Chronic kidney disease and severe mental illness: a scoping review

Background People who have severe mental illness experience higher rates of long-term conditions and die on average 15–20 years earlier than people who do not have severe mental illness, a phenomenon known as the mortality gap. Long-term conditions, such as diabetes, impact health outcomes for people who have severe mental illness, however there is limited recognition of the relationship between chronic kidney disease and severe mental illness. Therefore, the aim of this scoping review was to explore the available evidence on the relationship between chronic kidney disease and severe mental illness. Methods Electronic databases, including MEDLINE, Embase, CINAHL, and PsycINFO were searched. The database searches were limited to articles published between January 2000–January 2022, due to significant progress that has been made in the detection, diagnosis and treatment of both SMI and CKD. Articles were eligible for inclusion if they explored the relationship between SMI and CKD (Stages 1–5) in terms of prevalence, risk factors, clinical outcomes, and access to treatment and services. Severe mental illness was defined as conditions that can present with psychosis, including schizophrenia, schizoaffective disorder, bipolar disorder, and other psychotic disorders. Thirty articles were included in the review. Results The included studies illustrated that there is an increased risk of chronic kidney disease amongst people who have severe mental illness, compared to those who do not. However, people who have severe mental illness and chronic kidney disease are less likely to receive specialist nephrology care, are less likely to be evaluated for a transplant, and have higher rates of mortality. Conclusion In conclusion, there is a dearth of literature in this area, but the available literature suggests there are significant health inequalities in kidney care amongst people who have severe mental illness. Further research is needed to understand the factors that contribute to this relationship, and to develop strategies to improve both clinical outcomes and access to kidney care. Graphical abstract


Introduction
People with severe mental illness (SMI) experience significant health inequalities [1][2][3]. Severe mental illnesses are defined as mental health conditions that impair a person's ability to function in their daily life, typically conditions that can present with psychosis, such as schizophrenia and bipolar disorder [3]. People who have an SMI diagnosis die, on average, 15-20 years earlier than people who do not [4]. One issue that contributes substantially to this inequality is the high prevalence of long-term conditions among people with SMI, and the poor clinical outcomes that people with coexisting SMI and long term conditions experience in comparison to those without SMI [3].
There has been ongoing research exploring the relationship between long term conditions and SMI [4], including developing interventions to help support people with SMI manage these long term conditions, for example diabetes mellitus [5] and cardiovascular disease [6]. However, some long term conditions have received relatively limited attention in research, despite the high prevalence of risk factors and the potential significant impact on clinical outcomes for people with SMI. One long term condition that is growing in prevalence, yet receives limited recognition, is chronic kidney disease (CKD) [7]. The most common causes of CKD in the general population are diabetes mellitus and hypertension, conditions that are significantly more prevalent among people with SMI, in part because of the side-effects of antipsychotic medication [2,8,9].
The early stages of CKD (stages I-III) can be difficult to detect due to lack of symptoms, yet early detection and intervention can help prevent progression of CKD, development of complications, and mortality. When kidney failure is reached people require kidney replacement therapy to sustain life. Kidney replacement therapy involves replacement of the damaged kidneys through either transplantation or dialysis. These options involve a significant treatment burden, with intensive monitoring and medication regimens, alongside regular invasive treatments, dietary restrictions, and fluid restrictions for dialysis treatment [10]. Research exploring self-management of long term conditions amongst people with SMI has highlighted the difficulties these individuals have managing conditions such as diabetes mellitus, cardiovascular disease, or long term conditions in general [11,12]. However, kidney disease poses a unique and substantial burden in relation to selfmanagement, for example haemodialysis may require people to attend hospital three times a week for 4 h each time or restrict fluid intake to only 500 mLs per day [10]. It is poorly understood how people with SMI experience this burden or engagement in the treatment of CKD or kidney failure [13].
Despite the increased risk of CKD associated with lithium [14], a common medication for bipolar disorder and schizoaffective disorder, the high prevalence of risk factors associated with the development of CKD, CKD progression and poor outcomes, we know very little about this co-morbidity, including prevalence, outcomes, and access to services. Therefore, this scoping review is needed to describe the available literature on this relationship and present what is already known about CKD and SMI.

Aims
The aim of this review is to describe the available literature on the relationship between SMI and CKD, including prevalence, risk factors, clinical outcomes and access to treatment and specialist services.

Materials and methods
A scoping review approach, following the steps outlined in the PRISMA-ScR reporting guidelines [15], was used due to the breadth of the review questions and the need to map the available evidence. A protocol for the scoping review was created to guide the searching, screening, data extraction and synthesis process, however as scoping review protocols are not eligible for registration on the PROSPERO database this was not registered.

Information sources
Electronic databases, including MEDLINE, Embase, CINAHL, and PsycINFO were searched. The database searches were limited to articles published between January 2000-January 2022, due to significant progress that has been made in the detection, diagnosis and treatment of both SMI and CKD.

Search strategy
A search strategy was developed for each of the two conditions (CKD and SMI) which were reviewed and revised by subject librarians. The two search strategies were combined using the Boolean Operator 'AND' to capture literature that explored the co-existence of both CKD and SMI. Table 1 provides an exemplar search strategy used in MEDLINE.

Eligibility criteria
Articles were eligible for inclusion if they explored the relationship between SMI and CKD (Stages 1-5) in terms of prevalence, risk factors, clinical outcomes, and access to treatment and services. Severe mental illness was defined as conditions that can present with psychosis, including schizophrenia, schizoaffective disorder, bipolar disorder, and other psychotic disorders.
During the initial scoping of the literature, it was apparent that there is an extensive body of literature establishing the relationship between lithium treatment and CKD, including recent systematic literature reviews and meta-analyses [16]. To avoid duplication, articles that focused on lithium either in terms of the mechanisms, prevalence, or risk factors for CKD, were excluded unless they provided insight into other contributing risk factors. As this review is focused on CKD, articles focusing on acute kidney injury were excluded. A more comprehensive list of exclusion criteria can be found in Table 2.

Data extraction
A data extraction table was created to capture the main components of the included articles and chart the available evidence. The data items that were extracted included author names and date of publication, publication type, setting, aim, study design, start date-end date, participants, relevant findings, and conclusions. Data were extracted by CC and CC, and were reviewed and discussed by all authors to confirm accuracy and relevance.

Data synthesis
To address the aim of the scoping review, a descriptive narrative synthesis was carried out guided by the review question, providing an overview of the main areas of research to identify the current knowledge base and highlight gaps in the evidence. Due to the breadth of the review and heterogeneity of the included articles, no quantitative analysis was carried out.

Results
A total of 3581 articles were identified from the database search once duplicates were removed. Following title and abstract screening, 3365 records were excluded due to their focus on topics that were irrelevant to the review, and 216 records were identified for full text screening. Following full text screening, 185 articles were excluded. Most articles were excluded because the studies focused on the wrong patient population (n = 67) or focused on the role of lithium in CKD (n = 51). An overview of the screening process is available in Fig. 1.
Thirty studies were identified for inclusion, reported in journal articles (n = 24) and conference abstracts (n = 6).
Most included studies were retrospective cohort studies using national, regional, or local databases (n = 24), and cross-sectional analysis of healthcare databases (n = 5). One article reported a qualitative study. The studies were carried out in the USA (n = 11), UK (n = 5), Taiwan (n = 4), Denmark (n = 2), France (n = 2), Australia (n = 1), Canada (n = 1), India (n = 1), Ireland (n = 1), and Israel (n = 1). One study did not report the geographical location of study, although the authors were based in the UK.
An overview of the included studies is available in Table 3.

Risk of developing CKD amongst people with SMI
Several studies examined the risk of developing CKD for people with SMI [2,[17][18][19][20][21][22][23][24]]. There appears to be an increased risk for developing CKD amongst people with SMI [24], however the degree of risk may be different across SMI diagnoses. Tzur Bitan et al. [17] conducted a retrospective cohort study, comparing the prevalence of CKD amongst people with schizophrenia to matched controls without schizophrenia, over a 7-year period. Schizophrenia was found to be predictive of CKD above all other risk factors (OR = 1.62, 95% CI = 1.45-1.82, P < 0.0001) [17]. While other studies also found an increased risk of CKD amongst people with schizophrenia [21], the risk of CKD appeared to be higher for people with bipolar disorder. Smith et al. (2013) conducted a cross-sectional study exploring the prevalence of medical comorbidities in people with a diagnosis of bipolar disorder [21]. People with bipolar disorder had higher rates of CKD compared to controls (OR: 2.42, 95% CI = 2.04-2.86, P < 0.001). However, no studies carried out a direct comparison between SMI diagnoses.
The prevalence of CKD amongst people with SMI is increasing over time, according to a retrospective cohort study evaluating the prevalence of long-term conditions amongst people with SMI across a 12-year period [20]. The prevalence of CKD amongst people with SMI increased from 0.28% in 2000/2001 to 8.24% in 2011/2012. Whilst the prevalence of CKD also increased in people without SMI (from 0.26% in 2000/2001 to 4.16% in 2011/2012), the increase was at a slower rate, with the prevalence remaining lower in comparison.

Risk factors for developing CKD amongst people with SMI
Three studies explored the role psychiatric medications play in the development of CKD, specifically first-and secondgeneration antipsychotics [25,26], and anticonvulsants used as mood stabilisers in bipolar disorder [27].
Second-generation antipsychotics were associated with a small to moderate increase in risk of incident CKD in a case-control cohort study [25]. This study included 21,434 CKD cases and examined the association between secondgeneration antipsychotic use and incident CKD. They found an increased risk of CKD both in people who were using second-generation antipsychotics at the time of the study (OR 1.26, 95% CI: 1.12-1.42), and people who had ever used second-generation antipsychotics (OR 1.24, 95% CI:1.12-1.37). There did not appear to be a dose-response relationship, as both short-and long-term prescriptions were associated with an increased risk. All second-generation antipsychotics, except for aripiprazole, were associated with an increased risk of CKD, with the highest risk associated with clozapine [OR 1.81, 95% CI: 1.22-2.69]. In contrast, a study by Wang et al. [26] found there was no significant association between first-or second-generation antipsychotic use and CKD, but there was a significantly increased risk of CKD amongst people who used a combination of first-and second-generation antipsychotics (OR 1.28, 95% CI = 1.11-1.47). However, more than 85% of their sample were receiving a combination treatment of first-and second-generation antipsychotics, limiting their ability to draw definitive conclusions about the relative risk of different antipsychotics.
The role of anticonvulsants in the risk for CKD was explored by Kessing et al. [27] who compared the rates of CKD and kidney failure among people who had bipolar disorder who were treated with lithium, anticonvulsants, and people without bipolar disorder who had never been exposed to these drugs. Anticonvulsants and other mood stabilisers were associated with an increased risk of CKD. There was a dose-response relationship with anticonvulsant treatment, as people who received over 60 prescriptions of anticonvulsants were at a higher risk of developing CKD (HR = 2.30 [95% CI, 1.53-3.44]) compared to people who had only 1-2 prescriptions (HR = 1.23 [95% CI, 0.76-1.99]; P < 0.001). A similar pattern was also observed with kidney failure, as the rates of kidney failure increased with the number of anticonvulsant prescriptions a person had received (P = 0.002). This was in comparison to lithium, which did not have a significant association with the development of kidney failure.

Suicide
One study explored suicide in the context of physical multimorbidity and SMI. Chen et al. [28] explored the presence of physical comorbidities in the 3-month period before suicide mortality in individuals with bipolar disorder. People were more likely to have a higher comorbidity score and were more likely to have had a diagnosis of CKD in the 3-month period before they died by suicide.

Mortality
Several studies highlighted the increased risk of mortality amongst people who have co-existing SMI and kidney failure [29,30]. Over a 5-year period people with schizophrenia have a lower risk of commencing dialysis, but a significantly higher risk of death, compared to people without schizophrenia [31]. Kimmel et al. [32] examined the association between hospitalisations with psychiatric diagnoses and death in adult patients receiving dialysis. There was an increased risk of 1-year mortality for people with kidney failure who had been hospitalised with a primary (OR 1.23, 95% CI: 1.18-1.28) or secondary psychiatric diagnosis (OR 1.09, 95% CI: 1.08-1.11), compared to those hospitalised without a psychiatric diagnosis. A study by Abbott et al. [33] found that the cumulative mortality following hospitalisation for psychosis for people who have co-existing kidney failure was 53% at 1-year follow-up, 73% at 2-year follow-up, and 84% at 3-year follow-up, and hospitalisation for psychoses was independently associated with mortality risk (HR = 1.87, 95% CI: 1.77-1.98; P < 0.001).
One study explored risk factors for mortality in people who had schizophrenia and kidney failure [34]. There was a significantly increased risk of mortality amongst people who had co-existing schizophrenia and kidney failure if they were over the age of 35, used a central venous catheter for haemodialysis access, and had heart failure. These are also associated with an increased risk of mortality in people without co-existing SMI. However, there was a reduced risk of mortality amongst people with co-existing schizophrenia and kidney failure if they had hypertension, a history of suicidal ideation or drug dependence, had ischaemic cardiovascular disease or a history of being non-compliant with treatment. These factors are associated with an escalation in care, close monitoring, and regular follow-up, which was thought to account for the reduction [34].
Most research available exploring mortality for people with co-existing SMI and CKD focused on kidney failure, as opposed to earlier stages of CKD. This may under-estimate the role that CKD plays in mortality for people with coexisting SMI, as people with SMI may not progress to kidney failure as they die before reaching the final stages [31].

Hospitalisation
McPherson et al. [35] explored the relationship between co-existing SMI and CKD, and rates of hospitalisation and repeated hospitalisation. Rehospitalisation rates are significantly increased both in people who have CKD, and people who have SMI, however there is a higher rate amongst people who have co-existing SMI and CKD (HR = 1.55, 95% CI = 1.40-1.73, P < 0.001), particularly those admitted through the emergency department. In contrast, while there was not a significantly increased risk of repeated hospitalisation for people with co-existing SMI and CKD, there was a significantly increased risk of emergent rehospitalisation. This risk was higher for those with affective disorders, such as bipolar, compared to those with schizophrenia [35,36].
None of the included studies found an increased risk of poor clinical outcomes for people who had a diagnosis of SMI and received a kidney transplantation, although there may be a risk of exacerbation of SMI symptoms in the immediate post-transplant period. Kofman et al. [39] found that following transplantation, 23 patients from their sample had a relapse of their psychiatric disorder [39], while 13 people required psychiatric hospitalisation.

Access to specialist nephrology care
Despite the increased prevalence of CKD amongst people with SMI, there is evidence that people with SMI had lower rates of accessing specialist nephrologist care [17,   Participants with a diagnosis of severe mental illness and chronic kidney disease (Stages 1-5), or participants who have experience providing care for people with severe mental illness and chronic kidney disease Studies exploring the relationship between mental health and CKD, where mental health encompasses common mental disorders such as anxiety and depression Articles published in the English language Studies and articles focused exclusively on acute kidney injury Empirical articles, including quantitative and qualitative studies Studies and articles focused exclusively on the risk of CKD in lithium treatment Participants over the age of 18 Participants under the age of 18 Articles focused on the relationship between SMI and CKD, including prevalence, risk factors, clinical outcomes, and access to services Articles that are not reporting empirical research, for example editorial or news articles Hsu et al. [31] found that people with schizophrenia and co-existing kidney failure had a lower chance of receiving an appointment with a nephrologist (OR = 0.6, 95% CI 0.4-0.8, P < 0.001), of receiving an erythropoietin prescription (OR = 0.7, 95% CI 0.6-0.9, P < 0.05), and were more likely to be hospitalised within the first year of commencing dialysis (OR = 1.4, 95% CI 1.0-1.8, P < 0.05).
Most studies exploring access to nephrology care focused on access to kidney replacement therapy. People with SMI and kidney failure were less likely to have received a kidney transplant (OR = 5.43, 95% CI = 2.84-10.38, P < 0.001) [17,19], be on the register to receive a kidney transplant (only 1.4% of people registered for a kidney transplant had a psychiatric disorder, compared to 5.3% of people who were not registered) [43] or be scheduled for a kidney transplant evaluation appointment (only 6% patients with schizophrenia, who met the eligibility criteria for a kidney transplant, were scheduled for a transplant evaluation appointment) [44] compared to people without SMI. While people with SMI were more likely to commence haemodialysis than to receive a kidney transplant [19], there is evidence that in general they are less likely to commence kidney replacement therapy [17,31]. Tzur Bitan et al. [17] conducted a retrospective cohort study and found dialysis treatment was more commonly received among patients with CKD who did not have a comorbid schizophrenia diagnosis (11.1%), compared to CKD patients who did have the diagnosis (8.5%).

Barriers and facilitators of care
One study explored the experiences of renal nurses providing care to people with SMI receiving haemodialysis [45]. The main barriers to providing effective care were staff shortages, and a lack of staff education on mental illness. Nurses at times perceived people with SMI as difficult and challenging, particularly when they displayed aggressive behaviour. However, some facilitators included having effective communication skills and empathy. Renal nurses  Significant differences between suicide cases and living controls in relation to employment and Charlson comorbidity index score. Patients who had died by suicide were more likely to have a higher comorbidity score CKD was more commonly present in the 3-month period pre-suicide in women, and only CKD and sepsis were associated with suicidality in male patients CKD was more common in people with bipolar disorder, both men and women, who completed suicide in comparison to healthy controls   The risks for CKD for those who used SGAs longer cumulatively were higher than those who did not. In addition, those who used only FGAs and those who used both SGAs and FGAs seemed to have lower risk for CKD or transplantation) among individuals exposed to lithium, individuals exposed to anticonvulsants and other drugs used to treat bipolar disorder and individuals not exposed to these drugs The rates of end-stage CKD did not increase with the number of lithium prescriptions whereas they significantly increased with the number of prescriptions of anticonvulsants (P for trend = 0.002) The results suggest that CKD is associated with bipolar disorder, independent of medication. Both lithium and anticonvulsant treatment are associated with an increased risk of CKD, with anticonvulsant treatment being associated with kidney failure  It is likely that this population is at an increased risk of higher rates of CKD in the future due to their relatively young age reported that they wanted more support from mental health teams, with a collaborative approach to patient care between teams.

Discussion
This scoping review is the first to identify and describe the available evidence on the relationship between CKD and SMI. While several studies provided insight into the risk of CKD for people with SMI, contributing factors, clinical outcomes, and access to care, overall, there is a dearth of high-quality research focussed on co-existing SMI and CKD. The evidence demonstrates there may be a significantly increased risk of CKD for people who have SMI [17][18][19][20][21]31]. The level of risk appears to differ depending on the SMI diagnosis [19,21,27], although there is little research available examining the different prevalence rates across diagnoses. Additionally, there are few studies examining the impact of risk factors for CKD amongst people who have SMI. People who have SMI are more likely to have co-morbid health conditions such as type 2 diabetes and cardiovascular disease [2], are more likely to smoke [46], have metabolic syndrome [4], and live a sedentary lifestyle [47]. These factors all contribute to the risk of CKD, however none of the studies identified in this review examined the role of these risk factors in the development of CKD amongst people with SMI.
An interesting finding from the review was the relationship between psychiatric medication, other than lithium, and the risk of CKD. Two studies implicated both first-and second-generation antipsychotics in the risk of CKD [25,26], with mixed findings, and clozapine was associated with the highest risk. Clozapine is a highly effective atypical antipsychotic with an extensive side effect profile and, like most antipsychotic medication [9], is associated with the development of metabolic syndrome, type 2 diabetes, and cardiovascular disease [48]. Yet there is limited research acknowledging the potential impact of medication for SMI on kidney health, with the exception of lithium treatment [16].
This review also highlights potential inequality in access to specialist care for CKD among people with SMI, particularly kidney transplantation. People who have SMI appear to have similar outcomes following transplantation, with no evidence of increased rejection rates, graft loss, or mortality [38,39,41]. Despite this, people with SMI were less likely to receive a kidney transplant, be on the transplant register, or be assessed for transplantation [17,43]. This inequality in access to transplantation services for people with psychiatric diagnoses has been highlighted across specialties [49]. This lack of access could result from stigma towards people with SMI, a lack of knowledge of the guidelines and evidence, and therapeutic pessimism [50]. Further research is needed to explore the role of stigma as a barrier to transplant services for those with SMI, in order to address this inequality [49]. This inequality is compounded by worse clinical outcomes overall. The evidence highlights higher rates of mortality and hospitalisation for people who have CKD and SMI [31,33]. While people who have SMI are less likely to reach kidney failure, this is most likely a result of them dying too early for their disease to progress [31]. However, while people with SMI are not reaching the end stages of CKD, this does not mean that CKD does not contribute to the mortality gap. Cardiovascular mortality is a significant contributor to the mortality gap for people with SMI [51], while people with CKD are at a significantly raised risk of cardiovascular mortality, even in the earlier stages of the disease [30,52].
There are a few strengths and limitations to this review. Articles were restricted to those published in the English language, meaning there may be relevant articles published in other languages that would have contributed to our understanding of the relationship between CKD and SMI, but were not accessed. Studies were included on the basis that they reported information related to people who had been diagnosed with a mental health condition classed as an SMI, however the reliability of diagnosis was not assessed. Additionally, since this is a scoping review the quality of the studies was not formally assessed using a risk of bias tool as the aim of the review was to describe the evidence, as opposed to evaluating the reliability of results. However, this review is the first to provide an overview of the literature available on the relationship between CKD and SMI. Moreover, using the scoping review methodology enabled a broad overview of the evidence to identify key gaps in the research. Table 4 illustrates some of these gaps in the evidence.
In conclusion, there is a dearth of research exploring the prevalence, risk factors, outcomes, and available care for people with SMI and CKD. There was a significant gap in relation to qualitative research, particularly studies exploring the associated care pathways for people with SMI and CKD, how this is monitored and managed in primary or secondary care mental health services, and how this compares to other long-term conditions associated with SMI, such as diabetes, cardiovascular disease and chronic obstructive pulmonary disease. The available evidence suggests there is an increased risk of CKD amongst people who have SMI, that people with SMI have poorer clinical outcomes from CKD, and that they are not accessing specialist renal care to the same degree as people without SMI. There is a need for further research in this area to fully understand this relationship and inform strategies to improve clinical outcomes and close the mortality gap.