Chronic Kidney Disease (CKD) poses a substantial global health burden, with an estimated prevalence of over 800 million individuals worldwide and over 35 million in the United States (1–7). Beyond its clinical and economic burden, CKD is increasingly recognized as a condition influenced by complex interactions between physical and psychological health (8–13).

Among psychiatric conditions, depression is notably prevalent, with rates ranging from 20-40% in CKD populations which are substantially higher than in the general population (14–18). Anxiety, while also common and frequently co-morbid with depression, represents a distinct construct with different symptom patterns and clinical trajectories (17, 19). To maintain conceptual clarity, the present study focuses primarily on depression while acknowledging its frequent co-occurrence with anxiety in CKD.

Although depression is well characterized in advanced CKD and end-stage kidney disease (ESKD), its role in earlier CKD stages remains insufficiently understood despite the potential for meaningful intervention during this period (20–25). Recent work, including large-scale observational studies and KDIGO 2024 guidance, emphasizes the importance of early identification and management of psychosocial factors that may accelerate disease progression (25–27). Early-stage CKD (stages 1–3) represents a critical window in which modifiable behavioral and biological contributors may shape long-term outcomes, yet epidemiologic and mechanistic evidence in this population is limited.

Multiple pathways may plausibly link depression to CKD progression. Inflammation is a key mechanism: depression is associated with elevated pro-inflammatory cytokines such as IL-6 and TNF-α, which may contribute to renal microvascular injury and faster decline in estimated glomerular filtration rate (eGFR) (17, 19, 28). Neuroendocrine dysregulation, including chronic activation of the hypothalamic–pituitary–adrenal (HPA) axis and heightened sympathetic drive, can worsen hypertension, metabolic dysfunction, and endothelial injury which are all known accelerators of CKD progression. Additional behavioral mechanisms include reduced medication adherence, decreased engagement in self-management behaviors, suboptimal dietary adherence, and delayed follow-up, all of which may amplify renal risk (17, 19, 28). These mechanistic pathways collectively provide a strong theoretical basis for evaluating depression as a potentially modifiable risk factor for CKD progression.

Importantly, accumulating evidence across psychiatric research reinforces the interplay between mental health and physical disease. For example, real-world outpatient studies have illustrated the bidirectional relationships between psychiatric illness and metabolic comorbidities, highlighting the need for integrated care models (29). Likewise, emotional dysregulation, as demonstrated in studies examining guilt and affective symptoms in eating disorders, can exert broad effects on health-related behaviors and outcomes (30). The complexity of psychiatric diagnoses in real-world settings, including diagnostic variability and ethical considerations, further underscores the need for high-quality longitudinal data when evaluating depression’s impact on chronic conditions such as CKD (30). These insights parallel the challenges and opportunities inherent in using electronic health record (EHR) data, as seen in other retrospective real-world studies (8, 9).

Given these biological, behavioral, and clinical considerations, understanding the influence of depression during the earlier stages of CKD may have substantial implications for prevention, risk stratification, and timely mental health intervention. However, existing research remains limited, particularly regarding the temporal relationship between incident depression and subsequent kidney function decline. To address these gaps, we conducted a real-world, retrospective cohort study using electronic health record data to evaluate (1) whether patients diagnosed with depression after CKD experience accelerated kidney function decline, and (2) whether more advanced CKD stages are associated with a higher likelihood of developing new-onset depression.

This study was a retrospective cohort analysis utilizing data from the TriNetX electronic medical record (EMR) database, covering the years 2007 to 2022.

The findings of this study underscore a significant association between depression and CKD progression, particularly in the earlier stages. Our results show that patients with a diagnosis of depression following their CKD diagnosis had a higher likelihood of kidney disease progression compared to those without depression. Although no prior studies have specifically examined this relationship in earlier stages of CKD, existing literature has identified associations between depression and outcomes such as dialysis initiation and mortality. For example, a recent study by Shen et al. found that CKD patients with depressive symptoms were more likely to initiate dialysis or experience all-cause mortality, pointing to a potential link between mental health and disease trajectory (37). Similarly, Zhang et al. observed that depressive symptoms were associated with a 2.25-fold higher risk of progression to dialysis over a three-year period (38). These associations may be partly explained by reduced treatment adherence and behavioral changes commonly seen in individuals with depression (39).

Nonetheless, prior studies have produced conflicting findings (40, 41). Some concluded that depressive symptoms were not significantly related to disease progression or adverse outcomes like cardiovascular events (40, 42). While our hazard ratio analysis identified a statistically significant relationship between depression and progression of kidney disease, prior studies have reported varying magnitudes of association, possibly due to differences in study populations, access to mental health care, or measurement definitions (15). For instance, while we used eGFR decline and a threshold-based definition for kidney disease progression, other studies defined progression based on dialysis initiation or ESRD diagnosis. Population differences may also contribute: Eveleens et al., who focused on individuals aged 65 and older, reported contrasting findings. In our study, although the hazard ratio for CKD progression was numerically lower in those over 65, the difference was not statistically significant (41). Fischer et al. examined African American patients, whereas our cohort included individuals of multiple races, albeit predominantly White individuals, may explain some variation in our findings (40). Our analysis also identified a statistically significant association between CKD stage and subsequent depression diagnoses. Individuals in later CKD stages (Stages 4 and 5) were more likely to be diagnosed with depression than those in earlier stages. The Kaplan–Meier analysis indicated shorter time to depression onset in more advanced stages, suggesting a possible increasing psychological burden as kidney function declines. These results are consistent with prior studies reporting higher depression prevalence among individuals with advanced CKD. For example, Palmer et al.’s meta- analysis estimated depression prevalence at 22.8% to 39.3% in individuals with CKD, compared to lower rates in earlier stages (21.4% to 26.5%) (15).

Contrary findings have also been reported. For example, Balogun et al. found that depression was more prevalent among elderly male patients and in earlier CKD stages. They also reported that the prevalence of depression was higher in the earlier stages of CKD (29% to 47%) than in later stages (30%)⁴⁷. However, these differences may be attributable to sample characteristics, such as Balogun’s use of a Veterans Affairs cohort composed primarily of older adults.

Additionally, their analysis focused on prevalence, while our study examined time-to-event and hazard ratios, limiting direct comparison.

From a clinical standpoint, implementing systematic depression screening could meaningfully change CKD care pathways. The patient health questionnaire (PHQ)-2 is a recommended first-line tool due to its brevity and sensitivity, with positive screens followed by the PHQ-9 to guide diagnostic confirmation and severity assessment (43, 44). For patients in CKD Stages 1–3, screening every 6–12 months may be appropriate, while those in Stages 4–5 may benefit from screening every 3–6 months due to higher psychological burden and increased risk of symptom progression. For individuals identified with clinically significant depressive symptoms, interventions such as collaborative care models, psychotherapy (e.g., CBT), antidepressant pharmacotherapy with renal-adjusted dosing, social work involvement, and structured adherence-support programs have demonstrated benefit in chronic medical conditions. Embedding mental health professionals within nephrology practices or using telehealth-supported behavioral health could further enhance access. These targeted strategies may help address the biological, behavioral, and socioeconomic pathways through which depression contributes to poorer kidney outcomes.

Strengths of our study include a large sample size, the use of propensity score matching and time-to-event modeling methods, an extended retrospective follow-up period, and inclusion of data from rural Appalachian populations, an often-underrepresented group. It is important to note that although depression was associated with accelerated CKD progression in our cohort, these findings do not establish causality. The observational nature of our study limits the ability to determine whether depression directly contributes to kidney function decline or whether unmeasured confounding, shared biological pathways, or the psychological burden of worsening CKD may explain part of this relationship. While inflammatory, behavioral, and socioeconomic mechanisms provide biologically plausible pathways, the observed associations should be interpreted as indicative rather than causal.

This study relied on ICD-9-CM and ICD-10-CM codes to identify depression, anxiety, and CKD, which is standard in large EHR-based analyses but introduces several important limitations. ICD-coded psychiatric diagnoses generally capture only clinically recognized and documented conditions and may underestimate milder, subclinical, or undiagnosed cases. Furthermore, because depression and anxiety frequently co-occur and cannot always be reliably distinguished using administrative codes, these conditions were combined into a single exposure variable. While this approach aligns with prior real-world database studies, it may obscure clinically meaningful differences in symptomatology and severity.

Similarly, CKD was identified using ICD codes and staged using eGFR values obtained from routine clinical practice, which are subject to variability in measurement timing, frequency, and clinical context. Although we excluded patients with insufficient laboratory data or physiologically implausible values, residual misclassification of CKD stage may still occur. Additionally, to reduce confounding, patients with major cancers, prevalent mental illnesses, or insufficient follow-up were excluded using ICD codes. While these decisions improved internal validity, they may limit generalizability and do not eliminate unmeasured confounding inherent in retrospective EHR research.

Several additional limitations merit discussion. First, the absence of socioeconomic variables limited our ability to adjust for key determinants that influence both mental health and CKD outcomes. Socioeconomic status affects access to primary and specialty care, medication affordability, health literacy, transportation, environmental exposures, and chronic psychosocial stress, all of which may shape the onset or progression of both depression and kidney disease. The lack of these variables may therefore have contributed to residual confounding and could partly explain some of the associations observed.

Second, our study population was predominantly White, which may limit generalizability to more racially and ethnically diverse populations who often experience different burdens of CKD, variation in mental health diagnosis and treatment patterns, and distinct structural or socioeconomic barriers.

Third, because the dataset was derived largely from inpatient encounters, outpatient psychiatric diagnoses, antidepressant prescriptions, and routine eGFR monitoring may have been under-ascertained. This may lead to misclassification of both exposure and outcome, particularly for individuals with milder or intermittently managed depression.

Fourth, although propensity score matching helped reduce observable selection bias, unmeasured confounding remains possible, especially in relation to socioeconomic factors that were unavailable in the dataset. As such, some of the observed associations may reflect structural or environmental disparities rather than direct clinical effects, and no causal inference can be made in a retrospective design.

Finally, our models relied on the documented timing of depression diagnoses and eGFR measurements within the EHR. In routine clinical practice, laboratory testing and diagnostic coding occur at irregular intervals, and depressive symptoms may precede formal documentation by weeks or months. This limits our ability to determine the precise temporal sequence between exposure and outcome, and the resulting interval censoring may influence hazard estimates. Future studies incorporating more granular socioeconomic data and analytic methods designed to account for irregular observation intervals may improve precision and interpretability.

Future research should also examine interventional strategies aimed at addressing the bidirectional relationship between depression and CKD. Approaches such as nephrology-adapted collaborative care models, behavioral activation or CBT-based programs, and adherence-support interventions may hold promise for improving both psychological well-being and renal outcomes. Pragmatic or randomized trials that integrate behavioral health services within CKD clinics, including telehealth-enabled models. are needed to determine whether treating depression can meaningfully enhance patient quality of life or potentially slow CKD progression.

In summary, while these limitations reflect the inherent constraints of real-world data and administrative coding practices, our study contributes to the growing body of literature demonstrating a significant association between depression and CKD progression and an increased likelihood of depression among individuals with more advanced CKD. Because this study was retrospective and observational, the temporal patterns identified should be interpreted as associations rather than causal effects. Although our findings suggest that depression is associated with a higher probability of CKD progression and that more advanced CKD stages are associated with a greater likelihood of depression, these relationships do not establish causation. Instead, they highlight clinically meaningful bidirectional associations that warrant further investigation through prospective or interventional studies. Collectively, these findings underscore the importance of incorporating routine mental health assessment into CKD care and support the need for targeted research aimed at improving both psychological and renal outcomes.