Prevalence and treatment gap of comorbid depression and anxiety in elderly patients with hemodialysis

Abstract

BACKGROUND

Elderly patients undergoing hemodialysis are highly susceptible to depression and anxiety; however, their treatment status remains inadequately characterized.

AIM

To investigate the prevalence of comorbid depression and anxiety disorders and to identify associated factors among elderly hemodialysis patients admitted to our hospital between June 2022 and June 2025, and to evaluate the treatment gap.

METHODS

Clinical data were retrospectively collected from 300 elderly patients undergoing maintenance hemodialysis during the study period. Descriptive analyses assessed the prevalence of comorbid depression and/or anxiety and the treatment gap. Univariate and multivariate logistic regression analyses identified factors associated with comorbid depression and/or anxiety and the presence of a treatment gap.

RESULTS

Among the 300 patients, 114 met the diagnostic criteria for depression and/or anxiety disorders, corresponding to a prevalence of 38.0% (114/300). Of these, 75 (65.8%) were untreated or inadequately treated, indicating a considerable treatment gap. Multivariate analysis revealed that female sex [odds ratio (OR) = 2.15, 95% confidence interval (CI): 1.32-3.51], living alone (OR = 2.89, 95%CI: 1.72-4.85), dialysis duration ≥ 3 years (OR = 1.98, 95%CI: 1.21-3.24), serum albumin < 35 g/L (OR = 2.41, 95%CI: 1.45-4.01), and Charlson Comorbidity Index ≥ 5 (OR = 2.67, 95%CI: 1.63-4.38) were independent risk factors for comorbid depression and/or anxiety (P < 0.05). Advanced age (OR = 1.12, 95%CI: 1.03-1.21) and education level of primary school or below (OR = 3.01, 95%CI: 1.23-7.36) were independent risk factors for a treatment gap, whereas having a confirmed psychiatric diagnosis was protective (OR = 0.21, 95%CI: 0.08-0.53; P < 0.05).

CONCLUSION

Elderly haemodialysis patients are very likely to have depression and anxiety as well as their main health problem, and there is still a big difference between the number of people who receive treatment and the number of people who need it. If a woman lives alone, if she has had dialysis for a long time, if she has low albumin levels, and if her Charlson Comorbidity Index score is high, she is more likely to have depression and/or anxiety. Older age and lower education can mean that she does not get the treatment she needs. A confirmed psychiatric diagnosis markedly reduces this gap. This population’s outcomes and quality of life can be improved by enhanced psychological screening, timely intervention for high-risk individuals and multidisciplinary collaboration.

Key Words: Elderly; Hemodialysis; Depression; Anxiety disorders; Prevalence; Treatment gap

Core Tip: Among elderly patients with hemodialysis, the prevalence of depression and anxiety was 38.0%, with 65.8% exhibiting a treatment gap. Female sex, living alone, longer hemodialysis duration, hypoalbuminemia, and higher Charlson Comorbidity Index scores were risk factors, whereas a confirmed psychiatric diagnosis significantly reduced the treatment gap.

INTRODUCTION

In the face of a global demographic shift towards an ageing population and the persistent rise in the incidence of chronic ailments such as diabetes mellitus and hypertension, end-stage renal disease has emerged as a grave public health concern[1,2]. So, haemodialysis is the main way to treat kidney failure, right? It basically gets rid of all the metabolic waste and makes sure you don’t die, but it’s a pretty tough process. It’s non-stop and it’s really tiring for the patients. It puts a lot of strain on their bodies, minds and their wallets too. The challenges faced by elderly haemodialysis patients are numerous[3,4]. In addition to the complications of end-stage renal disease, there are also age-related declines in physiological function, polypharmacy and cognitive impairment. All of these factors have a marked effect on quality of life and increase susceptibility to psychological disorders[5,6]. Meta-analyses have reported pooled prevalence rates of depression. These range from approximately 20%-40% among hemodialysis patients. Meta-analyses also report pooled anxiety disorders rates. These range from about 15%-30% among hemodialysis patients. Even higher rates are observed in the elderly[7,8]. Nevertheless, a considerable proportion of patients remain unaware of their psychological afflictions, whilst the inequitable distribution of medical resources, in conjunction with the dearth of accessibility to specialised mental health services, has engendered treatment rates that are considerably lower than the prevalence rates, thereby engendering a substantial treatment gap that further exacerbates the disease burden and curtails survival chances[9,10].

These challenges have led to this retrospective medical record study, which set out to do two things: First, to assess the prevalence of depression and anxiety disorders, and second, to analyse the magnitude and determinants of the treatment gap among elderly haemodialysis patients. The hope is that this will provide evidence for targeted interventions and the optimal allocation of healthcare resources.

MATERIALS AND METHODS

Study population

These challenges have led to this retrospective medical record study, which set out to do two things: First, to assess the prevalence of depression and anxiety disorders, and second, to analyse the magnitude and determinants of the treatment gap among elderly haemodialysis patients. The hope is that this will provide evidence for targeted interventions and the optimal allocation of healthcare resources.

Inclusion criteria: (1) The age of the patient is ≥ 65 years; (2) The patient undergoes regular haemodialysis (2-3 sessions per week); (3) The patient has undergone maintenance dialysis for more than 3 months; and (4) The patient has complete and traceable medical records.

Exclusion criteria: (1) A serious problem with thinking, memory, or understanding; (2) A serious illness that cannot be cured; (3) Life expectancy less than 6 months; (4) A sudden problem with the kidneys that needs temporary dialysis; and (5) Missing key information (e.g., medication or test results) in more than 20% of the medical record.

Data collection

A standardized form was used to extract the following information from the electronic medical record system.

Sociodemographic data: (1) Age; (2) Sex; (3) Marital status (married/not married); (4) Living arrangement (living alone/Living with family); (5) Educational level (primary or below/secondary/college or above); (6) Body mass index (BMI); and (7) Method of medical payment (medical insurance/public/self-pay).

Clinical data: (1) Primary renal disease; (2) Dialysis vintage; (3) Charlson Comorbidity Index (CCI); and (4) Comorbidities.

Laboratory and dialysis-related indices: (1) Hemoglobin; (2) Serum albumin; (3) Serum creatinine; (4) Blood urea nitrogen; (5) Estimated glomerular filtration rate; (6) Serum calcium; (7) Serum phosphate; (8) Serum potassium; (9) Intact parathyroid hormone (iPTH); and (10) Single-pool Kt/V.

Definition of comorbid depression and anxiety

Comorbid depression and anxiety were defined as either a documented diagnosis of “depression”, “anxiety disorder”, “depressive episode”, or “generalized anxiety disorder” made by a psychiatrist or after psychiatric consultation, or continuous prescription (≥ 3 months) of antidepressant and/or anxiolytic medication.

Definition of treatment gap

Among patients who met the above diagnostic criteria, those who had received neither structured psychological intervention nor psychiatric medication were considered to have a treatment gap. Adequate treatment was defined as follows: (1) Prescription of antidepressant or anxiolytic drugs with documented continuous use; and (2) Completion of at least four sessions of structured psychotherapy or counseling.

Outcomes

Descriptive analysis of the prevalence of comorbid depression and anxiety and of the treatment gap. Univariate and multivariate logistic regression analyses of factors associated with comorbid depression and anxiety. Univariate and multivariate logistic regression analyses of factors associated with the treatment gap among patients with comorbid depression and anxiety.

Statistical analysis

Statistical data analysis was performed using SPSS version 26.0 (IBM Corp., Armonk, NY, United States). The normality of continuous variables was assessed using the Shapiro-Wilk test. Continuous variables meeting normality requirements were expressed as mean ± SD, with intergroup comparisons performed using independent samples t-tests. Non-normally distributed variables were presented as median (interquartile range), with intergroup comparisons conducted using the Mann-Whitney U test. Categorical variables were reported as case n (%), with intergroup comparisons performed using χ² tests or Fisher’s exact probability test.

Collinearity diagnosis: Prior to multivariate logistic regression, Pearson correlation coefficients were calculated among all continuous independent variables. Variance inflation factor (VIF) and tolerance values were computed to assess multicollinearity, with the aim of determining whether there was any multicollinearity present. Things that can change were checked carefully if they had a VIF of 10 or more or were very similar (r of 0.70 or more). Serum phosphorus and iPTH exhibited moderate collinearity (r = 0.72, VIF > 10). After including both in the multivariate model, neither remained significant (P > 0.05). Given the clinical relevance of serum albumin as a comprehensive nutritional indicator and its lower collinearity with other variables, serum phosphorus and iPTH were excluded from the final model to ensure model stability.

RESULTS

Prevalence of comorbid depression and anxiety

Among the 300 elderly hemodialysis patients, 114 met the diagnostic criteria for depression and/or anxiety, yielding an overall prevalence of 38.0% (114/300). Details are presented in Figure 1.

Figure 1  Distribution of the prevalence of comorbid depression and anxiety disorders among elderly hemodialysis patients.

Univariate analysis of factors associated with comorbid depression and anxiety

Compared with the non-depressed/non-anxious group, the depressed/anxious group had significantly higher proportions of female patients, individuals living alone, and patients with an educational level of primary school or below. In addition, this group had longer dialysis vintage (≥ 3 years), higher CCI ≥ 5 lower serum albumin, and higher serum phosphate and iPTH levels (P < 0.05). No significant differences were found in age, marital status, method of payment, primary renal disease, BMI, hemoglobin, serum creatinine, blood urea nitrogen, estimated glomerular filtration rate, serum calcium, serum potassium, single-pool Kt/V, or the presence of comorbid hypertension, diabetes, coronary heart disease, or cerebrovascular disease (P > 0.05; Table 1).

Table 1 Univariate analysis of comorbid depression and anxiety disorders among elderly hemodialysis patients, n (%)/mean ± SD/median (interquartile range).

Variable	Depression and anxiety group (n = 114)	Non-depressed/non-anxious group (n = 186)	Statistic	P value
Age (years)	73.1 ± 6.0	72.0 ± 5.6	t = 1.613	0.108
Female	62 (54.4)	60 (32.3)	χ² = 15.237	< 0.001
Married	76 (66.7)	139 (74.7)	χ² = 2.354	0.125
Living alone	38 (33.3)	29 (15.6)	χ² = 13.576	< 0.001
Education ≤ primary school	68 (59.6)	74 (39.8)	χ² = 11.856	0.001
Medical payment: Insurance	105 (92.1)	176 (94.6)	χ² = 0.749	0.387
Primary renal disease			χ² = 4.212	0.239
Diabetic nephropathy	46 (40.4)	62 (33.3)
Hypertensive nephropathy	33 (28.9)	64 (34.4)
Chronic glomerulonephritis	14 (12.3)	31 (16.7)
Other/unknown	21 (18.4)	29 (15.6)
Dialysis vintage (years)	3.2 (1.5, 5.5)	2.1 (1.0, 4.0)	Z = -3.362	0.001
Dialysis vintage ≥ 3 years	68 (59.6)	75 (40.3)	χ² = 11.213	0.001
Body mass index (kg/m2)	22.1 ± 3.2	22.6 ± 3.5	t = -1.215	0.225
Charlson Comorbidity Index ≥ 5	78 (68.4)	82 (44.1)	χ² = 17.632	< 0.001
Comorbidities
Hypertension	102 (89.5)	159 (85.5)	χ² = 0.974	0.324
Diabetes mellitus	58 (50.9)	95 (51.1)	χ² = 0.001	0.970
Coronary heart disease	51 (44.7)	72 (38.7)	χ² = 1.116	0.291
Cerebrovascular disease	29 (25.4)	38 (20.4)	χ² = 1.056	0.304
Hemoglobin (g/L)	104.3 ± 13.1	106.4 ± 11.8	t = -1.430	0.154
Serum albumin (g/L)	35.5 ± 4.0	37.6 ± 4.1	t = -4.437	< 0.001
Albumin < 35 g/L	49 (43.0)	45 (24.2)	χ² = 12.115	0.001
Serum creatinine (μmol/L)	678.5 (555.8, 812.3)	712.0 (580.5, 845.5)	Z = -1.105	0.269
Blood urea nitrogen (mmol/L)	24.8 ± 6.1	25.5 ± 5.7	t = -1.013	0.312
Estimated glomerular filtration rate (mL/minute/1.73 m2)	6.5 ± 1.8	6.7 ± 1.6	t = -1.007	0.315
Serum calcium (mmol/L)	2.18 ± 0.21	2.21 ± 0.19	t = -1.265	0.207
Serum phosphorus (mmol/L)	1.98 ± 0.52	1.84 ± 0.48	t = 2.428	0.016
Intact parathyroid hormone (pg/mL)	325.0 (188.5, 486.8)	268.0 (156.0, 398.0)	Z = -2.149	0.032
Serum potassium (mmol/L)	4.7 ± 0.6	4.6 ± 0.5	t = 1.538	0.125
Single-pool Kt/V	1.38 (1.25, 1.52)	1.41 (1.28, 1.55)	Z = -1.486	0.137

Independent-samples t-test, Mann-Whitney U test, or χ² test were applied as appropriate.

Multivariate logistic regression analysis of factors associated with comorbid depression and anxiety

Using the presence of comorbid depression and anxiety (yes = 1, no = 0) as the dependent variable, and sex (female = 1, male = 0), living arrangement (living alone = 1, living with family = 0), educational level (primary school or below = 1, secondary school or above = 0), dialysis vintage (≥ 3 years = 1, < 3 years = 0), CCI (≥ 5 = 1, < 5 = 0), serum albumin (< 35 g/L = 1, ≥ 35 g/L = 0), serum phosphate (continuous, per 1 mmol/L increase), and iPTH (continuous, per 100 pg/mL increase) as independent variables, multivariate logistic regression analysis (forward logistic regression method) revealed that female sex, living alone, dialysis vintage ≥ 3 years, serum albumin < 35 g/L, and CCI ≥ 5 were independent risk factors for comorbid depression and anxiety (P < 0.05; Table 2).

Table 2 Multivariate logistic regression analysis of comorbid depression and anxiety disorders among elderly hemodialysis patients.

Variable	β	SE	Wald χ²	P value	Odds ratio	95%CI
Sex (female vs male)	0.765	0.248	9.524	0.002	2.15	1.32-3.51
Living status (living alone vs living with others)	1.061	0.264	16.143	< 0.001	2.89	1.72-4.85
Education (primary school or below vs secondary school or above)	0.312	0.231	1.824	0.177	1.37	0.87-2.15
Dialysis vintage (≥ 3 years vs < 3 years)	0.683	0.248	7.589	0.006	1.98	1.21-3.24
Charlson Comorbidity Index (≥ 5 vs < 5)	0.982	0.250	15.426	< 0.001	2.67	1.63-4.38
Serum albumin (< 35 g/L vs ≥ 35 g/L)	0.880	0.255	11.914	0.001	2.41	1.45-4.01
Serum phosphorus (per 1 mmol/L increase)	0.285	0.198	2.071	0.150	1.33	0.90-1.96
Intact parathyroid hormone (per 100 pg/mL increase)	0.103	0.074	1.937	0.164	1.11	0.96-1.28
Constant	-3.521	0.452	60.673	< 0.001	0.030

Collinearity diagnosis: Correlation analysis revealed a moderate-to-high correlation between serum phosphorus and iPTH (r = 0.72, P < 0.001). VIF values of 12.28 and 11.52 were indicated for serum phosphorus and iPTH, respectively, suggesting the presence of multicollinearity (VIF > 10). When both variables were included in the multivariate model, serum phosphorus lost statistical significance. The odds ratio (OR) was 1.33, 95% confidence interval (CI): 0.90-1.96, P = 0.150. Similarly, iPTH lost statistical significance. The OR was 1.11, 95%CI: 0.96-1.28, P = 0.164. Serum albumin (< 35 g/L) showed a moderate correlation with serum phosphorus (r = 0.55) but an acceptable VIF (1.82). Therefore, serum albumin was retained in the final model as a more comprehensive nutritional indicator, whereas serum phosphorus and iPTH were excluded to prevent collinearity-induced instability.

Treatment gap profile in patients with comorbid depression and anxiety

Among the 114 patients who met the diagnostic criteria, 75 (65.8%) had a treatment gap, whereas 39 (34.2%) received guideline-concordant care. The distribution is presented in Figure 2.

Figure 2  Distribution of treatment gaps among elderly hemodialysis patients with comorbid depression and anxiety disorders.

Univariate analysis of factors associated with the treatment gap

Patients in the treatment-gap group were significantly older, more likely to live alone, had an educational level of primary school or below, lower serum albumin levels, and were less frequently given a definite psychiatric diagnosis (P < 0.05). No significant differences were observed between the two groups in sex, marital status, method of payment, primary renal disease, dialysis vintage, BMI, CCI ≥ 5, or other laboratory indices and comorbidities (P > 0.05; Table 3).

Table 3 Univariate analysis of treatment gaps among elderly hemodialysis patients with comorbid depression and anxiety disorders, n (%)/mean ± SD/median (interquartile range).

Variable	Treatment group (n = 39)	Treatment-gap group (n = 75)	Statistic	P value
Age (years)	70.2 ± 4.8	73.8 ± 6.1	t = -3.415	0.001
Female	20 (51.3)	42 (56.0)	χ² = 0.243	0.622
Married	28 (71.8)	48 (64.0)	χ² = 0.763	0.382
Living alone	10 (25.6)	33 (44.0)	χ² = 4.912	0.027
Education ≤ primary school	15 (38.5)	44 (58.7)	χ² = 5.678	0.017
Medical payment: Insurance	37 (94.9)	68 (90.7)	χ² = 0.703	0.402
Primary renal disease			χ² = 1.892	0.595
Diabetic nephropathy	18 (46.2)	28 (37.3)
Hypertensive nephropathy	10 (25.6)	23 (30.7)
Chronic glomerulonephritis	5 (12.8)	9 (12.0)
Other/unknown	6 (15.4)	15 (20.0)
Dialysis vintage (years)	3.0 (1.3, 5.2)	3.3 (1.6, 5.7)	Z = -0.823	0.411
Dialysis vintage ≥ 3 years	24 (61.5)	44 (58.7)	χ² = 0.094	0.759
Body mass index (kg/m2)	22.3 ± 3.0	22.0 ± 3.3	t = 0.467	0.641
Charlson Comorbidity Index ≥ 5	26 (66.7)	52 (69.3)	χ² = 0.095	0.758
Comorbidities
Hypertension	36 (92.3)	66 (88.0)	χ² = 0.539	0.463
Diabetes mellitus	21 (53.8)	37 (49.3)	χ² = 0.222	0.637
Coronary heart disease	19 (48.7)	32 (42.7)	χ² = 0.402	0.526
Cerebrovascular disease	8 (20.5)	21 (28.0)	χ² = 0.806	0.369
Psychiatrist-confirmed diagnosis	32 (82.1)	40 (53.3)	χ² = 9.782	0.002
Hemoglobin (g/L)	105.8 ± 12.5	103.6 ± 13.4	t = 0.872	0.385
Serum albumin (g/L)	36.8 ± 3.5	34.9 ± 4.1	t = 2.521	0.013
Albumin < 35 g/L	12 (30.8)	37 (49.3)	χ² = 4.115	0.043
Serum creatinine (μmol/L)	665.0 (548.5, 795.5)	685.0 (560.0, 820.0)	Z = -0.538	0.591
Blood urea nitrogen (mmol/L)	24.5 ± 5.8	25.0 ± 6.3	t = -0.427	0.670
Estimated glomerular filtration rate (mL/minute/1.73 m2)	6.6 ± 1.7	6.5 ± 1.8	t = 0.289	0.773
Serum calcium (mmol/L)	2.20 ± 0.20	2.17 ± 0.21	t = 0.734	0.465
Serum phosphorus (mmol/L)	2.01 ± 0.50	1.96 ± 0.53	t = 0.487	0.627
Intact parathyroid hormone (pg/mL)	315.0 (185.0, 455.0)	330.0 (190.0, 495.0)	Z = -0.634	0.526
Serum potassium (mmol/L)	4.7 ± 0.5	4.7 ± 0.6	t = 0.000	1.000
Single-pool Kt/V	1.39 (1.26, 1.53)	1.38 (1.24, 1.51)	Z = -0.315	0.753

Independent-samples t-test, Mann-Whitney U test, or χ² test were applied as appropriate.

Multivariate logistic regression analysis of factors associated with the treatment gap

Using the presence of a treatment gap (yes = 1, no = 0) as the dependent variable, and age (continuous, per 1-year increase), living arrangement (living alone = 1, living with family = 0), educational level (primary school or below = 1, secondary school or above = 0), serum albumin (continuous, per 1 g/L increase), and definite psychiatric diagnosis (yes = 1, no = 0) as independent variables, multivariate logistic regression analysis (forward logistic regression method) revealed that advanced age and low educational level (primary school or below) were independent risk factors for a treatment gap (P < 0.05), whereas a definite diagnosis by a psychiatrist was a protective factor (P < 0.05; Table 4).

Table 4 Multivariate logistic regression analysis of treatment gaps among elderly hemodialysis patients with comorbid depression and anxiety disorders.

Variable	β	SE	Wald χ²	P value	Odds ratio	95%CI
Age (per 1-year increase)	0.112	0.041	7.465	0.006	1.12	1.03-1.21
Education (primary school or below vs secondary school or above)	1.102	0.456	5.841	0.016	3.01	1.23-7.36
Psychiatrist-confirmed diagnosis (yes vs no)	-1.562	0.478	10.676	0.001	0.21	0.08-0.53
Living status (living alone vs living with others)	0.654	0.469	1.945	0.163	1.92	0.77-4.82
Serum albumin (per 1 g/L increase)	-0.085	0.058	2.147	0.143	0.92	0.82-1.03
Constant	-10.235	3.252	9.901	0.002	0.000

DISCUSSION

Quality of life in elderly hemodialysis patients depends not only on dialysis adequacy but also on the absence of psychological morbidity. Many older patients are stuck in a monotonous cycle of hospital and home, spending hours on end three times a week attached to a dialysis machine. They’re forced to follow stringent dietary and fluid rules, and their social lives and physical abilities gradually decline. A combination of ongoing worries about how illnesses progress, the chance of death and money troubles is a perfect setting for depression and anxiety to take root[11,12]. A depressed person will typically experience a consistent low mood and a loss of interest in things that they would usually enjoy. This is often accompanied by disturbances in sleep and appetite, fatigue, feelings of worthlessness, and suicidal thoughts. Anxiety disorders, on the other hand, are characterised by excessive fear, worry, and autonomic symptoms. It is common for these two conditions to occur together, which can result in a variety of complex and interconnected clinical symptoms[13,14].

The current study found that 38.0% of patients had depression and anxiety, and 65.8% of these patients did not receive treatment, which is similar to previous reports[15-18]. These results show that mental health problems are common among elderly dialysis patients and are often not recognized. A study of many things found that women, people who live on their own, people who have been on dialysis for 3 years or more, people with low albumin levels (less than 35 g/L) and people with a CCI score of 5 or more had a higher risk of feeling depressed and anxious (P < 0.05). Psychological distress is generally more likely to be reported by women, who are more sensitive to it. Additional burdens may be imposed by role transitions within the family. The 2.15-fold higher risk observed in this study is consistent with the findings of da Silva et al[19]. Older adults living alone lack immediate emotional and practical support, and may feel more helpless when faced with physical symptoms and existential fears. This pattern is consistently associated with poorer mental health outcomes[20-23]. A longer dialysis vintage means a patient is exposed to loss of autonomy for a longer period of time. This, in turn, can lead to a number of complications and ongoing uncertainty. All of these factors can gradually erode a patient’s psychological resilience. The presence of hypoalbuminemia can indicate a condition involving protein-energy wasting. Low serum albumin levels have been linked to impaired neurotransmitter synthesis and exacerbated fatigue, which can in turn contribute to the development of depressive symptoms[24-26]. A CCI ≥ 5 shows more than one illness, a bigger drop in how well the body works, and more complicated treatment, all of which make psychological stress worse[27].

What we found was really interesting, and it’s something we hadn’t expected. We saw that when people were given a formal psychiatric diagnosis, there was a much smaller chance of a treatment gap (OR = 0.21). However, the causal direction of this association requires cautious interpretation, as it is not clear whether the effect is caused by the factor under investigation or whether it is a result of some other factor. The act of diagnosis itself may facilitate access to targeted treatments, such as pharmacotherapy or psychotherapy, which can be effective in addressing specific mental health concerns. Conversely, patients who proactively seek and obtain a psychiatric diagnosis may possess intrinsic characteristics that facilitate treatment engagement, including superior health literacy, heightened motivation for help-seeking, enhanced social support, and diminished internalised stigma towards mental illness. Due to the cross-sectional and retrospective design of this study, it is not possible to disentangle these possibilities. There is a possibility that both mechanisms function at the same time: The diagnostic label validates the necessity for treatment within the healthcare system, while the individual qualities that contributed to acquiring the diagnosis also encourage compliance with treatment. This shows that just improving how we check for problems may not be enough. We also need to deal with the emotional and mental problems that stop patients from getting a diagnosis.

Treatment gaps have been found to be independently predicted by advanced age and low educational attainment (primary school level or below), whereas a potent protective factor has been identified as a clear psychiatric diagnosis. This may be closely linked to the phenomena of ‘stigma surrounding mental illness’ and ‘symptom somatisation’. Older patients, particularly those with lower educational attainment, may be more inclined to express or attribute depressive and anxious emotional experiences as physical discomfort (such as fatigue, loss of appetite, insomnia). These symptoms overlap with manifestations of uraemia or inadequate dialysis, leading clinicians to attribute psychological distress to physical causes. Concurrently, both internalised and externalised stigma surrounding mental illness hinders their willingness to disclose symptoms and seek assistance[28,29]. A formal psychiatric diagnosis can break this cycle. It not only clarifies the nature of symptoms and reduces patients' shame from inappropriate labelling but also paves the way for subsequent pharmacological or psychological treatment, thereby significantly narrowing the treatment gap[30]. Consequently, interventions for this population should prioritise reducing stigma through public education and training healthcare professionals to recognise somatised manifestations of depression and anxiety.

This study has several limitations. First, the definition of depression and anxiety that are also present, or comorbid, was based on previous psychiatric diagnoses or prescriptions for medications that affect the mind. This method of looking at past data, which is based on what has already happened, is practical, but probably doesn’t show how common the problem really is. Many elderly hemodialysis patients do not undergo formal psychiatric assessment. They also do not receive appropriate treatment. This is especially true of those presenting primarily with somatic complaints. It is also true of those presenting with cognitive impairment. And it is true of those presenting with stigma-related barriers to help-seeking. Therefore, the reported prevalence of 38.0% should be regarded as a lower-bound estimate. Future prospective studies employing standardized diagnostic interviews (e.g., Mini-International Neuropsychiatric Interview, Structured Clinical Interview for Diagnostic and Statistical Manual for Mental Disorders) are needed to capture the full spectrum of psychological morbidity in this population. It is worth noting that this study’s operational definition of the “treatment gap” relies entirely on treatment behaviours recorded within the healthcare system (pharmacotherapy or structured psychotherapy). This definition may overlook the emotional solace and psychological coping mechanisms patients derive from informal channels such as family support, religious beliefs, or community activities. Among the elderly population, the role of such ‘non-professional support’ should not be underestimated. It may to some extent mitigate psychological distress or influence patients’ willingness to seek professional help. Simply categorising this as ‘untreated’ may be overly absolute. Future research should explore how to more comprehensively assess the level of psychosocial support patients receive, and consider its potential impact on treatment gaps and clinical outcomes. This would enable the development of more precise, stratified intervention strategies. Therefore, based on the identified risk factors, we propose a structured, risk-stratified, multidisciplinary, and collaborative intervention roadmap to translate these findings into actionable clinical practice.

Nutrition-psychology integrated management for patients with hypoalbuminemic

Given that serum albumin < 35 g/L was identified as an independent risk factor for depression and anxiety (OR = 2.41), we recommend establishing a Nutrition-Psychology Dual-Track Screening Protocol. For all elderly hemodialysis patients with serum albumin < 35 g/L, nephrologists should initiate automatic referrals for psychological assessment using standardized tools (e.g., Patient Health Questionnaire 9, 7-item Generalized Anxiety Disorder Scale) during routine nutritional evaluations. Dietitians should be trained to recognize somatic manifestations of depression – such as appetite changes and fatigue – that overlap with malnutrition and to flag high-risk patients for psychiatric consultation. A pilot programme could include checking albumin levels in the blood every month and doing short mental health checks. After that, different medical experts could meet to talk about the patient’s case. This could include doctors who specialise in kidney problems, dietitians (nutrition experts), and psychiatrists (doctors who treat mental health problems). This would be for patients who have had low albumin levels for a long time and have high scores on depression tests.

Social support intervention for patients living alone

Living alone was associated with the highest risk for depression and anxiety. The OR for this was 2.89. It was also associated with a higher likelihood of a treatment gap. The OR for this was 1.92. This was not statistically significant in multivariate analysis. We propose the following components for a Community-Social Work-Nephrology Linkage Programme.

Risk identification: Patients living alone should be marked in the computer system with automatic messages reminding healthcare workers to ask them about their mood at each dialysis session.

Structured home visits: Social workers should assess patients living alone every month by using brief, validated instruments to evaluate both their functional capacity and psychological well-being.

Peer support networks: Set up programmes where volunteers act as “Dialysis Buddies”, pairing patients living alone with either stable dialysis patients or trained community volunteers.

Emergency linkage: Develop 24-hour helplines. These should connect patients living alone with dialysis units. They should also connect them with mental health crisis intervention services.

Sex-specific and age-stratified screening

Female patients require tailored screening and intervention strategies. The same is true for older patients. For them, the treatment gap increases by 1.12 per year. Female patients should have their mental health screened as part of routine gynaecological health checks. Structured peer support groups can help to reduce the caring responsibilities they face. For patients aged 75 and over who have not had much education, pictures or simple tools should be used to check for signs of illness. Family members (when they are available) should be involved in decisions about treatment to deal with problems related to people not understanding health information well.

Institutionalizing nephrologist-psychiatrist collaboration

The most important factor protecting against the treatment gap was a confirmed psychiatric diagnosis (OR = 0.21). We recommend the following: (1) Include mental health services in places where people have kidney problems, with a psychiatrist visiting every month; (2) Make it so that people can be sent to the right place for treatment, with clear rules about what to look for (for example, if the Patient Health Questionnaire 9 score is 10 or more, or if the 7-item Generalised Anxiety Disorder Scale score is 10 or more, or if the doctor thinks there is a strong chance of that); and (3) Create shared electronic health records that let psychiatrists see information about kidney problems and let kidney doctors check psychiatric treatment plans.

Quality metrics and implementation monitoring

To make sure that the process is carried out successfully, dialysis centres should keep an eye on the following indicators: (1) How often people with high risk are checked for psychological problems every month; (2) How long it takes from the result of the screening to the patient being seen by a psychiatrist; (3) How many patients with low albumin levels are given psychological assessments; and (4) How often patients living alone are contacted by a social worker. Dialysis units should make these indicators part of their plans to improve quality. This risk-stratified, multidisciplinary strategy goes beyond basic suggestions to provide specific and quantifiable actions that tackle the particular vulnerabilities pinpointed in this investigation.

CONCLUSION

Elderly haemodialysis patients are commonly affected by depression and anxiety, yet these conditions are often not addressed adequately. The presence of certain characteristics has been demonstrated to engender an elevated risk of the development of psychological maladies, including, but not limited to, female gender, solitary residence, prolonged dialysis vintage, hypoalbuminemia, and elevated CCI scores. Concurrently, advanced age and diminished educational attainment have been identified as contributing factors to the treatment gap. The most effective way to reduce that gap and improve overall clinical outcomes is to make a confirmed psychiatric diagnosis.