenia. Mediators of inflammation, 2016, 1438686. https://doi.org/10.1155/2016/1438686
15. Pérez-Baos, S., Prieto-Potin, I., Román-Blas, J. A., Sánchez-Pernaute, O., Largo, R., Herrero-Beaumont, G. Mediators and Patterns of Muscle Loss in Chronic Systemic Inflammation. Frontiers in physiology; 2018;9, pp. 409. https://doi.org/10.3389/fphys.2018.00409

16. Kalinkovich, A., Livshits, G. Sarcopenic obesity or obese sarcopenia: A cross talk between age-associated adipose tissue and skeletal muscle inflammation as a main mechanism of the pathogenesis. Ageing Res Rev, 2016;35, pp. 200–21. https://doi.org/10.1016/j.arr.2016.09.008
17. Yu, S., Umapathysivam, K., Visvanathan, R. Sarcopenia in older people. International Journal of Evidence-Based Healthcare, 2014;12(4), 227–243. https://doi.org/10.1097/xeb.0000000000000018
18. Polyzos, S. A., Margioris, A. N. Sarcopenic obesity. Hormones, 2018;17(3), pp. 321-331. https://doi.org/10.1007/s42000-018-0049-x
19. Tian, S., Xu, Y. Association of sarcopenic obesity with the risk of all-cause mortality: A meta-analysis of prospective cohort studies. Geriatrics & Gerontology International, 2015;16(2), 155–166. https://doi.org/10.1111/ggi.12579
20. Roubenoff, R. Sarcopenic Obesity: The Confluence of Two Epidemics. Obesity Research, 2004;12(6), pp. 887-8. https://doi.org/10.1038/oby.2004.107
21. Batsis, J. A., Villareal, D. T. Sarcopenic obesity in older adults: aetiology, epidemiology and treatment strategies. Nature reviews. Endocrinology, 2018;14(9), pp. 513–537. https://doi.org/10.1038/s41574-018-0062-9
22. Ul-Haq Z., Mackay D.F., Fenwick E. & Pell J.P. Meta-analysis of the association between body mass index and health-related quality of life among adults, assessed by the SF-36. Obesity (Silver Spring), 2013;21(3), E322–E327. https://doi.org/10.1002/oby.20107
23. Cho, Y., Shin, S.Y, Shin, M.J. Sarcopenic obesity is associated with lower indicators of psychological health and quality of life in Koreans. Nutr. Res., 2015;35(5), pp. 384-92. https://doi.org/10.1016/j.nutres.2015.04.002
24. Hamer, M., Batty, G. D., & Kivimaki, M. Sarcopenic obesity and risk of new onset depressive symptoms in older adults: English Longitudinal Study of Ageing. International journal of obesity (2005), 2015;39(12), pp. 1717–1720. https://doi.org/10.1038/ijo.2015.124
25. Keller, K., Engelhardt, M. Strength and muscle mass loss with aging process. Age and strength loss. Muscles, ligaments and tendons journal, 2014;3(4), pp. 346–350. PMID: 24596700; PMCID: PMC3940510.
26. Hong, Q. N., FÀBregues, S., Bartlett, G., Boardman, F., Cargo, M., Dagenais, P., Gagnong, M., Griffithsd , F., Nicolauh, B., O’Cathaini, A., Rousseauj, M., Vedela, I.,Pluye, P. The Mixed Methods Appraisal Tool (MMAT) version 2018 for information professionals and researchers. Education for Information, 2018;1–7. https://doi.org/10.3233/efi-180221
27. Kokkeler, K. J., van den Berg, K.S., Comijs, H.C., Oude Voshaar, R.C., Marijnissen, R.M. Sarcopenic obesity predicts nonremission of late-life depression. International Journal of Geriatric Psychiatry, 2019;34(8), pp. 1226-1234. https://doi.org/10.1002/gps.5121
28. Rantanen, T., Penninx, B.W., Masaki, K., Lintunen, T., Foley, D., Guralnik, J.M. Depressed mood and body mass index as predictors of muscle strength decline in old men. Journal of the American Geriatrics Society, 2000;48(6), pp. 613-617. https://doi.org/10.1111/j.1532-5415.2000.tb04717.x
29. Byeon, C. H., Kang, K.Y., Kang, S.H. Kim, H. K., Bae, E. J. Sarcopenia Is Not Associated with Depression in Korean Adults: Results from the 2010-2011 Korean National Health and Nutrition Examination Survey. Korean Journal of Family Medicine, 2016;37(1), pp. 37-43. https://doi.org/10.4082/kjfm.2016.37.1.37
30. Smith, L., White, S., Stubbs, B., Hu., L., Veronese, N., Vancampfort, D., Hamer, M., Gardner, B., Yang, L. Depressive symptoms, handgrip strength, and weight status in US older adults. Journal of Affective Disorders, 2018;238, pp. 305-310. https://doi.org/10.1016/j.jad.2018.06.016
31. Ishii S, Chang C, Tanaka T, Kuroda A, Tsuji T, Akishita M, Iijima, K. The Association between Sarcopenic Obesity and Depressive Symptoms in Older Japanese Adults. PLoS ONE, 2016;11(9): e0162898. https://doi.org/10.1371/journal.pone.0162898
32. Irwin, M., Artin, K.H., Oxman, M.N. Screening for depression in the older adult: criterion validity of the 10-item Center for Epidemiological Studies Depression Scale (CES-D). Arch Intern Med, 1999;159(15), pp. 1701–1704. https://doi.org/10.1001/archinte.159.15.1701
33. Radloff, L.S. The CES-D scale. A self-report depression scale for research in the general population. Appl Psycho1 Meas, 1977;1, pp. 385-401. https://doi.org/10.1177/014662167700100306
34. Schreiner, A.S., Hayakawa, H., Morimoto, T., Kakuma, T. Screening for late life depression: cut-off scores for the Geriatric Depression Scale and the Cornell Scale for Depression in Dementia among Japanese subjects. Int J Geriatr Psychiatry, 2003;18(6), pp. 498–505. https://doi.org/10.1002/gps.880
35. Kroenke, K., Spitzer, R.L., Williams, J.B. The PHQ-9: validity of a brief depression severity measure. J. Gen. Intern. Med., 2001;16(9), pp. 606–613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x
36. Wittchen, H.U., Robins, L.N., Cottler, L.B., Sartorius, N., Burke, J.D., Regier, D. Cross-cultural feasibility, reliability and sources of variance of the Composite International Diagnostic Interview (CIDI). The Multicenter WHO/ADAMHA Field Trials. Br J Psychiatry, 1991;159, pp. 645-653. https://doi.org/10.1192/bjp.159.5.645
37. Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gøtzsche, P.C., Ioannidis, J., Clarke, M., Devereaux, P. J., Kleijnen, J., Moher, D. The PRISMA statement for reporting systematic and meta-analyses of studies that evaluate interventions: explanation and elaboration. PLoS Medicine, 2009;6(7):e1000100. https://doi.org/10.1371/journal.pmed.1000100
38. Chen, L., Nelson, D. R., Zhao, Y., Cui, Z., & Johnston, J. A. Relationship between muscle mass and muscle strength, and the impact of comorbidities: a population-based, cross-sectional study of older adults in the United States. BMC Geriatrics, 2013;13(1). https://doi.org/10.1186/1471-2318-13-74
39. Newman, A., Haggerty, C.L., Goodpaster, B., Harris, T., Kritchevsky, S., Nevitt, M., Miles, T.P., Visser, M.; Health Aging And Body Composition Research Group. Strength and muscle quality in a cohort of well-functioning older adults: The Health Aging and Body Composition Study. J Am Geriatr Soc., 2003;51(3), pp. 323-330. https://doi.org/10.1046/j.1532-5415.2003.51105.x
40. De Simone, G., Pasanisi, F., Ferrara, A. L., Roman, M. J., Lee, E. T., Contaldo, F., Howard, B., Devereux, R. B. Relative fat-free mass deficiency and left ventricular adaptation to obesity: The Strong Heart Study. International Journal of Cardiology, 2013;168(2), 729–733. https://doi.org/10.1016/j.ijcard.2012.09.055
41. Fukumori, N., Yamamoto, Y., Takegami, M., Yamazaki, S., Onishi, Y., Sekiguchi, M., Otani, K., Konno S., Kikuchi S., Fukuhara, S. Association between hand-grip strength and depressive symptoms: Locomotive Syndrome and Health Outcomes in Aizu Cohort Study (LOHAS). Age and Ageing, 2015;44(4), 592–598. https://doi.org/10.1093/ageing/afv013
42. Ashdown-Franks, G., Stubbs, B., Koyanagi, A., Schuch, F., Firth, J., Veronese, N., & Vancampfort, D. Handgrip strength and depression among 34,129 adults aged 50 years and older in six low- and middle-income countries. Journal of Affective Disorders, 2018. https://doi.org/10.1016/j.jad.2018.09.036
43. Cugini, P., Cilli, M., Salandri, A., Ceccotti, P., Di Marzo, A., Rodio, A., Fontana, S., Pellegrino, A.M., De Francesco, G.P., Coda, S., De Vito, F., Colosi, L., Petrangeli, C.M., Giovannini, C. Anxiety, depression, hunger and body composition: III. Their relationships in obese patients. Obesity, 1999;4(3), pp. 115-120. https://doi.org/10.1007/bf03339726
44. Rolland, Y., Lauwers-Cances, V., Cristini, C., Abellan van Kan, G., Janssen, I., Morley, J., Vellas, B. Difficulties with physical function associated with obesity, sarcopenia, and sarcopenic-obesity in community-dwelling elderly women: the EPIDOS (EPIDemiologie de l’OSteoporose) Study, The American Journal of Clinical Nutrition, 2009;89(6); pp. 1895–1900. https://doi.org/10.3945/ajcn.2008.26950
45. Tolea, M.I., Chrisphonte, S., Galvin, J.E. Sarcopenic obesity and cognitive performance. Clin Interv Aging; 2018;13, pp. 1111–1119. https://doi.org/10.2147/CIA.S164113
46. Park, S., Cho, J., Kim, D. et al. Handgrip strength, depression, and all-cause mortality in Korean older adults. BMC Geriatr 2019;19, 127. https://doi.org/10.1186/s12877-019-1140-0
47. Kim, N.H., Kim, H.S., Eun, C.R., Seo, J.A., Cho, H.J., Kim, S.G., Choi, K.M., Baik, S.H., Choi, D.S., Park, M.H., Han, C., Kim, N.H. Depression is associated with sarcopenia, not central obesity, in elderly Korean men. J Am Geriatr Soc, 2011;59(11), pp. 2062–8. https://doi.org/10.1111/j.1532-5415.2011.03664.x
48. Zhao, G., Ford, E. S., Li, C., Tsai, J., Dhingra, S., & Balluz, L. S. Waist circumference, abdominal obesity, and depression among overweight and obese U.S. adults: national health and nutrition examination survey 2005-2006. BMC Psychiatry, 2011;11(1). https://doi.org/10.1186/1471-244x-11-130
49. Choo, V. WHO reassesses appropriate body-mass index for Asian populations. The Lancet, 2002;360(9328), 235. https://doi.org/10.1016/s0140-6736(02)09512-0
50. Stegenga, H., Haines, A., Jones, K., Wilding, J. Guideline Development Group. Identification, assessment, and management of overweight and obesity: summary of updated NICE guidance. BMJ, 2014;27;349:g6608. https://doi.org/10.1136/bmj.g6608
51. McGee, D. L. Body mass index and mortality: a meta-analysis based on person-level data from twenty-six observational studies. Annals of Epidemiology; 2005;15(2), pp. 87–97. https://doi.org/10.1016/j.annepidem.2004.05.012
52. Northrup, D.A. The Problem of the Self- Report in Survey Research. Institute for Social Research. 1996;11(3).
53. Trouwborst, I., Verreijen, A., Memelink, R., Massanet, P., Boirie, Y., Weijs, P., & Tieland, M. Exercise and Nutrition Strategies to Counteract Sarcopenic Obesity. Nutrients; 2018;10(5), pp. 605. https://doi.org/10.3390/nu10050605
54. De Rosa, E., Santarpia, L., Marra, M., Sammarco, R., Amato, V., Onufrio, M.,De Simone, G.,Pasanisi, F. Preliminary evaluation of the prevalence of sarcopenia in obese patients from Southern Italy. Nutrition, 2015;31(1), 79–83. https://doi.org/10.1016/j.nut.2014.04.025



M. Cesari1, B. Vellas2


1. Geriatric Unit, IRCCS Istituti Clinici Scientifici Maugeri, University of Milan, Milan, Italy; 2. Gerontopole – Inspire Program, UMR INSERM 1295, Toulouse University Hospital, University of Toulouse Paul Sabatier, Toulouse, France

Corresponding Author: Matteo Cesari, MD, PhD. IRCCS Istituti Clinici Scientifici Maugeri; via Camaldoli 64, 20138 Milan – Italy. Email: macesari@gmail.com; Twitter: @macesari

J Frailty Aging 2021;10(4)308-309
Published online September 17, 2021, http://dx.doi.org/10.14283/jfa.2021.37


Key words: Vaccines, geriatrics, frailty, integrated care, social care.


The COVID-19 pandemic has substantially changed our lives. It has also acted as a sort of stress test for care systems, letting emerge all the inconsistencies, weaknesses, and contradictions of them. In particular, frail persons have shown to be those paying the most severe consequences of the general disservices (1). To date, in the absence of specific drugs against the SARS-CoV-2, preventive measures against infection and vaccination represent the only available weapons. Social distancing, hand hygiene, and protective personal equipment have been immediately put in place since the very first phases of the pandemic. Starting in December 2020, vaccines against the SARS-CoV2 infection have been made available and mass campaigns of vaccinations have started worldwide.
Given the high-risk profile exhibited by older persons with frailty, these have been usually prioritized in the vaccination campaigns. Today, the vaccine administration is primarily focused to adults and young persons. Whereas it is generally assumed that the older population has now been vaccinated, a recent survey has suggested that many persons aged 80 years and older (almost 20%) are not yet (2). Indeed, as reported, “statistics from the Centers for Disease Control and Prevention showed this population’s vaccination rates soaring through the spring, then hitting a plateau”.
Which are the barriers precluding the vaccination of so many persons at risk of the most severe consequences of COVID-19? Several reasons can be hypothesized:
– Older persons may refuse the vaccination for personal opinions or because influenced by their proxies. In this context, the presence of cognitive impairment and difficulties in judgment might affect the capacity to decide, relying on what younger persons (potentially less concerned by the severity of the virus and more exposed to fake news) choose for them. In this context, it is important to consider that the many no-vax messages might have, directly and indirectly, influenced the most vulnerable ones (due to their frail status and/or low socio-cultural conditions).
– The frailty status of many older persons can complicate the access to the vaccination. Difficulties in the use of technologies for scheduling an appointment, mobility impairment and/or social isolation hampering the possibility to reach the vaccination site, cognitive disorders affecting the capacity to take and retain the appointment… are all examples of potential underestimated barriers.
– The pandemic has made clear that the hospital-centered design of our healthcare systems is not suitable for many persons living with frailty (3). Their protection implies the adoption of a more comprehensive approach, leaving the traditional standalone disease-model in favor of a holistic vision of the individual inclusive of his/her environment. In this context, it cannot be ignored how most of the vaccination campaigns are centered on hubs in the community where persons can go to receive their vaccine dose. However, relatively low interest has been put for supporting primary care and facilitate the vaccination of the frailest individuals who are home-bound. Indeed, the COVID-19 pandemic has exposed the extreme paucity of resources and infrastructures devoted to older persons where they live and age (i.e., in the community). More research is needed to better understand how many and why older persons are still “lost” to our care systems. This is pivotal to develop future strategies allowing the provision of preventive care in the community to the most vulnerable persons.

Access to care has been extremely difficult for many persons over the past months, not only because of the restrictions and lockdowns applied by governments during the hardest moments of the pandemic. Older persons have specially suffered the fragmentation of care and the prolonged disruptions of services (often motivated by the need of facing the COVID-19 emergency). The procrastination of routine clinical evaluations, often combined with the older person’s fear of being infected, has uphold many interventions that were instead needed (4). Furthermore, the lifestyle modifications forcedly brought by the pandemic have negatively impacted on the health status of the most vulnerable persons, worsening their functions and clinical conditions (5,6). The functional loss and social isolation developed by older persons over the past months will likely result in major consequences in the next future, both in terms of 1) frailer and more complex patients, and 2) incapacity of services to adequately address the increasing demands.
Interestingly, a recent study by Ankuda and colleagues (7) has recently described an exponential increase of community-dwelling older persons who have become home-bound (i.e., leaving the house once a week or less) during these months of pandemics. These persons are exposed to particularly high risk of negative outcomes. Their risk profile is further enhanced by their social isolation preventing them from prompt access to care. A further example is coming from Italy. During the vaccination campaign, almost 500-thousand persons (that is about 1% of the Italian population) were untraceable and difficult to reach. They are socially isolated, tend to live in rural areas, have no internet/phone connection, and/or move frequently across the country. In other words, the COVID-19 pandemic is showing the existence of a population of frail individuals for which a completely different model of care is needed. The usual reactive approach is evidently not working for them, and proactive/preventive strategies are needed.
Under the current COVID-19 situation, we would like to stress the importance of the following points:
1. With the aging of our population and the increasing number of socially isolated individuals, the system cannot anymore just wait for the incoming request. The continuation of this obsolete approach will contribute at accelerating the collapse of the systems which are designed for late interventions. It is necessary to reshape our clinical and public health strategies for anticipating the problems and act when the case is still reversible (for the benefit of the person and the community)(8).
2. Instead of waiting that the problem arrives to the attention of clinical and social services, it is necessary to identify the early signs of future issues to preventively intervene. This means the building of multidisciplinary bridges facilitating the sharing of relevant information across settings for the development of person-centered actions.
3. In this context, it is noteworthy the work conducted by the World Health Organization (WHO) to promote the integration and continuum of care (9). The WHO has repeatedly recommended over the past years to modify the approach to older persons by implementing preventive strategies and personalization of interventions (e.g., ICOPE Program)(10). Every point of contact between the individual and the care system should become an opportunity for estimating the residual reserves (i.e., intrinsic capacity) and abilities (i.e., functional ability) (11). The resulting information may then be used to track his/her trajectories and identify deviations from the normality.
4. Finally, the adoption of shared technologies is not an option anymore. Indeed, in a world dominated by technologies, it is not anymore acceptable that persons are “lost” to the care system. It is time to take advantage of technologies as exemplified by the ICOPE Monitor, an innovative digital healthcare program designed for community-dwelling older persons with frailty with the final aim of remote monitoring their health status (via nurse assistance) and facilitating access to preventive services (including COVID-19 vaccination) (12, 13).



1. Merchant R. COVID-19: role of integrated regional health system towards controlling pandemic in the community, intermediate and long-term care. J Frailty Aging. 2020:1-2. doi:10.14283/jfa.2020.39.
2. Span P. More Than 80 Percent of Seniors Are Vaccinated. That’s ‘Not Safe Enough.’ The New York Times. https://www.nytimes.com/2021/09/02/health/covid-vaccines-seniors.html. Published September 2, 2021. Accessed September 3, 2021.
3. Astrone P, Cesari M. Integrated Care and Geriatrics: A Call to Renovation from the COVID-19 Pandemic. J Frailty Aging. October 2020:1-2. doi:10.14283/jfa.2020.59.
4. Briguglio M, Giorgino R, Dell’Osso B, et al. Consequences for the Elderly After COVID-19 Isolation: FEaR (Frail Elderly amid Restrictions). Front Psychol. 2020;11:565052. doi:10.3389/fpsyg.2020.565052.
5. Kirwan R, McCullough D, Butler T, Perez de Heredia F, Davies IG, Stewart C. Sarcopenia during COVID-19 lockdown restrictions: long-term health effects of short-term muscle loss. Geroscience. 2020;42(6):1547-1578. doi:10.1007/s11357-020-00272-3.
6. Canevelli M, Valletta M, Toccaceli Blasi M, et al. Facing Dementia During the COVID-19 Outbreak. J Am Geriatr Soc. 2020;68(8):1673-1676. doi:10.1111/jgs.16644.
7. Ankuda CK, Leff B, Ritchie CS, Siu AL, Ornstein KA. Association of the COVID-19 Pandemic With the Prevalence of Homebound Older Adults in the United States, 2011-2020. JAMA Internal Medicine. August 2021. doi:10.1001/jamainternmed.2021.4456.
8. Barusch A, Waters D. Social engagement of frail elders. J Frailty Aging. 2012;1(4):189-194.
9. World Health Organization, Department of Ageing and Life Course. Integrated Care for Older People.; 2017. http://www.ncbi.nlm.nih.gov/books/NBK488250/. Accessed March 1, 2019.
10. Integrated Care for Older People (ICOPE): Guidance for Person-Centred Assessment and Pathways in Primary Care. World Health Organization; 2019.
11. World Health Organization (WHO). Decade of Healthy Ageing.; 2020. https://www.youtube.com/watch?v=ShmemfpkVLQ&list=PL1F160112BFDBC1D5&index=2. Accessed April 27, 2021.
12. González-Bautista E, De Souto Barreto P, Virecoulon Giudici K, et al. Frequency of Conditions Associated with Declines in Intrinsic Capacity According to a Screening Tool in the Context of Integrated Care for Older People. J Frailty Aging. August 2020. doi:10.14283/jfa.2020.42.
13. Tavassoli N, Piau A, Berbon C, et al. Framework Implementation of the INSPIRE ICOPE-CARE Program in Collaboration with the World Health Organization (WHO) in the Occitania Region. J Frailty Aging. 2021;10(2):103-109. doi:10.14283/jfa.2020.26.



A.R. Orkaby1,2, A.B. Dufour3, L. Yang3, H.D. Sesso2, J.M. Gaziano1,2, L. Djousse1,2, J.A. Driver1,2, T.G. Travison3


1. New England GRECC, VA Boston Healthcare System, Boston, MA, USA; 2. Brigham & Women’s Hospital and Harvard Medical School, Boston, MA, USA; 3. Marcus Institute for Aging Research, Hebrew Senior Life, and Harvard Medical School, Boston, MA, USA.

Corresponding Author: Ariela Orkaby, MD MPH, Assistant Professor of Medicine, Harvard Medical School, Geriatric Research, Education and Clinical Center, VA Boston Healthcare System, 150 South Huntington St, Boston, MA 02130, USA, aorkaby@bwh.harvard.edu

J Frailty Aging 2021;in press
Published online September 17, 2021, http://dx.doi.org/10.14283/jfa.2021.36



Background: Mobility limitation is a component of frailty that shares a bidirectional relationship with cardiovascular disease (CVD). Data are limited on the role of established CVD prevention therapies, such as aspirin, for prevention of frailty and mobility limitation.
Objectives: Examine the association between long-term aspirin use and walking speed.
Design, Setting, Participants: Prospective cohort of 14,315 men who participated in the Physicians’ Health Study I, a completed randomized controlled trial of aspirin (1982-1988), with extended post-trial follow-up.
Measurements: Annual questionnaires collected data on aspirin use, lifestyle and other factors. Average annual aspirin use was categorized for each participant: ≤60 days/year and >60 days/year. Mobility was defined according to self-reported walking pace, categorized as: don’t walk regularly (reference), easy/casual <2mph, normal ≥2-2.9mph, or brisk/very brisk ≥3mph. Propensity scoring balanced covariates between aspirin categories. Multinomial logistic regression models estimated odds of being in each self-reported walking category.
Results: Mean age was 70±8 years; mean aspirin use 11 years. There were 2,056 (14.3%) participants who reported aspirin use ≤60 days/year. Aspirin use >60 days/year was associated with drinking alcohol, smoking, hypertension, heart disease and stroke, while ≤60 days/year was associated with anticoagulation use and bleeding history. In all, 13% reported not walking regularly, 12% walked <2 mph, 44% walked ≥2-2.9 mph, and 31% walked ≥3 mph. After propensity score adjustment, regular aspirin use was associated with a faster walking speed. Odds ratios (95% confidence intervals) were 1.16 (0.97 to 1.39), 1.24 (1.08 to 1.43), and 1.40 (1.21 to 1.63) for <2 mph, ≥2-2.9 mph and ≥3 mph, respectively, compared to not walking regularly (p-trend<0.001).
Conclusions: In this cohort of older men, long-term aspirin use is associated with a greater probability of faster walking speed later in life.

Key words: Aspirin, mobility, gait speed, frailty.



One of the most feared consequences of aging is loss of function and resultant loss of independence (1). Mobility is a critical aspect of function (l) independence, and limitations in mobility – specifically as measured by usual walking speed – are associated with an increased risk of morbidity and mortality (2-4). To date, the only proven therapy to prevent mobility limitation is exercise and leisure-time activity (5).
Mobility limitation is often considered a component of frailty, a state of decreased resilience to stressors that is more common in older adults (6) We have previously demonstrated that slow walking speed is associated with cardiovascular disease (CVD) events (7), and it is thought that frailty and dismobility have a bidirectional relationship with CVD, including peripheral vascular disease (PVD) (8-10). This suggests that pharmacologic therapies targeting CVD may prevent frailty and, specifically, mobility limitation, but this has yet to be investigated.
Among such therapies, aspirin is an intriguing option as it has both anti-inflammatory and anti-thrombotic properties which can improve large and small vessel blood flow and muscle function (11-13). Data from the ASPirin in Reducing Events in the Elderly (ASPREE) trial, demonstrated a reduction in persistent ADL disability among older adults randomized to aspirin vs placebo (14). However, in older adults, data on aspirin are limited and conflicting, with increasing risks of major bleeding largely outweighing benefit for those without a history of CVD (15, 16). In prior work we have shown that long term aspirin use in men is associated with a lower risk of frailty (17). We hypothesized that long term aspirin use, especially if started in middle age when risks of aspirin are minimal, would be associated with a lower risk of mobility limitation measured according to walking speed. We used 15-year follow-up data from men enrolled in the extended observational phases of the Physician’s Health Study (PHS) I.




PHS I is a completed double-blind, 2×2 factorial placebo-controlled trial that randomized 22,071 male physicians to either aspirin or placebo, and to beta-carotene or placebo, beginning in 1982 (18-20). At the start of the trial, participants were free of cancer and CVD. The aspirin intervention component of the trial ended early in 1988 at the recommendation of the PHS Data Safety Monitoring Board, due to the highly significant reduction in the rate of total myocardial infarctions among those assigned to aspirin versus placebo (19). Observational follow-up was extended after completion of the trial with annual questionnaires.
Detailed questionnaires were sent to participants annually to collect information on clinical variables, medications, and lifestyle from 1982 through 2012. Questions regarding walking speed were added to the annual questionnaire returned from 2001-2003. Participants responding to this questionnaire were eligible for the cross-sectional analyses described here, which compares participants to one another on the basis of cumulative aspirin exposure irrespective of randomized beta-carotene assignment. Of 14,896 participants who responded to the self-reported walking speed question, 581 were excluded due to missing covariate data. All participants were aged ≥58 years at the time of the walking speed assessment.
All participants executed written informed consent, and the PHS trial and subsequent follow-up were approved by the Institutional Review Board at Brigham and Women’s Hospital in Boston, MA, USA.


For the duration of the aspirin trial, participants were randomized to 325 mg aspirin or placebo every other day. The aspirin arm of the trial was stopped after an average of 60 months of follow-up (19). Once the aspirin arm of the trial ended, treatment crossover to aspirin use among those who had been assigned to placebo was greater than 70% (21). Ongoing use of aspirin was asked on every annual questionnaire, with the following question: “Over the past 12 months, on how many days have you taken aspirin or medication containing aspirin? 0 days, 1-13 days, 14-30 days, 31-60 days, 61-90 days, 91-120 days, 121-180 days, 180+days.” For the analyses reported here, aspirin exposure was computed from responses to the questionnaires sent out 1988-2001, corresponding to the year that walking pace was added. Average annual use of aspirin was summed and categorized for each participant as follows: ≤60 days/year (low use), >61 days/year (regular use) for the follow up years post-trial, as has been done previously in PHS (22). Information on the actual dose of aspirin was not available once the trial had ended.


A previously validated self-assessment of average walking speed was added to the questionnaire in 2001 (23, 24). The question was asked as follows: “What is your usual walking pace?” with the following possible responses: “Don’t walk regularly; Easy, casual (<2 mph); Normal, average (2-2.9mph); Brisk pace (3-3.9mph); Very brisk, striding (4mph or faster)”. After examining the data distribution, we categorized walking pace into 4 categories: don’t walk regularly, easy casual <2mph, normal ≥2-2.9mph, and brisk or very brisk ≥3mph.

Other Baseline Covariates

Information on demographics, comorbidities, and health, including history of bleeding, heart disease, stroke, peripheral artery disease, arthritis, migraine or headache, atrial fibrillation, or anticoagulation use, was drawn from all prior questionnaires to ensure complete capture of data. Smoking status was quantified as “never, past, or current”. Alcohol consumption was categorized as “daily, weekly, monthly, or rarely”. Cumulative non-aspirin non-steroidal anti-inflammatory drug use (NSAID) was quantified as the average number of years of self-reported NSAID use that were >60 days per year.

Statistical Analysis

Because of significant crossover to aspirin after completion of the intervention phase of the trial and dissolution of its randomized groupings, we developed a propensity score to balance the degree of aspirin exposure over a maximum 15 years of the post-intervention follow-up by relevant covariates. Estimation of the propensity score accounted for the initial PHS trial randomization; other variables that might be related to post-intervention aspirin use (e.g. history of bleeding, heart disease) and health-related risk factors and outcomes (e.g. smoking history). Exercise and other potential mediators or moderators of the influence of aspirin on walking speed were excluded from the propensity scoring model. We examined the distribution of propensity scores using kernel density plots to determine whether there was sufficient overlap between exposed vs unexposed (25, 26). We initially considered aspirin exposure in 2 or 3 categories of use (≤60 days vs >60 days or ≤60 days vs 61-180 vs >180 days per years). Two categories had the best overlap and were used for all analyses, as in our prior work (17, 22). Estimation employed inverse probability of treatment weighting (27-29). We compared the standardized differences between baseline covariates before and after propensity score adjustment to ensure <10% difference (25).
Descriptive characteristics were obtained after adjustment for propensity scores. Multinomial logistic regression models estimated the odds of prevalent mobility defined according to self-reported walking speed category among aspirin users relative to non-users (≤60 d/yr) while contrasting each level of walking speed to “Don’t walk regularly” category. Potential effect modification by age, history of heart disease, arthritis, bleeding and exercise frequency were examined by inspection and application of interaction tests, stratification and subgroup analysis.
Model-based estimates of association were accompanied by 95% confidence intervals (95% CI). Prespecified hypothesis testing was performed at an α=0.05 level. All analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC, USA) and R version 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria).



In total, 14,315 male, predominantly white (93%), physicians were included in this study. At time of walking speed assessment, the mean (standard deviation; SD) age was 70 (8) years with range 58-100 years; mean (SD) duration of aspirin use was 11 (5) years (range 0-18 years). Aspirin use was reported as ≤60 days/year by 2,056 (14.4%) participants. For walking speed, 13.1% reported not walking regularly, 12.2% walked <2 mph, 44.1% walked ≥2-2.9 mph, and 30.7% walked ≥3 mph.
Prior to propensity score adjustment, those who reported greater aspirin use were more likely to report drinking alcohol daily, prior smoking, and a history of atrial fibrillation, diabetes, hypertension, coronary heart disease (CHD), and stroke. Lower frequency of aspirin use was associated with greater anticoagulation use and bleeding history such as gastritis and melena. Following propensity score IPTW adjustment, there were no significant differences between groups (Table 1). With adjustment for propensity score, we found a graded association between regular aspirin use and faster walking speed. The multiplicative increase (95% CI) attributable to high aspirin use in odds of faster walking speed as compared to not regularly walking was 1.16 (0.97 to 1.39), 1.24 (1.08 to 1.43), and 1.40 (1.21 to 1.63) for those reporting walking speeds of <2 mph, ≥2-2.9 mph and ≥3 mph, respectively, (p-trend<0.001) (Table 2). In a sensitivity analysis, after adjusting for the propensity score in the crude model, results were unchanged.

Table 1. Cohort characteristics of 14,315 PHS participants according to average annual aspirin use, before and after propensity score adjustment

Table 2. Association between regular aspirin use and self-reported walking pace in 14,315 PHS participants before and after propensity score weighting

Legend: Sample sizes are provided for each category to allow for calculation of the raw ORs. For example, comparing “normal” walkers with those who “don’t walk” the estimated unadjusted OR is 1.18, as shown in the table. This is interpreted as: for an individual in the high-aspirin group, the relative odds of being in the “normal» walking group than the «don’t walk” group are18% greater with high aspirin exposure vs low aspirin exposure.


Subgroup analyses generally showed little evidence of differential effects across pre-specified categories of risk factors, including age, arthritis, bleeding or exercise frequency. The only subgroup in which there was a statistically significant interaction was among those with history of CHD (p<0.0008). The OR (95% CI) for those with prior CHD and high aspirin use (vs low) was 1.15 (0.80 to 1.64), 1.72 (1.26 to 2.35), and 2.84 (1.93 to 4.16) for those reporting <2 mph, ≥2-2.9 mph and 3+ mph, respectively, compared to 1.18 (0.96 to 1.45), 1.15 (0.98 to 1.34), and 1.24 (1.05 to 1.46) for those without CHD. (Figures 1 and 2).

Figure 1. Forest plot showing the relative odds of increased mobility according to level of aspirin use, stratifying by age and and history of heart disease. The referent mobility group is those who do not engage in regular walking

Figure 2. Forest plot showing the relative odds of increased mobility according to level of aspirin, stratifying on history of arthritis, bleeding, and weekly exercise. The referent mobility group is those who do not engage in regular walking



In this 15-year post-trial follow up study of 14,315 male physicians, long term aspirin use, started in middle age, was associated with a greater probability of faster walking speed in late life. Our results suggest that decreased walking speed might be prevented through regular aspirin use, independent of medical history and health behaviors. Importantly, however, results suggest that after consideration of these factors, the associations due to aspirin exposure are more strongly manifested among older individuals with a history of heart disease. This observation is in line with the hypothesis that those cardiovascular morbidity and frailty have shared underlying pathophysiology and may be targeted by regular aspirin use.
Aspirin lowers the risk of cardiovascular events by both preventing platelet aggregation and reducing inflammation. In patients with PVD, aspirin is an established part of the treatment plan, either alone or in combination with other antiplatelets or anticoagulants (30, 31). However, clinical trial data has not consistently demonstrated a benefit of aspirin of intermittent claudication. Although aspirin has anti-inflammatory and vasodilatory properties which might improve walking speed, prior clinical data in those with PVD does not support this hypothesis. While most older adults do not have clinical PVD, the microvascular changes that occur with aging contribute to decreased blood flow to leg muscles and may explain in part the natural slowing in walking and increased sarcopenia that is seen with physiologic aging (3, 32). We hypothesize that regular use of aspirin over years, in late middle age and early old age, may improve blood flow and circulation to distal muscles particularly in those with CVD, and therefore lead to less mobility limitation in later decades, in those with heart disease in particular. This is likely mediated by the vasodilatory and anti-inflammatory properties of aspirin.Randomized trials of aspirin for mobility are limited. The ASPREE trial examined the effect of aspirin on disability-free survival, CV events, and mortality in adults aged 70 and older, over 5-years (15, 33, 34). In secondary analyses there was a signal for a lower rate of persistent physical disability (HR 0.85, 95% CI 0.70 to 1.03) (33). and possible benefit for mobility. However, there are serious concerns regarding the safety of aspirin that need to be considered, particularly related to fatal bleeding events (15). It is possible that any benefit of aspirin for prevention of mobility limitation would need to begin earlier in life, however we found that the association between long-term use of aspirin and lower risk of slow walking remained even after accounting for age.
In subgroup analyses we explored whether aspirin use is more effective in certain disease states which may be impacted by aspirin use. Although overall results were similar, there was a statistically significant interaction for those with a history of CHD and a non-significant suggestion of greater benefit among those who do not exercise regularly. Cardiovascular disease has multiple causes, including increased inflammation and thrombotic milieu. That those with CHD had evidence of greater benefit from aspirin may suggest that those using aspirin for secondary prevention of CVD might improve walking speed later in life by reducing their burden of CVD over the lifespan by preventing future CV events and resultant disability. On the other hand, those who exercise may have sufficient benefit from improving blood flow to extremities (35) that the addition of aspirin does not further improve walking. Exercise has been shown to be one of the few modalities that can improve function and frailty at all stages (36), and it is possible that the benefits of aspirin are most beneficial in those who are inactive or who have biologically active CVD.
There are several important limitations to consider. First, this cohort is entirely male and largely made up of individuals identifying as white within a US context, and the study should be repeated in a cohort of women. However, women are also more likely to outlive men by 6-8 years and data that is sex specific is needed in aging research. Second, walking speed was assessed using self-report which could either over or under estimate actual walking pace, although the question used has been well validated (23, 24). Although the end-point mixes aspects of physical activity, function, and ability or disability, we interpret this as an important measure of actualized functional capacity in the real-world setting (21). Third, even though we used propensity score methods to address confounding by indication, reverse causality remains possible. Fourth, information on dose of aspirin was not available after completion of the trial and we are unable to examine the role of aspirin dose on the outcome. Future work to understand the relationship between aspirin and mobility could include plasma analysis for inflammatory biomarkers and mediation analysis to understand what role aspirin may play in reducing inflammation and preventing frailty.
In conclusion, in this cohort of older men, we found that long-term aspirin use started in late midlife was associated with lower odds of mobility limitation, as defined by self-reported walking speed. Future work should replicate these findings in women and further explore the protentional role of aspirin as a preventive strategy for frailty and mobility limitation.


Acknowledgments: The authors gratefully acknowledge the patients and research staff who participated in the Physicians’ Health Study. We are grateful for Natalya Gomelskaya for her statistical support. All authors contributed to study conception, writing, and editing the manuscript. L.Y., A.B.D., and T.G.T. conducted the statistical analysis. The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.

Funding: This work was supported by a Career Development award to A.R.O. from the Boston Claude D. Pepper Older Americans Independence Center, National Institute on Aging grant P30-AG013679. A.R.O. is also supported by Veterans Affairs CSR&D CDA-2 IK2-CX001800 and National Institute on Aging grant R03-AG060169. The Physicians’ Health Study is funded by grants CA-34944, CA-40360, and CA-097193 from the National Cancer Institute and grants HL-26490 and HL-34595 from the National Heart, Lung, and Blood Institute, Bethesda, MD.

Conflicts of Interest: J.M.G. reports serving as a consultant and receiving honoraria for speaking for Bayer.

Ethical standards: All participants provided written informed consent. The trial was approved by the Institutional Review Board at Brigham and Women’s Hospital.



1. Parry SW, Finch T and Deary V. How should we manage fear of falling in older adults living in the community? BMJ (Clinical research ed). 2013;346:f2933. doi: 10.1136/bmj.f2933
2. Studenski S, Perera S, Patel K, Rosano C, Faulkner K, Inzitari M, Brach J, Chandler J, Cawthon P, Connor EB, Nevitt M, Visser M, Kritchevsky S, Badinelli S, Harris T, Newman AB, Cauley J, Ferrucci L and Guralnik J. Gait speed and survival in older adults. Jama. 2011;305:50-8. doi: 10.1001/jama.2010.1923
3. Fritz S and Lusardi M. White paper: «walking speed: the sixth vital sign». Journal of geriatric physical therapy (2001). 2009;32:46-9.
4. Middleton A, Fritz SL and Lusardi M. Walking speed: the functional vital sign. Journal of aging and physical activity. 2015;23:314-22. doi: 10.1123/japa.2013-0236
5. Brown CJ and Flood KL. Mobility limitation in the older patient: a clinical review. Jama. 2013;310:1168-77. doi: 10.1001/jama.2013.276566
6. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, Seeman T, Tracy R, Kop WJ, Burke G and McBurnie MA. Frailty in older adults: evidence for a phenotype. The journals of gerontology Series A, Biological sciences and medical sciences. 2001;56:M146-56. doi: 10.1093/gerona/56.3.M146
7. Imran TF, Orkaby A, Chen J, Selvaraj S, Driver JA, Gaziano JM and Djoussé L. Walking pace is inversely associated with risk of death and cardiovascular disease: The Physicians’ Health Study. Atherosclerosis. 2019;289:51-56. doi: 10.1016/j.atherosclerosis.2019.08.001
8. Afilalo J, Alexander KP, Mack MJ, Maurer MS, Green P, Allen LA, Popma JJ, Ferrucci L and Forman DE. Frailty assessment in the cardiovascular care of older adults. Journal of the American College of Cardiology. 2014;63:747-62. doi: 10.1016/j.jacc.2013.09.070
9. Bouillon K, Batty GD, Hamer M, Sabia S, Shipley MJ, Britton A, Singh-Manoux A and Kivimaki M. Cardiovascular disease risk scores in identifying future frailty: the Whitehall II prospective cohort study. Heart (British Cardiac Society). 2013;99:737-42. doi: 10.1136/heartjnl-2012-302922
10. Newman AB, Gottdiener JS, McBurnie MA, Hirsch CH, Kop WJ, Tracy R, Walston JD and Fried LP. Associations of subclinical cardiovascular disease with frailty. The journals of gerontology Series A, Biological sciences and medical sciences. 2001;56:M158-66. doi: 10.1093/gerona/56.3.M158
11. Vane JR and Botting RM. The mechanism of action of aspirin. Thrombosis research. 2003;110:255-8. doi: 10.1016/S0049-3848(03)00379-7
12. Oh J, Sinha I, Tan KY, Rosner B, Dreyfuss JM, Gjata O, Tran P, Shoelson SE and Wagers AJ. Age-associated NF-kappaB signaling in myofibers alters the satellite cell niche and re-strains muscle stem cell function. Aging. 2016;8:2871-2896. doi: 10.18632/aging.101098
13. Morris T, Stables M, Hobbs A, de Souza P, Colville-Nash P, Warner T, Newson J, Bellingan G and Gilroy DW. Effects of low-dose aspirin on acute inflammatory responses in humans. Journal of immunology (Baltimore, Md : 1950). 2009;183:2089-96. doi: 10.4049/jimmunol.0900477
14. Woods RL, Espinoza S, Thao LTP, Ernst ME, Ryan J, Wolfe R, Shah RC, Ward SA, Storey E, Nelson MR, Reid CM, Lockery JE, Orchard SG, Trevaks RE, Fitzgerald SM, Stocks NP, Williamson JD, McNeil JJ, Murray AM and Newman AB. Effect of Aspirin on Activities of Daily Living Disability in Community-Dwelling Older Adults. The journals of gerontology Series A, Biological sciences and medical sciences. 2020. doi: 10.1093/gerona/glaa316
15. McNeil JJ, Wolfe R, Woods RL, Tonkin AM, Donnan GA, Nelson MR, Reid CM, Lockery JE, Kirpach B, Storey E, Shah RC, Williamson JD, Margolis KL, Ernst ME, Abhayaratna WP, Stocks N, Fitzgerald SM, Orchard SG, Trevaks RE, Beilin LJ, Johnston CI, Ryan J, Radziszewska B, Jelinek M, Malik M, Eaton CB, Brauer D, Cloud G, Wood EM, Mahady SE, Satterfield S, Grimm R and Murray AM. Effect of Aspirin on Cardiovascular Events and Bleeding in the Healthy Elderly. The New England journal of medicine. 2018;379:1509-1518. doi: 10.1056/NEJMoa1805819
16. Bibbins-Domingo K. Aspirin Use for the Primary Prevention of Cardiovascular Disease and Colorectal Cancer: U.S. Preventive Services Task Force Recommendation Statement. Annals of internal medicine. 2016;164:836-45. doi: 10.7326/M16-0577
17. Orkaby AR, Yang L, Dufour AB, Travison TG, Sesso HD, Driver JA, Djousse L and Gaziano JM. Association Between Long-Term Aspirin Use and Frailty in Men: The Physicians’ Health Study. The journals of gerontology Series A, Biological sciences and medical sciences. 2020. doi: 10.1093/gerona/glaa233
18. Findings from the aspirin component of the ongoing Physicians’ Health Study. The New England journal of medicine. 1988;318:262-4. doi: 10.1056/NEJM198801283180431
19. Final report on the aspirin component of the ongoing Physicians’ Health Study. Steering Committee of the Physicians’ Health Study Research Group. The New England journal of medicine. 1989;321:129-35. doi: 10.1056/NEJM198907203210301
20. Hennekens CH, Buring JE, Manson JE, Stampfer M, Rosner B, Cook NR, Belanger C, LaMotte F, Gaziano JM, Ridker PM, Willett W and Peto R. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. The New England journal of medicine. 1996;334:1145-9. doi: 10.1056/NEJM199605023341801
21. Sturmer T, Glynn RJ, Lee IM, Manson JE, Buring JE and Hennekens CH. Aspirin use and colorectal cancer: post-trial follow-up data from the Physicians’ Health Study. Annals of internal medicine. 1998;128:713-20. doi:10.7326/0003-4819-128-9-199805010-00003
22. Driver JA, Logroscino G, Lu L, Gaziano JM and Kurth T. Use of non-steroidal anti-inflammatory drugs and risk of Parkinson’s disease: nested case-control study. BMJ (Clinical research ed). 2011;342:d198. doi: 10.1136/bmj.d198
23. Wolf AM, Hunter DJ, Colditz GA, Manson JE, Stampfer MJ, Corsano KA, Rosner B, Kriska A and Willett WC. Reproducibility and validity of a self-administered physical activity questionnaire. International journal of epidemiology. 1994;23:991-9. doi: 10.1093/ije/23.5.991
24. Hu FB, Sigal RJ, Rich-Edwards JW, Colditz GA, Solomon CG, Willett WC, Speizer FE and Manson JE. Walking compared with vigorous physical activity and risk of type 2 diabetes in women: a prospective study. Jama. 1999;282:1433-9. doi: 10.1001/jama.282.15.1433
25. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Statistics in medicine. 2009;28:3083-107. doi: 10.1002/sim.3697
26. Crump RK, Hotz VJ, Imbens GW and Mitnik OA. Dealing with limited overlap in estimation of average treatment effects. 2009;96:187-199. doi: 10.1093/biomet/asn055
27. Elze MC, Gregson J, Baber U, Williamson E, Sartori S, Mehran R, Nichols M, Stone GW and Pocock SJ. Comparison of Propensity Score Methods and Covariate Adjustment: Evaluation in 4 Cardiovascular Studies. Journal of the American College of Cardiology. 2017;69:345-357. doi: 10.1016/j.jacc.2016.10.060
28. Curtis LH, Hammill BG, Eisenstein EL, Kramer JM and Anstrom KJ. Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Medical care. 2007;45:S103-7. doi: 10.1097/MLR.0b013e31806518ac
29. Kim DH, Pieper CF, Ahmed A and Colon-Emeric CS. Use and Interpretation of Propensity Scores in Aging Research: A Guide for Clinical Researchers. Journal of the American Geriatrics Society. 2016;64:2065-2073. doi: 10.1111/jgs.14253
30. Gerhard-Herman MD, Gornik HL, Barrett C, Barshes NR, Corriere MA, Drachman DE, Fleisher LA, Fowkes FG, Hamburg NM, Kinlay S, Lookstein R, Misra S, Mureebe L, Olin JW, Patel RA, Regensteiner JG, Schanzer A, Shishehbor MH, Stewart KJ, Treat-Jacobson D and Walsh ME. 2016 AHA/ACC Guideline on the Management of Patients With Lower Extremity Peripheral Artery Disease: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 2017;135:e686-e725. doi: 10.1161/CIR.0000000000000470
31. Hess CN and Hiatt WR. Antithrombotic Therapy for Peripheral Artery Disease in 2018. Jama. 2018;319:2329-2330. doi: 10.1001/jama.2018.5422
32. Proctor DN, Shen PH, Dietz NM, Eickhoff TJ, Lawler LA, Ebersold EJ, Loeffler DL and Joyner MJ. Reduced leg blood flow during dynamic exercise in older endurance-trained men. J Appl Physiol (1985). 1998;85:68-75. doi: 10.1152/jappl.1998.85.1.68
33. McNeil JJ, Woods RL, Nelson MR, Reid CM, Kirpach B, Wolfe R, Storey E, Shah RC, Lockery JE, Tonkin AM, Newman AB, Williamson JD, Margolis KL, Ernst ME, Abhayaratna WP, Stocks N, Fitzgerald SM, Orchard SG, Trevaks RE, Beilin LJ, Donnan GA, Gibbs P, Johnston CI, Ryan J, Radziszewska B, Grimm R and Murray AM. Effect of Aspirin on Disability-free Survival in the Healthy Elderly. The New England journal of medicine. 2018;379:1499-1508. doi: 10.1056/NEJMoa1800722
34. McNeil JJ, Nelson MR, Woods RL, Lockery JE, Wolfe R, Reid CM, Kirpach B, Shah RC, Ives DG, Storey E, Ryan J, Tonkin AM, Newman AB, Williamson JD, Margolis KL, Ernst ME, Abhayaratna WP, Stocks N, Fitzgerald SM, Orchard SG, Trevaks RE, Beilin LJ, Donnan GA, Gibbs P, Johnston CI, Radziszewska B, Grimm R and Murray AM. Effect of Aspirin on All-Cause Mortality in the Healthy Elderly. The New England journal of medicine. 2018;379:1519-1528. doi: 10.1056/NEJMoa1803955
35. Hildebrandt W, Schwarzbach H, Pardun A, Hannemann L, Bogs B, König AM, Mahnken AH, Hildebrandt O, Koehler U and Kinscherf R. Age-related differences in skeletal muscle microvascular response to exercise as detected by contrast-enhanced ultrasound (CEUS). PLoS One. 2017;12:e0172771. doi: 10.1371/journal.pone.0172771
36. Clegg A, Young J, Iliffe S, Rikkert MO and Rockwood K. Frailty in elderly people. Lancet. 2013;381:752-62. doi: 10.1016/S0140-6736(12)62167-9



S. Abe1, O. Ezaki2, M. Suzuki1


1. Day Care SKY, Yokohama, Japan; 2. Institute of Women’s Health Science, Showa Women’s University, Tokyo, Japan

Corresponding Author: Osamu Ezaki, M.D. Institute of Women’s Health Science, Showa Women’s University, 1-7-57 Taishido, Setagaya-ku, Tokyo 154-8533, Japan, Tel: +81-3-3411-7450; Fax: +81-3-3411-7450, E-mail: ezaki1952@yahoo.co.jp

J Frailty Aging 2021;in press
Published online September 2, 2021, http://dx.doi.org/10.14283/jfa.2021.33



Objectives: Supplementation with 6 g/day of medium-chain triglycerides (MCTs) at dinnertime increases muscle function and cognition in frail elderly adults relative to supplementation with long-chain triglycerides. However, suitable timing of MCT supplementation during the day is unknown.
Design: We enrolled 40 elderly nursing home residents (85.9 ± 7.7 years) in a 1.5-month randomized intervention trial. Participants were randomly allocated to two groups: one received 6 g/day of MCTs at breakfast (breakfast group) as a test group and the other at dinnertime (dinner group) as a positive control group.
Measurements: Muscle mass, strength, function, and cognition were monitored at baseline and 1.5 months after initiation of intervention.
Results: Thirty-seven participants completed the study and were included in the analysis. MCT supplementation in breakfast and dinner groups respectively increased right arm muscle area from baseline by 1.1 ± 0.8 cm2 (P<0.001) and 1.6 ± 2.5 cm2 (P<0.001), left arm muscle area by 1.1 ± 0.7 cm2 (P<0.001) and 0.9 ± 1.0 cm2 (P<0.01), right knee extension time by 39 ± 42 s (P<0.01) and 20 ± 32 s (P<0.05), leg open and close test time by 1.74 ± 2.00 n/10 s (P<0.01) and 1.67 ± 2.01 n/10 s (P<0.01), and Mini-Mental State Examination score by 1.5 ± 3.0 points (P=0.06) and 1.0 ± 2.1 points (P=0.06). These increases between two groups did not differ statistically significantly.
Conclusion: Supplementation with 6 g MCTs/day for 1.5 months, irrespective of ingestion at breakfast or dinnertime, could increase muscle mass and function, and cognition in frail elderly adults.

Key words: FIM, frailty, MCT oil, MMSE, sarcopenia.


Sarcopenia (loss of skeletal muscle mass, strength, and function) and dementia (or cognitive impairment) are common in elderly adults but difficult to treat. Previously, in a randomized, controlled trial with a time course, we found that supplementation with medium-chain triglycerides (MCTs) (6 g/day) at dinnertime for 3 months in frail elderly adults increased their muscle strength, function, activities of daily living (ADL) (1), and cognition (2) relative to supplementation with long-chain triglycerides (LCTs). The muscle and cognitive functions in the LCT group gradually decreased with time, whereas those in the MCT group increased (1, 2). However, suitable timing of MCT supplementation during the day is unknown.
Our previous studies (1, 2) suggested that O-n-octanoylation of ghrelin by C8:0 in MCTs could be a mechanism by which dietary MCTs increase muscle function and cognition , rather than in its role as a ketogenic meal (3). O-n-octanoylation by ghrelin o-acyltransferase in the stomach is essential for the activation of ghrelin, which stimulates the release of growth hormone (GH) in rats (4). Also, the injection of acyl-ghrelin in humans released GH in a dose-dependent manner (5, 6). A decrease in GH secretion accompanies aging and may contribute to the sarcopenia that develops in older adults (7). Ingestion of MCTs leads to the activation (i.e., acylation) of ghrelin in mice (8), which may lead to an increase in the level of GH secretion and a subsequent increase in muscle mass (the MCTs/Ghrelin/GH hypothesis). In humans, ingestion of MCTs also leads to an increase in blood acyl-ghrelin concentration (9-11). However, it has not been shown that ingestion of MCTs can increase GH secretion.
Both ghrelin and GH are secreted with circadian rhythmicity. Plasma ghrelin levels rise before a meal and drop after feeding (12). Inter-meal ghrelin levels displayed a diurnal rhythm, rising throughout the day to a zenith at 1:00 a.m., then falling overnight to a nadir at 9:00 a.m. (12). This circadian rhythm of ghrelin secretion may affect GH secretion. The stimulatory actions of GH-releasing hormone and ghrelin on GH secretion from pituitary and the tonic inhibitory influence of somatostatin generate a pulsatile pattern of GH secretion (13), which is stimulated during sleep (14). A major secretory episode occurs shortly after sleep onset, in association with the first period of slow-wave sleep (15). However, during aging, slow-wave sleep and GH secretion decrease with the same chronology (16). Therefore, to increase GH secretion during sleep, the timing of MCT supplementation may be important.
In our previous studies (1, 2, 17, 18), MCTs were given at dinnertime. If MCTs are given at breakfast time, the favorable effects of MCTs might not be observed because the effects of MCTs might not be long enough to increase ghrelin and GH secretion during sleep. In the present study, to determine suitable timing of MCT supplementation during the day, we compared the effects of MCTs given either at breakfast or dinnertime for 1.5 months, as both times are suitable to add MCTs in served foods for most elderly adults.


Materials and methods


This trial was announced in early April 2019 at the Day Care SKY facility in Yokohama, Japan. All participants who resided in this nursing home and who required special care from a helper were targeted (n = 68; mean age, 86.1 ± 7.7 years) (Figure 1). The registration was started on April 6, 2019 and ended on April 18, 2019. During this interval, 21 participants were excluded by the criteria of a body mass index (BMI) of >23 kg/m2 (to avoid a further increase in body weight); <65 years of age; parenteral nutrition; difficulty in swallowing; severe heart failure, lung, liver, kidney, or blood disease; a fasting blood glucose level of ≥200 mg/dL; a blood creatinine level of ≥1.5 mg/dL; or a CRP level of ≥2.0 mg/dL, and 7 participants were moved to other facilities, as described in Figure 1. Thus, 40 participants (5 men, 35 women; mean age, 85.9 ± 7.7 years) were enrolled and allocated to each group on April 18, 2019. Data collection at baseline was started on April 19, 2019 and ended by May 2, 2019. The intervention took place from May 3, 2019 to June 14, 2019 at Day Care SKY. Data collection after the intervention was started on June 16, 2019 and ended by June 20, 2019.

Figure 1. Trial profile

Participants (n = 40) were randomly allocated to two groups: the breakfast group (6 g/day of medium-chain triglycerides at breakfast time, n =20) and dinner group (6 g/day of medium-chain triglycerides at dinnertime, n =20). Thirty-seven participants completed the study and were included in the analysis. FIM = Functional Independence Measure; MCT = medium-chain triglycerides; RSST = repetitive saliva swallowing test.


The participants and their family members were informed of the nature of the experimental procedures before their written informed consent was obtained. In patients with cognitive decline or difficulty in writing (n = 12), informed consent was obtained from the patient’s family members. The present study was approved by the Human Ethics Committee of Japan Society of Nutrition and Food Science (Approval No. 87). The procedures were conducted in accordance with either the ethical standards of the institutional committee on human experimentation or the Helsinki Declaration of 1975 (as revised in 2000).

Study design

We performed a 1.5-month randomized, single blinded, parallel group intervention trial in which the 40 participants were randomly allocated into two groups (Figure 1). Sealed envelopes containing the written informed consent of the individual participants (or their family members) were thoroughly shuffled. Twenty participants (envelopes) each were allocated to the group that received 6 g/day MCTs at breakfast time (7:30~8:15 a.m.) and the group that received 6 g/day MCTs at dinnertime (6:00~6:45 p.m.). Allocation was conducted by a non-member of this study.
The intervention duration was 1.5 months, which was half that of the previous trial design (3 months) (1, 2, 17, 18), because significant dropouts in participants had been expected after the 1.5-month intervention. Due to legal limitations on the period of residency in this nursing home, some of the participants were required to leave this nursing home during the study period.
The participants’ body weight, appendicular muscle mass, strength, function, cognition, and ADL as described previously (1, 2) were assessed at baseline and 1.5 months after the initiation of the intervention. To avoid the possible acute effects of MCTs, assessment of the MCT intervention was conducted 2 days after the last MCT supplementation. Additionally in the present study, body composition (body fat percentage, fat mass, and muscle mass) in the whole body was measured by bioelectrical impedance method (ONE SMARTDIET, South Korea).


The participants in both groups were blinded for group allocation. They did not know whether they belonged to the breakfast or dinner group because they could not detect the foods containing MCTs. This might be due to the faint smell of the MCTs oil itself, the relatively small amounts of MCTs added to the foods, or the decreased sense of taste in the elderly participants.
To assess the outcomes, the examiners who oversaw the walking speed test and undertook the Functional Independence Measure (FIM) and the Nishimura geriatric rating scale for mental status (NM scale) were unaware of each participant’s group (blinded). Other assessments (anthropometric measurements, hand grip strength, knee leg extension time, leg open and close test, peak expiratory flow [PEF] test, repetitive saliva swallowing test [RSST], and Mini-Mental State Examination [MMSE]) were conducted by an expert with a certificate of training but who was aware of the participant’s group assignment (unblinded).

Study products

The MCTs (75% 8:0 and 25% 10:0 from total fatty acids) were purchased from Nisshin OilliO Group Ltd. (Kanagawa, Japan). Six grams of MCTs (50 kcal; 8.3 kcal/g) per day was mixed with foods such as steamed rice or miso soup at breakfast or dinnertime.

Daily time schedule of the participants

The daily time schedule in this nursing home was as follows: hour of rising, 5:30~6:30 a.m.; breakfast, 7:30~8:15 a.m.; lunch, 12:00~12:45 p.m.; snack, 3:00~3:30 p.m.; dinner, 6:00~6:45 p.m.; bedtime, 8:00~9:00 p.m.

Dietary intake

Breakfast, lunch, and dinner were served daily in the nursing care home. The habitual daily energy and nutrient intake of the individual participants during the baseline and intervention periods was measured as described previously (17). Then, the mean daily energy and macronutrient intakes of the two groups were calculated based on the daily energy and nutrient intakes of the individual participants (Table 1).

Table 1. Habitual energy and macronutrients intake at baseline and during the 1.5-month intervention and their changes in the breakfast and dinner groups (excluding MCT supplement)

Values are mean + SD. †P values represent the differences in the changes of variables between the two groups as assessed by 1-factor ANCOVA adjusted by each baseline value. MCFA = medium-chain fatty acid.


Daily activity and rehabilitation/exercise

Daily activities of this nursing care home were follows: at 7:15 a.m., exercises for the mouth were started for 10 min. At 10:00 a.m., exercises for the arms and fingers were started for 20 min. At 3:30 p.m., recreational therapy was started for 1 hour as described previously (1). At other times, residents spent their free time watching TV, lying in bed, and doing other activities. In addition, rehabilitation/exercise protocols were individually conducted. Several types of exercises such as walking, resistance training, leg stretches, stair stepping, or balance training were individually conducted for 20 min twice a week. The individual daily activities and rehabilitation/exercise were not changed during the baseline and intervention periods. The conductors of the daily activity and individual rehabilitation/exercise were unaware of the group to which each participant was assigned.


Medical drugs (antihypertensive, antiplatelet, antipsychotic, antilipemic, antidiabetic, antiosteoporosis, laxative, and hypnotic drugs), which were used by some of the participants, were not changed during the baseline or intervention periods.

Participants excluded from the analysis

Two participants in the breakfast group had paralysis of the right hand and another participant suffered from Parkinson’s disease and were excluded from the analysis of right-hand grip strength (Figure 1). One participant in the breakfast group had paralysis of the right leg and was excluded from the analysis of right knee extension time. One participant in the breakfast group had paralysis of the left leg and was excluded from the analysis of left knee extension time.
Three participants in the breakfast group had right foot paralysis, a fourth participant had low-back pain, and another participant refused to undergo the analysis of walking speed. One participant from the dinner group had difficulty in standing up due to increased body weight. All 6 participants were excluded from this test.
Three participants in the breakfast group and one from the dinner group had difficulty in understanding how to perform the PEF test and were excluded from the analysis of PEF. One participant in the breakfast group had motor apraxia and was excluded from the analysis of RSST. One participant in the dinner group changed from use of a walker to use of a wheelchair during the intervention (i.e., ADL were changed) and was excluded from the analysis of the FIM score.

Primary and secondary outcome variables

The primary outcome of the trial was the result of right knee extension time, which showed the largest increase (26 s, 43.3%, P < 0.001) in the 1.5-month intervention from baseline in the MCT group among the muscle tests conducted in our previous study (1). The other test results were considered the secondary outcomes. For the primary efficacy measure of right knee extension time, 36 participants were required in one group (n = 72 in two groups) for a power of 80% at a two-sided P of 0.05 to detect a treatment difference of 26 s with an SD of 39 s between the two groups by t-test.

Statistical analysis

All data are expressed as the mean + SD. In the within-group analysis, values at baseline and at the end of the intervention in each group were compared by Wilcoxon signed-rank test.
In the between-group analysis, to compare the groups with respect to the effects of MCTs on the measurements, differences in the change (change value = intervention value – baseline value) between the groups were assessed with analysis of covariance (ANCOVA), with adjustment for the baseline values in each measurement, age, sex, and BMI as covariates. The variances of change in all measurements were homogeneous between the groups by Levene’s test.
The percentage of relative change (% change) was calculated as follows: % change = (mean of the intervention value – mean of the baseline value) / mean of baseline value × 100. This value was then used to describe the degree of effect.
Missing data (data that could not be collected at baseline and/or after the intervention due to difficulty in performing tests) were not included in the analyses. A P level of 0.05 was used to determine statistical significance. All statistical analyses were performed with the SPSS 20.0 software program (IBM, Chicago, IL).



Participants and compliance

We enrolled 40 participants in the trial (Figure 1). Three participants dropped out during the study: one participant in the breakfast group due to bone fracture following a fall and 2 participants in the dinner group due to signs of acute heart failure. Thus, the remaining 37 participants (breakfast group: n = 19; 2 men, 17 women; mean age, 85.6 ± 7.9 years and the dinner group: n = 18; 2 men, 16 women; mean age, 86.4 ± 8.3 years) completed the study, and their data were used for the following analysis. No side effects, including diarrhea, or any other claims were reported.

Dietary intake (excluding supplemental MCTs)

Habitual intakes of energy and macronutrients at baseline and during the intervention period for the two groups are shown in Table 1. The MCTs that were administered are not included in this table. No differences between the baseline and intervention period were observed in either group with regard to the habitual intake of energy, protein, fat, and carbohydrate.

Anthropometric measures

Table 2 shows the anthropometric measures obtained at baseline and at the end of the 1.5-month intervention and their changes from baseline in the two groups. MCT supplementation did not affect body weight and BMI in either group. However, MCTs increased whole-body muscle mass from baseline by 0.3 ± 0.7 kg (non-significant, P = 0.11) in the breakfast group and 0.4 ± 0.5 kg (P < 0.01) in the dinner group and decreased whole-body fat mass by -0.9 ± 1.4 kg (P < 0.01) and –1.0 ± 1.5 kg (P < 0.01), respectively. In both groups, in agreement with the changes in body composition, increases in calculated right/left arm muscle area (AMA) and decreases in right/left triceps skinfold thickness (TSF) were noted. However, the increase in the right calf circumference (CC) was significant in the dinner group (0.4 ± 0.5 cm, P < 0.01) but not in the breakfast group (0.1 ± 0.6 cm, P = 0.23), whereas the increase in the left CC was significant in the breakfast group (0.2 ± 0.4 cm, P < 0.05) but not in the dinner group (0.3 ± 0.5 cm, P = 0.06). These changes between the two groups did not differ statistically significantly.

Table 2. Anthropometric measurements at baseline and 1.5-month intervention and their changes from baseline in the breakfast and dinner groups

Values are expressed as the mean ± SD. Asterisks indicate a statistically significant difference from baseline *P < 0.05, **P < 0.01, ***P < 0.001 by Wilcoxon signed-rank test. †P value represents the difference in the change in a variable between the two groups as assessed by a 1-factor ANCOVA adjusted for the baseline value in each measurement, age, sex, and BMI. AC = arm circumference; AMA = arm muscle area; BMI = body mass index; CC = calf circumference; TSF = triceps skinfold thickness.


Muscle strength and function

Muscle strength and function at baseline and at the end of the 1.5-month intervention and their changes from baseline in the two groups are shown in Table 3. MCT supplementation in the breakfast and dinner groups increased the right knee extension time from baseline by 39 ± 42 s (P < 0.01) and 20 ± 32 s (P < 0.05); the left knee extension time by 37 ± 36 s (P < 0.01) and 28 ± 32 s (P < 0.01); and the leg open and close test results by 1.74 ± 2.00 n/10s (P < 0.01) and 1.67 ± 2.01 n/10s (P < 0.01), respectively. However, the increase in the right-hand grip strength was significant only in the dinner group (0.9 ± 1.8 kg, P < 0.05) and not in the breakfast group (0.9 ± 2.3 kg, P = 0.24). These increases between the two groups did not differ statistically significantly.

Table 3. Muscle strength and function at baseline and 1.5-month intervention and their changes from baseline in the breakfast and dinner groups

Values are expressed as the mean ± SD. Asterisks indicate a statistically significant difference from baseline *P < 0.05, **P < 0.01 by Wilcoxon signed-rank test. †P value represents the difference in the change in a variable between the two groups as assessed by a 1-factor ANCOVA adjusted for the baseline values in each measurement, age, sex, and BMI. RSST = repetitive saliva swallowing test.


ADL and cognition

The FIM, MMSE, and NM scale scores at baseline and at the end of the 1.5-month intervention and their changes from baseline in the two groups are shown in Table 4. In these tests, a high score indicates better function. MCT supplementation in the breakfast and dinner groups increased the FIM score from baseline by 5.1 ± 6.0 points (P < 0.01) and 3.7 ± 5.0 points (P < 0.01); the MMSE score by 1.5 ± 3.0 points (P = 0.06) and 1.0 ± 2.1 points (P = 0.06); and the NM scale score by 3.5 ± 2.7 points (P < 0.01) and 2.2 ± 2.8 points (P < 0.01), respectively. These increases between the two groups also did not differ statistically significantly.

Table 4. FIM, MMSE, and NM scale scores at baseline and 1.5-month intervention and their changes from baseline in the breakfast and dinner groups

Values are means ± SD. In these tests, a high score indicates better function. Asterisks indicate a statistically significant difference from baseline *P < 0.05, **P < 0.01, ***P < 0.001 by Wilcoxon signed-rank test. †P values represent the differences in the changes of variables between the two groups as assessed by 1-factor ANCOVA adjusted for the baseline values in each measurement, age, sex, and BMI. FIM = Functional Independence Measure; MMSE = Mini-Mental State Examination; NM scale = Nishimura geriatric rating scale for mental status.



The present study showed that irrespective of the timing of MCT supplementation during the day (either at breakfast or at dinnertime), supplementation with 6 g MCTs/day for 1.5 months increased muscle mass and function, cognition, and ADL of frail elderly adults from baseline measurements. These results suggested that ingestion of MCT at either time could lead to similar activation of the MCTs/Ghrelin/GH axis.
A negative control (or placebo) group could not be included due to the small number of participants in the present trial. However, because our previous studies included an unsupplemented control group (inert control group) (17, 18) or negative placebo group (LCT group) (1, 2), it is clear that 6 g/day MCT supplementation increased muscle strength, function, ADL, and cognition relative to the control groups (inert or placebo).
The results are promising, but a future larger randomized study is needed. In addition, considering the reported adverse effects of the administration of ghrelin (19) and GH (20), long-term studies of MCT supplementation will be required to examine potential adverse effects, including risks of diabetes mellitus, cancer, and hyperplasia in some tissues.

Effects of MCTs on dietary intake

We did not observe any differences in habitual energy or nutritional intake between baseline and the intervention period in either group (Table 1), in agreement with our previous studies (1, 17). These results suggest that intakes of other daily nutrients did not alter the effects of MCT supplementation.
MCTs have been shown to be more satiating (i.e., decrease appetite) and promote weight loss compared with LCTs (21, 22), whereas ghrelin administration increases appetite (19). We speculated that either the dose of 6 g/day MCTs in our studies was not enough to promote weight loss, or an increase in appetite generated by ghrelin might counteract the effect of MCTs to decrease appetite.

Possible reasons for similar effects of MCT supplementation at either breakfast or dinnertime

The circadian profile of acyl-ghrelin and GH concentrations in blood after MCT supplementation, in which frequent blood sampling (i.e., more than 30 times throughout a 24-h period) was required (12, 16), have not been examined. However, blood acyl-ghrelin concentration after acute or chronic MCT supplementation, in which blood sampling is required several times throughout a 24-h period, has been reported in several human studies (9-11).
A time lag from oral MCT supplementation to an increase in ghrelin concentration was observed in two studies (8, 9). In one study in mice, the n-octanoyl ghrelin concentration in the stomach increased significantly at 3 h after glyceryl trioctanoate supplementation (8). In another experiment in mice from the same study, plasma acyl-ghrelin, which does not naturally occur in mammals, appeared after 4 h following supplementation of either glyceryl triheptanoate or n-heptanoic acid (7:0) (8). In a clinical study, 2–5 h was required to observe an increase in acyl-ghrelin concentration after the single oral ingestion of 3.0 g of a MCT supplement in an energy-containing formula in cachectic patients relative to patients without administration of the formula (9). Surprisingly, this increase in blood acyl-ghrelin concentration was maintained for at least another 10 h (9). Considering the shorter half-life of acyl-ghrelin in blood (10 min) (23), the longer activation of ghrelin by MCT supplementation suggested that medium-chain fatty acids (MCFAs), which are stored as triglyceride (TG) in X/A-like cells of the stomach, might be used for the production of acyl-ghrelin.
Although a part of ingested MCFAs and MCTs are directly used for ghrelin acyl modification in X/A-like cells of the stomach (8), the fate of MCFAs in these cells is not clear. Most MCFAs entering the cells may be used for ATP production in mitochondria or synthesis of TG in cytosol (24, 25). MCFAs stored as TG in cells and MCFAs directly entering from the circulation could also be used for acyl-ghrelin synthesis. Therefore, a time lag to the increase in acyl-ghrelin concentration in blood is expected after MCT supplementation due to the time necessary for formation of acyl-ghrelin, and the duration of this increased acyl-ghrelin concentration in blood may depend on the bio-availability of MCFAs (mostly from storage in the TG form) in the X/A-like cells of the stomach.
In the present study, MCTs were given in the dinner group at 6:00~6:45 p.m., which was 1–3 h before sleeping time (8:00~9:00 p.m.). Blood acyl-ghrelin concentration may increase at the initiation of sleeping time when GH secretion also increases. MCTs in the breakfast group were given at 7:30~8:15 a.m., which was 11–14 h before sleeping time (8:00~9:00 p.m.), and blood acyl-ghrelin concentrations that derived from TG in cells of the stomach, may still be elevated at this time. Therefore, irrespective of the timing of MCT supplementation during the day (at breakfast or dinner), supplementation with 6 g MCTs/d for 1.5 months could increase muscle mass and function, cognition, and ADL of the frail elderly adults from baseline.


Limitations of the present study are as follows. 1) The number of participants might be too small to observe significant effects of MCTs in some measures and non-significant differences in measures between the breakfast and dinner groups due to a lack of statistical power. A large-scale intervention study or meta-analysis may clarify these issues. 2) The examiners were not completely blinded to treatment allocation, which could bias the results. There was a chance that the examiner might favorably judge cognition in some groups, although the examiner kept in mind the responsibility of precisely assessing cognition. 3) Changes in the circadian rhythm of ghrelin and GH after MCT supplementation were not measured. However, frequent blood sampling will be necessary to elucidate the mechanisms of the effects of MCTs in the future. 4) Because this study targeted only frail elderly Japanese individuals, we did not address whether similar favorable effects of MCTs would be observed in western populations with a larger body size or in non-frail subjects.



Whether 6 g MCTs/day was supplemented either at breakfast or dinnertime, the increases in muscle mass and muscle function, cognition, and ADL of frail elderly individuals by MCT supplementation did not differ. MCTs are promising nutrients for elderly frail adults.


Acknowledgements: The authors thank all our study participants and all personnel at the Day Care SKY nursing home for collaborating with us.

Author contributions: SA, OE, and MS designed the research; SA and MS conducted the research; SA and OE analyzed the data or performed the statistical analysis; OE wrote the manuscript; and SA has the primary responsibility for the final content. All of the authors read and approved the final manuscript.

Funding sources: This study was supported by a grant from the Policy-Based Medical Services Foundation, 2019 (to SA).

Author disclosures: SA, OE, and MS have nothing to disclose.

Ethical Standards: The authors declare that the study procedures comply with the current ethical standards for investigation involving human participants in Japan.



1. Abe, S.; Ezaki, O.; Suzuki, M. Medium-chain triglycerides (8:0 and 10:0) are promising nutrients for sarcopenia: a randomized controlled trial. The American journal of clinical nutrition 2019, 110, 652-665, doi:10.1093/ajcn/nqz138.
2. Abe, S.; Ezaki, O.; Suzuki, M. Medium-Chain Triglycerides (8:0 and 10:0) Increase Mini-Mental State Examination (MMSE) Score in Frail Elderly Adults in a Randomized Controlled Trial. J Nutr 2020, 150, 2383-2390, doi:10.1093/jn/nxaa186.
3. Cunnane, S.C.; Courchesne-Loyer, A.; Vandenberghe, C.; St-Pierre, V.; Fortier, M.; Hennebelle, M.; Croteau, E.; Bocti, C.; Fulop, T.; Castellano, C.A. Can Ketones Help Rescue Brain Fuel Supply in Later Life? Implications for Cognitive Health during Aging and the Treatment of Alzheimer’s Disease. Frontiers in molecular neuroscience 2016, 9, 53, doi:10.3389/fnmol.2016.00053.
4. Kojima, M.; Hosoda, H.; Date, Y.; Nakazato, M.; Matsuo, H.; Kangawa, K. Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 1999, 402, 656-660, doi:10.1038/45230.
5. Takaya, K.; Ariyasu, H.; Kanamoto, N.; Iwakura, H.; Yoshimoto, A.; Harada, M.; Mori, K.; Komatsu, Y.; Usui, T.; Shimatsu, A., et al. Ghrelin strongly stimulates growth hormone release in humans. J Clin Endocrinol Metab 2000, 85, 4908-4911, doi:10.1210/jcem.85.12.7167.
6. Hataya, Y.; Akamizu, T.; Takaya, K.; Kanamoto, N.; Ariyasu, H.; Saijo, M.; Moriyama, K.; Shimatsu, A.; Kojima, M.; Kangawa, K., et al. A low dose of ghrelin stimulates growth hormone (GH) release synergistically with GH-releasing hormone in humans. J Clin Endocrinol Metab 2001, 86, 4552, doi:10.1210/jcem.86.9.8002.
7. Liu, H.; Bravata, D.M.; Olkin, I.; Nayak, S.; Roberts, B.; Garber, A.M.; Hoffman, A.R. Systematic review: the safety and efficacy of growth hormone in the healthy elderly. Ann Intern Med 2007, 146, 104-115, doi:10.7326/0003-4819-146-2-200701160-00005.
8. Nishi, Y.; Hiejima, H.; Hosoda, H.; Kaiya, H.; Mori, K.; Fukue, Y.; Yanase, T.; Nawata, H.; Kangawa, K.; Kojima, M. Ingested medium-chain fatty acids are directly utilized for the acyl modification of ghrelin. Endocrinology 2005, 146, 2255-2264, doi:10.1210/en.2004-0695.
9. Ashitani, J.; Matsumoto, N.; Nakazato, M. Effect of octanoic acid-rich formula on plasma ghrelin levels in cachectic patients with chronic respiratory disease. Nutr J 2009, 8, 25, doi:10.1186/1475-2891-8-25.
10. Kawai, K.; Nakashima, M.; Kojima, M.; Yamashita, S.; Takakura, S.; Shimizu, M.; Kubo, C.; Sudo, N. Ghrelin activation and neuropeptide Y elevation in response to medium chain triglyceride administration in anorexia nervosa patients. Clin Nutr ESPEN 2017, 17, 100-104, doi:10.1016/j.clnesp.2016.10.001.
11. Yoshimura, Y.; Shimazu, S.; Shiraishi, A.; Nagano, F.; Tominaga, S.; Hamada, T.; Kudo, M.; Yamasaki, Y.; Noda, S.; Bise, T. GHRELIN ACTIVATION BY INGESTION OF MEDIUM-CHAIN TRIGLYCERIDES IN HEALTHY ADULTS: A PILOT TRIAL. Journal of Aging Research & Clinical Practice 2018, 7.
12. Cummings, D.E.; Purnell, J.Q.; Frayo, R.S.; Schmidova, K.; Wisse, B.E.; Weigle, D.S. A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 2001, 50, 1714-1719, doi:10.2337/diabetes.50.8.1714.
13. Murray, P.G.; Higham, C.E.; Clayton, P.E. 60 YEARS OF NEUROENDOCRINOLOGY: The hypothalamo-GH axis: the past 60 years. The Journal of endocrinology 2015, 226, T123-140, doi:10.1530/JOE-15-0120.
14. Veldhuis, J.D.; Keenan, D.M.; Pincus, S.M. Motivations and methods for analyzing pulsatile hormone secretion. Endocrine reviews 2008, 29, 823-864, doi:10.1210/er.2008-0005.
15. Van Cauter, E.; Plat, L.; Copinschi, G. Interrelations between sleep and the somatotropic axis. Sleep 1998, 21, 553-566.
16. Van Cauter, E.; Leproult, R.; Plat, L. Age-related changes in slow wave sleep and REM sleep and relationship with growth hormone and cortisol levels in healthy men. Jama 2000, 284, 861-868, doi:10.1001/jama.284.7.861.
17. Abe, S.; Ezaki, O.; Suzuki, M. Medium-Chain Triglycerides in Combination with Leucine and Vitamin D Increase Muscle Strength and Function in Frail Elderly Adults in a Randomized Controlled Trial. J Nutr 2016, 146, 1017-1026, doi:10.3945/jn.115.228965.
18. Abe, S.; Ezaki, O.; Suzuki, M. Medium-Chain Triglycerides in Combination with Leucine and Vitamin D Benefit Cognition in Frail Elderly Adults: A Randomized Controlled Trial. J Nutr Sci Vitaminol (Tokyo) 2017, 63, 133-140, doi:10.3177/jnsv.63.133.
19. Yanagi, S.; Sato, T.; Kangawa, K.; Nakazato, M. The Homeostatic Force of Ghrelin. Cell metabolism 2018, 27, 786-804, doi:10.1016/j.cmet.2018.02.008.
20. Colon, G.; Saccon, T.; Schneider, A.; Cavalcante, M.B.; Huffman, D.M.; Berryman, D.; List, E.; Ikeno, Y.; Musi, N.; Bartke, A., et al. The enigmatic role of growth hormone in age-related diseases, cognition, and longevity. Geroscience 2019, 41, 759-774, doi:10.1007/s11357-019-00096-w.
21. St-Onge, M.P.; Jones, P.J. Physiological effects of medium-chain triglycerides: potential agents in the prevention of obesity. J Nutr 2002, 132, 329-332, doi:10.1093/jn/132.3.329.
22. French, S.; Robinson, T. Fats and food intake. Curr Opin Clin Nutr Metab Care 2003, 6, 629-634, doi:10.1097/00075197-200311000-00004.
23. Tong, J.; Dave, N.; Mugundu, G.M.; Davis, H.W.; Gaylinn, B.D.; Thorner, M.O.; Tschop, M.H.; D’Alessio, D.; Desai, P.B. The pharmacokinetics of acyl, des-acyl, and total ghrelin in healthy human subjects. Eur J Endocrinol 2013, 168, 821-828, doi:10.1530/EJE-13-0072.
24. Friedman, M.I.; Ramirez, I.; Bowden, C.R.; Tordoff, M.G. Fuel partitioning and food intake: role for mitochondrial fatty acid transport. Am J Physiol 1990, 258, R216-221, doi:10.1152/ajpregu.1990.258.1.R216.
25. Papamandjaris, A.A.; MacDougall, D.E.; Jones, P.J. Medium chain fatty acid metabolism and energy expenditure: obesity treatment implications. Life Sci 1998, 62, 1203-1215.




G. Kojima1, Y. Taniguchi2, T. Urano3


1. Department of Research, Dr. AGA Clinic, Tokyo, Japan; 2. Center for Health and Environmental Risk Research, National Institute for Environmental Studies, Tsukuba, Japan; 3. Department of Geriatric Medicine, International University of Health and Welfare, Chiba, Japan

Corresponding Author: Tomohiko Urano, MD, PhD, Department of Geriatric Medicine, International University of Health and Welfare, Kozunomori 4-3, Narita City, Chiba Prefecture, 286-8686, Japan, Phone: +81 (0)476-20-7701, Fax: +81 (0)476-20-7702, Email: turano@iuhw.ac.jp

J Frailty Aging 2021;in press
Published online September 2, 2021, http://dx.doi.org/10.14283/jfa.2021.32



Background: There is limited evidence regarding associations between fruit and vegetable consumption and incident frailty risk among older people.
Objectives: The objective of this study was to conduct a systematic review and meta-analysis regarding the association between fruit and vegetable consumption and incident frailty risk among older adults.
Methods: A systematic search of the literature was conducted according to the PRISMA guidelines using PubMed in January 2021 for studies that prospectively examined risk of incident frailty in relation to fruit and vegetable consumption in older adults aged 60 and older. Methodological quality and heterogeneity were assessed. Odds ratios (OR) were pooled using random-effects or fixed-effects meta-analysis, depending on the presence of heterogeneity.
Results: Among three studies included in this review, data of four cohorts were provided by two studies and used in meta-analysis. The highest fruit and vegetable consumption was significantly associated with lower risk of incident frailty compared with the lowest consumption (pooled OR=0.38, 95%CI=0.24-0.59, p=<0.001).
Conclusions: This study provided the pooled evidence that high fruit and vegetable consumption may be beneficial for preventing the development of frailty in older adults. Increasing fruit and vegetable consumption can be a relevant strategy to prevent frailty.

Key words: Frailty, fruit, vegetables, diet, Nutrition, meta-analysis.



Fruit and vegetables contain various nutrients, including micronutrients with anti-inflammatory and antioxidant properties and are important parts of healthy diet (1). Their benefits for human health have been extensively studied and well documented in the literature (1). Sufficient intake of fruit and vegetables is associated with reduced risk of cardiovascular diseases, certain cancers, and premature death (1, 2). Therefore, fruit and vegetables are widely recommended by most authorities and nutritional guidelines (1, 2). A joint report by The World Health Organization and The Food and Agriculture Organization of the United Nations recommends at least 400g of fruit and vegetables a day to prevent chronic diseases (3). Unfortunately, these health benefits of fruit and vegetable consumption are relatively understudied in the older adults (4).
Frailty is a state of vulnerability to stressors and age-related decreased functioning of various physiological systems (5). It is considered as one of the problematic expression of aging. Previous studies showed approximately 10% of older adults aged 65 or more are affected by frailty and the prevalence increases with age (6). Those who are frail are susceptible to multiple negative health outcomes, including falls (7), disabilities (8), dementia (9), and death (10). Frailty have also been shown to increase the use of healthcare resources, such as hospitalization (11), institutionalization (12), and emergency department visits (13). Given these detrimental negative impacts of frailty on older people as well as on societies, frailty has been recognized as one of the most important public health priorities (14, 15).
Inadequate intake of fruit and vegetable is common among older people due to various reasons (4), such as age-related physiological and social changes (16), and may be related to increased inflammation and oxidative stress. Evidence suggests that oxidative stress and inflammation play important roles in pathogenesis of sarcopenia, which is the decline in muscle mass and function with age and a core feature of frailty (17). Therefore, increasing fruit and vegetable consumption may be effective for preventing frailty. However, few studies focused specifically on fruit and vegetables in relation to frailty (18), although multiple studies have investigated associations between certain dietary patterns and frailty risk, i.e., Mediterranean diet (19). Although a previous systematic review on associations between fruit and vegetable intake and frailty found little evidence (18), there have been more evidence published in the literature since then. Therefore, the objectives of the current study were to perform an updated systematic literature search and to attempt to conduct a meta-analysis to pool the evidence regarding the association between fruit and vegetable consumption and incident frailty risk among community-dwelling older adults.



Search strategy and study selection

The protocol of this systematic review was developed according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statements (20) and registered at PROSPERO (https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021225161). The PubMed was searched by one investigator (KG) in January 2021 using prespecified search terms, which included Fruit (MeSH) OR Vegetables (MeSH) OR Fruit and Vegetable Juices (MeSH) OR Antioxidants (MeSH) OR Diet (MeSH) OR Diet Therapy (MeSH) OR Nutrition Therapy (MeSH) OR fruit* OR vegetable* OR anti-oxidant* OR antioxidant* OR diet* OR nutrition* AND Frailty (MeSH) OR Frail Elderly (MeSH) OR OR frailty. Study may have been eligible if they were prospectively cohort studies examining baseline fruit and/or vegetable consumption and subsequent risk of developing frailty in community-dwelling older people with a mean age of 60 or over. Reference lists of included and relevant articles were also examined. The search period was from 2000 or later, since the most widely used frailty criteria, the Cardiovascular Health Study (CHS) criteria were published in 2001 (21). No language restriction was applied.
For meta-analysis, eligible studies prospectively examined risk of incident frailty in relation to consumption of fruit, vegetables, or both at baseline in samples of older adults with a mean age of 60 or over. Studies including selected samples, such as patients with dementia or hospitalized patients, or focusing on dietary patterns were not considered. Frailty should be defined by validated criteria designed to measure frailty. Studies defining frailty status as a continuous score or index, such as the Frailty Index, were not included. Reviews, conference abstracts, editorials, comments, or letters were not included.
Two investigators (GK, YT) screened the titles, abstracts, and full-texts of the identified studies for eligibility. Any disagreement was solved by discussion.

Data Extraction

The data extracted from the included studies were first author, publication year, cohort name, location, sample size, proportion of women in the cohort, mean age, age range, dietary assessment tool, frailty criteria, follow-up period, and findings.

Methodological Quality Assessment

The methodological quality of the included studies was assessed using the Newcastle-Ottawa scale for cohort studies (22). This tool has nine criteria and a study meeting five or more criteria was considered to have adequate methodological quality.

Statistical Analysis

If two or more prospective studies provided the same effect measures, such as odds ratio (OR) or hazard ratio (HR), they were pooled by meta-analysis using the inverse variance method. The heterogeneity across the studies were examined using chi-square, and the degree of heterogeneity was assessed using the I2 statistic. A random-effects model was used when significant heterogeneity was present, and a fixed-effects model was used when absent. If a study provided both unadjusted and adjusted effect measures, the ones with full adjustment were used for meta-analysis. Publication bias was assessed by visually inspecting the funnel plots. Sensitivity and subgroup analyses were not possible due to the small number of included studies.



Selection Processes

The results of the systematic review and the study selection processes were presented in Figure 1. The systematic search of the literature found 290 studies, of which 278 studies were excluded by title screening. Among twelve studies left, 8 studies and 1 study were further excluded due to not providing sufficient data and using the same cohort, respectively. Finally, three studies were included for this review, among which two studies were used for meta-analysis. One study was excluded since it used frequency of fruit and vegetable consumption.

Figure 1. Flow chart of systematic literature review


Study Characteristics

Table 1 presents the summary of included studies. Three studies used quantity of fruit and vegetable consumption. A recent US study used data of 78,336 older female nurses from the Nurses’ Health Study in the US and examined risk of incident frailty (23). They found that those who consumed 7 or more servings of fruit and vegetables per day had a significantly decreased risk than those who consumed less than 3 servings per day over 20 years (23). Another study from the England investigated the associations between fruit and vegetable consumption at baseline and risk of developing frailty 4 years later in 2,634 non-frail older men and women, however, failed to find any significant associations (24). The third study used three cohorts of older adults (1 from Spain and 2 from France), and two of them showed that those consuming 5 or more portions of fruit and vegetables a day had a significantly higher risk of developing frailty compared with those consuming less than 1 portions a day, and the other showed non-significant results (25). The food assessments were either self-administered questionnaires (23-25) or questionnaires with a help of trained research assistant (25).

Table 1. Summary of included studies on associations between fruit and vegetable consumption and incident frailty risk

aHR: Adjusted hazard ratio; aOR: Adjusted odds ratio; BMI: Body mass index; ELSA: English Longitudinal Study of Ageing; FFQ: Food Frequency Questionnaire; MMSE: Mini-Mental State Examination; Senior-ENRICA: Study on Nutrition and Cardiovascular Risk Factors in Spain


Methodological quality assessment

Three studies (23-25) were evaluated using the Newcastle-Ottawa scale for cohort studies and were considered to have adequate methodological quality, with the mean score of 6.8 (range=6-8).

Fruit and vegetable consumption and incident frailty risk

OR of incident frailty risk according to fruit and vegetable consumption (combined) were provided from four cohorts (3 cohorts in 1 study (25) and 1 cohort in another study (24); Figure 2A). The OR for the highest category of fruit and vegetable consumption compared with the lowest category were combined using the fixed-effects meta-analysis. The highest fruit and vegetable consumption was significantly associated with lower risk of incident frailty (pooled OR=0.38, 95%CI=0.24-0.59, p=<0.001) than the lowest consumption. The degree of heterogeneity across the included cohort data was low (I2=0%).

Figure 2. Forest plots of incident frailty risk according to A: fruit and vegetables, B: fruit only, and C: vegetables only

AMI: Integrated Multidisciplinary Approach cohort, CI: confidence interval, IV: inverse variance, SE: standard error


The meta-analysis was repeated for fruit and vegetable alone separately (Figures 2B and 2C). Although the highest vegetable consumption was significantly associated with decreased risk of incident frailty (pooled OR=0.54, 95%CI=0.32-0.91, p=0.02; Figure 2C), there was no significant association observed between fruit and incident frailty risk (pooled OR=0.75, 95%CI=0.35-1.60, p=0.46; Figure 2B). It was not possible to properly assess the publication bias due to the small number of the included cohorts.



This study systematically searched the literature for associations between fruit and vegetable consumption and incident frailty risk and was able to provide the pooled evidence that higher fruit and vegetable consumption is associated with lower risk of incident frailty among community-dwelling older adults.
We found a few studies focusing on frequency of consuming fruit and vegetable, or vegetable alone (26-28). Two studies using the British Whitehall II study cohort, consisting of middle-aged civil servants aged 35-55 at recruitment showed similar results. In the first study, consuming fruit and vegetables at least twice a day when 50 years old was associated with a decreased risk of frailty 20 years later compared with consuming less than once a day (adjusted HR=0.70, 95%CI=0.53-0.92) (26). It should be noted that this study did not examine frailty status at baseline, although prevalence of frailty at the age of 50 would have been very low (26). The second study of the Whitehall II cohort showed that those consuming fruit and vegetables less than daily at the age of 45-55 had significantly higher risk of developing frailty approximately 18 years later compared with those consuming more than daily (adjusted OR=1.29, 95%CI=1.05-1.58). A Swedish study conducted a secondary data analysis of 371 independent community-dwelling Swedish men and women aged 80 years or older who participated in a randomized trial designed to evaluate the effects of a preventive home visit and multi-professional senior group meetings (28). According to this study, frequency of vegetable intake at baseline was not significantly associated with frailty risk at any time points: baseline, 12 months, or 24 months (Unadjusted OR=1.22, 95%CI=0.86-1.73 at 1 year, unadjusted OR=1.03, 95%CI=0.71-1.49 at 2 years) (28). Potential limitations of the discussed studies may include a small sample size, unadjusted effect measures, selection bias, and a relatively short study period.
“Inflammaging” is referred to an age-related low-grade chronic inflammation status and has been considered to be associated with the development of frailty (29). Multiple studies showed that those who are frail have significantly higher levels of inflammatory markers than those who are not (29). Another possible cause of frailty is oxidative stress. The oxidative stress plays an important role in the aging process and is associated with accelerated aging. Previous studies showed that frailty was associated with higher oxidative stress and lower anti-oxidative parameters (30). Fruit and vegetables are rich in flavonoids, which form a group of natural products with various phenolic structures and are known for their health-beneficial effects, including anti-inflammatory and anti-oxidative characteristics (31). These properties may be attributed to the decreased risk of developing frailty.
In the meta-analysis, fruit consumption only was not associated with incident frailty. Since high degree of heterogeneity was observed, a random-effect meta-analysis was used. In a sensitivity analysis, excluding one study showing very high OR (24) significantly decreased the heterogeneity from 69% to 0%. A meta-analysis involving the remaining three cohorts showed that the highest fruit consumption group had a significantly lower risk of incident frailty (pooled OR=0.51, 95%CI=0.31-0.83, p<0.01). These findings suggest that the results of the meta-analysis of the fruit consumption and incident frailty may not be robust and that it is not necessarily true that fruit is not associated with a lower incident frailty risk.
The current review has potential limitations. First, the number of cohorts included in the meta-analysis were relatively small. However, the result of the meta-analysis showed the significant inverse association between fruit and vegetable consumption and incident frailty risk was persistent and that the degree of heterogeneity was low. Second, a systematic search of the literature was conducted using PubMed only, therefore, some important studies may have been missed. Nonetheless, the chance may be limited since the comprehensive search strategy with an extensive array of search terms, including MeSH terms, was applied and the reference list of the relevant papers was scrutinized.
This study has multiple strengths. First, the robust search protocol was developed following the PRISMA statements. Second, the comprehensive search strategy, including screening done by two researchers independently, assessment of heterogeneity and methodological quality. In addition, it was possible to perform the meta-analysis and show the significant inverse association between fruit and vegetable consumption and incident frailty risk.



This systematic review and meta-analysis provided the pooled evidence that high fruit and vegetable consumption may be beneficial for preventing the development of frailty in older adults. As increasing fruit and vegetable consumption is feasible and relatively cheap without significant side effects, this can be a relevant strategy to prevent frailty. Randomized controlled trials are warranted to confirm the positive effects against frailty (32, 33).


Funding: We had no specific funding to support this study.

Conflict of interests: All authors declare no conflict of interest.

Acknowledgment: We are grateful to the authors who provided additional data.

Ethical standards: This systematic review and meta-analysis study was conducted according to the PRISMA guidelines and the prespecified protocol was registered at PROSPERO.





1. Fruit and Vegetables for Health Initiative, Fruit and Vegetables for Health Initiative; 2018. http://www.fao.org/3/i6807e/i6807e.pdf. Accessed 21 March 2021.
2. Slavin, JL, Lloyd, B. Health benefits of fruits and vegetables. Adv Nutr 2012;3(4):506-516.
3. Who, J, Consultation, FE. Diet, nutrition and the prevention of chronic diseases. World Health Organ Tech Rep Ser 2003;916(i-viii).
4. Nicklett, EJ, Kadell, AR. Fruit and vegetable intake among older adults: a scoping review. Maturitas 2013;75(4):305-312.
5. Clegg, A, Young, J, Iliffe, S, et al. Frailty in elderly people. Lancet 2013;381(9868):752-762.
6. Collard, RM, Boter, H, Schoevers, RA, et al. Prevalence of frailty in community-dwelling older persons: a systematic review. J Am Geriatr Soc 2012;60(8):1487-1492.
7. Kojima, G. Frailty as a Predictor of Future Falls Among Community-Dwelling Older People: A Systematic Review and Meta-Analysis. J Am Med Dir Assoc 2015;16(12):1027-1033.
8. Kojima, G. Frailty as a predictor of disabilities among community-dwelling older people: a systematic review and meta-analysis. Disabil Rehabil 2017;39(19):1897-1908.
9. Kojima, G, Taniguchi, Y, Iliffe, S, et al. Frailty as a Predictor of Alzheimer Disease, Vascular Dementia, and All Dementia Among Community-Dwelling Older People: A Systematic Review and Meta-Analysis. J Am Med Dir Assoc 2016;17(10):881-888.
10. Vermeiren, S, Vella-Azzopardi, R, Beckwée, D, et al. Frailty and the Prediction of Negative Health Outcomes: A Meta-Analysis. J Am Med Dir Assoc 2016;17(12):1163.e1161-1163.e1117.
11. Kojima, G. Frailty as a predictor of hospitalisation among community-dwelling older people: a systematic review and meta-analysis. J Epidemiol Community Health 2016;70(7):722-729.
12. Kojima, G. Frailty as a Predictor of Nursing Home Placement Among Community-Dwelling Older Adults: A Systematic Review and Meta-analysis. J Geriatr Phys Ther 2018;41(1):42-48.
13. Kojima, G. Frailty as a Predictor of Emergency Department Utilization among Community-Dwelling Older People: A Systematic Review and Meta-Analysis. J Am Med Dir Assoc 2019;20(1):103-105.
14. Cesari, M, Prince, M, Thiyagarajan, JA, et al. Frailty: An Emerging Public Health Priority. J Am Med Dir Assoc 2016;17(3):188-192.
15. Kojima, G, Liljas, AEM, Iliffe, S. Frailty syndrome: implications and challenges for health care policy. Risk Manag Healthc Policy 2019;12:23-30.
16. Azzolino, D, Passarelli, PC, De Angelis, P, et al. Poor Oral Health as a Determinant of Malnutrition and Sarcopenia. Nutrients 2019;11(12).
17. Meng, SJ, Yu, LJ. Oxidative stress, molecular inflammation and sarcopenia. Int J Mol Sci 2010;11(4):1509-1526.
18. Kojima, G, Avgerinou, C, Iliffe, S, et al. Fruit and Vegetable Consumption and Frailty: A Systematic Review. J Nutr Health Aging 2018;22(8):1010-1017.
19. Kojima, G, Avgerinou, C, Iliffe, S, et al. Adherence to Mediterranean Diet Reduces Incident Frailty Risk: Systematic Review and Meta-Analysis. J Am Geriatr Soc 2018;66(4):783-788.
20. Moher, D, Liberati, A, Tetzlaff, J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Bmj 2009;339:b2535.
21. Fried, LP, Tangen, CM, Walston, J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56(3):M146-156.
22. Wells, G, Shea, B, O’connell, D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa: Ottawa Hospital Research Institute. oxford. asp; 2011.
23. Fung, TT, Struijk, EA, Rodriguez-Artalejo, F, et al. Fruit and vegetable intake and risk of frailty in women 60 years old or older. Am J Clin Nutr 2020;112(6):1540-1546.
24. Kojima, G, Iliffe, S, Jivraj, S, et al. Fruit and Vegetable Consumption and Incident Prefrailty and Frailty in Community-Dwelling Older People: The English Longitudinal Study of Ageing. Nutrients 2020;12(12).
25. García-Esquinas, E, Rahi, B, Peres, K, et al. Consumption of fruit and vegetables and risk of frailty: a dose-response analysis of 3 prospective cohorts of community-dwelling older adults. Am J Clin Nutr 2016;104(1):132-142.
26. Gil-Salcedo, A, Dugravot, A, Fayosse, A, et al. Healthy behaviors at age 50 years and frailty at older ages in a 20-year follow-up of the UK Whitehall II cohort: A longitudinal study. PLoS Med 2020;17(7):e1003147.
27. Brunner, EJ, Shipley, MJ, Ahmadi-Abhari, S, et al. Midlife contributors to socioeconomic differences in frailty during later life: a prospective cohort study. Lancet Public Health 2018;3(7):e313-e322.
28. Johannesson, J, Rothenberg, E, Gustafsson, S, et al. Meal frequency and vegetable intake does not predict the development of frailty in older adults. Nutr Health 2019;25(1):21-28.
29. Marcos-Pérez, D, Sánchez-Flores, M, Proietti, S, et al. Association of inflammatory mediators with frailty status in older adults: results from a systematic review and meta-analysis. Geroscience 2020;42(6):1451-1473.
30. Soysal, P, Isik, AT, Carvalho, AF, et al. Oxidative stress and frailty: A systematic review and synthesis of the best evidence. Maturitas 2017;99:66-72.
31. Panche, AN, Diwan, AD, Chandra, SR. Flavonoids: an overview. J Nutr Sci 2016;5:e47.
32. Goisser, S, Guyonnet, S, Volkert, D. The Role of Nutrition in Frailty: An Overview. J Frailty Aging 2016;5(2):74-77.
33. Manal, B, Suzana, S, Singh, DK. Nutrition and Frailty: A Review of Clinical Intervention Studies. J Frailty Aging 2015;4(2):100-106.




S. Risbridger1, R. Walker1,2, W.K. Gray2, S.B. Kamaruzzaman3, C. Ai-Vyrn3, N.N. Hairi4, P.L. Khoo5, T.M.Pin3


1. Faculty of Medical Sciences, Newcastle University, UK; 2. Northumbria Healthcare NHS Foundation Trust, UK; 3. Department of Medicine, Faculty of Medicine, University of Malaya, Malaysia; 4. Department of Social and Preventive Medicine, Faculty of Medicine, University of Malaya, Malaysia; .5. Sports Centre, University of Malaya, Malaysia

Corresponding Author: Tan Maw Pin, Department of Medicine, Faculty of Medicine, University of Malaya, Malaysia, mptan@ummc.edu.my

J Frailty Aging 2021;in press
Published online August 27, 2021, http://dx.doi.org/10.14283/jfa.2021.31



Background: The global population is ageing rapidly, with the most dramatic increases in developing countries like Malaysia. Older people are at increased risk of multimorbidity, frailty and falls.
Objectives: In this study we aimed to determine the relationship between social participation, frailty and falls in Malaysia.
Design, Setting, and Participants: This was a cross-sectional study of individuals aged 55 years and above selected from the electoral rolls of three Klang Valley parliamentary constituencies through stratified random sampling. They were invited to take part in a questionnaire and physical assessment as part of the Malaysian Elders Longitudinal Research (MELoR) study.
Measurements: Fallers were individuals who had fallen in the previous year. Frailty was defined as meeting ≥3 of: low body mass index, reduced cognition, low physical activity, low hand-grip strength, and slow walking speed. Social participation was determined from employment status, social network, and community activity. Binomial logistic regression multivariant analysis was performed to identify links between the measures of social participation and falls and frailty.
Results: The mean age of the 1383 participants was 68.5 years, with 57.1% female. Within the population, 22.9% were fallers and 9.3% were frail. Social isolation (OR= 2.119; 95% CI=1.351-3.324), and non-engagement in community activities (OR=2.548; 95% CI=1.107-5.865) were associated with increased frailty. Falls increased with social isolation (OR=1.327; 95% CI=1.004-1.754).
Conclusions: Previous studies have shown social participation to be linked to frailty and falls risk, and social isolation to be a predictor of falls. In this study frailty was associated with all three social participation measures and history of falls was associated with social isolation.

Key words: Healthy ageing, frailty, falls, social isolation, community.



The world’s population is ageing rapidly and Asian countries are expected to see a much greater increase than their more developed counterparts (1). Already, the proportion of over-65s in the Malaysian population rose from 3.9% in 2000 to 6.7% in 2018This demographic transition is likely to lead to increased dependency, due to frailty, falls, and other age-related conditions, and requires a shift in health and social care provision (2-4).
Advancing age is associated with frailty as the result of global, cumulative, cyclical physiological decline. There is no single concept or phenotype to describe frailty due to its complex mixture of increasing vulnerability, decreasing cognition causing confusion and restricted mental capacity, and reduced physical abilities. Frailty can be completely reversible or improved through intervention (5). Frailty-related syndromes include falls, delirium, lack of mobility, incontinence and susceptibility to medication side effects (6). The prevalence of frailty in Malaysia has been reported as being between 5.7- 9.4% in the over 65s (3, 7-9). The longer individuals live, the greater their risk developing age-related conditions, with the population aged 80 years and above increasing the most rapidly and representing individuals at highest risk of falls and frailty (10).
A bidirectional relationship is likely to exist between falls and frailty. As one of the frailty-related syndromes, falls have multiple associated risk factors including: sarcopenia, impaired sensorimotor function, environmental factors such as home hazards, medication side effects, and cardiovascular diseases (11-13). Falls have been identified as the second leading cause of unintentional death globally (14). The prevalence of falls in over 55’s in Malaysia was reported as 18.9%, with many studies in South-East Asia reporting comparatively lower prevalence than in more developed countries (2). In their discussion on healthy ageing, the World Health Organization highlighted the multifactorial causes of falls; social interaction and community support was highlighted as an area for helping prevent falls (15). They described social participation as one of the key pillars of ‘Active Ageing’ and highlighted its necessity for healthy ageing (16, 15).
Whilst no universal definition of social participation exists it is often described as ‘a person’s involvement in activities providing interactions with others in society or the community’ (17, 18). The continuum of participation ranges from those with ‘active’ roles running organisations, to solely participating in these activities, and extends to employment. These active roles often require higher levels of interaction with the local community (19). This social participation is recognised as a major contributor to health with evidence demonstrating reduction in all-cause mortality (18).
Social participation may be reduced with increasing age (20). Social isolation is associated with mortality, hospitalisation, mild cognitive impairment (MCI), dementia, and many other health conditions (21-24). Protective mechanisms such as social inclusion and cohesion may decrease frailty prevalence and reduce the incidence of falls (25).
Social participation, frailty, and falls are related and in turn may influence healthy ageing and quality of life for older adults (26). In this study we aimed to gain insights into the relationship between social participation, frailty, and falls in the elderly Malaysian population.




This was an exploratory analysis using cross sectional data.


The MELoR study is a longitudinal cohort study based in Klang Valley, Kuala Lumpur. The initial study consisted of a 14-section questionnaire, which was undertaken at the participant’s home, and a clinical assessment completed at the University of Malaya Medical Centre. The questionnaire was administered by trained interviewers in English, Malay, Chinese, or Tamil, as appropriate to the participant.
Data collection occurred between November 2013 and October 2015. The electoral rolls of the Parliamentary constituencies of Petaling Jaya North, Petaling Jaya South, and Pantai Valley were used to determine the base population for the study. Individuals aged 55 years and above in 2013 were selected through simple random sampling, stratified by age deciles and ethnicity, and were included if they could provide informed consent. An age cut-off of 55 years was used as this was the mandatory retirement age in Malaysia at study initiation, though it has since increased to 60 years.
Stratification by ethnicity was for the three main ethnic groups in Malaysia: Malay, Chinese, and Indian. The exclusion criteria included those who were institutionalised, bed-bound, unable to access the research centre, or those who had communication difficulties, including severe cognitive impairment meaning they lacked the ability to answer the questionnaire. These initial cross-sectional data were used for this study, as the longitudinal follow up data were not yet available.
A total of 8769 individuals were invited to take part in the study, of whom 5815 were initially contactable, and 3334 met all the eligibility criteria.


An adapted version of the Fried frailty phenotype was operationalised based on the available data. The ‘shrinking’ element was defined in line with the HAALSI cohort study (27), using body mass index (BMI); a BMI lower than the WHO recommended 18.5k/m2 was given a positive score (28). The ‘poor endurance and energy’ element was replaced by a measure of cognitive impairment, the Montreal cognitive assessment (MoCA); any individual scoring ≤18 on the MoCA after educational adjustment was determined to have MCI, in line with the validation of the Bahasa Malaysia MoCA (29). The ‘low physical activity’ element was measured using the international physical activity questionnaire (IPAQ); the bottom quintile were given a positive score, ≤292.68 for males and ≤216.00 for females. The ‘slowness’ element was a test of walking 15ft; a cut-off of ≥7 seconds was used for males and females as this included all the bottom quintile. The ‘weakness’ element was a test of hand-grip strength (HGS); the bottom quintile was any male with right-handed HGS ≤21.83kg and any female with right-handed HGS ≤14.50kg. In keeping with the original methods, a score of three or more out of five was considered to signify a diagnosis of frailty (5). Adjustments to the original frailty phenotype were undertaken partly due to data availability. The decision to include the MoCA as a measure of cognitive frailty was in recognition of the fact that cognition does have a major effect on frailty. Lack of measurement of cognition is a key issue with Fried’s original score (27, 30). Participants were determined to be a faller if they had reported at least one fall in the previous 12 months.
Social participation was measured in three ways; as these do not form a coherent score, each were analysed separately and discussed individually where appropriate. Employment status was self-disclosed within the questionnaire: whether they had ever worked, and whether they were currently working. ‘Working’ included self-employment, family businesses, and private or governmental employment We wanted to highlight the difference within the population between those who were retired, as is common in this age group in Malaysia, and those who were unemployed and have never worked who may be at greater risk of conditions associated with ageing. The Lubben Social Network Scale-6 (LSNS-6) was used to determine which individuals were at risk of social isolation, as per the published guidelines (31). Levels of community activities (such as sports groups, clubs, societies, and involvement with non-governmental/ voluntary organisations) were separated into three groups: those who had no community activities ‘none’, those who had relatively ‘inactive roles’ or simply stated membership with no role attached, and those with ‘active roles’ in positions of responsibility such as a committee member or tutor. In the discussion, ‘social participation’ is used as an overarching concept of inclusion in society, measured through these three scales.

Statistical methods and data analysis

Data were analysed using SPSS Version 26 with the significance threshold set at p <0.05. Univariate analysis was performed using Pearson’s chi-squared statistical test of homogeneity. Significant results were then taken on to post hoc analysis with multiple z-tests of two proportions, with Bonferroni corrections, and binomial logistic regression multivariant analysis.
Participants missing demographic data were excluded from analysis as they could not be correctly differentiated within populations. Individuals missing data either for frailty or falls, but not both, were included in the analysis for the sections they had completed. The same process was applied for the three measures of social participation. Participants who did not have every item completed for the frailty score fell into two categories: those with enough data to assign ‘frail’ or ‘not frail’; those without enough available data who were therefore excluded from this part of the analysis. Assignment to a category was determined according to one of three criteria: they had already scored three so were ‘frail’; they had scored zero and were missing one or two items of data so were ‘not frail’; they had scored one and were missing one piece of data so were ‘not frail’.



The total study population was 1383 after removal of incomplete entries, duplications, and those who did not meet eligibility criteria. The cohort which were excluded in data cleaning had similar age, sex, ethnicities, levels of social isolation and participation, and number of fallers to the final sample used, according to the data available.
As shown in Table 1, the average age was 68.5 years, with the eldest participant aged 93 years. The largest ethnic group was Chinese. Only 2.9% had no formal education, and 73.9% had secondary education or higher. The mean BMI was 25.33 kg/m2, ranging from 13.62 to 59.63. Most of the population were not currently employed, although they had been previously. 28% were found to be at risk of social isolation. 33.1% reported being a member of a society, 10.6% were members of Non-Governmental Organisations or charities, and 18% had active roles in either activity.

Table 1. Basic characteristics and outcome data from the study population


As shown in Table 1, 126 (9.3%) of the study population were identified as ‘frail’, while for 2% of the population there were not enough data to determine their status. According to the specific frailty score calculated, 591 (45.1%) were deemed non-frail, 603 (46.0%) were pre-frail, and 117 (9%) frail. 311 (22.9%) were identified as a ‘faller’ (1.8% of the sample not completing this section) of whom 201 (66.8%) had only fallen once in the preceding 12 months.
117 (8.5%) participants had never worked, of whom 33 (28.9%) were fallers and 26 (23.2%) were frail. Of the 284 (20.7%) participants who were still currently employed, 53 (18.9%) were fallers and 12 (4.2%) were frail. There was a statistically significant difference in proportions of those who were frail when grouped by employment status, Χ2= 34.949, p <0.001. Regression analysis showed employment status no longer had a statistically significant effect on diagnosis of frailty when other variables had been considered, mainly educational attainment.
382 (28%) participants were deemed to be at risk of social isolation when using the LSNS-6; 101 (26.9%) were identified as fallers and 60 (16.3%) were frail. There was a statistically significant difference in frailty between the isolation classes, Χ2= 31.436, p <0.001. Regression analysis showed those at risk of social isolation had 2.119 (95% CI 1.351-3.324) times higher odds of being classed as frail, p =0.001, as shown in Fig 1.

Figure 1. Adjusted regression analysis for frailty. Odds ratio showing increased frailty compared to index category. Error bars representing 95% CI


In relation to social isolation, there was a statistically significant difference in the proportion of those who fell compared to those who did not, Χ2= 4.579, p=0.032. Regression analysis showed those at risk of social isolation had 1.327 (95% CI 1.004-1.754) times higher odds of falling, p =0.047, as shown in Fig 2.

Figure 2. Adjusted regression analysis for falls. Odds ratio showing increased falls compared to index category. Error bars representing 95% CI


843 (62%) participants were not involved in community groups; 201 (24.1) of these were fallers and 99 (12%) were frail. 244 (28%) had at least one active role with 48 (19.9%) of these being fallers and 7 (2.9) being frail. There was a statistically significant difference in proportions of those who were frail when grouped by involvement in community groups, Χ2= 26.139, p <0.001. Regression analysis showed that those not participating in community groups had 2.548 (95% CI 1.107-5.865) times higher odds of being frail than those with active roles, p =0.028 (seen in Fig 1).



Reduced social participation, measured through social isolation according to the LSNS-6 and absence of participation in community groups, was associated with frailty in this study. Falls were also associated with reduced social participation in this study specifically through the measure of increased social isolation.
The prevalence of frailty in this cohort was 9.3%, in keeping with prevalence estimates of between 5.7% and 9.4% from other studies (3, 7-9, 32). As this cohort included individuals aged 55 years and over, rather than the usually employed lower age cut-offs of 60 or 65 years, the prevalence was expected to be lower than previously reported. The relatively high prevalence may be due to the high proportion of both females and those aged ≥75 in the study, and the way in which Fried’s criteria were operationalised in our adapted assessment. As expected, age had a large impact on the diagnosis of frailty; 24.5% of those aged ≥75 were frail, with only 8.5% of individuals aged 70 to 74 years being considered frail. Badrasawi et al in 2016 studied a demographically similar population in Klang Valley, in people ≥60. They found a frailty prevalence of 8.9% in a smaller sample of 473 participants[8].
Frailty was found to be significantly associated with social isolation, according to the LSNS-6, and lack of participation in community groups. These both caused the odds of being frail to be two times higher in the adjusted models. Once adjusted for education, the relationship between employment status and frailty was no longer statistically significant. Community activities are often physically or mentally stimulating, which appears to maintain healthier levels of activity and cognitive function (18, 33).
Being involved in the community and maintaining social ties will provide a support system for older people, improve their mental wellbeing, and allow them to stay active; having objectives and a purpose in life has been shown to improve happiness and health (24, 23). The link between social activity and frailty has not been as well studied as other aspects of frailty. Few studies exist globally, but the positive nature of social participation is shown often (18, 34, 35). Developing systems to increase social participation, particularly in the oldest age groups, is likely to improve the lives of Malaysia’s ageing population.
Social isolation was associated with increased reported falls after adjusting for age, gender, and ethnicity. This fits with research by Pohl et al in the USA who determined that social isolation is a good predictor of falls, though participants had an average age 10 years older than in this study (22). No similar studies have been performed in Malaysia or within South-East Asia. Although our cross-sectional study does not suggest a causal link between social isolation and falling, social isolation may increase fear of falling or the consequences of a fall may be more severe, making the experience easier to recall. Many studies have highlighted the chain effect of falling, fearing falling, decreasing social activities, increasing isolation, and then having a greater risk of falling again[36]. An improved social circle, and regular contact with friends and family, could mean the individuals have better quality of life and reduced healthcare burden (34, 35).
The importance of the effects of social participation, and the risks of social isolation could not be more relevant to the current global situation. COVID-19 has led to many countries introducing social-distancing and ‘lockdown’ measures; it was estimated that in March 2020 over one-third of the global population had some form of restricted movement (37). Individuals who are older, have chronic health conditions, or are otherwise vulnerable are most likely to be facing the longest periods of isolation; social isolation in the elderly is already known to put them at risk of neurocognitive decline, cardiovascular disease, autoimmune problems, depression and anxiety, as well as an increase in mortality of nearly one-third (37-39). To add to that, an increase in frailty and falls will place increased demands on services. However, COVID-19 was not the trigger for social isolation of the older populations (but definitely worsened the problem), insights into the root causes of disintegration within communities and lack of opportunities for participation in the older age groups may inform future interventions and improve all aspects of health for the older population. Protecting this population from the future health risks caused by the social isolation from this pandemic would allow health systems to better prepare for similar future challenges. The WHO Global Network for Age-friendly Cities and Communities is an important example of current policy striving to better provide for our older populations in this way, and more initiatives such as this could have a major impact in improving quality of life (40).


Due to the retrospective nature of the study questionnaire, falls prevalence may have been underestimated, as recall bias can underestimate falls by 13-32% (12). This bias would also limit recall of engagement in community activities. As this study was performed cross-sectionally, reverse causation cannot be assessed, and so temporal relationships can be suggested but not proved.
We used an adapted Fried frailty score which has not been validated, and this may mean that the predictive value demonstrated by the original score is not reflected in this study. The frailty phenotype has several documented limitations including its reliance on physical traits and ignoring psychological and cognitive decline (41). The lack of availability of data about ‘exhaustion’ is the most limiting aspect of this operationalisation. Previous researchers have suggested links between the physical frailty described in Fried’s phenotype and cognitive decline, and the usefulness of adding a cognitive test to the phenotype, but there is no evidence that this could replace the missing element (42).
The younger cohorts included in this study may also have biased our findings. The older an individual is, the more likely they are to be affected by falls and frailty, and so including down to the ages of 55 may have hidden some key details.

Recommendations for future work

A longitudinal cohort study could help determine whether social isolation in fact leads to frailty or if frail individuals are less able to participate in current society. Whether greater participation in society before a person retires could reduce their future risk of frailty and falls is an important link to the multifactorial aetiology of frailty.
Creation of a universally agreed measurement of social participation and frailty would make research more relatable to the global population. A score which identifies the key areas of social participation, while being accessible to countries with poor healthcare infrastructure and funding, would best provide a global measurement of participation to allow results to be directly compared between populations. A universal consensus for the diagnosis of frailty would also mean results could be directly compared globally.



Reduced social participation, measured with reduced social network and absence of participation in community groups, is associated with the frailty phenotype in individuals aged 55 years and over included in this study. Falls, as an outcome of frailty, also shows an association with reduced social participation through increased social isolation. Within developing nations, social participation may be an affordably modifiable risk factor for frailty and its associated syndromes, which would in turn improve quality of life and support healthy ageing.


Ethics approval: The MELoR study was approved by the University of Malaya Medical Centre Medical Ethics Committee (Ref: 925.4).

Authors’ contributions: Material preparation and data collection were performed by Shahrul B Kamaruzzaman, Chin Ai-Vyrn, Noran N Hairi, Phaik Lin Khoo, and Tan Maw Pin. Analysis was performed by Sophie Risbridger, under the guidance of Richard Walker and William Keith Gray. The manuscript was written by Sophie Risbridger and all authors commented on versions of the manuscript.

Funding: This work was supported by the Ministry of Health, Malaysia, Long-term Research Grant Scheme [grant LR005-2019]. The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.



1. United Nations Department of Economic and Social Affairs. World Population Ageing 2017- Highlights2017 2017.
2. Alex D, Khor HM, Chin AV, Hairi NN, Othman S, Khoo SPK et al. Cross-sectional analysis of ethnic differences in fall prevalence in urban dwellers aged 55 years and over in the Malaysian Elders Longitudinal Research study. BMJ Open. 2018;8(7):e019579. doi:10.1136/bmjopen-2017-019579.
3. Ahmad NS, Hairi NN, Said MA, Kamaruzzaman SB, Choo WY, Hairi F et al. Prevalence, transitions and factors predicting transition between frailty states among rural community-dwelling older adults in Malaysia. PLoS One. 2018;13(11):Online. doi:10.1371/journal.pone.0206445.
4. Crome P, Lally F. Frailty: joining the giants. Canadian Medical Association Journal. 2011;183(8):889-90. doi:10.1503/cmaj.110626.
5. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J et al. Frailty in Older Adults: Evidence for a Phenotype. The Journals of Gerontology: Series A. 2001;56(3):M146-57. doi:10.1093/gerona/56.3.M146.
6. British Geriatrics Society, Age UK, Royal College of General Practitioners. Fit for frailty- Consensus best practice guidance for the care of older people living in community and outpatient settings. London2014.
7. Sathasivam J, Kamaruzzaman SB, Hairi F, Ng CW, Chinna K. Frail Elders in an Urban District Setting in Malaysia: Multidimensional Frailty and Its Correlates. Asia Pacific Journal of Public Health. 2015;27(8 Suppl):52S-61S. doi:10.1177/1010539515583332.
8. Badrasawi M, Shahar S, Singh DKA. Risk Factors of Frailty Among Multi-Ethnic Malaysian Older Adults. International Journal of Gerontology. 2017;11(3):154-60. doi:10.1016/j.ijge.2016.07.006.
9. Siriwardhana DD, Hardoon S, Rait G, Weerasinghe MC, Walters KR. Prevalence of frailty and prefrailty among community-dwelling older adults in low-income and middle-income countries: a systematic review and meta-analysis. BMJ Open. 2018;8(3):e018195. doi:10.1136/bmjopen-2017-018195.
10. Tinetti M, Huang A, Molnar F. The Geriatrics 5M’s: A New Way of Communicating What We Do. J Am Geriatr Soc. 2017;65(9):2115. doi:10.1111/jgs.14979.
11. Caird FI, Andrews GR, Kennedy RD. Effect of posture on blood pressure in the elderly. British Heart Journal. 1973;35(5):527-30. doi:10.1136/hrt.35.5.527.
12. Masud T, Morris RO. Epidemiology of falls. Age and Ageing. 2001;30 Suppl 4:3-7. doi:10.1093/ageing/30.suppl_4.3.
13. Goh CH, Ng SC, Kamaruzzaman SB, Chin AV, Poi PJ, Chee KH et al. Evaluation of Two New Indices of Blood Pressure Variability Using Postural Change in Older Fallers. Medicine (Baltimore). 2016;95(19):e3614. doi:10.1097/MD.0000000000003614.
14. WHO. Falls. In: Fact Sheet. World Health Organisation, Online. 2018. https://www.who.int/news-room/fact-sheets/detail/falls. Accessed 08/04/2020 2020.
15. WHO. World Report on Ageing and Health. Geneva: World Health Organisation; 2015 30/09/2015.
16. WHO. Active Ageing: A Policy Framework. Geneva: World Health Organisation;2002 04/2002.
17. Eriksson BG. Ordinal dispersion of ratings of social participation as a function of age from 70 years of age among the H-70 panel, Gothenburg, Sweden. Archives of Gerontology and Geriatrics. 2008;47(2):229-39. doi:10.1016/j.archger.2007.08.007.
18. Okura M, Ogita M, Yamamoto M, Nakai T, Numata T, Arai H. Community activities predict disability and mortality in community-dwelling older adults. Geriatrics and Gerontology International. 2018;18(7):1114-24. doi:10.1111/ggi.13315.
19. Aw S, Koh G, Oh YJ, Wong ML, Vrijhoef HJM, Harding SC et al. Explaining the continuum of social participation among older adults in Singapore: from ‘closed doors’ to active ageing in multi-ethnic community settings. Journal of Aging Studies. 2017;42:46-55. doi:10.1016/j.jaging.2017.07.002.
20. Liu JY. The severity and associated factors of participation restriction among community-dwelling frail older people: an application of the International Classification of Functioning, Disability and Health (WHO-ICF). BMC Geriatriatrics. 2017;17(1):43. doi:10.1186/s12877-017-0422-7.
21. Rawtaer I, Gao Q, Nyunt MS, Feng L, Chong MS, Lim WS et al. Psychosocial Risk and Protective Factors and Incident Mild Cognitive Impairment and Dementia in Community Dwelling Elderly: Findings from the Singapore Longitudinal Ageing Study. Journal of Alzheimer’s Disease. 2017;57(2):603-11. doi:10.3233/JAD-160862.
22. Pohl JS, Cochrane BB, Schepp KG, Woods NF. Falls and Social Isolation of Older Adults in the National Health and Aging Trends Study. Research in Gerontological Nursing. 2018;11(2):61-70. doi:10.3928/19404921-20180216-02.
23. Courtin E, Knapp M. Social isolation, loneliness and health in old age: a scoping review. Health and Social Care in the Community. 2017;25(3):799-812. doi:10.1111/hsc.12311.
24. Uchino BN. Social support and health: a review of physiological processes potentially underlying links to disease outcomes. Journal of Behavioural Medicine. 2006;29(4):377-87. doi:10.1007/s10865-006-9056-5.
25. Chon D, Lee Y, Kim J, Lee KE. The Association between Frequency of Social Contact and Frailty in Older People: Korean Frailty and Aging Cohort Study (KFACS). Journal of Korean Medical Sciences. 2018;33(51):e332. doi:10.3346/jkms.2018.33.e332.
26. Hayashi T, Umegaki H, Makino T, Huang CH, Inoue A, Shimada H et al. Combined Impact of Physical Frailty and Social Isolation on Rate of Falls in Older Adults. The journal of nutrition, health & aging. 2020;24(3):312-8. doi:10.1007/s12603-020-1316-5.
27. Payne CF, Wade A, Kabudula CW, Davies JI, Chang AY, Gomez-Olive FX et al. Prevalence and correlates of frailty in an older rural African population: findings from the HAALSI cohort study. BMC Geriatriatrics. 2017;17(1):293. doi:10.1186/s12877-017-0694-y.
28. Global Health Observatory. Mean Body Mass Index (BMI). In: GHO Data. World Health Organisation. 2009. https://www.who.int/gho/ncd/risk_factors/bmi_text/en/. Accessed 01/04/2020 2020.
29. Normah CD, Shahar S, Zulkifli BH, Razali R, Chin A, Omar A. Validation and Optimal Cut-Off Scores of the Bahasa Malaysia Version of the Montreal Cognitive Assessment (MoCA-BM) for Mild Cognitive Impairment among Community Dwelling Older Adults in Malaysia. Sains Malaysiana. 2016;45(9):1337-43.
30. Lewis EG, Coles S, Howorth K, Kissima J, Gray WK, Urasa S et al. The prevalence and characteristics of frailty by frailty phenotype in rural Tanzania. BMC Geriatrics. 2018;18(1). doi:10.1186/s12877-018-0967-0.
31. Lubben J, Blozik E, Gillmann G, Iliffe S, von Renteln Kruse W, Beck JC et al. Performance of an abbreviated version of the Lubben Social Network Scale among three European community-dwelling older adult populations. The Gerontologist,. 2006;46(4):503-13. doi:10.1093/geront/46.4.503.
32. Gray WK, Richardson J, McGuire J, Dewhurst F, Elder V, Weeks J et al. Frailty Screening in Low- and Middle-Income Countries: A Systematic Review. Journal of the American Geriatrics Society. 2016;64(4):806-23. doi:10.1111/jgs.14069.
33. Okura M, Ogita M, Yamamoto M, Nakai T, Numata T, Arai H. The relationship of community activities with cognitive impairment and depressive mood independent of mobility disorder in Japanese older adults. Arch Gerontol Geriatr. 2017;70:54-61. doi:10.1016/j.archger.2016.12.010.
34. Pereira GN, Morsch P, Lopes DG, Trevisan MD, Ribeiro A, Navarro JH et al. Social and environmental factors associated with the occurrence of falls in the elderly. Ciencia and Saude Coletiva. 2013;18(12):3507-14. doi:10.1590/s1413-81232013001200007.
35. Hajek A, Konig HH. The association of falls with loneliness and social exclusion: evidence from the DEAS German Ageing Survey. BMC Geriatriatrics. 2017;17(1):204. doi:10.1186/s12877-017-0602-5.
36. Petersen N, König H-H, Hajek A. The link between falls, social isolation and loneliness: A systematic review. Archives of gerontology and geriatrics. 2020;88:104020.
37. Lippi G, Henry BM, Bovo C, Sanchis-Gomar F. Health risks and potential remedies during prolonged lockdowns for coronavirus disease 2019 (COVID-19). Diagnosis (Berlin). 2020;7(2):85-90. doi:10.1515/dx-2020-0041.
38. Armitage R, Nellums LB. COVID-19 and the consequences of isolating the elderly. Lancet Public Health. 2020;5(5):e256. doi:10.1016/S2468-2667(20)30061-X.
39. Douglas M, Katikireddi SV, Taulbut M, McKee M, McCartney G. Mitigating the wider health effects of covid-19 pandemic response. BMJ. 2020;369:m1557. doi:10.1136/bmj.m1557.
40. WHO. WHO Global Network for Age-friendly Cities and Communities. World Health Organisation; , Online. 2016. https://www.who.int/ageing/projects/age_friendly_cities_network/en/. Accessed 07/01/2021 2021.
41. Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. The Lancet. 2013;381(9868):752-62. doi:10.1016/S0140-6736(12)62167-9.
42. Ma L, Zhang L, Sun F, Li Y, Tang Z. Cognitive function in Prefrail and frail community-dwelling older adults in China. BMC Geriatrics. 2019;19(1):53. doi:10.1186/s12877-019-1056-8.




D.E. Brazier1, N. Perneta2, F.E. Lithander1, E.J. Henderson1,2


1. Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, United Kingdom; 2. Royal United Hospital Bath NHS Foundation Trust, Bath, United Kingdom;

Corresponding Author: Danielle Brazier, Population Health Sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol, BS8 1NU, United Kingdom;Email : danielle.brazier@bristol.ac.uk

J Frailty Aging 2021;in press
Published onlineAugust 27, 2021, http://dx.doi.org/10.14283/jfa.2021.30



COVID-19 disproportionately affects older people, with higher rates of infection and a higher risk of adverse outcomes. A brief review of literature was undertaken to inform development of a protocol describing the indications and process of prone positioning to aid the management of COVID-19 infection in non-mechanically ventilated, awake older adults. PubMed was searched up to 14th January 2021 to identify English language papers that described prone positioning procedures used in non-mechanically ventilated patients. Data were pooled to inform the development of a prone positioning protocol for use in hospital ward environments. The protocol was trialled and refined during routine clinical practice. Screening of 146 articles yielded five studies detailing a prone positioning protocol. Prone positioning is a potentially feasible and tolerated treatment adjunct for hypoxaemia in older adults with COVID-19. Future studies should further establish the efficacy, safety, and tolerability in respiratory illnesses in non-intensive care settings.

Key words: Prone position, COVID-19, older adults, non-pharmaceutical interventions, pandemic.



Older adults are disproportionately affected by COVID-19 infection (1). Age is an independent predictor of mortality (2). The presence of co-morbidities (3) as well as age-related changes in physiology (4) contribute to this risk. Furthermore, older adults are less likely to present with the typically described symptoms of dyspnoea, anosmia, cough and fever but rather with atypical symptoms such as delirium (5) which may negatively impact time to diagnosis and subsequent treatment (6). The World Health Organisation issued guidance in March 2020 (updated May 2020) on the management of COVID-19 infection in older adults (7), which has been mirrored in UK guidance (8, 9). It recommended that all older adults are screened for COVID-19 when accessing healthcare and medications should be reviewed. It advocates establishing whether an advance care plan is in place and working within a multidisciplinary team. The mainstay of current management in hospital ward environments comprises oxygen, dexamethasone (7, 10), and potentially remdesivir if disease is severe (11). Respiratory decompensation occurs around day 10 of illness (12) whereby COVID-19 associated Acute Respiratory Distress Syndrome (ARDS) manifests as profound hypoxaemia often in the absence of apparent dyspnoea (13, 14).
Prone positioning was first described as a therapeutic strategy to relieve hypoxia in the 1970s (15). Randomised controlled trials have established that it is associated with improved outcomes in those with ARDS and respiratory failure (16–18). Prone positioning is associated with increased end-expiratory lung volume, alveolar recruitment and oxygenation (19), reduction in pressure differential across anterior and posterior lung tissue and improvement in ventilation/perfusion mismatch, which is exacerbated by ARDS (20). It is commonly used during anaesthesia and during the immediate post-operative period in patients who are intubated and ventilated (21). Cost effectiveness has been established in the Intensive Care Unit (ICU) setting in those with severe ARDS (22). During the COVID-19 pandemic, small studies, specifically in awake patients, suggest that prone ventilation is feasible in Emergency Department (ED) settings (23, 24), ICU (25–32), and in ward-based settings (33–37) with participants receiving standard oxygen therapy, high-flow nasal oxygen, or non-invasive ventilation in the form of continuous positive airway pressure.
Prone positioning has potential utility as an adjunct to a rather limited repertoire of treatment strategies available for older adults in hospital with COVID-19 infection. We sought to perform a brief review of the literature to develop a protocol for prone positioning that would assimilate existing evidence, particularly considering the specific needs and requirements of older adults. On this basis, we developed a prone positioning protocol designed to be used specifically in older adults with COVID-19 infection cared for in standard ward environments. We tailored this protocol specifically to the needs of older adults receiving supplemental oxygen via nasal cannuale, face mask air-entrainment mask or non-rebreathe mask in normal ward settings recognising that this will be relevant to a significant proportion of older people. Additional measures already described elsewhere (25–32) for proning in settings with facilities and expertise that permit more invasive respiratory support.



A search of PubMed was carried out up to 14th January 2021 using the following search criteria:
(prone position[MeSH terms]) OR (Prone [MeSH Terms]) AND (Aged [MeSH Terms]) OR (Frail Older adult [MeSH terms]) OR (Elderly [MeSH terms]). Abstracts were searched to determine papers that described the process of proning, efficacy and/or considerations in non-intubated older adults outside of ICU settings and potential complications. Combinations using COVID-19 or coronavirus, sars [MeSH terms] did not retrieve any additional relevant results. Further articles were identified through reference and citation lists. Results were collated to inform the development of a proning protocol suitable for use in older adults.



Of the 146 articles on the use of prone positioning in patients with respiratory conditions that were identified, 32 were relevant to non-mechanically ventilated patients but only five provided details of a proning procedure. Of the remaining 27 studies, two focused on nerve injury relating to prone positioning; a case study (38) and a case series (39). The case series found that 15% of 83 patients admitted to inpatient rehabilitation facilities post-COVID-19 were diagnosed with peripheral nerve injury, 92% of whom had been proned in acute care. Five studies examined patients with ARDS (22, 40–43).
The remaining 25 studies included patients with COVID-19 infection supported with non-invasive ventilation. One study sought to describe sputum characteristics of patients with severe COVID-19, and to determine the effect of airway clearance methods on outcomes in these patients (47) whilst another examined the expansion of the lungs in supine versus prone in COVID-19 using CT scan (48).
The optimal duration of proning has not been determined. Three studies reported a duration of 3 hours or fewer (24, 28, 32), 5 studies reported a duration over 3 hours per day (26, 29, 30, 36, 50), and 3 reported a range across patients: <1hour, 1-3 hours and >3 hours (34), 1-16 hours (25), up to 24 hours a day (35) and 5 studies did not report the duration of proning (27, 31, 33, 37, 49). The results of the studies which did report duration of proning did not provide conclusive evidence to suggest a consensus on optimal duration in the COVID-19 population, although studies that compared duration show a trend towards longer duration being of greater benefit (34, 35).
Five of the 32 studies provided significant detail on the methods of a proning procedure (24, 36, 44–46) and were used to develop our protocol. Two tested the protocol in patients with non-invasive ventilation (24, 36), and 3 described a protocol but did not test it in patients (44–46). Using the data from these 5 articles and from relevant clinical guidelines (51–54), coupled with clinical experience, a protocol was designed by a multidisciplinary team with expertise in caring for older adults with acute COVID-19 infection.

Table 1. Characteristics of the five studies used to inform the development of the proning protocol

Abbreviations: N/A: not applicable; ARDS: acute respiratory distress syndrome; ICU: intensive care unit; US: ultrasound scan; CT: computerised tomography; NIV: non-invasive ventilation; CPAP: continuous positive airway pressure; PP: prone positioning; ED: emergency department; COVID-19: Coronavirus disease 2019


This protocol, shown in Figure 1, covers the indications, proning manoeuvre and requisite monitoring. The proning manoeuvre to assist an individual from supine to prone is shown in Figure 2.

Figure 1. Prone positioning of older adults with COVID-19 infection

Figure 2. Prone positioning manoeuvre

Step 1 – Patient selection and suitability

Prone positioning can be trialled for patients who require supplemental oxygen to maintain saturations ≥92% (or ≥88% in the presence of hypercapnic respiratory failure), without severe delirium or impairment in cognition that would preclude compliance with the procedure. Contraindications include spinal instability, unstable pelvic or facial fracture, anterior open wounds or burns, and raised intracranial pressure. In the presence of any relative contraindications, including head injury, uncontrolled seizures, raised intraocular pressure, cardiovascular instability, delirium and morbid obesity, clinical judgement should be used to balance the potential risks and benefits which should be discussed with the patient wherever possible. The relative risks and benefits should be explained, and consent gained where feasible.

Step 2 – Assemble equipment

Up to five staff members may be necessary depending on the degree of assistance required, and appropriate personal protective equipment (PPE) should be worn. The patient’s ability to rotate their cervical spine and head to 45-90° should be checked. All clothing and jewellery should be removed to minimise the risk of a subsequent pressure injury. In advance of the manoeuvre, adequate tubing length and positioning should be checked anticipating the eventual prone position. Suction should be available and functioning. Pre-oxygenation with high flow oxygen via a non-rebreathe mask may be considered where exertion leads to critical desaturation. Given the higher risk to this population of developing pressure injuries, it is strongly recommended that patients should be on an air mattress. Bladder catheter tubing and bags should be placed on the bed, between the legs, rather than affixed to the bedside.

Step 3 – Recording of observations

Older adults are more prone to pressure injury and therefore specific care should be taken of the face, malar area, ears, eyes, shoulders, elbows, breasts, iliac crests, knees and toes (55). Particular attention should be paid to the bridge of the nose for those wearing a face mask, the columella for those wearing nasal cannulae, and the tops of the ears in both instances. Prior to the proning manoeuvre, it is recommended that anterior electrocardiographic electrodes are removed if in situ. These can be reapplied to the posterior chest post-manoeuvre if electrocardiographic monitoring is required. Skin integrity and heart rate, blood pressure, oxygen saturations, temperature, respiratory rate and conscious level should be monitored (51, 54).

Step 4 – Proning manoeuvre

Patients should be encouraged to position themselves independently where they are able. If assistance is required, one individual should co-ordinate the manoeuvre which is achieved by means of a slidesheet-assisted manual handling technique (Figure 2). Once positioning, secretions should be suctioned, and bed angled to 10° (reverse Trendelenburg) which increases adherence, and reduces aspiration risk (52).

Step 5 – Monitoring

Once in the prone position, observations should be recorded at 15 minute intervals for the first hour and according to clinical judgement thereafter. The optimal duration of proning is uncertain with studies of proning in COVID-19 infection reporting a duration of 1-21 hours per day (34, 36). Establishing treatment goals early on will allow the team to assess whether the patient is responding to the proning protocol. There is no strong evidence base to guide the duration after which an individual can be deemed to have responded. Therefore response should be assessed according to clinical judgement with regular proning cycles implemented thereafter in those with satisfactory response. Our experience has been to prone patients as tolerated with shorter recurrent periods prone often facilitating adequate nutrition and hydration at mealtimes. Hourly repositioning, alternating flexion and extension of the arms, with head turned “face facing hand” should be undertaken. If the patient is unable to comply with the positioning requirements we do not advocate the use of physical restraints. Eyes should be lubricated, and face skin should be protected with hyper-oxygenated fatty acids and silicone dressings. Pillows can offload bony prominences such as shoulders, knees, toes and iliac crests and support the chest (52). Ensure oxygen delivery systems are correctly fitted and not too tight across pressure areas.

Step 6 – Post-manoeuvre

After repositioning, observations should be recorded, monitoring particularly for hypotension which can occur during sudden positional changes (56, 57).

Risks and special considerations

If the patient is receiving nutrition via nasogastric (NG) tube, this should be discontinued or aspirated at least one hour prior to the manoeuvre and prone, feed can be restarted at 10ml/h.
Recognised complications of proning include brachial plexus injury (38, 39), pressure ulcers (58, 59) and hypotension on returning to supine (57). As such, regular assessment of pain using a recognised pain score is useful. Proning precludes anterior chest wall observation and there is a risk that recognition of deterioration can be delayed. Whilst successful resuscitation is described in the prone position (60) we advocate that the patient is immediately returned to supine should this need arise. Staffing levels may be a significant factor in limiting feasibility in ward environments, and as such proning should be scheduled to allow for sufficient monitoring.



To the best of our knowledge, this is the first prone positioning protocol that has been specifically designed for older adults with COVID-19 who are cared for on geriatric or general medical wards. Our approach has been to combine a rapid review of the literature with our pragmatic experience of utilising proning as an adjunctive nonpharmacological therapy for treating older inpatients with COVID-19. The articles identified in the literature search provided information on patient inclusion and exclusion criteria (45, 46), patient positioning (24, 46), duration of proning (36), and physiological monitoring (44, 46) but these were not specific to older adults nor hospital ward settings. Relevant data were assimilated and integrated with specific knowledge of the care of older people to develop a protocol that specifically considered the high likelihood of concurrent delirium, cognitive impairment, limitations in mobility, comorbid conditions (such as fractures), pressure injury risk and tolerability.
When compared with other protocols, our protocol provides a comprehensive checklist, five detailed images which display the patient, equipment and the manoeuvre itself, and a clear description of indications and requisite monitoring pre-, peri- and post-maneuvre. Mitchell & Secka (2018) provide a thorough checklist (44), and Bastoni et al., (2020) provide two images and video footage of the procedure being undertaken (24). None of the five studies however combine all aspects. In addition, whilst Raoof et al., (2020) highlighted considerations such as pressure points (45), our protocol includes specific considerations in older adults, highlighting the complications which may arise including pressure injury.
We recognise the limitations of our approach has been to adapt protocols utilised in ICU settings (44), ED (46) or in ward-based settings (24, 36) not specific to older people. We anticipate that the protocol may be utilised as an adjunctive treatment for other respiratory conditions and that the optimal timing of initiation, duration of proning and effectiveness will be further established with future research.



Our experience suggests that with careful patient selection, proning is a feasible, well-tolerated procedure for those who are alert, willing and able to comply. Its use is not precluded by the presence of delirium or cognitive impairment so long as clinical judgement is exercised. We anticipate that this protocol may improve the safety and efficacy of prone positioning in older people.


Acknowledgements: We would like to acknowledge the constructive and helpful comments provided by the Reviewers.

Funding: DB, EJH and FEL receive salary support from The Gatsby Foundation. EJH has received grants from Parkinson’s UK, NIHR and The Gatsby Foundation, and has received travel consultancy and honoraria from Profile, Bial, Abbvie, Luye, Ever and Simbec Orion. The funders played no role in the design, execution, analysis and interpretation of the data, or writing of the study.

Conflicts of Interest: DEB & FEL report grants from The Gatsby Foundation during the conduct of the study; NP has nothing to disclose; EJH reports grants from The Gatsby Foundation, The Dunhill Trust, and National Institute of Health Research, personal fees from Bial, Abbvie, Ever, Profile pharma, and Luye, outside the submitted work.



1. Docherty AB, Harrison EM, Green CA, et al. Features of 20 133 UK patients in hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: Prospective observational cohort study. BMJ. 2020;369(March):1-12. doi.org/10.1136/ bmj.m1985.
2. Mehraeen E, Karimi A, Barzegary A, et al. Predictors of mortality in patients with COVID-19-a systematic review. Eur J Integr Med. 2020;40:101226. doi.org/10.1016/j.eujim.2020.101226.
3. Davis JW, Chung R, Juarez DT. Prevalence of comorbid conditions with aging among patients with diabetes and cardiovascular disease. Hawaii Med J. 2011;70(10):209-213.
4. Boss GR, Seegmiller JE. Age-related physiological changes and their clinical significance. West J Med. 1981;135(6):434-440.
5. Gan JM, Kho J, Akhunbay-Fudge M, et al. Atypical presentation of COVID-19 in hospitalised older adults. Ir J Med Sci. Published online 2020. doi.org/10.1007/s11845- 020-02372-7.
6. Davis P, Gibson R, Wright E, et al. Atypical presentations in the hospitalised older adult testing positive for SARS-CoV-2: a retrospective observational study in Glasgow, Scotland. Scott Med J. Published online 2020. doi.org/10.1177/0036933020962891.
7. World Health Organisation. Clinical Management of COVID-19 – WHO Interim Guidance.; 2020.
8. National Institute For Health and Care Excellence (NICE). COVID-19 rapid guideline: critical care in adults. Natl Inst Heal Care Excell. 2020;(March):2020.
9. BMJ. Coronavirus Disease 2019 (Covid-19) – BMJ Best Practice. Vol 2019.; 2020.
10. National Institute for Health and Care Excellence. COVID-19 Prescribing Briefing : Corticosteroids COVID-19 Prescribing Briefing : Corticosteroids.; 2020.
11. National Institute for Health and Care Excellence. COVID-19 Rapid Evidence Summary : Remdesivir for Treating Hospitalised Patients with Suspected or Confirmed COVID-19 Key Messages.; 2020.
12. Ye Q, Wang B, Mao J. The pathogenesis and treatment of the ‘Cytokine Storm’’ in COVID-19.’ J Infect. 2020;80(6):607-613. doi.org/10.1016/j.jinf.2020.03.037.
13. Ragab D, Salah Eldin H, Taeimah M, Khattab R, Salem R. The COVID-19 Cytokine Storm; What We Know So Far. Front Immunol. 2020;11(June):1-4. doi.org/10.3389/fimmu.2020.01446.
14. Tobin MJ, Laghi F, Jubran A. Why COVID-19 silent hypoxemia is baffling to physicians. Am J Respir Crit Care Med. 2020;202(3):356-360. doi.org/10.1164/rccm.202006-2157CP.
15. Piehl MA, Brown RS. Use of extreme position changes in acute respiratory failure. Crit Care Med. 1976;4(1):13-14. doi.org/10.1097/00003246-197601000-00003.
16. Taccone P, Pesenti A, Latini R, et al. Prone positioning in patients with moderate and severe acute respiratory distress syndrome: A randomized controlled trial. JAMA – J Am Med Assoc. 2009;302(18):1977-1984. doi.org/10.1001/jama.2009.1614.
17. Guérin C, Reignier J, Richard J-C, et al. Prone Positioning in Severe Acute Respiratory Distress Syndrome. N Engl J Med. 2013;368(23):2159-2168. doi.org/10.1056/nejmoa1214103.
18. Mancebo J, Fernández R, Blanch L, et al. A multicenter trial of prolonged prone ventilation in severe acute respiratory distress syndrome. Am J Respir Crit Care Med.2006;173(11):1233-1239. doi.org/10.1164/rccm.200503-353OC.
19. Dirkes S, Dickinson S, Havey R, O’Brien D. Prone positioning: Is it safe and effective? Crit Care Nurs Q. 2012;35(1):64-75. doi.org/10.1097/CNQ.0b013e31823b20c6.
20. Koulouras V, Papathanakos G, Papathanasiou A, Nakos G. Efficacy of prone position in acute respiratory distress syndrome patients: A pathophysiology-based review. World J Crit Care Med. 2016;5(2):121. doi.org/10.5492/wjccm.v5.i2.121.
21. Furukawa H, Sato H, Hashizume K, et al. [Clinical Efficacy of Prone Positioning in Elderly Patients with Respiratory Failure after Thoracic Aortic Surgery]. Kyobu Geka. 2018;71(8):583-586.
22. Baston CM, Coe NB, Guerin C, Mancebo J, Halpern S. The Cost-Effectiveness of Interventions to Increase Utilization of Prone Positioning for Severe Acute Respiratory Distress Syndrome. Crit Care Med. 2019;47(3):e198-e205. doi.org/10.1097/ CCM.0000000000003617.
23. Caputo ND, Strayer RJ, Levitan R. Early Self-Proning in Awake, Non-intubated Patients in the Emergency Department: A Single ED’s Experience During the COVID-19 Pandemic. Acad Emerg Med. 2020;27(5):375-378. doi.org/10.1111/acem.13994.
24. Bastoni D, Poggiali E, Vercelli A, et al. Prone positioning in patients treated with non-invasive ventilation for COVID-19 pneumonia in an Italian emergency department. Emerg Med J. 2020;37(9):565-566. doi.org/10.1136/emermed-2020-209744
25. Despres C, Brunin Y, Berthier F, Pili-Floury S, Besch G. Prone positioning combined with high-flow nasal or conventional oxygen therapy in severe Covid-19 patients. Crit Care. 2020;24(1):256. doi.org/10.1186/s13054-020-03001-6.
26. González-Castro A, Escudero-Acha P, Arnaiz F, Ferrer D. High-flow oxygen therapy with spontaneous breathing prono position in SARS-CoV-2 pneumonia. Rev Esp Anestesiol Reanim. 2020;67(9):529-530. doi.org/10.1016/j.redar.2020.05.014.
27. Hallifax RJ, Porter BM, Elder PJ, et al. Successful awake proning is associated with improved clinical outcomes in patients with COVID-19: single-centre high- dependency unit experience. BMJ open Respir Res. 2020;7(1):1-7. doi.org/10.1136/bmjresp-2020-000678.
28. Taboada M, González M, Álvarez A, et al. Effectiveness of Prone Positioning in Nonintubated Intensive Care Unit Patients With Moderate to Severe Acute Respiratory Distress Syndrome by Coronavirus Disease 2019. Anesth Analg. 2021;132(1):25-30. doi.org/10.1213/ANE.0000000000005239.
29. Slessarev M, Cheng J, Ondrejicka M, Arntfield R. Patient self-proning with high- flow nasal cannula improves oxygenation in COVID-19 pneumonia. Can J Anesth. 2020;67(9):1288-1290. doi.org/10.1007/s12630-020-01661-0.
30. Ferrando C, Mellado-Artigas R, Gea A, et al. Awake prone positioning does not reduce the risk of intubation in COVID-19 treated with high-flow nasal oxygen therapy: A multicenter, adjusted cohort study. Crit Care. 2020;24(1):1-11. doi.org/10.1186/s13054-020-03314-6.
31. Damarla M, Zaeh S, Niedermeyer S, et al. Prone positioning of nonintubated patients with COVID-19. Am J Respir Crit Care Med. 2020;202(4):604-606. doi.org/10.1164/ rccm.202004-1331LE.
32. Retucci M, Aliberti S, Ceruti C, et al. Prone and Lateral Positioning in Spontaneously Breathing Patients With COVID-19 Pneumonia Undergoing Noninvasive Helmet CPAP Treatment. Chest. 2020;158(6):2431-2435. doi.org/10.1016/j.chest.2020.07.006
33. Taboada M, Rodríguez N, Riveiro V, Baluja A, Atanassoff PG. Prone positioning in awake non-ICU patients with ARDS caused by COVID-19. Anaesth Crit Care Pain Med. 2020;39(5):581-583. doi.org/10.1016/j.accpm.2020.08.002.
34. Elharrar X, Trigui Y, Dols A-M, et al. Use of Prone Positioning in Nonintubated Patients With COVID-19 and Hypoxemic Acute Respiratory Failure. JAMA. 2020;323(22):2336-2338. doi.org/10.1001/jama.2020.8255.
35. Thompson AE, Ranard BL, Wei Y, Jelic S. Prone Positioning in Awake, Nonintubated Patients With COVID-19 Hypoxemic Respiratory Failure. JAMA Intern Med. 2020;180(11):1537. doi.org/10.1001/jamainternmed.2020.3030
36. Ng Z, Tay WC, Benjamin Ho CH. Awake prone positioning for non-intubated oxygen dependent COVID-19 pneumonia patients. Eur Respir J. 2020;56(1):0-5. doi.org/10.1183/13993003.01198-2020.
37. Ripoll-Gallardo A, Grillenzoni L, Bollon J, Della Corte F, Barone-Adesi F. Prone Positioning in Non-Intubated Patients with COVID-19 Outside of the Intensive Care Unit: More Evidence Needed. Disaster Med Public Health Prep. 2020;14(4):E22-E24. doi.org/10.1017/dmp.2020.267.
38. Jiang LG, LeBaron J, Bodnar D, et al. Conscious Proning: An Introduction of a Proning Protocol for Nonintubated, Awake, Hypoxic Emergency Department COVID-19 Patients. Acad Emerg Med. 2020;27(7):566-569. doi:10.1111/acem.14035
39. Mitchell DA, Seckel MA. Acute Respiratory Distress Syndrome and Prone Positioning. AACN Adv Crit Care. 2018;29(4):415-425. doi.org/10.4037/aacnacc2018161.
40. Raoof S, Nava S, Carpati C, Hill NS. High-Flow, Noninvasive Ventilation and Awake (Nonintubation) Proning in Patients With Coronavirus Disease 2019 With Respiratory Failure. Chest. 2020;158(5):1992-2002. doi.org/10.1016/j.chest.2020.07.013.
41. Bamford AP, Bentley A, Dean J. ICS Guidance for Prone Positioning of the Conscious COVID Patient 2020. Intensive Care Soc. Published online 2020. https:// emcrit.org/wp-content/uploads/2020/04/2020-04-12-Guidance-for-conscious-proning.pdf.
42. Madathil S. Proning in the Ward-Based Awake Self-Ventilating Patient with COVID-19.; 2020.
43. Africa Centres for Disease Control and Prevention. Guidance for awake prone ventilation in the non-intubated conscious patient.; 2020. https://africacdc.org/download/guidance-for-awake-prone-ventilation-in-the-non-intubated-conscious-patient/
44. Tomalin L, Metcalfe A. NDDH Guidance for Prone Positioning Self – Ventilating Patients ± CPAP For COVID-19 2020.; 2020.
45. Hahnel E, Lichterfeld A, Blume-Peytavi U, Kottner J. The epidemiology of skin conditions in the aged: A systematic review. J Tissue Viability. 2017;26(1):20-28. doi.org/10.1016/j.jtv.2016.04.001.
46. Tabara Y, Tachibana-Iimori R, Yamamoto M, et al. Hypotension associated with prone body position: A possible overlooked postural hypotension. Hypertens Res. 2005;28(9):741-746. doi.org/10.1291/hypres.28.741.
47. Manohar N, Ramesh VJ, Radhakrishnan M, Chakraborti D. Haemodynamic changes during prone positioning in anaesthetised chronic cervical myelopathy patients. Indian J Anaesth. 2019;63(3):212. doi.org/10.4103/ija.IJA_810_18.
48. Malik GR, Wolfe AR, Soriano R, et al. Injury-prone: peripheral nerve injuries associated with prone positioning for COVID-19-related acute respiratory distress syndrome. Br J Anaesth. 2020;125(6):e478-e480. doi.org/10.1016/j.bja.2020.08.045.
49. Sánchez-Soblechero A, García CA, Sáez Ansotegui A, et al. Upper trunk brachial plexopathy as a consequence of prone positioning due to SARS-CoV-2 acute respiratory distress syndrome. Muscle and Nerve. 2020;62(5):E76-E78. doi.org/10.1002/ mus.27055.
50. Moore Z, Patton D, Avsar P, et al. Prevention of pressure ulcers among individuals cared for in the prone position: lessons for the COVID-19 emergency. J Wound Care. 2020;29(6):312-320. doi.org/10.12968/jowc.2020.29.6.312.
51. Ibarra G, Rivera A, Fernandez-Ibarburu B, Lorca-García C, Garcia-Ruano A. Prone position pressure sores in the COVID-19 pandemic: The Madrid experience. J Plast Reconstr Aesthetic Surg. 2021. doi.org/10.1016/j.bjps.2020.12.057.
52. Douma MJ, MacKenzie E, Loch T, et al. Prone cardiopulmonary resuscitation: A scoping and expanded grey literature review for the COVID-19 pandemic. Resuscitation. 2020;155(July 2020):103-111. doi.org/10.1016/j.resuscitation.2020.07.010.