Abstract
Purpose
Older adults receiving hemodialysis experience geriatric problems that affect quality of life and contribute to high healthcare utilization but are often underrecognized in dialysis care. Scalable approaches to identifying these problems are needed. This study evaluated the agreement between a self-administered screening battery and a geriatrician telehealth evaluation and assessed the feasibility of implementing the screening battery in dialysis clinics.
Patients and Methods
Participants aged ≥55 receiving hemodialysis completed the Geriatric screen for Older adults receiving Dialysis (GOLD), a screening battery assessing seven geriatric problems: depression, anxiety, functional impairment, cognitive impairment, falls risk, limited social support, and low nutritional status, followed by a blinded geriatrician telehealth evaluation. Intra-problem agreement between the GOLD and geriatrician evaluation was assessed using Cohen’s κ, and analyses evaluated alternative screening thresholds. Acceptability and feasibility of the GOLD were evaluated through participant and staff surveys.
Results
The cohort included 50 participants [76% (n=38) Black race, 62% (n=31) male, 16% (n=8) Hispanic/Latino ethnicity] with a mean (SD) age of 68 (7.8) years and a median dialysis vintage of 3.5 [IQR (1, 12.5)] years. Geriatric problems were common, including fall risk (78% geriatrician; 90% GOLD), functional impairment (58% geriatrician; 44% GOLD), and low nutritional status (50% geriatrician; 40% GOLD). Agreement between the GOLD and geriatrician evaluation was substantial for anxiety (κ=0.74), functional impairment (κ=0.65), low nutritional status (κ=0.64), and limited social support (κ=0.65), but not for cognitive impairment, falls risk, or depression. Most participants (37 of 45) reported the GOLD was easy to complete, and a minority (16 of 50) required assistance.
Conclusion
The GOLD demonstrated agreement with geriatrician evaluation for many geriatric problems and was feasible to implement in dialysis clinics. These findings suggest potential utility for scalable geriatric screening in dialysis care, although further work is needed to confirm these findings in larger cohorts and to refine screening for cognitive impairment and other domains.
Keywords: geriatric assessment, hemodialysis, screening
Introduction
Older adults receiving hemodialysis experience a high burden of geriatric problems that influence functional independence, quality of life, and healthcare utilization. Specifically, 50% have mild or moderate cognitive impairment,1 47% experience falls,2 40% have activity of daily living impairments,3 and 22% have depression.4 These geriatric problems can contribute to the high rates of hospitalization and poor quality of life among a severely ill population with limited life expectancy.5,6 These findings highlight the complex geriatric care needs of older adults with kidney failure which are not routinely addressed in traditional nephrology care.7 Because older adults represent ~70% of the dialysis population and geriatric problems occur early among patients receiving dialysis,8,9 there is an urgent need for feasible strategies that identify geriatric problems in the hemodialysis population.
Comprehensive geriatric assessment (CGA), a multidimensional, interprofessional, person-centered approach to identifying and managing geriatric problems, is considered the gold standard for the detection and management of geriatric problems.10 CGA in the dialysis setting has been studied, but it is time- and resource-intensive and not widely available.11 In other clinical subspecialities, such as oncology, self-administered screening batteries have been implemented as an alternative to CGA to identify potential geriatric concerns.12–14 While these tools cannot replace a CGA, they may offer a practical solution to identifying geriatric problems for older adults receiving dialysis. However, practical strategies for implementing geriatric screening in dialysis clinics remain understudied.
Towards a goal of a scalable self-administered geriatric screening approach in dialysis clinics, we compared patient-completed screening with a pragmatic telehealth geriatrician evaluation. The primary objectives of this study were: 1) to evaluate agreement between a self-administered screening battery and a geriatrician telehealth evaluation, and 2) to assess the feasibility and acceptability of implementing a self-administered geriatric screening battery within dialysis clinics. A secondary objective was to examine whether the level of agreement between the self-administered battery and the geriatrician evaluation differed depending on the presence of severe cognitive impairment.
Materials and Methods
Study Population
We recruited community-dwelling adults aged ≥ 55 receiving dialysis at seven dialysis clinics around Durham, North Carolina and San Francisco, California from May 2024 to June 2025. We selected age ≥55 because of the earlier onset of geriatric syndromes commonly observed among individuals receiving dialysis. Exclusion criteria included residents of long-term care, hospice patients, and people with severe cognitive impairment who had no caregiver who could participate with them. This study was approved by the Duke University IRB (Pro00112850). Informed consent was obtained from each participant prior to study commencement. This study complied with the Declaration of Helsinki.
Self-Administered Screening Battery: The GOLD
We compiled nine self-administered geriatric screening instruments into a screening battery, the “Geriatric screen for OLder adults receiving Dialysis” (GOLD). The instruments for the GOLD were selected because of previous validation in older multimorbid populations, widespread use in clinical practice, low participant burden, previous use in a dialysis cohort, or correspondence with a geriatric problem common in the older dialysis population.15–31 The GOLD screened for seven geriatric problems: depression, anxiety, cognitive impairment, functional impairment, fall risk, limited social support, and low nutritional status. The selected instruments are summarized in Table 1.
Table 1.
GOLD Instruments and Corresponding Geriatrician Evaluation
| Geriatric Problem | GOLD Instrument | Geriatrician Evaluation |
|---|---|---|
| Depression | Patient Health Questionnaire 9 (PHQ-9)27 | PHQ-2 followed by PHQ-9 if indicated, and further medical questions at geriatrician’s discretion |
| Anxiety | Geriatric Anxiety Inventory - Short Form (GAI-SF)24 | Generalized Anxiety Disorder 2-item (GAD-2) followed by GAD-732 if indicated, further medical questions |
| Cognitive impairment | Cognitive Change Index (CCI; participant or caregiver)16,17 | Montreal Cognitive Assessment (MoCA) Blind,33 further medical questions |
| Functional impairment | Instrumental Activities of Daily Living,29 Lower Extremity Physical Activities of Daily Living,26 Upper Extremity Physical Activities of Daily Living26 | Older American Resources and Service questionnaire (OARS),29 further medical questions |
| Fall risk | Stay Independent checklist,31 and single-item question: fall in the last year | Pre-selected questions about falls, further medical questions |
| Limited social support | Modified Medical Outcomes Study Social Support Survey (mMOS-SS)18 | Pre-selected questions about living situation and support network |
| Low nutritional status | Simplified Nutritional Appetite Questionnaire (SNAQ)19 | SNAQ, further medical questions |
Notes: This table describes instruments included in the GOLD, a self-administered battery for geriatric problem screening, and the geriatrician’s approach to screening each geriatric problem.
The GOLD was administered through REDCap electronic surveys. Participants who had access to the internet and were comfortable accessing the GOLD online were given the option to complete their surveys remotely. Otherwise, surveys were completed on tablets by participants while they were receiving dialysis. Participants were given a tutorial on how to complete the GOLD and left to complete it on their own, with study staff available if needed, unless an impairment was identified that necessitated study staff assistance for completion.
Participant scores for the GOLD instruments were categorized into two groups: (1) problem not present or (2) problem present. “Problem present” screening thresholds (scores indicating a positive screen) for each instrument were based on published literature, including studies in which the instrument score was associated with adverse outcomes. When multiple GOLD instruments assessed the same geriatric problem, meeting the screening threshold on any one instrument was sufficient to classify the problem as present. More information about the rationale for each screening threshold is included in Supplemental Table 1.
Geriatrician Evaluation
Each participant subsequently underwent a telehealth evaluation by a geriatrician. In-person comprehensive geriatric assessments (or even a modified geriatrician evaluation) were not feasible within the dialysis clinic workflow, which motivated the evaluation of telehealth and self-administered approaches that could be more readily integrated into routine dialysis care. The telehealth geriatrician evaluation occurred while the participant received dialysis or at a separate time scheduled at the participant’s convenience. This evaluation involved assessment of the same seven geriatric problems as the GOLD and included medical history taken at the geriatrician’s discretion as well as administration of instruments used routinely in comprehensive geriatric assessment, some of which were identical to instruments in the GOLD. All geriatrician evaluations were completed by a single geriatrician, who was blinded to the GOLD results. Details of the instruments used in the geriatrician evaluation are summarized in Table 1.
After each geriatrician evaluation, the geriatrician utilized a pre-determined set of criteria (Supplemental Table 2) to assign one of three designations: (1) problem not present, (2) problem present but with no indication to change current management, or (3) problem present and change in management indicated. To match the categories utilized for the GOLD instruments, the two problem present categories were combined into one category for analysis of agreement, yielding (1) problem not present and (2) problem present.
Acceptability and Feasibility
A survey regarding acceptability of the GOLD was given to all participants. This survey included a question on ease of completion, “How easy was the GOLD for you to complete?”, answered on a 1 (not at all easy) to 5 (very easy) Likert scale, and a question on length of time, “Was the amount of time it took to complete the GOLD too long, too short, or just right?”. A separate feasibility survey was completed by study staff and included questions on whether the participant needed any help with completion, and if so, why they needed help.
Additional Measures
Study coordinators interviewed participants to obtain demographics, education level, marital status, employment status, comorbidities (Charlson score survey),34 years receiving dialysis, and mini Montreal Cognitive Assessment (mini MoCA) score.35 Study staff reviewed medications with participants via a list or bottle review to obtain the medication count.
Statistical Analysis and Sample Size
We performed descriptive statistics of demographics, social and clinical characteristics, and survey responses. We assessed agreement between the GOLD instrument scores and the geriatrician evaluation using Cohen’s κ, with a target level of > 0.60 to indicate “substantial agreement”.36 For each geriatric problem, a κ statistic and corresponding 95% confidence interval (CI) were reported. If agreement was below the target, alternative screening thresholds for the GOLD instruments were explored, and κ statistics were recalculated. Sensitivity and specificity were computed for each geriatric problem using the updated screening thresholds to describe their operating characteristics.
Pre-specified subgroup analyses were conducted to assess whether agreement varied by cognitive status. Participants were classified as having severe cognitive impairment (Mini MoCA ≤10) or not (>10). κ statistics and corresponding 95% CIs were generated for each subgroup across all geriatric problems using the updated screening thresholds.
For feasibility reasons, the target sample size was set at 50. We performed a precision analysis: across a range of expected probabilities of agreement between the GOLD and geriatrician evaluation and a sample size of 50, we expected 95% confidence interval widths of 0.39–0.42 when the true kappa was 0.60 and narrower if the estimated kappa was larger.
Results
Cohort Characteristics
Across all clinics, there were 154 adults age ≥55 receiving dialysis, of whom 104 met all eligibility criteria, 45 did not provide consent, 59 consented, and 50 completed both the GOLD and the geriatrician evaluations (Figure 1). Among the 50 participants, the cohort was predominantly male (62%), Black or African American (76%), and not Hispanic or Latino (84%). Most had a high school education or less (62%), were unemployed (80%), and were unmarried (40%), with a mean age of 68 years (SD: 7.8). Median dialysis duration was 3.5 years (Q1, Q3: 1.0, 12.5). Mean Charlson score was 4.9 (SD: 2.8), mini MoCA score was 11.2 (SD: 2.2, maximum 30 points with scores ≥ 26 typically indicating normal cognitive function), and participants were prescribed an average of 11.3 medications (SD: 4.2). Demographic details are summarized in Table 2.
Figure 1.
Flow diagram of study participants. This diagram illustrates the flow of participants through screening, consent, and study completion phases.
Table 2.
Cohort Demographics
| N = 50 | |
|---|---|
| Age (years), Mean (SD) | 67.86 (7.79) |
| Gender | |
| Female | 19 (38.0%) |
| Race | |
| White | 6 (12.0%) |
| Black or African American | 38 (76.0%) |
| Other | 4 (8.0%) |
| Unknown or not reported | 2 (4.0%) |
| Ethnicity | |
| Hispanic or Latino | 8 (16.0%) |
| Education level | |
| Less than High School | 5 (10.0%) |
| High School | 26 (52.0%) |
| College | 9 (18.0%) |
| Graduate School | 10 (20.0%) |
| Marital status | |
| Married | 18 (36.0%) |
| Not Married | 20 (40.0%) |
| Widowed | 12 (24.0%) |
| Employment status | |
| Yes | 10 (20.0%) |
| Years on dialysis, Median (Q1, Q3) | 3.5 (1.0, 12.5) |
| Charlson Score, Mean (SD) | 4.6 (2.7) |
| Mini MoCA scorea | |
| ≤10 | 15 (30.0%) |
| >10 | 35 (70.0%) |
| Medication count, Mean (SD) | 11.3 (4.2) |
Notes: aParticipants were deemed to have severe cognitive impairment if the Mini MoCA score was ≤10.
Agreement Between GOLD and Geriatrician Evaluation
Table 3 presents each geriatric problem, with their original and revised thresholds and their corresponding κ. Six geriatric problems did not achieve the target κ using the pre-specified screening thresholds, but alternative screening thresholds yielded κ >0.60 for all problems except depression, falls, and cognitive impairment. Functional impairment reached target agreement after removing the Physical Lower Extremity Activities of Daily Living instrument and lowering the screening threshold for Physical Upper Extremity Activities of Daily Living to ≥2, while retaining ≤11 for Instrumental Activities of Daily Living (κ; 95% CI: 0.65; 0.45, 0.85). Limited social support met agreement after removing the emotional support instrument and lowering the tangible support screening threshold to ≤4 (κ; 95% CI: 0.65; 0.19, 1.00). Low nutritional status achieved κ >0.60 by adjusting the Simplified Nutritional Assessment Questionnaire screening threshold from ≤14 to ≤13 (κ; 95% CI: 0.64; 0.43, 0.85). Agreement for anxiety improved to 0.74 (95% CI: 0.46, 1.00) with a revised Geriatric Anxiety Inventory-Short Form screening threshold of ≥4. Given the modest sample size, the confidence intervals for some analyses are wide. For sensitivity and specificity data, see Supplemental Table 3.
Table 3.
Summary of Agreement Between the GOLD and Geriatrician Evaluations with Original and Updated GOLD Instrument Screening Thresholds
| GOLD Instrument | Original Screening Threshold | Updated Screening Threshold | ||
|---|---|---|---|---|
| Problem Present | κ (95% CI) | Problem Present | κ (95% CI) | |
| Depression | ||||
| Patient Health Questionnaire-9 | ≥10 | 0.50 (0.19, 0.82) | ≥5 | 0.50 (0.25, 0.75) |
| Anxiety | ||||
| Geriatric Anxiety Inventory-Short Form | ≥3 | 0.61 (0.30, 0.92) | ≥4 | 0.74 (0.46, 1.00) |
| Cognitive Impairmenta | ||||
| Cognitive Change Index Participant or Caregiver Score | ≥20 | −0.06 (−0.31, 0.19) | ≥30 | 0.18 (0.05, 0.30) |
| Functional Impairmenta | ||||
| Instrumental ADLs or | ≤11 | 0.57 (0.36, 0.78) | ≤11 | 0.65 (0.45, 0.85) |
| Lower extremity physical ADLs or | ≥22 | – | ||
| Upper extremity physical ADLs | ≥8 | ≥2 | ||
| Fall Riska | ||||
| Stay Independent checklist or | ≥4 | 0.38 (0.08, 0.68) | ≥1 | 0.42 (0.11, 0.74) |
| Single-item question: fall in last year | Yes | Yes | ||
| Limited Social Support | ||||
| Modified Medical Outcomes Tangible support or | ≤11 | 0.19 (−0.09, 0.47) | ≤4 | 0.65 (0.19, 1.00) |
| Emotional support | ≤11 | – | ||
| Low Nutritional Status | ||||
| Simplified Nutritional Appetite Questionnaire | ≤14 | 0.56 (0.33, 0.79) | ≤13 | 0.64 (0.43, 0.85) |
Notes: aA participant was deemed to have the “problem present” if ≥1 of the GOLD Instrument scores met the corresponding screening threshold.
Abbreviation: ADLs, activities of daily living.
Although target agreement was not reached for depression, the updated screening thresholds for the Patient Health Questionnaire-9 yielded high sensitivity and specificity, suggesting reasonable performance for depression screening. No screening thresholds for fall risk or the Cognitive Change Index produced a κ >0.60, and both showed limited screening performance.
Figure 2 shows the number and percent of participants identified by the GOLD and the geriatrician evaluation for each geriatric problem, along with the corresponding κ statistic, using the updated screening thresholds. Fall risk was the most frequently identified problem, identified in 78% by the geriatrician and 90% of participants by the GOLD. Other problems commonly identified by both methods include functional impairment (58% by geriatrician and 44% by GOLD) and low nutritional status (50% by geriatrician and 40% by GOLD). In contrast, cognitive impairment was identified in 72% by the geriatrician but only 20% by the GOLD. Anxiety was identified in 16% by the geriatrician and in 10% by the GOLD, while depression was identified in 18% by the geriatrician and in 34% by the GOLD. Limited social support was identified by 6% by both the geriatrician and the GOLD. Cross tabulations of this data are shown in Supplemental Tables 4–10.
Figure 2.
Prevalence of geriatric problems across GOLD and geriatrician evaluations with corresponding agreement (κ statistics). This figure shows that the geriatric problem prevalence as identified by a GOLD instrument (blue bars) or the geriatrician evaluation (red bars). Problems with the lowest agreement between approaches had lower κ values (cognitive impairment, depression, and fall risk).
Analyzing Level of Agreement by Presence of Cognitive Impairment
Based on Mini MoCA scores, 15 (30%) of the 50 participants showed evidence of severe cognitive impairment. Results of the subgroup analyses exploring whether agreement varied by presence of severe cognitive impairment can be found in Table 4. Agreement between the geriatrician and GOLD was higher among participants without cognitive impairment than those with cognitive impairment across most geriatric problems, though confidence intervals were overlapping.
Table 4.
Agreement Between the GOLD and Geriatrician Evaluations by Cognitive Impairment Status
| Geriatric Problem | κ (95% CI) | |
|---|---|---|
| Severe Cognitive Impairment | No Cognitive Impairment | |
| Anxiety | 1 | 0.72 (0.43, 1.00) |
| Cognitive Impairment | 0.07 (−0.07, 0.21) | 0.18 (0.02, 0.34) |
| Depression | 0.24 (−0.07, 0.54) | 0.68 (0.39, 0.97) |
| Fall Risk | 0.63 (−0.01, 1.00) | 0.36 (0.01, 0.71) |
| Functional Impairment | 0.59 (0.21, 0.97) | 0.66 (0.42, 0.90) |
| Limited Social Support | 0.63 (−0.01, 1.00) | 0.65 (0.03, 1.00) |
| Low Nutritional Status | 0.32 (−0.15, 0.80) | 0.77 (0.57, 0.98) |
Feasibility and Acceptability
Most participants (88%) completed the GOLD without a caregiver present. However, 16 (32%) participants without a caregiver present for the GOLD needed study staff assistance with GOLD completion. Study staff assistance was needed for a variety of reasons, most commonly visual (n=6) or hand impairment (n=6), followed by lack of familiarity with technology (n=3), language barrier (n=2), and/or cognitive impairment (n=1). Acceptability was high among participants, with 82% favorably rating ease of GOLD completion (scores of 4 or 5 on a Likert 1–5 scale) and 91% reporting the length of time for GOLD completion as “just right”. Most (72%) completed the geriatrician evaluation during dialysis. The median days between GOLD completion and geriatrician evaluation was 22 (IQR 12, 32) days.
Discussion
In this study of older adults receiving hemodialysis, we investigated agreement between the GOLD, a battery of self-reported screening instruments, and a geriatrician evaluation across several common geriatric problems and evaluated the feasibility of incorporating self-administered geriatric screening into dialysis care. The GOLD demonstrated substantial agreement with the geriatrician evaluation for anxiety, functional impairment, social support, and nutrition. However, the GOLD may be refined to improve its performance for screening for falls, depression, and cognitive impairment. Also, alternative approaches for screening geriatric problems may need to be considered for patients with severe cognitive impairment. Further, the GOLD was feasible to implement and acceptable to most older adults receiving dialysis. While there is room for improvement, this study suggests the potential for a scalable approach to the geriatric screening in hemodialysis, which is the first necessary step toward managing these problems in this seriously ill population.
Similar to other studies in the dialysis population,1–4 our population had a high prevalence of geriatric problems, including cognitive impairment, functional impairment, fall risk, and depression. Prior studies have conducted geriatric assessments for older patients on dialysis,11 for older patients before or right after starting dialysis,37 for possible transplant candidates,38 and for pre-dialysis older patients,39 and there is growing recognition of the importance of incorporating geriatric assessment and improving “nephrogeriatric care”.40 However, to our knowledge, our study is the first to assess the agreement between a battery of self-administered screening instruments and a geriatrician evaluation in the older dialysis population. Our exploratory analyses suggested that alternative screening thresholds were associated with higher agreement for several domains. Because of the small sample size, these thresholds should be considered hypothesis-generating and require evaluation in larger cohorts. Still, these findings suggest that dialysis populations may require different screening thresholds than those developed in other patient populations.
Our findings highlight an important opportunity to make geriatric assessment more accessible for the dialysis population. Currently, the dialysis multidisciplinary team has limited time and expertise to identify geriatric problems.7 Furthermore, geriatrician access is limited, especially in rural communities.41 However, the observed agreement between the GOLD and geriatrician evaluation suggests that reliable, clinically relevant information can be gathered directly from patients and/or caregivers. Incorporating a brief, self-administered screening battery into dialysis workflows could allow teams to flag potential geriatric problems. Identifying these geriatric problems could prompt clinicians to initiate management strategies such as referrals or social services. Our ongoing work, a dialysis geriatric care model, involves understanding how the GOLD should be integrated into clinical practice to facilitate geriatric problem management by the dialysis team.
The lack of agreement for the problems of cognitive impairment, depression, and fall risk likely occurred for several reasons. First, there were differences in measurement, with dissimilar instruments and questions used by the geriatrician and GOLD for screening for fall risk and cognitive impairment. For example, for identifying cognitive impairment, participants only self-rated their cognition, while the geriatrician evaluation included both self-report and a validated cognitive test. Self-report can be less reliable in people with significant cognitive impairment, as a lack of insight into their difficulties is often a feature of dementia.42,43 Future work in improving the identification of cognitive impairment could involve testing new instruments, such as the self-administered MoCA Xpresso,44 in this population. In addition, participants’ written self-reports may have differed from verbal reports to the geriatrician, with underreporting due to social desirability bias.45 Indeed, several studies have found that patient scores on depression scales are higher when the scale is completed by the patient alone vs when it is administered by an interviewer or a clinician.46–48 Finally, the geriatrician evaluation allowed for a more nuanced discussion and assessment of geriatric problems, which could contribute to discordant findings. Discordant results also have clinical implications. False-negative screening results may lead to missed opportunities to identify and address important geriatric problems, whereas false-positive results may result in additional evaluation or patient burden. However, given that the goal of screening is to identify individuals who may benefit from further evaluation rather than establish diagnoses, some false positives may be acceptable. While the screening instruments for these geriatric problems cannot replace evaluation by a healthcare professional, they may aid in identifying people who need further evaluation by a professional. For example, identifying fall risk may lead to a referral for an evaluation by a physical therapist.44
When we examined agreement by the presence of severe cognitive impairment, we found that agreement was low for several problems – depression, functional impairment, cognitive impairment, and low nutritional status – in participants with severe cognitive impairment. Importantly, this finding suggests that the GOLD may perform less well in detecting potential geriatric problems as well in this segment of the dialysis population, among whom likely have the highest need for identification (and subsequent management) of additional geriatric problems. Because sufficient executive function optimizes the validity of the GOLD, future work may focus on evaluating the agreement between care partner impressions (as a proxy for completion of the GOLD) and geriatrician evaluation.
The successful use of technology during dialysis is similar to other studies that have implemented data collection and/or delivered interventions through tablets.49,50 However, a sizeable minority of participants in this study needed some assistance; while it is unlikely that all dialysis participants could have completed the GOLD independently given the extent of cognitive, motor, and/or visual impairments, feasibility may have been further enhanced with a tutorial on use of the technology, larger tablet font, or an improved tablet setup. Overall, the findings suggest that the GOLD could be implemented in routine clinical practice, with most older adults being able to complete the GOLD independently.
Strengths of our study include the use of videoconferencing to accommodate dialysis schedules and enable geriatric evaluation within routine dialysis care, the use of a standardized evaluation protocol conducted by a geriatrician, and the inclusion of a racially diverse cohort with a high proportion of Black participants, a group disproportionately affected by kidney failure. However, there are several limitations. The geriatrician telehealth evaluation served as a pragmatic comparator rather than a diagnostic gold standard. The evaluation relied on a single geriatrician and inter-rater reliability was not assessed, and it did not include in-person physical performance testing. As such, the results should be interpreted as agreement between two assessment approaches rather than validation of the GOLD battery against a comprehensive geriatric assessment. In addition, the median interval of 22 days between GOLD completion and the geriatrician evaluation may have influenced agreement for several problems, such as depression and anxiety, that can fluctuate substantially over short periods of time. Additional limitations include the exclusion of nine participants who did not complete either the GOLD or the geriatrician evaluation, which may have introduced selection bias. Participants lived in or near only two cities in the US, so the results may not be generalizable to dialysis populations in other parts of the US or globally. Participants completed the GOLD during a dialysis session, when cognitive performance may be impaired and environmental distractions are common.51 Additionally, while we identified agreement between the GOLD and the geriatrician evaluation with respect to problem identification, we were not able to determine if the GOLD could differentiate between problems that were already being addressed and those that were unaddressed and required management; we prioritized sensitivity and erred on the side of over-identifying rather than under-identifying geriatric problems. The modest sample size limited the precision of the agreement estimates across geriatric domains. Additionally, the revised screening thresholds were developed within the same cohort and therefore may overestimate agreement because of overfitting. Therefore, the thresholds derived in this study should be considered hypothesis-generating, with need for confirmation in larger cohorts, but they may provide a practical starting point for implementing geriatric screening in dialysis populations. Finally, as the goal of this study was to create a decision rule, we used Cohen’s κ to assess binary agreement between the GOLD and geriatrician evaluations, foregoing a weighted κ with multiple categories; future studies could collect data to explore more nuanced concordance metrics.
Conclusions
In summary, our findings showed that the GOLD showed agreement with a telehealth geriatrician evaluation for many important geriatric problems and was easily implemented. While these results support the potential utility of the GOLD as a pragmatic screening approach, larger studies, including more diverse cohorts, are needed to confirm these findings and define its role in routine dialysis care. Our findings have informed the next step in our work: testing a care model that both screens for and addresses geriatric problems in the older dialysis population. Given the high prevalence of geriatric problems in this population, further investigation to refine an effective, scalable approach for identifying and addressing geriatric problems is essential.
Acknowledgments
The authors wish to acknowledge dialysis clinic staff who provided invaluable support. Part of this work was presented as a poster at the American Geriatrics Society conference in 2025.
Funding Statement
This study is supported by the National Institute on Aging Grant Number RF1AG076918. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Abbreviations
ADL, activities of daily living; CCI, Cognitive Change Index; CGA, Comprehensive Geriatric Assessment; CI, confidence interval; GAD, Generalized Anxiety Disorder; GAI-SF, Geriatric Anxiety Inventory – Short Form; GOLD, Geriatric Screen for Older Dialysis Patients; IQR, interquartile range; mMOS-SS, modified Medical Outcomes Study Social Support; MOCA, Montreal Cognitive Assessment; PHQ9, Patient Health Questionnaire-9; Q1, first quartile; Q3, third quartile; REDCap, Research Electronic Data Capture; SD, standard deviation; SNAQ, Simplified Nutritional Appetite Questionnaire.
Data Sharing Statement
The datasets used and/or analyzed during the current study are not publicly available due to participant privacy and institutional restrictions but are available from the corresponding author on reasonable request. It would be in a de-identified format upon regulatory approval.
Ethics Approval and Informed Consent
This study was approved by the Duke University IRB under Protocol Number Pro00112850. Consent for publication: Not applicable.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
Dr Cathleen Colón-Emeric reports royalties or licenses from Wolters Kluwer UptoDate in Medicine, outside the submitted work. Dr Sarah Peskoe reports Consulting fees from PCORI, JAMA Network Open, Spiral Med Dev, outside the submitted work. The other authors declare they have no competing interests.
References
- 1.Murray AM, Tupper DE, Knopman DS, et al. Cognitive impairment in hemodialysis patients is common. Neurology. 2006;67(2):216–12. doi: 10.1212/01.wnl.0000225182.15532.40 [DOI] [PubMed] [Google Scholar]
- 2.Cook WL, Tomlinson G, Donaldson M, et al. Falls and fall-related injuries in older dialysis patients. Clin J Am Soc Nephrol. 2006;1(6):1197–1204. [DOI] [PubMed] [Google Scholar]
- 3.McAdams-Demarco MA, Law A, Garonzik-Wang JM, et al. Activity of daily living disability and dialysis mortality: better prediction using metrics of aging. J Am Geriatr Soc. 2012;60(10):1981–1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Alencar SBV, de Lima FM, Dias LDA, et al. Depression and quality of life in older adults on hemodialysis. Braz J Psychiatry. 2020;42(2):195–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hall RK, Toles M, Massing M, et al. Utilization of acute care among patients with ESRD discharged home from skilled nursing facilities. Clin J Am Soc Nephrol. 2015;10(3):428–434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tamura MK, Tan JC, O’hare AM. Optimizing renal replacement therapy in older adults: a framework for making individualized decisions. Kidney Int. 2012;82(3):261–269. doi: 10.1038/ki.2011.384 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hall R, Rutledge J, Colon-Emeric C, Fish LJ. Unmet needs of older adults receiving in-center hemodialysis: a qualitative needs assessment. Kidney Med. 2020;2(5):543–51 e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Anand S, Johansen KL, Kurella Tamura M. Aging and chronic kidney disease: the impact on physical function and cognition. J Gerontol Biol Sci Med Sci. 2014;69(3):315–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.United States Renal Data System. 2024 USRDS Annual Data Report: Epidemiology of Kidney Disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2024. [Google Scholar]
- 10.Pilotto A, Cella A, Pilotto A, et al. Three decades of comprehensive geriatric assessment: evidence coming from different healthcare settings and specific clinical conditions. J Am Med Dir Assoc. 2017;18(2):.e192.1–e.11. [DOI] [PubMed] [Google Scholar]
- 11.Parlevliet JL, Buurman BM, Pannekeet MM, et al. Systematic comprehensive geriatric assessment in elderly patients on chronic dialysis: a cross-sectional comparative and feasibility study. BMC Nephrol. 2012;13:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kikuchi R, Broadwater G, Shelby R, et al. Detecting geriatric needs in older patients with breast cancer through use of a brief geriatric screening tool. J Geriatr Oncol. 2019;10(6):968–972. [DOI] [PubMed] [Google Scholar]
- 13.McCleary NJ, Wigler D, Berry D, et al. Feasibility of computer-based self-administered cancer-specific geriatric assessment in older patients with gastrointestinal malignancy. Oncologist. 2013;18(1):64–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wildes TM, Tuchman SA, Klepin HD, et al. Geriatric assessment in older adults with multiple myeloma. J Am Geriatr Soc. 2019;67(5):987–991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Moser A, Stuck AE, Silliman RA, Ganz PA, Clough-Gorr KM. The eight-item modified medical outcomes study social support survey: psychometric evaluation showed excellent performance. J Clin Epidemiol. 2012;65(10):1107–1116. doi: 10.1016/j.jclinepi.2012.04.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Rattanabannakit C, Risacher SL, Gao S, et al. The cognitive change Index as a measure of self and informant perception of cognitive decline: relation to neuropsychological tests. J Alzheimers Dis. 2016;51(4):1145–1155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Risacher SL, Tallman EF, West JD, et al. Olfactory identification in subjective cognitive decline and mild cognitive impairment: association with tau but not amyloid positron emission tomography. Alzheimers Dement. 2017;9:57–66. doi: 10.1016/j.dadm.2017.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med. 1991;32(6):705–714. doi: 10.1016/0277-9536(91)90150-B [DOI] [PubMed] [Google Scholar]
- 19.Wilson MM, Thomas DR, Rubenstein LZ, et al. Appetite assessment: simple appetite questionnaire predicts weight loss in community-dwelling adults and nursing home residents. Am J Clin Nutr. 2005;82(5):1074–1081. [DOI] [PubMed] [Google Scholar]
- 20.McCusker J, Bellavance F, Cardin S, Belzile É. Validity of an Activities of Daily Living Questionnaire among older patients in the emergency Department. J Clin Epidemiol. 1999;52(11):1023–1030. doi: 10.1016/S0895-4356(99)00084-0 [DOI] [PubMed] [Google Scholar]
- 21.Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary care evaluation of mental disorders. Patient Health Questionnaire. JAMA. 1999;282(18):1737–1744. [DOI] [PubMed] [Google Scholar]
- 22.Levis B, Benedetti A, Thombs BD; DEPRESsion Screening Data (DEPRESSD) Collaboration. Accuracy of Patient Health Questionnaire-9 (PHQ-9) for screening to detect major depression: individual participant data meta-analysis. BMJ. 2019;365:l1476. doi: 10.1136/bmj.l1476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rubenstein LZ, Vivrette R, Harker JO, Stevens JA, Kramer BJ. Validating an evidence-based, self-rated fall risk questionnaire (FRQ) for older adults. J Saf Res. 2011;42(6):493–499. doi: 10.1016/j.jsr.2011.08.006 [DOI] [PubMed] [Google Scholar]
- 24.Byrne GJ, Pachana NA. Development and validation of a short form of the geriatric Anxiety Inventory-the GAI-SF. Int Psychogeriatr. 2011;23(1):125–131. [DOI] [PubMed] [Google Scholar]
- 25.Johnco C, Knight A, Tadic D, Wuthrich VM. Psychometric properties of the Geriatric Anxiety Inventory (GAI) and its short-form (GAI-SF) in a clinical and non-clinical sample of older adults. Int Psychogeriatr. 2015;27(7):1089–1097. doi: 10.1017/S1041610214001586 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Magaziner J, Hawkes W, Hebel JR, et al. Recovery from hip fracture in eight areas of function. J Gerontol Biol Sci Med Sci. 2000;55(9):M498–507. [DOI] [PubMed] [Google Scholar]
- 27.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. doi: 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Notomi S, Kitamura M, Yamaguchi K, et al. Anorexia assessed by simplified nutritional appetite questionnaire and association with medication in older patients undergoing hemodialysis. Biol Pharm Bull. 2023;46(2):286–291. [DOI] [PubMed] [Google Scholar]
- 29.Fillenbaum GG. Multidimensional Functional Assessment of Older Adults: The Duke Older Americans Resources and Services Procedures. Psychology Press; 2013. [Google Scholar]
- 30.Friedman AN, Kim J, Kaiser S, Pedersen TL, Newman JW, Watkins BA. Association between plasma endocannabinoids and appetite in hemodialysis patients: a pilot study. Nutr Res. 2016;36(7):658–662. doi: 10.1016/j.nutres.2016.03.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Centers for Disease Control and Prevention. Stay Independent. 2023. [updated 2023]. Available from: https://www.cdc.gov/steadi/pdf/STEADI-Brochure-StayIndependent-508.pdf. Accessed October 1, 2026.
- 32.Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092–1097. doi: 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
- 33.Wittich W, Phillips N, Nasreddine ZS, Chertkow H. Sensitivity and specificity of the montreal cognitive assessment modified for individuals who are visually impaired. J Vis Impair Blind. 2010;104(6):360–368. doi: 10.1177/0145482X1010400606 [DOI] [Google Scholar]
- 34.Katz JN, Chang LC, Sangha O, Fossel AH, Bates DW. Can comorbidity be measured by questionnaire rather than medical record review? Med Care. 1996;34(1):73–84. doi: 10.1097/00005650-199601000-00006 [DOI] [PubMed] [Google Scholar]
- 35.Mai LM, Oczkowski W, Mackenzie G, et al. Screening for cognitive impairment in a stroke prevention clinic using the MoCA. Can J Neurol Sci. 2013;40(2):192–197. [DOI] [PubMed] [Google Scholar]
- 36.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–174. doi: 10.2307/2529310 [DOI] [PubMed] [Google Scholar]
- 37.Goto NA, van Loon IN, Morpey MI, et al. Geriatric assessment in elderly patients with end-stage kidney disease. Nephron. 2019;141(1):41–48. doi: 10.1159/000494222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Hall RK, McAdams-DeMarco MA. Breaking the cycle of functional decline in older dialysis patients. Semin Dial. 2018;31(5):462–467. doi: 10.1111/sdi.12695 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Hall RK, Haines C, Gorbatkin SM, et al. Incorporating geriatric assessment into a nephrology clinic: preliminary data from two models of care. J Am Geriatr Soc. 2016;64(10):2152–2154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Litjens EJR, Dani M, Verberne WR, Van Den Noortgate NJ, Joosten HMH, Brys ADH. Geriatric assessment in older patients with advanced Kidney disease: a key to personalized care and shared decision-making-a narrative review. J Clin Med. 2025;14(5):1749. doi: 10.3390/jcm14051749 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Health USDo, Human Services HR, Services Administration NCfHWA. National and regional projections of supply and demand for geriatricians: 2013–2025. Rockville: US Department of Health and Human Services; 2017. [Google Scholar]
- 42.Buelow MT, Tremont G, Frakey LL, Grace J, Ott BR. Utility of the cognitive difficulties scale and association with objective test performance. Am J Alzheimers Dis Other Demen. 2014;29(8):755–761. doi: 10.1177/1533317514539032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Tremont G, Alosco ML. Relationship between cognition and awareness of deficit in mild cognitive impairment. Int J Geriatr Psychiatry. 2011;26(3):299–306. doi: 10.1002/gps.2529 [DOI] [PubMed] [Google Scholar]
- 44.Klil-Drori S, Bodenstein KC, Sun S, et al. Montreal cognitive assessment (MoCA) XpressO: validation of a digital self-administered cognitive prescreening tool. J Am Geriatr Soc. 2024;72(8):2516–2522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Rickwood DJ, Coleman-Rose CL. The effect of survey administration mode on youth mental health measures: social desirability bias and sensitive questions. Heliyon. 2023;9(9):e20131. doi: 10.1016/j.heliyon.2023.e20131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Lysandrou M, Quaye N, Landes K, et al. Evaluating self- vs interviewer-administered screening for depression in sickle cell disease. Blood Adv. 2024;8(3):699–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Batra EK, Tuan WJ, Sekhar D, Merai R, Shi T, Fogel BN. Comparison of adolescent depression screening using orally administered versus written self-report scores. J Prim Care Community Health. 2025;16:21501319251374583. doi: 10.1177/21501319251374583 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Cannon BJ, Thaler T, Roos S. Oral versus written administration of the geriatric depression scale. Aging Ment Health. 2002;6(4):418–422. doi: 10.1080/1360786021000022930 [DOI] [PubMed] [Google Scholar]
- 49.Guerraoui A, Prezelin-Reydit M, Kolko A, et al. Patient-reported outcome measures in hemodialysis patients: results of the first multicenter cross-sectional ePROMs study in France. BMC Nephrol. 2021;22(1):357. doi: 10.1186/s12882-021-02551-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Dember LM, Hsu JY, Mehrotra R, et al. Pain coping skills training for patients receiving hemodialysis: the HOPE consortium randomized clinical trial. JAMA Intern Med. 2025;185(2):197–207. doi: 10.1001/jamainternmed.2024.7140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Murray AM, Pederson SL, Tupper DE, et al. Acute variation in cognitive function in hemodialysis patients: a cohort study with repeated measures. Am J Kidney Dis. 2007;50(2):270–278. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analyzed during the current study are not publicly available due to participant privacy and institutional restrictions but are available from the corresponding author on reasonable request. It would be in a de-identified format upon regulatory approval.