Submitted:
22 October 2026
Posted:
23 October 2026
You are already at the latest version
Abstract
Objectives: To determine the prevalence of possible sarcopenia and examine its associations with malnutrition and other related factors among community-dwelling older adults attending public primary care clinics in Malaysia. Methods: A cross-sectional study was conducted among participants aged ≥60 years from January to November 2024 in 11 public primary care clinics in Malaysia. Possible sarcopenia was determined based on handgrip strength according to the Asian Working Group for Sarcopenia 2019. Firth’s penalized likelihood logistic regression was performed to determine the associated factors. Results: A total of 551 participants were included in the analysis, of whom 51.4% had possible sarcopenia. Malay ethnicity (adjusted odds ratio [aOR] 2.17, 95% confidence interval [CI] 1.09–4.34), Indian ethnicity (aOR 2.77, 95% CI 1.05–4.34), frailty (aOR 2.31, 95% CI 1.45–3.66), being at risk of malnutrition (aOR 1.68, 95% CI 1.07–2.64), and malnutrition (aOR 3.40, 95% CI 1.08–10.70) were independently associated with higher odds of possible sarcopenia. Conversely, having previously worked in a professional occupation (aOR 0.55, 95% CI 0.32–0.96) was independently associated with lower odds of possible sarcopenia. Conclusion: Possible sarcopenia was highly prevalent among community-dwelling older adults attending public primary care clinics in Malaysia. Key independent correlates of possible sarcopenia were non-professional occupational status, Malay and Indian ethnicities, malnutrition risk or overt malnutrition and frailty. Routine primary care screening using handgrip dynamometry and frailty assessments is recommended to enable early detection and targeted interventions, focusing on nutritional support, adequate hydration, and frailty management.
Keywords:
possible sarcopenia
; older adults
; community dwelling
; risk assessment
; malnutrition
; primary care
1. Introduction
Sarcopenia is globally recognized not merely as a physiological byproduct of aging, but as a distinct skeletal-muscular disease (ICD-10-CM code M62.84), a conceptual shift formalized by the European Working Group on Sarcopenia in Older People 2 (EWGSOP2) consensus. Globally, sarcopenia affects an estimated 10% to 27% of community-dwelling older adults, presenting a major public health challenge linked to falls, frailty, loss of independence, and increased all-cause mortality. Marking a major paradigm shift in global consensus, the EWGSOP2 and the Global Leadership Initiative in Sarcopenia (GLIS) redefined sarcopenia by elevating low muscle strength, rather than reduced muscle mass alone as its primary operational entry point. Within Asian populations, rapid demographic aging has accelerated this challenge, with regional prevalence rates ranging from 6.8% to 25.7% depending on the diagnostic criteria and settings evaluated. In alignment with this global shift toward early identification, the Asian Working Group for Sarcopenia (AWGS 2019) consensus operationalized this strength-focused model for Asian populations through the construct of possible sarcopenia, defined by low handgrip strength (<28.0 kg for men, <18.0 kg for women aged ≥ 65) or poor physical performance (≥ 12 seconds on the 5-times chair stand test) [1]. This construct serves as the gold-standard, front-line screening metric, empowering primary care clinicians to detect early neuromuscular impairment and initiate targeted lifestyle and nutritional strategies without immediate reliance on advanced muscle mass quantification via Dual-energy X-ray Absorptiometry (DXA) or Bioelectrical Impedance Analysis (BIA). While the recent AWGS 2025 Consensus Update emphasized on life-course muscle health framework, by leveraging muscle health into the global implementation of WHO Integrated Care For Older People (ICOPE) framework for healthy aging, it explicitly retained the construct of possible sarcopenia for primary care and community settings.
In Malaysia, research on muscle health among older adults has expanded in recent years, yet existing epidemiological evidence remains constrained by specific methodological limitations and selection biases. Recent local studies have predominantly evaluated sarcopenia within restricted cohorts, including healthy Chinese adults in a single geographic region [2], specialized hospital outpatient clinics managing chronic respiratory illnesses [3], or high-risk individuals selected specifically for elevated body mass index and fall histories [4]. Similarly, while Lipoeto et al. (2025) evaluated cognitive risks among sarcopenic individuals, their design focused on secondary neurocognitive outcomes rather than front-line screening [5]. Within primary care, landmark work by Sazlina et al. (2020) established sarcopenia profiles among older attendees, but their sample was restricted strictly to patients with Type 2 Diabetes Mellitus [6].
Malaysia, like many other middle-income countries, is experiencing a rapid demographic transition, characterized by a rising proportion of older adults. By 2040, it is projected that approximately 14.5% of the Malaysian population will be aged 65 years and above, with the urban areas having the highest percentage of elderly residents [7]. This demographic shift underscores the urgent need for strategies to address aging-related conditions such as sarcopenia, particularly among community-dwelling older adults. Consequently, robust baseline data reflecting the prevalence of possible sarcopenia among the broader, heterogeneous population of older adults presenting for routine outpatient care in Malaysian public primary care clinics (Health Clinics) is crucial. Malaysia National Health and Morbidity Survey 2025, as household-based surveys suffer from healthy-volunteer bias, whereas hospital-based studies overrepresent severe disease states. Public primary care clinics occupy a vital middle ground, serving a high volume of community-dwelling older adults who harbor significant multi-morbidity but remain functionally independent enough to seek routine primary care. Establishing setting-specific baseline evidence in this population is imperative to justify integrating rapid functional screening into routine national health assessments for older adults. The existing national older adults’ health screening in these clinics does not include standardized screening for possible sarcopenia and nutritional status.
Emerging evidence highlights the coexistence of sarcopenia and malnutrition, commonly referred to as malnutrition-sarcopenia syndrome [8]. Adequate nutrition is a fundamental determinant of skeletal muscle architecture and function in aging. According to the AWGS 2019 consensus, muscle strength is the primary operational marker for possible sarcopenia, which relies on neuromuscular function, myofibrillar protein synthesis (MPS), and cellular homeostasis, all of which are shaped by diet and biochemical markers. Fruit and vegetable consumption plays a critical protective role against sarcopenia through two key mechanisms: dietary antioxidants (polyphenols, carotenoids, vitamin C and E) scavenge reactive oxygen species (ROS) to prevent age-related mitochondrial dysfunction and muscle fiber atrophy, while organic potassium and magnesium salts buffer low-grade metabolic acidosis, preventing acid-induced protein degradation via the ubiquitin-proteasome pathway. Fluid intake regulates cellular hydration; skeletal muscle is 75% water, and hypo-hydration triggers stress-activated protein kinases (p38 MAPK) that accelerate proteolysis while disrupting sarcolemmal electrolyte balances (Na+, K+, Ca2+) necessary for electromechanical coupling during muscle contraction. At the biochemical level, serum total protein and albumin reflect circulating amino acid availability required to stimulate the mechanistic target of rapamycin complex 1 (mTORC1) pathway for MPS, while low levels signal systemic low-grade inflammation (elevated IL-6, TNF-alpha) that accelerates muscle degradation. Ultimately, these biological inputs are governed by individual nutrition knowledge, which acts as the behavioral driver dictating overall diet quality.
Therefore, this study aimed to determine the prevalence of possible sarcopenia, using AWGS consensus criteria, and identify its associated risk factors, focusing on nutritional parameters, among community-dwelling older adults attending public primary care clinics in Malaysia. Understanding these relationships is crucial for promoting healthy aging and improving nutrition at the community level. The findings will help guide the development of comprehensive prevention and intervention strategies, particularly for individuals at risk of or already experiencing sarcopenia, to mitigate its progression and associated health complications.
2. Materials and Methods
2.1. Study Design and Setting
A cross-sectional survey to determine possible sarcopenia and its association with malnutrition, clinical and demographic data, functional capacity, and nutritional knowledge among community-dwelling older adults aged ≥ 60 years who attended public primary care clinics in Malaysia. This survey employed a two-stage stratified cluster sampling design to recruit older adults from multiple public primary care clinics across Malaysia. Geographically, the 14 states of Malaysia were categorized into five main regions: Northern (Perlis, Kedah, Perak, Penang), Eastern (Terengganu, Kelantan, Pahang), Central (Selangor, Kuala Lumpur and Negeri Sembilan) and Southern regions of Peninsular Malaysia (Johor and Melaka), and East Malaysia (Sabah and Sarawak). In the first stage of sampling, states were randomly selected from each region, resulting in the inclusion of two states from each Northern (Perak and Kedah) and Central regions of Peninsular Malaysia (Negeri Sembilan and Kuala Lumpur), one state from each Southern (Johor) and Eastern regions of Peninsular Malaysia (Terengganu) and one state from the East Malaysia region (Sarawak). The second stage of sampling involved selecting public primary care clinics within the chosen states, where the clinics were considered clusters. A total of 11 clinics were then randomly selected (Supplement Figure A1) and eligible older adults attending these clinics during the study period were recruited consecutively until the required sample size was achieved. The site investigators from each study site were briefed on how to administer the questionnaire to ensure uniformity in data collection.
The study was approved by the Medical Research Ethics Committee of the Ministry of Health Malaysia (NMRR: 21-01937-QUG, 11 January 2023). Written permission was obtained from the district health department for each study site. All participants provided written informed consent, and confidentiality and anonymity of the data were strictly maintained.
2.2. Participants and Sample
We recruited participants aged ≥ 60 years who attended public primary care clinics in Malaysia from January to November 2024. The sample size for this study was determined using the single proportion formula: n = [DEFF*Np(1-p)]/ [(d2/Z21-α/2*(N-1)+p*(1-p)], where DEFF is the design effect (1.5), p is the estimated prevalence of possible sarcopenia (0.25) [9], d is the margin of error or absolute precision (0.05), and Z is the Z-score for a 95% confidence level (1.96). The calculation was performed using OpenEpi Version 2.3.1 [10]. Additionally, a 10% non-response rate was considered to ensure an adequate sample size. Based on these parameters, the final calculated sample size for the study was 520 participants.
Participants from long-term care facilities, cognitively impaired (Abbreviated Mental Test-4 score of 3 or less), illiterate, deaf, or visually impaired without the presence of a caregiver or family member were excluded. However, when those had caregivers or family members present, they would act as surrogate respondents or translators, ensuring the accuracy and comprehensiveness of the data collection. Participants with pre-existing physical limitations that might affect clinical findings, such as unilateral hemiparesis with muscle wasting, were excluded from the study. Excluding this group ensures that the data collected is of high quality and accurately reflects the participants’ health status.
2.3. Instruments
The data collection instrument contains nine parts. Parts (A) to (H) were completed by the site investigators via face-to-face interview to obtain the required data, while Part (I) was self-administered by participants. The contents are elaborated as follows:
- Part (A) & (B) - Sociodemographic and clinical characteristics.
Sociodemographic data, including age, gender, marital status, education level, monthly household income, and living arrangement, were recorded. Clinical characteristic data include self-reported current medical comorbidity, ongoing medication(s) use, history of recent hospitalization, smoking, and dental status.
- Part (C) - Clinical Frailty Scale (CFS)
CFS is adopted to determine the frailty status [11]. Frailty status was determined by the site investigators at each study site based on participants’ functional capability explored during the clinic visit. Given the wide heterogeneity among frail participants, we classified them into two degrees of frailty: “non-frail” (CFS 1-3) versus “frail” (CFS 4-9) for analysis. CFS has a sensitivity of 56% and a specificity of 98.4% in detecting frailty among community-dwelling older adults [12].
- Part (D) - Activities of Daily Living (ADL) status
A validated Katz ADL questionnaire was used to evaluate the participant’s ADL [13]. However, no formal reliability and validity reports are available in the literature.
- Part (E) - Anthropometric assessment
The participants’ height and weight were measured using a portable medical stadiometer (SECA 213) and a calibrated digital weighing scale (SECA 874), respectively. The readings were taken to the nearest 0.1cm for height and 0.1 kg for weight. Body mass index (BMI) is calculated by dividing weight in kilograms by the square of height in meters (kg/m2).
- Part (F) – Possible sarcopenia screening and assessment
The participant’s handgrip strength of the dominant hand was measured using a validated manual hand dynamometer (i.e., Jamar Hydraulic Dynamometer) to assess for possible sarcopenia [1]. The training comprised theoretical modules on the AWGS 2019 protocol and hands-on practical sessions evaluating measurement techniques. Physical measurements were strictly standardized across all primary care study sites according to AWGS 2019 recommendations:
- Handgrip Strength Assessment: Handgrip strength (kg) was measured using a calibrated digital hand dynamometer (i.e., Jamar Hydraulic Dynamometer). All site investigators completed a mandatory 1-hour standardized virtual training session led by a geriatrician prior to study commencement. Standardized AWGS-aligned handgrip strength protocol required the dynamometer to be zero-calibrated before each testing session. Participants were seated comfortably in a standard armless chair with their feet flat on the floor, shoulders adducted and neutrally rotated, elbow flexed at 90°, forearm in a neutral position, and wrist between 0° and 30° of extension, without resting their arm on a support surface. Handle size was adjusted to fit the participant’s hand size (the middle phalanges of index to little fingers wrap around the movable handlebar, while the thumb wraps around the opposite fixed handle). Participants were given a brief practice trial at submaximal effort to ensure understanding. During testing, investigators provided standardized verbal encouragement (“Squeeze as hard as you can!”) maintained for 3 seconds. Three maximum-effort trials were conducted for the dominant hand. The handgrip strength measurement protocol recommended by AWGS 2019 is to take the maximum reading of the 3 readings. Based on the AWGS 2019, a low handgrip strength (less than 28 kg for males, or less than 18 kg for females) was considered as possible sarcopenia [1].
- Part (G) - Biochemical indicators within the last 12 months
The associations between low total protein, serum albumin and haemoglobin with sarcopenia have been studied [14,15]. Biochemical parameters (total protein, albumin, and hemoglobin) were extracted from clinic medical records where available within a 12-month window prior to enrollment, reflecting routine annual chronic disease monitoring in public primary care clinics. All participating clinics utilized accredited Ministry of Health Malaysia laboratories operating under standardized quality assurance protocols and automated assay methods. Because exact blood collection dates were not recorded, laboratory data were analyzed as categorical/continuous baseline indicators of general nutritional and metabolic status.
- Part (H) - Nutritional status- screening and assessment
Nutritional screening was performed using the validated six-item Mini Nutritional Assessment-Short Form (MNA-SF) [16] to categorize overall nutritional risk. In addition, specific dietary behaviors including daily meal frequency, intake of protein sources (dairy, legumes/eggs, meat/fish/poultry), fruit and vegetable consumption, and daily fluid intake, were collected as supplementary dietary variables adapted from the full MNA tool to examine specific nutritional correlates of muscle strength.
- Part (I) - Questionnaire on Nutrition Knowledge
The questionnaire on knowledge related to malnutrition was self-constructed based on a study by Agbozo et al. and was available in Malay, English, and Mandarin languages [17]. The questionnaire consists of 13 questions. Some of the questions were negatively worded to avoid response bias among participants. Responses to the questions in this part have options of ‘Correct’, ‘Wrong’, and ‘Do not know’, which were scored 3, 2, and 1, respectively, for positive items, and the scores are reversed accordingly for negative statements. The scores were calculated in percentages and interpreted as good knowledge if scores ≥70%, average knowledge if scores 51% to 69% and poor knowledge if scores ≤50% [18]. The questionnaires in three languages were evaluated by the experts before performing a reliability test on each. The reliability analysis for each language showed Cronbach’s alpha values of 0.686 for Malay, 0.644 for English, and 0.638 for Mandarin.
2.4. Statistical Analysis
Statistical analysis was conducted to determine associations between possible sarcopenia status (normal versus possible sarcopenia) and various sociodemographic, clinical, and nutritional variables. Descriptive statistics summarized categorical variables as frequencies and percentages, while continuous variables were presented as medians with interquartile ranges (IQR) due to non-normal distributions. Pearson’s Chi-square or Fisher’s exact tests were used for categorical data, and the Mann-Whitney U test was applied for continuous variables. Simple and multiple logistic regression analyses were performed to identify factors associated with possible sarcopenia among the elderly respondents. In the simple logistic regression, each independent variable was individually tested for its association with possible sarcopenia, yielding crude odds ratios (OR) and 95% confidence intervals (CI). Variables with a p-value <0.25 in the simple logistic regression were included in the multiple logistic regression analysis. A liberal cut-off threshold of p <0.25 was applied to ensure clinically or practically relevant variables were not missed, recognizing that variables with weak individual associations can contribute significantly to the outcome when evaluated in combination [19]. To determine the most significant independent factors while controlling for confounding factors, a backward (stepwise) logistic regression method was applied, where variables were systematically removed from the model if they did not have significant association with possible sarcopenia (p-value > 0.05). After identifying the significant independent factors of the standard logistic regression, Firth’s penalized likelihood logistic regression was performed because several independent factors categories contained very few participants without possible sarcopenia, including underweight participants in the BMI category (non-possible sarcopenia, n = 7) and participants with malnutrition in the nutritional status category (non-possible sarcopenia, n = 4). Firth’s penalized likelihood logistic regression is applied when the binary outcome contained few events (or few non-events), when one outcome category had very few observations, or when complete separation was present. Standard logistic regression relies on maximum likelihood estimation, which may produce biased parameter estimates in small samples or with sparse outcome data and can yield infinite coefficient estimates and inflated standard errors in the presence of complete or quasi-complete separation. Firth’s penalized likelihood approach reduces small-sample bias and overcomes separation, providing stable, finite parameter estimates and more reliable confidence intervals. Following the application of logistic regression, the goodness-of-fit (GOF) of the model was assessed using the Hosmer-Lemeshow test, and model classification was evaluated through classification accuracy. The model’s performance was measured by using sensitivity, specificity, positive predictive value and negative predictive value. Additionally, the model’s discriminatory power was assessed using the area under the receiver operating characteristic curve (AUC-ROC).
The report of this cross-sectional study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [20].
3. Results
A total of 658 community-dwelling older adults were screened during the study period. Before the univariable analysis, 107 participants were excluded because of missing data, including 51 with missing data on possible sarcopenia only, 29 with missing data on both possible sarcopenia and total monthly household income, and 27 with missing data on both total monthly household income and BMI classification. Consequently, 551 participants were included in the univariable analyses (response rate of 83.7%). As total monthly household income was missing for 30 participants in this analytical sample, analyses involving this variable were based on 521 participants. For the multivariable analysis, a complete-case (listwise deletion) approach was applied. Accordingly, the final analytical sample comprised 521 participants. The participant selection process is illustrated in Supplementary Figure 2.
The mean age was 71.5 years (SD=6.0), with most aged 65-74 years (65.9%). Females (55.0%) outnumbered males (45.0%). The majority were Malay ethnicity (54.6%). Most were married (71.9%). Education levels were predominantly primary (43.2%) and secondary (33.4%). A large portion were unemployed/ retirees/ homemakers (89.5%), and 83.3% had non-professional past professions. Most participants lived with family/ children/ relatives (65.7%), while 25.6% lived with a spouse and 8.7% lived alone. Household income was largely below MYR 4850 (USD 1193) (95.0%), with few earning above MYR 4850 (5.0%) (Table 1).
Table 2 shows the clinical characteristics of the participants. Among the participants (n=551), 13.1% had abnormal cognition, and comorbid illnesses were prevalent, including hypertension (93.1%), diabetes mellitus (63.0%), and heart diseases (16.3%). Polypharmacy (≥5 medications) was observed in 64.6% of participants, while 8.5% had a recent hospital admission. Dental problems were present in 39.2%, and 50.6% wore dentures. Most participants were non-smokers (74.8%). About 51.4% of the participants were classified as having possible sarcopenia based on their hand-grip strength. Nutritional status assessments revealed that 29.8% were at risk of malnutrition, and 4.7% were malnourished. The majority were non-frail (66.8%) and independent in ADLs (96.4%). The study showed that 40.7% consumed daily dairy, while 81.5% had daily meat, fish, or poultry as a protein source. Fruit/vegetable intake was insufficient in 21.2% of participants, and only 55.2% consumed more than 5 cups of fluid daily. Nearly all self-feed without issues (99.8%), yet 9.4% viewed themselves as malnourished. Nutritional knowledge was good in 66.8% of participants. The median of total protein, albumin, and haemoglobin levels were 75.0 g/L, 42.0 g/L, and 12.8 g/dL, respectively.
The prevalence of possible sarcopenia was 51.4% (283/551). The Pearson’s chi-square and Fisher’s exact test analyses showed that the prevalence of possible sarcopenia was significantly associated with various sociodemographic and clinical characteristics (Table 3). Older age groups (75–84 years: 62.5%; ≥85 years: 65.0%) had a higher prevalence compared to younger groups (p=0.013). Lower education levels, such as no formal education (58.3%) and primary education (56.7%), were associated with higher rates of possible sarcopenia (p=0.025). Non-professionals (53.4%) and individuals with monthly incomes below MYR 4850 (USD 1193) (52.1%) were also more affected (p<0.05). Clinically, possible sarcopenia was significantly linked to stroke (75.0%, p=0.006), frailty (67.8%, p<0.001), and malnutrition (84.6%, p<0.001). Serum albumin and haemoglobin levels were significantly lower among respondents with possible sarcopenia (p<0.001), whereas serum protein levels did not differ significantly between the possible sarcopenia and normal groups. Possible sarcopenia was more prevalent among those consuming fewer than two servings of fruits/vegetables daily (62.4%, p<0.001), those with a fluid intake of less than three cups per day (70.0%, p=0.001), and individuals who were uncertain about their nutritional status or viewed themselves as malnourished (60.0% and 65.4%, respectively, p=0.004). Poor nutritional knowledge (75.8%, p=0.011) and lower BMI (p=0.007) were also significantly associated with possible sarcopenia.
Multivariable logistic regression analyses identified several factors independently associated with possible sarcopenia (Table 4). Overall, the findings were largely consistent across the three modelling approaches (standard logistic regression Model 1, standard logistic regression Model 2 adjusted for age, sex and BMI, and Firth’s penalized logistic regression), demonstrating the robustness of the observed associations.
Compared with participants from other ethnic groups, Malays (Model 1: aOR=2.25, 95% CI: 1.11–4.54; Firth’s: aOR=2.17, 95% CI: 1.09–4.34) and Indians (Model 1: aOR=3.19, 95% CI: 1.20–8.48; Firth’s: aOR=2.77, 95% CI: 1.05–4.34) had significantly higher odds of possible sarcopenia. Being previously employed in a professional occupation was associated with lower odds of possible sarcopenia in Model 2 (aOR=0.54, 95% CI: 0.31–0.95) and the Firth’s model (aOR=0.55, 95% CI: 0.32–0.96), although the association did not reach statistical significance in Model 1. Higher household income (≥MYR 4,850/month) was associated with lower odds of possible sarcopenia and was significant only in Model 1 (aOR=0.34, 95% CI: 0.12–0.92).
Nutritional status showed a strong dose-response relationship with possible sarcopenia. Compared with participants with normal nutritional status, those at risk of malnutrition had approximately 1.7- to 2.0-fold higher odds for possible sarcopenia (Model 1: aOR= 1.99, 95% CI: 1.29-3.06; Firth’s: aOR=1.68, 95% CI:1.07-2.64), while participants with malnutrition had 3.4- to 5.0-fold higher odds across all models (Model 1: aOR= 5.02, 95% CI:1.56-16.22; Firth’s: aOR= 3.40, 95% CI: 1.08-10.70). Frailty was also consistently associated with approximately twofold increased odds of possible sarcopenia (Model 1: aOR=2.36, 95% CI: 1.53-3.64; Firth’s: aOR=2.31, 95% CI: 1.45-3.66).
Regarding lifestyle factors, consuming more than five cups of fluids daily was associated with lower odds of possible sarcopenia in Model 1 (aOR=0.33, 95% CI: 0.11-0.97) and Model 2 (aOR: 0.32, 95% CI: 0.11-0.99) but not significant in the Firth model (aOR=0.36, 95% CI: 0.13-1.03). Daily consumption of 3–5 cups of fluids and daily fruit or vegetable intake of fewer than two servings was not significantly associated with possible sarcopenia in all logistic regression model.
After adjustment for age, sex and BMI in Model 2, none of these covariates were independently associated with possible sarcopenia, and the inclusion of these variables had minimal impact on the estimates of the primary risk factors.
The three multivariable models demonstrated comparable overall performance (Table 4). Model 2 showed marginally better discrimination than Model 1 (AUC: 0.737 [95% CI: 0.695–0.779] vs. 0.729 [95% CI: 0.686–0.772]), while the Firth logistic regression achieved an identical AUC (0.737 [95% CI: 0.695–0.779]). The standard logistic regression models showed acceptable calibration based on the Hosmer–Lemeshow goodness-of-fit test (Model 1: p=0.261; Model 2: p=0.320). Classification accuracy was similar across the three models (67.9%, 67.0%, and 65.6% for Models 1, 2, and the Firth model, respectively). At the optimal probability cut-off of 0.50, the Firth model demonstrated substantially higher specificity (87.1%) and positive predictive value (78.3%) than the standard logistic regression models, albeit with lower sensitivity (44.9%). In contrast, Models 1 and 2 provided a more balanced trade-off between sensitivity (64.9% and 64.5%) and specificity (71.1% and 69.5%), respectively. Despite the slightly lower classification accuracy, the Firth logistic regression was retained as the final model because it reduced potential estimation bias associated with sparse data in some nutritional status and BMI categories while maintaining comparable discriminative ability.
4. Discussion
This nationwide, multi-center study highlights a high prevalence of possible sarcopenia (51.4%) among community-dwelling older adults aged ≥ 60 years attending public primary care clinics in Malaysia. This observed burden is comparable to high prevalence rates reported among community-dwelling older populations in regional Asian studies, such as in China [21]. However, direct comparisons across populations must account for differences in recruitment settings, as primary care clinic attendees represent an opportunistic cohort with a higher burden of chronic multi-morbidity compared to general household-sampled populations.
Socioeconomic status and occupational background emerged as key independent correlates of muscle health. In Model 1, higher household income (≥MYR4850) was significantly associated with lower odds of possible sarcopenia, consistent with epidemiological studies in Tehran and China [22,23]. However, this association lost statistical significance after adjusting for age, sex, and BMI in Model 2 and Firth’s model. Having a professional background remained robustly associated with reduced odds for possible sarcopenia across the fully adjusted model. Professional occupations are generally related to higher educational background and better financial income. Although our study did not directly measure food expenditure, food affordability, or dietary supplement usage, established literature demonstrates that higher household income strongly predicts greater diet quality and nutrient density [24]. Higher financial capacity facilitates regular access to protein-rich foods (e.g., fresh lean meats, fish, and dairy) and nutrient-dense products, which carry higher costs per calorie than energy-dense, refined carbohydrate staples [24,25]. Adequate intake of high-quality protein and micronutrients is necessary to stimulate myofibrillar protein synthesis and attenuate muscle breakdown [26]. While these dietary and economic pathways offer a plausible mechanism, they represent hypothesized drivers supported by external literature rather than directly measured variables in our dataset. Overall, screening for possible sarcopenia in Malaysian public primary care clinics must account for these socioeconomic disparities, advocating for targeted nutritional support among lower-income older adults who face financial barriers to purchasing high-quality foods.
Our findings demonstrate that older adults who were malnourished or at risk of malnutrition had significantly higher odds of possible sarcopenia. The odds of possible sarcopenia increased progressively from those at risk of malnutrition to those with overt malnutrition, reinforcing the established link between poor nutritional status and muscle decline in aging [27,28]. Although the prevalence of nutritional vulnerability in our primary care sample was higher than estimates reported in Singapore [29], these differences should be viewed through the lens of sampling and setting heterogeneity. Other international study, such as research in Brazil using the SARC-F questionnaire with calf circumference measurement [30], have similarly observed strong associations between malnutrition or at nutritional risk and sarcopenia risk. The strength of association varies across studies depending on the screening instruments and diagnostic criteria applied.
Ethnicity was significantly associated with possible sarcopenia in multivariable models. Compared to the “Others” reference group, Malay participants and Indian participants demonstrated higher odds of possible sarcopenia across all regression models, whereas Chinese ethnicity did not reach statistical significance. These observed disparities reflect documented variations in body composition phenotypes, muscle mass indices, and physical performance thresholds across diverse Asian ethnic groups [31]. Lifelong variations in dietary composition, physical activity habits, and metabolic traits among Malaysia’s multi-ethnic population may further account for these differences, warranting ethnic-specific consideration in muscle health screening.
To date, based on a systematic review and meta-analysis, cases of sarcopenia in older adults may be related to insufficient protein consumption [32]. However, our study did not find a significant association between insufficient protein consumption and possible sarcopenia. This could be due to limitations in our data related to protein intake, whereby we did not objectively measure daily protein intake and the sources of protein, either animal or plant-based. In contrast, the Newcastle 85+ Study indicated that dietary patterns high in red meat with protein intake >1 g/kg/day were associated with an increased risk of sarcopenia [33]. Furthermore, physical activity levels or structured exercise prescriptions which are the critical co-factors required alongside protein intake to retain lean body mass and function (1.0 to 1.2 g protein/kg body weight/day) [34], were not measured in our dataset.
Our multivariable analysis revealed that daily fruit and vegetable consumption below two servings per day was not an independent correlate of possible sarcopenia, contrasting with previous observational literature that reported significant protective effects of higher plant-based food intake on muscle strength and physical performance [35,36]. Several methodological factors may explain this lack of independent statistical association. First, our study evaluated binary daily serving frequency (< 2 vs. ≥ 2 servings per day) via a self-reported screening item rather than a comprehensive, validated Food Frequency Questionnaire (FFQ) or 24-hour dietary recall. Consequently, key quantitative metrics such as total daily micronutrient density, specific antioxidant capacities (e.g., vitamins C and E, carotenoids, and polyphenols), and dietary fiber content were not captured. Biologically, these specific bioactive compounds mitigate skeletal muscle catabolism by suppressing systemic oxidative stress and neutralizing metabolic acid loads [37,38]; however, crude self-reported serving thresholds may lack the sensitivity needed to detect these physiological pathways. Secondly, because fruit and vegetable intake often correlate strongly with overall diet quality, higher income, and healthier lifestyle behaviors, the crude trend observed in univariate analyses may have been attenuated after adjusting for major socioeconomic and clinical confounders such as household income, occupational background, nutritional risk, and frailty status. Future prospective studies utilizing detailed nutritional profiling and circulating biomarker assays (e.g., serum carotenoids or inflammation markers) are required to clarify whether specific components of fruit and vegetable intake directly preserve muscle mass and function in older adults.
Biologically, skeletal muscle consists of nearly 75% water [39], and age-related factors including blunted thirst mechanisms, impaired renal function, or diuretic use often predispose older adults to fluid deficits [39]. Such deficits could theoretically compromise muscle mechanical and metabolic performance [39]. Even though previous population-based studies showed an association between sarcopenia and inadequate water intake [40], our findings showed no significant association between fluid intake and possible sarcopenia. This could be due to limitations in our study whereby we only use self-reported fluid intake frequency, but we did not objectively quantify total water balance, clinical hydration parameters, or renal function. Therefore, this association should be interpreted with caution. Nevertheless, whether fluid consumption acts as a protective factor against muscle decline requires confirmation through prospective studies utilizing objective hydration biomarkers and detailed clinical profiles.
Frail older adults, as evaluated by the Clinical Frailty Scale (CFS), exhibited more than twofold higher odds of possible sarcopenia across all models, consistent with existing Asian literature [41,42,43,44]. Critically, frailty, ADL limitations, and reduced muscle strength represent overlapping domains of functional decline rather than distinct, isolated risk factors. The less pronounced association for frailty in our study compared to prior literature likely stems from tool selection: whereas previous studies used physical phenotype measures that directly include grip strength [45], we employed the Clinical Frailty Scale (CFS), which relies on clinician judgment of ADL performance and cognition [11,12]. Although the CFS may be less sensitive for isolated physical sarcopenia, it provides a safe, rapid screening instrument suitable for busy primary care settings [46]. Given that ADL limitations directly reduce physical activity levels and exacerbate muscle wasting [47], frailty and ADL dependence should be viewed as interconnected markers of a shared spectrum of physical frailty and sarcopenia.
Our findings demonstrate a significant burden of possible sarcopenia among older adults attending public primary care clinics in Malaysia, underscoring the value of front-line functional screening. Implementing routine assessment in primary care could enable early identification of muscle strength loss and allow for timely lifestyle interventions to preserve physical function. Major strengths of this study include its large sample of primary care clinic attendees from multiple regions in Malaysia and the use of standardized handgrip dynamometry as a simple, objective screening instrument. Importantly, handgrip dynamometry evaluates muscle strength to identify possible sarcopenia and is complementary to, rather than a substitute for, muscle mass quantification tools such as DXA. Future research should evaluate the implementation and predictive validity of primary care screening protocols for muscle health in Malaysia. However, there are limitations in our study. Firstly, this was a cross-sectional study that may limit the ability to define the causal relationship between the risk factors and possible sarcopenia. Secondly, although this study included various covariates for possible sarcopenia, many of which were not included in our study, such as physical activity [47], polypharmacy with potentially inappropriate medications [48], or uncontrolled comorbidities, such as diabetes with complications [49], or chronic kidney disease (CKD) stage [50], that influence muscle metabolism. Further research is needed to study these factors concerning possible sarcopenia in the elderly. Thirdly, biochemical markers (total protein, albumin, and hemoglobin) were obtained from routine clinic records within a 12-month window, and the exact interval between blood sampling and functional assessment was not recorded. While acute illness was excluded at enrollment and serum proteins remain reasonably stable in community-dwelling older adults under routine chronic disease care, unmeasured temporal fluctuations or acute intercurrent events occurring between blood draws and physical assessments cannot be entirely ruled out. Future prospective studies should incorporate simultaneous baseline blood draws alongside physical functional testing. Additionally, daily water, protein, and fibre intake data used in our study were based on a self-reported questionnaire, which could lead to bias in absolute dietary intake amounts.
5. Conclusions
In conclusion, this nationwide, multi-center study demonstrates a high prevalence of possible sarcopenia among community-dwelling older adults attending public primary care clinics across Malaysia. Through multivariable regression, lower occupational status, Malay and Indian ethnicities, nutritional risk or overt malnutrition and frailty were identified as key independent correlates of muscle strength impairment. Conversely, specific protein intake quantities, fluid intake and fruit and vegetable consumption did not retain independent statistical significance after controlling for major socioeconomic and clinical confounders.
These findings highlight the critical importance of implementing routine, opportunistic functional screening such as handgrip dynamometry paired with CFS assessments in primary care settings to identify older adults at risk of muscle decline. Furthermore, interventions targeting muscle health in primary care must adopt a multi-faceted approach, incorporating targeted nutritional support for lower-income populations and early management of physical frailty. Future prospective studies utilizing longitudinal designs, objective physical activity tracking, precise dietary intake, fluid consumption and biological biomarker profiling are required to establish causal pathways and evaluate the predictive validity of primary care screening protocols for muscle health in Malaysia.
Supplementary Materials
The following supporting information can be downloaded at the website of this paper posted on Preprints.org.
Author Contributions
Conceptualization, N.C.C.; Funding acquisition, N.C.C.; Methodology, N.C.C. K.C.C., N.F.G, T.C.L; Project administration, N.C.C. Resources, N.C.C. K.C.C., supervision, N.C.C., K.C.C., N.F.G., T.C.l.; Investigation, K.S.Z., D.A.L, K.L.B., L.C.K.T., T.P., S.R., K.B.M.S, Z.J.; Writing—original draft preparation, N.C.C., K.C.C., N.F.G., T.C.L., K.S.Z., D.A.L, K.L.B., L.C.K.T., T.P., S.R., K.B.M.S, Z.J; Writing—review and editing, N.C.C., K.C.C., N.F.G., T.C.L. All authors have read and agreed to the published version of the manuscript.
Funding
This research was funded by the National Institutes of Health, Ministry of Health Malaysia, grant number P42005001171008, and the APC was funded by the National Institutes of Health, Ministry of Health Malaysia.
Institutional Review Board Statement
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Medical Research and Ethics Committee, Ministry of Health Malaysia (NMRR: 21-01937-QUG, 11 Jan 2023).
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
The data for this study are not publicly available due to government policies on patient data sharing. However, they may be obtained from the corresponding author upon reasonable request and with permission from the Director General of Health, Malaysia.
Acknowledgments
We would like to thank the Director-General of Health Malaysia for granting permission to publish this paper. We would also like to acknowledge the contribution by the National Institute of Health (NIH) for funding the study’s instrument. During the preparation of this work, the authors used Google Gemini to refine the language, enhance text readability, and improve narrative clarity; following these adjustments, the authors have reviewed the content and assume full responsibility for the final publication.
Conflicts of Interest
The authors declare no conflicts of interest.
Abbreviations
The following abbreviations are used in this manuscript:
| ADL | Activities of Daily Living |
| aOR | Adjusted Odds Ratio |
| AWGS 2019 | Asian Working Group for Sarcopenia 2019 |
| BMI | Body Mass Index |
| CFS | Clinical Frailty Scale |
| CI | Confidence Interval |
| DXA | Dual-energy X-ray Absorptiometry |
| EWGSOP2 | European Working Group on Sarcopenia in Older People 2 |
| GLIS | Global Leadership Initiative in Sarcopenia |
| ICOPE | Integrated Care For Older People |
| IQR | Interquartile Range |
| MNA-SF | Mini Nutritional Assessment-Short Form |
| NPV | Negative Predictive Value |
| PPV | Positive Predictive Value |
| WHO | World Health Organization |
Appendix A
Figure A1. Distribution of the study samples.
Figure A2. Process of data cleaning.
References
- Chen LK, Woo J, Assantachai P, Auyeung TW, Chou MY, Iijima K, et al. Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. J Am Med Dir Assoc. 2020 Mar;21(3):300-307.e2. [CrossRef]
- Foo LH, Wen YS, Kadir AA. Assessments of sarcopenia and its associated factors in community-dwelling middle-aged and older Chinese adults in Kelantan, Malaysia. Sci Rep. 2023 May 9;13(1):7498. [CrossRef]
- Mohamad Zani RA, Ahmad Yusof H, Azizan N ‘Ain, Hyder Ali IA, Ismail S, Mohd Shariff N. Sarcopenia and it’s influencing factors among adults with asthma, chronic obstructive pulmonary disease, and tuberculosis in Penang, Malaysia. BMC Public Health. 2025 Apr 28;25(1):1572. [CrossRef]
- Hashim NNA, Mat S, Myint PK, Delibegovic M, Kioh SH, Kamaruzzaman SB, et al. Sarcopenia and physical performance among older adults with increased body mass index and falls. Sci Rep. 2026 Apr 21;16(1):18420. [CrossRef]
- Lipoeto NI, Vanoh D, Desmawati D, Ishak WRW, Mohamed R. Determinants of dementia risk among older adults with probable sarcopenia and sarcopenia. BMC Public Health. 2025 Mar 25;25(1):1139. [CrossRef]
- Sazlina SG, Lee PY, Chan YM, A. Hamid MS, Tan NC. The prevalence and factors associated with sarcopenia among community living elderly with type 2 diabetes mellitus in primary care clinics in Malaysia. PLoS One. 2020 May 20;15(5):e0233299. [CrossRef]
- Department of Statistics Malaysia (DOSM). Population Projection (Revised), Malaysia, 2010-2040 [Internet]. 2016 [cited 2025 Oct 3]. Available from: https://www.dosm.gov.my/portal-main/release-content/population-projection-revised-malaysia-2010-2040.
- Vandewoude MFJ, Alish CJ, Sauer AC, Hegazi RA. Malnutrition-Sarcopenia Syndrome: Is This the Future of Nutrition Screening and Assessment for Older Adults? J Aging Res. 2012;2012:1–8. [CrossRef]
- Iskandar I, Joanny A, Azizan A, Justine M. The Prevalence of Sarcopenia and Its Impact on Quality of Life in Elderly Residing in the Community. Malaysian Journal of Medicine and Health Sciences. 2021.
- Dean AG, Sullivan K, Soe M. ScienceOpen. 2013. OpenEpi: Open source epidemiologic statistics for public health, version 2.3.1 .
- Rockwood K, Theou O. Using the Clinical Frailty Scale in Allocating Scarce Health Care Resources. Canadian Geriatrics Journal. 2020 Aug 24;23(3):254–9. [CrossRef]
- Sukkriang N, Punsawad C. Comparison of geriatric assessment tools for frailty among community elderly. Heliyon. 2020 Sep;6(9):e04797. [CrossRef]
- Yu R, Wong M, Chang B, Lai X, Lum CM, Auyeung TW, et al. Trends in activities of daily living disability in a large sample of community-dwelling Chinese older adults in Hong Kong: an age-period-cohort analysis. BMJ Open. 2016 Dec 15;6(12):e013259. [CrossRef]
- Yin M, Zhang H, Liu Q, Ding F, Deng Y, Hou L, et al. Diagnostic Performance of Clinical Laboratory Indicators With Sarcopenia: Results From the West China Health and Aging Trend Study. Front Endocrinol (Lausanne). 2021 Dec 10;12. [CrossRef]
- Yoshimura Y, Wakabayashi H, Nagano F, Bise T, Shimazu S, Shiraishi A. Low Hemoglobin Levels are Associated with Sarcopenia, Dysphagia, and Adverse Rehabilitation Outcomes After Stroke. Journal of Stroke and Cerebrovascular Diseases. 2020 Dec;29(12):105405. [CrossRef]
- Phillips BNutrDiet MB, Foley AL, Barnard MEd R, Isenring EA, Miller MD. Nutritional screening in community-dwelling older adults: a systematic literature review. Asia Pac J Clin Nutr. 2010.
- Agbozo F, Amardi-Mfoafo J, Dwase H, Ellahi B. Nutrition knowledge, dietary patterns and anthropometric indices of older persons in four peri-urban communities in Ga West municipality, Ghana. Afr Health Sci. 2018 Aug 15;18(3):743. [CrossRef]
- Ifediora CO, Azuike EC. Tackling breast cancer in developing countries: insights from the knowledge, attitudes and practices on breast cancer and its prevention among Nigerian teenagers in secondary schools. J Prev Med Hyg. 2018 Dec 1;59(4):E282–300. PubMed PMID: 30656231. [CrossRef]
- Chowdhury MZI, Turin TC. Variable selection strategies and its importance in clinical prediction modelling. Fam Med Community Health. 2020 Feb 16;8(1):e000262. [CrossRef]
- von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. The Lancet. 2007 Oct;370(9596):1453–7. [CrossRef]
- Yao J, Wang Y, Yang L, Ren M, Li L, Wang H. Prevalence of possible sarcopenia in community-dwelling older Chinese adults: a cross-sectional study. BMJ Open. 2022 Dec 14;12(12):e067425. [CrossRef]
- Dorosty A, Arero G, Chamar M, Tavakoli S. Prevalence of sarcopenia and its association with socioeconomic status among the elderly in Tehran. Ethiop J Health Sci. 2016 Jul 15;26(4):389. [CrossRef]
- Wan H, Hu YH, Li WP, Wang Q, Su H, Chenshu JY, et al. Quality of life, household income, and dietary habits are associated with the risk of sarcopenia among the Chinese elderly. Aging Clin Exp Res. 2024 Feb 9;36(1):29. [CrossRef]
- Darmon N, Drewnowski A. Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutr Rev. 2015 Oct;73(10):643–60. [CrossRef]
- Kwon DH, Park HA, Cho YG, Kim KW, Kim NH. Different Associations of Socioeconomic Status on Protein Intake in the Korean Elderly Population: A Cross-Sectional Analysis of the Korea National Health and Nutrition Examination Survey. Nutrients. 2019 Dec 19;12(1):10. [CrossRef]
- Paddon-Jones D, Rasmussen BB. Dietary protein recommendations and the prevention of sarcopenia. Curr Opin Clin Nutr Metab Care. 2009 Jan;12(1):86–90. [CrossRef]
- Tan VMH, Pang BWJ, Lau LK, Jabbar KA, Seah WT, Chen KK, et al. Malnutrition and Sarcopenia in Community-Dwelling Adults in Singapore: Yishun Health Study. J Nutr Health Aging. 2021 Mar;25(3):374–81. [CrossRef]
- Gao Q, Hu K, Yan C, Zhao B, Mei F, Chen F, et al. Associated Factors of Sarcopenia in Community-Dwelling Older Adults: A Systematic Review and Meta-Analysis. Nutrients. 2021 Nov 27;13(12):4291. [CrossRef]
- Wei K, Nyunt MSZ, Gao Q, Wee SL, Ng TP. Frailty and Malnutrition: Related and Distinct Syndrome Prevalence and Association among Community-Dwelling Older Adults: Singapore Longitudinal Ageing Studies. J Am Med Dir Assoc. 2017 Dec;18(12):1019–28. [CrossRef]
- Sato PHR, Ferreira AA, Rosado EL. The prevalence and risk factors for sarcopenia in older adults and long-living older adults. Arch Gerontol Geriatr. 2020 Jul;89:104089. [CrossRef]
- Woo J, Arai H, Ng TP, Sayer AA, Wong M, Syddall H, et al. Ethnic and geographic variations in muscle mass, muscle strength and physical performance measures. Eur Geriatr Med. 2014 Jun;5(3):155–64. [CrossRef]
- Coelho-Junior H, Calvani R, Azzolino D, Picca A, Tosato M, Landi F, et al. Protein Intake and Sarcopenia in Older Adults: A Systematic Review and Meta-Analysis. Int J Environ Res Public Health. 2022 Jul 18;19(14):8718. [CrossRef]
- Granic A, Mendonça N, Sayer AA, Hill TR, Davies K, Siervo M, et al. Effects of dietary patterns and low protein intake on sarcopenia risk in the very old: The Newcastle 85+ study. Clinical Nutrition. 2020 Jan;39(1):166–73. [CrossRef]
- Bauer J, Biolo G, Cederholm T, Cesari M, Cruz-Jentoft AJ, Morley JE, et al. Evidence-Based Recommendations for Optimal Dietary Protein Intake in Older People: A Position Paper From the PROT-AGE Study Group. J Am Med Dir Assoc. 2013 Aug;14(8):542–59. [CrossRef]
- Montiel-Rojas D, Nilsson A, Santoro A, Franceschi C, Bazzocchi A, Battista G, et al. Dietary Fibre May Mitigate Sarcopenia Risk: Findings from the NU-AGE Cohort of Older European Adults. Nutrients. 2020 Apr 13;12(4):1075. [CrossRef]
- Hong SH, Bae YJ. Association of Dietary Vegetable and Fruit Consumption with Sarcopenia: A Systematic Review and Meta-Analysis. Nutrients. 2024 May 30;16(11):1707. [CrossRef]
- Kao CC, Yang ZY, Chen WL. The association between dietary fiber intake and sarcopenia. J Funct Foods. 2023 Mar;102:105437. [CrossRef]
- Mueller NT, Zhang M, Juraschek SP, Miller ER, Appel LJ. Effects of high-fiber diets enriched with carbohydrate, protein, or unsaturated fat on circulating short chain fatty acids: results from the OmniHeart randomized trial. Am J Clin Nutr. 2020 Mar;111(3):545–54. [CrossRef]
- Lorenzo I, Serra-Prat M, Yébenes JC. The Role of Water Homeostasis in Muscle Function and Frailty: A Review. Nutrients. 2019 Aug 9;11(8):1857. [CrossRef]
- Yoo JI, Choi H, Song SY, Park KS, Lee DH, Ha YC. Relationship between water intake and skeletal muscle mass in elderly Koreans: A nationwide population-based study. Nutrition. 2018 Sep;53:38–42. [CrossRef]
- Nishiguchi S, Yamada M, Fukutani N, Adachi D, Tashiro Y, Hotta T, et al. Differential Association of Frailty With Cognitive Decline and Sarcopenia in Community-Dwelling Older Adults. J Am Med Dir Assoc. 2015 Feb;16(2):120–4. [CrossRef]
- Therakomen V, Petchlorlian A, Lakananurak N. Prevalence and risk factors of primary sarcopenia in community-dwelling outpatient elderly: a cross-sectional study. Sci Rep. 2020 Nov 11;10(1):19551. [CrossRef]
- Pham LAT, Nguyen BT, Huynh DT, Nguyen BMLT, Tran PAN, Van Vo T, et al. Community-based prevalence and associated factors of sarcopenia in the Vietnamese elderly. Sci Rep. 2024 Jan 2;14(1):17. [CrossRef]
- Ukegbu U, Maselko J, Malhotra R, Perera B, Østbye T. Correlates of HandGrip Strength and Activities of Daily Living in Elderly Sri Lankans. J Am Geriatr Soc. 2014 Sep 20;62(9):1800–1. [CrossRef]
- 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 Mar 1;56(3):M146–57. [CrossRef]
- Oviedo-Briones M, Laso ÁR, Carnicero JA, Cesari M, Grodzicki T, Gryglewska B, et al. A Comparison of Frailty Assessment Instruments in Different Clinical and Social Care Settings: The Frailtools Project. J Am Med Dir Assoc. 2021 Mar;22(3):607.e7-607.e12. [CrossRef]
- Sánchez-Sánchez JL, He L, Morales JS, de Souto Barreto P, Jiménez-Pavón D, Carbonell-Baeza A, et al. Association of physical behaviours with sarcopenia in older adults: a systematic review and meta-analysis of observational studies. Lancet Healthy Longev. 2024 Feb;5(2):e108–19. [CrossRef]
- Tanaka T, Akishita M, Kojima T, Son BK, Iijima K. Polypharmacy with potentially inappropriate medications as a risk factor of new onset sarcopenia among community-dwelling Japanese older adults: a 9-year Kashiwa cohort study. BMC Geriatr. 2023 Jun 26;23(1):390. [CrossRef]
- Veronese N, Pizzol D, Demurtas J, Soysal P, Smith L, Sieber C, et al. Association between sarcopenia and diabetes: a systematic review and meta-analysis of observational studies. Eur Geriatr Med. 2019 Oct 5;10(5):685–96. [CrossRef]
- Chatrenet A, de Müllenheim PY, Torreggiani M, Hernández JN, Arronte RU, Espinoza AG, et al. Quality matters: chronic kidney disease progression is associated with reduced muscle strength independently of changes in skeletal muscle mass: an observational study. Clin Kidney J. 2025 Mar 4;18(3). [CrossRef]
Table 1.
Sociodemographic characteristics of the participants (n=551).
| Sociodemographic variables | Frequency (n) | Percentage (%) | |
|---|---|---|---|
| Age, mean (SD) | 71.5 (6.0) | ||
| Age groups (years) | 60-64 | 40 | 7.3 |
| 65-74 | 363 | 65.9 | |
| 75-84 | 128 | 23.2 | |
| 85 and above | 20 | 3.6 | |
| Sex | Female | 303 | 55.0 |
| Male | 248 | 45.0 | |
| Ethnicity | Malay | 301 | 54.6 |
| Chinese | 153 | 27.8 | |
| Indian | 47 | 8.5 | |
| Others | 50 | 9.1 | |
| Marital status | Never married | 12 | 2.2 |
| Married | 396 | 71.9 | |
| Divorced/ widowed | 143 | 26.0 | |
| Highest Education level | No formal education | 72 | 13.1 |
| Primary | 238 | 43.2 | |
| Secondary | 184 | 33.4 | |
| Tertiary | 57 | 10.3 | |
| Working status | Unemployed/ retiree/ homemaker | 493 | 89.5 |
| Employed | 58 | 10.5 | |
| Past profession | Non-professional | 459 | 83.3 |
| Professional | 92 | 16.7 | |
| Living status | Living alone | 48 | 8.7 |
| Living with spouse | 141 | 25.6 | |
| Living with family/ children/ relatives | 362 | 65.7 | |
|
Household monthly income (n=521) |
< MYR 4850 (USD 1193) | 495 | 95.0 |
| ≥MYR 4850 (USD 1193) | 26 | 5.0 | |
Table 2.
Clinical characteristics and nutritional status of the participants (n=551).
| Frequency | Percentage (%) | ||
|---|---|---|---|
| Cognitive Status | Abnormal Cognition | 72 | 13.1 |
| Normal Cognition | 479 | 86.9 | |
| Comorbid illness | Diabetes mellitus | 347 | 63.0 |
| Hypertension | 513 | 93.1 | |
| Heart Diseases (Ischaemic heart disease/Heart failure) | 90 | 16.3 | |
| Chronic kidney disease | 91 | 16.5 | |
| Chronic lung disease | 42 | 7.6 | |
| Stroke / Transient Ischemic attack | 32 | 5.8 | |
| Osteoporosis | 15 | 2.7 | |
| Cancer | 10 | 1.8 | |
| Dementia | 4 | 0.7 | |
| Other chronic diseases | 26 | 4.7 | |
| Number of prescribed medications | < 5 | 195 | 35.4 |
| ≥ 5 | 356 | 64.6 | |
| Recent admission in past 6 months | No | 504 | 91.5 |
| Yes | 47 | 8.5 | |
| Dental problem | No | 335 | 60.8 |
| Yes | 216 | 39.2 | |
| Denture | No | 272 | 49.4 |
| Yes | 279 | 50.6 | |
| Smoking status | Non-smoker | 412 | 74.8 |
| Current smoker | 39 | 7.1 | |
| Former smoker | 100 | 18.1 | |
| Possible Sarcopenia | No | 268 | 48.6 |
| Yes | 283 | 51.4 | |
| Nutritional status | Normal | 361 | 65.5 |
| At Risk of Malnutrition | 164 | 29.8 | |
| Malnutrition | 26 | 4.7 | |
| Clinical Frailty Scale (CFS) | Non-frail | 368 | 66.8 |
| Mild to moderate frail | 178 | 32.3 | |
| Severely frail | 5 | 0.9 | |
| ADL status | Dependent | 10 | 1.8 |
| Semi-dependent | 10 | 1.8 | |
| Independent | 531 | 96.4 | |
| BMI status | Underweight | 25 | 4.5 |
| Normal | 225 | 40.8 | |
| Overweight | 179 | 32.5 | |
| Obese | 122 | 22.1 | |
| Biomedical indicators, median (IQR) | Total Protein (g/L) | 75.0 (6.8) | |
| Albumin (g/L) | 42.0 (4.5) | ||
| Haemoglobin (g/dL) | 12.8 (2.3) | ||
| No. of full meals daily | 1 | 72 | 13.1 |
| 2 | 246 | 44.6 | |
| 3 | 233 | 42.3 | |
| Protein intake | |||
| ≥ 1 serving dairy product daily | Yes | 224 | 40.7 |
| No | 327 | 59.3 | |
| ≥ 2 legumes or eggs per week | Yes | 448 | 81.3 |
| No | 103 | 18.7 | |
| Meat, fish or poultry daily | Yes | 449 | 81.5 |
| No | 102 | 18.5 | |
| Fruit/vegetable intake daily | < 2 servings | 117 | 21.2 |
| ≥2 servings | 434 | 78.8 | |
| Fluid intake daily | < 3 cups | 50 | 9.1 |
| 3-5 cups | 197 | 35.8 | |
| >5 cups | 304 | 55.2 | |
| Mode of feeding | |||
| Unable to eat without assistance | 1 | 0.2 | |
| Self-fed without any problems | 550 | 99.8 | |
| Self-view of nutritional status | |||
| Views self as being malnourished | 52 | 9.4 | |
| Uncertain of nutritional state | 120 | 21.8 | |
| Views self as having no nutritional problem | 379 | 68.8 | |
| Nutritional Knowledge score | |||
| Poor | 33 | 6.0 | |
| Average | 150 | 27.2 | |
| Good | 368 | 66.8 | |
Table 3.
Prevalence of possible sarcopenia by sociodemographic and clinical characteristics (n=551).
Table 3.
Prevalence of possible sarcopenia by sociodemographic and clinical characteristics (n=551).
| Sociodemographic variables | Normal n (%) |
Possible sarcopenia n (%) |
p value | ||
|---|---|---|---|---|---|
| Age groups (years) | 60-64 | 20 (50.0) | 20 (50.0) | 0.013 | |
| 65-74 | 193 (53.2) | 170 (46.8) | |||
| 75-84 | 48 (37.5) | 80 (62.5) | |||
| 85 and above | 7 (35.0) | 13 (65.0) | |||
| Sex | Female | 146 (48.2) | 157 (51.8) | 0.814 | |
| Male | 122 (49.2) | 126 (50.8) | |||
| Ethnicity | Malay | 139 (46.2) | 162 (53.8) | 0.145 | |
| Chinese | 80 (52.3) | 73 (47.7) | |||
| Indian | 19 (40.4) | 28 (59.8) | |||
| Others | 30 (60.0) | 20 (40.0) | |||
| Marital status | Never married | 5 (41.7) | 7 (58.3) | 0.470 | |
| Married | 199 (50.3) | 197 (49.7) | |||
| Divorced/ widowed | 64 (44.8) | 85 (55.2) | |||
| Highest Education level | No formal education | 30 (41.7) | 42 (58.3) | 0.025 | |
| Primary | 103 (43.3) | 135 (56.7) | |||
| Secondary | 103 (56.0) | 81 (44.0) | |||
| Tertiary | 32 (56.1) | 25 (43.9) | |||
| Working status | Unemployed/retiree/homemaker | 227 (48.1) | 256 (51.9) | 0.438 | |
| Employed | 31 (53.4) | 27 (46.6) | |||
|
Past profession |
Non-professional | 214 (46.6) | 245 (53.4) | 0.034 | |
| Professional | 54 (58.7) | 38 (41.3) | |||
|
Living status |
Living alone | 24 (50.0) | 24 (50.0) | 0.107 | |
| Living with spouse | 79 (56.0) | 62 (44.0) | |||
| Living with family/ children/relatives | 165 (45.6) | 197 (54.4) | |||
| Household monthly income (n=521) | < MYR 4850 (USD 1193) | 237 (47.9) | 258 (52.1) | 0.012 | |
| ≥MYR 4850 (USD 1193) | 19 (73.1) | 7 (26.9) | |||
| Clinical Characteristics | |||||
| Cognitive Status | Abnormal Cognition | 28 (38.9) | 44 (61.1) | 0.076 | |
| Normal Cognition | 240 (50.1) | 239 (49.9) | |||
| Comorbid illness | Diabetes mellitus | 164 (47.3) | 183 (52.7) | 0.3999 | |
| Hypertension | 244 (47.6) | 269 (52.4) | 0.063 | ||
| Heart Diseases | 49 (54.4) | 41 (45.6) | 0.228 | ||
| Chronic kidney disease | 43 (47.3) | 48 (52.7) | 0.772 | ||
| Chronic lung disease | 24 (57.1) | 18 (42.9) | 0.251 | ||
| Stroke / Transient Ischemic attack | 8 (25.0) | 24 (75.0) | 0.006 | ||
| Osteoporosis | 7 (46.7) | 8 (53.3) | 0.877 | ||
| Cancer | 5 (50.0) | 5 (50.0) | 1.0 | ||
| Dementia | 0 (0) | 4 (100) | 0.124 | ||
| Others chronic diseases | 11 (42.3) | 15 (57.7) | 0.508 | ||
|
Number of prescribed medications |
< 5 | 103 (52.8) | 92 (47.2) | 0.146 | |
| ≥ 5 | 165 (46.3) | 191 (53.7) | |||
| Recent admission in past 6 months | No | 248 (49.2) | 256 (50.8) | 0.383 | |
| Yes | 20 (42.6) | 27 (57.4) | |||
| Dental problem | No | 172 (51.3) | 163 (48.7) | 0.114 | |
| Yes | 96 (44.4) | 120 (55.6) | |||
| Denture | No | 131 (48.2) | 141 (51.8) | 0.825 | |
| Yes | 137 (49.1) | 142 (50.9) | |||
| Smoking status | Non-smoker | 198 (48.1) | 214 (51.9) | 0.231 | |
| Current smoker | 24 (61.5) | 15 (38.5) | |||
| Former smoker | 46 (46.0) | 54 (54.0) | |||
| Nutritional status | Normal | 203 (56.2) | 158 (43.8) | <0.001 | |
| At Risk of Malnutrition | 61 (37.2) | 103 (62.8) | |||
| Malnutrition | 4 (15.4) | 22 (84.6) | |||
| Clinical Frailty Scale (CFS) | Non-frail | 209 (56.8) | 159 (43.2) | <0.001 | |
| Frail | 59 (32.2) | 124(67.8) | |||
| ADL status | Dependent & semi-dependent | 0 (0.0) | 20 (100.0) | <0.001 | |
| Independent | 268 (50.5) | 263 (49.5) | |||
| BMI classifications | Underweight | 7 (28.0) | 18 (72.0) | 0.007 | |
| Normal | 97 (43.1) | 128 (56.9) | |||
| Overweight | 102 (57.0) | 77 (43.0) | |||
| Obese | 62 (50.8) | 60 (49.2) | |||
|
Biomedical indicators, median (IQR) |
Total Protein (g/L) | 75.0(6.0) | 74.8 (7.1) | 0.094 | |
| Albumin (g/L) | 42.5 (4.5) | 41.0 (4.1) | <0.001 | ||
| Haemoglobin (g/dL) -Male -Female |
13.2 (2.2) 14.0 (1.9) 12.4(1.8) |
12.4 (2.2) 13.1 (2.6) 12.1 (1.8) |
<0.001 <0.001 0.010 |
||
| Number of full meals daily | 1 | 33 (45.8) | 39 (54.2) | 0.130 | |
| 2 | 110 (44.7) | 136 (55.3) | |||
| 3 | 125 (53.6) | 108 (46.4) | |||
| Dietary protein intake | |||||
| ≥1 serving dairy product daily |
Yes | 113 (50.4) | 111 (49.6) | 0.482 | |
| No | 155 (47.4) | 172 (52.6) | |||
| ≥2 legumes or eggs per week | Yes | 226 (50.4) | 222 (49.6) | 0.077 | |
| No | 42 (40.8) | 61 (59.2) | |||
| Meat, fish or poultry daily | Yes | 227 (50.6) | 222 (49.4) | 0.059 | |
| No | 41 (40.2) | 61 (59.8) | |||
|
Fruits/vegetable intake daily |
< 2 servings | 44 (37.6) | 73 (62.4) | <0.001 | |
| ≥2 servings | 224 (51.6) | 210 (48.4) | |||
| Fluid intake daily | < 3 cups | 15 (30.0) | 35 (70.0) | 0.001 | |
| 3-5 cups | 85 (43.1) | 112 (56.9) | |||
| >5 cups | 168 (55.3) | 136 (44.7) | |||
| Mode of feeding | |||||
| Unable to eat without assistance | 1 (100.0) | 0 (0) | 0.486 | ||
| Self-fed without any problems | 267 (48.5) | 283 (51.4) | |||
| Self-view of nutritional status | |||||
| Views self as being malnourished | 18 (34.6) | 34 (65.4) | 0.004 | ||
| Uncertain of nutritional state | 48 (40.0) | 72 (60.0) | |||
| Views self as no nutritional problem | 202 (53.3) | 177 (46.7) | |||
| Nutritional Knowledge score | |||||
| Poor | 8 (24.2) | 25 (75.8) | 0.011 | ||
| Average | 71 (47.3) | 79 (52.7) | |||
| Good | 189 (51.4) | 179 (48.6) | |||
Continuous variables were presented as mean (standard deviation) or median (interquartile range), while categorical variables were presented as frequency and percentage. The Mann-Whitney test was used for continuous data, and Pearson’s Chi-square test or Fisher’s exact test was applied for categorical data, as appropriate.
Table 4.
Associated factors of possible sarcopenia among the participants (n=521).
| Standard Logistic Regressiona (Model 1) |
Standard Logistic Regressionb (Model 2) |
Firth’s Logistic Regression | |
| Variables | Adjusted odds ratio aOR (95% CI) |
Adjusted odds ratio aOR (95% CI) |
Adjusted odds ratiob aOR (95% CI) |
| Ethnicity | |||
| Malays | 2.25 (1.11, 4.54)* | 2.26 (1.11, 4.60)* | 2.17 (1.09, 4.34)* |
| Chinese | 1.71 (0.81, 3.58) | 1.47 (0.68, 3.15) | 1.44 (0.68, 3.03) |
| Indians | 3.19 (1.20, 8.48)* | 2.85 (1.04, 7.78)* | 2.77 (1.05, 4.34)* |
| Others | 1 | 1 | |
| Past Professional | |||
| Non-professional | 1 | 1 | 1 |
| Professional | 0.63 (0.36, 1.08) | 0.54 (0.31, 0.95)* | 0.55 (0.32, 0.96)* |
| Household monthly income | |||
| < MYR 4850 (USD 1193) | 1 | 1 | 1 |
| ≥MYR 4850 (USD 1193) | 0.34 (0.12, 0.92)* | 0.37 (0.13, 1.0) | 0.39 (0.14, 1.04) |
| Chronic lung disease or asthma | |||
| No | 1 | 1 | 1 |
| Yes | 0.54 (0.26, 1.12) | 0.50 (0.23, 1.07) | 0.52 (0.25, 1.09) |
| Heart Diseases | |||
| No | 1 | 1 | 1 |
| Yes | 0.62 (0.37, 1.04) | 0.63 (0.37, 1.07) | 0.64 (0.38, 1.08) |
| Stroke or Transient Ischemic attack | |||
| No | 1 | 1 | 1 |
| Yes | 2.28 (0.86, 6.07) | 2.60 (0.96, 7.04) | 2.39 (0.92, 6.19) |
| Nutritional status | |||
| Normal | 1 | 1 | 1 |
| At Risk of Malnutrition | 1.99 (1.29, 3.06)** | 1.72 (1.08, 2.74)* | 1.68 (1.07, 2.64)* |
| Malnutrition | 5.02 (1.56, 16.22)** | 3.93 (1.16, 13.29)* | 3.40 (1.08, 10.70)* |
| Clinical Frailty Scale (CFS) | |||
| Non-frail | 1 | 1 | 1 |
| Frail | 2.36 (1.53, 3.64)** | 2.41 (1.50, 3.86)** | 2.31 (1.45, 3.66)** |
| Fruits or vegetable intake daily | |||
| ≥2 servings | 1 | 1 | 1 |
| < 2 servings | 1.50 (0.94, 2.39) | 1.52 (0.94, 2.47) | 1.50 (0.93, 2.40) |
| Fluid intake daily | |||
| < 3 cups | 1 | 1 | 1 |
| 3-5 cups | 0.41 (0.14, 1.24) | 0.39 (0.13, 1.20) | 0.42 (0.14, 1.24) |
| >5 cups | 0.33 (0.11, 0.97)* | 0.32 (0.11, 0.99)* | 0.36 (0.13, 1.03) |
| Biochemical indicator | |||
| Total serum protein level (g/L) | 0.97 (0.95, 1.0) | 0.97 (0.95, 1.01) | 0.98 (0.95, 1.01) |
| Age groups | |||
| 60-64 | 1 | 1 | |
| 65-74 | 0.86 (0.39, 1.93) | 0.87 (0.40, 1.89) | |
| 75-84 | 1.30 (0.53, 3.17) | 1.28 (0.54, 3.05) | |
| 85 and above | 0.92 (0.24, 3.43) | 0.91 (0.25, 3.26) | |
| Sex | |||
| Female | 0.73 (0.49, 1.09) | 1.34 (0.91, 1.99) | |
| Male | 1 | 1 | |
| BMI Status | |||
| Underweight | 1 | 1 | |
| Normal | 0.80 (0.27, 2.37) | 0.83 (0.28, 2.33) | |
| Overweight | 0.52 (0.17, 1.59) | 0.54 (0.18, 1.60) | |
| Obese | 0.58 (0.18, 1.86) | 0.61 (0.20, 1.86) | |
| Model’s Goodness-of-fit | |||
| Hosmer and Lemeshow Test (p value) | 0.261 | 0.320 | - |
| Classification table (%) | 67.9 | 67.0 | 65.6 |
| Area under the ROC curve (95%CI) | 0.729, (0.686–0.772) | 0.737 (0.695, 0.779) | 0.737 (0.695, 0.779) |
| Model performance measures | |||
| Cut-off |Sensitivity | Specificity | PPV | NPV | 0.50 | 64.9% | 71.1% | 69.9% | 66.2% | | 0.50 | 64.5% | 69.5% | 68.7% | 65.4% | | 0.50 | 44.9% | 87.1% | 78.3%| 60.4% | |
*p <0.05, ** p<0.01, aMultiple variables logistic regression with backward selection method (Model 1). bModel 2 adjusted for age, sex, and BMI classifications. Firth’s penalized likelihood logistic regression was performed. There was no multicollinearity (VIF < 5) or interaction effects (p>0.05) among the independent factors in the final model.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.