1 Introduction
Sarcopenia, a progressive skeletal muscle disease, is characterized by the degenerative loss of muscle strength and mass [
1]. It exhibits age-dependent incidence patterns and varies across racial groups, leading to increased risks of falls, functional decline, and mortality [
2–
4]. Sarcopenia significantly reduces quality of life in older adults and increases hospitalization costs by 34%–58%, contributing to annual medical expenses exceeding $18.5 billion in the United States [
5,
6].
In recent years, there has been ongoing debate regarding the relationship between sarcopenia, immune system inflammation, and infections [
7,
8]. Exposure to certain pathogens or a compromised immune system significantly elevates the risk of acute or chronic infections, including both infectious diseases and non-communicable infections, with substantial health consequences. Skeletal muscle plays a vital regulatory role in immune function, acting as a critical integrator linking sarcopenia and immune senescence within the aging biological system. Age-related muscle degeneration impairs the body’s immune response to pathogens [
9]. Consequently, several clinical observational studies have demonstrated that individuals with sarcopenia who undergo colorectal cancer (CRC) surgery have a higher susceptibility to non-surgical-site infections [
10,
11]. Similar findings have been reported in postoperative patients with other cancers, such as adenocarcinoma and penile cancer [
12,
13]. Additionally, the association between sarcopenia and infections has been confirmed in elderly individuals and patients with type 2 diabetes [
7,
14]. However, conflicting results exist regarding the relationship between sarcopenia and postoperative infections or inflammatory markers [
15,
16], indicating that the underlying mechanisms require further investigation.
Several limitations are evident in previous research. First, most studies have focused on specific infections or demographic groups, limiting their generalizability [
12,
13]. Second, prior studies are often constrained by small sample sizes, retrospective designs, and short follow-up periods [
7,
14]. Lastly, few studies have explored the relationship between sarcopenia and infectious diseases comprehensively.
Recently, Yu and colleagues analyzed 243 097 White European participants from the UK Biobank (UKBB) and showed that both probable and confirmed sarcopenia were prospectively associated with a higher long-term risk of infection-related hospitalization across multiple organ-specific, bacterial, and viral infection subtypes, indicating that sarcopenia is an important marker of severe infections [
17]. However, their analysis was restricted to infection-related hospitalizations in White European adults and did not distinguish non-communicable infections from classical infectious diseases or assess the independent contributions of muscle strength and muscle mass.
To address these remaining gaps, we utilized prospective cohort data from the UKBB to investigate the associations between sarcopenia and the long-term risk of overall infections and infectious diseases in the general adult population, and further validated the findings in an independent US cohort.
2 Methods
2.1 Study participants
The design, data access policies, and statistical approaches of the UKBB have been detailed previously [
18]. Briefly, from 2006 to 2010, the UKBB recruited over half a million individuals aged 40–69 years from 22 assessment centers across England, Scotland, and Wales. Approximately 9.2 million potentially eligible individuals aged 40–69 years who lived within 40 km of one of the 22 assessment centers were invited to participate in UKBB, of whom 5.5% attended the baseline assessment [
19]. Comprehensive phenotypic and genotypic data were collected during recruitment, including sociodemographic information, dietary patterns, lifestyle factors, and medical histories through touchscreen surveys. Physical measurements were obtained via direct assessments, while blood, urine, and saliva samples were collected at baseline. Written informed consent was obtained for the collection of questionnaire and biological data. The study was conducted in accordance with ethical approval from the UK North West Multi-Centre Research Ethics Committee (11/NW/0382) under UKBB project number 93 044. The research followed the STROBE guidelines [
20].
Fig. S1 provides an overview of the study design. We excluded participants without sarcopenia data or those lost to follow-up (n = 12 143). We also excluded participants diagnosed with infection or infectious diseases either at baseline or within the first year of follow-up (n = 29 303). In addition, participants with a history of malignancy or conditions suggestive of immunosuppression at baseline (e.g., autoimmune disease, chronic use of systemic corticosteroids, or chronic use of immunosuppressive agents) were excluded (n = 2609). Ultimately, 458 332individuals were included in the primary analysis. In addition, 9163 participants from the NHANES database were included to validate the main results of the study.
2.2 Assessment of overall infections and infectious diseases
Infections (non-contagious) and infectious diseases (contagious) were defined according to diagnosis records in the UKBB, coded using the International Classification of Diseases (ICD)-10 and ICD-9 [
21]. During the follow-up period (median follow-up: 12.5 years), only the first occurrence of each outcome (overall infections and infectious diseases) was included. For a given outcome, person-time was censored at the date of the first recorded infection or infectious disease, and participants did not contribute further observation time after that event. Based on the coding criteria, a total of 92 590 cases (19.9%) were identified, with 47 203 classified as infections and 70 159 as infectious diseases. The relevant ICD-10 and ICD-9 codes and NHANES data on related infectious diseases are provided in Table S1.
2.3 Assessment of sarcopenia
Sarcopenia was assessed according to the 2019 revised consensus of the European Working Group on Sarcopenia in Older People (EWGSOP2), which considers muscle strength, appendicular lean mass (ALM) index, and physical performance [
1]. In the UKBB, muscle strength was measured using a Jamar J00105 hydraulic hand-held dynamometer. ALM was assessed using segmental bioelectrical impedance analysis (BIA) with a Tanita BC-418MA body composition analyzer (Tanita Corp., Tokyo, Japan). This 8-electrode device provides segmental estimates of fat mass and fat-free mass for the right and left arms, right and left legs, and trunk based on impedance measurements and the manufacturer’s prediction equations incorporating height, weight, age, and sex [
22]. All BIA measurements were performed by trained staff following the standardized UKBB body composition protocol, which includes central quality control procedures for Tanita BC-418MA devices. In a subsample of UKBB participants, BIA-derived ALM shows a very high correlation with dual-energy X ray absorptiometry-derived ALM (Pearson’s
r ≈ 0.96), supporting the validity of this measure [
23]. Furthermore, independent validation studies of the Tanita BC-418MA have reported a same-day test–retest coefficient of variation of approximately 1.4% for percentage body fat, with minimal mean bias between measurements, indicating good short-term measurement reliability of this instrument [
24]. Gait speed was assessed via self-reported question: “How would you describe your usual walking speed?” Weak grip strength was defined as < 16 kg for women and < 27 kg for men. Low muscle mass was classified as an appendicular lean mass-to-body mass index (ALM/BMI) ratio of < 0.789 kg/m
2 for men and < 0.512 kg/m
2 for women [
25]. Poor physical performance was defined as walking less than 1.3 m/s.
For the NHANES dataset, body composition was assessed using dual-energy X-ray absorptiometry (DXA), and appendicular skeletal mass (ASM) was calculated as the sum of lean mass of the arms and legs. In line with the Foundation for the National Institutes of Health (FNIH) Sarcopenia Project, we defined low muscle mass as ASM divided by BMI (ASM/BMI) < 0.789 kg/m
2 in men and < 0.512 kg/m
2 in women [
26]. Participants who met these FNIH low-muscle-mass criteria were classified as having sarcopenia in the NHANES validation analyses, whereas those above these thresholds were classified as non-sarcopenic. The EWGSOP2 categories of probable, confirmed, and severe sarcopenia were not applied to the NHANES data; instead, the NHANES sarcopenia definition captures the low muscle mass component of sarcopenia, which is one of the core elements used to confirm sarcopenia under the EWGSOP2 framework [
27,
28].
Participants with low muscle strength but without low muscle mass or poor physical performance were classified as having probable sarcopenia. Those with both low muscle strength and low muscle mass, but not poor physical performance, were classified as having confirmed sarcopenia. Individuals with low muscle strength, low muscle mass, and poor physical performance were classified as having severe sarcopenia. Due to the limited number of severe sarcopenia cases (n = 121), these participants were included in the confirmed sarcopenia group for subsequent analyses.
2.4 Assessment of covariates
This study included various covariates, including demographic characteristics, lifestyle factors, physiologic and biochemical traits, and disease histories, to address potential confounders in the association between sarcopenia and infectious diseases.
Demographic factors included age, sex, and ethnicity. Sociological factors included the Townsend deprivation index and educational attainment (e.g., college degree, A-levels, or equivalent). Lifestyle factors included physical activity measured using the abbreviated International Physical Activity Questionnaire (IPAQ), smoking status (current, former, or never), and alcohol consumption. Medical history included the presence of prevalent hypertension and diabetes, defined as present if diagnosed at or before baseline. Biochemical factors included C-reactive protein (CRP), neutrophil-to-lymphocyte ratio (NLR), and estimated glomerular filtration rate (eGFR) calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula [
29].
2.5 Mediation analysis
We conducted mediation analyses to evaluate whether systemic inflammation partially mediates the association between sarcopenia categories and incident infections. Baseline CRP and NLR were log-transformed to reduce skewness. First, associations of sarcopenia categories with ln(CRP) and ln(NLR) were examined using multivariable linear regression adjusted for the same covariates as in Model 2 (excluding the mediator of interest). Second, Cox proportional hazards models were fitted for sarcopenia and infection risk, with and without additional adjustment for each inflammatory biomarker. The proportion mediated was calculated as the attenuation in the log hazard ratio for each sarcopenia category after inclusion of ln(CRP) or ln(NLR) in the Cox model.
2.6 Statistical analyses
Continuous variables were summarized using the mean (standard deviation), while categorical variables were expressed as counts (percentages). Comparisons of categorical variables were made using the Chi-square test, while continuous variables were analyzed using the t-test, Mann–Whitney U test, or Kruskal–Wallis H test.
Cox proportional hazards regression was used to investigate the risk of overall infections and infectious diseases associated with sarcopenia. Hazard ratios (HRs) with 95% confidence intervals (95% CIs) were reported. The proportional hazards assumption was assessed using the Schoenfeld residuals test. Individuals without sarcopenia served as the reference group. Hospital inpatient data were censored on different dates depending on the region: September 30, 2021, for England, July 31, 2021, for Scotland, and February 28, 2018, for Wales. Follow-up started at the date of the baseline assessment and ended at the earliest of: (1) the first diagnosis of infections or infectious diseases, (2) death, (3) the recorded date of loss to follow-up (including emigration and withdrawal from further record linkage), or (4) the region-specific censoring date for hospital inpatient data, whichever occurred first.
Multivariable model 1 was adjusted for age, sex, and ethnicity. Model 2 additionally adjusted for BMI, Townsend deprivation index, educational attainment, physical inactivity, smoking and drinking status, diabetes, hypertension, eGFR, CRP, and NLR. Because repeated measurements of BMI, inflammatory markers, and lifestyle factors were available only in a subset of participants, we did not fit time-updated Cox models. Instead, we descriptively assessed the stability of selected covariates (BMI, CRP, NLR, and lifestyle factors) across available repeat assessment time points, and the results are shown in Table S2.
Among the 458 332 UKBB participants included in the primary analysis, 122 060 individuals (26.6%) had at least one missing value in baseline covariates. We assumed that baseline covariates were missing at random (MAR), conditional on the observed variables included in the imputation models. To reduce bias and loss of precision, we used multiple imputation by chained equations to impute missing baseline covariates. Imputation models were run separately by sex and included all covariates from the analysis models. We generated 20 imputed datasets and combined regression coefficients and standard errors across datasets using Rubin’s rules. In addition, as a complete-case sensitivity analysis, we repeated all Cox models restricting the sample to participants with no missing values in any covariate (n = 336 272).
Several sensitivity analyses were conducted to validate the robustness of the findings. (1) Subgroup analyses stratified by age (< 65 years or ≥ 65 years), gender, race, Townsend deprivation index (quintiles), BMI (< 30 kg/m
2 or ≥ 30 kg/m
2), educational attainment, smoking and drinking status, IPAQ, diabetes, hypertension, and eGFR (< 60 mL/min/1.73m
2 or ≥ 60 mL/min/1.73m
2). (2) Repeating analyses after excluding participants with missing covariates. (3) Excluding participants who developed infections within the first 2 years of follow-up to minimize reverse causation. (4) Changing time scale to age [
30]. (5) Validation using data from the US NHANES dataset. The primary analysis was repeated using NHANES data, excluding infection-related data due to availability constraints. All statistical analyses were conducted using R (version 4.2.2). A two-tailed
P value < 0.05 was considered statistically significant.
3 Results
3.1 Participant characteristics
This study included 458 332 participants with a mean age of 56.8 years, of whom 54.6% (250 410) were female. Among the cohort, 4.9% (22 419) were identified as having probable sarcopenia, and 0.2% (947) were classified as having confirmed sarcopenia. Baseline characteristics by sarcopenia categories are presented in Table 1. Compared with participants without sarcopenia, individuals with probable or confirmed sarcopenia were older, more likely to be male, less physically active, current smokers, and had higher prevalence rates of hypertension and diabetes.
3.2 Association between sarcopenia and the long-term risk of overall infections and infectious diseases
During a median follow-up of 12.5 years (IQR: 11.5–13.3), 87 130 new cases of overall infections or infectious diseases were documented. These included 80 253 cases (18.5%) in the non-sarcopenia group, 6483 cases (28.9%) in the probable sarcopenia group, and 394 cases (41.6%) in the confirmed sarcopenia group (Fig. 1). The cumulative hazard for overall infections and infectious diseases was significantly higher in the probable sarcopenia group than the non-sarcopenia group, with the highest hazard observed in the confirmed sarcopenia group (Fig. 2; log-rank P < 0.001). Subtype analyses of infections and infectious diseases further supported the primary findings, showing distinct patterns of risk. Similarly, participants with reduced handgrip strength or low muscle mass exhibited elevated cumulative infection hazards compared to controls (Fig. 3; both log-rank P < 0.001).
Cox proportional hazards models, with the non-sarcopenia group as the reference, corroborated these results. In multivariable Model 1, the HRs for overall infections and infectious diseases were 1.43 (95% CI 1.40–1.47) in the probable sarcopenia group and 1.96 (95% CI 1.78–2.17) in the confirmed sarcopenia group (Fig. 4). After additional adjustments in Model 2, these HRs were attenuated to 1.18 (95% CI 1.15–1.21) and 1.34 (95% CI 1.21–1.48), respectively.
For infections, the adjusted HRs (Model 2) were 1.13 (95% CI 1.09–1.17) and 1.26 (95% CI 1.10–1.45) for the probable and confirmed sarcopenia groups, respectively (Fig. 4). For infectious diseases, the adjusted HRs were 1.22 (95% CI 1.18–1.25) and 1.40 (95% CI 1.25–1.56) for the probable and confirmed sarcopenia groups, respectively (Fig. 4).
Participants with reduced handgrip strength or low muscle mass exhibited significant associations with an increased risk of overall outcomes, including infections and infectious diseases, in the fully adjusted Cox models (Model 2) (Fig. 4).
3.3 Mediation by inflammatory biomarkers
In mediation analyses, inclusion of ln(CRP) attenuated the association between sarcopenia and overall infection outcomes, corresponding to a mediation proportion of 7.28% for probable sarcopenia and 5.72% for confirmed sarcopenia (Fig. 5A and 5B). Similar attenuation was observed with ln(NLR) (2.47% and 3.15%, respectively) (Fig. 5C and 5D). These findings suggest that baseline systemic inflammation partially explains the elevated infection risk associated with sarcopenia.
3.4 Sensitivity analyses
Subgroup analyses for overall infections and infectious diseases showed that participants who were younger, had a high BMI, a low Townsend deprivation index, and no history of hypertension exhibited higher risks (P for interaction < 0.05, Fig. 6). Similar results were observed in the subgroup analyses when infections and infectious diseases were considered as separate outcomes (Tables S3 and S4). As a complete-case analysis, we repeated the Cox models after excluding participants with any missing covariates (n = 122 060), and we also repeated the analyses after excluding participants who developed infections within the first 2 years of follow-up (n = 3882). In both sets of sensitivity analyses, the results remained robust and were very similar to those of the primary multiple imputation analyses (Tables S5 and S6).
To address potential residual confounding by age, we conducted sensitivity analyses using chronological age as the underlying time scale in the Cox models (Table S7). For the composite endpoint of infections and infectious diseases, probable sarcopenia was associated with a higher risk (Model 2: HR 1.17, 95% CI 1.14–1.20), and confirmed sarcopenia showed a stronger association (Model 2: HR 1.40, 95% CI 1.27–1.54). Similar patterns were observed for infections and infectious diseases.
3.5 Validation using NHANES data
Validation was performed using data from the US NHANES cohort, comprising 9136 participants with a mean age of 38.9 ± 11.7 years, of whom 48.6% (4455) were female and 9.7% (891) were diagnosed with sarcopenia (Table S8). Sarcopenia participants were more likely to be female, have a lower BMI, higher education levels, and be unmarried, and were less likely to engage in high-intensity physical activity. Among participants without sarcopenia, 1299 cases (15.7%) of infectious diseases were recorded, compared to 187 cases (30.1%) among those with sarcopenia. In the unadjusted model (Model 1), sarcopenia was significantly associated with a higher risk of infectious diseases (HR 1.42, 95% CI 1.18–1.70, P < 0.001). After adjusting for confounders (Model 2), the association remained significant (HR 1.28, 95% CI 1.06–1.54, P = 0.011). These results confirm that sarcopenia is independently associated with an increased risk of infectious diseases (Table 2).
4 Discussion
In this large prospective cohort study of 465 592 adults from the UKBB, supplemented by external validation in NHANES, we found that probable and confirmed sarcopenia were associated with 25% and 44% higher long-term risks, respectively, of overall infections and infectious diseases compared with participants without sarcopenia after extensive multivariable adjustment. Low handgrip strength and low muscle mass were each independently associated with increased risk, and the associations were robust across multiple sensitivity and subgroup analyses. We further showed that these associations could be replicated in an independent US population in which body composition was assessed using DXA, supporting the generalisability of the findings across measurement methods and populations.
While our manuscript was under review, Yu et al. published a related analysis of 243 097 White European participants from the UKBB, showing that probable and confirmed sarcopenia, defined using the EWGSOP2 algorithm, were associated with a higher long-term risk of infection-related hospitalisation across multiple organ-specific, bacterial, and viral infections, supporting sarcopenia as an important risk factor for serious infections. Our study differs in several important ways. We examined a broader spectrum of outcomes, including overall infections and infectious diseases and separately analyzed non-communicable infections and classical infectious diseases, rather than focusing only on hospitalisations. We applied an EWGSOP-based definition incorporating muscle strength, ALM/BMI, and physical performance, and also evaluated handgrip strength and muscle mass as individual risk factors. We analyzed the full UKBB adult cohort with adjustment for ethnicity and externally validated the associations in an independent, younger US NHANES sample with DXA-derived body composition. Although sarcopenia in NHANES was defined using the FNIH ASM/BMI cut-points rather than the full EWGSOP2 algorithm, and the age and lifestyle distributions differ from those in UKBB, this heterogeneity is a strength for validation, as it tests whether the association between low muscle mass and infectious diseases is robust across different populations and operational definitions. The observation of a consistent positive association between sarcopenia and infectious diseases in NHANES (HR 1.28, 95% CI 1.06–1.54) supports the generalisability of our main findings. In addition, extensive subgroup analyses revealed particularly elevated relative risks among younger and obese individuals and those without hypertension. Collectively, these features indicate that our work complements and extends the findings of Yu et al. by describing the impact of sarcopenia on infection risk across a wider range of outcomes, populations, and definitions.
Our findings, together with those of Yu
et al., fit within a broader literature linking sarcopenia to susceptibility to infection. Several meta-analyses have identified sarcopenia as a risk factor for pneumonia and surgical complications in patients with esophageal or CRC [
31,
32]. However, other studies have reported conflicting findings. For instance, a meta-analysis found no significant association between sarcopenia and postoperative infection rates in esophageal cancer surgery [
33]. Most prior studies have focused on specific populations and infections, such as postoperative infections in cancer patients. Zhang
et al. also identified sarcopenia as a risk factor for infections in individuals with type 2 diabetes, with validation performed in a cohort with a median follow-up of 1.84 years [
7]. However, extended studies on sarcopenia and infection risk in broader populations remain scarce. Our results, based on a general adult population and externally validated, therefore add important complementary evidence.
While previous research suggested no direct association between variations in BMI or body fat and infection risk among patients with sarcopenia [
7], individuals with sarcopenic obesity—characterized by coexisting obesity, reduced bone density, and diminished muscle mass—face significantly worse health outcomes. These include increased risks of fractures, functional limitations, reduced insulin sensitivity, weakened immune responses, longer hospitalizations, and shorter life expectancy [
34]. Sarcopenic obesity has been linked to higher infection rates, worsened infection outcomes, and increased infection-related mortality [
35,
36]. Consistent with these findings, our subgroup analyses revealed a marked increase in infection risk among obese patients (BMI > 30 kg/m
2) with sarcopenia.
After adjusting for confounders, our analysis showed that sarcopenic individuals with low physical activity faced significantly higher infection risks. Reduced physical activity is a well-established functional consequence of poor muscle health [
37]. Sarcopenia and frailty are often interconnected, with sarcopenia recognized as a core component of frailty in older populations [
38]. Longitudinal studies have also demonstrated a temporal relationship between sarcopenic obesity and the development of frailty [
39]. Sudden declines in physical capability may signal the onset of subclinical sarcopenia or early-stage infections [
40,
41]. Conversely, higher physical activity levels are associated with reduced hospitalization rates, disease severity, and mortality [
42,
43]. This relationship supports the hypothesis that interventions targeting physical activity and muscle strength might help reduce infection risk in sarcopenic individuals; however, interventional and causal inference studies are needed to test this possibility.
Our subgroup analyses revealed an unexpected trend: younger participants with sarcopenia exhibited a higher incidence of infectious diseases compared to older individuals. While older adults are typically more susceptible to infections due to immunosenescence and the higher prevalence of sarcopenia [
1,
21], this pattern suggests that sarcopenia in younger adults may represent a marker of premature biological aging, more severe underlying pathology, or unrecognized chronic conditions. Younger adults with sarcopenia may also have higher levels of exposure to community and occupational contacts, leading to greater infection pressure, and survival bias in older age groups could attenuate age-related differences. This finding suggests that younger adults with sarcopenia may represent an important group for future research, including studies that clarify causal pathways and evaluate whether targeted interventions such as exercise and muscle-strengthening programmes can reduce infection susceptibility.
The underlying mechanisms remain partially explained but are likely multifactorial. Initially, sarcopenia often coexists with malnutrition and frailty, placing the body in a catabolic state that impairs immune function and reduces the ability to respond to inflammatory triggers, increasing infection vulnerability [
44]. Sarcopenia is associated with increased inflammatory markers such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α), which accelerate muscle deterioration and exacerbate systemic inflammation [
45,
46]. Sarcopenic individuals often have reduced levels of omega-3 fatty acids (ω-3 FAs)—which possess anti-inflammatory properties—and elevated levels of pro-inflammatory omega-6 fatty acids, contributing to heightened inflammation and infection susceptibility [
47]. Recent studies implicate the NLRP3 inflammasome in sarcopenia development, with targeted treatments showing promise in mitigating its effects [
48]. While our findings strongly suggest a link between sarcopenia and infections, causality remains difficult to establish in observational studies. Furthermore, experimental models have demonstrated that inflammation adversely affects muscle function through increased proteolysis, further supporting this bidirectional relationship [
49].
This study has several strengths, including its large sample size, extended follow-up period, and rigorous sensitivity analyses. To our knowledge, this is the first large-scale prospective study to examine sarcopenia in relation to both infections and infectious diseases, treated as a distinct category, in the general adult population and to validate the findings in an independent cohort. Taken together, these features make our work one of the most comprehensive investigations to date on the association between sarcopenia and infection risk in the general population.
However, several limitations should be noted. First, we used BIA instead of the more accurate dual-energy X-ray absorptiometry, which may introduce measurement bias. Second, despite adjusting for multiple covariates, the observational nature of the study leaves room for residual confounding. Third, all covariates (including BMI, CRP, lifestyle factors, and comorbidities) were assessed at baseline, and we did not model them as time-dependent covariates. Therefore, we could not fully account for changes in adiposity, inflammation or lifestyle during follow-up, which may have introduced residual confounding and potentially attenuated some associations. However, in the subset of participants with repeat assessments, key covariates appeared broadly stable across available time points (Table S2). Future studies with more complete repeated measurements and time-updated models are warranted. Fourth, usual walking speed was assessed using a self-reported question rather than an objective timed walking test. Self-reported measures are prone to social desirability and information bias, which may have led to non-differential misclassification of physical performance and potentially attenuated the observed associations. Lastly, UKBB had a low participation rate (~5.5%) and participants are generally healthier and more socioeconomically advantaged than the general UK population in the same age range, reflecting a “healthy volunteer” selection bias [
19]. This likely leads to underestimation of the absolute prevalence of sarcopenia and infections and may limit the generalisability of absolute risk estimates to the wider population. Nevertheless, previous work comparing risk factor–mortality associations in UKBB with multiple nationally representative cohort studies has shown broadly similar hazard ratios for a wide range of established risk factors, suggesting that relative associations such as those reported here for sarcopenia and infection outcomes are likely to be more robust to this selection bias [
50].
5 Conclusions
In conclusion, in this large population-based cohort of UK adults, probable and confirmed sarcopenia were significantly associated with an increased long-term risk of overall infections and infectious diseases. These findings indicate that sarcopenia and its components are important markers of elevated infection risk, particularly in younger and obese individuals. Further research, including causal inference studies and intervention trials, is needed to determine whether preventing or delaying sarcopenia through exercise and muscle-strengthening programms can reduce infection burden and improve health outcomes.
5.0.0.0.1 Acknowledgements
We are grateful to UKBB participants. This research was conducted using the UKBB resource under application number 93044. This work was supported by grants from the National Natural Science Foundation of China (No. 82373337), the Science and Technology Innovation Program of Hunan Province (No. 2025RC1017), and the Research Project on Graduate Educational Reform of Central South University (No. 2025JGB179).
PDF
(3025KB)
