Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care

Carla White; Clíona Ní Cheallaigh; Brenda Griffin

doi:10.12688/hrbopenres.13947.1

Home Browse Homeless patients requiring haemodialysis: incidence, caseload characteristics,...

ALL Metrics

-

Views

18

Downloads

Get PDF

Get XML

Export

▬

✚

Research Article

Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care

[version 1; peer review: 1 approved with reservations]

Carla White ¹, Clíona Ní Cheallaigh^2,3^*, Brenda Griffin^1,2^*

^* Equal contributors

PUBLISHED 03 Sep 2024

Author details Author details

¹ Department of Nephrology, St James's Hospital, Dublin 8, Ireland, D08 NHY1, Ireland
² Department of Medicine, The University of Dublin Trinity College, Dublin, Leinster, D02 R590, Ireland
³ Department of Inclusion Health, St James's Hospital, Dublin, Leinster, D08 NYH1, Ireland

Carla White
Roles: Conceptualization, Formal Analysis, Investigation, Methodology, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

Clíona Ní Cheallaigh
Roles: Conceptualization, Methodology, Supervision, Writing – Review & Editing

Brenda Griffin
Roles: Conceptualization, Methodology, Supervision, Writing – Review & Editing

OPEN PEER REVIEW

REVIEWER STATUS

Abstract

Background

Homelessness is a recognized risk factor for the development of kidney failure. However, the risk of requiring renal replacement therapy (RRT) in homeless populations remains unknown. Furthermore, access to and provision of RRT for homeless populations have not been described.

Methods

We performed a retrospective study of all patients who newly commenced in-centre haemodialysis at St. James’s Hospital, Dublin, from January 2021 to December 2023. Housing status was captured using registered addresses and a review of social workers’ assessments. Homelessness was defined according to the European Typology on Homelessness and Housing Exclusion Criteria. Using publicly available census and homelessness data, we estimated the risk of commencing haemodialysis among homeless and non-homeless populations in the catchment area.

Results

The overall prevalence of homelessness among adults in the catchment area was approximately 2000/270,000 (0.74%). Of the 74 patients who commenced haemodialysis over the 3-year period, 13 (17.6%) were homeless. There was a significant difference in the incidence of commencing haemodialysis between non-homeless and homeless individuals in the catchment area (p<0.001, chi-square test).

Among patients who commenced haemodialysis, the mean age was 53 years in homeless patients and 62 years in non-homeless patients (p<0.05, one-way ANOVA). 3/13 homeless patients (23%) had received nephrology healthcare at least 3 months prior to haemodialysis, compared with 39/61 non-homeless patients (64%) (p <0.05, chi-square). Of the three homeless patients who had received three months of nephrology healthcare prior to haemodialysis, all three patients commenced haemodialysis via an arteriovenous fistula.

Mental health comorbidities were more frequent among homeless patients than among non-homeless patients (23% versus 8%) (not statistically significant), as were substance use disorders (38% versus 7%) (p<0.05, chi-squared test).

Conclusion

The incidence of kidney failure requiring haemodialysis was higher in homeless adults than in their non-homeless peers. Homeless patients were younger and started haemodialysis more urgently. There is an unmet need for nephrology healthcare for the homeless population.

Keywords

Homelessness, people experiencing homelessness, inclusion health, haemodialysis, renal replacement therapy, end-stage kidney disease.

Corresponding author: Carla White

Competing interests: No competing interests were disclosed.

Grant information: The author(s) declared that no grants were involved in supporting this work.

Copyright: © 2024 White C et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: White C, Ní Cheallaigh C and Griffin B. Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care [version 1; peer review: 1 approved with reservations]. HRB Open Res 2024, 7:56 (https://doi.org/10.12688/hrbopenres.13947.1) First published: 03 Sep 2024, 7:56 (https://doi.org/10.12688/hrbopenres.13947.1) Latest published: 03 Sep 2024, 7:56 (https://doi.org/10.12688/hrbopenres.13947.1)

Introduction

Kidney disease, including chronic kidney disease (CKD) and episodes of acute kidney injury (AKI), is a major public health challenge, affecting more than 10% of the general population¹ and more than 20% of acute hospital admissions², which increases the risk of morbidity and mortality throughout the course of the disease. At its most severe, kidney disease progresses to end-stage kidney disease (ESKD), which is associated with the greatest risk of adverse outcomes and mortality, and necessitates renal replacement therapy (RRT), placing a large burden on patients, and causing large financial expenses.

Healthcare for patients undergoing haemodialysis is typically highly complex, involving hospital visits for several hours three times per week, frequent hospital admissions, procedures, and invasive lifestyle factors such as dietary and fluid restrictions to ensure safety between haemodialysis sessions. Those who commence haemodialysis urgently in an unplanned fashion (sometimes called “crash landers”) have been reported to have even higher risk profiles with increased morbidity and mortality than when dialysis is started in a planned way^3–5. Healthcare outcomes in patients on RRT can be improved by planning RRT in advance so that it is commenced under optimal, individualized circumstances that utilise shared-decision making and minimize risks⁶.

Socioeconomic disadvantage is a major risk factor for adverse kidney health outcomes across the spectrum of kidney disease⁷. Socioeconomically disadvantaged individuals are more likely to develop CKD⁸, to progress to ESKD⁹, and to die, even in countries with universal free healthcare⁷. Homelessness can be conceptualised as a state of extreme socioeconomic disadvantage¹⁰ and encompasses a range of living conditions, including rooflessness, living in homeless hostels and sofa-surfing¹¹. Homelessness is associated with extremely poor health. While it is well known that homelessness is associated with increased risks of infectious diseases, psychiatric illnesses and substance misuse, more recently, it has been documented that homelessness is also associated with a greater burden of adverse outcomes from noncommunicable diseases such as cardiovascular diseases¹² and kidney disease¹³. Homelessness is a recognised risk factor for end-stage kidney disease^13,14.

There are limited data on the incidence of ESKD among homeless populations or the prevalence of homelessness among populations receiving RRT¹⁵. Furthermore, access to and provision of RRT to homeless individuals has to date been reported only in case reports and series^15,16. There is a need for research into the relationship between homelessness, ESKD, and nephrology healthcare delivery and access^17,18.

We aimed to ascertain the risk of kidney failure requiring haemodialysis among a homeless population in the inner city of Dublin. We hypothesised that homelessness is associated with an increased risk of haemodialysis initiation. Furthermore, we hypothesised that homeless patients would be more likely to start haemodialysis urgently without a preceding period of nephrology specialist care or planning. Accordingly, we aimed to ascertain the duration of nephrology care preceding haemodialysis initiation and the types of initial haemodialysis access most commonly used in homeless and non-homeless patients in our cohort. Finally, we aimed to ascertain the characteristics of the presentations of kidney failure, including demographic data, comorbidities, prescribing, and length of stay in the inpatient hospital setting.

Methods

Study design and setting

We performed a retrospective study of all patients who newly commenced haemodialysis at St James’s Hospital, Dublin, during a three-year study period from 1^st January 2021 to 31^st December 2023. St James’s Hospital is a large university teaching hospital that serves adults residing in the south-inner city of Dublin. The Haemodialysis Unit in St James’s Hospital is an acute unit that provides intermittent haemodialysis to all hospital adult inpatients who require it either for the progression of CKD or for AKI, as well as to some outpatients who are newly commencing haemodialysis. Data were collected from the Electronic Patient Record. All patients who commenced haemodialysis in the Unit during the study period were included.

Definition and identification of homelessness amongst patient cohort

Housing status was determined using the patients’ registered addresses. In addition, to capture homelessness not otherwise captured by registered addresses, a manual review of social workers’ assessments was performed for all patients included in the study.

Homelessness was defined according to the European Typology of Homelessness and Housing Exclusion¹¹. As such, patients were classified as homeless if they were either roofless, houseless, living in insecure housing or living in inadequate housing. Specifically, roofless refers to individuals who are sleeping rough or in emergency overnight shelters. Houseless refers to individuals living in accommodation designated for homeless people such as shelters, refuges, or immigration centres. Insecure housing refers to individuals living in temporary abodes such as with family or friends, without any legal subtenancy and without another accommodation option, or who are being evicted from their accommodation without any other accommodation option. Inadequate housing refers to individuals living in housing that is unfit for habitation or is extremely overcrowded.

Data extracted

For all patients who commenced haemodialysis during the study period, we recorded their housing status, age, sex, duration of nephrology follow-up prior to haemodialysis (less than 3 months versus greater than 3 months), first haemodialysis access type, haemoglobin level at the time of commencing haemodialysis, whether they had been prescribed erythropoietin-stimulating agents prior to commencing haemodialysis, smoking status, substance misuse comorbidities and mental health comorbidities. For all patients, we also recorded whether haemodialysis was commenced as an inpatient or an outpatient, as well as the date that the patient was recorded in the medical notes as being fit for discharge from a medical perspective, and whether they experienced any delay in discharge from the inpatient hospital setting due to nonmedical or social reasons (referred to as “delayed inpatient discharge”).

Data on catchment area and local homelessness rates

The catchment population of St. James’s Hospital was obtained from previous publications¹⁹. The rate of homelessness in the catchment area was estimated using 2023 data published by the Dublin Regional Homeless Executive, and included individual adults who either resided in emergency or homeless accommodation²⁰ or were sleeping rough²¹.

Ethical approval

The collection of patient data was approved as a service audit by the Service Director for the Nephrology Department in St. James’s Hospital. Data regarding the catchment area size and homelessness figures are publicly available. As no patient identifiable information was collected and the study was approved as a departmental audit in line with the Health Service Executive audit policy, formal ethical approval was not required and patient consent was not required.

Statistical analysis

To compare differences in the incidence of starting dialysis between homeless and non-homeless populations in the catchment area, the chi-square test was used. To compare differences in age and blood haemoglobin levels between homeless and non-homeless patients who started haemodialysis during the study period, one-way analysis of variance (ANOVA) was used. To compare the differences between homeless and non-homeless patients’ duration of nephrology follow-up prior to commencing haemodialysis, their haemodialysis access methods, mental health comorbidities, smoking status, and substance misuse comorbidities, chi-square tests were used. Statistical analyses were performed using SPSS software.

Results

1. Incidence of starting haemodialysis amongst homeless versus non-homeless residents in the catchment area

The catchment area of St. James’s Hospital includes approximately 270,000 adult residents. In the larger Dublin region, 6,288 adults were living in emergency or homeless accommodation in April 2023. A minimum of 118 people were sleeping rough in Dublin as of November 2023. Using these figures, the number of homeless adults in the St James’s Hospital catchment area was estimated to be 2,000. Thus, the overall prevalence of homelessness in adults in the St. James’s Hospital catchment area was approximately 2000/270,000 (0.74%). Of the 74 patients who commenced haemodialysis at St James’s Hospital over the 3-year period, 13 (17.6%) were homeless. There was a statistically significant difference in the incidence of commencing haemodialysis between non-homeless and homeless residents in the catchment area (p<0.001, chi-square test).

The prevalence of homelessness among patients commencing haemodialysis at St. James’s Hospital increased year-on-year during the study period: 2 of 25 patients (8%) commencing haemodialysis in 2021 were homeless, 4 of 25 patients (16%) in 2022, and 7 of 24 patients (29%) in 2023, although this trend was not statistically significant.

2. Age and sex of patients commencing haemodialysis

Among patients who commenced haemodialysis, the mean age was 53 years (standard deviation 15.2) in homeless patients versus 62 years (standard deviation 14.8) in non-homeless patients. Homeless patients who commenced haemodialysis were significantly younger than non-homeless patients (p<0.05, one-way ANOVA) (Figure 1). 12 of the 13 homeless patients (92%) were male, whereas 36 of the 61 non-homeless patients (59%) were male (p<0.05, chi-square test).

Figure 1. Mean age of homeless and non-homeless patients commencing haemodialysis during the study period.

Error bars represent standard deviation.

3. Subcategories of homelessness experienced by patients commencing haemodialysis

Homelessness categories as defined by the European Typology of Homelessness and Housing Exclusion include Roofless, Houseless, Unstable Housing, or Inadequate Housing. Using reviews of social work assessments for all patients who commenced haemodialysis during the study period, we captured the subtypes of homelessness experienced by patients commencing haemodialysis. Of the 13 homeless patients who commenced haemodialysis during the study period, 2 patients were categorised as roofless, 4 patients as houseless, 3 patients as having unstable housing and 4 patients as having inadequate housing (Table 1).

Table 1. Categories of homelessness experienced by patients commencing haemodialysis.

Total homeless patients commencing haemodialysis (n=13)
Subcategory of homelessness:
Roofless (n=2)
• Sleeping in car (n=2)
Houseless (n=4)
• Homeless hostel (n=2)
• Immigration centre (n=2)
Insecure housing (n=4)
• Staying with family/friends without other accommodation option (n=2)
• Received eviction notice without other accommodation option (n=2)
Inadequate housing (n=3)
• Major structural issues/unfit for habitation (n=2)
• Major social issues/unfit for habitation (n=1)

4. Duration of nephrology follow up prior to commencing haemodialysis

We sought to measure the proportion of homeless and non-homeless patients who commenced haemodialysis in a planned fashion, which we defined as having received nephrology specialist care for at least three months prior to starting haemodialysis. 3 of the 13 homeless patients (23%) had been receiving nephrology healthcare at least 3 months prior to haemodialysis, compared with 39 of the 61 non-homeless patients (64%) (p <0.05, chi-square test) (Figure 2).

Figure 2. Duration of nephrology follow up prior to commencing haemodialysis in homeless versus non-homeless patients.

5. Inpatient hospital stay and delayed inpatient discharge

All 13 homeless patients (100%) commenced haemodialysis as hospital inpatients, and 57 of 61 non-homeless patients (93%) commenced haemodialysis as hospital inpatients. There was no statistically significant difference in the rates of commencing haemodialysis as an inpatient between homeless and non-homeless patients.

We then sought to measure the proportion of patients who had commenced haemodialysis as inpatients and were stated to be medically fit for discharge, who then remained in hospital for social reasons. 5 of the 13 homeless patients experienced delayed inpatient discharges compared to 5 of the 61 non-homeless patients in our cohort. This difference was statistically significant (p<0.05, chi-squared test). Among patients who experienced delayed hospital discharge, the median number of additional days spent in hospital due to social reasons was 27 days (range, 14–129 days) for homeless patients and 3 days (range, 2–11 days) for non-homeless patients.

6. Haemodialysis access types

For all patients, we recorded whether the first haemodialysis access was an arteriovenous fistula (AVF), an arteriovenous graft (AVG), a tunnelled haemodialysis catheter (TC) or a non-tunnelled haemodialysis catheter (NTC) (Table 2). Of the 13 homeless patients who commenced haemodialysis during the study period, 5 (38%) commenced haemodialysis using a NTC, 4 (31%) patients using a TC, 4 patients using an AVF (31%), and no patients using an AVG. Of the 61 non-homeless patients who commenced haemodialysis during the study period, 25 patients (41%) were using a NTC, 27 patients (44%) using a TL, 7 patients (11%) using an AVF and 2 patients (3%) using an AVG. There were no statistically significant differences in the rates of commencing haemodialysis using an AVF or via a NTL between homeless and non-homeless patients.

Table 2. Initial haemodialysis access types.

Data expressed as number (percentage).

Haemodialysis access types
	All homeless patients (n=13)	All non-homeless patients (n=61)
Non-tunnelled haemodialysis catheter	5 (38%)	25 (41%)
Tunnelled haemodialysis catheter	4 (31%)	27 (44%)
Arteriovenous fistula or graft	4 (31%)	9 (23%)
	Homeless patients who had been receiving nephrology care for >3months (n=3)	Non-homeless patients who had been receiving nephrology care for >3months (n=39)
Non-tunnelled haemodialysis catheter	0 (0%)	13 (33%)
Tunnelled haemodialysis catheter	0 (0%)	17 (44%)
Arteriovenous fistula or graft	3 (100%)	9 (23%)

Of the 3 homeless patients who had been receiving nephrology care for at least 3 months prior to commencing haemodialysis, all 3 patients commenced haemodialysis via an AVF. Of the 39 non-homeless patients who had been receiving nephrology care for at least 3 months prior to commencing haemodialysis, 13 (33%) commenced haemodialysis via a NTL, 17 patients (44%) via a TL, and 9 patients (23%) via an AVF. When restricting our analysis to patients who had been receiving nephrology specialist care for at least 3 months prior to commencing haemodialysis, there remained no statistically significant differences in the rates of commencing haemodialysis using a NTL between homeless and non-homeless patients.

7. Prescribing of erythropoietin stimulating agents

Mean blood haemoglobin level at the time of commencing haemodialysis was 9.6g/dL in homeless patients and 9.2g/dL in non-homeless patients (not statistically significant, one-way ANOVA).

Fewer homeless patients were on erythropoietin-stimulating agents than non-homeless patients, although the numbers were low in both groups (1 of 12 homeless patients (7.5%) versus 12 of 61 non-homeless patients (19.6%)).

8. Smoking, substance misuse and mental health comorbidities

Unfortunately, smoking data were missing for many of the patients in our cohort. Only 7 of 13 homeless patients (54%) had their smoking status recorded in the Electronic Patient Record, and 33 of 61 (54%) of non-homeless patients. There was no statistically significant difference between the rate of smoking status having been recorded between homeless and non-homeless patients. 2 of the 7 homeless patients whose smoking status was recorded (28%) were active smokers, as compared with 6 of the 33 non-homeless patients whose smoking status was recorded (18%), which was not statistically significant.

Mental health comorbidities were more frequent among homeless patients than among non-homeless patients (23% versus 8%) (not statistically significant), as were substance use disorders (38% versus 7%) (p<0.05, chi-squared test).

Discussion

Homelessness is a state of extreme socioeconomic deprivation associated with an increased risk of developing ESKD¹⁴. In the catchment area of our study, we observed that homeless adults had a more than 10-fold increased risk of needing to start haemodialysis compared to non-homeless adults. The rates of homelessness among patients commencing haemodialysis have not previously been described, and access and provision of haemodialysis for patients who are homeless has not been well documented in the literature. Here, we report that 17% of the patients (1 in 6) who commenced haemodialysis at our centre during the study period were homeless. Homeless patients in our cohort were predominantly male and were, on average, 9 years younger than their non-homeless counterparts at the time of commencing haemodialysis.

Homelessness is associated with increased use of emergency and unscheduled healthcare¹⁹, and it has been reported in the USA that homeless individuals with CKD are less likely to access specialist nephrology care¹⁴. In prior studies of patients commencing haemodialysis, the highest risk of poor outcomes has been observed in individuals who have not accessed a period of nephrology specialist care prior to commencing haemodialysis, and therefore start haemodialysis more urgently^3–5. In our cohort of patients commencing haemodialysis, we observed that homeless patients were significantly less likely than their non-homeless counterparts to have been receiving specialist nephrology care for at least 3 months prior to commencing haemodialysis.

While RRT options include home haemodialysis, peritoneal dialysis and renal transplantation, in-centre haemodialysis is usually the only feasible initial RRT option for someone without a stable fixed abode. We chose patients commencing in-centre haemodialysis at St James’s Hospital as the population for our study. In patients with progressive kidney disease who access nephrology specialist care, planning for haemodialysis typically involves multiple hospital visits, patient education, shared decision-making, investigations such as vein mapping, and procedures such as creation of AVFs or AVGs. AVFs and AVGs, when feasible and acceptable to patients, provide definitive haemodialysis access, which has a lower risk of bloodstream infection than either TCs or NTCs. NTCs are used temporarily (typically for up to 1 week), usually in urgent clinical scenarios where haemodialysis needs to be commenced without time to plan for a TC, AVF or AVG. In our cohort, the rates of non-tunnelled catheters were relatively high in both homeless and non-homeless patients (38% and 41%, respectively), with relatively low rates of commencing haemodialysis via an AVF or AVG. This may represent the caseload of the haemodialysis unit in St. James’s Hospital, which provides haemodialysis mainly to acute hospital inpatients. However, it should be noted that among the 3 homeless patients in our cohort who had been receiving nephrology specialist care at least 3 months prior to commencing haemodialysis, all 3 patients commenced haemodialysis using an AVF. While these are small numbers, they demonstrate that elements of best practice in nephrology care can be achieved in the setting of homelessness.

Cohorts of homeless patients may differ. For the current study, we selected all patients who were commenced on haemodialysis during the study period. We observed increased rates of mental health diagnoses and substance misuse comorbidities (38% and 23%, respectively) among homeless patients in our cohort compared with their non-homeless counterparts. These proportions within our homeless cohort were lower than have been described in other homeless patient cohorts, where mental health diagnoses and substance misuse rates have been even higher^14,22. This may represent differences in collection methods since we determined these diagnoses based on what was either coded or documented in the medical notes of the Electronic Patient Record. On the other hand, there may be true differences between homeless patients who are commencing haemodialysis and the general population of homeless patients. Further work to characterise groups of homeless patients commencing RRT is needed in order to address the needs of these groups.

Commencing haemodialysis adds a substantial burden of care to patients, their families, and caregivers. We observed that homeless patients who had commenced haemodialysis were significantly more likely to remain in hospital for prolonged periods after they were medically stable than their non-homeless counterparts. This is in keeping with previous work demonstrating that homelessness is associated with a disproportionate reliance on inpatient hospital beds¹⁹. Commencing haemodialysis may make it even more difficult for homeless patients and the healthcare team to ensure that they have a safe place to go to after discharge from the hospital.

To our knowledge, this study is the first to describe the prevalence of homelessness among a cohort of patients on commencing haemodialysis. Homelessness can be hidden, for example, if homeless patients provide the address of friends or families as their registered address¹⁹. The current study addressed this issue by using individual reviews of social workers’ assessments for all patients commencing haemodialysis during the study period; therefore, our observation of the number of homeless patients is likely to be relatively accurate. Our study also has a number of weaknesses. Estimates of homelessness in the catchment area of St. James’s Hospital, although based on official figures, are crude. Official figures may underrepresent the true number of homeless adults in the catchment area due to hidden homelessness and people living in insecure or inadequate housing. Therefore, our results on the rates of haemodialysis initiation among homeless patients in the catchment area should be viewed as estimates. Nevertheless, based on these estimates, we observed a strong signal of an increased risk of starting haemodialysis among the homeless population in the catchment area. Large-scale studies are warranted to validate these findings.

In conclusion, we demonstrated that homelessness was common, affecting 1 in 6 patients who commenced haemodialysis in this cohort. Homeless individuals who commenced haemodialysis were younger and more likely to commence haemodialysis urgently. Our findings emphasise the drastic effects of homelessness on kidney health and healthcare access, resulting in a high risk of requiring haemodialysis urgently, a burdensome and costly resource for patients, and healthcare services. Interventions are needed both to address homelessness itself and to increase access to and provision of nephrology services to meet the needs of homeless patients.

Consent

Patient consent was not required for this study. The study was approved as a service audit by the Service Director for the Nephrology Department in St. James’s Hospital in line with Health Service Execute audit policy. No patient identifiable information was collected.

Ethics and consent

The collection of patient data for this study was approved as a service audit by the Service Director for the Nephrology Department in St James’s Hospital. Data regarding the catchment area size and homelessness figures are publicly available. As no patient identifiable information was collected and the study was approved as a departmental audit in line with the Health Service Executive audit policy, formal ethical approval was not required and patient consent was not required.

Data availability

The raw data for this article are not suitable for publication or sharing under any circumstances. This was a small study involving data from 74 patients who received haemodialysis in a single haemodialysis unit. Data were collected in line with the HSE Clinical Audit Policy. Although no direct patient identifiable data were collected, data such as age, sex, housing status, and date of commencing haemodialysis could potentially be used to identify individual patients. While the data analyses as presented are entirely anonymised and can be published; the raw data are not suitable for sharing or publication because of the risk of individual patients being identified.

Reporting guidelines

This article has been prepared in accordance with STROBE reporting guidelines for observational studies.

Faculty Opinions recommended

References

1. Stack AG, Casserly LF, Cronin CJ, et al.: Prevalence and variation of Chronic Kidney Disease in the Irish health system: initial findings from the national kidney disease surveillance programme. BMC Nephrol. 2014; 15(1): 1–12, 185. PubMed Abstract | Publisher Full Text | Free Full Text
2. Wang HE, Muntner P, Chertow GM, et al.: Acute Kidney Injury and mortality in hospitalized patients. Am J Nephrol. 2012; 35(4): 349–355. PubMed Abstract | Publisher Full Text | Free Full Text
3. Ratcliffe PJ, Phillips RE, Oliver DO: Late referral for maintenance dialysis. Br Med J (Clin Res Ed). 1984; 288(6415): 441–3. PubMed Abstract | Publisher Full Text | Free Full Text
4. Cass A, Cunningham J, Arnold PC, et al.: Delayed referral to a nephrologist: outcomes among patients who survive at least one year on dialysis. Med J Aust. 2002; 177(3): 135–138. PubMed Abstract | Publisher Full Text
5. Stack AG: Impact of timing of nephrology referral and pre-ESRD care on mortality risk among new ESRD patients in the United States. Am J Kidney Dis. 2003; 41(2): 310–8. PubMed Abstract | Publisher Full Text
6. Chan CT, Blankestijn PJ, Dember LM, et al.: Dialysis initiation, modality choice, access, and prescription: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) controversies conference. Kidney Int. 2019; 96(1): 37–47. PubMed Abstract | Publisher Full Text
7. Sawhney S, Blakeman T, Blana D, et al.: Care processes and outcomes of deprivation across the clinical course of kidney disease: findings from a high-income country with universal healthcare. Nephrol Dial Transplant. 2023; 38(5): 1170–1182. PubMed Abstract | Publisher Full Text | Free Full Text
8. Choi AI, Weekley CC, Chen SC, et al.: Association of educational attainment with chronic disease and mortality: the Kidney Early Evaluation Program (KEEP). Am J Kidney Dis. 2011; 58(2): 228–234. PubMed Abstract | Publisher Full Text | Free Full Text
9. Hsu CY, Iribarren C, McCulloch CE, et al.: Risk factors for End-Stage Renal Disease: 25-year follow-up. Arch Intern Med. 2009; 169(4): 342–350. PubMed Abstract | Publisher Full Text | Free Full Text
10. Marmot M: Inclusion health: addressing the causes of the causes. Lancet. 2018; 391(10117): 186–188. PubMed Abstract | Publisher Full Text
11. Feantsa: On the way home? FEANTSA monitoring report on homelessness and homeless policies in Europe. Feantsa. 2012; 1–92. Reference Source
12. Baggett TP, Liauw SS, Hwang SW: Cardiovascular disease and homelessness. J Am Coll Cardiol. 2018; 71(22): 2585–2597. PubMed Abstract | Publisher Full Text
13. Maziarz M, Chertow GM, Himmelfarb J, et al.: Homelessness and risk of end-stage renal disease. J Health Care Poor Underserved. 2014; 25(3): 1231–1244. PubMed Abstract | Publisher Full Text | Free Full Text
14. Hall YN, Choi AI, Himmelfarb J, et al.: Homelessness and CKD: a cohort study. Clin J Am Soc Nephrol. 2012; 7(7): 1094–1102. PubMed Abstract | Publisher Full Text | Free Full Text
15. Podymow T, Turnbull J: Management of chronic kidney disease and dialysis in homeless persons. Kidney Int Suppl. 2013; 3(2): 230–235. PubMed Abstract | Publisher Full Text | Free Full Text
16. Blackmond N, Loh T, Kanke JD, et al.: Challenges in management of kidney failure in a free clinic setting: a case report. Cureus. 2022; 14(6): e26352. PubMed Abstract | Publisher Full Text | Free Full Text
17. Novick TK, King B: Addressing housing issues among people with kidney disease : importance, challenges, and recommendations. Am J Kidney Dis. 2024; 84(1): 111–119. PubMed Abstract | Publisher Full Text | Free Full Text
18. Novick TK, Gadegbeku CA, Crews DC: Dialysis for patients with end-stage renal disease who are homeless: importance, challenges and recommendations. JAMA Intern Med. 2018; 178(12): 1581–1582. PubMed Abstract | Publisher Full Text | Free Full Text
19. Cheallaigh CN, Cullivan S, Sears J, et al.: Usage of unscheduled hospital care by homeless individuals in Dublin, Ireland: a cross-sectional study. BMJ Open. 2017; 7(11): e016420. PubMed Abstract | Publisher Full Text | Free Full Text
20. Executive DRH: Monthly report to Dublin City councillors on homelessness. 2023; 1–8. Reference Source
21. Executive DRH: Week long assessment of rough sleepers in the Dublin Region, winter count 2023. 2023. Reference Source
22. O ’Reilly F, Barror S, Hannigan A, et al.: Homelessness: an unhealthy state. Health status, risk behaviours and service utilisation among homeless people in two Irish Cities. 2015. Reference Source

Comments on this article Comments (0)

Version 1

VERSION 1 PUBLISHED 03 Sep 2024

Author details Author details

¹ Department of Nephrology, St James's Hospital, Dublin 8, Ireland, D08 NHY1, Ireland
² Department of Medicine, The University of Dublin Trinity College, Dublin, Leinster, D02 R590, Ireland
³ Department of Inclusion Health, St James's Hospital, Dublin, Leinster, D08 NYH1, Ireland

Carla White
Roles: Conceptualization, Formal Analysis, Investigation, Methodology, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

Clíona Ní Cheallaigh
Roles: Conceptualization, Methodology, Supervision, Writing – Review & Editing

Brenda Griffin
Roles: Conceptualization, Methodology, Supervision, Writing – Review & Editing

Competing interests

No competing interests were disclosed.

Grant information

The author(s) declared that no grants were involved in supporting this work.

Article Versions (1)

version 1

Published: 03 Sep 2024, 7:56

https://doi.org/10.12688/hrbopenres.13947.1

Copyright

© 2024 White C et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

VIEWS

284

downloads

18

Citations

0

SEE MORE DETAILS

CITE

how to cite this article

White C, Ní Cheallaigh C and Griffin B. Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care [version 1; peer review: 1 approved with reservations]. HRB Open Res 2024, 7:56 (https://doi.org/10.12688/hrbopenres.13947.1)

NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Version 1

VERSION 1

PUBLISHED 03 Sep 2024

Views

7

Reviewer Report 30 Oct 2024

Wael Hussein, Stanford University School of Medicine, San Jose, CA, USA; Stanford University School of Medicine, Stanford, CA, USA

Approved with Reservations

https://doi.org/10.21956/hrbopenres.15303.r42535

Review of [Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care]

29 October 2024

The authors of this paper observed a high proportion of people experiencing homelessness among those starting dialysis ... Continue reading

Review of [Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care]

29 October 2024

The authors of this paper observed a high proportion of people experiencing homelessness among those starting dialysis at an inner-city hospital in Dublin, Ireland. The proportion was more than 20x higher than the estimated proportion of the homeless population in the catchment area.
The investigators compared the proportion of homeless population in the catchment area (estimated at 0.74%) to the proportion of patients who started RRT over a three year-period while homeless (17.6%). The proportions were significantly different using a Chi square test. They concluded that the incidence of RRT was higher among the homeless population.
The authors further compared those starting dialysis while homeless to those with no housing problems in terms of characteristics, pre-dialysis care, and post-dialysis care utilization.
The authors are to be commended on exploring such an important health-related issue in a vulnerable population.
I have the following comments about the report:
The premise of reporting on “risk” using the paper’s methodology is problematic.
While this report highlights the relative perspective on the risk of RRT, note that:

The analysis is cross-sectional in terms of capturing homelessness. Calculation of the “risk” or “incidence” would require follow up over time.
The assumption of this analysis is that “homelessness” is a static condition. The assumption ignores that homelessness is a dynamic status.
Proper reporting requires capturing a cohort of people with homelessness and following them over time until start of RRT or completing a follow up duration. It needs to capture and account for competing outcomes (eg death before requiring or starting RRT) and censoring outcomes (eg change in status or leaving the catchment area).
In addition to reporting the risk in time units, proper analysis would adjust for confounding variables.
Study period: Jan 2021 to Dec 2023: What changes took place during this COVID period that could have affected this analysis? Examples are: projects to reduce homelessness, and competing outcomes that could have affected the incidence of RRT (eg death from other causes, changes in the incidence of RRT in general – as observed by others in the literature).
The assumption that homelessness is evenly distributed across the city is a potential source of error. Certain areas, like the inner city, may have disproportionately higher rates of homelessness, which could explain why the number of homeless in the catchment area might exceed what a simple proportional estimate suggests. How would different assumptions change the estimates and conclusions?

While the use of chi-square and ANOVA is ok given the available data, the reporting of confidence intervals (CIs) for key outcomes, such as the incidence of haemodialysis and the significance of comorbidities, would improve the robustness of the statistical analysis.

Furthermore, note the following:

Can you describe the dialysis service at the dialysis unit observed? There is a note that it is an acute unit. Are there patients who start at a different unit particularly if stable or did the analysis capture all RRT starts from the catchment area even if they started elsewhere?
The report is limited by the small sample size, limiting the interpretation of the findings.
Please reflect on the surprising finding that the population with homelessness had higher than expected follow up and permanent dialysis access. Homelessness is known to be associated with reduced adherence to healthcare appointments in the literature.
This is a single center study.
Please discuss the generalizability (or lack of it) given the limitations. Extrapolation of results to other regions with varying healthcare systems and homelessness profiles might not be accurate without further validation.

The paper nicely highlights the longer hospital stays among patients with homelessness, pointing to the need for further attention to the problem. Given the small cohort, would the investigators explore other outcomes starting from start of RRT, such as dialysis adherence, modality over time, and transplantation? A discussion on these endpoints would enrich the findings and provide a fuller picture of the care needs for homeless patients on haemodialysis.
What were the causes of ESKD in these patients?
Given the limitations in assessment of the “risk”, the manuscript needs to tone that part down and clearly highlight the simple observation of the much higher than expected proportion of patients with homelessness starting RRT. It should instead emphasize the finding related to long hospital stays.
Minor comment: NTL and NTC – consider using one abbreviation.
With large proportion of missing data re smoking, it might be better to skip reporting it.
Be careful with the language while interpreting the results; eg, can’t say that kidney disease affects them younger with this methodology.
The discussion could be strengthened by including specific, actionable recommendations (even if brief) for healthcare providers, policymakers, or homeless support services to address these gaps.
Did these patients have any other healthcare encounters in the months prior to ESKD? Particularly among those not seen by a nephrologist – indicating problems with referrals from other providers. Note: In the general population, homelessness is associated with higher mortality, hospitalizations, and health care costs. [Novick AJKD 2024]
Are there any other information that should be included in the manuscript for international readers? Eg, Healthcare system in Ireland, policies that impact homelessness, policies related to homeless people access to healthcare.

Is the work clearly and accurately presented and does it cite the current literature?

Yes
Is the study design appropriate and is the work technically sound?

Partly
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

No
Are the conclusions drawn adequately supported by the results?

Partly

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Dialysis

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

CITE

Report a concern

Respond or Comment

Comments on this article Comments (0)

Version 1

VERSION 1 PUBLISHED 03 Sep 2024

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1
Version 1 03 Sep 24	read

Wael Hussein, Stanford University School of Medicine, San Jose, CA, USA; Stanford University School of Medicine, Stanford, CA, USA

Comments on this article

All Comments(0)

Add a comment

Sign up for content alerts

Back to all reports

Reviewer Report

7 Views

30 Oct 2024 | for Version 1

Wael Hussein, Stanford University School of Medicine, San Jose, CA, USA; Stanford University School of Medicine, Stanford, CA, USA

7 Views Cite this report Responses(0)

Approved With Reservations

Review of [Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care]

29 October 2024

The authors of this paper observed a high proportion of people experiencing homelessness among those starting dialysis at an inner-city hospital in Dublin, Ireland. The proportion was more than 20x higher than the estimated proportion of the homeless population in the catchment area.
The investigators compared the proportion of homeless population in the catchment area (estimated at 0.74%) to the proportion of patients who started RRT over a three year-period while homeless (17.6%). The proportions were significantly different using a Chi square test. They concluded that the incidence of RRT was higher among the homeless population.
The authors further compared those starting dialysis while homeless to those with no housing problems in terms of characteristics, pre-dialysis care, and post-dialysis care utilization.
The authors are to be commended on exploring such an important health-related issue in a vulnerable population.
I have the following comments about the report:
The premise of reporting on “risk” using the paper’s methodology is problematic.
While this report highlights the relative perspective on the risk of RRT, note that:

The analysis is cross-sectional in terms of capturing homelessness. Calculation of the “risk” or “incidence” would require follow up over time.
The assumption of this analysis is that “homelessness” is a static condition. The assumption ignores that homelessness is a dynamic status.
Proper reporting requires capturing a cohort of people with homelessness and following them over time until start of RRT or completing a follow up duration. It needs to capture and account for competing outcomes (eg death before requiring or starting RRT) and censoring outcomes (eg change in status or leaving the catchment area).
In addition to reporting the risk in time units, proper analysis would adjust for confounding variables.
Study period: Jan 2021 to Dec 2023: What changes took place during this COVID period that could have affected this analysis? Examples are: projects to reduce homelessness, and competing outcomes that could have affected the incidence of RRT (eg death from other causes, changes in the incidence of RRT in general – as observed by others in the literature).
The assumption that homelessness is evenly distributed across the city is a potential source of error. Certain areas, like the inner city, may have disproportionately higher rates of homelessness, which could explain why the number of homeless in the catchment area might exceed what a simple proportional estimate suggests. How would different assumptions change the estimates and conclusions?

While the use of chi-square and ANOVA is ok given the available data, the reporting of confidence intervals (CIs) for key outcomes, such as the incidence of haemodialysis and the significance of comorbidities, would improve the robustness of the statistical analysis.

Furthermore, note the following:

Can you describe the dialysis service at the dialysis unit observed? There is a note that it is an acute unit. Are there patients who start at a different unit particularly if stable or did the analysis capture all RRT starts from the catchment area even if they started elsewhere?
The report is limited by the small sample size, limiting the interpretation of the findings.
Please reflect on the surprising finding that the population with homelessness had higher than expected follow up and permanent dialysis access. Homelessness is known to be associated with reduced adherence to healthcare appointments in the literature.
This is a single center study.
Please discuss the generalizability (or lack of it) given the limitations. Extrapolation of results to other regions with varying healthcare systems and homelessness profiles might not be accurate without further validation.

The paper nicely highlights the longer hospital stays among patients with homelessness, pointing to the need for further attention to the problem. Given the small cohort, would the investigators explore other outcomes starting from start of RRT, such as dialysis adherence, modality over time, and transplantation? A discussion on these endpoints would enrich the findings and provide a fuller picture of the care needs for homeless patients on haemodialysis.
What were the causes of ESKD in these patients?
Given the limitations in assessment of the “risk”, the manuscript needs to tone that part down and clearly highlight the simple observation of the much higher than expected proportion of patients with homelessness starting RRT. It should instead emphasize the finding related to long hospital stays.
Minor comment: NTL and NTC – consider using one abbreviation.
With large proportion of missing data re smoking, it might be better to skip reporting it.
Be careful with the language while interpreting the results; eg, can’t say that kidney disease affects them younger with this methodology.
The discussion could be strengthened by including specific, actionable recommendations (even if brief) for healthcare providers, policymakers, or homeless support services to address these gaps.
Did these patients have any other healthcare encounters in the months prior to ESKD? Particularly among those not seen by a nephrologist – indicating problems with referrals from other providers. Note: In the general population, homelessness is associated with higher mortality, hospitalizations, and health care costs. [Novick AJKD 2024]
Are there any other information that should be included in the manuscript for international readers? Eg, Healthcare system in Ireland, policies that impact homelessness, policies related to homeless people access to healthcare.

Is the work clearly and accurately presented and does it cite the current literature?

Yes
Is the study design appropriate and is the work technically sound?

Partly
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

No
Are the conclusions drawn adequately supported by the results?

Partly

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Dialysis

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Respond to this report

Responses (0)

[1] 1. Stack AG, Casserly LF, Cronin CJ, et al.: Prevalence and variation of Chronic Kidney Disease in the Irish health system: initial findings from the national kidney disease surveillance programme. BMC Nephrol. 2014; 15(1): 1–12, 185. PubMed Abstract | Publisher Full Text | Free Full Text

[2] 2. Wang HE, Muntner P, Chertow GM, et al.: Acute Kidney Injury and mortality in hospitalized patients. Am J Nephrol. 2012; 35(4): 349–355. PubMed Abstract | Publisher Full Text | Free Full Text

[3] 3. Ratcliffe PJ, Phillips RE, Oliver DO: Late referral for maintenance dialysis. Br Med J (Clin Res Ed). 1984; 288(6415): 441–3. PubMed Abstract | Publisher Full Text | Free Full Text

[4] 4. Cass A, Cunningham J, Arnold PC, et al.: Delayed referral to a nephrologist: outcomes among patients who survive at least one year on dialysis. Med J Aust. 2002; 177(3): 135–138. PubMed Abstract | Publisher Full Text

[5] 5. Stack AG: Impact of timing of nephrology referral and pre-ESRD care on mortality risk among new ESRD patients in the United States. Am J Kidney Dis. 2003; 41(2): 310–8. PubMed Abstract | Publisher Full Text

[6] 6. Chan CT, Blankestijn PJ, Dember LM, et al.: Dialysis initiation, modality choice, access, and prescription: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) controversies conference. Kidney Int. 2019; 96(1): 37–47. PubMed Abstract | Publisher Full Text

[7] 7. Sawhney S, Blakeman T, Blana D, et al.: Care processes and outcomes of deprivation across the clinical course of kidney disease: findings from a high-income country with universal healthcare. Nephrol Dial Transplant. 2023; 38(5): 1170–1182. PubMed Abstract | Publisher Full Text | Free Full Text

[8] 8. Choi AI, Weekley CC, Chen SC, et al.: Association of educational attainment with chronic disease and mortality: the Kidney Early Evaluation Program (KEEP). Am J Kidney Dis. 2011; 58(2): 228–234. PubMed Abstract | Publisher Full Text | Free Full Text

[9] 9. Hsu CY, Iribarren C, McCulloch CE, et al.: Risk factors for End-Stage Renal Disease: 25-year follow-up. Arch Intern Med. 2009; 169(4): 342–350. PubMed Abstract | Publisher Full Text | Free Full Text

[10] 10. Marmot M: Inclusion health: addressing the causes of the causes. Lancet. 2018; 391(10117): 186–188. PubMed Abstract | Publisher Full Text

[11] 11. Feantsa: On the way home? FEANTSA monitoring report on homelessness and homeless policies in Europe. Feantsa. 2012; 1–92. Reference Source

[12] 12. Baggett TP, Liauw SS, Hwang SW: Cardiovascular disease and homelessness. J Am Coll Cardiol. 2018; 71(22): 2585–2597. PubMed Abstract | Publisher Full Text

[13] 13. Maziarz M, Chertow GM, Himmelfarb J, et al.: Homelessness and risk of end-stage renal disease. J Health Care Poor Underserved. 2014; 25(3): 1231–1244. PubMed Abstract | Publisher Full Text | Free Full Text

[14] 14. Hall YN, Choi AI, Himmelfarb J, et al.: Homelessness and CKD: a cohort study. Clin J Am Soc Nephrol. 2012; 7(7): 1094–1102. PubMed Abstract | Publisher Full Text | Free Full Text

[15] 15. Podymow T, Turnbull J: Management of chronic kidney disease and dialysis in homeless persons. Kidney Int Suppl. 2013; 3(2): 230–235. PubMed Abstract | Publisher Full Text | Free Full Text

[16] 16. Blackmond N, Loh T, Kanke JD, et al.: Challenges in management of kidney failure in a free clinic setting: a case report. Cureus. 2022; 14(6): e26352. PubMed Abstract | Publisher Full Text | Free Full Text

[17] 17. Novick TK, King B: Addressing housing issues among people with kidney disease : importance, challenges, and recommendations. Am J Kidney Dis. 2024; 84(1): 111–119. PubMed Abstract | Publisher Full Text | Free Full Text

[18] 18. Novick TK, Gadegbeku CA, Crews DC: Dialysis for patients with end-stage renal disease who are homeless: importance, challenges and recommendations. JAMA Intern Med. 2018; 178(12): 1581–1582. PubMed Abstract | Publisher Full Text | Free Full Text

[19] 19. Cheallaigh CN, Cullivan S, Sears J, et al.: Usage of unscheduled hospital care by homeless individuals in Dublin, Ireland: a cross-sectional study. BMJ Open. 2017; 7(11): e016420. PubMed Abstract | Publisher Full Text | Free Full Text

[20] 20. Executive DRH: Monthly report to Dublin City councillors on homelessness. 2023; 1–8. Reference Source

[21] 21. Executive DRH: Week long assessment of rough sleepers in the Dublin Region, winter count 2023. 2023. Reference Source

[22] 22. O ’Reilly F, Barror S, Hannigan A, et al.: Homelessness: an unhealthy state. Health status, risk behaviours and service utilisation among homeless people in two Irish Cities. 2015. Reference Source

Homeless patients requiring haemodialysis: incidence, caseload characteristics, and implications for care

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Methods

Study design and setting

Definition and identification of homelessness amongst patient cohort

Data extracted

Data on catchment area and local homelessness rates

Ethical approval

Statistical analysis

Results

1. Incidence of starting haemodialysis amongst homeless versus non-homeless residents in the catchment area

2. Age and sex of patients commencing haemodialysis

Figure 1. Mean age of homeless and non-homeless patients commencing haemodialysis during the study period.

3. Subcategories of homelessness experienced by patients commencing haemodialysis

Table 1. Categories of homelessness experienced by patients commencing haemodialysis.

4. Duration of nephrology follow up prior to commencing haemodialysis

Figure 2. Duration of nephrology follow up prior to commencing haemodialysis in homeless versus non-homeless patients.

5. Inpatient hospital stay and delayed inpatient discharge

6. Haemodialysis access types

Table 2. Initial haemodialysis access types.

7. Prescribing of erythropoietin stimulating agents

8. Smoking, substance misuse and mental health comorbidities

Discussion

Consent

Ethics and consent

Data availability

Reporting guidelines

References

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Competing Interests Policy

Stay Updated

Are you a HRB-funded researcher?

Thank you!