Functional impairment as predictors of renal outcomes and mortality in elderly patients: a retrospective cohort study

Article information

Korean J Nephrol. 2026;.j.krcp.25.098

Publication date (electronic) : 2026 March 6

doi : https://doi.org/10.23876/j.krcp.25.098

Bokeung Peun^1,2, Seunghyun Won³, Kwang-il Kim^2,4, Jung-Yeon Choi^2,4, Ho Jun Chin^,2,4

¹Department of Psychiatric Nursing, College of Nursing, Seoul National University, Seoul, Republic of Korea

²Department of Internal Medicine, Seoul National University Bundang Hospital, Seongnam, Republic of Korea

³Medical Research Collaborating Center, Seoul National University Bundang Hospital, Seongnam, Republic of Korea

⁴Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

Correspondence: Ho Jun Chin Department of Internal Medicine, Seoul National University Bundang Hospital, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 13620, Republic of Korea. E-mail: mednep@snubh.org

Received 2025 April 13; Revised 2025 December 15; Accepted 2025 December 16.

Abstract

Introduction

Functional decline in older adults is a well-recognized marker of frailty and an independent predictor of morbidity and mortality. Activities of daily living (ADL) and instrumental activities of daily living (IADL) scores not only reflect an individual’s ability to perform essential self-care and complex daily tasks but also provide prognostic insights into clinical outcomes [1,2]. Elderly individuals with reduced ADL or IADL face a higher risk of hospitalization, disability, and mortality, highlighting the importance of functional assessment in geriatric care [3,4].

Additionally, aging is naturally associated with a decline in renal function, and the presence of multiple comorbidities such as diabetes mellitus (DM), hypertension, and cardiovascular disease accelerates renal deterioration, thereby increasing the risk of adverse health outcomes [5–7]. In addition to traditional risk factors, as measured by ADL and IADL, is emerging as a key determinant of health trajectories in elderly patients with chronic kidney disease (CKD) [3,8].

Despite a growing body of evidence, limited research has specifically examined the role of functional status in predicting renal outcomes, including the need for renal replacement therapy (RRT) and mortality in elderly patients [9]. Although impaired renal function correlates with poor physical and cognitive performance, the extent to which functional impairments influence CKD progression and survival remains unclear, particularly in patients who have not yet undergone dialysis. Moreover, sex-specific differences in functional status assessments add another layer of complexity, as men and women may experience and adapt differently to functional impairments.

Understanding these relationships is critical for identifying at-risk populations and developing targeted interventions to improve patient outcomes. This study aimed to address these gaps by investigating the association between ADL and IADL scores and renal outcomes, including RRT initiation and mortality, in elderly patients. By analyzing a retrospective cohort of elderly patients who underwent comprehensive geriatric assessments (CGAs), we aimed to determine the prognostic value of functional status and to identify clinical subgroups that may benefit from early intervention strategies. Moreover, it remains unclear whether ADL and IADL provide prognostic information beyond traditional variables such as age, body mass index (BMI), or comorbidity burden; this study seeks to clarify this through multivariate analysis.

This study aimed to examine the differences in RRT rates and mortality according to the levels of ADL and IADL in elderly patients and to determine whether ADL or IADL had a greater impact on RRT rates and mortality in this population.

Methods

Study design and study subjects

This retrospective cohort study analyzed data from 8,179 elderly patients who underwent a CGA at Seoul National University Bundang Hospital between 2016 and 2020. Eligible participants were those aged ≥65 years who had undergone ADL and IADL assessments, had an estimated glomerular filtration rate (eGFR) of >15 mL/min/1.73 m², were not undergoing dialysis at the time of assessment, and had a follow-up duration of more than 3 months (Fig. 1).

Figure 1.

Selection of patients.

ADL, activities of daily living; eGFR, estimated glomerular filtration rate; IADL, instrumental activities of daily living.

A total of 6,087 participants met the inclusion criteria and were included in the analysis (2,737 males and 3,350 females). Considering the differences in the IADL scoring systems (5 points for males and 8 points for females), the analyses were conducted separately for men and women.

Clinical data, including RRT initiation, were extracted from electronic health records (EHR). Over a median follow-up of 42 months, 2,108 participants (34.6%) died, and 103 (1.7%) required RRT. Data completeness is presented in Supplementary Table 1 (available online). Mortality data were requested from the Ministry of Interior and Safety of Korea and merged with data from the EHR using unique personal identifiers of all Korean patients.

Results

Characteristics and follow-up status of the participants

Participants were categorized into ADL groups (ADL1 vs. ADL2) and IADL groups (IADL1 vs. IADL2) according to their functional status (Table 1). Across both sexes, impaired groups (ADL2/IADL2) consistently demonstrated older age, lower BMI, reduced hemoglobin, albumin, and protein levels, and a higher prevalence of major comorbidities, including DM, hypertension, and cardiovascular disease. These trends corresponded with small-to-moderate effect sizes for most clinical variables, indicating that functional impairment was accompanied by a measurable degree of metabolic and physiological decline.

Table 1.

Demographic and clinical characteristics of the study population

Although cancer prevalence did not differ meaningfully between functional groups, this pattern may reflect survivorship or selection bias, as individuals with advanced malignancy may have been less likely to undergo outpatient geriatric assessment.

Sex-specific patterns were also evident. In males, functional impairment was associated with larger effect sizes for BMI, serum protein, and inflammatory markers, suggesting a closer link between nutritional burden and functional deterioration. In contrast, females exhibited greater effect sizes in renal parameters such as serum creatinine and eGFR, indicating heightened renal vulnerability among women with reduced functional capacity. While the overall direction of differences was consistent, the magnitude varied by sex, supporting the need for sex-specific interpretation of functional decline.

Notably, despite a higher prevalence of hypertension among impaired groups, both systolic and diastolic blood pressure values were lower in these participants. This paradoxical pattern, frequently observed in frail older adults, likely reflects intensive pharmacologic management together with age-related autonomic dysregulation.

Taken together, effect-size patterns highlight a shared clinical phenotype of metabolic frailty among individuals with functional impairment, while also revealing distinct physiological correlates by sex. These findings suggest a complex interplay between comorbidity burden, aging processes, and functional capacity in elderly populations.

Impact of activities of daily living and instrumental activities of daily living on renal replacement therapy and mortality

Kaplan-Meier survival curves illustrated visually distinct trajectories for both RRT incidence and overall survival between participants with normal and abnormal ADL or IADL. These curves are presented for descriptive purposes only and are not intended for inferential comparison. In both males and females, impairment in ADL or IADL was associated with a reduced survival rate. Notably, among males, the incidence of RRT increased more prominently in those with IADL impairment, suggesting a stronger association with progression to kidney failure (Fig. 2).

Figure 2.

Kaplan-Meier survival curves for RRT-free and overall survival.

(A) RRT-free survival curves according to ADL/IADL status by sex. (B) Overall survival curves according to ADL/IADL status by sex.

ADL, activities of daily living; IADL, instrumental activities of daily living; RRT, renal replacement therapy.

The associations between functional decline (assessed by ADL and IADL) and clinical outcomes were explored using Cox proportional hazards models adjusted for demographic and clinical covariates (Table 2). Subgroup analyses were additionally performed to evaluate effect modification by age, comorbidity burden, and renal function, with corresponding interaction terms formally tested within a single-model framework (Table 3).

Table 2.

Impact of ADL and IADL on outcomes

Table 3.

Subgroup analyses of functional impairment for renal outcomes and mortality

For the outcome of RRT initiation, the predictive value of functional impairment varied by sex. In males, both abnormal IADL and ADL scores were significantly associated with an increased risk of RRT initiation. Specifically, males with IADL <5 had a 2.3-fold higher risk (95% CI, 1.3–4.4; p = 0.008), and those with ADL impairment showed a moderate trend (HR, 1.9; 95% CI, 1.0–3.5; p = 0.05). In the subgroup analysis, younger males (<75 years) with abnormal IADL scores exhibited the highest risk (HR, 14.2; 95% CI, 4.6–44.3; p < 0.001), while older males (≥75 years) also had elevated risk (HR, 3.4; 95% CI, 2.0–6.1; p < 0.001). Importantly, the interaction term between age group and IADL status was statistically significant (p = 0.02), confirming that age significantly modifies the predictive utility of IADL with RRT risk in males. Similar interaction-based subgroup differences were observed among those with preserved renal function (GFR, ≥75 mL/min/1.73 m²), where abnormal IADL conferred a higher risk (HR, 5.3; 95% CI, 2.0–13.8; p < 0.001) compared to abnormal ADL (HR, 4.2; 95% CI, 1.6–11.0; p = 0.004). Males with lower comorbidity burden (CCI, <3) also showed stronger associations with IADL impairment (HR, 5.0; 95% CI, 2.1–11.9; p < 0.001) than ADL (HR, 2.9; 95% CI, 1.1–7.4; p = 0.03). In contrast, among females, the association between functional impairment and RRT initiation was more modest. Although abnormal ADL scores showed a tendency toward an increased risk of RRT (HR, 2.0; 95% CI, 1.0–4.4; p = 0.07), this association did not reach statistical significance. Likewise, abnormal IADL impairment was not significantly associated with RRT risk (HR, 1.5; 95% CI, 0.7–3.3; p = 0.30). Interaction terms for age and hypertension were statistically significant (p = 0.03 and p = 0.03, respectively), suggesting potential modification effects in these subgroups. For mortality, both ADL and IADL impairments were strongly associated with increased risk in males and females. In males, group-level IADL impairment was associated with an increased risk of death (HR, 1.9; 95% CI, 1.6–2.2; p < 0.001), while abnormal ADL also showed a significant association (HR, 1.7; 95% CI, 1.4–2.0; p < 0.001). Subgroup analyses showed that these effects persisted across age and renal function strata. Significant interaction effects were observed by DM status, cancer history, and comorbidity burden (all p < 0.001), suggesting differential mortality risk patterns based on these clinical characteristics. In females, both ADL (HR, 2.3; 95% CI, 1.9–2.7; p < 0.001) and IADL (HR, 2.1; 95% CI, 1.8–2.5; p < 0.001) impairments were significantly associated with higher mortality, with ADL showing a slightly stronger effect. Subgroup analyses revealed that younger elderly females (<75 years) with ADL impairment had a higher mortality risk (HR, 4.2; 95% CI, 3.0–5.9; p < 0.001) compared to those aged ≥75 years (HR, 3.0; 95% CI, 2.6–3.4; p < 0.001), although the age interaction was not statistically significant (p = 0.23). Both ADL and IADL impairments showed significant interaction effects with DM status, cancer history, and comorbidity burden (ADL: all p < 0.001; IADL: p < 0.001 and p = 0.004), indicating variable mortality risks according to these clinical factors.

Taken together, these findings highlight that functional decline assessed by ADL or IADL should be interpreted in a sex- and outcome-specific manner. While both ADL and IADL impairments are relevant predictors of adverse outcomes, their relative impact differs by sex, age, and comorbidity status. IADL was a stronger predictor of RRT in males, especially in younger and healthier subgroups, whereas ADL was more predictive of mortality in females. These differences underscore the importance of incorporating interaction terms in modeling to capture nuanced effect modifications and emphasize the clinical value of routine IADL assessment, particularly in males at risk of kidney failure.

ROC analysis demonstrated that IADL had higher or comparable predictive power for mortality compared to ADL in both sexes. In males, IADL showed slightly better performance in predicting mortality (AUC_IADL, 0.625; 95% CI, 0.603–0.647) than ADL (AUC_ADL, 0.609; 95% CI, 0.587–0.631; p < 0.001). In females, IADL also demonstrated superior predictive accuracy for mortality (AUC_IADL, 0.704; 95% CI, 0.683–0.725) compared to ADL (AUC_ADL, 0.688; 95% CI, 0.666–0.709; p < 0.001) (Figure 3).

Figure 3.

ROC curves of ADL and IADL for predicting RRT and mortality.

(A) ROC curves for RRT between ADL and IADL by sex. (B) ROC curves for mortality between ADL and IADL by sex.

ADL, activities of daily living; AUC, area under the curve; IADL, instrumental activities of daily living; ROC, receiver operating characteristic; RRT, renal replacement therapy.

However, for RRT initiation, there was no statistically significant difference in predictive ability between IADL and ADL in either males (AUC_IADL, 0.630 [95% CI, 0.553–0.708] vs. AUC_ADL, 0.615 [95% CI, 0.536–0.694]; p = 0.26) or females (AUC_IADL, 0.654 [95% CI, 0.562–0.746] vs. AUC_ADL, 0.654 [95% CI, 0.564–0.744]; p > 0.99).

Discussion

Functional impairment, measured by ADL and IADL, independently predicted RRT initiation and mortality, with distinct sex-specific patterns. In males, IADL impairment was strongly associated with RRT risk (HR, 2.3; p = 0.008), particularly in younger males (<75 years; HR, 14.2) and those with preserved renal function (eGFR, ≥75 mL/min/1.73 m²; HR, 5.3). ADL showed a weaker, borderline association (p = 0.05). In females, functional status was not significantly associated with RRT in categorical models, with significant interactions by age and hypertension. For mortality, both ADL and IADL impairments predicted a higher risk in both sexes. In males, IADL (HR, 1.9) and ADL (HR, 1.7) were significant; in females, ADL had a stronger association with mortality (HR, 2.3) than IADL (HR, 2.1). Mortality risk appeared higher among younger elderly (<75 years; HR, 4.2) compared to those aged ≥75 years (HR = 3.0), although the interaction by age was not statistically significant (p = 0.234). These findings support the use of ADL and IADL as outcome-specific, sex-sensitive indicators for geriatric risk stratification.

These findings confirm that functional decline contributes to adverse health outcomes in older adults. Previous research has demonstrated that both frailty and functional limitations predict mortality [5,15]. A study on frailty components in older Chinese adults reported that although the cumulative frailty index was the strongest predictor of mortality, impairments in ADL and IADL were also significant individual contributors [16]. Building on these findings, the present study differentiates the roles of ADL and IADL impairments, emphasizing not only their distinct contributions but also their sex-specific and interaction-dependent effects on both RRT initiation and mortality.

While both impairments were associated with adverse outcomes, their predictive roles differed by sex. IADL impairment significantly predicted RRT initiation in men (p = 0.008), whereas in women, functional status was not significantly associated with RRT initiation, although significant interactions by age and hypertension were observed. This finding may partly reflect the smaller number of RRT events and the lower progression rate to end-stage renal disease among females, leading to limited statistical power and caution in interpretation. These findings align with previous research indicating that functional decline is more closely linked to multimorbidity and frailty than with isolated disease progression [17–20]. ADL, rather than IADL, has been associated with increased mortality in patients undergoing chronic hemodialysis [21,22], while both impairments have been identified as mortality predictors in community-dwelling older adults [3]. In peritoneal dialysis populations, functional dependence across both ADL and IADL is prevalent, with many requiring assistance in daily living [23]. These findings underscore the importance of assessing both ADL and IADL impairments with respect to the individual’s health trajectory: IADL limitations may indicate early functional decline, whereas ADL impairments may reflect advanced disease severity or frailty [24].

This study also highlights the sex-based differences in the association between functional impairment and clinical outcomes. Older males with IADL impairment were more likely to undergo RRT, whereas older females with ADL impairment showed a stronger association with mortality. This aligns with research suggesting that frailty has a greater impact on survival in women, whereas men tend to struggle more with IADL-related tasks such as cleaning and shopping; women demonstrate resilience in these areas [23,25]. This may explain why IADL impairment predicted RRT only in males, while ADL impairment had a stronger link to mortality in females. These results emphasize the need for sex-specific geriatric risk stratification and intervention.

Notably, the study also identifies the vulnerability of younger elderly males (<75 years) with preserved renal function (eGFR, ≥75 mL/min/1.73 m²) to RRT initiation. This underscores the need for early functional screening in seemingly healthy older adults, as functional impairments may indicate a risk of adverse clinical events. This trend was especially pronounced in males under 75 years-old, with significant interaction effects observed (p = 0.02). Functional decline can manifest early in CKD and contribute to disease progression, even in patients with preserved renal function [26–29]. Additionally, in individuals with preserved renal function (eGFR, ≥75 mL/min/1.73 m²), IADL impairments were associated with RRT initiation, whereas in those with reduced renal function (eGFR, <75 mL/min/1.73 m²), ADL impairments were linked to mortality. This highlights the need for tailored assessments and intervention strategies based on baseline renal function.

Comorbidities play a role in shaping clinical outcomes in older adults. In this study, IADL impairment was associated with RRT initiation in individuals with a lower comorbidity burden (CCI, <3). This suggests that functional impairments in relatively healthier older adults may indicate early health deterioration even before significant comorbidities develop. Conversely, among individuals with higher comorbidity burdens (CCI, ≥3), both ADL and IADL impairments were similarly associated with clinical outcomes. As comorbidity burden increased, the distinction between ADL and IADL effects diminished, suggesting converging trajectories of functional decline in frailer populations. Previous research supports the idea that as multimorbidity increases, functional decline becomes a predictor of health deterioration rather than being linked to a specific outcome [30,31]. Psychiatric comorbidities, such as depression, further contribute to this decline, worsening daily functioning and quality of life in patients with CKD [32]. These findings highlight the importance of integrating functional assessments into routine geriatric care, particularly for individuals with fewer comorbidities, in whom early functional decline may serve as a clinical warning sign.

Given these findings, integrating routine ADL and IADL assessments into geriatric nephrology and primary care is essential for the early identification of high-risk older adults and for enabling timely interventions. Additionally, structured rehabilitation programs—including physical and occupational therapy—can help mitigate functional decline and reduce the risk of adverse outcomes in this population [29,32,33]. Psychological resilience has been linked to better functional and health outcomes, suggesting that interventions aimed at enhancing resilience may support long-term well-being [34]. The incorporation of interaction-based risk stratification may improve the precision of interventions targeting functional decline. By distinguishing the impact of ADL and IADL impairment and highlighting their sex-specific associations, this study provides insights into functional decline as a marker of systemic health vulnerability. These findings underscore the importance of individualized care strategies to improve longevity and quality of life in aging populations.

This study has several strengths. The inclusion of an entire population, a large sample size, and an extended follow-up period enhanced the statistical power and reliability, providing robust evidence of the association between functional impairments and health outcomes in elderly patients. The sex-specific analyses further highlighted differences in functional decline, emphasizing the need for tailored approaches in geriatric and nephrology care. By identifying distinct patterns in males and females, this study contributes to a deeper understanding of functional deterioration and its clinical implications.

However, this study had some limitations. First, the retrospective design may have introduced selection bias and limited causal inference. Second, sex-based differences in IADL scores may have affected the comparability of results between males and females, leading to potential measurement bias. Third, unmeasured variables—such as psychosocial support, caregiving roles, and socioeconomic status—were not included, despite their potential influence on functional status and outcomes.

In addition, ADL and IADL were assessed only once at baseline, without clarity on whether the scores reflected chronic disability or temporary decline due to acute illness. This raises the possibility of misclassification, which may have compromised the accuracy of prognostic assessments. Furthermore, since ADL and IADL consist of multiple components, future research should investigate which specific items are most predictive of renal outcomes, potentially enabling the development of a unified or adjusted functional score for risk stratification.

Lastly, the number of RRT events was relatively small (n = 103, 1.7%), which may have contributed to model instability and rare event bias, particularly in subgroup analyses. This limited statistical power may reduce the reliability and generalizability of the RRT-related findings.

This study demonstrated that functional impairment plays a critical role in predicting adverse health outcomes in older adults. Specifically, IADL impairment was associated with increased RRT initiation, particularly among younger elderly males with preserved renal function. In contrast, ADL impairment was more strongly associated with mortality in females, and this effect was more pronounced in subgroups with DM and a higher comorbidity burden. These sex-specific and renal function-dependent patterns underscore the importance of incorporating both ADL and IADL assessments into routine geriatric evaluations. Functional screening can serve as an early warning system, identifying individuals at increased risk before clinical deterioration becomes apparent. To address this risk, tailored interventions—including structured rehabilitation and multidisciplinary geriatric care—should be prioritized. Integrating functional assessments into clinical workflows may enhance risk stratification, support independence, and ultimately improve health outcomes and quality of life in aging populations.

Supplementary Materials

Article information Continued