This cross-sectional study was conducted in several geriatric care centres in different islands of Indonesia, which were Cipto Mangunkusumo National General Hospital in Central Jakarta, Atma Jaya Private Hospital in North Jakarta, Dr Mohammad Hoesin General Hospital in Palembang, Siti Khadijah Islamic Hospital in Palembang, Haji Adam Malik General Hospital in Medan, Dr Wahidin Sudirohusodo General Hospital in Makassar, Dr Soetomo General Hospital in Surabaya, Dr Hasan Sadikin General Hospital in Bandung, Dr Saiful Anwar General Hospital in Malang, Dr Mohammad Djamil General Hospital in Padang, Dr Kariadi General Hospital in Semarang, and Dr Moewardi General Hospital in Surakarta. We obtained the data from multi-ethnic Indonesian adults aged 60 years and older. The interviewers were trained physicians caring for older adults. Data collection process was done from April to October 2022. The subjects provided their written informed consent to participate in this study. We excluded patients with acute illness(es) and/or incomplete data, as we believed that acute illness(es) may affect the examination results related to malnutrition, sarcopenia, and/or frailty. Ethical approval was obtained from the Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia, with registration number KET-369/UN2.F1/ETIK/PPM.00.02/2020.

Frailty state was assessed using the FRAIL scale. One point was given if the older adult reported each diagnostic tool criterion. The criteria consist of fatigue, resistance (defined as the ability to climb one flight of stairs), ambulation (walk a block), number of co-morbid illnesses <5, and weight loss of more than 5% in the previous year. If the total score was 3 or higher, the patient was categorized as frail. If the total score was 0, the patient was categorized as robust, whereas the total score of 1 or 2 indicates pre-frailty.¹²⁾

Possible sarcopenia in this study was based on possible sarcopenia diagnostic criteria from Asian Working Group for Sarcopenia (AWGS) 2019 consensus.¹³⁾ Diagnosis began with case finding with the measurement of calf circumference (<34 cm for male and <33 cm for female) or completing SARC-F questionnaire. The latter consists of five components, namely strength, assistance in walking, rise from a chair, climb stairs, and fall. The SARC-F has been adapted and validated in Indonesian population,¹⁴⁾ and its performance was highly sensitive and specific for detecting sarcopenia based on the criteria from European Working Group for Sarcopenia in Older People (EWGSOP) and from AWGS in 2019.¹⁴⁾

Following case finding, the diagnosis of possible sarcopenia requires the assessment of muscle strength or physical performance. The former was assessed by handgrip strength, whereas the latter was assessed by five times sit to stand (FTSTS) test. Handgrip strength was considered low if the result was <28 kg for male and <18 kg for female. Physical performance was considered abnormal if the result of FTSTS test was 12 seconds or higher.¹³⁾ We obtained 3 readings of handgrip strength for each person using JAMAR hand dynamometer (Sammons Preston Rolyan, Bolingbrook, IL, USA). All measurements are standardised across different clinics.

Malnutrition in this study was defined as the Mini Nutritional Assessment–Short Form (MNA-SF) score <11. The assessment tool was selected for its simplicity and fair sensitivity and specificity to diagnose malnutrition based on the Global Leadership Initiative on Malnutrition (GLIM) criteria. MNA-SF consists of objective features of malnutrition, such as body mass loss, decreased food intake, and the presence of an acute illness or stress during the preceding three months. Additionally, it also assesses the subject’s body mass index, presence of any neuropsychological disorders (depression or dementia), mobility and limitations.¹⁵⁾

We collected details from each individuals as follows. Sex was categorized as male or female. Age was categorized as 60–69 years or 70 years and older. Age cut-off point was based on previous report of community-dwelling older adults in Indonesia¹⁶⁾ and Malaysia.¹⁷⁾ Ethnic groups were categorised into Sundanese, Batak, Betawi, Javanese, Malay, Minang, Chinese and others. In this study Malay ethnic group included people native to the regions on the east coast of Sumatra island, such as Deli Malay and Palembang Malay. Smoking history was documented as never-smoker, former or current smoker. Information regarding past medical history was collected through history taking and from medical records. Diagnoses were made by physicians. We took into account past medical history such as diabetes mellitus, chronic kidney disease, cardiovascular disease, transient ischemic attack (TIA) and cerebrovascular accident, dementia, hypertension, osteoarthritis, dyslipidaemia, falls and recurrent falls. Falls were defined as history of one fall or more during the previous 12 months, whereas recurrent falls were defined as history of two falls or more during the preceding 12-month period.¹⁸⁾ In comprehensive geriatric assessment (CGA), we collected data regarding cognitive function, depression and functional status. We used 10-point abbreviated mental test (AMT or AMT-10) and classified cognitive impairment with cut-off point of 6. AMT was known for its reasonable test performance.¹⁹⁾ Functional status was categorized as independent or dependent, depending on Barthel Index of activity of daily living (ADL) questionnaire. Score of below 20 in the ADL questionnaire indicates dependent functional status.¹⁶⁾ Depression variable was determined by the result of five items of Geriatric Depression Scale (GDS-5) at cut-off point 2.²⁰⁾ Multimorbidity was defined as the co-occurrence of more than one chronic conditions.^21,22) We defined polypharmacy as the administration of five or more medications daily.²³⁾

We analyzed the data with SPSS version 21 (IBM, Armonk, NY, USA). We provided descriptive data of frailty, sarcopenia, malnutrition, and the overlap of the evaluated geriatric syndromes. We used chi-square test to perform the bivariate analysis, followed by multivariable analysis to assess the association between the overlap of syndromes and the covariates. Variables with p-value <0.25 in bivariate analysis were included in multivariable analysis using multiple logistic regression method.²⁴⁾ The alpha level for the study was set at 0.05.

The prevalence of FRAIL scale-based frailty in this multicentre study population was 13.6%. Previous multicentre studies also suggested similar epidemiological data in community-dwelling older adults in the region.^16,25) Frailty is linked to functional dependence, depression, history of fall, polypharmacy, malnutrition,¹⁶⁾ and possible sarcopenia. Even without sarcopenia diagnosis, frailty itself is a strong predictor of mortality,²⁶⁾ and is associated with hospitalization and low quality of life.²⁷⁾ Frailty assessment results in our study showed that 55.0% and 31.4% of older adults were pre-frail and robust, respectively.

The prevalence of possible sarcopenia in this study was 45.5%. The finding was in line with the reported prevalence of sarcopenia in Asia using AWGS definition, which ranged from 2.5 to 45.7%.^28,29) Asian population may have higher prevalence of sarcopenia than non-Asian community. In this study, we decided to use possible sarcopenia criteria due to its excellent diagnostic accuracy. Previous study involving Japanese older adults suggested that for male population the sensitivity and specificity to predict sarcopenia were 89.3% and 99.0%, respectively. In addition, the sensitivity and specificity in female population were 92.1% and 87.0%, respectively.³⁰⁾ The possible sarcopenia criteria included SARC-F questionnaire in the case finding stage, which was a validated tool in Indonesian population, with high sensitivity and specificity for detecting sarcopenia based on the criteria from AWGS and EWGSOP.¹⁴⁾ In order to address the problem of obstacles to disseminating the diagnosis and treatment of sarcopenia based on criteria utilizing dual-energy X-ray absorptiometry (DXA) or bioelectrical impedance analysis (BIA) method, the AWGS 2019 has added possible sarcopenia criteria with simple assessments. This potentially promotes early detection and intervention of sarcopenia, especially in resource-limited settings.

The prevalence of malnutrition in this study was 5.3%. This finding was supported by the previous systematic review of observational studies suggesting that the prevalence of malnutrition on the basis of MNA result in Indonesia ranged from 2.1 to 14.6%.³¹⁾

Therefore, approximately 1 out of 10 had frailty, 1 out of 2 had possible sarcopenia, and 1 out of 20 had malnutrition. Thus, primary care and specialized outpatient clinics should be involved in the detection of these prevalent geriatric syndromes, as the evaluated syndromes may be reversible if addressed properly.³²⁾ If patients were proven not to have these preventable geriatric syndromes, prevention should be included in the interdisciplinary care for the older adult.

The prevalence of overlap of frailty, possible sarcopenia and malnutrition in our outpatient study was 3.3%. It was lower than the result of similar study in nursing home setting (14.5%)⁷⁾ and in medical inpatient setting (approximately 8%).⁶⁾

As the coexistence of three evaluated geriatric syndromes have implications for outcome and proper treatment, appropriate screening may be required in older population. This may suggest the need for comprehensive geriatric assessment (CGA). When applied to community-dwelling older adults, outpatient CGA may delay the progression frailty and contribute to the improvement of frail older adults with multimorbidity.³³⁾ CGA remained the gold standard for caring for frail older adults with frailty, as stated in the best practice guidelines on the management of frailty by several medical societies.

The detection of disease(s) and/or geriatric syndrome(s) through CGA can be followed by appropriate management of the problems. The substantial overlap of three evaluated syndromes may also indicate that older outpatients are likely to benefit from nutritional support with a protein-enriched diet as appropriate.³⁴⁾ A systematic review of observational studies in the region also identified a high prevalence of protein inadequacy among Indonesian community-dwelling older adults.³¹⁾

Interestingly, the three syndromes were found in parallel in 24.3%, 7.2%, and 61.1% older adults with frailty, possible sarcopenia, and malnutrition, respectively. This indicates that a majority of malnourished older adults also had frailty and possible sarcopenia. Recently, sarcopenia was detected to be the mediating factor that could explain the indirect relationship between malnutrition and frailty.³⁵⁾ However, malnutrition may be directly linked to frailty as well. Previous study showed that malnutrition-sarcopenia syndrome was four times more likely to die than other nonsarcopenic adults with normal nutritional status.³⁶⁾ Sarcopenia and malnutrition may be present in parallel via a combination of several factors, including inflammation, decreased body weight, decreased food intake, disease burden, as well as altered endocrine and immune systems. Even without malnutrition, a recent systematic review and meta-analysis suggested that adults with sarcopenia had two times higher risk of mortality than adults without sarcopenia, irrespective of sarcopenia definition and population.³⁷⁾

In this study, older adults with three overlapping syndromes were characterized by a substantial proportion of female sex, older adults with low educational attainment, diabetes mellitus, hypertension, cognitive impairment, multimorbidity, and dependent functional status. However, not all of the covariates were associated with the overlap of frailty, possible sarcopenia and malnutrition.

Our data suggested that overlap of frailty, possible sarcopenia, and malnutrition was associated with TIA and cerebrovascular accident (OR=4.65, 95% CI 1.17–18.47). A recent systematic review and meta-analysis of 18 studies reported the prevalence of frailty in individuals with stroke as 22%.³⁸⁾ Our study, also suggested similar finding, in which 10 out of 41 patients with diagnosis of TIA and cerebrovascular accident were frail (24.4%). In patients with stroke, the prevalence of frailty to be two-fold compared to those without stroke. Further research is required to prove potential bi-directional relationship between stroke and frailty, that might lead to a self-propagating cycle suggested by Evans and colleagues.³⁹⁾ In the cycle, frail older adults may have lower response to treatment, leading to poorer stroke recovery and increased number of stroke events. Following TIA and cerebrovascular accident, individuals with greater frailty may have worse outcomes, including greater length of stay, higher readmission, higher 90-day mortality, and worse health-related quality of life. In addition, frailty is a major factor influencing cessation of therapy in patients with atrial fibrillation who are on anticoagulant regimen.⁴⁰⁾ This may lead to worsening of stroke risk factors and higher severity of stroke.

The statistically significant relationship between TIA and cerebrovascular accident, and the overlap of frailty, possible sarcopenia and malnutrition can also be explained by the potential negative impacts of cerebrovascular accident on nutrient intake. Numerous studies have consistently demonstrated that patients who had stroke could not achieve the recommended nutritional intake.⁴¹⁾ Poor nutritional status is associated with worse stroke outcomes. The combination of poor nutritional status, immobilization and inactivity after cerebrovascular accident could lead to possible sarcopenia and negatively impact the recovery after stroke.⁴²⁾ The presence of possible sarcopenia prior to a stroke may also be more common than suspected and lead to worse functional recovery,⁴²⁾ which may indicate suspected bidirectional relationship between possible sarcopenia and cerebrovascular accident.

Cognitive impairment based on AMT-10 in our study is linked to the overlap of frailty, possible sarcopenia, and malnutrition (OR=3.55, 95% CI 1.16–10.89). A systematic review and meta-analysis of six observational studies concluded that baseline frailty was associated with higher risk of cognitive disorders (including dementia and mild cognitive impairment [MCI]).⁴³⁾

Experts have also suggested the role of nutritional status as the mediating variable in the association between cognitive decline and sarcopenia, although the relationship between the two are bidirectional and complex. Firstly, pro-inflammatory factors such as interleukin-6 (IL-6) is higher in patients with cognitive impairment⁴⁴⁾ and sarcopenia.⁴⁵⁾ Secondly, nitrosative and oxidative stress products accumulate during ageing process and potentially cause cognitive impairment.⁴⁶⁾ Thirdly, older adults with cognitive decline often forget to eat, become less able to prepare and access food, all of which potentially impair food intake per oral and subsequently result in malnutrition. Not only do older adults with cognitive impairment demonstrate lower dietary intake but also reduced physical activity that potentially triggers excessive muscle loss and accelerate sarcopenia. On the other hand, malnutrition may also promote cognitive decline and cause subsequent increase in the risk of sarcopenia and functional dependence.⁴⁷⁾

Dependent functional capacity is also associated with the overlap of three syndromes in our study (OR=12.16, 95% CI 3.50–42.29). The combination of frailty and possible sarcopenia in an older adult may lead to physical function impairment characterized by weak muscle strength, poor balance, and slow walking speed. We hypothesized that the relationship may be bidirectional, as sarcopenia leads to impaired ability to perform ADL, and inactivity may also accelerate sarcopenia.⁴⁸⁾ Taken all in all, we provided the illustration of findings related to the relationship between statistically significant covariates and the overlap of three evaluated syndromes, including the hypothesized bi-directional link (Fig. 2).

Multinational experts previously suggested standardized screening for these conditions at hospital admission.³⁴⁾ In addition, we suggest standardized screening procedure in older population at the community level to detect frailty, possible sarcopenia and malnutrition alone or in combination. This may help physicians detect geriatric syndromes and acts as secondary preventive strategy to stop age-associated disability. Early intervention following the detection of geriatric syndromes may decrease hospitalization and mortality.⁴⁹⁾

One feasible strategy is by using Rapid Geriatric Assessment (RGA) tool established at St. Louis University, which only takes less than 5 minutes to screen the aforementioned geriatric problems. The tool has been used in Singapore with promising benefit, both in the form of administered RGA and self-administered RGA.⁵⁰⁾ Thus, we also suggest validation study for this tool in local older population.

We managed to find out three alarming conditions associated with the occurrence of three geriatric giants. Practice nurses and general practitioners are all well-placed to screen for frailty, sarcopenia, and malnutrition.¹¹⁾ We suggest a holistic referral pathway for individuals who might benefit from geriatric medicine referral. We suggest older adults aged 60 years and older with new, recurrent, or worsening cerebrovascular disease to be referred to consultant geriatrician for thorough assessment of patient’s condition. In addition, we also suggest older adults aged 60 years and older with cognitive impairment to be referred to consultant geriatrician for thorough assessment of patient’s condition. Dependent functional capacity should be an alarm, too.

To the best of our knowledge, this multicentre observational study was the first to evaluate the overlap of frailty, possible sarcopenia and malnutrition in Indonesian archipelago. By involving various centres on different islands, our multicentre study may represent the real population better. Multivariable analysis in this multicentre study helped adjust study variables appropriately. However, we acknowledge limitations of the study that need to be considered. First, causal links could not be assessed by this study design. Future longitudinal study in this field may be necessary. Second, we only relied on possible sarcopenia criteria, based on AWGS 2019 consensus on sarcopenia. Possible sarcopenia diagnostic criteria had excellent accuracy in Asian population and could promote early detection and intervention of sarcopenia,³⁰⁾ especially in Indonesia where sophisticated BIA and DXA facilities are not widely available in healthcare centres. Third, the lack of statistical significance in our study could be due to an insufficient sample size, although it was precalculated appropriately.

Among community-dwelling older outpatients, 1 out of 10 had frailty, 1 out of 2 had possible sarcopenia, and 1 out of 20 had malnutrition. Approximately 1 in 30 in the same cohort had overlapping frailty, possible sarcopenia, and malnutrition. The co-occurrence of the aforementioned geriatric problems is characterized by a substantial proportion of female sex, older adults with low educational attainment, diabetes mellitus, hypertension, cognitive impairment, multimorbidity, and dependent functional status. The condition is associated with TIA and cerebrovascular accident, cognitive impairment, and dependent functional capacity. We suggest standardized screening procedure in older population at the community level to detect frailty, possible sarcopenia and malnutrition alone or in combination.