Abstract
Background
Frailty is an increasingly recognized comorbidity in chronic kidney disease and is associated with adverse outcomes. Although extensively studied in dialysis populations, its prevalence and associated factors in nondialysis chronic kidney disease remain less well-defined. This study aimed to estimate frailty prevalence and identify associated demographic, clinical, and laboratory factors in adults with nondialysis chronic kidney disease.
Methods
This cross-sectional study used data from the National Health and Nutrition Examination Survey 2007–2018, including 9156 adults with nondialysis chronic kidney disease. Frailty status (robust, prefrail, and frail) was assessed using a modified Fried phenotype. Survey-weighted multinomial logistic regression was applied to explore associations with demographic, clinical, and laboratory variables.
Results
The robust group comprised 30.3% of participants, the prefrail group 67.2%, and the frail group 2.5%. Most participants were in early-stage chronic kidney disease (G1–G2). Frailty was associated with older age, hypertension, cardiovascular disease, and lower education level. Laboratory findings associated with frailty included lower albumin and hemoglobin, higher blood urea nitrogen, lower bilirubin, reduced globulin, decreased lymphocyte count, and elevated monocyte count.
Conclusions
Frailty and prefrailty are highly prevalent even in early-stage chronic kidney disease. These findings highlight the need for early screening and integrated care strategies targeting modifiable clinical and sociodemographic factors.
Keywords
Introduction
Chronic kidney disease (CKD), increasingly prevalent due to demographic shifts and lifestyle-related conditions such as hypertension and diabetes, has emerged as a major global public health challenge. 1 Affecting millions worldwide, CKD not only compromises individual health but also places a substantial burden on healthcare systems. It is characterized by a progressive loss of kidney function, declining estimated glomerular filtration rate (eGFR), and often accompanied by proteinuria.2,3 The progression of CKD is closely interlinked with comorbidities such as hypertension and metabolic syndrome, further complicating patient management. In nondialysis CKD, metabolic alterations and dietary restrictions frequently result in nutritional deficits that may contribute to malnutrition and physical decline. 4
Frailty is a multidimensional clinical syndrome characterized by diminished physiological reserve and heightened vulnerability to stressors. It originated within geriatric medicine as a framework to capture biological aging beyond chronological age.5,6 It is now recognized as a nonlinear consequence of cumulative dysregulation across multiple physiological systems, rather than the result of dysfunction in any single organ or disease. Age-related alterations in inflammatory pathways, endocrine regulation, hematologic function, neuromuscular integrity, and metabolic homeostasis have all been implicated in the development and progression of frailty.
This multisystem perspective is particularly relevant in chronic disease populations, including individuals with CKD, in whom chronic inflammation, anemia, metabolic disturbances, and nutritional impairment frequently coexist. Such overlapping biological mechanisms may predispose patients with CKD to early functional vulnerability even before the onset of advanced organ failure, highlighting frailty as a critical construct for risk stratification in aging CKD populations. In nondialysis CKD, the prevalence of frailty has been reported to range from 7.9% to 16%, increasing to more than 20% in advanced CKD stages,7,8 and frailty in this context is associated with approximately a twofold higher risk of mortality. 9 With the growing number of older adults diagnosed with CKD, identifying potentially modifiable risk factors for frailty has become an urgent clinical and public health priority.
Although the prognostic significance of frailty is well-established in end-stage renal disease (ESRD) populations, 10 determinants of frailty in nondialysis CKD remain less well understood. Factors such as systemic inflammation, nutritional deficits, and reduced physical activity suggest a complex interplay that warrants further investigation. To address this gap, we analyzed a nationally representative US cohort to examine associations between frailty, socioeconomic status, and nutritional biomarkers in patients with nondialysis CKD. Our study aimed to identify key predictors that may inform strategies for early intervention and improved patient outcomes.
Methods
Study design and patients
This cross-sectional study used data from the National Health and Nutrition Examination Survey (NHANES), an ongoing program designed to assess the health and nutritional status of the US population through a complex, multistage probability sampling design. Six NHANES cycles (2007–2018) were included in the analysis. Of 59,842 participants, adults aged ≥18 years with nondialysis CKD were identified, and participants who reported receiving dialysis treatment at the time of the survey were excluded. After excluding individuals with incomplete covariate data, 9156 participants were eligible for the final analysis (Figure 1). NHANES protocols were approved by the National Center for Health Statistics (NCHS) Research Ethics Review Board, and written informed consent was obtained from all participants at the time of data collection. This secondary analysis used publicly available, fully deidentified NHANES data and was therefore deemed exempt from Institutional Review Board (IRB) review in accordance with institutional policies. The present study was conducted in accordance with the principles of the Declaration of Helsinki, as revised in 2024.
Flow chart of sample selection. NHANES: National Health and Nutrition Evaluation Survey.
The study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 11
Chronic kidney disease
Serum creatinine was used to measure eGFR using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.
12
Participants with an eGFR <60 mL/min/1.73 m2 or a urinary albumin-to-creatinine ratio (UACR) ≥30 mg/g were classified as having CKD according to established criteria.13,14 CKD stages were categorized as follows: G1. eGFR ≥90 mL/min/1.73 m2, indicative of kidney disease despite normal eGFR; G2. eGFR 60–89 mL/min/1.73 m2, mild loss of kidney function; G3a. eGFR 45–59 mL/min/1.73 m2, mild-to-moderate loss of kidney function; G3b. eGFR 30–44 mL/min/1.73 m2, moderate-to-severe loss of kidney function; G4. eGFR 15–29 mL/min/1.73 m2, severe loss of kidney function; G5. eGFR <15 mL/min/1.73 m2, kidney failure.
Modified frailty assessment
Frailty was assessed using a modified Fried frailty phenotype, consistent with prior NHANES-based studies.15–16 The following five criteria were used.
Weakness. Self-reported difficulty lifting or carrying 10 pounds, defined as responses 2–4 on item PFQ061E;
Low physical activity. The lowest quintile of self-reported moderate-to-vigorous physical activity within the study population (PAQ670);
Fatigue. Self-reported feelings of tiredness and low energy, indicated by responses 2 or 3 on the DPQ040 survey item;
Slow walking speed. Difficulty walking between rooms on the same floor, as indicated by responses 2, 3, or 4 on the PFQ061H survey item;
Underweight. Determined by a body mass index (BMI) of less than 18.5 kg/m2.
Participants meeting ≥3 criteria were categorized into the frail group, those meeting 1–2 criteria into the prefrail group, and those meeting none of the criteria into the robust group.
Covariates
Potential confounders included age, sex, race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, and other), education level, marital status, BMI, hypertension, diabetes, cardiovascular disease (CVD), and relevant laboratory variables. Hypertension was defined as blood pressure ≥140/90 mmHg, self-reported hypertension, or the use of antihypertensive medications. Diabetes was defined as fasting blood glucose ≥126 mg/dL, self-reported diabetes, or the use of hypoglycemic agents or insulin. CVD was defined as self-reported diagnosis of heart failure, coronary heart disease, angina, heart attack, or stroke.
BMI was calculated as weight in kilograms divided by height in meters squared. Weight and height conversions were performed using the standard NHANES conversion factors. Detailed information on variable definitions and data collection procedures is publicly available on the NHANES website (https://www.cdc.gov/nchs/nhanes/).
Statistical analysis
In accordance with Centers for Disease Control and Prevention (CDC) guidelines for analyzing multistage cluster survey data, 17 all analyses accounted for the complex sampling design and applied appropriate sampling weights to ensure national representativeness. Continuous variables were presented as mean ± SD and categorical variables as frequencies and percentages. Differences across frailty groups (robust, prefrail, and frail) were assessed using survey-weighted regression models for continuous variables and Rao–Scott chi-square tests for categorical variables. Effect sizes, including Cohen’s d for continuous variables and Cramer’s V for categorical variables, were calculated to facilitate the interpretation of clinical relevance.
Multivariable survey-weighted logistic regression models were constructed to identify factors independently associated with frailty, with robust participants serving as the reference group. Candidate covariates included age, sex, race/ethnicity, BMI, hypertension, diabetes, CVD, and selected laboratory parameters. To reduce the risk of overfitting, variables with p <0.10 in univariate analyses or those supported by prior literature were retained in the final multivariable models. Multicollinearity was assessed using variance inflation factors (VIFs), with values <10 indicating acceptable levels of collinearity. Results were presented as odds ratios (ORs) with 95% confidence intervals (CIs), and statistical significance was defined as a two-sided p-value <0.05.
All analyses were performed using Statistical Package for Social Sciences (SPSS) 25.0 (IBM Corporation; Armonk, NY, USA) and R 4.1.1.18,19
Results
Demographic and clinical characteristics of participants
A total of 9156 adults with nondialysis CKD were included in the analysis, of whom 68.9% were male. Participants were categorized into three age groups (18–59, 60–74, and ≥75 years). According to the frailty criteria, 2767 participants (30.3%) were categorized into the robust group, 6158 (67.2%) into the prefrail group, and 231 (2.5%) into the frail group (Table 1).
Baseline characteristics of participants.
CKD: chronic kidney disease.
Significant differences in demographic and clinical characteristics were observed across frailty groups (Table 1). Compared with participants in the robust group, those in the frail group were older, more likely to be female, less likely to be married, and to have lower educational attainment. Participants in the frail group also exhibited lower BMI, hemoglobin, and serum albumin levels along with higher blood urea nitrogen (BUN) concentrations. In addition, significant differences were observed across frailty groups for several biochemical markers, including serum calcium, cholesterol, glucose, total bilirubin, triglycerides, sodium, globulin, white blood cell count, and glycated hemoglobin (HbA1c). By contrast, no significant differences were detected in serum creatinine, liver enzymes, phosphorus, total protein, uric acid, potassium, or lymphocyte percentage (all p > 0.05). Monocyte count demonstrated a significant increasing trend across frailty groups (p < 0.001). Furthermore, the distribution of CKD stages differed significantly across frailty groups (p < 0.001), with a higher proportion of advanced CKD (G4–G5) observed among participants in the frail group, whereas early-stage CKD (G1–G2) was more prevalent in the robust group.
Logistic regression analysis
Multivariable logistic regression analysis (Table 2) identified several demographic, clinical, and laboratory factors independently associated with frailty among participants with nondialysis CKD. Age was strongly associated with frailty. Compared with participants aged ≥75 years, those aged 60–74 years demonstrated significantly lower odds of frailty (OR = 0.63, 95% CI: 0.56–0.71), whereas participants aged 18–59 years exhibited an even greater reduction in frailty risk (OR = 0.15, 95% CI: 0.10–0.23), demonstrating a clear age gradient in frailty likelihood.
Results of multivariate logistic regression analysis of frailty.
Participants in the robust group were used as the reference group. Logistic regression models were adjusted for demographic, clinical, and laboratory covariates. Data are presented as odds ratios (OR) with 95% confidence intervals (CI). *p < 0.05, **p < 0.01, ***p < 0.001.
CKD: chronic kidney disease; CVD: cardiovascular disease; BMI: body mass index; ref: reference.
Socioeconomic factors demonstrated stage-specific associations with frailty (Table 2). Marital status was not independently associated with either prefrailty or frailty after multivariable adjustment. In contrast, education level was significantly associated with prefrailty but not with frailty. Participants with low education level had higher odds of prefrailty compared with those with high education (OR = 1.48, 95% CI: 1.28–1.71, p < 0.001), whereas medium education showed no significant association. Education level was not independently associated with frailty in the fully adjusted model.
Among laboratory variables, several markers were independently associated with frailty in the fully adjusted models (Table 2). Serum albumin was inversely associated with frailty (OR = 0.32, 95% CI: 0.22–0.46), whereas its association with prefrailty was not statistically significant. BUN was positively associated with frailty (OR = 1.16, 95% CI: 1.12–1.20) but not with prefrailty. Total bilirubin was not associated with prefrailty; however, lower bilirubin levels were observed among participants in the frail group, consistent with an inverse association in the adjusted model (OR = 0.32, 95% CI: 0.19–0.53). Triglycerides were associated with higher odds of prefrailty (OR = 1.15, 95% CI: 1.01–1.32). The white blood cell count was inversely associated with prefrailty (OR = 0.974, 95% CI: 0.958–0.989). Globulin and lymphocyte counts were inversely associated with both prefrailty and frailty (globulin: OR = 0.62 and 0.76; lymphocyte count: OR = 0.35 and 0.43, respectively), whereas monocyte count was significantly associated with higher odds of frailty (OR = 1.88, 95% CI: 1.26–2.81). Hemoglobin levels were not independently associated with prefrailty after multivariable adjustment, whereas lower hemoglobin levels were significantly associated with a higher likelihood of frailty (OR = 0.66, 95% CI: 0.59–0.75).
Comorbidities also showed significant associations with frailty. Hypertension was independently associated with higher odds of frailty (OR = 1.55, 95% CI: 1.12–2.18), whereas cardiovascular disease was associated with a substantially increased likelihood of frailty (OR = 1.94, 95% CI: 1.45–2.60).
Finally, the CKD stage was significantly associated with frailty status. Compared with participants with advanced CKD (G4–G5), those with early-stage CKD (G1–G2) had substantially lower odds of frailty (OR = 0.26, 95% CI: 0.14–0.47), and individuals with moderate CKD (G3a–G3b) also showed reduced odds of frailty (OR = 0.41, 95% CI: 0.22–0.76). These findings suggest a stepwise increase in frailty risk with more advanced CKD stages.
Discussion
This study, using data from the NHANES, demonstrates a high prevalence of frailty and prefrailty among patients with CKD, with approximately two-thirds of participants classified as prefrail or frail. These findings underscores the importance of recognizing frailty as a critical yet underappreciated comorbidity in routine nephrology practice. Consistent with disease progression, frailty risk increased in a graded manner with advancing CKD stage, with the highest odds observed among patients with advanced CKD (G4–G5). Nevertheless, a substantial proportion of frailty and prefrailty was already present in earlier CKD stages, suggesting that vulnerability to functional decline may emerge well before the onset of end-stage kidney disease. Together, these findings indicate that frailty in CKD is not solely a consequence of advanced renal dysfunction but may reflect the cumulative impact of metabolic disturbances, chronic inflammation, and the comorbidity burden that begin earlier in the disease course, highlighting the need for proactive frailty screening and integrated management strategies from the early stages of CKD.
Frailty prevalence and comparisons with previous studies
Our findings are broadly consistent with previous NHANES-based investigations examining frailty among patients with CKD. Blodgett et al. 20 reported prevalences of frailty and prefrailty of 3.6% and 26.8%, respectively, using NHANES data from earlier survey cycles. Similarly, Wilhelm-Leen et al. 7 identified a frailty prevalence of 2.8% based on a modified Fried frailty phenotype. By incorporating more recent NHANES cycles spanning 2007–2018, the present study extended these earlier observations and confirmed a comparable prevalence of frailty (approximately 3%) and revealed a substantially higher prevalence of prefrailty (approximately 67%). 21
Differences in reported prevalence across studies are likely attributable to variations in survey periods, operational definitions of frailty components, and underlying population characteristics. In addition, the inclusion of more recent NHANES cycles, which capture an aging population with a higher burden of metabolic disorders and comorbidities, may partially explain the increased detection of prefrailty. Together, these findings suggest that although overt frailty prevalence has remained relatively stable over time, prefrailty represents an increasingly common and clinically relevant condition among patients with CKD.
Frailty and related factors: clinical, demographic, and laboratory associations
Our findings confirm a strong and graded association between age and frailty among patients with CKD. Compared with participants aged ≥75 years, younger age groups exhibited substantially lower odds of frailty, highlighting the well-established role of aging in the development of frailty. This age-related vulnerability is likely amplified in CKD because of cumulative physiological stress, metabolic disturbances, and reduced physiological reserve associated with declining renal function.22,23
With respect to kidney disease severity, frailty risk increased progressively with advancing CKD stage. Participants with early-stage CKD (G1–G2) and moderate CKD (G3a–G3b) exhibited significantly lower odds of frailty compared with those with advanced CKD (G4–G5), indicating a graded increase in frailty burden as kidney function deteriorates. Importantly, despite the lower relative risk in earlier stages, a substantial proportion of prefrailty and frailty was already present among patients with mild-to-moderate CKD, underscoring that functional vulnerability may emerge well before the onset of end-stage kidney disease.7,24 These findings support the need for frailty assessment and intervention across the entire CKD spectrum, rather than focusing solely on advanced stages.
Socioeconomic factors demonstrated differential associations with frailty status. Education level was independently associated with prefrailty but not with frailty in the fully adjusted models, suggesting that lower education may contribute to earlier functional vulnerability rather than established frailty. This pattern is consistent with the notion that limited health literacy and reduced access to preventive resources may predispose patients with CKD to early functional decline.25–27 In contrast, marital status was not independently associated with either prefrailty or frailty after multivariable adjustment, indicating that its apparent effects in the unadjusted analyses may be mediated through other demographic or clinical factors.
Biological and inflammatory markers: albumin, BUN, and inflammation
In this study, several laboratory biomarkers were independently associated with frailty among patients with CKD, highlighting the complex interplay between nutritional status, metabolic burden, and immune dysregulation. Serum albumin, a well-established marker of nutritional reserve and systemic inflammation, was inversely associated with frailty, with higher albumin levels conferring significantly lower odds of frailty (OR = 0.32, 95% CI: 0.22–0.46); however, no significant association was observed for prefrailty. This pattern is consistent with prior studies in patients with CKD, in which hypoalbuminemia has been repeatedly identified as a robust marker of frailty, disability, and adverse clinical outcomes.28–30 In the context of CKD, serum albumin reflects not only nutritional status but also chronic inflammation and the malnutrition–inflammation complex syndrome, both of which are central to frailty pathophysiology.
BUN, reflecting both renal function and protein metabolism, demonstrated a positive association with frailty but not with prefrailty. Higher BUN levels were associated with increased odds of frailty, suggesting that a greater metabolic burden and impaired nitrogen handling may contribute to advanced functional decline in patients with CKD. 31 In contrast, BUN did not demonstrate a significant association with prefrailty, indicating that its impact may be more pronounced at later stages of vulnerability. Total bilirubin exhibited an inverse association with frailty; however, no significant relationship was observed for prefrailty. Although bilirubin is traditionally recognized as a marker of hepatic function, emerging evidence suggests that it may also reflect antioxidant capacity. 32 The observed protective association may indicate that lower bilirubin levels are associated with increased oxidative stress and frailty susceptibility, although the underlying mechanisms warrant further investigation.
Markers related to immune and inflammatory status also demonstrated distinct associations. White blood cell count was inversely associated with prefrailty but not with frailty, suggesting stage-specific inflammatory responses during early functional decline. Globulin levels were inversely associated with both prefrailty and frailty, indicating that reduced immune protein availability or altered inflammatory regulation may play a role in frailty development. 33 Similarly, lymphocyte count showed protective associations with both prefrailty and frailty, consistent with the concept of immunosenescence contributing to frailty risk. 34 In contrast, monocyte count was positively associated with frailty, but not with prefrailty, supporting the role of chronic low-grade inflammation and innate immune activation in advanced frailty.
Hemoglobin levels were not independently associated with prefrailty, whereas lower hemoglobin levels were significantly associated with increased odds of frailty. This pattern suggests that hemoglobin may be more strongly related to advanced functional decline rather than early vulnerability. Previous studies in patients with CKD have shown that hemoglobin levels are strongly influenced by disturbances in iron metabolism and renal pathology, which may limit their ability to capture subtle physiological changes in earlier stages of frailty.35,36 In addition, the relatively small number of cases of participants in the frail group in our study may have reduced statistical power to detect more nuanced associations, particularly for prefrailty.
Taken together, these findings suggest that laboratory markers reflect stage-specific biological processes in frailty progression, with metabolic burden and immune dysregulation becoming increasingly prominent as frailty progresses. The differential associations observed across prefrailty and frailty underscore the complexity of frailty biology in CKD and highlight the need for longitudinal studies to clarify causal pathways.
Comorbidities and frailty
In this study, comorbidities demonstrated distinct associations with frailty status. Hypertension was independently associated with higher odds of frailty after multivariable adjustment, a finding consistent with evidence from a systematic review and meta-analysis reporting a high prevalence of frailty among patients with hypertension and a generally positive association between the two conditions in observational studies. 37 Given the cross-sectional design of most available studies, including ours, this association likely reflects the cumulative burden of vascular risk, long-term antihypertensive treatment, and shared pathophysiological pathways, rather than a direct causal relationship.
CVD demonstrated a substantially stronger association with frailty. This observation is consistent with longitudinal evidence from large prospective cohorts showing that the progression of frailty markedly increases the risk of incident CVD, whereas improvement in frailty status is associated with reduced cardiovascular risk. 38 Taken together, these findings support the concept that frailty and CVD are closely intertwined conditions characterized by bidirectional, dynamic interactions, particularly in patients with CKD.
Study strengths and limitations
The strengths of this study include the use of nationally representative NHANES data, a large sample of patients with nondialysis CKD, and comprehensive adjustment for demographic, socioeconomic, clinical, and laboratory covariates using survey-weighted methods, all of which together enhance the robustness and population-level generalizability of our findings. Several limitations should also be acknowledged. First, the cross-sectional design precludes causal inference and limits the ability to establish temporal relationships among CKD progression, biological markers, and frailty. Second, frailty was assessed using a modified Fried phenotype based on variables available in NHANES rather than on objective physical performance tests. Particularly, weakness and slowness were determined using self-reported functional difficulty rather than direct performance-based measures. Although this approach has been widely used in NHANES-based frailty research, it may be less sensitive to subtle impairments, particularly in the early stages of functional decline. Finally, although extensive covariates were included, residual confounding from unmeasured factors cannot be entirely excluded.
Conclusions
In conclusion, this nationally representative study demonstrates that frailty and prefrailty are highly prevalent among patients with nondialysis chronic kidney disease and are evident even in the early stages of CKD. Frailty was independently associated with demographic, clinical, and biological factors, highlighting it as a clinically meaningful condition rather than as a complication limited to advanced kidney dysfunction. Although causal relationships cannot be inferred due to the cross-sectional design, these findings underscore the importance of early screening for frailty and integrated management in CKD care. Longitudinal studies are warranted to clarify underlying mechanisms, particularly the roles of inflammation, nutrition, and cardiovascular health, and to evaluate whether early targeted interventions can attenuate frailty progression and improve outcomes in this vulnerable population.
Footnotes
Availability of data and materials
Competing interests
The authors have disclosed no potential conflicts of interest, financial or otherwise.
Contributions
Conceptualization: Y.Y; Investigation: Y.Y, F.Y, Y.Z, L.Z; Methodology: F.Y; Resources: L.Z, Z.Y; Writing-original draft: Y.Y; Writing-review and editing: F.Y, L.Z. All authors have read and agreed to the published version of the manuscript.
Ethics declarations
The survey was approved by the National Center for Health Statistics (NCHS) Ethics Review Board (protocol number: 98–12). All experiments were performed in accordance with relevant guidelines and regulations.
Clinical trial number
Not applicable.
Funding
No Funding.