Associations between depressive symptoms and clinical factors in hemodialysis patients may differ by age. This study investigated age-related differences in factors linked to depressive symptoms. We analyzed 783 hemodialysis outpatients (582 aged ≥65; 201 <65). Depressive symptoms were defined as a Geriatric Depression Scale-5 score ≥2. Clinical data included CRP, grip strength, Short Physical Performance Battery (SPPB), Kt/V, Geriatric Nutritional Risk Index (GNRI), Intact-PTH, and self-reported stress. Prevalence was compared by Chi-square test. Associations in each age group were assessed by logistic regression. Depressive symptoms were less common in elderly (45.5%) than non-elderly patients (59.7%). Stress was significantly associated with depressive symptoms in non-elderly patients [OR: 2.42 (1.25–4.67)]. Low SPPB [OR: 0.92 (0.87–0.97)] and stress [OR: 2.01 (1.39–2.89)] were significant in elderly patients. Stress management was an important issue for both elderly and non-elderly hemodialysis patients. In particular, it was suggested that the importance of exercise therapy for reducing depressive symptoms may differ between elderly and non-elderly hemodialysis patients.
Depressive symptoms in hemodialysis (HD) patients represent a critical therapeutic target. Previous research has indicated a high prevalence of depression among HD patients [1], with HD itself identified as an independent risk factor for depressive symptoms when compared to patients with other chronic illnesses [2]. The progression of chronic kidney disease and the comorbidities associated with end-stage renal disease are posited to significantly influence the prevalence of depressive symptoms in this population. Moreover, depressive symptoms in HD patients have been shown to predict reduced survival rates [3], with recent depressive episodes specifically linked to an increased mortality risk [4]. These findings underscore the importance of early intervention in addressing depressive symptoms in HD patients.
Interventions for depression in HD patients should be tailored to address specific contributing factors. While psychological, social, lifestyle, and environmental factors are commonly associated with depressive symptoms in non-HD populations [5–7], HD patients present with unique challenges. These include elevated levels of inflammatory cytokines [8], accumulation of uremic toxins [9], abnormalities in parathyroid hormone levels [10], chronic systemic inflammation [11], impaired physical function [12], and malnutrition [11]. Such factors are often consequences of kidney disease or the dialysis process and may contribute significantly to the onset of depressive symptoms. Despite these associations, the precise mechanisms linking these factors to depressive symptoms in HD patients remain unclear, highlighting the need to identify modifiable factors and develop targeted interventions.
The factors associated with depressive symptoms in HD patients are likely to vary by age. Previous research has indicated that age influences the factors linked to depressive symptoms, with younger patients showing associations with high body mass index (BMI), smoking, and obesity [13]. In contrast, older patients are more likely to experience depressive symptoms in relation to chronic diseases, sleep disturbances, motor dysfunction, dependence on daily activities, and visual impairments [14]. Physical function decline, particularly in elderly patients, is a common issue and has been found to be significantly associated with depressive symptoms [15]. It is plausible that the factors requiring intervention for depressive symptoms differ between elderly and non-elderly HD patients, although these factors have not been comprehensively identified.
This study aims to investigate the factors associated with depressive symptoms in elderly and non-elderly HD patients and to elucidate the generational differences in these contributing factors.
This study employed a multicenter, cross-sectional, exploratory design. The study population consisted of outpatient HD patients from 21 facilities within the Kaikoukai Group who participated in an intradialytic exercise therapy program between April 2018 and April 2022. Data from the intradialytic exercise therapy program were recorded in a clinical database, and baseline measurements were retrospectively extracted for secondary analysis. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethical Committee of Nagoya Kyoritsu Hospital (K145-01). As the data were collected as part of routine clinical practice and used retrospectively, the requirement for written informed consent was waived. Instead, the patients’ right to decline participation was ensured through an opt-out approach.
The inclusion criteria were: (1) patients who voluntarily agreed to participate in the intradialytic exercise therapy program and (2) patients with sufficient cognitive function to complete the questionnaire. Exclusion criteria included: (1) patients who could not undergo physical function assessment, (2) patients who declined evaluation, and (3) patients for whom depressive symptoms could not be assessed due to the absence of Geriatric Depression Scale-5 (GDS-5) measurements.
Data Collection
GDS-5 was used to assess depressive symptoms [16], and patient characteristics, grip strength, and the Short Physical Performance Battery (SPPB) were measured as physical functions.
The GDS-5 is a brief five-item scale for assessing depressive symptoms: 1) Are you basically satisfied with your life? 2) Do you often get bored? 3) Do you often feel helpless? 4) Do you prefer to stay at home, rather than going out and doing new things? 5) Do you feel pretty worthless the way you are now?
The scoring method for the GDS-5 is as follows: if the respondent answers no to the 1st item and yes to the 2nd and subsequent items, one point is added, for a total score of 5 points. This scale is a shortened version of the GDS-15 and has been reported to be useful in screening for depressive symptoms in patients 18 years of age and older. The GDS-5 has also been used as a screening tool for depressive symptoms in HD patients [17]. The predictive validity of the GDS-5 for depressive symptoms is best balanced at a cutoff value of 2, with a reported sensitivity of 0.98 (95% CI: 0.96–1.00) and specificity of 0.83 (95% CI: 0.79–0.87) [18]. In this study, patients with a score of 2 or higher on the GDS-5 were defined as having depressive symptoms.
The patient characteristics were age, gender, exercise habits, smoking habits, drinking habits, dry weight, BMI, presence of diabetes mellitus, dialysis vintage, hemodialysis induction disease (diabetic nephropathy, nephrosclerosis, chronic glomerulonephritis, other, unknown), and stress status and blood data were collected.
Stress status was assessed using a self-administered questionnaire in which participants answered “yes” or “no” to the question “Are you currently feeling stressed?”, regardless of the degree or nature of their stress. This questionnaire was used to assess the presence or absence of stress in the subjects at the time of the study.
Blood data included albumin (Alb), blood urea nitrogen (BUN), creatinine (Cre), total cholesterol (T-cho), phosphorus (P), calcium (Ca), hemoglobin (Hb), Intact-PTH, C-reactive protein (CRP), creatinine production rate (%CGR), normalized Protein catabolic rate (nPCR), and Kt/V. Geriatric Nutritional Risk Index (GNRI) was also collected. Intact-PTH was treated as an indicator of abnormal parathyroid function and CRP as an indicator of inflammation; Kt/V, an indicator of adequacy for HD, was collected using the method of Shinzato et al [19]. The GNRI is a tool used to assess the risk of nutritional disorders based on height, weight, and albumin levels [20]. In this study, it was used as an indicator of nutritional status. The GNRI was calculated using the following formula.
GNRI = {14.89×serum albumin (g/dl)}+{41.7× (DW/IBW)}
IBW (male) = height (cm)–100–[(height)–150]/4]
IBW (female) = height (cm)–100–[(height)–150]/2.5]
DW: dry weight, IBW: ideal body weight
Grip strength was used as an indicator of muscle strength [21], and was measured using a digital grip strength meter (Grip-D, Takei Kikikogyo Co., Ltd., Kamo) in the standing or sitting position with the elbow joint in extension. SPPB was used as an index of mobility [22], and the standing balance, 4-m walking speed, and chair standing tests were scored and evaluated on a 12-point scale.
Statistical Analysis
First, missing data were addressed using multiple imputation, generating 20 imputed datasets to minimize bias. Participants were categorized into two age groups: non-elderly (<65 years) and elderly (≥65 years). These groups were further stratified into four subgroups based on the presence or absence of depressive symptoms: non-elderly with depressive symptoms, non-elderly without depressive symptoms, elderly with depressive symptoms, and elderly without depressive symptoms. Comparisons of patient characteristics were conducted between the elderly and non-elderly groups, as well as within the subgroups: non-elderly with depressive symptoms versus non-elderly without depressive symptoms, and elderly with depressive symptoms versus elderly without depressive symptoms. Pearson’s χ² test was used for categorical variables, while continuous variables were analyzed using either the independent t-test or the Kruskal-Wallis test. Categorical variables were expressed as frequencies (%), and continuous variables as means with standard errors. The GDS-5 compared total scores, prevalence of depressive symptoms (total score of 2 or more), and sub-items between the elderly and non-elderly groups and subgroups.
Logistic regression analysis was performed to identify factors associated with depressive symptoms among all the participants, as well as within the elderly and non-elderly subgroups. Depressive symptoms were treated as the dependent variable, while grip strength, SPPB, Kt/V, GNRI, presence of stress, intact-PTH, and CRP were included as independent variables. A propensity score analysis was conducted as a sensitivity analysis to adjust for potential confounders. Propensity scores were calculated based on eight factors: age, sex, presence of diabetes, dialysis vintage, drinking habits, smoking habits, exercise habits, and the HD induction disease. Three models were analyzed: Model 1 included unadjusted data without propensity scores or adjustment for confounders; Model 2 incorporated propensity scores adjusted for sex, presence of diabetes, and dialysis vintage; and Model 3 included propensity scores calculated using sex, presence of diabetes, dialysis vintage, drinking habits, smoking habits, exercise habits, and dialysis-inducing disease as covariates. The threshold for statistical significance was set at a 5% risk level.
The subjects were 906 patients who received HD at 21 facilities and participated in an intradialytic exercise therapy program from April 2018 to December 2022; the final analysis included 783 patients (non-elderly with depressive symptoms group: n=120, non-elderly without depressive symptoms group: n=81, elderly with depressive symptoms group: n=265, elderly without depressive symptoms group: n=317) after excluding 123 patients who did not consent to the GDS-5 assessment (Figure 1).
Table 1 presents a comparison of patient backgrounds between the non-elderly and elderly groups. The elderly group had a significantly lower prevalence of smoking, dry weight, BMI, Alb, Cre, T-cho, P, GNRI, grip strength, and SPPB, while stress and Kt/v were significantly higher (Table 1).
| Overall subjects | Non-elderly group | Elderly group | ||
|---|---|---|---|---|
| n=783 | n=201 | n=582 | ||
| Age (years) | 71.17 (0.42) | 55.02 (0.51) | 76.73 (0.27) | <0.001 |
| Sex (male%) | 440 (56.2) | 112 (55.7) | 328 (56.4) | 0.934 |
| Exercise habit [Yes, n(%)] | 287 (37.0) | 70 (35.0) | 217 (37.7) | 0.498 |
| Smoking habits (yes, %) | 194 (26.4) | 61 (32.3) | 133 (24.4) | 0.044 |
| Drinking habits (yes, %) | 148 (20.3) | 38 (20.3) | 110 (20.3) | 1.000 |
| Dry weight (kg) | 55.36 (0.46) | 62.97 (1.13) | 52.74 (0.42) | <0.001 |
| BMI (kg/m2) | 21.9 (0.15) | 23.98 (0.38) | 21.18 (0.14) | <0.001 |
| Diabetes mellitus [Yes, n(%)] | 384 (49.0) | 107 (53.2) | 277 (47.6) | 0.190 |
| Hemodialysis duration (months) | 61.69 (3.29) | 58 (5.07) | 63.21 (4.07) | 0.487 |
| Hemodialysis Induction Disease | ||||
| Diabetic nephropathy (%) | 295 (37.7) | 90 (44.8) | 205 (35.2) | <0.001 |
| Nephrosclerosis (%) | 143 (18.3) | 19 (9.5) | 124 (21.3) | |
| Chronic glomerulonephritis (%) | 97 (12.4) | 19 (9.5) | 78 (13.4) | |
| Unknown (%) | 91 (11.6) | 25 (12.4) | 66 (11.3) | |
| Other (%) | 157 (20.1) | 48 (23.9) | 109 (18.7) | |
| Stress [Yes, n(%)] | 302 (41.5) | 104 (55.3) | 341 (63.3) | <0.001 |
| Laboratory data | ||||
| Alb (g/dl) | 3.5 (0.01) | 3.57 (0.03) | 3.47 (0.02) | 0.001 |
| BUN (mg/dl) | 57.41 (0.61) | 58.73 (1.11) | 56.96 (0.73) | 0.190 |
| Cre (mg/dl) | 8.91 (0.09) | 10.07 (0.2) | 8.51 (0.1) | <0.001 |
| T-cho (mg/dl) | 154.27 (1.39) | 161.52 (2.91) | 151.77 (1.56) | 0.002 |
| P (mg/dl) | 5.09 (0.05) | 5.31 (0.09) | 5.01 (0.05) | 0.005 |
| Ca (mg/dl) | 8.58 (0.11) | 8.56 (0.06) | 8.59 (0.15) | 0.904 |
| Hb (g/dl) | 11.00 (0.04) | 11.04 (0.08) | 10.98 (0.05) | 0.471 |
| INT-PTH (pg/dl) | 177.78 (6.22) | 191.42 (12.02) | 177.36 (7.36) | 0.078 |
| CRP (mg/dl) | 0.40 (0.03) | 0.34 (0.05) | 0.43 (0.04) | 0.243 |
| nPCR (g/kg/day) | 0.83 (0.01) | 0.83 (0.01) | 0.82 (0.01) | 0.614 |
| %CGR (%) | 87.63 (1) | 85.65 (1.93) | 88.41 (1.17) | 0.222 |
| Kt/v | 1.53 (0.01) | 1.48 (0.02) | 1.55 (0.01) | 0.005 |
| GNRI | 92.95 (0.31) | 96.66 (0.67) | 91.68 (0.33) | <0.001 |
| Physical function assessment | ||||
| Grip strength (kg) | 22.7 (0.29) | 26.7 (0.59) | 21.34 (0.32) | <0.001 |
| SPPB (points) | 9.01 (0.12) | 9.81 (0.22) | 8.73 (0.14) | <0.001 |
BMI: Body Mass Index, Alb: albumin, BUN: blood urea nitrogen, Cre: creatinine, T-cho: total cholesterol, P: phosphorus, Ca: calcium, Hb: hemoglobin, INT-PTH: intact-parathormone, CRP: C-reactive protein, nPCR: normalized protein catabolic rate, %CGR: % creatinine generation rate, GNRI: Geriatric Nutritional Risk Index, SPPB: Short Physical Performance Battery.
Table 2 presents the comparisons among the non-elderly with depressive symptoms, non-elderly without depressive symptoms, elderly with depressive symptoms, and elderly without depressive symptoms groups. Compared with the non-elderly without depressive symptoms group, the non-elderly with depressive symptoms group were younger, had a higher prevalence of diabetes and stress, and showed significantly lower %CGR and SPPB. In the elderly population, the group with depressive symptoms included more men and more patients with a history of smoking, had a higher prevalence of diabetes and CRP levels, and showed significantly lower Alb, dry weight, nPCR, and SPPB (Table 2).
| Non-elderly group | P-value | Elderly group | P-value | |||
|---|---|---|---|---|---|---|
| Without depressive symptoms | With depressive symptoms | Without depressive symptoms | With depressive symptoms | |||
| n=81 | n=120 | n=317 | n=265 | |||
| Age (years) | 56.94 (0.77) | 53.73 (0.66) | 0.002 | 76.75 (0.38) | 76.71 (0.39) | 0.946 |
| Prevalence of depressive symptoms (Yes, %) | ||||||
| All ages | 120 (100) | 265 (100) | ||||
| 30–39 years | – | 2 (1.7) | – | – | – | – |
| 40–49 years | – | 34 (28.3) | – | – | – | – |
| 50–59 years | – | 51 (42.5) | – | – | – | – |
| 60–69 years | – | 33 (27.5) | – | – | 36 (14.7) | – |
| 70–79 years | – | – | – | – | 142 (53.6) | – |
| 80–89 years | – | – | – | – | 79 (29.8) | – |
| Over 90 years | – | – | – | – | 5 (1.9) | – |
| Sex (male%) | 42 (51.9) | 70 (58.3) | 0.364 | 165 (52.1) | 163 (61.5) | 0.022 |
| Exercise habit [Yes, n(%)] | 32 (39.5) | 39 (32.5) | 0.308 | 128 (40.3) | 92 (34.7) | 0.154 |
| Smoking habits (yes, %) | 24 (29.1) | 42 (35.0) | 0.492 | 68 (21.5) | 76 (28.7) | 0.036 |
| Drinking habits (yes, %) | 19 (23.5) | 23 (18.9) | 0.385 | 59 (18.5) | 59 (22.2) | 0.251 |
| Dry weight (kg) | 60.65 (1.63) | 64.54 (1.53) | 0.082 | 51.79 (0.57) | 53.87 (0.62) | 0.014 |
| BMI (kg/m2) | 23.45 (0.52) | 24.33 (0.54) | 0.237 | 21.07 (0.2) | 21.32 (0.2) | 0.384 |
| Diabetes mellitus [Yes, n(%)] | 36 (44.4) | 71 (59.2) | 0.045 | 139 (43.8) | 138 (52.1) | 0.048 |
| Hemodialysis duration (months) | 65.04 (8.68) | 53.24 (6.14) | 0.267 | 64.51 (5.95) | 61.66 (5.43) | 0.724 |
| Hemodialysis Induction Disease | ||||||
| Diabetic nephropathy (%) | 27 (33.3) | 63 (52.5) | 0.074 | 103 (32.5) | 102 (38.5) | 0.48 |
| Nephrosclerosis (%) | 8 (9.9) | 11 (9.2) | 71 (22.4) | 53 (20.0) | ||
| Chronic glomerulonephritis (%) | 9 (11.1) | 10 (8.3) | 48 (15.1) | 30 (11.3) | ||
| Unknown (%) | 11 (13.6) | 14 (11.7) | 35 (11.0) | 31 (11.7) | ||
| Other (%) | 26 (32.1) | 22 (18.3) | 60 (18.9) | 49 (18.5) | ||
| Stress [Yes, n(%)] | 32 (39.5) | 79 (65.8) | <0.001 | 93 (29.4) | 121 (45.7) | < 0.001 |
| Laboratory data | ||||||
| Alb (g/dl) | 3.58 (0.04) | 3.56 (0.03) | 0.641 | 3.5 (0.02) | 3.43 (0.02) | 0.024 |
| BUN (mg/dl) | 60.85 (1.8) | 57.29 (1.41) | 0.119 | 57.51 (0.84) | 56.31 (1.25) | 0.427 |
| Cre (mg/dl) | 10.32 (0.32) | 9.9 (0.25) | 0.305 | 8.51 (0.13) | 8.51 (0.14) | 0.979 |
| T-cho(mg/dl) | 168.25 (4.11) | 157.06 (3.95) | 0.05 | 152.22 (2.05) | 151.24 (2.41) | 0.722 |
| P (mg/dl) | 5.36 (0.14) | 5.27 (0.13) | 0.637 | 5.05 (0.07) | 4.96 (0.08) | 0.368 |
| Ca (mg/dl) | 8.6 (0.09) | 8.53 (0.07) | 0.517 | 8.69 (0.28) | 8.46 (0.05) | 0.412 |
| Hb (g/dl) | 11.19 (0.12) | 10.94 (0.11) | 0.132 | 10.93 (0.06) | 11.04 (0.06) | 0.209 |
| INT-PTH (pg/dl) | 175.64 (17.05) | 202.07 (16.57) | 0.27 | 181.69 (9.01) | 172.17 (10.93) | 0.477 |
| CRP (mg/dl) | 0.34 (0.07) | 0.35 (0.06) | 0.889 | 0.33 (0.04) | 0.54 (0.07) | 0.007 |
| nPCR (g/kg/day) | 0.87 (0.03) | 0.81 (0.02) | 0.055 | 0.84 (0.01) | 0.81 (0.01) | 0.026 |
| %CGR(%) | 91.29 (3.23) | 81.85 (2.33) | 0.018 | 89.06 (1.61) | 87.64 (1.71) | 0.586 |
| Kt/v | 1.54 (0.04) | 1.44 (0.03) | 0.062 | 1.56 (0.02) | 1.54 (0.02) | 0.25 |
| GNRI | 96.56 (1.05) | 96.73 (0.87) | 0.934 | 91.79 (0.48) | 91.55 (0.46) | 0.707 |
| Physical function assessment | ||||||
| Grip strength (kg) | 27.74 (0.92) | 26 (0.75) | 0.145 | 21.47 (0.44) | 21.19 (0.47) | 0.654 |
| SPPB (points) | 10.49 (0.32) | 9.34 (0.3) | 0.008 | 9.3 (0.17) | 8.05 (0.24) | <0.001 |
BMI: Body Mass Index, Alb: albumin, BUN: blood urea nitrogen, Cre: creatinine, T-cho: total cholesterol, P: phosphorus, Ca: calcium, Hb: hemoglobin, INT-PTH: intact-parathormone, CRP: C-reactive protein, nPCR: normalized protein catabolic rate, %CGR: % creatinine generation rate, GNRI: Geriatric Nutritional Risk Index, SPPB: Short Physical Performance Battery.
Depressive symptoms were present in 385 (49.2%) of the patients. The total GDS-5 score was significantly higher in the non-elderly group [non-elderly group: 2.10 (1.55) vs. the elderly group: 1.62 (1.37) points. The prevalence of depressive symptoms was higher in the non-elderly group [non-elderly group: n=120 (59.7%), elderly group: n=265 (45.5%), P<0.001]. The non-elderly group showed a significantly higher percentage of “yes” responses to the GDS-5 item “Are you basically satisfied with your life?” (Non-elderly group: 75.0% vs. elderly group: 49.4%) and “Do you often feel helpless?” (Non-elderly group: 59.2% vs. elderly group: 45.2%). Significant differences were not found in the other sub-items (Table 3). In addition, subgroup analysis showed that the non-elderly with depressive symptoms group was significantly more likely than the elderly group with depressive symptoms to answer “Yes” to the question “Are you basically satisfied with your life? (the elderly with depressive symptoms group: 49.4% vs. the non-elderly with depressive symptoms group: 75%, P<0.001) (Table 3).
|
Over all subject n=783 |
Non-elderly group | elderly group |
Difference between non-elderly and elderly groups P-value |
Difference between non-elderly group with depressive symptoms and elderly groups with depressive symptoms P-value |
|||||
|---|---|---|---|---|---|---|---|---|---|
| all | Without depressive symptoms | With depressive symptoms | all | Without depressive symptoms | With depressive symptoms | ||||
| n=201 | n=81 | n=120 | n=582 | n=317 | n=265 | ||||
| GDS-5 (Points) |
1.75 (1.44) |
2.10 (1.55) |
0.53 (0.06) |
3.17 (0.09) |
1.62 (1.37) |
0.56 (0.03) |
2.9 (0.06) |
<0.001 | 0.011 |
| Prevalence of depression symptoms |
385 (49.2) |
120 (59.7) |
– | – |
265 (45.5) |
– | – | <0.001 | – |
| Are you basically satisfied with your life? [No, n (%)] |
247 (31.6) |
95 (47.3) |
5 (6.2) |
90 (75.0) |
152 (26.2) |
21 (6.6) |
131 (49.4) |
<0.001 | <0.001 |
| Do you often get bored? [Yes, n (%)] |
231 (29.6) |
69 (34.3) |
1 (1.2) |
68 (56.7) |
162 (27.9) |
24 (7.6) |
138 (52.0) |
0.089 | 0.441 |
| Do you often feel helpless? [Yes, n (%)] |
382 (48.8) |
119 (59.2) |
16 (19.8) |
103 (83.3) |
263 (45.2) |
46 (14.5) |
217 (81.9) |
0.001 | 0.380 |
| Do you prefer to stay at home, rather than going out and doing new things? [Yes, n (%)] |
306 (39.1) |
84 (41.8) |
18 (22.2) |
66 (55.0) |
222 (38.1) |
73 (23.0) |
149 (56.2) |
0.402 | 0.826 |
| Do you feel pretty worthless the way you are now? [Yes, n (%)] |
202 (25.8) |
56 (27.9) |
3 (3.7) |
53 (44.1) |
146 (25.1) |
13 (4.0) |
133 (50.2) |
0.455 | 0.272 |
GDS-5: geriatric depression scale 5.
Figure 2 shows the models that examined the factors associated with depressive symptoms in the overall, non-elderly, and elderly groups, respectively. In the all-subjects group, SPPB [Model 1, OR: 0.91(0.86 to 0.95); Model 2, 0.92(0.87 to 0.97)], presence of stress [Model 1, OR: 2.18(1.61 to 2.96); Model 2, OR: 2.12(1.56 to 2.89)], and CRP [Model 1, OR: 1.24(1.00 to 1.52); Model 2, OR: 1.23(1.00 to 1.52)] showed significant associations; in Model 3, SPPB [OR: 0.92(0.88 to 0.97)], and stress status [2.14 (1.56 to 2.91)] were significantly associated (P<0.05). In the non-elderly group, Kt/V [OR: 0.31 (0.10 to 0.93)] and presence of stress [OR: 2.79 (1.48 to 5.24)] were significantly associated in Model 1. However, in Models 2 and 3, adjusted for confounders, only the presence of stress was significantly associated [Model 2, OR: 2.62 (1.37 to 5.02); Model 3, OR: 2.42 (1.25 to 4.67)] (P<0.05). In contrast, in the elderly group, the SPPB [Model 1, OR: 0.90 (0.85 to 0.95); Model 2, OR: 0.91 (0.86 to 0.97); Model 3, OR: 0.92(0.87 to 0.97)] and presence of stress in all models [Model 1, OR: 1.85(1.29 to 2.64); Model 2, OR: 1.97(1.36 to 2.84); Model 3, OR: 2.01(1.39 to 2.89)] were significantly related factors (Figure 2).
This study is the first to investigate factors associated with depressive symptoms in non-elderly and elderly HD patients. The findings revealed that depressive symptoms were significantly associated with the presence of stress and SPPB scores across all participants. Furthermore, low SPPB scores were specifically linked to depressive symptoms in elderly HD patients, but not in non-elderly patients. This finding suggests that the factors contributing to depressive symptoms may vary by age group. These results provide crucial evidence supporting the need for age-specific and individualized interventions to address depressive symptoms in HD patients.
Depressive symptoms in patients undergoing HD constitute a significant clinical concern, necessitating further evidence to establish targeted intervention strategies. In the present study, 385 patients (49.2%) exhibited depressive symptoms, with a significantly higher prevalence observed in the non-elderly group (59.7%) compared to the elderly group (45.5%). A previous study reported a 39.3% prevalence of depression in HD patients [1], which is higher than that observed in non-HD populations. Research on HD patients with a mean age in the 50s has indicated an association between younger age and depression [23], with prevalence rates of 15.9% among those aged 18–44 years and 14.1% among those aged 45–62 years, compared to 12.6% in patients over 75 years [24]. Conversely, studies involving HD patients with a mean age in the 60s have suggested that advanced age is a contributing factor to depressive symptoms [25]. These discrepancies indicate that variations in the age composition of study populations may preclude definitive conclusions. The findings of the present study align with previous research demonstrating a higher prevalence of depressive symptoms in non-elderly HD patients [24].
Aging has generally been reported to be associated with depressive symptoms [25], and the existence of factors specific to hemodialysis patients cannot be ruled out, making it important to investigate factors that are unique to hemodialysis patients and present only in non-elderly individuals. A systematic review on the prevalence of depressive symptoms in chronic kidney disease patients reported a mean patient age of 54.1 years across observational studies [1], underscoring the paucity of evidence concerning elderly individuals. Nonetheless, the management of depressive symptoms remains crucial for both middle-aged and elderly patients, necessitating intervention strategies tailored to the distinct pathophysiological characteristics of each age group.
The non-elderly with depressive symptoms group were younger, had a higher prevalence of diabetes and stress levels, and exhibited significantly lower %CGR and SPPB compared with the non-elderly without depressive symptoms group. Logistic regression analysis identified psychological stress as the only statistically significant factor associated with depressive symptoms in the non-elderly group. Although patients with depressive symptoms exhibited significantly lower SPPB scores compared to those without depressive symptoms, this variable was not retained as an independent factor in the multivariate model. This suggests that the association between low physical function and depressive symptoms may be confounded by other variables, and that stress remained the most robust explanatory factor after adjustment.
Regarding the sub-item analysis of the GDS-5, the proportion of affirmative responses to the question, “Are you basically satisfied with your life?” was notably lower. This finding may be attributed to the restriction of life events and lifestyle activities unique to non-elderly patients due to chronic kidney disease and HD treatment, as compared to healthy individuals. Such restrictions likely contribute to the accumulation of psychological stress, which is associated with reduced life satisfaction and depressive symptoms. Previous studies have identified fluid and dietary restrictions, along with fatigue resulting from HD treatment, as major stressors in younger HD patients [26]. Additionally, non-elderly HD patients often report disruptions linked directly to life events, such as diminished work capacity and sexual dysfunction [27–29]. These factors likely contribute to reduced life satisfaction and the development of depressive symptoms. Despite the reductions in physical function and skeletal muscle mass observed in patients with depressive symptoms, these parameters were not directly associated with depressive symptoms in non-elderly HD patients. One potential explanation is that non-elderly HD patients may have better access to compensatory strategies and are less likely to experience functional dependence, even with impaired physical function. Thus, it is imperative to prioritize interventions targeting the psychological dimensions of depressive symptoms in non-elderly HD patients.
The elderly with depressive symptoms group had a higher proportion of men, a greater history of smoking, a higher prevalence of diabetes, and high CRP levels compared to the elderly without depressive symptoms group. Additionally, Alb, dry weight, nPCR, and SPPB were significantly lower in this group. Logistic regression analysis revealed significant associations between stress, SPPB scores, and depressive symptoms. These findings suggest that stress is more closely linked to depressive symptoms in HD patients than objective measures of dialysis adequacy, nutritional status, or inflammatory markers. Furthermore, physical function, as indicated by SPPB scores, appears to play a key role in the presence of depressive symptoms among elderly HD patients. SPPB, an index of mobility in HD patients, has been associated with poorer prognostic outcomes and higher hospitalization rates [22, 30]. One potential explanation is that reduced mobility and subsequent dependence on caregivers may lead to increased stress and the onset of depressive symptoms. Lower SPPB scores are associated with diminished physical activity, which could exacerbate depressive symptoms by reducing serotonin secretion [31]. It has also been reported that levels of brain-derived neurotrophic factor (BDNF) below 220 pg/ml and levels of IL-6 around 20.47 pg/ml are associated with the onset of depressive symptoms [32, 33].
Alternatively, depressive symptoms themselves may drive stress accumulation and physical dysfunction [34]. Evidence indicates that elderly individuals with depressive symptoms are at higher risk for physical decline and increased functional dependence due to reduced physical activity and diminished social engagement [35]. Ambulatory HD patients with depressive symptoms, particularly older adults, often struggle with activities of daily living. This suggests that depressive symptoms may directly contribute to declines in physical function through mechanisms involving reduced physical activity and social participation.
Addressing depressive symptoms in elderly HD patients requires psychological interventions targeting stress management and exercise therapy to enhance physical function. Educational interventions focused on emotional regulation have demonstrated effectiveness as stress management strategies. Chronic and persistent stress in HD patients necessitates continuous coping mechanisms. Accordingly, stress management training and interventions based on the transactional model have proven effective in fostering emotional control in HD patients [36, 37]. These interventions not only teach the nature of stress and coping strategies but also promote emotional regulation through the development of self-confidence and self-esteem. Such approaches may prevent the onset of depressive symptoms in HD patients, with particular effectiveness observed in enhancing emotional regulation skills in both elderly and younger HD populations, thereby mitigating the psychological burden of stress. These interventions are conducted in a group setting by nurses and clinical psychologists, and due to their format and flexibility, they are highly feasible for implementation in general clinical settings. On the other hand, cognitive behavioral therapy (CBT) is an individual-oriented psychotherapy that targets cognitive distortions, focusing on the interaction between thoughts, emotions, and behaviors [38]. CBT requires ongoing therapist-led sessions and specialized training. While both approaches aim to provide mental health support, they differ in their objectives, implementation methods, and clinical applications.
Exercise therapy has also been reported to be an effective intervention for improving physical function and alleviating depressive symptoms. However, the results of this study suggest that exercise therapy may be positioned differently as a preferred intervention for depressive symptoms in elderly HD and non-elderly HD patients, and we believe that this has important implications for elderly HD patients. Numerous studies, including meta-analyses, have indicated that supervised aerobic exercise can significantly reduce depressive symptoms in HD patients [39]. Sub analyses have revealed that aerobic exercise is more effective than combined or resistance training alone in alleviating depressive symptoms. Furthermore, sustained aerobic exercise over a duration exceeding six months, with sessions lasting 60 minutes or more, has been identified as the most effective regimen for reducing depressive symptoms. These findings suggest that a structured program of aerobic exercise for at least six months, with a session duration of one hour, may substantially benefit HD patients [39].
When implementing exercise therapy in a clinical setting, however, it is necessary to consider the patient’s condition and tolerance, and gradually adjust the duration and intensity of the exercise. In elderly HD patients, exercise therapy has also demonstrated efficacy in improving the SPPB scores [12]. Effective exercises include resistance training using tubing and aerobic activities involving ergometers [40, 41]. Encouraging regular exercise as part of HD treatment protocols may help maintain and improve SPPB scores while concurrently reducing depressive symptoms. However, a direct correlation between SPPB improvement and depressive symptom alleviation remains uncertain, necessitating further research to clarify this relationship.
The study has several limitations. First, the data used in this study were collected from a previously measured database; therefore there is a possibility of unmeasured confounding factors. Specific variables or factors were not included in the data, and it remains possible that these factors could influence the factors associated with depressive symptoms. Second, depressive symptoms were assessed in a simplified manner using only the GDS-5. Depressive symptoms need to be assessed comprehensively because of their diversity. Third, we were unable to collect information on the subjects’ history of mental disorders. Mental disorders may influence the onset of depressive symptoms and may be confounding factors. It will be necessary to collect data including the history of mental disorders and re-examine the results in the future. Fourth, the assessment of stress in this study was based on a binary question, and did not capture detailed information such as the nature, intensity, or duration of stress. Therefore, there are limitations to the stress assessment, and future studies will require a more detailed assessment that includes the characteristics and degree of stress. Fifth, since this study is a secondary analysis using a database of subjects who participated in an intradialytic exercise therapy program, it is possible that there were relatively few elderly participants with depressive symptoms, and the possibility of selection bias remains.
In conclusion, factors associated with depressive symptoms in HD patients differed by age group: stress was a significant factor in non-elderly HD patients, while both stress and low SPPB values were associated in elderly HD patients. These findings suggest that the factors influencing depressive symptoms may vary by age. This study provides important evidence that highlights the need for age-specific interventions to address depressive symptoms in HD patients.
We would like to thank the patients who participated in this study and the hospital staff of the Kaikoukai health care group for their tremendous cooperation during the study period.
None
None
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Research idea and study design: RT, HY; data acquisition: RT, TH, YT, YM; data analysis/interpretation: RT, HY; statistical analysis: RT, HY; supervision or mentorship: KK, YM, HI, TY.
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethical Committee of the Nagoya Kyoritsu Hospital (4/21/2022 approval number. K145-01).
All data from the exercise program during dialysis were recorded in a database, and study data were retrospectively extracted from this database for secondary use, thus waiving the requirement for written informed consent. The patients’ right to decline enrollment was ensured through an opt-out method.
