Background

Chronic kidney disease (CKD) among older patients is common in the United States and the prevalence among those aged ≥65 has increased from approximately 3 to 14% in the last two decades [1]. Among patients who ultimately develop end-stage renal disease (ESRD), many are referred by their nephrologists for educational sessions to learn about different treatment options such as dialysis. However, ESRD patient education programs tend to focus on modality type, vascular access, and care setting rather than on how the different treatment modalities may affect patient lifestyles and goals of care [2,3,4]. Few patients with advanced CKD are aware of treatment options such as conservative kidney management or advance care planning and seldom have an opportunity to discuss preferences with their clinicians or loved ones [5,6,7]. One recent study of older patients with advanced CKD and ESRD showed that many did not fully understand dialysis and perceived it as the only alternative to death [6].

Emerging models of care demonstrate that integration of palliative care within nephrology practices can help promote conservative kidney management and advance care planning and also improve patient satisfaction among older patients; however, implementation has proved to be challenging [8, 9]. Previous research has demonstrated numerous clinician barriers to providing conservative kidney management and facilitating advance care planning including prognostic uncertainty, inconsistent collaboration between nephrologists and primary care physicians (PCP), and limited knowledge of ESRD treatment options [10,11,12]. Although there is some literature that explores perceptions of ESRD treatment and end-of-life care options among older patients with advanced CKD and ESRD, it is unclear whether their views are consistent with their clinicians [6, 13,14,15,16,17,18,19]. To fill this knowledge gap, we sought to explore the alignment between clinician views of their responsibilities and patient clinical experiences via qualitative interviews with nephrologists, PCPs, patients, and caregivers regarding ESRD treatments and advance care planning.

Methods

Study design and participant population

We conducted this qualitative study between 3/2017 and 5/2018 at Massachusetts General Hospital in Boston, Massachusetts. Eligibility criteria included a convenience sample of PCPs and nephrologists who cared for older patients with advanced CKD (aged ≥65 with estimated glomerular filtration rate < 20 as defined by the CKD Epidemiology Collaboration for estimation of GFR) who were not on dialysis [20].

PCPs and nephrologists were recruited via email from primary care and nephrology clinics at Massachusetts General Hospital. All clinicians had a professional working relationship with N.D.E and all clinicians who were approached agreed to participate in the study. Written informed consent was subsequently obtained by the primary investigator (N.D.E, MD – nephrologist, female]). A telephone interview was then scheduled by A.K.L, (PhD – psychologist, female). A study coordinator (S.P, BA - study coordinator, female) approached patients with advanced CKD and their caregivers on the day they were scheduled to receive in-person dialysis education from trained nurses in the outpatient nephrology clinic. None of the patients had a relationship with N.D.E. prior to recruitment. Out of 15 patients who were approached, 3 patients were not interested in the study. Twelve patients were consented, however 2 were withdrawn from the study as one was hospitalized and the other was lost to follow-up. Out of 10 enrolled patients, 5 patients had at least one caregiver who was interested in participating in the study. The study coordinator obtained written informed consent for all participants and scheduled the phone interview. All phone interviews took place in a quiet setting of the participant’s choice. Patient and caregivers were interviewed separately. No other individuals were present during the phone interviews outside of the study participant and interviewer (A.K.L).

Investigative team members, all trained in qualitative methods, (N.D.E, A.K.L and E.R.P [PhD – psychologist, female]) developed both clinician and patient/caregiver semi-structured interview guides (See Additional file 1). Both guides were pilot tested among clinicians (n = 3) and patients (n = 5) and modified based on feedback prior to use in the study. All interviews were conducted and audiotaped by the same member of the investigative team (A.K.L). There were no field notes recorded during the interview as they were conducted by telephone. One clinician interview (nephrologist) was repeated due to audiotape malfunction. Prior to the interview, none of the participants knew A.K.L or had any knowledge of her research interests. Interviews were conducted among the convenience sample who represented varied perspectives. The Institutional Review Boards at Partners HealthCare and the University of Pennsylvania approved this study.

Analyses

TranscribeMe transcribed all audio recordings [21]. Transcripts were not returned to participants for comment or correction. Team members N.D.E, A.K.L, and T.A.K (PhD – researcher trained in qualitative methods, female,) initially reviewed all the transcripts and created two codebooks (clinician and patient/caregiver) based on content that arose in the interviews. Transcripts were analyzed using Nvivo (version 12; QSR international, Melbourne, Australia). The codebook was iteratively revised after independent coding of initial transcripts by team members N.D.E, A.K.L, and T.L.S (MS – researcher trained in qualitative methods, female). After consensus meetings with all team members, the final codebooks were used to code the remaining transcripts (clinician transcripts – N.D.E and A.K.L; patient and caregiver transcripts – N.D.E and T.L.S). Emergent themes beyond those included in the interview guide were identified and discussed during the coding process. After reviewing all transcripts, themes were generally consistent across varied perspectives within the sample [22]. We determined that we had reached thematic saturation at this point [14, 22, 23]. Coding discrepancies were resolved via consensus. Participants did not provide feedback on findings.

Results

Mean time for clinician interviews (nephrologists n = 8, primary care clinicians n = 8) was 28 min and 3 s; 12 min and 18 s for patient interviews (n = 10); and 11 min and 22 s for caregiver interviews (n = 5). Demographic data for all participants are displayed in Table 1. We found a clear difference in alignment of experiences and expectations between clinicians and patients and their caregivers (Fig. 1).

Table 1 Participant characteristics
Full size table
Fig. 1
figure 1

Alignment of clinician and patient/caregiver expectations and experiences

Full size image

Themes

Three major themes emerged from the data.

Dialysis discussions

Nephrologists and primary care clinicians felt that nephrologists’ primary role was to manage all aspects of care specific to kidney disease and initiate dialysis discussions (Table 2). Several nephrologists remarked that they educated all of their patients about dialysis regardless of whether this would be beneficial to the patient. Consideration of lifestyle changes and quality of life were usually not strongly emphasized in dialysis discussions.

Table 2 Dialysis discussions
Full size table

In terms of the role of PCPs, many PCPs and nephrologists commented that PCPs valued providing continuity of care in their patients’ lives – especially in the setting of those who started dialysis and had to establish relationships with new nephrologists. Some PCPs and nephrologists felt the PCP’s primary role was to ensure their patients understood how different treatments would impact their livelihood in addition to advocating for them as they navigated the healthcare system.

With regard to dialysis discussions, patients and their caregivers expressed positive sentiments about the quality of information that was provided and that they also trusted their clinicians’ specific recommendations for dialysis. However, most only commented on being educated about the technical aspects of dialysis. These included modality and dialysis setting, access placement, and frequency of treatments. Many were not aware of alternatives to dialysis treatments and felt hopeless about their lack of options.

Dialysis decision-making

Often, nephrologists and PCPs felt that nephrologists were the most knowledgeable about which patients would not do well with dialysis therapy. Many nephrologists used frameworks that included age, and the presence of frailty and/or life-limiting illnesses when considering conservative kidney management (Table 3). Many nephrologists remarked that selecting treatments for renal disease based on these parameters was a central role in their clinical practice. In terms of the role of PCPs, nephrologists deferred to PCPs to integrate treatment decision making with patients’ goals and also provide continuity of care.

Table 3 Dialysis decision-making
Full size table

However, although PCPs agreed that they should primarily help patients navigate their renal disease, several PCPs remarked that they struggled with ESRD treatment decisions for their older and frail patients with advanced CKD. For example, some PCPs felt that dialysis may not be the best treatment for their patients, but PCPs also did not feel that this decision was not for them to make. Many commented that deferring this decision to the nephrologist made the most sense. Others felt that decisions about dialysis or conservative kidney management should be made in conjunction with the nephrologist.

Although clinicians agreed about their respective roles in dialysis decision-making with their patients, patients and caregivers did not have a clear understanding of the implications of dialysis and expectations varied. For instance, many expressed concerns about the general effects of dialysis on their lifestyles. Other respondents were not sure what to anticipate from dialysis. Many expressed hopes that dialysis would make them feel better while others longed to live longer and normal lives.

Processes of advance care planning

Nephrologists felt that advance care planning discussions should primarily be led by PCPs, and many remarked that they did not feel comfortable leading these discussions (Table 4). Some felt that advance care planning discussions should be conducted in the setting of an interdisciplinary meeting that included the patient, caregivers, and the clinicians most involved in that patient’s care.

Table 4 Processes of advance care planning
Full size table

PCPs agreed with nephrologists about the role of discussing advance care planning and many PCPs felt they should initiate and drive goals of care discussions. Some PCPs also remarked that conducting goals of care conversations in conjunction with nephrologists would be ideal. Some PCPs felt that discussions should occur before or at the same time as renal replacement therapy discussions.

In contrast to clinicians’ clarity about the process of discussing advance care planning, many patients and caregivers had not discussed their end-of-life wishes with their clinicians. Some patients had discussed their preferences with their caregivers and had also completed advance directives. Of those who had completed a health care proxy or living will, many reported they had done so during a hospitalization or before an elective surgery. Regardless of why they completed the form, only a few remarked on the value of having someone speak on their behalf and honoring their wishes if they became incapacitated. Patients and their caregivers also desired goals of care discussions that were honest and included the chances of survival if they were to choose dialysis. Some felt that these discussions should occur earlier in the disease course – before the time dialysis would be needed. Several also described wanting to have discussions with their family members at their nephrologists’ office or in their homes.

Discussion

Among nephrologists, primary care clinicians, older patients with advanced CKD and their caregivers, perspectives and expectations about ESRD treatments and advance care planning varied. Furthermore, although clinicians were mostly aligned on their respective roles with regard to communication and care delivery, patients and caregivers were unclear about several aspects of dialysis, alternatives to treatment, and advance care planning. Lastly, patients and caregivers specifically desired more information about treatment choices and advance care planning from their nephrologists.

In the United States, the infrastructure to effectively deliver nephrology care to older patients with advanced chronic kidney disease continues to evolve [24]. Conservative kidney management, which incorporates non-dialytic therapy that aims to slow progression of disease and treat a wide range of symptoms, is an approach that many clinicians are still learning about [5, 25, 26]. We found that both nephrologists and PCPs agreed that nephrologists should take the initiative in determining which patients are eligible for dialysis. However, PCPs struggled with dialysis decisions that had been made for patients who they personally felt would not fare well. Our results are consistent with Parvez et al. who found that, compared to nephrologists, PCPs were less knowledgeable about conservative kidney management treatment effectiveness and less confident about selection of suitable candidates [10]. However, this study also showed that nephrologists had difficulty with determining which of their patients would benefit from conservative kidney management. Ladin et al., described the “interpretative” approach to conservative kidney management (e.g., nephrologists serve as a treatment decision navigator for patients) as the best example of shared decision-making given that patients’ values and goals are incorporated into the process [27]. However, in this same study, a minority of nephrologists actually presented conservative kidney management as an option to their patients. As we found that PCPs felt strongly about guiding their patients through the decision-making process, a collaborative model in which both the PCP and nephrologist assume distinct roles (e.g., information provider and treatment decision navigator, respectively) could potentially improve ESRD decision-making for this patient population [12, 28,29,30].

Integrative models of primary and specialty care have also been described to improve the rates of goals of care discussions and advance care planning among seriously ill patients [31, 32]. In our study, nephrologists and PCPs felt that although advance care planning was an important aspect of care for their older patients, most believed that PCPs should lead these discussions. These findings counter work from O’Hare et al. where they found that among physicians from different specialties (e.g. nephrology, cardiology, primary care, etc.), there was an unclear understanding of who should carry out advance care planning [11]. We also found that the few patients and caregivers who had goals of care discussions or had completed advanced directives, had done so in an inpatient setting. However, a recent study showed that standard initiation of advance care planning in primary care practices was effectively implemented across a large integrated health care system using staff training and electronic medical record enhancements [33]. Notably, most clinicians in our study described primarily communicating about their shared patients through the electronic medical record or email. Therefore, structured primary care interventions that utilize telecommunication as an adjunct to in-person clinic visits could promote interdisciplinary collaboration and circumvent systemic barriers to advance care planning such as lack of time and fragmentation of care [11, 34,35,36,37].

Perhaps the most striking finding from our study was the lack of clarity about ESRD treatment options and subsequent effects on livelihood as well as advance care planning among patients and their caregivers. The majority of patients and their caregivers seemed confident only in their knowledge of the process of dialysis treatments and advance directive completion. In particular, older patients with advanced CKD have been shown to have knowledge gaps with regard to prognostic awareness and disease progression [17]. However, our findings showed that patients were also in need of more information about the direct effects of ESRD treatments on their quality of life and longevity. Patients and caregivers also desired more information from their nephrologists regarding advance care planning – especially in the setting of dialysis decisions. In one recent study of older adults with advanced CKD, patients described poor communication with their nephrologists about their disease as a source of worry about their future health and misconceptions about their illness [17]. Another study showed that skewed perceptions about ESRD and end-of-life treatments could lead to more intense care at the end of life for patients with kidney disease [14]. These findings, in addition to our own, highlight the need for improved communication between nephrologists, PCPs, and their patients to fulfill the psychosocial needs of patients in addition to technical information requests. Current clinician training programs on how to effectively conduct goals of care discussions with seriously ill patients with kidney disease show self-reported improvement and confidence in communication skills [38]. However, more research is needed to also evaluate the quality of communication and whether concerns regarding illness and treatment effects are adequately addressed among older patients and their caregivers.

This study has a few limitations. First, the modest sample size limits transferability of our findings to clinicians, patients, and caregivers in different settings. Second, although we reached thematic saturation, it is possible that we could have identified additional observations with more interviews. Third, conversations about ESRD treatments with patients and caregivers mostly centered around dialysis options and we were thus unable to glean insights into perspectives about non-dialytic treatments. Lastly, as we only included nephrologists and PCPs, we were unable to make inferences about experiences and perceptions among other members of the health care team.

Conclusions

This is the first study to our knowledge to comprehensively examine experiences around ESRD treatments and advance care planning discussions between clinicians, patients with advanced CKD and their caregivers. Our findings identify an important tension between clinicians’ perceptions of communication and care delivery and unmet needs of patients and their caregivers. Further research is needed to develop and implement collaborative models of care to holistically care for older patients with advanced CKD and improve understanding of conservative kidney management and advance care planning.