Article Text

Serious Illness Conversations in advanced kidney disease: a mixed-methods implementation study
  1. Natanong Thamcharoen1,2,
  2. Pitchaphon Nissaisorakarn3,
  3. Robert A Cohen3 and
  4. Mara A Schonberg4
  1. 1Cheewabhibaln Palliative Care Center, King Chulalongkorn Memorial Hospital, Bangkok, Thailand
  2. 2Nephrology Division, Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
  3. 3Nephrology Division, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA
  4. 4Division of General Medicine and Primary Care, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Natanong Thamcharoen, Cheewabhibaln Palliative Care Center, King Chulalongkorn Memorial Hospital, Bangkok 10330, Thailand; natthamcharoen{at}


Objective Advanced kidney disease is associated with a high risk of morbidity and mortality. Consequently, invasive treatments such as dialysis may not yield survival benefits. Advance care planning has been encouraged. However, whether such discussions are acceptable when done earlier, before end-stage kidney treatment decision-making occurs, is unclear. This pilot study aimed to explore whether use of the Serious Illness Conversation Guide to aid early advance care planning is acceptable, and to evaluate the information gained from these conversations.

Methods Patients with advanced kidney disease (stage 3B and above) and high mortality risk at 2 years were enrolled in this mixed-methods study from an academic nephrology clinic. Semi-structured interviews were conducted using the adapted Serious Illness Conversation Guide. Thematic analysis was used to assess patients’ perceptions of the conversation. Participants completed a questionnaire assessing conversation acceptability.

Results Twenty-six patients participated, 50% were female. Participants felt that the conversation guide helped them reflect on their prognosis, goals of care and treatment preferences. Most did not feel that the conversation provoked anxiety (23/26, 88%) nor that it decreased hopefulness (24/26, 92%). Some challenges were elicited; patients expressed cognitive dissonance with the kidney disease severity due to lack of symptoms; had difficulty conceptualising their goals of care; and vocalised fear of personal failure without attempting dialysis.

Conclusions Patients in this pilot study found the adapted Serious Illness Conversation Guide acceptable. This guide may be used with patients early in the course of advanced kidney disease to gather information for future advanced care planning.

  • chronic conditions
  • renal failure
  • communication
  • clinical decisions

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

What was already known?

  • Advance care planning (ACP) has been encouraged in advanced kidney disease (AKD).

  • The Serious Illness Conversation Guide (SICG) is a conversation guide to assist providers in ACP.

What are the new findings?

  • The use of the adapted SICG earlier in the course of AKD did not significantly affect participants’ emotional state. Important informations to aid future ACP were gained from the conversations.

  • Cognitive dissonance of kidney disease severity, difficulties in conceptualising goals of care and fear of earlier mortality if dialysis is not attempted are challenges for ACP using SICG in these participants.

What is their significance?

  • Clinical: Clinicians should routinely assess patient’s readiness to ACP.

  • Research: Future studies should aim to enrol a larger group of participants to confirm the acceptability of the SICG use in AKD population.


Advanced kidney disease (AKD) is a serious illness associated with heightened mortality and morbidity. Older adults with AKD are more likely to die before reaching end-stage kidney disease (ESKD), with the risk of death reportedly 40 times greater than the risk of reaching ESKD in patients older than 85 years with AKD. For individuals older than 80 years with multimorbidity, dialysis may not confer a survival advantage compared with conservative care (CC).1 Further, dialysis may not improve quality of life (QOL) or functional status. Individuals who reach ESKD are more susceptible to increased hospitalisations and functional setbacks while on dialysis, especially older patients who often have multiple comorbidities.2 Unfortunately, providers rarely incorporate this prognostic information when ESKD treatment is being discussed and patients often choose dialysis or other invasive treatments without considering their values, goals or priorities. Discordance between providers’ perceptions of patients’ priorities and patients’ actual reported priorities exists.3 Despite data indicating patient care preferences that centre on pain relief, QOL and the wish to die at home,4 most older adults with ESKD spend the final month of life in the hospital and fewer use hospice compared with patients with heart failure or cancer.5

Advance care planning (ACP) is a clinician-led process in which patients ponder future health states and care concordant with elicited goals and values.6 Such conversations have been associated with a higher likelihood of reduced hospitalisations, dying in a patient’s setting of choice and less aggressive care at the end of life.7 Experts agree that ACP is integral to the care and medical decision-making process of patients with AKD.8 9 ACP early in the course of AKD to enhance shared decision-making has been advised.4 9 Unfortunately, the majority of patients with ESKD do not have ACP documented10 and ACP appears to occur less frequently than in patients with comparable serious illnesses.10

The Serious Illness Conversation Guide (SICG),11 an ACP tool developed by the Dana Farber Cancer Institute and Ariadne Labs, is a conversation guide that was developed to assist providers in exploring patient preferences of care when they become seriously ill. It includes structured open-ended questions which providers and patients review together. It has been tested and found acceptable in various clinical settings including in oncology clinics, long-term acute care hospitals (LTACHs) and in primary care practices.12 13 However, it is unclear whether the use of the SICG to aid earlier ACP is acceptable to patients with AKD before the need to discuss ESKD arises and how much information may be gleaned from these patients using the SICG. To address these questions, we piloted the use of the SICG among patients with AKD before decisions about ESKD treatment occured.


Study design

Investigators planned a mixed-methods study of patients with AKD to better understand their perceptions when discussing their goals of care using the SICG. The SICG was slightly adapted for use with this population (figure 1). Specifically, a question asking participants to describe their current quality of life was added while information on the patient’s prognosis and a treatment recommendation was not offered since the interviewer was not participating in the patient’s direct care. Participants completed this adapted SICG in person with a study investigator (NT) who was a nephrology and palliative care trainee at the time of the study and had completed a SICG training course.

Figure 1

Adapted Serious Illness Conversation Guide.

After the interview, each participant completed an acceptability questionnaire (online supplemental 1). Participants were followed via their electronic medical records (EMRs) for vital status through 30 June 2020, which was at least 24 weeks from the last patient interview. EMR data were reviewed for whether a healthcare proxy or medical order for life-sustaining treatment (MOLST) form had been completed.


A list of patients followed in a nephrology clinic at a large academic medical centre in Boston, MA were screened from February 2019 through January 2020. Eligible patients were English-speaking adults (>18 years of age) with chronic kidney disease (CKD) stage 3B–5 (eGFR ≤45 mL/min/1.73 m2). They were also required to have advanced organ failure or metastatic cancer (box 1), indicating high mortality risk in 2 years based on two validated risk models.14 15 Based on EMR data, patients with moderate to advanced cognitive impairment or who lacked capacity to make medical decisions were excluded. Eligible patients were sent a study informational letter with information on how to opt-out of being contacted for the study. Patients who did not opt out were contacted by telephone to obtain verbal consent.

Box 1

Eligibility criteria

Patients who are older than 18 years old, English speaking with CKD stage ≥ 3B (eGFR ≤45 mL/min/1.73 m2) with the following criteria:

  1. Age ≥80 years


  1. Age ≥70 years with diabetes or cardiovascular disease


  1. Any age with other advanced stage organ diseases

    1. Heart failure with New York Heart Association class III or IV

    2. Severe chronic obstructive pulmonary disease (COPD)

    3. Cirrhosis with child class C or Model for End-Stage Liver Disease (MELD) score ≥17


  1. Any age with metastatic cancer

  2. Any age with CKD stage 4 or 5

Data collection

Interviews occurred in a private clinical space, immediately after a regularly scheduled nephrology clinic appointment. Participants’ family members or friends were permitted to accompany the patient during the interview. All interviews were timed, audio-recorded and transcribed verbatim. After the interview, participants completed the acceptability questionnaire. Using 5-point Likert scales, the questionnaire assessed how the conversation affected participant’s anxiety level (decreased a lot to increased a lot), sense of hopefulness and peacefulness (decreased a lot to increased a lot), how helpful each question from SICG was (not at all helpful to extremely helpful), whether they liked the conversation (disliked a lot to liked a lot) and whether the conversation was worthwhile (not at all to extremely worthwhile). This questionnaire had been administered previously to assess the acceptability of SICG in tracheostomy patients.12 Participants were also asked about their activities of daily living (ADLs) and instrumental activities of daily living (IADLs).16 17 Participant demographic data, CKD stage and other comorbidities were abstracted from the EMR and their Charlson Comorbidity Index was calculated.18

Data analysis

Descriptive statistics were used to analyse quantitative data from patient surveys. To identify themes in participants’ comments during SICG interviews, a thematic analysis using both a deductive and inductive approach was performed from March 2020 to May 2020. A thematic analysis is a qualitative research method for identifying, analysing, organising, describing and reporting themes found within a data set.19 The deductive approach was informed by the inherent structure of the conversation guide while the inductive themes emerged from the comments of this unique group of patients with AKD. Themes were initially grouped into three categories based on the three domains for ACP (prognosis awareness, goals of care and factors impacting patient’s treatment decision-making). Then within each of these categories, new themes that emerged from the text were noted. NVivo V.12, a qualitative data analysis software, was used for coding. Initially, two investigators (NT, PN) independently reviewed five interview transcripts to generate a preliminary list of codes. Next, the team reviewed and finalised the codebook, with discrepancies in the meaning of codes resolved through consensus (online supplemental 2). Using the final codebook, NT and PN independently coded all interviews. To verify reliability, NT reviewed PN’s coded scripts. Any disagreements or possible emerging themes were discussed until consensus was reached. In interpretive analyses, the codes were regrouped by participants’ demographics (age, education level) and CKD stage (CKD 3B and 4–5) to determine if themes differed by these factors. Thematic saturation was achieved after reviewing 24 interview scripts; two additional participants were interviewed for confirmation of thematic saturation. Direct quotes from the data are used to illustrate themes.


One hundred and fifty-one potentially eligible patients were contacted through mail and 26 agreed to participate (figure 2). Patients who declined participation were similar in age, race and sex to participants (online supplemental 3). Mean age was 78 years (SD 7.8 years; age range 58–92), 19 (73%) were white and 13 (50%) had at minimum a college degree. Thirteen (50%) had CKD stage 4, seven (27%) had CKD stage 3B and six (23%) had CKD stage 5 (table 1). Eight participants (31%) had a family member join the interview. Mean interview duration was 22 min (SD 9.6, range 11–53 min). At the time of the interview, 24 participants (92%) had a healthcare proxy assigned and three participants (12%) had a completed MOLST. On average, participants were followed for 45 weeks (SD 20 weeks) via the EMR. Four of the 26 participants died within 24 weeks’ follow-up, none of whom had a MOLST on the interview day but two had a MOLST completed during the hospitalisation in which they died (comfort focused measures and DNR/DNI noted).

Table 1

Participant characteristics

SICG acceptability

Of the 26 participants, 22 (84%) liked the way the SICG conversation was introduced, 23 (88%) did not feel that the conversation provoked anxiety and 24 (92%) reported a sense of hopefulness about their future QOL. Most participants found all questions from the SICG helpful, ranging from 24 to 26 participants (92%–100%) for each question (table 2). Twenty-three participants (88%) found it at least somewhat helpful to be asked about their illness understanding and to discuss their personal goals for the future. However, only 13 (50%) reported having discussed health values and priorities with their doctors. Almost all participants (n=25, 96%) indicated that this was the right time for an ACP conversation and every participant reported it was worthwhile talking about these issues. Twenty-two participants (84%) wanted to continue this conversation with their doctors.

Table 2

Participants’ acceptability to the serious illness conversation

Thematic analysis

Themes were categorised based on the three domains for ACP (prognosis awareness, goals of care and factors impacting patient’s treatment decision-making).

Prognosis awareness

Asymptomatic nature of CKD resulting in unawareness of the disease severity

Unlike other diseases possessing a comparable poor prognosis such as advanced heart disease, for which patients frequently have uncomfortable symptoms (eg, dyspnoea, oedema or chest pain), AKD is often asymptomatic. Although participants were typically aware that AKD is a serious condition, the absence of symptoms called into question its severity and therefore created cognitive dissonance.

How I feel is terrific. From what I understand from doctors and nurses, I'm sick. I just don't feel sick. (Patient 10)

Mortality acceptance

Several participants reflected on their own mortality and expressed acceptance. However, these participants were all ≥70 years. These patients also acknowledged that ageing contributes to functional decline and that some conditions are not curable.

As you get old you imagine all sorts of things going wrong. (Patient 22)

These patients were also more likely to acknowledge that they were approaching mortality.

I'm 88 years old. I think that’s a blessing. You don't live forever, so I mean you're always worried about the endgame. (Patient 18)

A few participants attributed determination of eventual mortality to a higher power. God would determine their longevity, not doctors. Hence, they were less likely to discuss plans for the future during the conversation.

It’s up to Him when I go. I'll go to bed and He'll take me in the quiet of night. No pulling the plug for me. (Patient 24)

Goals of care

Conceptualising goals of care proved challenging

Participants frequently reported longevity as a priority when asked the question about most important goals. However, when later in the interview asked to identify the most critical ability or what matters most, participants regarded focusing on quality of life as their primary goal reflected by such stated priorities as the ability to think clearly, meaningfully interact with loved ones, being independent, being at home (or not being in a facility) and not being a burden to family. Many patients expressed not wanting their life to be extended if they were in a vegetative state.

I want to live. But I would prefer to be not just be laying there. Nobody wants to be a vegetable. (Patient 26)

As well as longevity, focusing on improving present health were also common initial answers when questioned about their most important goal or priority.

If it was something that I could fix by changing my lifestyle, I'd like to take that extra step. (Patient 10)

For a few participants, despite the emphasis on quality of life expressed, they insisted on living as long as possible regardless of the seriousness of their condition.

Put me on ice. You keep me alive until they find the treatment. I don’t want to go any faster than I have to (Patient 24)

Factors impacting patient’s treatment decision-making

Contending with a fear of failure

Dialysis was described as the inevitable kidney failure treatment option by participants, with the caveat that everything should be done to prevent the need for this intervention. Participants felt uncomfortable with perceptions that they might be giving up easily if they did not accept dialysis when facing ESKD. They wanted to fight to maintain the status quo.

If there’s anything I can do to remedy or make it better, I'll think about what could be done. (Patient 9)

However, if dialysis initiation is under consideration, they would agree to it with the belief or hope of its life-extending capacity despite their negative impressions of dialysis.

If I had dialysis, that would be extremely unhappy for me. But I would do it if it’s necessary.” (Patient 3).

On the other hand, participants recognised that they may experience worsening function. Participants were aware of the potential trade-offs with dialysis and specifically focused on a worry that it might worsen QOL or functional status. However, some participants said they might adjust to a diminished functional level within certain limits.

Even if I have to stay in a wheelchair, sometimes I just have to get around. If I can take care of myself to the limit, all right. But it’s not like I’d totally give up. I’m not one to totally throw in the towel and just say ‘The hell with it. The day’s over’. (Patient 17)

Some participants believed their doctors described dialysis too favourably.

He sort of made it sound like it wasn’t all that bad. But I’m not sure that’s true. (Patient 18).

Trial of treatment as an acceptable approach

Participants would not mind invasive or aggressive treatments if there is a chance to recover an acceptable functional status.

I don't know what dialysis is like. But I’d obviously try it. If it makes you more miserable than staying alive, I will think about it again. (Patient 18)

However, if a treatment was deemed futile or they were unable to recover to an acceptable functional level, participants felt it was acceptable to consider stopping the treatment.

I would not want extra measures taken simply to prolong life—if the quality of life just wasn’t there. (Patient 16)


In this small pilot study, nearly all patients with AKD found using the adapted SICG helpful for discussing ACP and perceived that it did not worsen anxiety or affect the level of hopefulness or peacefulness. These findings are comparable with what has been reported in studies of other illness populations, including in patients with cancer and patients with tracheostomy in a LTACH.12 13 However, low participation rate (26 out of 151) in this study raises the concern that most patients are either not familiar with or have a unfavourable view of ACP, despite several studies reporting that most patients wish to have ACP discussions with their doctors.20 Nevertheless, this study suggests that patients who agree to have these conversation find them acceptable. Therefore, clinicians should routinely assess patient’s readiness to ACP and not hesitate to initiate it in patients who are ready.8 20 Time constraints have been reported as a barrier for discussing ACP; however, the mean interview duration was 22 min, suggesting the feasibility of using the SICG during an encounter. MOLST forms were completed in only three participants in our study, reflecting the low rate of ACP in this AKD population. Four out of 26 participants died during follow-up and two required ICU care. These data highlight the high mortality risk of this population and the potential benefit of ACP in the AKD population.

Use of the adapted SICG enabled participants to share their prognosis awareness, priorities, hopes and worries, which are ACP primary components. This pilot study suggests that SICG is an effective tool for aiding clinicians in initiating ACP conversations earlier in the course of AKD.8 In addition, survey results indicated that most participants found the conversations worthwhile and without negative emotional impact. It prompted consideration of life priorities in the context of worsening health for these participants, most of whom had never contemplated these issues previously. Only half had a prior ACP discussion with their doctors. Since ACP is an iterative process, such information gained from these early conversations serves as a vital foundation for future discussions.

This study also identified challenges in using the SICG for ACP conversations in patients with AKD. Participants had difficulty reconciling the serious nature of CKD due to the often asymptomatic nature of this disease until late in its course.21 Hence, patients may face difficulty acknowledging that AKD affects mortality risk when having ACP conversations earlier in the disease trajectory, particularly if clinicians avoid discussing prognosis. Studies have indicated that patients with AKD tend to be more optimistic about their prognosis than clinician and thus may be unprepared when they need to make a treatment decision when illness worsens to a more serious stage.22 23 In this study, older participants (>75 years) acknowledged the toll of ageing on their function and were more accepting of their mortality, suggesting that these patients may be more comfortable with ACP conversations.

Addressing care goals early in the course of AKD might be additionally challenging since participants tended, initially in the conversation, to anchor on current health status and priorities. One possible explanation is this tendency represents a coping mechanism by which discussion of unpleasant circumstances is avoided.23 However, when participants’ goals were explored further by specifically asking them as part of the SICG to consider life priorities in the context of future worsening health, or asking about the abilities critical to a meaningful life, responses shifted accordingly. These data suggest that by following the SICG format, clinicians might gain more complete and deeper information related to ACP, compared with discussing it without a conversation guide.

Clinicians often present dialysis as a beneficial treatment option, while minimising potential burdens.24 Few share that it may not improve longevity or QOL in certain patients. Interestingly, none of the participants in our study mentioned conservertive care (CC) as an option. This suggests that patients usually regard dialysis as the norm and that few are aware of CC.21 Admittedly, multiple barriers preclude routine discussion of CC as a possible treatment option for ESKD. One obstacle is the dearth of education about and communication skills for effective conversations regarding CC.25 Another potential barrier is the patient and clinician perception that death is a failure to be avoided, resulting in overly aggressive or futile treatment that disregards patients’ care goals.26 The emotional burden associated with discussing end of life may also prevent some clinicians from presenting CC to patients with the default instead to recommend dialysis, as in stating, “just do it” or at least “just try it”.26

In this study, participants expressed some reluctance to continue treatments that are unlikely to help them achieve their priorities or desired QOL. Hence, in the circumstance when patient prognosis is uncertain or when a strong desire to achieve a certain goal is expressed, a trial of dialysis might be proposed as a means of attempting to reach this outcome while simultaneously acknowledging that it can be stopped if QOL is compromised or more setbacks occur. For patients in the study who worried that not doing dialysis would amount to personal failure, framing this intervention as a time-limited trial might help them overcome this negative perception.

Shared decision-making, a process by which patients are provided information about prognosis and the burdens and benefits of certain treatments and also have priorities and concerns elicited with the goal of incorporating this information into treatment decisions, is considered a fundamental approach to be applied by nephrologists who care for patients with AKD.9 The SICG is a valuable tool that enables shared decision-making.8 Although designed for use with minimal training, another barrier to implementation of SICG is the need for more widespread education of nephrologists in basic communication skills necessary for shared decision-making conversations.

This study has several limitations. First, this is a pilot study in a single centre, limiting its generalisability. Also, since we aimed to recruit patients with high mortality risk who could benefit from ACP, only two of our patients were <70. Thus, the acceptability result and information from the conversation might not represent responses from younger patients. Second, interviews were conducted by only one individual trained in ACP communication and the use of the SICG. In other settings, the information gained might be different depending on interviewer skill level. Third, although we interviewed participants until saturation had been reached, the number of participants was small. Therefore, the rate of perceived acceptability of the SICG might be different if assessed in a larger cohort of patients with AKD. We also anticipate the impact of selection bias on acceptability, as participants who agreed to be interviewed may be more inclined to accept ACP conversations at baseline. Lastly, because the SICG was conducted with an investigator and not a care provider, patient prognosis was not discussed. Including discussions of prognosis may change how participants respond to questions about priorities and goals of care and how they perceive these conversations. Future studies should aim to enrol a larger group of participants to confirm the acceptability of the SICG use in AKD population.

In summary, this pilot study found that use of the adapted SICG for ACP earlier in the course of AKD did not significantly affect patient’s anxiety level, sense of peacefulness or hopefulness, and resulted in patients sharing valuable information about their values and goals. Since ACP is an interative process, such conversations should serve as a foundation for future ones, particularly those in which ESKD treatments need to be addressed. Some challenges for these conversations in patients with AKD, however, were identified. These included cognitive dissonance around kidney disease severity due to the lack of symptoms, difficulties in conceptualising goals of care and fear of earlier mortality if dialysis is not attempted. Use of the adapted SICG earlier in the trajectory of AKD has the promise of unearthing these issues and also intiating the process of learning patient preferences. This should ultimately assist our patients in making value concordant treatment decisions for ESKD and end-of-life approaches.


We would like to thank Anna Johansson, PhD and Rachel Bernacki, MD for the recommendation for the study. We would like to thank the participants and the clinic staffs who supported the study.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • RAC and MAS are joint senior authors.

  • Contributors Research idea and study design: RAC, NT. Data acquisition: NT, PN. Data analysis/interpretation: NT, PN, RAC, MAS. Statistical analysis: NT: Supervision or mentorship: RAC, MAS. RAC and MAS are holding equal co-senior authorship. Each author contributed important intellectual content during manuscript drafting or revision, accepts personal accountability for the author’s own contributions, and agrees to ensure that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The Beth Israel Deaconess Medical Center Institutional Review Board approved this study.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on reasonable request.