Introduction
To sustain life, patients with end-stage kidney disease (ESKD) require renal replacement therapy to take the place of their damaged kidneys. Dialysis, the most common form of renal replacement therapy, comprises 97% of incident ESKD cases and 70% of prevalent ESKD cases in the United States (US). Dialysis may be conducted in-center (i.e., in-center hemodialysis) or at home (i.e., peritoneal dialysis (PD) and home hemodialysis) [
1]. In-center hemodialysis is administered by a nurse or a dialysis technician within a clinic setting, three times a week, 3–5 h at a time [
1,
2]. In the US, PD is self-managed and conducted at home, and home hemodialysis requires a care partner to administer dialysis to the patient [
1,
3,
4].
US patients undergoing home dialysis experience benefits such as flexible dialysis schedules, reduced need for transportation to clinics, and greater self-reported quality of life [
1,
3‐
5]. In addition, Medicare cost savings per-patient-per-year is over $14,000 for PD compared to hemodialysis [
1]. Although medical eligibility for in-center hemodialysis and home hemodialysis is comparable, patients who undergo home hemodialysis still experience quality of life benefits [
1,
4,
5]. However, home hemodialysis is the least utilized form of dialysis, with about 1.3% of prevalent ESKD cases undergoing home hemodialysis [
1].
Ideally, the decision for which dialysis modality to undergo should mostly depend on patient preferences, and occur within a shared decision-making context with a nephrology provider, in which nephrology providers inform and advise patients on their treatment options to facilitate a collaborative approach with patients to reach a final dialysis decision that aligns with patient needs and preferences [
6‐
8]. However, despite benefits associated with home dialysis, prevalence rates for patients on dialysis in the US are overwhelmingly skewed towards in-center (87.5%) over both home dialysis modalities combined (12.5%) [
1]. Given these rates, there may be eligible patients who would potentially prefer home dialysis, but are instead served in center.
While many factors may influence dialysis modality decision-making, past interventions and clinical recommendations aimed at supporting patients who must choose a renal replacement therapy that is right for them have focused on educating patients about their options and promoting shared decision-making between patients and nephrology providers [
9,
10]. However, interventions intended to increase patient knowledge about treatment options, and to improve the shared decision-making process between healthcare providers and patients almost always occur right before the end stage of the disease is reached [
11‐
16]. In other words, conversations between healthcare providers and patients regarding their options often occur too late to be effective or do not occur at all. In addition, efforts to improve patient knowledge about treatment options and shared decision-making processes have not generated enough change, and have not involved patients with early-stage chronic kidney disease (CKD) [
13,
15‐
18]. Taken together, these factors may suggest that deficiencies in shared decision-making processes, and the overall low rates of home dialysis, are beyond the control of just patients and nephrology providers. Using a social-ecological lens to examine this phenomenon may help clarify what circumstances lead patients to be informed about their dialysis options.
The Social Ecological Framework (SEF) is used to understand the dynamic between various personal and environmental factors [
19]. Under the SEF, the individual-, interpersonal-, organizational/community-, and policy/national-levels are all considered. In health-related research, this framework recognizes that while individuals have agency in making their own health-related decisions, their environment can either help or hinder the decision-making process as well as its outcomes [
20,
21]. For example, relying on an individual patient approaching ESKD to either independently educate themselves about their dialysis options or find opportunities for dialysis education on their own dismisses the important role that nephrology providers could play in guiding them to a decision that works best with their overall health. Similarly, expecting shared decision-making processes to be the sole method in which optimal dialysis outcomes are reached ignores outside factors that may prevent patients from being referred to nephrology with enough time before they reach ESKD.
Little is known regarding the extent to which organizational/community- and/or policy/national-level factors, otherwise known as macro-level factors (e.g., lack of CKD screening recommendations, lack of hospital access to decision aid materials) affect patients’ dialysis modality decisions. Exploring dialysis modality decision-making through a social-ecological lens may shed light on challenges and barriers during the decision-making process that influence lack of patient knowledge and lack of a shared decision-making process.
The purpose of this qualitative study was to explore both micro- and macro-level factors surrounding dialysis decision-making among patients undergoing in-center and home dialysis to examine what factors influence the final dialysis decision as well as what leads patients to select home dialysis over in-center hemodialysis, henceforth referred to as “in-center dialysis.”
Methods
Design, participants, and setting
The present study employed an exploratory qualitative approach using in-depth individual interviews [
22,
23] that later underwent content analysis [
24]. This study design was utilized because factors associated with the dialysis decision-making process, such as macro-level factors, are largely unexplored. The Consolidated criteria for Reporting Qualitative Research (COREQ) checklist was used and reported in the
supplementary material [
25].
Eligible participants were 18 years or older, undergoing dialysis for ESKD, and had the cognitive ability to provide consent. A purposive sampling approach was employed, using referrals from nephrology providers. A diverse sample with respect to racial/ethnic background, rural/urban status, and time since dialysis initiation was recruited while achieving equal numbers of patients undergoing in-center and home dialysis.
Patients already undergoing dialysis were selected as the ideal candidates to shed light on dialysis modality decision-making processes, in lieu of patients who were approaching ESKD, because the former group had already undergone the decision-making experience, and the aim of the present study was to explore factors related to a final dialysis decision without interfering. Furthermore, both typical case sampling and intensity sampling were utilized [
26,
27]. Typical case sampling is conducted to illustrate the average experience of participants (i.e., patients with ESKD undergoing in-center dialysis) [
27]. Intensity sampling is a form of positive deviant case sampling, which is used to highlight cases that are uncommon but not necessarily unusual [
26]. The purpose of this sampling strategy was to learn from “successful” cases, because these patients may have knowledge and insights regarding resources or strategies that could be applied to others. In this study, patients who were undergoing home dialysis were considered intensity cases, since home dialysis is vastly underutilized in the US [
1].
Qualitative interviews were conducted with patients in a nephrology clinic at an academic medical center in the Midwestern region of the US. In-center dialysis patients participated during their dialysis session, and home dialysis patients participated in a private clinic room before or after their monthly appointments with nephrology providers. Home dialysis patients were only seen in the clinic once a week. Both patients undergoing PD and home hemodialysis were recruited from the home dialysis group. Every participant received $25 in appreciation of their time. All data were deidentified and the study was approved by the university’s institutional review board.
Data collection
Semi-structured, individual interviews were conducted from September 2019 to December 2020 in English and Spanish (MVB) followed by a short questionnaire assessing demographic and clinical factors. There was a three-month hiatus in recruitment from March 2020 to June 2020 due to the COVID-19 pandemic. Most patients approached for recruitment agreed to participate. All patients undergoing home dialysis who were approached agreed to participate. About 75% of approached patients undergoing in-center dialysis agreed to participate. The primary reported reason patients declined participation was because they felt too weak to participate while they were dialyzing. Before participation, the interviewer introduced herself, described the study aims, and explained what participant involvement entailed.
The interview guide was informed by literature review of dialysis decision-making and common healthcare-related determinants outlined by the SEF [
21]. The guide contained questions covering topics in the following areas: 1) participants’ journeys while learning they had CKD; 2) their pre-nephrology healthcare providers; 3) their accessibility to healthcare services prior to nephrology care; 4) their transition into nephrology care; 5) learning about their dialysis options; 6) how they made a final decision; 7) perceptions of their current dialysis modality; and 8) what they would have done differently during this process. Given that most interviews were conducted during the COVID-19 pandemic, participants were also asked how the pandemic had affected their lives as patients on dialysis. Although the primary topics remained the same throughout the study, questions and prompts of the interview guide underwent an iterative process based on what came up in interviews and how participants responded to questions. When necessary, participants were probed to clarify and provide more detail. Interviews were audio recorded and transcribed verbatim.
Questionnaire items asked participants to report their age, gender, race/ethnicity, relationship status, years of education attained, household income, zip code, general health status rating, comorbid diseases, and time since their CKD diagnosis, ESKD diagnosis, and dialysis initiation. Zip code was used to measure rurality using US Department of Agriculture Rural Urban Commuting Area codes [
28,
29].
Sample size was guided by data saturation, meaning recruitment stopped once interviewing more participants did not provide additional insights regarding the study’s main objectives. In other words, recruitment ended once participants’ stories, which included different details, all began to allude to the same themes [
30,
31].
Analysis – Qualitative data
Interview transcripts were audited for clarity and removal of identifying information, then imported into Dedoose, a qualitative data analysis software package [
32]. Qualitative analysis team members included two PhD candidates (MVB, JLA) and one faculty member (NMA). They had backgrounds in behavioral medicine, health psychology (MVB, JLA), clinical psychology (JLA), and community and behavioral health (MVB, NMA). All had advanced degrees, and two were experienced in qualitative research (MVB, NMA). Two analysis team members (MVB, JLA) used a blend of deductive then inductive-dominant content analysis [
24]. The preliminary code list included some a priori codes based on the study aims, then other codes were added after reading the first few interview transcripts repeatedly and forming first impressions. Each interview transcript was analyzed independently using the code list (MVB, JLA), then jointly to resolve differences on codes and themes identified. The code list was iteratively adapted as interviews and analysis progressed. The third qualitative analysis team member (NMA) served as an advisor throughout the analysis process and provided input on codes and overarching themes. Qualitative analysis team members maintained rigor by record keeping within the code list, including code definitions, inclusion/exclusion criteria, adaptations, dates of changes, and examples using excerpts from participant responses. Overarching themes were derived from the finalized coded transcripts. Since data collection occurred in a single visit, transcripts and coded data were not returned to participants for correction.
Analysis – Questionnaire data
Means, standard deviations, and ranges were calculated for continuous variables, and frequencies and percentages were calculated for categorical variables. All event information was recorded in years. If an event had occurred less than a year ago, it was recorded as a fraction out of 12 to convert to a year-value to obtain relevant descriptive statistics. Chi-square analyses and Pearson correlations were conducted to test within-sample differences between modality groups and significant findings are reported herein.
Discussion
This qualitative study describes and compares the experiences, knowledge, and attitudes of forty participants undergoing home and in-center dialysis, and aimed to identify macro-level factors associated with dialysis decision-making. Healthcare access and engagement before entering nephrology care, after entering nephrology care, and following dialysis initiation influenced patients’ awareness regarding their CKD status, progression toward ESKD, and dialysis options. Interactions with nephrology providers seemingly influenced awareness the most, but the totality of these experiences affected which dialysis modality patients would eventually undergo. Potential modifiers of these outcomes include race, ethnicity, and language barriers. Most participants adopted a passive role in the decision-making process, following through with the dialysis modality suggested to them by nephrology providers. Finally, fatigue, concerns regarding one’s dialyzing schedule, and problems with fistula/catheter access sites contributed to overall satisfaction with one’s dialysis modality. Because these findings highlighted both micro- and macro-level factors associated with the final dialysis modality (Fig.
2), the SEF seems useful for understanding this phenomenon.
Data-driven analysis identified statistically significant within-sample findings using Pearson’s correlation and Chi-square analysis. The home dialysis cohort were more likely to be White, rural, in a relationship, have more years of education, rate their health as better, and initiate dialysis more recently compared to the in-center cohort. These results, however, should only be interpreted in terms of this small, exploratory study. For instance, home dialysis rates in other clinics may not have rural-majority patient samples; thus, this finding may not be generalizable. On the other hand, the fact that patients undergoing home dialysis were more likely to be in a relationship compared to those undergoing in-center dialysis may suggest that being in a relationship implicitly contributes to one’s self-efficacy to independently manage dialysis at home, even though that finding did not emerge from qualitative interviews. Further research is warranted.
The study sample came from a well-resourced Midwestern hospital with an established home dialysis program. This makes the within-sample findings unique, since the nephrology clinic was equipped with a home dialysis team that provided pre-dialysis education and training for home dialysis use. Thus, the present sample theoretically had the resources to learn about dialysis options prior to and following dialysis initiation. Given the availability of these resources, these findings show that healthcare access and engagement before and after entering nephrology care played a critical role in what dialysis modality patients would eventually undergo.
Findings from the present study are consistent with other qualitative studies outside of the US. One study from Canada (
N = 12) that interviewed patients undergoing in-center hemodialysis (
n = 4), home hemodialysis (
n = 4), and PD (
n = 4) discussed the difficulty in reaching a quality decision when faced with a limited timeframe [
33]. Another study from the Netherlands that interviewed thirteen participants who followed through with a dialysis decision described the important role of nephrology providers in the provision of dialysis information, training, and support when initiating home-based dialysis methods [
34]. A qualitative study describing the shared decision-making experiences among patients undergoing in-center dialysis in Australia (
N = 35) found that participants expressed passivity during their involvement with healthcare-related decisions. Further, primarily Arabic-speaking participants seemed to experience some difficulties with quality decision-making due to language barriers [
35]. In conjunction with the present study, these findings suggest there is a need to ensure patients undergoing dialysis are experiencing high quality decision-making even if their approach is passive. Patients who experience communication barriers due to language differences may require greater language interpretation support from professionals in healthcare settings.
Race and ethnicity emerged as a category associated with dialysis modality and differing healthcare experiences. White and non-Hispanic patients made up 75% of the present home dialysis sample but only 30% of the in-center dialysis sample. Based on the qualitative responses, Black and Latine participants experienced less preventive general and kidney-related healthcare compared to White and non-Hispanic participants. Considering the organizational support available at this clinic, these results allude to barriers associated with healthcare access and engagement among patients with CKD who are Black and Latine, and broadly, racial/ethnic inequities in healthcare [
36‐
41]. Although reflective of a small sample, these findings are consistent with national trends. Among national incident dialysis cases, 12.04% of White patients utilized home dialysis, versus 9.56% of Black patients, and 11.79% of non-Hispanic patients, versus 10.60% of Hispanic patients. Among prevalent dialysis cases, 14.08% of White patients utilize home dialysis versus 9.29% of Black patients, and 13.07% of non-Hispanic patients versus 9.90% of Hispanic patients [
1]. All five Spanish-speaking patients of limited English proficiency were undergoing in-center dialysis (25% of in-center study sample), which hints at potential obstacles getting patients with limited English proficiency interested in home dialysis. This finding is consistent with other studies identifying language barriers within kidney care, and a broad lack of translation and interpretation support in healthcare [
42,
43].
Limitations
Given the cross-sectional, exploratory nature of this study, causal statements cannot be made, but relationships between factors may be suggested. Because the study required participants to reflect on their experiences retrospectively, their responses were susceptible to recall bias. Although one of the qualitative findings describe the patient approach to decision-making as passive, the interviewer was not present during any patient-provider conversations and could only draw conclusions based on participant descriptions and impressions of the decision-making process. Therefore, these findings should be interpreted with caution. Not every eligible patient within the clinic was interviewed, however, the demographic characteristics of those who declined were similar to the present sample. Some patients within the in-center dialysis clinic declined participation because they felt too tired to participate in an interview while they were dialyzing, so their perspectives were not reflected in the findings. Furthermore, only two participants undergoing home hemodialysis were interviewed. While this low rate included in the sample is consistent with national trends [
1], and their responses were consistent with the rest of the home dialysis group, it may be worthwhile for future qualitative studies to confirm whether patients undergoing home hemodialysis and PD have comparable quality of life and decision-making experiences.
All participants came from a tertiary care research hospital in a Midwestern, largely rural US state, where they had access to dialysis education via home dialysis providers and the Kidney Options class offered by the nephrology clinic, even though many dialysis clinics in the US do not offer home dialysis nor do they offer dialysis education to their patients [
44]. Therefore, findings may not generalize to other US patients undergoing dialysis. However, these results may be used to generate hypotheses and quantify these relationships in larger samples of patients within cross-sectional, longitudinal and/or multi-site studies.
Clinical implications
Overall, these findings lay a foundation for future directions for improving patient outcomes in nephrology care. Participant responses suggested that healthcare access affected one’s future modality selection, and furthermore, healthcare access during pre-nephrology care, pre-ESKD nephrology care, and following dialysis initiation were potentially equally influential. Although past interventions have prioritized pre-ESKD nephrology care as the setting for dialysis education to occur [
11,
13,
15‐
18], these findings suggest that broadening the timeframe of messaging surrounding dialysis options to include the other healthcare stages (e.g., Fig.
1) may be effective in increasing home dialysis selection. Post-dialysis initiation may be a particularly pivotal stage because patients can better evaluate whether their fatigue levels, fistula/catheter access sites, and dialyzing schedules are compatible with their lifestyles, and be better equipped to engage more proactively during shared decision-making processes. Providing dialysis information throughout other stages of healthcare may improve patient agency in decision-making, since urgency will be lower as it will not be occurring just before ESKD is reached, thus giving patients more time to weigh their preferences alongside nephrology providers’ recommendations. Special consideration and additional resources should be prioritized for patients of underrepresented races and ethnicities, particularly within nephrology care, since they may have faced greater healthcare challenges leading up to dialysis initiation.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.