Introduction

Chronic kidney disease (CKD) is a common disease affecting about 700 million people globally [1]. It is a significant cause of mortality worldwide [2, 3]. The burden of CKD is higher in low- and middle-income countries than in high-income countries [3]. It causes a substantial economic, physical, and psychosocial burden for the patient [4, 5]. Psychosocial problems such as depression and anxiety are more commonly encountered in patients with CKD and their caregivers compared to the general population [6, 7]. These problems adversely affect the quality of life and contribute to deterioration in kidney function, progression to an advanced stage of disease, hospitalization, and mortality in individuals with CKD [8, 9]. Depression may contribute to poor adherence to medications, fluid, and dietary prescriptions of individuals with CKD [10, 11]. It also plays a pivotal role in the development of cardiovascular disease in CKD [12]. Psychosocial problems adversely affect the quality of life of individuals with CKD and contribute to overall poor outcomes of CKD [6, 13].

Despite the significant impact of psychosocial problems on the overall outcomes of CKD, they have not been given adequate priority in the management of individuals with CKD. Depression is not routinely screened for and managed in these patients, and mental health professionals were rarely mentioned in their multidisciplinary management [14]. Although there is inertia in the use of medications to manage psychosocial problems such as depression in CKD due to safety concerns by most clinicians, evidence supporting the safety of some medications and associated beneficial effects is now emerging [15]. In addition, non-pharmacological therapy, such as psychosocial interventions, has been established to improve the quality of life and reduce depression and anxiety in individuals with CKD [16]. Routine evaluation for common psychosocial problems, especially depression in this at-risk group, will aid in prompt diagnosis and allow those affected to be managed with consequent improvement in their overall outcomes.

This study, an update of an earlier review [17], assessed the prevalence of clinical depression in the global chronic kidney disease population. In addition, we reported the prevalence of depression in the pediatric CKD population and documented the geographic differences and temporal trends in depression prevalence in the CKD population. The findings of the study will provide helpful information on the burden of depression and also serve as empirical evidence to include its routine evaluation and treatment in the management of CKD, especially in countries where this is not routinely done.

Methods

This systematic review and meta-analysis includes studies that document the prevalence of clinical depression among patients with CKD. The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guideline [18] was used in reporting this study (Supplementary Table 1). This study was registered with PROSPERO with registration number CRD42022382708.

Literature search

A systematic literature search was done on published articles involving individuals with CKD who had clinically diagnosed depression. We performed a search on PubMed, African Journal Online, and Embase using terms related to depression in chronic kidney disease such as "chronic kidney disease", "CKD", "chronic renal failure", "CRF", "end-stage renal disease, "ESRD", "end-stage kidney failure", "ESKF", "dialysis-dependent chronic kidney disease", "renal failure", " non-dialysis dependent chronic kidney disease", "pre-dialysis chronic kidney disease", "chronic renal insufficiency", "dialysis population" "depression", "depressive symptom", mental health disorder ", psychiatric disorder' in conjunction with all the names of continents. The search strategy, created in April 2023, is detailed in Supplementary Tables 2A and 2B. There was no language limitation. A hand search of the reference list of articles of interest was also undertaken.

Study selection

Inclusion criteria were cross-sectional and prospective studies reporting on clinically diagnosed depression in adult and pediatric populations and pre-dialysis and dialysis populations across all the World Health Organization (WHO) regions. Exclusion criteria were studies that used screening tools to diagnose depression because it has been shown that the latter overestimate depression prevalence [19]; studies that involved kidney transplant patients; abstract papers on depression in CKD without full text; studies that determined clinical diagnosis of depression but did not clearly define the study population in terms of those on peritoneal dialysis (PD), hemodialysis (HD) or who had transplant; and randomized controlled trial studies on depression in CKD. Two investigators (OAA and OOI) independently screened records for eligibility based on titles and abstracts. Full texts of articles deemed potentially eligible were retrieved and screened by the same investigators (OAA and OOI) for final inclusion. All conflicts were resolved by a third investigator (OEY).

Data extraction and management

The following variables were extracted from selected studies: the last name of the first author, year of publication and country and continent in which the study was carried out, sample size of the study, duration of the study, study design, mean age of the study participants, stage of CKD, type of renal replacement therapy, the proportion of subjects with depression, clinical depression diagnostic criteria used [Diagnostic Statistical Manual for Mental Disorders III (DSM III), DSM IV, DSM V, International Classification of Disease-9 (ICD-9), ICD-10], and severity of depression. The data extraction form was developed with input from all the investigators, including Mental Health and Nephrology specialists. The data were extracted by two different investigators, and a third investigator resolved areas of conflict. We categorized study location based on the WHO regions [(African Region, AFR), (Region of the Americas, AMR), (South East Asian Region, SEAR), (European Region, EUR), (Eastern Mediterranean Region, EMR), and (Western Pacific Region, WPR)]. Studies were also categorized based on the year of publication of the study, severity of CKD and dialysis modality.

Methodological quality

The Joanna-Briggs Institute Critical Appraisal Checklist for Studies Reporting Prevalence Data was used to assess the methodological quality of the constituent studies [20]. Studies scored 1 for each of the nine questions with a "yes" response. Studies with scores 0–3 were regarded as poor quality, 4–6 as intermediate or medium quality, and 7–9 as high quality. Two different investigators did the quality assessment, and areas of disagreement were resolved by consensus.

Ethical consideration

Ethical approval was not required. The study protocol was registered with PROSPERO (CRD42022382708).

Statistical analysis

Stata 17.0 (Stata Corp., 2021. Stata Statistical Software: Release 17, College Station, TX) was used for statistical analysis. The pooled prevalence of depression in the global CKD population was determined using meta-analytic techniques. The study-specific estimates derived from the DerSimonian-Laird random effects model [21] were pooled to estimate the prevalence of depression in this population. To minimize the effect of extreme values, the Freeman-Tukey double arcsine transformation [22] was used to stabilize the individual study variances before using the random effects model to obtain the pooled estimates. Publication bias was assessed using the Egger test [23]. We also undertook a subgroup analysis of pooled prevalence by continent, HD versus PD population, pre-dialysis versus dialysis CKD population, pediatric versus adult population with CKD, among studies that used different clinical diagnostic criteria such as DSM III, DSM IV, DSM V, ICD-9 and ICD-10 and among studies that fell into different range of years. Subgroup analysis was performed using the Q-test based on ANOVA. The I2 statistic was used to determine the between-study heterogeneity.

Results

Study selection and characteristics

The systematic literature search identified 9240 articles, of which 137 were selected for full-text review after duplicate removal and title and abstract screening. Finally, 65 articles [24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88] were eligible and therefore included in this systematic review (Fig. 1), with publication years ranging from 1987 to 2023. Included studies reported on 80,932 CKD and kidney failure patients from 27 countries. There were 249 participants (3 studies) from the African Region [32, 60, 72]; 4,295 (14 studies) from the European Region [33, 35, 36, 38, 41, 45, 58, 63, 68, 70, 77, 80, 82, 84]; 4,160 (19 studies) from the Region of the Americas [24,25,26,27,28, 31, 34, 37, 40, 47, 55, 57, 64, 67, 69, 71, 79, 85, 87]; 679 (7 studies) from the Eastern Mediterranean Region [42, 51, 53, 62, 66, 76, 83]; 846 (7 studies) from the South East Asian Region [30, 49, 52, 61, 73, 75, 81], and 70,703 (15 studies) from the the Western Pacific Region [29, 39, 43, 45, 46, 48, 50, 54, 56, 59, 65, 74, 78, 86, 88]. Figure 2 shows the geographical distribution of the included studies.

Fig. 1
figure 1

PRISMA flow diagram—Identification and Screening of Articles

Full size image
Fig. 2
figure 2

Geographical distribution of included studies

Full size image

The overwhelming majority of the studies (60 studies, 92.3%) used the DSM clinical diagnostic criteria or one of its derivatives, while five studies used the ICD criteria. The sample size of the component studies ranged from 20 [72] to 67,866 [39] patients, with the proportion of women ranging from 0% [27] to 68% [76]. The mean age of the participants ranged from 11.0 years [62] to 76.3 years [79]

Most of the studies had medium methodological quality (70.8%, n = 46) (Table 1); 14 studies (21.5%) were of high quality, including three studies from the Americas [25, 27, 79], one study from the African region [32], two studies from the Western Pacific region [44, 77], three studies from the Eastern Mediterranean region [66, 76, 83], three from the European region [63, 77, 82] and two studies from South East Asia [73, 81]. Tables 1 and 2 summarize the data extracted from the constituent articles and the sub-populations studied. The majority (39, 60.0%) of the articles reported on severe or major depressive illness only; 5 (7.7%) reported on the whole spectrum of depression; 2 (3.1%) reported only on mild or moderate depression, while 15 (23.1%) did not state depression severity.

Table 1 Summary of Extracted Data from Included Articles
Full size table
Table 2 Summary of sub-population characteristics and pooled prevalence
Full size table

Prevalence of depression

The overall pooled prevalence of depression was [26.5% 95% CI 23.1–30.1%), N = 65 studies, I2 = 97.2%, p < 0.001 for heterogeneity] irrespective of the clinical depression diagnostic criteria used. Studies using DSM and ICD returned a pooled prevalence of [25.5% (21.3–30.0%), n = 60 studies, I2 = 95.9%, p < 0.001 for DSM] versus [39.6% (95% CI 28.1–51.7%), n = 5 studies, I2 = 98.7%, p < 0.001 for ICD], p = 0.03 for difference between the different diagnostic criteria (Fig. 3). The p-value for the Egger test was 0.83, suggesting no small study effects.

Fig. 3
figure 3

Pooled prevalence of depression in the CKD population by diagnostic criteria

Full size image

The South East Asian Region had the highest pooled prevalence of 43.2% (32.2–54.5%, I2 = 90.2%, p < 0.001, n = 7 studies) compared to the Region of the Americas 19.9% (16.7–23.3%, n = 19 studies, I2 = 82.2%, p < 0.001), the Western Pacific Region 23.3% (18.8–28.1%, n = 15 studies, I2 = 93.3%), the African Region 27.2% (14.2–42.5%, n = 3 studies, I2 = 78.4%, p = 0.01), and the European Region 25.7% (14.8–38.3%, n = 14 studies, I2 = 97.5%, p < 0.001); p = 0.01 for difference across groups (Supplementary Fig. 1).

There was a significantly higher pooled prevalence of clinical depression among the patients on chronic HD compared to those in the pre-dialytic stages of CKD 29.9% (24.4–35.8%, n = 44 studies, I2 = 95.6%, p < 0.001) versus 18.5% (12.9–24.9%, n = 10 studies, I2 = 93.1%, p < 0.001); p = 0.01 for difference across the groups (Supplementary Fig. 2).

There was no significant difference in the pooled prevalence among adults compared with children (26.8% vs 15.9%; p = 0.21), Supplementary Fig. 3. There was a higher prevalence of depression among the HD than the PD population, 30.6% (25.0–36.6%, n = 43 studies, I2 = 95.5%, p < 0.001) versus 20.4% (13.1–28.7%, n = 3 studies, I2 = 41.7%, p = 0.18); p = 0.047 for difference between the groups, Supplementary Fig. 4. There appeared to be an increasing temporal trend in depression prevalence. However, this increase did not achieve statistical significance (p = 0.16, Supplementary Figs. 5 and 6). The pooled prevalence for the studies published in 2000 and prior was 17.7% (8.9–28.6%) compared to 26.5% (22.2–31.1%) for those published between the years 2001 and 2020 and 34.4% (20.7–49.5%) for those published after 2020. This trend persisted when considering all studies irrespective of the diagnostic criteria used (Supplementary Fig. 5) or when only the DSM criteria were used (Supplementary Fig. 6).

Discussion

This systematic review and meta-analysis determined the pooled prevalence of clinical depression among 80,932 individuals with CKD from 27 countries spread across the globe. The pooled prevalence of clinical depression was 26.5%. The prevalence varied with the CKD population's geographical location, CKD stage, age group of the individuals with CKD, and dialysis modality (HD versus PD).

The pooled prevalence of depression in the CKD population in this study is higher than the prevalence of 8.4% and 6.9% reported in the general adult population in the United States and Norway, respectively [89, 90]. Similarly, the prevalence of depression in the pediatric population with CKD is higher than the 3.2% reported in the general pediatric population in the United States [91]. These findings showed that depression is a common mental health problem in the CKD population compared to the general population. The prevalence of depression in this study is higher than 21.9% and 16.3% reported among individuals with epilepsy and cancer, respectively [92, 93], suggesting that the magnitude of depression in CKD is higher than in some other chronic illnesses.

The pooled prevalence of depression in this study is higher than 21.4% and 22.8% reported in individuals with CKD who were in pre-dialysis and dialytic stages, respectively, in a systematic review and meta-analysis by Palmer et al.[17] conducted a decade ago. It is, however, lower than 62% reported in a systematic review and meta-analysis involving a population with CKD in Iran [94]. The wide difference in the prevalence rates may be due to the method of assessment of depression and the stage of individuals with CKD that were studied. While the diagnosis of depression was made in the present study using standard clinical diagnostic criteria, the study in Iran used a validated depression screening tool for the diagnosis of depression. The Iranian study also involved only patients on maintenance HD, while the present study included individuals with CKD who were both in the pre-dialysis and dialysis stages.

There was no significant difference between the pooled prevalence of depression in the adult and pediatric population. However, this finding should be interpreted with caution as only two studies involving the pediatric population were included in this analysis. In addition, the prevalence of depression from these studies could have been underestimated as depression may have been unrecognized in children who are largely unable to express themselves [95] adequately. Also, the presentation of depression in children may not follow the typically known presentation as commonly seen in the adult population [95]. This underscores the need for physicians to consider these peculiarities in managing the pediatric CKD population.

The pooled prevalence of clinical depression across the various continents ranges between 19.9 and 43.2%, with the South East Asian Region having the highest while the Region of the Americas had the lowest. The African Region had the second-highest pooled prevalence of depression (27.2%). There was a significant difference in the pooled prevalence of depression in CKD across the various continents. The difference in the severity of CKD, age and socioeconomic status of individuals with CKD, available psychosocial support, governmental support, type of available health care financing, level of health care sophistication, type of maintenance HD and race may be partly responsible for the significant variations in prevalence of depression across the continents. The African continent had the lowest number of studies, while the Region of the Americas had the highest number of studies. This underscores the need for more studies on depression in the African continent in order to ascertain the magnitude of the problem in the region.

Among the studies included in this review, DSM diagnostic criteria were more commonly used for the diagnosis of clinical depression. This may be due to the fact that DSM has a better application in research settings than ICD [96]. Although the DSM was primarily designed for use by psychiatrists in the United States, unlike ICD, which is a global disease classification designed for use by all health practitioners, especially in low and middle-income countries, its use has been reported in many other countries [96]. The pooled prevalence of depression in the studies that used ICD criteria was 39.6%, which was significantly higher than 25.5% in the studies that used DSM criteria. This finding is similar to the report of Wittchen et al. [97], which showed a higher depression prevalence of 11.3% with ICD-10 when compared with 4.2% using DSM 4 in the same population. This finding may be because ICD has greater sensitivity in the diagnosis of mild depression than DSM [98].

There was a significantly higher pooled prevalence of depression among individuals with end-stage kidney disease (ESKD) on HD compared to those in the pre-dialysis stage (29.9% versus 18.5%) in this study. The demands of dialysis treatment on patients with ESKD may partly account for this significant difference. Other factors, such as dietary restriction, inflammation, hormonal changes, poor sleep quality, and reduced quality of life associated with dialysis treatment, may also contribute to a higher frequency of depression in patients on HD compared to individuals with pre-dialysis CKD [15, 99]. A qualitative study by Avdal et al. [100] reported that patients with ESKD experienced despair and decreased social support from relatives and caregivers after the commencement of PD or HD. This may effectively contribute to a higher prevalence of depression in them compared to individuals with pre-dialysis CKD. This study also showed that about one in five individuals with pre-dialysis CKD had depression. This underscores the need to screen individuals with CKD, even in the early stages, for depression so that prompt treatment can be instituted. This may consequently improve their overall outcomes.

The pooled prevalence of depression was significantly higher in individuals with ESKD on maintenance HD compared with those on PD (30.6% versus 20.4%) in this study. There are conflicting reports from existing literature on the relationship between depression and mode of dialysis. The finding of this study is, however, supported by Martin et al. [101] but differs from some previous reports that showed a significantly higher prevalence of depression in PD patients compared to those on HD [102, 103]. Although some other studies reported a higher prevalence of depression in HD patients compared with PD patients, the difference was not statistically significant [104, 105].

There was an increasing temporal trend in depression prevalence after the year 2020 (34.4%) compared to 2001–2020 (26.5%) and before 2000 (17.7%). This may be a reflection of the increase in the level of awareness of mental health problems globally with improvement in campaigns geared towards sensitization of the global community. This is supported by a review by Foulke et al. [106], where it was reported that there has been increased recognition and diagnosis of mental health disorders over the years.

The use of clinical methods of assessment of depression is more specific than the use of validated screening tools in the CKD population. Depression screening tools have been reported to over-diagnose depression in individuals with CKD, especially those at advanced stages [17]. The study by Palmer et al. [17] is instructive in this regard, as it showed that the prevalence of depression was higher in the CKD population when depression screening tools were used compared to when clinical diagnostic criteria were deployed. This was also buttressed by Smith et al. [107] in their study, which showed prevalence rates of depression in the same individuals on maintenance HD as 47% and 5% using Beck's depression index and DSM III, respectively. Also, in cases where depression screening tools are being used in the assessment of depression, individuals with advanced CKD may have some uremic symptoms such as poor sleep, anorexia, fatigue, and lack of concentration, which may overlap and be regarded as somatic symptoms of depression [15, 99]. Another limitation of the use of depression screening tools lies in the fact that a higher score is commonly used as a cut-off in advanced CKD compared to the general population [15, 99]. These cut-off values may vary in different studies using the same screening tool, introducing major inconsistencies.

Depression has a significant adverse impact on the overall outcomes of CKD. Hence, it deserves adequate attention by being promptly diagnosed and managed by the clinician. Depression is associated with poor compliance with treatment follow-up, increased suicidal tendency, withdrawal from dialysis, non-adherence to medications, dietary and fluid restrictions and malnutrition-inflammatory-atherosclerosis [10, 11, 108,109,110]. These may account for the increased hospitalization and disease progression, reduced quality of life, and increased mortality in individuals with CKD who have depression [6, 8, 9, 13, 111,112,113]. Prompt diagnosis and management of depression may mitigate these adverse consequences and improve the overall outcome of these patients.

There is clinician inertia in treating depression in CKD with antidepressants because of uncertainty about the pharmacokinetic properties of the medications and safety profile in people with reduced kidney function. This is supported by a report that showed that only about one-third of individuals with CKD diagnosed with depression received treatment [114]. Furthermore, most randomized controlled trials on the efficacy and safety of antidepressants exclude individuals with CKD. Despite the above, there is evidence, though limited, to support the effectiveness of antidepressants in reducing depression and improving QoL in individuals with CKD [115]. This present systematic review showed that clinical depression is common in CKD and lends credence to the need for the inclusion of the CKD population in clinical trials on the efficacy of antidepressant treatment. Management of depression also involves non-pharmacological treatments. There are reports of some randomized controlled trials that showed the efficacy of non-pharmacological treatment of depression, such as cognitive-based therapy and relaxation techniques, in the treatment of depression in CKD [16, 116,117,118]. A multidisciplinary team-based approach that includes mental health professionals will, therefore, be highly valuable in the management of individuals with CKD with depression.

A limitation of this study is that only a few studies in this review determined the prevalence of depression in the early stages of CKD. Secondly, there was limited information on the prevalence of clinically diagnosed depression in the pediatric population, being reported in only two studies. The strength of this systematic review lies in the fact that the pooled prevalence of depression found in the CKD population is a fairly true representation of the magnitude of the disease. This is particularly true because the study included only articles that used clinical interviews, which are more specific than screening tools.

Conclusion

This systematic review and meta-analysis showed that depression is a relatively common mental health disorder in the CKD population. It has brought to the fore the need for clinicians to make deliberate efforts to evaluate individuals with CKD, especially those with advanced stages of the disease for depression. The findings of this review have also given credence to the call to include the CKD population in large, randomized controlled trials on the safety and efficacy of antidepressants because they are potential beneficiaries of the findings of such studies.