Abstract
Background
There is a clear need for more effective screening tools for patients undergoing hemodialysis, especially considering the wide variation in depression screening options and the lack of a gold standard assessment tool for this specific population. This study aims to examine the validity and reliability of the Turkish version of the Depression Inventory for Maintenance Hemodialysis Patients (DI-MHD).
Methods
This was a methodological study of the translation, cultural adaptation, and psychometric validation of the DI-MHD. Two hundred eighty-three patients from two hemodialysis units in Turkey were recruited for the study using convenience sampling. The following data were collected from the participants: descriptive characteristics, DI-MHD scores, Beck Depression Inventory (BDI) scores, and inflammatory biomarkers. The BDI was used as a reference instrument to evaluate the concurrent validity of the DI-MHD. Exploratory and confirmatory factor analyses were implemented to test the construct validity. The test‒retest method was used to test the reliability and consistency of the scale over time. IBM SPSS version 22.0 and AMOS 24.0 were used for analysis.
Results
Exploratory factor analysis revealed a dominant single factor explaining 37.21% of the total variance (extracted variance = 33.52%). Although confirmatory factor analysis demonstrated acceptable fit for the correlated four-factor and second-order models, high latent factor correlations (φ = 0.677–1.052) and strong higher-order loadings (0.708–1.041) indicated limited discriminant validity among dimensions. Therefore, the one-factor model was retained as the most parsimonious and theoretically coherent solution.
Conclusion
The Turkish version of the DI-MHD is a valid and reliable tool that can be used to evaluate and classify the depression levels of patients in hemodialysis units.
Clinical trial registration
ClinicalTrials.gov NCT. First registration date: 2025-02-20.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12882-026-04979-x.
Keywords: Depression, Hemodialysis, Reliability, Validity, Management, Health
Introduction
Depression is frequently observed among patients undergoing hemodialysis; however, it causes an impaired quality of life, a decrease in adherence to treatment, and an increase in hospitalization and mortality rates [1–4]. According to the literature, the prevalence rate of depression in patients undergoing hemodialysis varies between 25% and 73% [5–7]. Despite its high prevalence rate and adverse effects, only a low percentage of patients with depressive symptoms are diagnosed and treated. This is primarily associated with the fact that depressive symptoms in patients are not adequately recognized and examined [3, 8].
Depression is assessed through structured clinical interviews or diagnostic scales. Structured clinical interviews are often regarded as the “gold standard” for the diagnosis of depressive disorders since they are based on specific diagnostic criteria. However, since it is not always possible for patients to be evaluated by a psychiatrist, some scales are used instead. Even though these scales are not diagnostic criteria, they are helpful in identifying patients at risk [9]. However, screening may also result in false negatives, where cases of depression are not detected and, consequently, go untreated. Patients with kidney failure differ from the general population because they experience higher rates of depression.
Depression is up to five times more prevalent in patients undergoing dialysis than in the general population, with approximately 40% of individuals with end-stage renal disease requiring hemodialysis experiencing this condition [10]. In our country, the prevalence of depression in patients undergoing hemodialysis varies between 22.5% and 58.1%. Global prevalence rates have been reported to range between approximately 22.2% and 68% [11, 12]. The prevalence observed in our country is consistent with these global estimates. They often exhibit symptoms that resemble the physical symptoms of depression, which are connected to their underlying medical conditions and treatments [13]. Patients undergoing hemodialysis experience depression as a result of physiological factors such as cerebrovascular disease, inflammation, malnutrition, changes in autonomic activity, and behavioral and sex-related factors. Depression is characterized by symptoms such as loss of interest, sleep disturbances, feelings of worthlessness and guilt, fatigue, difficulty concentrating, changes in weight and appetite, restlessness or slowed movements, and thoughts of suicide [14]. Despite the prevalence of these symptoms, there is no standardized screening process for depression, resulting in inconsistent assessment methods. In addition to antidepressant medications, nonpharmacological approaches such as cognitive behavioral therapy, exercise, music therapy, light therapy, counseling, and a healthy diet are commonly used to treat depression [14, 15].
In the literature, subjective diagnostic tools such as the Beck Depression Inventory, Hospital Anxiety and Depression Scale, Geriatric Depression Scale-15, Patient Health Questionnaire-9, Cognitive Depression Index, Center for Epidemiologic Studies–Depression Scale, and Hamilton Depression Scale are used to diagnose depression in patients undergoing hemodialysis [13, 16]. Self-contained assessment scales can be prioritized for screening and clinical research, as they reduce patient burden and minimize the influence of researchers’ subjective factors. However, there are significant differences between various diagnostic tools. When self-assessment questionnaires are used with individuals with chronic illnesses, confusion can arise, as symptoms such as fatigue, sleep disturbances, and loss of appetite are common in both end-stage renal disease and depression. Patients who experience significant uremic symptoms may score higher on depression assessments. This can make it challenging for researchers to differentiate between somatic symptoms caused by depression, the underlying disease, or a combination of both, potentially leading to an increased diagnosis of depression. Self-reported scales may contribute to overdiagnosis by attributing symptoms commonly associated with end-stage renal disease, such as fatigue, sleep disturbances, and poor appetite, to somatic symptoms of depression. Structured clinical interviews conducted by experienced psychiatrists are considered the gold standard for definitively diagnosing major depression in patients undergoing hemodialysis [14, 17]. While the Beck Depression Inventory (BDI) is widely used in research and mental health settings, it was developed on the basis of outdated DSM-IV criteria and may be less effective for screening and assessment because of its emphasis on somatic symptoms [13]. Other validated tools for depression screening are more commonly used in the general population.
The scales used to assess depression rely on subjective data reported by patients, with the exception of one scale [18], which is designed to evaluate depression in the general population. A meta-analysis investigating the use of objective methods in the diagnosis of depression reported that circulating inflammatory biomarkers, such as interleukin-6, serum albumin, C-reactive protein (CRP), and tumor necrosis factor-alpha, in the general population [19, 20] were associated with depressive symptoms [21]. In another meta-analysis, some studies revealed that depressive symptoms were observed in patients with low serum albumin levels and high C-reactive protein (CRP) levels; however, others reported no correlation between CRP and serum albumin levels [16].
Indeed, there is a clear need for more effective screening tools for patients undergoing hemodialysis, given the wide variation in depression screening methods and the absence of a gold-standard assessment tool tailored to this specific population. The development of an easily administered, disease-specific self-report measure for depression could prove to be highly beneficial. Only one study has examined a depression tool specifically designed for patients on maintenance dialysis, known as the Depression Inventory–Maintenance Hemodialysis (DI-MHD) [18]. There is currently no Turkish-language, dialysis-specific instrument designed specifically to screen for depression in dialysis patients. When the psychometric properties of the DI-MHD were evaluated, it was found to have a higher Cronbach’s alpha than the HADS and BDI, indicating good internal consistency [22]. The DI-MHD may be a more effective depression screening tool than the BDI, as it has significantly higher sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Youden index values [18]. Additionally, the sensitivity (0.97) and specificity (0.86) of the DI-MHD cutoff scores determined via ROC analysis were good and higher than those of the BDI in our study and in previous research [23]. The objective of this methodological study was to assess the validity and reliability of the Turkish version of the “DI-MHD,” originally developed by Wang et al. [18], when applied to patients undergoing chronic hemodialysis treatment.
Methods
Design and setting
We employed the standard bilingual translation and validation methodology for the DI-MHD [24]. The study was conducted in a total of 2 hemodialysis units. The Private X Dialysis Center and the Y Dialysis Center are located in the two largest cities of Turkey between April and August 2022. To maintain methodological rigor, the study followed the Guidelines for Reporting Reliability and Agreement Studies by Kottner et al. [25] and the applicable sections of the Consensus-Based Standards for the Selection of Health Measurement Instruments (COSMIN) reporting guidelines for studies on measurement properties by Gagnier et al. [26]. During the development of the Turkish version of the DI-MHD, the following steps were followed in accordance with the study by Wild et al. [27], titled “Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: Report of the ISPOR Task Force for Translation and Cultural Adaptation”:
Preparation: The research team assessed the suitability of the scale items for Turkish culture.
Forward Translation: The scale was translated into Turkish by four linguists proficient in both English and Turkish.
Reconciliation: The research team reviewed the forward translations and synthesized the initial Turkish version of the scale.
Back Translation: The Turkish version was back-translated into the original language by two independent linguists.
Back Translation Review: The research team examined the back-translations to ensure conceptual consistency with the original English version.
Harmonization: The original version of the scale was compared with the version obtained during the back-translation process. It was determined that the DI-MHD items retained their semantic integrity.
Cognitive Debriefing: To establish content validity, opinions were obtained from a panel of five experts (See Table 1).
Review of Cognitive Debriefing Results and Finalization: The expert panel analyzed the items for semantic, idiomatic, experiential, and conceptual equivalence, adapting specific words to align with Turkish culture.
Proofreading: The research team corrected grammatical errors and finalized the Turkish version of the scale.
Final Report: The final Turkish version was prepared for administration to patients to perform the subsequent psychometric analyses.
Table 1.
Expert opinion results (N = 5)
| Item | Very relevant + Relevant | I-CVI* | Interpretation | S-CVI** |
|---|---|---|---|---|
| Item 1 | 5 | 1 | Appropriate | 0.97 |
| Item 2 | 5 | 1 | Appropriate | |
| Item 3 | 4 | 0.80 | Appropriate | |
| Item 4 | 5 | 1 | Appropriate | |
| Item 5 | 5 | 1 | Appropriate | |
| Item 6 | 5 | 1 | Appropriate | |
| Item 7 | 5 | 1 | Appropriate | |
| Item 8 | 5 | 1 | Appropriate | |
| Item 9 | 5 | 1 | Appropriate | |
| Item 10 | 5 | 1 | Appropriate | |
| Item 11 | 4 | 0.80 | Appropriate | |
| Item 12 | 5 | 1 | Appropriate | |
| Item 13 | 5 | 1 | Appropriate | |
| Item 14 | 5 | 1 | Appropriate | |
| Item 15 | 5 | 1 | Appropriate | |
| Item 16 | 5 | 1 | Appropriate | |
| Item 17 | 5 | 1 | Appropriate |
* Item content validity index
**Scale Content Validity Index
Participants
A total of 283 patients who underwent hemodialysis in two hemodialysis units in Turkey from April to August 2022 were selected via convenience sampling. Patients aged 18 years or older who had been receiving hemodialysis for at least 3 months, could communicate in Turkish, and agreed to participate in the study were included. Patients who suffered from psychiatric disorders such as chronic psychosis and were previously diagnosed with depression were excluded from the study. The sample sizes of the studies were calculated using the formula specified for scale development studies. Accordingly, it is recommended that participants (5–30 times the number of scale items) be included in the sample to determine the sample size [28]. Since the scale comprises 17 items, the study should include between 65 and 510 patients. The study population consisted of 320 patients receiving treatment at both dialysis centers. Between the specified dates, 5 patients died, 10 patients had been previously diagnosed with depression, and 22 patients declined to participate in the study. For this reason, the sample comprised 283 patients who met the inclusion criteria and consented to participate in the study during the scheduled study dates.
Instruments
Data collection form for the descriptive characteristics of the participants
The researchers created this form after conducting a thorough review of the relevant literature [2, 3, 9, 13, 15, 18]. The form includes questions about the patients’ sociodemographic and clinical characteristics. It took the researchers approximately 10 min to complete the questionnaire.
Depression inventory for maintenance hemodialysis patients
The inventory, developed by Wang et al. [18], consists of 17 items. In this four-point Likert-type scale, items are rated as never: 1 point, sometimes: 2 points, often: 3 points, and almost always: 4 points. The minimum and maximum scores on the inventory are 17 and 68, respectively. Higher scores indicate greater fatigue severity. A score of 25 or more from the inventory indicates the presence of depression. At this cutoff point, the inventory’s sensitivity and specificity are 97% and 86%, respectively. The Cronbach’s alpha value of the inventory is 0.893. The inventory has four subscales: a low mood subscale: 7 items (7–28 points), an inhibition of movement and thought subscale: 4 items (4–16 points), a change in cognition subscale: 4 items (4–16 points), and a low self-evaluation subscale: 2 items (2–8 points). Higher scores indicate that the situation reflected by the subscale has worsened. Since we did not receive any response to the e-mails we sent to Wang and other authors requesting permission to use the inventory, we contacted the journal in which the article was published and obtained the necessary permission by completing the agreement form [18].
Beck depression inventory (BDI)
The inventory was developed by Beck et al. [29] to assess depressive symptoms and was adapted into Turkish by Hisli [30]. This four-point scale consists of 21 items. Each item is rated on a scale of 0 to 3. The lowest and highest scores of the inventory are points “0” and “63”, respectively. In its Turkish version, translated by Hisli [30], the cutoff point was 17, and the Cronbach’s alpha was 0.80. The severity of depressive symptoms is evaluated based on the total score. The degree of depression symptoms was scored as follows: 0 to 9 points: minimal depression; 10 to 16 points: mild depression; 17 to 29 points: moderate depression; and 30 to 63 points: severe depression [30]. The BDI is preferred because it is frequently used to assess depression in dialysis patients [5–7]. The questions are intended to be answered by considering the person’s situation from the previous week. In this study, the inventory’s Cronbach’s alpha was 0.858.
Inflammatory biomarkers
Depression has a detrimental effect on the health of patients undergoing hemodialysis, resulting in a decrease in quality of life, noncompliance with medication, frequent hospitalizations, and increased rates of suicide and mortality. These issues are associated with the physiological changes caused by depression, including elevated levels of inflammatory markers such as C-reactive protein (CRP) and proinflammatory cytokines, such as tumor necrosis factor-alpha and interleukin-6 [14, 31]. The serum albumin and CRP levels of the patients, measured at the month the study started, were obtained from the patients’ medical files and recorded by the investigators.
Data collection
The researchers collected data through face‒to-face interviews with patients during the 2nd hour of treatment, when acute complications related to hemodialysis treatment were less common. The researchers obtained data on the participants’ clinical characteristics from their patient files. The mean interview duration was 20 min.
Translation and cultural adaptation of the DI-MHD scale into Turkish
Language equivalence of the scale
Linguistic equivalence of the scale was established via the forward–backward translation procedure. For this purpose, the following steps were followed:
To ensure language validity, Turkish adaptation studies were carried out by four linguists. For this reason, a standardized guideline was followed during the adaptation of the scale from its original language to Turkish [32].
The translators were provided with detailed information on the measurement tool, terminology, and the target population to maintain the original meaning of each item throughout the translation process.
The original scale was independently translated into Turkish by two bilingual translators, resulting in an initial Turkish version.
This initial Turkish version was subsequently back-translated into English by two independent translators who were blinded to the original version.
The items were then reviewed by a Turkish-language expert (Supplementary File), taking into account their English and Turkish meanings, and finalized by the researchers [33]. Since we did not receive any response to the e-mails we sent to Wang and other authors requesting permission to use the inventory, we contacted the journal in which the article was published and obtained the necessary permission by completing the agreement form.
Expert panel: content validity
The content validity index (CVI) was calculated via the Davis technique to evaluate expert opinions [34] for the DI-MHD. The CVI is a measure used to assess the relevance of items in a scale or questionnaire. Content validity was assessed by an expert panel of five academics, each with ≥ 10 years of professional experience. Two experts were from psychiatric/mental health nursing and had experience in dialysis care, while three were from internal medicine nursing and had conducted research in the dialysis field. Experts were purposively selected based on their expertise relevant to the construct and clinical context, and they independently evaluated the items for content validity to calculate the CVI. In this case, five experts were sent both the original scale and its translation by e-mail and asked to rate each item according to its relevance. The rating scale ranged from 1 to 4, with the following interpretations: 1 = not relevant: the item is not applicable or does not measure what it intends to assess. 2 = item needs some revision: The item has potential but requires some modifications to be relevant. 3 = relevant but needs minor revision: The item is appropriate but may benefit from minor adjustments to improve its relevance. 4 = very relevant: The item is entirely relevant and accurately measures what it intends to assess. The item-level content validity index (I-CVI) for each item was calculated as the proportion of content experts who rated it as 3 or 4 among the total number of content experts who reviewed it. The I-CVI indicates the proportion of agreement on the relevance of each item and ranges from 0 to 1. Higher I-CVI scores indicate greater agreement among experts regarding the item’s relevance. An I-CVI of 1.0 indicates unanimous agreement among all the experts that the item is relevant [35]. In this particular study, the CVI of the DI-MHD scale was 0.97 (Table 1).
Psychometric testing of the DI-MHD scale
Construct validity
Construct validity was assessed via both exploratory and confirmatory factor analyses.
Exploratory factor analysis (EFA)
Exploratory factor analysis was conducted using data from 283 participants. A total of 17 items were included in the analysis. Sampling adequacy was assessed using the Kaiser–Meyer–Olkin (KMO) measure and Bartlett’s test of sphericity. The number of factors was initially determined by eigenvalues greater than 1.0 and by inspection of the scree plot. Extraction was performed using Principal Axis Factoring (PAF), and oblique rotation was applied.
Confirmatory factor analysis (CFA)
Confirmatory Factor Analysis was conducted to evaluate the one-factor model suggested by EFA, as well as the correlated four-factor and second-order models based on the original theoretical structure. Multivariate normality was evaluated via multivariate kurtosis and the critical ratio. Because the multivariate normality assumption was not met, the bootstrap procedure and the unweighted least squares (ULS) estimation method were used. Model fit was evaluated via x²/df, GFI, AGFI, NFI, and SRMR.
Criterion validity (hypothesis testing: BDI)
Criterion (concurrent) validity was examined via hypothesis testing via the Beck Depression Inventory (BDI) administered concurrently with the DI-MHD. A significant positive correlation between DI-MHD and BDI scores was expected and tested via Spearman correlation.
Reliability
Internal consistency
The internal consistency of the DI-MHD was evaluated via Cronbach’s alpha coefficient. In addition, item analysis was performed by examining corrected item–total correlations to assess each item’s contribution to the overall scale.
Test–retest reliability (temporal stability)
The temporal stability of the DI-MHD was evaluated using the test–retest method. A total of 264 patients were reassessed after 14 days. Reliability was assessed using the intraclass correlation coefficient (ICC, two-way mixed-effects model with absolute agreement). Measurement error and agreement were further examined using SEM, MDC95, and Bland–Altman analysis.
Statistical analysis
Data analyses were conducted via IBM SPSS version 22.0 and AMOS version 24.0 (IBM Corp., Armonk, NY). Descriptive statistics are reported as frequencies (n), percentages (%), means±standard deviations, and medians, as appropriate. The normality of the data distribution was assessed using the Kolmogorov-Smirnov test. The results indicated that the Beck Depression Inventory, Depression Inventory for Maintenance Hemodialysis Patients, CRP, and age data were not normally distributed (p < 0,05). Psychometric testing of the DI-MHD scale was performed to evaluate construct validity, criterion validity, and reliability. EFA was performed using Principal Axis Factoring (PAF) and, an oblique rotation method was applied. CFA was conducted to test the fit of the unidimensional model suggested by EFA, the correlated four-factor model, and the second-order factor model derived from the original theoretical structure. Prior to CFA, multivariate normality was evaluated via multivariate kurtosis and the critical ratio (CR). As the data did not meet the multivariate normality assumption required for maximum likelihood estimation, CFA was performed via the bootstrap procedure and the unweighted least squares (ULS) estimation method with a 95% confidence interval [36–38]. Criterion (concurrent) validity was examined through hypothesis testing via the Beck Depression Inventory (BDI) administered concurrently with the DI-MHD, and the relationship between the two scales was evaluated via Spearman’s product–moment correlation coefficient. Reliability was assessed by examining internal consistency via Cronbach’s alpha and corrected item–total correlations. Test–retest reliability was evaluated using the intraclass correlation coefficient (ICC) based on a two-way mixed-effects model with absolute agreement (ICC [2,1]). Measurement error was further quantified using the standard error of measurement (SEM = SD×√(1 − ICC)) and the minimal detectable change at the 95% confidence level (MDC95 = 1.96×√2×SEM). Agreement between test and retest scores was additionally examined using Bland–Altman analysis, including estimation of the mean difference (bias) and 95% limits of agreement (mean difference ± 1.96 × SD of the differences). Proportional bias was assessed using Pearson correlation between the mean and difference scores. p-value of < 0.05 was considered to indicate statistical significance.
Results
Participant characteristics
A total of 283 patients participated in the study; 50.1% were female, with a mean age of 61.32 ± 14.33 years, and 82% had an additional chronic disorder. The patients received HD treatment for 36 (3-1411) months (Table 2).
Table 2.
Descriptive characteristics of the participants (N = 283)
| Variables | n | % | |
|---|---|---|---|
| Age (years) |
61.32 ± 14.33 (min:19-max:90) Median:63 IQR:19 |
||
| Gender | Female | 143 | 50.5 |
| Male | 140 | 49.5 | |
| Educational status | Not literate | 32 | 11.3 |
| Literate | 16 | 5.7 | |
| Secondary | 152 | 53.7 | |
| High school and above | 83 | 29.3 | |
| Income status | Income exceeds expenditure | 25 | 8.8 |
| Income equal to expenditure | 213 | 75.3 | |
| Income less than expenditure | 45 | 15.9 | |
| Chronic disease | Yes | 232 | 82.0 |
| No | 51 | 18.0 | |
| The type of vascular route | Arteriovenous fistula | 221 | 78.1 |
| Central venous catheter | 59 | 20.8 | |
| Graft | 2 | 0.7 | |
| Missing data | 1 | ||
| Dialysis vintage (month) | 36 (min:3- max:1411) | ||
| CRP (mg/L) |
20.83 ± 20.53 (min:0.20-max:124.40) Median:11.10 IQR: 32.50 |
||
| Albumin (g/dL) |
3.33 ± 1.01(min:1- max:5.30) Median:3.67 IQR: 0.57 |
||
Data was shown as the n, %, mean ± standard deviation or median (IQR)
CRP; C-Reactive Protein
Construct validity
Exploratory factor analysis (EFA)
Table 3 presents the results of the exploratory factor analysis (EFA) conducted using Principal Axis Factoring with Oblimin rotation (N = 283). The Kaiser–Meyer–Olkin (KMO) measure was 0.885, indicating excellent sampling adequacy. Bartlett’s Test of Sphericity was statistically significant (x²=1812.58, p < 0.001), confirming that the correlation matrix was suitable for factor analysis. The unrestricted solution initially suggested a three-factor structure based on the Kaiser criterion (initial eigenvalues: 6.325, 1.557, and 1.194). However, inspection of the extracted eigenvalues revealed that only the first factor retained a substantial contribution to common variance (extracted eigenvalue = 5.784), accounting for 34.02% of the total variance. In contrast, the second and third factors showed markedly lower extracted eigenvalues (0.988 and 0.628) and explained only 5.81% and 3.69% of the variance, respectively. This sharp decline indicated a dominant first factor. Factor loadings on the first factor ranged from 0.373 to 0.792, demonstrating moderate to strong item–factor associations.
Table 3.
Explanatory factor analysis results of the depression inventory (N = 283)
| Sampling Adequacy | |||||
| Kaiser–Meyer–Olkin (KMO) | 0.885 | ||||
| Bartlett’s Test of Sphericity | x²=1812.58, p < 0.001 | ||||
| Factor | Eigenvalue | Explained variance (%) | Cumulative (%) | Items | Factor Loadings |
| Factor 1 | 5.784 | 34.02 | 34.02 | Item_7 | 0.792 |
| Item_9 | 0.746 | ||||
| Item_3 | 0.685 | ||||
| Item_13 | 0.650 | ||||
| Item_10 | 0.535 | ||||
| Item_1 | 0.509 | ||||
| Item_4 | 0.499 | ||||
| Item_5 | 0.382 | ||||
| Item_6 | 0.373 | ||||
| Factor 2 | 0.988 | 5.81 | 39.83 | Item_16 | 0.689 |
| Item_15 | 0.592 | ||||
| Item_12 | 0.532 | ||||
| Item_11 | 0.525 | ||||
| Item_14 | 0.418 | ||||
| Factor 3 | 0.628 | 3.69 | 43.53 | Item_8 | 0.602 |
| Item_17 | 0.550 | ||||
| Item_2 | 0.412 | ||||
Note: Extraction Method: Principal Axis Factoring, Rotation Method: Oblimin with Kaiser Normalization
Given the clear predominance of the first factor and the limited contribution of the remaining factors, the analysis was repeated with the number of factors constrained to one to evaluate the robustness of a unidimensional structure. In this constrained solution, the single factor yielded an initial eigenvalue of 6.325, explaining 37.21% of the total variance; after extraction, it accounted for 33.52% of the variance (extracted eigenvalue = 5.699). All 17 items loaded positively on the general factor, with loadings ranging from 0.387 to 0.767. Taken together, these findings supported the presence of a dominant single underlying construct. Based on this result, confirmatory factor analyses (CFA) were subsequently conducted to compare the unidimensional model with the theoretically proposed correlated four-factor and second-order models.
Confirmatory factor analysis (CFA)
Table 4 presents the comparison of three competing CFA models for the DI-MHD. The one-factor model showed good fit to the data (x²=52.122, df = 119, x²/df = 0.44, GFI=0.968, AGFI=0.959, SRMR=0.074, NFI=0.949, RFI=0.942). Standardized factor loadings ranged from 0.353 to 0.737 (see Table 5), indicating moderate to strong item-factor associations. The correlated four-factor model yielded slightly improved fit statistics (x²/df = 0.42; GFI=0.971; AGFI=0.960; SRMR=0.071; NFI=0.954; RFI=0.944). Standardized loadings ranged from 0.352 to 0.799. However, inter-factor correlations were relatively high (φ = 0.677–1.052), suggesting substantial shared variance among latent dimensions and raising concerns regarding discriminant validity. Similarly, the second-order model demonstrated comparable fit (x²/df = 0.42; GFI=0.970; AGFI=0.961; SRMR=0.072; NFI=0.953; RFI=0.944). Higher-order standardized loadings ranged from 0.708 to 1.041, indicating that a general higher-order construct strongly represented the four first-order factors.
Table 4.
Comparison of Competing CFA Models for the DI-MHD (N = 283)
| Model | x² (CMIN) | df | x²/df | GFI | AGFI | SRMR | NFI | RFI | Std. λ (range) | Std. γ/φ (range) |
|---|---|---|---|---|---|---|---|---|---|---|
| One-Factor | 52.122 | 119 | 0.438 | 0.968 | 0.959 | 0.074 | 0.949 | 0.942 | 0.353–0.737 | - |
| First Order (Correlated) | 47.697 | 113 | 0.422 | 0.971 | 0.960 | 0.071 | 0.954 | 0.944 | 0.352–0.799 | 677–1.052a |
| Second-Order | 48.324 | 115 | 0.420 | 0.970 | 0.961 | 0.072 | 0.953 | 0.944 | 0.353–0.799 | 0.708–1.041b |
λ = standardized factor loadings; a=standardized latent factor correlations; b=standardized higher-order loadings
Table 5.
Depression inventory (DI-MHD) confirmatory factor analysis: the one-factor model results
| Item | λ (Std.) | 95% CI | p |
|---|---|---|---|
| Item_7 | 0.737 | 0.642 – 0.796 | 0.021 |
| Item_9 | 0.710 | 0.605 – 0.777 | 0.018 |
| Item_10 | 0.674 | 0.593 – 0.755 | 0.012 |
| Item_13 | 0.644 | 0.545 – 0.714 | 0.020 |
| Item_11 | 0.632 | 0.541 – 0.696 | 0.016 |
| Item_1 | 0.630 | 0.524 – 0.707 | 0.012 |
| Item_3 | 0.627 | 0.501 – 0.732 | 0.009 |
| Item_2 | 0.569 | 0.426 – 0.641 | 0.023 |
| Item_12 | 0.580 | 0.485 – 0.662 | 0.012 |
| Item_6 | 0.549 | 0.438 – 0.637 | 0.012 |
| Item_5 | 0.545 | 0.429 – 0.644 | 0.007 |
| Item_4 | 0.543 | 0.455 – 0.634 | 0.013 |
| Item_14 | 0.502 | 0.400 – 0.588 | 0.016 |
| Item_17 | 0.475 | 0.348 – 0.565 | 0.023 |
| Item_15 | 0.464 | 0.329 – 0.568 | 0.013 |
| Item_16 | 0.420 | 0.294 – 0.532 | 0.012 |
| Item_8 | 0.353 | 0.220 – 0.477 | 0.009 |
Although the correlated four-factor and second-order models demonstrated acceptable global fit indices, the four-factor model’s latent factor correlations were high, suggesting limited discriminant validity among the dimensions. In the second-order model, the higher-order loadings were also substantial, suggesting that a single general construct largely explained the first-order factors. These findings indicate that the multidimensional structures did not provide clear added value beyond a general depression factor. Taken together, the dominant one factor observed in the EFA and the cumulative scoring structure of the DI-MHD support retaining the one-factor model as the final measurement model. The unidimensional solution adequately represents the construct without introducing unsupported multidimensional complexity.
Concurrent validity
A statistically significant positive correlation was found between the depression scores of the DI-MHD and Beck Depression Inventory (r = 0.734; p < 0.001). There were no statistically significant correlations between DI-MHD and CRP or albumin levels (p > 0.05).
Reliability
Internal consistency analysis
Internal consistency reliability of the DI-MHD and its subscales was assessed using Cronbach’s alpha. The corrected item-total correlations for the 17 inventory items range from 0.387 to 0.719. The Cronbach’s alpha for the scale was 0.888 (Table 6).
Table 6.
Item analysis and internal consistency of DI-MHD
| Item | Corrected Item-Total Correlation | Cronbach’s Alpha if item Deleted |
|---|---|---|
| Item 1 (Feeling distressed or in a low mood) | 0.622 | 0.735 |
| Item 2 (Social withdrawal (unwillingness to communicate with others) | 0.580 | 0.738 |
| Item 3 (Loss of interest in self-dressing) | 0.611 | 0.738 |
| Item 4 (Lack of motivation) | 0.557 | 0.738 |
| Item 5 (Feeling guilty or self-blame) | 0.560 | 0.739 |
| Item 6 (Feeling like an economic or psychological burden on family) | 0.566 | 0.738 |
| Item 7 (Feeling worthless or like a failure) | 0.719 | 0.735 |
| Item 8 (Having recurring thoughts of death, including suicidal ideation) | 0.387 | 0.747 |
| Item 9 (Feeling unable to accomplish anything) | 0.690 | 0.736 |
| Item 10 (Feeling pessimistic or disappointed) | 0.653 | 0.735 |
| Item 11 (Feeling the need to force yourself to focus on something) | 0.624 | 0.735 |
| Item 12 (Loss of interest) | 0.586 | 0.736 |
| Item 13 (Feeling lonely) | 0.629 | 0.732 |
| Item 14 (Crying easily) | 0.521 | 0.735 |
| Item 15 (Slowed movement and lags in response) | 0.491 | 0.738 |
| Item 16 (Memory impairment) | 0.449 | 0.741 |
| Item 17 (Feeling undeserving of others care) | 0.495 | 0.744 |
| Total Score | 1.000 | 0.888 |
Test-retest reliability
In Table 7, test–retest reliability was evaluated in 264 participants using a two-way mixed-effects model with absolute agreement (ICC [2,1]). The mean DI-MHD total score was 24.59 ± 6.95 at baseline and 23.90 ± 6.70 at retest. The mean difference between measurements was 0.69 ± 5.38, indicating minimal systematic bias. The single-measure ICC was 0.686 (95% CI: 0.617–0.745), reflecting moderate stability. The standard error of measurement (SEM) was 3.89, and the minimal detectable change at the 95% confidence level (MDC95) was 10.78 (Fig. 1).
Table 7.
Test–Retest Reliability and Measurement Error of the Turkish DI-MHD
| Measure | Test Mean ± SD | Retest Mean ± SD | Mean Difference (Bias) ± SD | 95% LoA (Lower to Upper) | ICC (2,1) | 95% CI | SEM | MDC95 |
|---|---|---|---|---|---|---|---|---|
| DI-MHD | 24.59 ± 6.95 | 23.90 ± 6.70 | 0.69 ± 5.38 | −9.86 to 11.23 | 0.686 | 0.617–0.745 | 3.89 | 10.78 |
ICC (2,1) = two-way mixed-effects model, absolute agreement (single measures)
SEM = SD × √(1 − ICC)
MDC95 = 1.96 × √2 × SEM
LoA (Limits of Agreement)=Mean difference ± 1.96 × SD of differences (Bland–Altman), representing the range within which 95% of individual test–retest differences are
Fig. 1.
Confirmatory factor analysis measurement model of the depression scale
Bland–Altman analysis (Fig. 2) demonstrated 95% limits of agreement ranging from − 9.86 to 11.23. Visual inspection of the plot showed a random distribution of differences around the mean bias line, with no evidence of proportional bias (p = 0.416). Overall, the DI-MHD demonstrated acceptable test–retest reliability at the group level in this population.
Fig. 2.
Bland–altman plot for test–retest agreement of the Turkish DI-MHD total score
Discussion
The findings of this study indicate that the Turkish version of the DI-MHD is a valid and reliable instrument for assessing patients undergoing hemodialysis treatment. Content validity refers to the extent to which a measurement tool adequately represents or includes all the features that are intended to be measured in each statement. In the context of this study, Özdemir [39] defines content validity as ensuring that the Turkish version of the DI-MHD includes all relevant aspects to be measured. To evaluate content validity, the Davis technique was used. The CVI value is then calculated by dividing the number of experts selecting category (a) or (b) by the total number of experts [34]. A CVI value of 0.80 for the scale is desired [40]. In this study, a CVI value of 0.97 was obtained, indicating that the DI-MHD is considered appropriate for Turkish culture and has content validity.
The internal consistency of a measurement tool is typically assessed using Cronbach’s alpha [41]. This coefficient provides information about how well the items in the tool are measuring the same construct. Different classifications are available for interpreting the Cronbach’s alpha coefficient. A widely accepted approach is that a Cronbach’s alpha coefficient of 0.70 and above is considered desirable [42]. This finding indicates that the items in the measurement tool reliably measure the intended construct. According to Tavşancıl [43] and Özdamar [44], a Cronbach’s alpha coefficient in the range of 0.60–0.79 is a reliable measurement tool. This shows that the items are moderately consistent in measuring the construct. For even higher reliability, a Cronbach’s alpha coefficient in the range of 0.80-1.00 is considered to indicate high reliability [44]. This implies that the items are highly consistent in measuring the intended construct. The Cronbach’s alpha coefficient of the scale developed by Wang et al. [18] was 0.893, indicating high internal consistency [18]. Similarly, the Turkish version of the DI-MHD, used in this study, yielded a Cronbach’s alpha coefficient of 0.888. This value is also quite high, suggesting that the items in the DI-MHD are reliable and consistently measure depression in patients undergoing hemodialysis treatment in the Turkish cultural context.
The reliability of a measurement tool is considered an indicator of its consistency and accuracy [45]. To assess reliability, each item in the measurement tool is analyzed, and decisions are made regarding whether to keep or remove it from the scale. The Cronbach’s alpha coefficient is recalculated when item removal is considered. If the total scale maintains a high reliability coefficient even after an item is removed, that item is eliminated to avoid any negative impact on the measurement. A common approach currently involves ensuring that the difference in correlation coefficients between the specific item and the total score of the remaining items is non-negative and less than 0.25 [44, 46]. The DI-MHD comprises 17 items, and their correlations with the total score range from 0.387 to 0.719. As a result of this analysis, the 17 DI-MHD items were deemed reliable.
To conduct a factor analysis to assess the structural validity of a sample, certain criteria must be met. According to Berkün [47], the KMO test value should be above 0.60, and Bartlett’s test of sphericity should yield a statistically significant result. The KMO value ranges from 0.70 to 0.80 for moderate sample adequacy, and higher values indicate excellent sample adequacy [43]. In this study, the KMO value for the scale was 0.885, indicating excellent sample adequacy. Additionally, Bartlett’s test of sphericity was statistically significant, confirming the adequacy of the correlation matrix for the scale items in conducting a factor analysis. After confirming the data’s adequacy for factor analysis, the single-factor solution yielded an initial eigenvalue of 6.325, accounting for 37.21% of the total variance. After extraction, the factor explained 33.52% of the variance. Although the original version proposed a four-factor conceptual structure, the unrestricted EFA conducted did not yield clearly distinct and interpretable multidimensional factors. A dominant factor emerged, and the factor correlation matrix indicated moderate to high associations among the extracted factors. Therefore, the EFA was re-run under a one-factor constraint to examine the adequacy of a unidimensional structure. Notably, a variance of 30% and above is considered adequate for single-factor scales [48].
Subsequently, confirmatory factor analyses were conducted to compare the one-factor model suggested by the EFA with the original correlated four-factor model and a second-order factor model. The multidimensional CFA solutions did not clearly demonstrate separable latent dimensions, and the findings indicated that a single underlying factor largely accounted for the covariance structure among the items. Although the original scale includes multiple subdimensions, current evidence supports its unidimensional use based on the total score. Subscale scores may be used descriptively; however, they should be interpreted with caution due to their limited discriminant validity.
The test‒retest approach is commonly used to assess the consistency of a measurement tool over time [38]. To determine the reliability of the test‒retest method, researchers often look at the ICC, which reflects the consistency between the results of the initial test and the retest. In the literature, an ICC value of 0.80 is considered necessary for a reliable measurement tool [49]. In this study, the ICC was 0.686, indicating a strong, statistically significant relationship between the participants’ scale scores at the first and repeat administrations. These results demonstrate that the DI-MHD scale used in the study was consistent over time, making it a reliable measurement tool. Importantly, the developers of the DI-MHD scale, Wang et al. [18], did not evaluate the test-retest reliability of the scale themselves, according to the study. However, the researchers in the current study demonstrated the scale’s reliability through their test‒retest evaluation and the resulting ICC value.
Criterion validity is a measure of how well a new measurement tool correlates with an established criterion or gold standard. According to Şencan [49], a correlation coefficient of 0.30 or higher is generally considered necessary to ensure criterion validity. In the present study, the researchers conducted criterion-related analyses and reported a statistically significant correlation between the DI-MHD and BDI scores (r = 0.734, p < 0.001). Since the correlation coefficient is well above the 0.30 threshold, the DI-MHD measurement tool shows good criterion validity relative to the BDI score. The developers of the DI-MHD, Wang et al. [18], also used the BDI score to evaluate the criterion validity of the scale, reporting a correlation coefficient of 0.785. These results further support the criterion validity of the DI-MHD scale, demonstrating a strong relationship with the BDI score, an established criterion. As emphasized by Wang et al. [18], the quick and easy administration of the DI-MHD demonstrates its superiority over other screening scales. Although the DI-MHD demonstrated strong psychometric properties in the present study, it should be emphasized that the instrument reflects relative levels of depressive symptom severity rather than providing a definitive clinical diagnosis. In the original validation study, Wang et al. [18] established the clinical cutoff value of the DI-MHD. In contrast, the present study did not aim to determine a diagnostic threshold or to validate cutoff scores against clinical diagnoses. Therefore, the scores obtained from the DI-MHD should be interpreted as indicative of relative depression burden among patients undergoing hemodialysis rather than as diagnostic evidence of major depressive disorder.
Studies have shown different results regarding the correlation between serum albumin and CRP levels and depression [16]. In this study, no statistically significant correlation was found between serum albumin and CRP levels and DI-MHD. Chronic kidney disease is a proinflammatory condition, and CRP concentrations are often increased in hemodialysis patients [50, 51]. In the studies included by Gregg et al. [16] in their meta-analysis, they reported a correlation between low serum albumin levels and depression. However, the heterogeneous results of these studies likely reflect, in part, small sample sizes. Larger studies are more likely to detect differences than smaller studies may have been underpowered to demonstrate [16].
Strengths and limitations
Indeed, the study has several notable strengths that contribute to its significance and reliability. One strength of the study is that it introduced, for the first time in the country, a measurement tool specifically tailored to the DI-MHD. By developing and using a measurement tool to assess DI-MHD symptoms, this study addresses a gap in existing research and clinical practice, providing a valuable contribution to kidney disease management. Another strength of the study is the use of the test‒retest method to evaluate the consistency of the DI-MHD scale over time. By conducting a follow-up assessment with the same participants, the researchers were able to assess the stability of the scale’s results over time, demonstrating its reliability and consistency in capturing DI-MHD symptoms.
Additionally, this is a notable addition to the research, given the lack of such an evaluation in the original study by Wang et al. [18]. The study’s inclusion of two different dialysis centers in two different cities is another strength. This multicenter approach enhances the generalizability and external validity of the findings. By including participants from diverse geographic locations, this study provides more comprehensive insights into DI-MHD symptoms and their measurement in various dialysis settings. This study is limited in several ways. The first limitation is that depression was not diagnosed by a psychiatrist. In this study, the DI-MHD was not evaluated against a psychiatrist-confirmed clinical diagnosis of depression. The study sample was drawn from private dialysis centers. Differences in access to care, educational level, and economic status may influence how depressive symptoms are perceived and reported. In our country, within the scope of chronic disease management, patients can also benefit from private healthcare services under general health insurance coverage.
Consequently, the scale’s diagnostic utility could not be assessed. In this context, the DI-MHD should be regarded as a screening measure capable of capturing relative depressive symptom levels, rather than a tool for establishing clinical diagnoses. Future studies should therefore include psychiatrist-confirmed diagnostic assessments as a reference standard to examine the diagnostic performance of the DI-MHD. Conducting such diagnostic validation studies, including receiver operating characteristic (ROC) analyses, would enable the determination of population-specific cutoff values. This approach would clearly distinguish the screening function of the DI-MHD from its potential diagnostic role, thereby strengthening its clinical interpretability in hemodialysis populations. In future cross-cultural adaptation studies, we recommend including patients diagnosed with depression to demonstrate the scale’s discriminative power as a screening tool. In this context, it is essential to establish the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the DI-MHD. The scale contains a large number of somatic symptoms, which may make it difficult to differentiate between chronic somatic disease and psychiatric comorbidity; moreover, the use of only the BDI to assess depression represents a second limitation. A third limitation could be the development of depression as a side effect of the medications used to treat chronic diseases in our patients. The fourth limitation of this study is the temporal gap between the collection of biological measures and the administration of the DI-MHD scale. Future studies could benefit from synchronized timing to better capture the immediate relationship between acute inflammatory spikes and fluctuating depressive symptoms. Future studies with longitudinal designs and synchronized data collection are needed to better clarify the relationship between inflammatory markers and depressive symptoms in hemodialysis patients. Finally, the study’s results can be generalized to dialysis centers where the study was conducted.
Conclusion
The present study provides evidence that the Turkish version of the DI-MHD demonstrates adequate psychometric properties and supports a unidimensional (one-factor) structure in patients undergoing maintenance hemodialysis. Based on both exploratory and confirmatory factor analyses, the one-factor model was retained as the most parsimonious and conceptually coherent measurement model. As a screening tool, its ease of use makes it suitable for routine administration to dialysis patients, enabling early recognition of symptoms and facilitating better symptom management. However, to further validate the scale’s applicability and generalizability to all patients undergoing maintenance hemodialysis, it would be beneficial to conduct similar studies with more diverse and representative ethnicities. Investigating the scale’s performance across different populations can provide valuable insights into its robustness and applicability in various cultural and ethnic contexts. By using valid and reliable measurement tools such as the DI-MHD scale, healthcare professionals can effectively assess and manage symptoms in patients undergoing hemodialysis. Timely recognition and management of symptoms can improve patient outcomes and quality of life for individuals undergoing hemodialysis. In hemodialysis patients, many factors can lead to the emergence of depressive symptoms. When evaluating patients with the DI-MHD scale, it is also recommended to address conditions such as cerebrovascular disease, inflammation, and malnutrition, which can contribute to the development of these depressive symptoms.
As research in this area continues to evolve, it will contribute to the overall advancement of care for individuals with chronic kidney disease and those undergoing maintenance hemodialysis.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
Concept – NO, VU; Design – NO, VU; Supervision – NO; Resources – NO, VU, CNS, TC, BS; Materials - NO, VU; Data Collection and/or Processing – TC, BS, UU; and/or Interpretation – NO, VU, UU; Literature Search - NO; Writing Manuscript - NO, VU, CNS; Critical Review – All of the authors.
Funding
No.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
Since we did not receive any response from the e-mails we had sent to Wang and other authors for permission to use the inventory, we contacted the journal in which the article was published and obtained the necessary permission by filling out the agreement form. After obtaining ethical approval from the Acıbadem Mehmet Ali Aydınlar University Medical Research Ethics Committee (ATADEK) (decision date: 03/25/2022; decision no: 2022-06/19) and the required permissions from the participating dialysis centers, all eligible patients were approached prior to data collection. The study purpose, procedures, and the voluntary nature of participation (including the right to withdraw at any time) were explained in plain language, and participants were given the opportunity to ask questions. Written informed consent was then obtained from all participants before enrollment. This study was conducted in accordance with the principles of the Declaration of Helsinki.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Farrokhi F, Abedi N, Beyene J, Kurdyak P, Jassal SV. Association between depression and mortality in patients receiving long-term dialysis: a systematic review and meta-analysis. Am J Kidney Dis. 2014;63(4):623–35. [DOI] [PubMed] [Google Scholar]
- 2.Palmer S, Vecchio M, Craig JC, Tonelli M, Johnson DW, Nicolucci A, et al. Prevalence of depression in chronic kidney disease: systematic review and meta-analysis of observational studies. Kidney Int. 2013;84(1):179–91. [DOI] [PubMed] [Google Scholar]
- 3.Schouten RW, Harmse VJ, Dekker FW, van Ballegooijen W, Siegert CEH, Honig A. Dimensions of Depressive Symptoms and Their Association With Mortality, Hospitalization, and Quality of Life in Dialysis Patients: A Cohort Study. Psychosom Med. 2019;81(7):649–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Weisbord SD, Mor MK, Sevick MA, Shields AM, Rollman BL, Palevsky PM, et al. Associations of depressive symptoms and pain with dialysis adherence, health resource utilization, and mortality in patients receiving chronic hemodialysis. Clin J Am Soc Nephrol. 2014;9(9):1594–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ekramzadeh M, Mazloom Z, Sagheb M. Association of Depression With Selenium Deficiency and Nutritional Markers in the Patients With End-Stage Renal Disease on Hemodialysis. J Ren Nutr. 2015;25(4):381–7. [DOI] [PubMed] [Google Scholar]
- 6.Fan L, Sarnak MJ, Tighiouart H, Drew DA, Kantor AL, Lou KV, et al. Depression and all-cause mortality in hemodialysis patients. Am J Nephrol. 2014;40(1):12–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jong IC, Tsai HB, Lin CH, Ma TL, Guo HR, Hung PH, et al. Close correlation between the ankle-brachial index and symptoms of depression in hemodialysis patients. Int Urol Nephrol. 2017;49(8):1463–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hackett ML, Jardine MJ. We Need to Talk about Depression and Dialysis: but What Questions Should We Ask, and Does Anyone Know the Answers? Clin J Am Soc Nephrol. 2017;12(2):222–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Goh ZS, Griva K. Anxiety and depression in patients with end-stage renal disease: impact and management challenges - a narrative review. Int J Nephrol Renovasc Dis. 2018;11:93–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhang S, Liu SX, Wu QJ, Wang ZH, Liu H, Xiao P, et al. Association between handgrip strength and depressive symptoms in patients undergoing hemodialysis: a cross-sectional study from a single Chinese center. BMC Psychiatry. 2024;24(1):182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ye W, Wang L, Wang Y, Wang C, Zeng J. Depression and anxiety symptoms among patients receiving maintenance hemodialysis: a single center cross-sectional study. BMC Nephrol. 2022;23(1):417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Saglimbene V, Palmer S, Scardapane M, Craig JC, Ruospo M, Natale P, et al. Depression and all-cause and cardiovascular mortality in patients on haemodialysis: a multinational cohort study. Nephrol Dial Transpl. 2017;32(2):377–84. [DOI] [PubMed] [Google Scholar]
- 13.Kondo K, Antick JR, Ayers CK, Kansagara D, Chopra P. Depression Screening Tools for Patients with Kidney Failure: A Systematic Review. Clin J Am Soc Nephrol. 2020;15(12):1785–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Li Y, Zhu B, Shen J, Miao L. Depression in maintenance hemodialysis patients: What do we need to know? Heliyon. 2023;9(9):e19383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pearce CJ, Hall N, Hudson JL, Farrington K, Tucker MJR, Wellsted D, et al. Approaches to the identification and management of depression in people living with chronic kidney disease: A scoping review of 860 papers. J Ren Care. 2024;50(1):4–14. [DOI] [PubMed] [Google Scholar]
- 16.Gregg LP, Carmody T, Le D, Martins G, Trivedi M, Hedayati SS. A Systematic Review and Meta-Analysis of Depression and Protein-Energy Wasting in Kidney Disease. Kidney Int Rep. 2019;5(3):318–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chilcot J, Wellsted D, Da Silva-Gane M, Farrington K. Depression on dialysis. Nephron Clin Pract. 2008;108(4):c256–64. [DOI] [PubMed] [Google Scholar]
- 18.Wang YY, Zhang WW, Feng L, Gao D, Liu C, Zhong L, et al. Development and Preliminary Validation of a Depression Assessment Tool for Maintenance Hemodialysis Patients. Ther Apher Dial. 2019;23(1):49–58. [DOI] [PubMed] [Google Scholar]
- 19.Fan N, Luo Y, Ou Y, He H. Altered serum levels of TNF-α, IL-6, and IL-18 in depressive disorder patients. Hum Psychopharmacol. 2017;32(4). [DOI] [PubMed]
- 20.Haapakoski R, Mathieu J, Ebmeier KP, Alenius H, Kivimäki M. Cumulative meta-analysis of interleukins 6 and 1β, tumour necrosis factor α and C-reactive protein in patients with major depressive disorder. Brain Behav Immun. 2015;49:206–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pasco JA, Nicholson GC, Williams LJ, Jacka FN, Henry MJ, Kotowicz MA, et al. Association of high-sensitivity C-reactive protein with de novo major depression. Br J Psychiatry. 2010;197(5):372–7. [DOI] [PubMed] [Google Scholar]
- 22.Loosman WL, Siegert CE, Korzec A, Honig A. Validity of the Hospital Anxiety and Depression Scale and the Beck Depression Inventory for use in end-stage renal disease patients. Br J Clin Psychol. 2010;49(Pt 4):507–16. [DOI] [PubMed] [Google Scholar]
- 23.Preljevic VT, Østhus TB, Sandvik L, Opjordsmoen S, Nordhus IH, Os I, et al. Screening for anxiety and depression in dialysis patients: comparison of the Hospital Anxiety and Depression Scale and the Beck Depression Inventory. J Psychosom Res. 2012;73(2):139–44. [DOI] [PubMed] [Google Scholar]
- 24.Cruchinho P, López-Franco MD, Capelas ML, Almeida S, Bennett PM, Miranda da Silva M, et al. Translation, cross-cultural adaptation, and validation of measurement instruments: A practical guideline for novice researchers. J Multidiscip Healthc. 2024;17:2701–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kottner J, Audige L, Brorson S, Donner A, Gajewski BJ, Hróbjartsson A, et al. Guidelines for Reporting Reliability and Agreement Studies (GRRAS) were proposed. Int J Nurs Stud. 2011;48(6):661–71. [DOI] [PubMed] [Google Scholar]
- 26.Gagnier JJ, Lai J, Mokkink LB, Terwee CB. COSMIN reporting guideline for studies on measurement properties of patient-reported outcome measures. Qual Life Res. 2021;30(8):2197–218. [DOI] [PubMed] [Google Scholar]
- 27.Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee-Lorenz A, et al. Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and Cultural Adaptation. Value Health. 2005;8(2):94–104. [DOI] [PubMed] [Google Scholar]
- 28.Yong AG, Pearce S. A beginner’s guide to factor analysis: Focusing on exploratory factor analysis. Tutorials Quant Methods Psychol. 2013;9(2):79–94. [Google Scholar]
- 29.Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561–71. [DOI] [PubMed] [Google Scholar]
- 30.Hisli N. Beck Depresyon Envanteri’nin üniversite öğrencileri için geçerliği, güvenirliği. Psikoloji Dergisi. 1989;7(23):3–13. [Published in Turkish]. [Google Scholar]
- 31.Lotrich FE. Inflammatory cytokine-associated depression. Brain Res. 2015;1617:113–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sousa VD, Rojjanasrirat W. Translation, adaptation and validation of instruments or scales for use in cross-cultural health care research: a clear and user-friendly guideline. J Eval Clin Pract. 2011;17(2):268–74. [DOI] [PubMed] [Google Scholar]
- 33.Hançer M. Ölçeklerin yazım dilinden başka bir dile çevirileri ve kullanılan değişik yaklaşımlar. Balıkesir Üniversitesi Sosyal Bilimler Enstitüsü Dergisi. 2003;6:7–61. [Published in Turkish]. [Google Scholar]
- 34.Davis LL. Instrument review: Getting the most from a panel of experts. Appl Nurs Res. 1992;5:194–7. [Google Scholar]
- 35.Zamanzadeh V, Rassouli M, Abbaszadeh A, Majd HA, Nikanfar A, Ghahramanian A. Details of content validity and objectifying it in instrument development. Nurs Pract Today. 2014;1(3):163–71. [Google Scholar]
- 36.Karagöz Y. SPSS 23 ve AMOS 23 Uygulamalı İstatistiksel Analizler: Nobel Akademik Yayıncılık. 2019. [Published in Turkish].
- 37.Meydan CH, Şeşen H. Yapısal Eşitlik Modellemesi AMOS Uygulamaları. Detay Yayıncılık. 2011. [Published in Turkish].
- 38.Alpar R, Spor. Sağlık ve Eğitim Bilimlerinden Örneklerle Uygulamalı İstatistik ve Geçerlik-Güvenirlik. Detay Yayıncılık. 2010. [Published in Turkish].
- 39.Özdemir Z. Sağlık Bilimlerinde Likert Tipi Tutum Ölçeği Geliştirme. Hacettepe Üniversitesi Hemşirelik Fakültesi Dergisi. 2018;5:60–8. [Published in Turkish]. [Google Scholar]
- 40.Esin MN. Veri Toplama Araçlarının Güvenirlik ve Geçerliği. In: S. Erdoğan, N. Nahcivan, & M. N. Esin, editors. Hemşirelikte Araştırma Süreç, Uygulama ve Kritik. (pp. 216–229). Nobel Tıp Kitabevleri. 2014. [Published in Turkish].
- 41.LoBiondo-Wood G, Haber J. Realibility and Validity. In: LoBiondo-Wood G, Haber J, editors. Nursing Research. 8th ed. Elsevier Health Sciences; 2014. pp. 289–309.
- 42.George D, Mallery P. SPSS for Windows step by step: A simple guide and reference. 11.0 update. 4th ed. Allyn & Bacon; 2003.
- 43.Tavşancıl E. Tutumların Ölçülmesi ve SPSS İle Veri Analizi. Nobel Yayınları. 2010. [Published in Turkish].
- 44.Özdamar K. Paket Programlar İle İstatistiksel Veri Analizi. (4th ed.). Kaan Kitabevi. 2002. [Published in Turkish].
- 45.Koçak C, Albayrak SA, Duman NB. Hemşirelerin Bakım Verici Rollerine İlişkin Tutum Ölçeği Geliştirilmesi: Geçerlik ve Güvenirlik Çalışması. Hemşirelikte Eğitim ve Araştırma. Dergisi. 2014;11:16–21. [Published in Turkish]. [Google Scholar]
- 46.Yıldırım A, Şimşek H. Sosyal Bilimlerde Nitel Araştırma Yöntemleri (6th ed.). Seçkin Yayıncılık. 2008. [Published in Turkish].
- 47.Berkün S. Sosyal hizmet uzmanlarının çalışma hayatında karşılaştıkları mesleki sorunlar: Bursa örneği. Toplum ve Sosyal Hizmet Dergisi. 2010;21:99–109. [Published in Turkish]. [Google Scholar]
- 48.Ural A, Kılıç İ. Bilimsel Araştırma Süreci ve SPSS ile Veri Analizi (2nd Extended Ed.). Ankara: Detay Yayıncılık. 2006. [Published in Turkish].
- 49.Şencan H. Sosyal ve Davranışsal Ölçümlerde Güvenilirlik ve Geçerlilik. Seçkin Yayıncılık Sanayi ve Ticaret AŞ. 2005. [Published in Turkish].
- 50.Chilcot J, Friedli K, Guirguis A, Wellsted D, Farrington K, Davenport A. C reactive protein and depressive symptoms in hemodialysis patients: A questionable association. Hemodial Int. 2017;21(4):542–8. 10.1111/hdi.12500. [DOI] [PubMed] [Google Scholar]
- 51.Lee BT, Ahmed FA, Hamm LL, Teran FJ, Chen CS, Liu Y, et al. Association of C-reactive protein, tumor necrosis factor-alpha, and interleukin-6 with chronic kidney disease. BMC Nephrol. 2015;16:77. 10.1186/s12882-015-0068-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.